Issues, Theory, and Research in Industrial Organizational Psychology

ISSUES, THEORY, AND RESEARCH IN INDUSTRIAL /ORGANIZATIONAL PSYCHOLOGY

ADVANCES IN PSYCHOLOGY 82 Editors:

G . E. STELMACH

P. A. VROON

NORTH-HOLLAND AMSTERDAM LONDON NEW YORK * TOKYO

ISSUES, THEORY, AND RESEARCH IN

INDUSTRIAL/ORGANIZATIONAL PSYCHOLOGY Edited by

Kathryn KELLEY Department of Psychology University at Albany State University of New York Albany, New York, U . S . A .

1992

NORTH-HOLLAND AMSTERDAM LONDON NEW YORK TOKYO

NORTH-HOLLAND ELSEVIER SCIENCE PUBLISHERS B.V. Sara Burgerhartstraat 25 P.O. Box 21 1, 1000 AE Amsterdam, The Netherlands

Distributors for the United States and Canada: ELSEVIER SCIENCE PUBLISHING COMPANY, INC. 655 Avenue of the Americas New York, N.Y. 10010. U.S.A.

Ltbrary o f Congress Cataloging-in-Publication Data

I s s u e s , t h e o r y , a n d r s s e a r c h in i n d u s t r i a l / o r g a n i z a t i o n a l p s y c h o l o g y / e d i t e d by K a t h r y n K e l l e y . p. c m . -- ( A d v a n c e s in p s y c h o l o g y ; 82) Includes bibliographical references and indexes. ISBN 0-444-88777-6 1 . P s y c h o l o g y , I n d u s t r i a l . 2. O r g a n i z a t i o n a l b e h a v i o r . I. K e l l e y . K a t h r y n . 11. S e r i e s : A d v a n c e s in p s y c h o l o g y ( A m s t e r d a m . N e t h e r l a n d s ) ; 82. HF5548.8.178 1992 302.3'5--dc20

9 1-38439

CIP

ISBN: 0 444 88777 6 1992 ELSEVIER SCIENCE PUBLISHERS B.V. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the publisher, Elsevier Science Publishers B.V., Permissions Department, P.O. Box 521, 1000 A M Amsterdam, The Netherlands. Special regulations for readers in the U.S.A. - This publication has been registered with the Copyright Clearance Center Inc. (CCC), Salem, Massachusetts. Information can be obtained from the CCC about conditions under which photocopies of parts of this publication may be made in the U.S.A. All other copyright questions, including photocopying outside of the U.S.A., should be referred to the copyright owner. Elsevier Science Publishers B.V., unless otherwise specified. No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability. negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Printed in The Netherlands

V

PREFACE Industrial/Organizational psychologists are a rather diverse group of people bound together, simply, by a common interest in applying psychology to work settings. This is the conclusion reached by George Alliger in the opening chapter of this edited volume, and it sets the tone for the rest of the book, for this book does surely attempt to expand our view of what can be considered I/O psychology. The authors of the individual chapters come from a variety of backgrounds, not all of them associated with I/O psychology, and they discuss topics that range from more traditional I/O topics such as managerial success and training, to topics much more on the edge of I/O such as team-building and organizational theory. Thus, this volume makes an important statement about the potential diversity of our field and what it is becoming. At the same time, this volume will help move us towards that diversity by providing insights and information in areas that should be, and are becoming part of the realm of 110 psychology. These insights into non-traditional topics, as well as particularly interesting approaches to more traditional areas, make this volume of worthwhile and useful to almost anyone concerned with I/O psychology. As noted above, the volume opens with a chapter by George Alliger, which deals with the role of theory in I/O psychology. In it he notes that I/O psychology cannot be characterized by a "grand theory", but is instead, characterized by many smaller theories that apply to specific aspects of behavior. He neither condemns I/O psychologists for this, nor does he apologize. Instead, he simply notes that the field has always been characterized by diversity and that is quite healthy. Of course, this is exactly the proper opening note to sound because diversity is the hallmark of this volume. Alliger's chapter is followed by a chapter, by Byrne and Neuman, that discusses (within the clever framework of the development and breakup of a relationship) how attraction research can contribute to research in such areas as selection, socialization, performance appraisal, and turnover. The richly developed theoretical framework of interpersonal attraction should go far to further empirical work in those domains. The next chapter, by Richard Hall, presents a critical analysis of emergent institutional theory, a topic more typically considered by organizational theorists. But here, the author suggests that I/O psychologists who work in organizations to improve effectiveness may propose changes that unintentionally violate deeper, and more symbolic function of organizations. Such proposals, then, are bound

vi

Preface

to be met with resistance. Since I/O psychologists are typically not attuned to more macro-level topics, there is much here that we should be thinking about as we study and "tinker with" organizations. Again, this chapter provides a rich theory base that can aid researchers interested in workers' reactions to different structures at work. The book then moves to more traditional areas of concern for I/O psychologists, although here too there are new twists and ideas. The first chapter in this section, by Lillibridge and Williams, deals with the role of self-competence in behavior on the job. The authors point out that this construct has been receiving attention lately in the literature, but they adopt a more integrative perspective. Specifically, they note that a broader definition of self competence - one which includes self-efficacy - would allow us to integrate research from such areas as charismatic leadership and goal setting to the process of sustaining and reinforcing individuals' feeling of competence. Such attempts at integration are always worthwhile, and tend to enrich the questions researchers ask. The next chapter, by Tannenbaum, Beard, the Salas, deals with team building which, although it has been a hot topic i n the management area, has not received much attention from I/O psychologists. This is somewhat ironic, given that I/O psychologists have areas of expertise that should come to bear on the study of teams. I n fact, this chapter helps outline some of ways in which I/O psychologists can and should get involved in this area, especially through their expertise on groups and group processes. A chapter by Williams and Lillibridge, on personality and managerial effectiveness follows, and here they present a theoretical model as well as some data relating the personali ty constructs of surgency, agreeableness, conscientiousness, neuroticism and intellectual interests to various measures of managerial success. Debates over the role of trait-based models of leadership will probably go on as long as we study leadership, but this chapter makes a clear contribution to the debate by presenting a well-reasoned approach to the study of individual difference variables and leadership. This is followed by a chapter, by Racicot and Hall, dealing with managerial training - an area where I/O psychologists have presumably felt comfortable for some time. The chapter is more of a review of techniques and models of training, and deals less with implications for research, but the emphasis here on considering both the internal and external environments when designing and executing managerial training programs is an interesting and useful one. Finally, the section ends with a chapter by Caska, Kelley, and Christensen dealing with organizational needs assessment. These authors make the point that the success of any organi-

Preface

vii

zational intervention depends on the validity of the data obtained from the initial needs assessment. The authors then go on and provide guidelines and thoughts on how to select the approach that will yield the most useful data. The third and final section of the book deals with issues that concern individuals in organizations. The first chapter in this section, by Donna Ashcraft, deals with health issues in the workplace. The emphasis here is clearly on ways to control the health-care costs in organizations, but the author also raises important issues concerning the role of workplace environments in determining the health and well-being of employees. The next chapter in this section is by Kelley and Streeter, and deals with gender issues in the workplace. These authors present a rather comprehensive view of the potential causes, as well as the effects of sex segregation in the workplace. Included here are discussions of sex differences in performance and attitudes, work-family conflict, and sexual harassment. Overall, the chapter provides a large body of interesting material for both researchers and practitioners who must deal with an increasingly female workforce. Although there have been a number of other reviews dealing with some of these issues, none have been as comprehensive in integrating the different areas where sex segregation may play a role at work. The next chapter, by Bernard0 Ferdman, which deals with diversity in the workplace continues on a similar note. The author approaches the topic from the perspective of managing intergroup relations and, after reviewing two major (and quite relevant) streams of research in this area, concludes that ethnic diversity and interethnic differences are neither intrinsically good nor bad. He then concludes that the key for management in the future is to manage, rather than eliminate or strengthen these differences. The section, and the book, closes with a chapter by Stone, Stone, and Dipboye, that deals with the problems of stigmas in the workplace. This chapter really takes the theme of managing diversity a step further by discussing the problems that arise from treating "atypical" persons differently. Specifically, the authors review the literature on treatment based on race, handicap and physical attractiveness, and then suggest ways of trying to deal with the resulting bias. The material on differentia1 treatment based on physical attractiveness (and to a lesser extent handicap) is especially useful because here we are dealing with a bias that we all know exists, and yet is not illegal. This chapter has the potential to make a real contribution in its call for research and theory in these areas where stigmas affect a person's treatment at work.

...

Vlll

Preface

Clearly then, this volume is rather eclectic. Some might argue that all of these topics do not belong in a book about I/O psychology. Even those critics, however, would be hard pressed to argue that any one of these chapters does not deal with a topic that I/O psychologists are either familiar with, or should be familiar with. Perhaps it is in the latter area that this book will make its greatest contribution by informing I/O psychologists of different theories, models and perspectives that relate to the very topics we call our domain. By making I/O psychologists aware of these alternative views, the authors have the potential of enriching I/O psychology, and for showing I/O psychologists how their work should and does fit into the larger framework of organizational studies. Thus, whether scholar or practitioner, I am confident that every I/O psychologists will find something here that stimulates new thoughts and new ideas, and will come away with a broader perspective on the domain of I/O psychology.

Angelo DeNisi New Brunswick, NJ 1991

ix

About the Contributors George M. Alliger, Ph.D., is an Assistant Professor in the Department of Psychology, University at Albany, State University of New York. His research focuses on personnel selection, job analysis, training evaluation, and psychometrics. Donna M. Ashcraft, Ph.D., is currently an Assistant Professor at Clarion University of Pennsylvania. She is an active researcher in the areas of health psychology and psychology of women. Rebecca L. Beard, Ph.D., is a Senior Consultant with the Executive Consulting Group, Inc. She provides organizational, team, and management development services to a wide range of organizations. Donn Byrne, Ph.D., holds the position of Distinguished Research Professor of Psychology at the University at Albany, State University of New York, and is the Director of the SocialPersonality Program. His research interests focus on attraction theory and its application to a variety of areas including organizational, sexual, and political behavior. Barbara A. Cash, M.A., is an Instructor and doctoral student in the Department of Psychology at the University at Albany, State University of New York. Her research interests include the process of organizational development, concerns for older employees, and issues of unemployment. Edward W. Christensen, Ph.D., is Associate Professor of Counseling Psychology at the University at Albany, State University of New York. He is active i n organizational consultation. Angelo DeNisi, Ph.D., is Professor of Management and Labor Relations at Rutgers University. His research interests include: job analysis; cognitive approaches to performance appraisal; realistic job previews (and the role of met expectations); and work values.

X

Contributors

Robert L. Dipboye, Ph.D.! is Professor of Psychology and Administrative Science at Rice University. He conducts research on such topics as the effects of physical attractiveness on personnel decision-making, interviewing, performance appraisal, and performance feedback. Bemardo M. Ferdman, Ph.D., is Assistant Professor in the Departments of Psychology and of Latin American and Caribbean Studies at the University at Albany, State University of New York. His professional interests focus on multiculturalism and ethnic diversity in organizations and on ethnic identity; currently he is engaged i n research studies of multiculturalism in multinational businesses and in multiethnic schools. Richard H. Hall, Ph.D., is Distinguished Service Professor of Sociology and Interim Dean of the College of Social and Behavioral Sciences at t h e University at Albany, State University of New York. He is currently examining the applicability of Western organizational theory for East Asia, particularly China. Rosalie J. Hall is Assistant Professor of Psychology at the University of Akron. Her research interests include: training, technology and innovation; and work motivation. Kathryn Kelley, Ph.D., is Associate Professor of Psychology at the University at Albany, State University of New York. Her research interests focus on gender issues and risk behaviors. John R. Lillibridge, Ph.D., is Associate Psychologist at the Wilton Development Center in Wilton, N.Y. His research interests include leadership effectiveness, group processes, organizational development, the role of information systems in organizational change, and career development. Joel H. Neuman, Ph.D., holds the position of Assistant Professor of Management at the State University of New York College at New Paltz. His research interests include the application of work on interpersonal attraction to organizational issues. Bernadette M. Racicot, Ph.D., is Assistant Professor of Psychology at Rensselaer Polytechnic Institute. Her research interests include work motivation, organizational theory, and training.

Contributors

xi

Eduardo Salas, Ph.D., is a Senior Research Psychologist with the Naval Training Systems Center, where he conducts research on team training and performance. Dianna L. Stone, Ph.D., is Associate Professor of Management, Psychology, and Organizational Studies at the University at Albany, State University of New York. She conducts research on a variety of topics, including information privacy, drug testing, honesty testing, performance appraisal, performance feedback, and the stigmatizing effects of handicaps on personnel decision making. Eugene F. Stone, Ph.D., is Professor of Psychology and of Organizatinal Studies at the University at Albany, State University of New York. He conducts research on a number of topics including biases in performance ratings, reactions to appraisal feedback, privacy in work organizations, job design, and attachment to organizations. Deborah Streeter, M.A., is an Instructor and doctoral student in the Department of Psychology at the University at Albany, State University of New York. Her research interests include gender-related topics and fairness issues in organizations. Scott I. Tannenbaum, Ph.D., is an Assistant Professor in the Management Department at the University at Albany, State University of New York. His primary interests are in training effectiveness, team performance, and human resource information systems. Kevin J. Williams, Ph.D., is Assistant Professor of Psychology at the University at Albany, State University of New York. His research interests include performance evaluation, work-family linkages, job satisfaction, occupational stress, and the role of self-referent beliefs in human performance.

This Page Intentionally Left Blank

xiii

Acknowledgements Quotation on p. 16 reprinted from Cluster Analysis for Researchers by H. C. Romesburg by permission of Wadsworth Publishing. Copyright 1984 by Van Nostrand Reinhold. Quotation on p. 17 reprinted from "Scientific Method" by P. Caws. Used with permission of MacMillan Publishing Company from Encyclopedia of Philosophy, Paul Edwards, Editor in Chief. Vol. 7, p. 343, Copyright 1967 by MacMillan, Inc. Quotation on p. 19 reprinted from Personal Knowledge by M. Polanyi by permission of University of Chicago Press. Copyright 1958 by The University of Chicago Press. Quotation on p. 23 reprinted from "Psychology" by R. S. Peters and C. A. Mace. Used with permission of MacMillan Publishing Company from Encyclopedia of Philosophy, Paul Edwards, Editor in Chief, Vol. 7, p.26. Copyright 1967 by MacMillan, Inc. Quotation on p. 24 reprinted from "Five Parables" by I. Hacking published in Philosophy in History by permission of Cambridge University Press. Copyright 1984 by Cambridge University Press. Quotation on pp. 7 1-72 reprinted from "Organizational theory: Current controversies, issues, and directions" by A. G. Bedeian published in G. L. Cooper and I. T. Roberts (Eds.), International Review of Industrial and Organizational Psychology. Copyright 1987, John Wiley & Sons, Ltd. Quotations on p. 76, 77 and 78 reprinted from "The adolescence of institutional theory" by W. R. Scott published in Administrative Science Quarterly, Volume 32, by permission of Administrative Science Quarterly. Copyright 1987 by Cornell University. Quotation on p. 77 reprinted from "Institutionalized organizations: Formal structure as myth and ceremony by J. W. Meyer and B. Rowan published in American Journal of Sociology, Volume 83, by permission of University of Chicago Press. Copyright 1977 by The University of Chicago.

xiv

Acknowledgements

Quotation on pp. 80-81 reprinted from "Review essay: Overboard with myth and symbols" by C. Perrow published in American Journal of Sociology, Volume 9 1 , by permission of University of Chicago Press. Copyright 1985 by The University of Chicago Press. Quotation on p . 82 reprinted from " T h e development and institutionalization of subunit power in organizations" by W. Boeker published in Administrative Science Quarterly, Volume 34, by permission of Administrative Science Quarterly. Copyright 1989 by Cornell University. Quotation on p. 83 reprinted from "In search of rationality: The purposes behind the use of formal analysis in organizations" by A. Langley published in Administrative Science Quarterly, Volume 34, by permission of Administrative Science Quarterly. Copyright 1989 by Cornell University. Quotation on p. 84 reprinted from "Mimetic processes within an interorganizational field: An empirical test" by J. Galaskiewicz and S . Wasserman published i n Administrative Science Quarterly, Volume 34, by permission of Administrative Science Quarterly. Copyright 1989 by Cornell University. Quotation on p. 174 reprinted from "Socialization tactics, selfefficacy and newcomers' adjustment to organizations" by G . R. Jones published in Academy of Management Journal, Volume 29, by permission of Academy of Management. Quotation on p. 349 reprinted from In Common Predicament: Social Psychology of Intergroup Conflict and Cooperation by M. Sherif by permission of A. Sherif. Copyright 1966 by Houghton Mifflin. Quotation on p. 350 reprinted from "The role of cooperation in reducing intergroup conflict" by S . Worchel published in S . Worchel and W. G . Austin (Eds.), Psychology of intergroup relations (2nd ed.) by permission of Dorothy J. Anderson, Nelson-Hall Co. Copyright 1986 by Nelson-Hall Co.

xv

Contents Preface Written by Angelo DeNisi v About the Contributors xi Acknowledgements xiii Theoretical Approaches to the Industrial/OrganizationalField

I.

1.

The Theory and Structure of

Industrial/Organizational Psychology George M,Alliger 3 Introduction 3 Structure of the Field 4 Theory 11 Conclusion 2 1 References 25

2.

The Implications of Attraction Research for Organizational Issues Donn Byrne and Joel H. Neuman 29 The Role of Affect in Interpersonal Attraction 30 Attraction: Interpersonal and Organizational 32 Personnel Selection: The First Date 36 Organizational Entry and Beyond: For Better or Worse? 44 Organizational Exit: Separation and Divorce 53 Concluding Comments 57 References 58

Contents

xvi

3.

Taking Things a Bit Too Far: Some Problems with Emergent Institutional Theory Richard H. Hall 71 ITandOT 71 What is IT and Where Did It Come From 72 Some Problems with XT 79 Some Signs of Hope 83 References 85

11. 4.

Traditional Areas Of Research in Organizations Another Look at Personality and Managerial Potential: Application of the Five-factor Model John R. Lillibridge and Kevin J. Williams 91 The Williams and Lillibridge Model 92 Robust Dimensions of Personality: The Big Five 96 Exploring the Relation between Personality and Managerial Talent: A Preliminary Investigation 99 Conclusion 112 References 112

5.

Team Building and Its Influence on Team Effectiveness: An Examination of Conceptual and Empirical Developments Scott I. Tannenbaum,R. L. Beard, and Eduardo Sales 117 Teams and Team Building: Definitions 118 Factors that Influence Team Effectiveness 120 Team Building: Previous Reviews 128 Team Building Research since 1980 131 References 147

6.

Perceived Self-competence and Organizational Behavior Kevin J . Williams and John R. Lillibridge 155 Introduction 155 Conceptualization of Perceived Self-competence 156 Organizational Outcomes of Perceived Self-competence 159 Antecedents and Development of Perceived Competence in Organizations 166 Recent Developments 174 General Conclusion 177 References 178

xvii

Contents

7.

Current Trends in Managerial Training Bernadette M.Racicot and Rosalie J. Hall 185 Leadership versus Management: An Open Systems Viewpoint 186 Current Approaches to Managerial Training 198 Other Managerial Training Techniques 205 Current Trends in Managerial Training 209 Summary 218 References 2 19

8.

Organizational Needs Assessment: Technology and Use Barbara A. Caska,Kathryn Kelley, andEdward W. Christensen 229 Introduction 229 The Role of Need Assessment in Organizational Diagnosis 230 Organizational Needs Assessment and Scientific Research 233 Components of an Effective Needs Assessment 235 Influences on Technique Selection 236 Evaluation of Techniques 238 Techniques and Their Characteristics 241 Examples of Organizational Needs Assessments 247 Conclusions 253 References 255

111.

9.

Issues of Concern to the Individual in Organizations Health in the Workplace Donna Musialowski Ashcraft 259 Stress 261 Work Environment 261 The Impact of Changes in Technology 265 The Individual 267 The Aging Workforce 276 References 279

Contents

xviij

10.

The Role of Gender in Organizations Kathryn Kelley and Deborah Streeter 285 Trends in Work Patterns 285 Dynamics of Career Choice 289 Organizational Effects on Sex Segregation 292 Gender Effects on Career Development 294 Gender Differences in Compensation 296 Explanations for Gender-related Inequalities 299 Performance, Attitudes, and Perceptions 304 Work and Family 313 Health, Stress, and Working Women 318 Sexual Harassment 320 Conclusion 323 References 324

11.

The Dynamics of Ethnic Diversity in Organizations: Toward Integrative Models Bemardo M . Ferdman 339 Ethnicity and Organizational Psychology 34 1 Concepts of Ethnicity and Ethnic Diversity 342 Current Approaches 248 Toward More Integrated and Complex Approaches 363 Conclusion 366 References 368

12.

Stigmas in Organizations: Race, Handicaps, and Physical Unattractiveness Eugene F. Stone, Dianm L. Stone, and Robert L. Dipboye 385 Introduction 385 Stigmas and the Stigmatizing Process 388 Race as a Stigma in Organizations 400 Physical Unattractiveness as a Stigma in Organizations 412 Handicaps as a Stigma in Organizations 425 Dealing with Stigmas in Organizational Contexts 438 Needed Research on Stigmas of Various Types 442 Summary and Conclusions 443 References 444 Name Index

459

Subject Index 485

Part I: THEORETICAL APPROACHES TO THE

INDUSTRIAL/ORGANIZATIONALFIELD

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial1 Organizational Psychology - K . Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

3

CHAPTER 1 THE THEORY AND STRUCTURE OF

INDUSTRIAL/ORGANIZATIONALPSYCHOLOGY George M.Alliger

INTRODUCTION The generation and use of theory in Industrial/Organizational psychology has been often characterized as problematic. This chapter considers first the pedagogical and professional view of the structure of the field. I first show that although grand theory is not reflected in the structure of the field (consonant with the view of some critics), this does not contradict the original direction of the field. Next, I attempt to show that inductive approaches to research cannot be divorced from deduction. This is in line with what I term "modified positivism," which I argue to be the emerginig understanding of the scientific approach to reality which characterizes the field. One of the important features of this approach is the identification of previously implicit asssumptions; the work of authors in the field who attempt to help us identify such assumptions is discussed. Industrial/Organizationalpsychology has existed as an identifiable subdiscipline within the larger discipline of modern psychology long enough to generate substantial internal debate on its research approaches and practices. Appearing in diverse times and places, critiques of the field have argued that there are shortcomings in the science and practice of IndustriaUOrganizational psychology and that these shortcomings are generally unrecognized. These critiques sound several recurring themes or strains. Gordon, Kleiman, & Hanie (1978), in fact, suggest that critiques of the theory and praxis of Industrial/Organizational psychology tend to fall into four broad categories. Gordon et al. ' s first category subsumes those critiques which argue that the field is typified by problem with theory. Specifically, unlike many disciplines, Industrial/Organizational psychology tends

Alliger

4

to generate discrete studies on topics which are themselves peculiarly un-unified by over-arching theoretical frameworks. The second category of critiques contains those that suggest the field is preoccupied with methods -- that is, with issues of statistics and experimental design. The third category has as its members those critiques that suggest that Industrial/Organizational psychology as a field may not adequately train its future representatives -- included here are issues in the selection and training of Industrial/Organizationalpsychologists. Finally, the fourth category contains those critiques which attack the "publish or perish" system prevalent in academic circles. This system, of course, is identified as injurious to the development of profound research. A fifth category could be added to Gordon et al.'s taxonomy of critiques: one that would consist of those arguments that portray Industrial/Organizational psychologists as the uncritical assistants, if not actually tools or lackeys, of management (Shore, 1979; Stagner, 1982). Since some display of self-critical faculties is probably a sign of health for a field of study just as it is for an individual, written works evincing the action of such faculties can be expected from some minority of researchers within any vibrant discipline. To say that critiques are to be expected, of course, is not the same as saying that they should simply be catalogued or dismissed. Some of them, certainly, will be relevant and interesting, even challenging, and could be considered important reading for those being trained in the field. This chapter takes up the discussion of the first category: the nature and role of theory in Industrial/Organizational psychology. This is done through the consideration of two interrelated issues: the structure of the field of IndustriaVOrganizational psychology, and the role of theory within the field. The first section of the chapter, which serves to set the stage for the second, examines which of many areas within IndustriallOrganizational psychology, taken together, can be said to constitute the field, and how these areas may be organized into a greater whole. The second section then discusses the nature of theory within the field, taking its structure as the starting point. Finally, the chapter concludes with a discussion of the future integration of theory into those areas of study which comprise Industrial/Organizational psychology.

STRUCTURE OF THE FIELD IndustriallOrganizationalPsychology: Context-Centered Industrial/Organizational psychology is one of a number of psychological fields which could be considered context-centered.

Theory in IndutriaUOrganizational

5

That is, it is primarily characterized by theplace of application: in this case, the human workplace. Such a focus on context cannot easily be said to typify most other disciplines of psychology. Social psychology, for example, considers personal perceptions and interactions, wherever such occur; cognitive psychology examines, broadly, human rationation, where context again is interesting only as another independent variable; bio-psychology studies, generally, the complicated relationships between body and behavior; clinical psychology addresses problematic human behavior as the target of interest and study, in a general sense and across situations. These disciplines may be termed zopiccentered. IndustriaVOrganizationd psychology, however, pursues psychology in a different way. It does not so much study a topic as it does any topic that appears to pertain to the workplace. The rationale for its approach might be: "Everyday most people go to work. In the context of this broad activity known as work, people spend a large portion of their waking time. Psychology has lots of interesting observations to make about people. Therefore, a psychology of people at work is reasonable and should be useful. J oi n in g In d u stria 1/ 0r gan i z at i o n a1 psych olog y in t h i s cantextcentered approach are other psychological disciplines such as military and marketing psychology. In the first of these, military personnel and their activities are the major focus, or context; in the second, the context is the marketplace. This distinction between topic-centered psychology and psychology that centers on a context is not new; the rubric of basic versus applied psychology is a more common and somewhat similar categorization. In this scheme of things, Industrial/Organizational psychology usually comes under the applied grouping. However, since any of the basic findings of the other psychological fields may also become applied, a certain fuzziness is inherent in the basic/applied dichotomy. Moreover, much of the research in Industrial/Organizationat psychology is performed in the laboratory, and is justly termed basic. Such IndustriaVOrganizational laboratory research, however, whatever its particular focus (topic), still has as its ultimate focus human performance in the workplace (context). Some of the implications of having a context-centered rather than a topic-centered discipline are obvious only on examination of the discipline in some detail. One obvious problem is that there may appear to be little interlinking among different approaches to the context (the activity of work). That is, IndustrWOrganizationalcan appear to be an agglomeration of techniques and topics, each appearing in the field due to their relevance to the context of human work, I'

6

Alliger

but bearing among themselves no compelling interrelationships. Any linking that does occur may be only functional -- job analysis, for example, becomes a sort of linking pin or central concept in industrial psychology but this is largely due to its functional relationship to other areas of study, not because it represents a broad theoretical undergirding. This lack of a central concept is noted, for example, by Stagner (1982), who argued that Industrial/Organizational psychology today is without perspective or unity. Stagner drew the parralled between jobs which are fragmented into many seemingly unconnected tasks, and the field of Industrial/Organizational psychology which is fragmented into many seemingly unconnected specialities. There is thus a connection between the structure of the field and Gordon et al. 's first category of critiques -- the lack of a strong presence of theory in Industrial/Organizationalpsychology. Whether a lack of theory is prior to the structure, or whether the reign of "specific specialties" itself resists theory development, or whether the two exist in a relation of reciprocal causality, is not so clear. In any case, to the extent that one accepts the premise that the field exists in a state of disconnection, it may be reasonable to argue that to that extent theory is lacking, If there were grand, connecting theories in Industrial/Organizational, these would be obvious to most everyone, and the field could not be perceived as a group of discontinuous or discrete areas. However, before we accept the characterization of Industrial/Organizationalpsychology as a fragmented discipline, it might be worthwhile to consider whether this is in fact the case. What are the topics that supposedly exist in disarray? One way to examine the field with an eye toward answering this question is to consider what might be called the current pedagogical structure of the field.

The Current Pedagogical Structure When I speak about how a field of study is structured, or organized, I mean to convey how sub-areas of study within the larger field relate to each other, which sub-areas are closely identified with others, and so forth. It is as if one were to attempt to create a mental map or topography of a field: some sub-areas are geographically close to others because they address the same or similar questions, they use a similar vocabulary, a similar method or design, or fall under a single larger category of study. Therefore, for example, in Industrial/Organizational psychology recruitment and selection of personnel are closely linked; leadership and group behavior in organ-

Theory in Industrial/Organizational

7

izations are related. One way to begin to understand this geography is to consider how the field is taught. Organization for pedagogical purposes reflects an organization which those who have undertaken to teach the field believe will be best for teaching. This is certainly a primary, though not the only, source of information on how experts understand the field to be structured. In order to uncover the pedagogical structure of the field, I consider how four popular and recent Industrial/Organizational textbooks have organized the field. The four textbooks are Jewel1 and Seigall (1990), Schultz and Schultz (1990), Muchinsky (1990), and Seigel and Lane (1987). Each book was examined for chapter content. In cases of questions about whether a particular text covered a topic or not, subject indices were consulted. A list of covered subject areas by textbook was created. No attempt was made to match these areas to guidelines put out by Division 14 of the APA; instead, a listing of the subjects representing the field was "recreated" through textbook examination without any conscious pre-existing structure in mind. Table 1 shows subject areas arrived at, and the proportion of texts covering each area. These scores could be seen as "membership scores," showing to what extent each area "participates" in the field. An examination of Table 1 provides some interesting indications about the pedagogical approach to the field. Consumer and health psychology, for example, are each included only in a single text. This may make sense: both areas are often seen as disciplines in their own right. In addition, while they are applied, they are not specifically or solely strongly related to the general topic of work. More surprising is the low representation placed on job design and job evaluation, which many Industrial/Organizationalpsychologists would consider core topics. Perhaps the explanation for these and other questions which Table 1 may raise in the mind of the reader lies precisely in the pedagogical nature of the sources: what academics tell undergraduates a field looks like may to some extent diverge from the actual everyday conception of the field held by those academics. Textbooks, of course, often tend to have direct conceptual links with other textbooks in the field, since publishers require "market surveys" of competing texts, and fear leaving out any topic the survey indicates is important. This may create a somewhat artificial agreement among texts in the delineation of a field, and also retard somewhat the relaying of the image of the field as it is represented in the minds of researchers to the pages of a textbook.

Alliger

8

Thble 1 Area of Industr,,.'Organizational Psychology Covered in Four Undergraduate Texts

Employee selection History of IndustriallOrganizational Job analysis Leadership Motivation Organizational development Perfornance appraisal Psychological testing Statistics/research metholology Training Working conditions Group processes Human factors Job commitrnent/ernployee attitudes Labor relations Absenteeism Employee development Criteria Job evaluation Job design Consumer psychology Stresdhealth psychology

Membership Score

1.o 1.o 1.o 1.o 1.o 1.o 1.o 1.o 1.o 1.o 1.o .75 .75 .75 .75 .50 SO .50 .50 .50 .25 .25

Theory in Industrial/Organizational

9

The Professional Structure Howard (1990) presents not a pedagogical but a professional view of the field: about 2,000 members of the Society for Industrial and Organizational Psychology responded to a questionnaire which asked members, among other things, to identify which of 56 subareas of specialty "best described their current interests and involvement." After elimination of some of the very infrequently responded to specialties, the remaining 4 1 were clustered via the average linking method. The five clusters which resulted were: 1) Individual Evaluation (the "personnel psychology" specialty, including areas of selection, testing, job analysis, equal employment, legal issues, performance appraisal, statistics, research methods, and individual differences); 2) Training (engineering psychology, military psychology, training, and program evaluation); 3) Organizational Behavior (attitudes and surveys, job satisfaction, turnover, labor relations, quality of work life, job design, organizational behavior, work motivation, productivity, reward systems); 4) Organizational Development (Culture and climate, organizational development, leadership, management, decision making, problem solving, group processes, intergroup relations, conflict resolution, power, communication); and 5 ) Employee Development (career development, employment counseling, gender issues, work and family, stress, ethics). The results reported in Howard (1990) show a somewhat greater complexity than the pedagogical list in Table 1, but otherwise the two analyses do not sharply disagree. The only topic in n b l e 1 not reported in the cluster analysis results is consumer psychology: a topic only included in one of the four textbooks. Also of interest in the Howard (1990) cluster analysis results, not noted in the text of the report but obvious from examining the cluster dendrogram (p. 19), is that while Howard chose to stop at a five cluster solution, if one had proceeded to a two cluster solution then clusters one and two (Individual Evaluation and Training) would have been joined into a single cluster, as would have clusters three, four and five (Organizational Behavior, Organizational Development, and Employee Development). This would correspond roughly to the traditional division between Industrial psychology and Organizational psychology. Most importantly for our purposes, both Table 1 and Howard (1990) show a truly wide range of topics included within the purview of Industrial/Organizational psychology. It is in fact difficult to identify any theory which connects all these topics, or to say that the topics are theoretically dependent on one another. The use of cluster analysis by Howard (1990) may thus be particularly appropriate: we

A lliger

10

ask not how subareas relate theoretically, but how relations among them occur within the perceptions of Industrial/Organizational psychologists themselves. That is, we choose a pragmatic, databased approach to the very organization of the field. Thus, we reaffirm the view of Industrial/Organizational psychology as a context-centered discipline. The primary organizing force, which draws many disparate areas of study together, is to varying degrees the target or context of each of those areas: work.

Has Industrial/OrganizationalPsychology Diverged From Its Original Mission? The question might well be raised as to whether this apparent lack of theory, this apparent pragmatic grouping of subareas into a larger psychological identity, is recent: did Industrial/Organizational psychology at some point diverge from its original agenda? Or has it always been purely a context-centered field? The obvious way to answer this question is to consider the history of the role of theory in Industrial/Organizationalpsychology. Of course, this is not a simple thing to do in the course of a single chapter. Consequently, in order to examine this question, I choose a simple but illustrative test: how the topics outlined in Tmble 1 or the areas identified by Howard (1990) (or, for that matter, the courses typically listed in any Industrial/Organizational psychology program catalogue, or the list of areas of expertise identified by Division 14) compare with the original editorial statement of IndustriaVOrganizational's flagship journal, Journal of Applied Psychology. The first issue of this journal appeared in 1917 --a substantial time after Tmylor's first time-and-motion studies (188O's), the founding of the American Psychological Association (1892), and even Munsterberg's text Psychology and Industrial Eflciency (1913). That is, by the time the Journal of Applied Psychology appeared there was some sense of an identity surrounding the process of applying psychology to work. It was this sense of identity, presumably, which precipitated the journal's founding. For its debut in 1917, the Journal of Applied Psychology had on its editorial board G. Stanley Hall, John Wallace Baird, and L.R. Geissler. The editorial policy which these three men constructed (or at least to which they put their imprimatur) was succinct. It stated that the journal would contain articles addressing: a) the application of psychology to vocational activities such as law, art, public speaking, industrial and commercial work, and problems of business appeal, b) studies of individual mentalities, such as types of character, special talents, genius, and individual

Theory in Industrial/Organizational

11

and individual differences, including the problems o f mental diagnosis and vocational prognosis, c) the influence of general environmental conditions, such as climate, weather, humidity, temperature, also such conditions as nutrition, fatigue, etc., and d) the psychology of everyday activities such as reading, writing, speaking, singing, playing games or musical instruments, sports, and the like. @p. 1-2)

This editorial mandate is intriguing, and a little thought makes one realize that the identity of Industrial/Organizationd psychology as an empirical, context-centered discipline is not new. Indeed, the initial task the editors have set the field is explicitly contextual: "the application of psychology to vocational activities. 'I On the other hand, this context was not understood as strictly as it is currently. For example, section d), "everyday activities, would probably not be recognized as part of the research agenda of Industrial/Organiz.ational psychology today. Why? Because "everyday activities, 'I as stated in the above policy, are distinguishable from work and the workplace -- the editorial board would have us, I think, interpret these precisely as activities which characterize a context different than work, namely those activities carried out in leisure time.' The statement of the Journal of Applied Psychology editorial board also fails to show that Industrial/Organizational psychology was initially more theoretically driven than is the case today, in contradiction to Stagner (1982) cited previously, who speaks as if there were once a unified viewpoint in Industrial/Organizational psychology which has been lost. The areas of study listed in the original JAP policy are hardly less disparate or more connected than those in Table 1 or Howard (1990). The tendency has been for Industrial/Organizational psychology to become, not more theoretical, but simply more purely centered in the context of work. We can, I think, expect this trend to continue. Thus, Industrial/Organizational may end up completely spinning off most of military, consumer, and human factors psychology. These have in any case over the years developed into fields in their own right, and the central focus of each (the military, the marketplace, and machine-person interfaces, respectively) is substantially different than work in and of itself.

THEORY The Relation Between Structure and Theory I have suggested that the structure of a field and the role that theory plays within a field are inextricably linked. I have noted that

12

Alliger

Stagner (1982) argued that the field of Industrial/Organizational psychology appears to be disjointed and fragmented, even to the extent that "students of psychology are likely to perceive the industriaUorganizationa1 field as a set of specific specialties having no ineluctable interconnections." Thus, it appears to some that there is a dearth of theory (the first category of Gordon et al.'s taxonomy of criticisms). However, as we have also seen, a perception of "specific specialties" may be precisely what one would expect, given the initial trajectory of Industrial/Organizational psychology and its context-centered nature. Although, therefore, the structure of the field reflects a certain lack of theory, whether this should be bemoaned as a serious weakness is another question. It is probably true that for "ineluctable interconnections" between the topics of a field to exist, theory is required. This is the case because theory acts to bridge gaps and connect concepts, predictions, and outcomes. Theory predicts how variables should interrelate, and consequently serves to show something of the underlying unity of a field of study. This is true whether a theory is developed inductively (generalized from observed data) or deductively (specific interrelations are assumed to exist as part of a larger presumed structure). But, I think it may be fairly said, not that Industrial/Organizational psychology does not have theory, but that it does not have a system of grand theory, where such a system is a unifying vision that addresses and offers predictions for all or most subareas of a discipline. Physics, or even chemistry, can be seen as disciplines that are candidates for the label "unified." Thus, for example, the set of theories that describe intra-atomic and interatomic attraction are universally applied throughout the subfields of chemistry. Industrial/Organizational psychology, on the other hand, has small t theory (Winter, 1990): theories which operate within specific subareas, which may surface at surprising times in diverse places, and which are extremely useful for making specific predictions within a subarea.* Thus, we find many theories of leadership, of work motivation, of decision making, and so forth. Few of these are large theories, covering more than one of area of study. This forces textbooks simply to walk through one theory after another in succession, presenting pros and cons of each. For example, in the area of job satisfaction, students typically are exposed to expectancy theory, equity theory, need fulfillment theories, the opponent process theory, and so forth. The very way such small t theories are listed one after another indicates that they are not expected to "unify" a field. In fact, an awareness of the list of theories and the results of pertinent studies is taken precisely as that which constitutes knowledge of the field, as a perusal of the content of most undergraduate Industrial/Organizational courses will show.

Theory in Iru€wtrial/Organizational

13

I do not wish to suggest that small t theories are not important, although one assumes that they cannot all be simultaneously correct, assuming as they do competing explanations of causalities of human behavior. Nor do I mean to oversimplify by suggesting that theory in Industrial/Organizationalis never integrative: some researchers have attempted to unify large pieces of the field by theory. A largely unheralded entrant in this regard is Path-Goal theory (House, 1971), which attempts to link leadership, job satisfaction, and motivation. Similarly, systems theory is a broad attempt at combining most if not all typical psychological variables into a single framework.3

Inductive or Deductive meory-Building ? Having established this limited (small t ) nature of theory in Industrial/Organizationalpsychology (and reiterating that limited is descriptive and not pejorative), it is still left to us to consider the particular approach or approaches to theory taken by the field. One common dimension on which theory and theory-building is discussed is anchored on either end by the terms "deductive" and "inductive." Induction, it may be recalled, is defined as the process of reasoning from the level of the particular and individual to the level of the general and universal. Deduction, on the other hand, is, specifically, the deriving of a conclusion by reasoning about general principles; such a conclusion could then be tested by examination of particular effects. Dubin (1976) suggests that, for Industrial/Organizational psychology, an inductive approach is to be preferred, since It is exceedingly difficult to say something meaningful about the real world without starting in the real world. Observation and description of the real world are the essential points of origin for theories in applied psychology... Work is a human activity and any student of working behavior is simply forced to start with reality. @. 18) Dubin offers the example of a researcher who discovers that job autonomy and job satisfaction are related. Following an inductive process, the researcher begins to posit reasons for this observed relationship: perhaps workers expend energy fighting to obtain autonomy if they have less than some optimal amount, and hence have lower satisfaction than workers with more autonomy. The inductive development of a theory is thus initiated precisely through

14

Alliger

the observation of data. Dubin's example is a reasonable one and, I think, portrays how researchers often think. But, as far as providing a convincing argument for the primacy of induction in theorybuilding, both Dubin's discussion and example fall short. First, consider Dubin's contention that one must start in the "real world" in order to say something meaningful about it. Undoubtedly: but, of course, deductive theory is not in any way divorced from reality: its reality is simply in thecfirst instance the reality of the mind of the theorist. That is, all that is required to understand how deductive theory can be appropriate to a practical, context-centered discipline like Industrial/Organizational psychology is to recognize that people make theories, and people are part of the "real world". Any argument to the contrary harkens back to unmodified logical positivism, where the scientist adopts the (rather awkward) fiction that he or she does not exist: only "facts" exist. But, as shall be discussed in the next section, this brand of strict positivism no longer carries its former weight in the scientific community. Incidentally, I do not accuse Dubin himself of such a positivist stance. The point is that it is very easy to fail to realize that the "real world" includes everything: even our constructions of or theories about it. Second, the very existence of data presupposes a priori deduction. Consider again Dubin's example of the researcher who noted the relation between job satisfaction and autonomy. Curious about this observed relation, the researcher begins to think inductively, to attempt to generalize and explain. However, this example need not be seen only as a characterization of an inductive process. If, for example, we consider why the data were collected in the first place, we see that someone, either the researcher of our example or some other researcher, must have obtained and analyzed these data for some purpose. That is, it is likely that someone's theory (however ill-defined or vague) suggested that it would be reasonable to collect these data at the same time on the same group of workers. Even if it is argued that the data collection was purely fortuitous in this particular case, this certainly cannot be the case the majority of the time. Otherwise, we would be reduced to firing off questionnaires or collecting measures in a random fashion, later to detect patterns. This is not the primary method in which psychologists operate, although Gordon et al. (1978) would seem to disagree, for they suggest that the main concern of Industrial/Organizational psychologists has been simply to observe and record how behaviors at work COVW.~

My premise, then, is that deduction and induction are only artificially separated. If one starts deductively on a particular question,

Theory in Indmtrial/Organizational

15

that is, with a statement of a general nature, this does not rule out inductive, re-generalizing processes once relevant data are available on the question. On the other hand, if one begins inductively in a specific case, it may well be necessary to make new or revised predictions (that is, to operate deductively) once one has arrived at the level of theory. Viewed in this way, deduction and induction exist in a dynamic, reciprocating relationship, like the pistons in an internal combustion engine: you think inductively in order to be able to think deductively. Alternatively, you think deductively in order to think inductively. Neither approach has a primary claim on the "real world". In a sense, induction and deduction then exist together as a chicken-and-egg issue: neither can be definitively said to come first, either chronologically or in terms of importance for Industrial/Organizational psychology. Understanding science in this way should have a freeing effect for researchers: there is no "one best way" to develop theory. Having stated this, however, I should also state that I suspect that researchers differ in their predilections regarding theory, whether or not these predilections affect how they actually approach it. One is tempted to suggest that some are more Platonic (deductively-oriented) in their approach to psychological science, while others are more Aristotelian (inductively-oriented) in theirs.

Positivism and Modified Positivism The preceding discussion on the relation between inductive and deductive theory building can be understood in the light of Kuhn's paradigm model of scientific advance.' I have already mentioned one general scientific paradigm: positivism. Industrial/Organizational psychology, certainly, has primarily identified with positivism as a method. This identification would explain, for example, a preference for inductive as opposed to deductive science (Dubin, 1976). Sometimes called logical positivism, logical empiricism, objectivism, or functionalism, a modified version of this approach to research and reality typifies many scientists today. Briefly, positivism can be characterized as a belief that "all meaningful statements are either analytic or conclusively verifiable or at least confirmable by observation and experiment" (Webster's Seventh New Collegiate Dictionary, 1963). As suggested earlier, adherents of pure logical positivism, as a breed, are just about extinct. Pure logical positivists could be characterized as scientists who in no way concede that their assumptions or predilections in any way whatsoever influence their work, that facts are absolutely knowable in the most objective sense, that the act of scientific per-

A lliger

16

ception is unbiasing and hence irrelevant in and of itself. But, on the contrary, most psychological research now is done with a more or less explicit awareness that scientists are humans engaged in a human act, That act is simply the attempt to better understand the world, where a nonrigidly defined scientific method is the method guiding that act. In psychology, Rosenthal's (1963)concept of "experimenter demand effects," which outlines the effect of observation on the measurement of the observed, has been pretty well assimilated, despite contrary discussions of possible problems with Rosenthal's work (e.g. Barber & Silver, 1968). And, although one may still see attacks aganst a pure logical positivist position, that position must be set up like a "straw man," where the scientist is portrayed as an individual ignorant of himself as a determining force in scientific research (cf. Kukla, 1984). This portrayal of science and scientists cannot really be considered as representative of current psychological science. Instead, social science today, including psychology, could better be characterized as utilizing a modified positivistic approach to science. That is, the attempt to characterize reality in some objective way has not been abandoned, but the contingent nature of any attempt to do so is, at least to some extent, recognized. Thus, for example, while it is possible to say in some way that an objective body of knowledge is being built, it also has to be said that this occurs only in the crucible of the entire community of scientists, where individual commitment to theory (Mitroff, 1972), necessary as it is for the pursuit of science by individuals, is in some sense and at some point in time purged or canceled out (Alliger & Hanges, 1984). As Romesburg (1984)points out within a discussion about the use of subjective skill in proper use of cluster analysis, When a group of professionals follows these less rigid rules, a consensus may result, but usually there will be differences of opinion as to which guides are best. The guiding rules for subjective decisions are the norms that professional fields sanction. The norms map out a range of what is and is not acceptable, rather than constraining the decision to one possibility., . What comes after the equal sign in 'Let X is properly a subjective decision. The researcher risks being wrong, but this is not worrisome because, in the end ... the researcher will gain feedback indicating whether he was right or wrong. (p. ...,I

241-242)

Theory in Industn'al/Organizational

17

That is, the community of scientists recognizes the important and proper role of subjectivity, though each discipline will attempt to set up rules within which that subjectivity should operate. And, in the end, the community will provide "feedback," that is, judgments, about the acceptability of a particular scientist's subjectively determined work. We have then in operation a modified positivism, which recognizes an objective reality, but defines this in terms of the perceptions, rules, norms, and decisions of the scientific community. However, to say that we have to some extent modified the purely positivist paradigm is not to say that we have resolved exactly how we are to do science. In fact, in place of the logical positivist's desire to lay down a specific definition of science which all must follow to the letter, there is a wideranging debate on the precise character science is to take within each discipline. As Caws (1967) notes:

...the search for a unique scientific method seems less urgent than it once did, because by now it is clear that in general it has already been discovered, while in detail there is no such thing. The general conclusion is that the method of science is a mixture -- the proportions of which vary from one science to another -of logical construction and empirical observation, these components standing in a roughly dialectical relation. The details, on either side of this division, remain the object of a range of inquiries from the study of laboratory techniques to pure axiomatics. (p. 342, italics added). That is, there is really no single scientific method, although it may be suspected that the average graduate student of psychology might be surprised to hear this. Thus we find that our conclusion of the previous section, that there is no one best way to approach theorybuilding, is somewhat broadened into: there is no one best way to do science. Certainly there are methodologies, approaches, and ways of structuring research which may be particularly fruitful in providing findings and ideas which help us to structure our world. And, it is true that it is precisely widely-shared agreement on such approaches that permit the existence of a scientific community of psychologists. Modified positivism can then be said to have three broad characteristics. First, there is a recognition of the subjective element in experimentation. Modified positivism does not require that the scientist hide him or herself, or otherwise pretend not to exist.6

18

Alliger

Second, there is no single best way to "do science". For any given study, the induction-deduction cycle can be entered at any point. Researchers in different areas of study are free to develop methods and approaches appropriate for their own area. To be sure, any such development must be justified as appropriate, or the researcher risks a degradation in credibilit Third, the entire community of scientists, and their scrutiny o results presented to them, comprise the "objective" nature of science. The public nature of experimental results, and replicability, are important.' Every discipline will have, to varying degrees, remnants of unmodified positivism. Indeed, if one of the problems that we have in Industrial/Organizationalpsychology is a dearth of linking theories, this may in fact be due precisely to an incomplete adoption of what I have called the modified positivist approach to science.

y.

Identijjdn Underlying Assumptions in Industria /Organizational Psychology

f

One theme that has underlain our discussion of modified positivism is that scientists, like everyone else, have presuppositions and assumptions guiding their thinking, perceptions, and actions. On one level this can be a relatively trivial observation -- of course we have these assumptions, but the idea is not to let them affect one's scientific work. But if we accept Mitroff's (1972) or Polanyi's (1958) concept of the importance of commitment in science, then we see that it is precisely assumptions that give power and direction to research. The continuing struggle to understand and attain modified positivism is then naturally to a certain extent the attempt to understand how subjectivity operates in research, and to uncover the assumptions guiding research. So, we have writers who seem to do one of two things. First, they may actively encourage subjectivity in research. Second, they may attempt to identify (and sometimes condemn) underlying research assumption s. An example of the first group is Gergen, who pointed out in a 1978 article entitled Toward Generative Theory, that an emphasis on objectivity, contemporary knowledge and views, and uninvolved observation may retard the promulgation or adoption of new, fresh viewpoints in psychological research. That is, our positivistic inheritance can thwart theorizing.8 Sampson (1978) argues that since our theorizing cannot be part of a completely "objective" science, we might as well recognize this and direct our science by conscious commitments, rather than unconscious ones. Polanyi, in his justly famous Personal Knowledge (1958), antici-

Theory in Industrial/Organizational

19

pated Gergen and Sampson by suggesting that instead of emphasis on the objectivity of the scientist, the commitment of the scientist, or the scientist's "heuristic passion" is key. Such a passion interacts with scientific training and knowledge in the scientist him or herself to produce original research. As Polanyi (1958) puts it:

To accept commitment as the only relation in which we can believe something to be true, is to abandon all efforts to find strict criteria of truth and strict procedures for arriving at truth. A result obtained by applying strict rules mechanically, without committing anyone personally, can mean nothing to anybody. Desisting henceforth from the vain pursuit of a formalized scientific method, commitment accepts i n its place the person of the scientist as the agent responsible for conducting and accrediting scientific discoveries. The scientist's method is of course methodical. But his methods are but the maxims of an art which he applies in his own original way to the problem of his own choice. (p. 311-312) In Polanyi's words we again see the delineation of science as a real examination of the real world, and yet only real in process, that is, as it is instantiated in the living person and actions of the scientist. Indeed, that is why some writers have argued that forays into Artificial Intelligence will never result in true intelligent thinking: the lack of any ability to feel commitment means that computers will of necessity see the landscape of facts i n monocolor, as it were. Computers necessarily lack the ability to specify, through commitment, the world in a certain way prior to exploring it. The second approach to subjectivity in research has this agenda: identify and discuss the appropriateness of underlying research assumptions. I started this chapter by pointing out that in Gordon et al. 's critique taxonomy, the first category included those writers who argue that IndustriaUOrganizational has a lack of theory. Having discussed modified positivism, however, I am now in a position to picture these critiques in part as constructive attempts to identify underlying assumptions in guiding research in the field. I examine a few of these critiques briefly. Argyris (1976) argues, for example, that IndustriaVOrganizational psychologists use an implicit organizational theory in their work: namely, a) that organizations are structured in a hierarchical way and such a structure is appropriate, and b) that the goals of the organiza-

20

Alliger

tion as articulated by management must necessarily be optimal. For this reason, indices such as turnover, absenteeism, or volume of sales are seen as accurate indexes of an effective organization. The "steady-state" of the existing organization is favored. Acting on this implicit "steady-state" theory, Argyris argues, is likely to be dysfunctional. Organizations have to change to stay viable. In an intriguing twist, Argyris (1976) goes on to argue not that the Industrial/Organizationalpsychologist is being too mercenary in simply accepting management's status-quo without question, but that this acceptance may place him/her out of touch with management: "the result may be that industrial psychologists could become, unintentionally, to the field of human behavior, what the shiny-pants, green-shade bookkeeper is to finance" (p. 167). Thus, failing to adopt a modified positivism that sees the world as dynamic and constructed could lead Industrial/Organizationalpsychologists to be "left out". Stagner (1982) identifies a different implicit assumption held by Industrial/Organizational psychologists. He says that mylor, like Marx, accepted and preached the idea that motivation to work can be boiled down to basic economic motives. The acceptance of such a concept of motivation promotes, in turn, a mechanistic and ultimately cynical concept of workers. Nord (1982) quite explicitly argues that residual metaphors of logical positivistic science, namely quantification and objectivity and rationality, are not politically neutral tools. Their use can legitimate certain elite groups. Ignoring the political dimensions of these tools may cause Industrial/Organizationalpsychologists to underplay the role of conflict in the workplace. Unions, for example, may tend to be marginalized. Organizations, in short, are not economic entities with no political identity. An unmodified positivism would tend to rely on an objective (and hence, after Argyris, status-quo) approach to organizations, and fail to perceive this. Shore (1979), in a brief article entitled "Servants of Power," presented yet another twist on the don't-accept-the-organizationalstatus-quo theme; his article most clearly falls in the additional fifth category of critiques mentioned in the introduction of this chapter. That fifth category includes those critiques which argue that psychologists can too easily be the unwitting tools of management. Shore argued that Industrial/Organizationalpsychologists tell people to adapt to and accept stressful conditions in the workplace. That is, the solution is seen in self-management techniques rather than an attempt to change the organization to eliminate stressors. There are several purposes to be achieved by bringing our assumptions as psychologists to the fore. First, this is more honest

Theory in Indmtrial/Organizational

21

and reflective of the way real research is done than is the pretense of complete objectivity. Second, in addition to recognizing our a priori stances, we can avoid, to some extent, the influences of these in our research. The field does not always agree with this second point. There are proponents of some organizational theorizing, for example, who argue that a specific political agenda should inform research (see, for example, the discussion of Gioia & Pitre, 1990). But if we are aware of certain biases or predilections, it would seem axiomatic that it is our duty to see through them, however uncomfortable this may be. This is because modified positivism is still to some extent positivism: there is a reality which we attempt to describe. Third, recognition of assumptions can be a powerful heuristic. This seems to be what Mitroff (1972) and Polanyi (1958) are arguing. When we are able to explicate our deep seated beliefs concretely, we will be able to generate "tests" of these. More importantly, they will provide us with the critical edge we need to understand and judge the current state of the field. Theory-based research should not be a revered but opaque process, but instead the relatively straightforward application of beliefs to the observable world, testing, and reevaluation of beliefs.

CONCLUSION In this chapter I have considered several points. First, I asked whether it is true, as some internal criticism suggests,that the field of Industrial/Organizational psychology is i n fact a field comprised of theoretically unconnected topics. The answer as formulated here, from an examination of the structure and content of the field, is a very qualified yes. Industrial/Organizational psychology is contextcentered: it applies psychology to a major human context: work. In this, however, it neither deviates from its original direction nor does it renounce theory. Rather, it uses small t theory: theory about a given area that does not purport to be grand, overarching Theory. There are some possible candidates for grand Theory, however, emerging in the field. Systems theory was mentioned as one of the most prominent of these. In addition, I used the discussion of theory as a setting in which to argue that the approach by the field to theory should not be, indeed, it cannot be, simply inductive in nature. We all may have predilections in the type of theory-building and testing we employ, of course, which is as it should be. But induction and deduction exist together in ineluctable interdependence. Second, I proposed that what I have called "modified positivism" appropriately typifies t h e science occurring in t h e field of Industrial/Organizational psychology, and probably science in gener-

22

Alliger

a1 today. Modified positivism accepts the subjectivity of the researcher, posits no single way in which the scientific method must be operationalized, and sees objectivity of knowledge as being created, or discovered, by the community of scholars. However, while I have argued that unmitigated objectivism fails to be persuasive precisely because, ironically, of its exclusive emphasis on "facts," it should be noted that I have not abandoned the scientific enterprise to simple relativity. The latter can lead only to the endorsement of a type of "private knowledge" which will be detrimental to the real interests of the scientific community. That such relativity is actually promulgated by some as a the proper approach to science in the social sciences should not be doubted (cf. Kukla, 1982). Third, the first characteristic of modified positivism, namely the recognition of scientific subjectivity, was discussed r e IndustriaVOrganizational psychology using the concept of "heuristic passion". How some Industrial/Organizational psychologists have identified the status-quo subjectivity of the field through examination of some of its underlying assumptions was also discussed. Vis-a-vis this last point, it might not be fair to end this chapter without some discussion of just how the act of identifying the assumptions which guide our approaches to psychology of work is to be pursued. I think the short answer to this is that this is where science meets philosophy, history, literature, and even religion. That is, the act of identification of the assumptions guiding research involves the totality of the person, and perhaps a characteristic such as risk-taking, as well as, presumably, some actual perspicacity and learning. But practical steps in this direction are possible as well: the reading of what other disciplines have to say about work, for example, is of obvious benefit. Work as a human activity really is very interesting, and psychologists represent a small minority of people who have written and thought deeply and well about it. It is through such reading that we discover our own historical boundedness and why we were taught to pursue psychology in a certain way. I think that when we get this sense of perspective that not only do we see the limitations of our field, but begin to appreciate its contributions to the theoretical understanding of work.

NOTES 1 . Leisure or every-day non-working activities have, it is true, shown signs of coming to the fore as a topic of importance to Industrial/Organizationalpsychology, but only and precisely because i t is non-work, and as such has influences on work. Williams, Suls, Alliger, Learner & Wan (in press), for example, discuss the importance of understanding how workers "juggle"work and non-work tasks.

Theory in Indwtrial/Organizational

23

2. It is interesting to note that some writers outside of the discipline of psychology regard all of psychology as either atheoretical or what we might call "aphilosophical". This latter term would mean that, after all, psychology avoids the "basic" questions. A relatively extreme statement in this regard is made by Peters and Mace (1967): Psychology has always been, and may always remain, a parasitic discipline. F o r twenty centuries i t was j u s t a branch of philosophy. To gain emancipation, it entered into willing bondage to the established natural sciences. Increasingly it has claimed to be, and has been increasingly accepted as, a biological science.. . In the twentieth century psychologists and biologists have found a common approach, frame of reference, and interest in such new special studies as ethology, cybernetics, and information theory and a common lack of interest or only a peripheral interest in problems of the philosophy of the mind. (p. 26) Needless to say, this view may be taken with a grain of salt. However, it is not only other disciplines that consider psychology as a whole as lacking in unity or theory (e.g., Wapner, 1988; Yamamoto, 1988). 3. It can be noted that when a very broad theory is proposed it may be resisted by researchers, in part, apparently, for its very breadth. For example, Locke (1989) used the word "imperialist" in describing systems theory. The very characteristics of systems theory that suggest that it may be able to unify disparate areas of motivational psychology, one of which is goalsetting theory, prompted Locke to reject it. Here we may be up against the scientist's "heuristic passion" or "heuristic commitment" which we discuss, in a wholly favorable light, later in this chapter. However, it may be that this passion can act, not only as an important driving force in scientific research, but as a retardant to the acceptance of overarching theory within a field, as the above example shows. 4. One of the incorrect presumptions about the inductive method is that the data are determinative of theory. Actually, there will be many theories that could be induced from data. As Mulaik (1987) points out in the context of one of the limitations of factor analysis, namely that factors derived from a set of data can not be definitively named:

Factor indeterminacy is just a special case of a more general form of indeterminacy commonly encountered in science, known to philosophers of science as the empirical underdetermination of theory -- Data by themselves are never sufficient to determine uniquely theories for generalizing inductively beyond the data. (p. 298)

Alliger

24

A belief that one can make inductive inferences uniquely from data without a priori assumptions has been termed the "inductivist fallacy. "

5. Paradigm, for our purposes, can be thought of as a general accepted approach to the study of behavioral phenomena. This is how it appears generally to be used in the psychological literature (e.g., Gioia & Pitre, 1990). It may be that any such definition of the word should be amended to include "with apologies to Thomas Kuhn." Kuhn (1970), of course, is responsible for the ground-breaking examination of how "normal" science is periodically stressed to the breaking point by one or more experimental anomalies, and the concomitant appearance of a new "paradigm" or way to approach science. Less well known is the fact that while the social sciences have wholeheartedly accepted Kuhn's view of the way science works, natural science has proven much more resistant to it. This occurred despite Kuhn's intention to address only natural science, and to explicitly avoid addressing the social sciences. As Hacking (1984) musingly notes, however, the natural sciences have largely ignored Kuhn's work, while ...[i]t is quite the opposite in the social and psychological sciences. Kuhn's Structure had hardly appeared in print when presidential addresses to annual meetings of the American Psychological Association and American Sociological Association avowed their need for paradigms. It has always seemed to me that Kuhn was a good deal clearer about his use of his famous word than most of his readers, including presidents of learned societies.. . it may be that the impact of Kuhn on social sciences is a sign of their lack of self-understanding. " (p. 115)

6. One manifestation of this element of modified positivism in psychological research is editorial policy. Compare, for example, the explicit recognition of subjectivity in the most recent Publication Manual of the APA (1983):

Inappropriately or illogically attributing action in the name of objectivity can be misleading. For example, writing "the experimenter instructed the subjects" when "the experimenter" refers to yourself is at best ambiguous and may even give the impression that you disavow your own study. (p. 35) 7. The requirement of replicability in particular could be said to place limits on our second point, namely that there is no "one best way" to do science. To the extent a psychologist wishes to obtain results that are replicable by other psychologists in other situations, he or she is not free to pursue science in any way whatsoever. Even if it is agreed that the scientific method is not a unitary, fully developed, unalterable method, it is still a body of method and a coherent philosophy, and the hope of obtaining results that may be replicable depends on participation in that method and philosophy.

Theory in ZndustriallOrganizational

25

8. It may well be that a legacy of positivism and Gordon et al.'s second category of critiques of I/O psychology (too much emphasis on methods) are related. That is, the emphasis on research based on objectivity and noninvolvement to some extent requires an undisputed body of methods and statistics. An observation is objective if and only if it is carried out in a certain way. Relationships are only understandable if data are analyzed in a certain way. In this light, it is fascinating to read the work of Gigerenzer and colleagues (Gigerenzer & Murray 1987; Sedlmeier & Gigerenzer 1989). They suggest that it is only because psychology needed a single and undisputed set of methods and statistics that such a body was developed and taught. Out of the sharp theoretical disagreements between Fisher on the one hand and Neyman and Pearson on the other regarding significance testing and design psychologists forged a single statistical method. This has then been taught and passed down as the method.

REFERENCES Alliger, G.M., & Hanges, P.J. (1984). Objectivity and science: Reply to Kukla. Journal of Personality and Social Psychology, 47, 676-679. American Psychological Association. (1983). Publication manual of the American Psychological Association. (3rd ed .). Washington, DC: Author. Argyris, C. (1976). Problems and new directions for industrial psychology. I n M.D. Dunnette (Ed.), Handbook of industrial and organizational psychology. Chicago: Rand-McNally. Barber, T.X., & Silver, M.J. (1968). Pitfalls in data analysis and interpretation: A reply to Rosenthal. Psychological Bulletin: Monograph Supplement, 70, 48-62. Caws, P. (1967). Scientific Method. In P. Edwards (Ed.), The encyclopedia of philosophy, New York: Collin MacMillan. Dubin, R. (1976). Theory building in applied areas. In M.D. Dunnette (Ed.), Handbook of industrial and organizational psychology. Chicago: Rand-McNally. Gergen, K.J. (1978). Toward generative theory. Journal of Personality and Social Psychology, 36, 1344-1360. Gigerenzer, G., & Murray, D.Y. (1987). Cognition as Intuitive Statistics. Hillsdale, NJ: Erlbaum. Gioia, D.A., & Pitre, E. (1990). Multiparadigm perspectives on theory building. Academy of Management Review, 15, 584602. Gordon, M.E., Kleiman, L.S., & Hanie, C.A. (1978). IndustrialOrganizational Psychology: Open thy ears 0 house of Israel. American Psychologist, 33, 893-905.

26

Alliger

Hacking, I. (1984). Five parables. In R. Rorty, J.B. Schneewind, & Q. Skinner (Ed.), Philosophy in history. Cambridge: Cambridge University Press. Hall, G. S., Baird, J. W., & Geissler, L. R. (1917). Introduction. Journal of Applied Psychology, 1 , 1-4. House, R.J. (1971). A path-goal theory of leadership. Administrative Science Quarterly, 16, 321-338. Howard, A. (1990). The multiple facets of industrial-organizational psychology: Membership survey results. Arlington Heights, IL: Society for Industrial and Organizational Psychology, Inc. Jewell, L. N. & Siegall, M. (1990). Contermprary industrial/organizational psychology (2nd 4.).St. Paul, MN: West. Kuhn, T. (1970). The structure of scientijk revolutions. Chicago: University of Chicago Press. Kukla, A. (1982). Logical incoherence of value-free science. Journal of Personality and Social Psychology, 43, 1014-1017. Locke, E.A. (1989, April). In. H.J. Klein (Chair), Debate: Control theory and understanding motivated behavior. Symposium conducted at the annual meeting of the Society for Industrial and Organizational Psychology, Boston. Mitroff, I. (1972). The myth of objectivity. Management Science, 19, b613-b618. Muchinsky, P. M. (1990). Psychology applied to work (3rd ed.). Pacific Grove, CA: Brooks/Cole. Mulaik, S.A. (1987). A brief history of the philosophical foundations of exploratory factor analysis. Multivariate Behavioral Research, 22, 267-305. Munsterberg, H. (19 13). Psychology and industrial efficiency. Boston: Houghton Mifflin. Nord, W.R. (1982). Continuity and change in Industrial/Organizational psychology: Learning from previous mistakes. Professional Psychology, 13, 942-952. Peters, R.S., & Mace, C.A. (1967). Psychology. In P. Edwards (Ed.), The encyclopedia of philosophy. New York: Collin MacMillan. Polanyi, M. (1958). Personal knowledge. Chicago: The University of Chicago Press. Romesburg, H.C. (1984). Cluster anulysis for researchers. Belmont, CA: Wadsworth, Inc. Rosenthal, R. (1963). The experimental hypothesis as unintended determinant of experimental results. American Scientist, 51 , 268-283. Royce, J.R. (1982). Philosophic issues, Division 24, and the future. American Psychologist, 37, 258-266.

Theory in IndustriallOrganizational

27

Sampson, E.E. (1978). Scientific paradigms and social values: Wanted -- A scientific revolution. Journal of Personality and Social Psychology, 36, 1332-1343. Schultz, D.P., & Schultz, S.E. (1990). Psychology and industry today. New York: MacMillan. Sedlmeier, P., & Gigerenzer, G. (1989). Do studies of statistical power have an effect on the power of studies? Psychological Bulletin, 105, 309-3 16. Siegel, L., & Lane, I.M. (1987). Personnel and organizational psychology. Homewood, IL: Irwin. Shore, R. (1979). Servants of power. APA Monitor, 10, 2. Stagner, R. (1982). Past and future of Industrial/Organizational psychology. Professional Psychology, 13, 892-902. Wapner, S. (1988). Unifying psychology: Strategies from a holistic, developmental, systems theory. Hiroshima Forum for PsycholO ~ Y ,13, 1-15. Wbster's new collegiate dictionmy. (1963). (7th ed.). Springfield, MA: Meniam Webster Winter, S.L. (1990). Bull Durham and the uses of theory. Stanford Law Review, 42, 640-693. Williams, K.J., Suls, J:, Alliger, G.M., Learner, S.M. & Wan C.K. (in press). Multiple role juggling and daily mood states in working mothers: An experience sampling study. Manuscript submitted for publication. Journal of Applied Psychology. Yamamoto, T. (1988). How far can we go towards unifying psychology? -- Comments on Prof. Wapner's paper. Hiroshima Forum for Psychology, 13, 17-18.

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial/ Organizational Psychology - K . Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

29

CHAPTER 2 THE IMPLICATIONS OF ATTRACTION RESEARCH FOR ORGANIZATIONAL ISSUES

Donn Byrne and Joel H. Neuman

...in many decisions affect plays a more important role than we are willing to admit. We sometimes delude ourselves that we proceed in a rational manner and weigh all the pros and cons of the various alternatives. But this is probably seldom the actual case. Quite often 'I decided in favor of X' is no more than ' I liked X.' Most of the time, information collected about alternatives serves us less for making a decision than for justifying it afterward... We buy the cars we 'like,' choose the jobs and houses that we find ' attractive, ' and then justify those choices by various reasons that might appear convincing to others who never fail to ask us, 'Why this car?' or 'Why this house?' We need not convince ourselves. We know what we like. (Zajonc, 1980, p.155) Zajonc's analysis could easily be extended to include why this wife? Why this husband? Why this career? Why this job candidate? One could also add countless other whys. In some situations emotional decisions seem very appropriate and not in need of justification. In purchasing a work of art, selecting a wardrobe, decorating a house, or choosing a friend, affective responses are ordinarily considered sensible and even desirable. People are not, however, generally inclined to admit that emotional forces determine behavior in the "dollars-and-cents" world of business. Most people would agree that an individual's decision about an academic major, career and, eventually, occupation are critically important decisions. So, too, are the decisions that managers continually face: personnel selection, performance appraisals, worker compensation, task and

Byme and Neuman

30

plant design, and incentive programs, to name just a few. Poor decisions can result in substantial costs to individuals and to organizations. For this reason, decision making in these areas is presumably a rational process, devoid of emotionality. It is true that these are extremely important issues, but 'it-is probably not true that they are resolved with emotionless dispassion.

THE ROLE OF AFFECT IN INTERPERSONAL ATTRACTION We believe, as does Zajonc (1980), that "affect plays a more important role than we are willing to admit."

me Reinforcement-affect Model of Attraction One formulation of the attraction process stresses the role of affect as the primary determinant of evaluative responses (Byme, 1971; Byme & Clore, 1970; Clore & Byme, 1974).

Associated AfSect. The reinforcement-affect model of interpersonal attraction is based on a classical conditioning paradigm in which affect plays a czntral role: stimuli that evoke positive affect are liked while stimuli that evoke negative affect are disliked. Attraction to another person is based on the extent to which that person elicits positive affect or is merely associated with another stimulus that elicits positive affect. As an example, the effect of associated affect on interpersonal attraction was demonstrated by Griffitt (1970) in an experiment where subjects were required to evaluate a target stranger. Griffitt exposed one group of experimental subjects to unpleasant combinations of temperature and humidity while another group of subjects was exposed to pleasant levels of these environmental variables. Subjects i n the "pleasant" condition evaluated target strangers more positively than subjects in the "unpleasant" condition. In this instance, affect induced by temperature and humidity was associated with the target stranger and influenced the evaluation of the target accordingly. In another study, Gouaux (1971) induced both positive and negative affect by means of the Velten (1967) procedure. With this technique, the appropriate affective state is induced by having subjects read 60 self-referent statements which become increasingly more positive or negative, depending on the desired mood state the experimenter wishes to induce. Following mood-induction, subjects were then asked to evaluate a target stranger. As predicted, positive affect resulted in relatively positive

Implications of Attraction Research

31

evaluations of a target stranger and negative affect in relatively negative evaluations.

Attitude Similarity. Another procedure which induces affect involves exposing subjects to "simple" attitude statements. In one popular technique, known as the bogus stranger paradigm (Byrne, 1961), subjects are asked to complete an attitude questionnaire and indicate to what extent they agree or disagree with several attitude statements. At a later time (often several weeks later), subjects are presented with an attitude questionnaire purportedly filled out by a stranger unknown to the subject. Subjects are told that the study in which they are participating is designed to investigate interpersonal judgment. Participants are then asked to review the stranger's questionnaire and indicate how much they like that person and to what extent they would like working with him or her. In reality, no actual stranger exists. The "stranger's" questionnaire is created by the experimenter based on the subject's attitudes that were obtained at the beginning of the study. In this way, the experimenter is able to manipulate the amount of similarity or dissimilarity between the subject and the bogus stranger. Extensive research using this paradigm has demonstrated that as the proportion of similar attitudes increases, attraction toward a target person increases in a linear function (Byrne & Nelson, 1965). This relationship is mediated by an implicit affective response. Specifically, similar attitude statements elicit positive affect which, in turn, elicits positive evaluative responses. Support for this hypothesis was supplied in an experiment designed by Byrne and Clore (1970) in which subjects completed a semantic differential scale as part of a typical attraction study. Similar attitude statements elicited responses indicating positive affect, and dissimilar statements elicited responses indicative of negative affect. Dissimilar attitude statements also result in physiological arousal, as measured by changes in skin conductance (Clore & Gormly, 1969). The power of associated affect is not restricted to our evaluations of others but also to the evaluation of other discriminable stimulus objects in the situation. For example, within the context of a typical attraction experiment, Meadow (1971) demonstrated that perceived attitude similarity resulted in subjects rating the experiment as being more enjoyable than those in the dissimilar condition, and perceiving that time passed more quickly -- showing once more that "time flies when you're having fun." f i e Attraction Response Dimension. Although much of the research on attraction has focused on interpersonal relationships, we must not lose sight of the fact that we are referring to a much more general

Byme and Neuman

32

evaluative process. This model can be applied to any situation involving likes and dislikes leading to approach versus avoidance behavior. Rather than being restricted to interpersonal relationships, the same conceptualization applies equally well in evaluations of all manner of stimuli. The attraction response dimension depicted in Figure 1 can easily be applied to employees, supervisors, managers, books, movies, sports events, geographical settings, products, political candidates, careers, job offers, salaries, training, corporate cultures, workplace environment, shopping malls, or any other stimuli we are called on to evaluate. Implicit in the scale depicted in Figure 1 is the fact that attraction is not an all or nothing response. In fact, attraction is found to be a linear function of the amount of positive affect in relation to the combined total of positive and negative affect elicited (Byrne & Nelson, 1965). The implications of attraction research for organizational issues will be examined at two levels of analysis: First, we will draw a parallel between marital relationships (initial attraction, marriage, learning about one another, long-term maintenance of the relationship, and possible dissolution) and occupational relationships (career and job choice, organizational entry, on-the-job performance, and turnover). Second, we will present evidence that factors that mediate interpersonal attraction, such as attitude similarity, overt stimulus characteristics, and propinquity, are also operative in the highly "objective" world of business.

ATTRACTION: INTERPERSONAL AND ORGANIZATIONAL The evaluationsone makes of anotherperson areconceptuallyparallel to the evaluations made of an occupation, career, or organization.

Selecting a Mate and Selecting an Organization Considering the long-term (at least theoretically) implicationsof a marriage and an occupation, we find it useful to point out the factors common to each selection process.

Propinquity. With no opportunity for interaction there can be no opportunity for attraction. Although self-evident, most of us proba-

bly underestimate the importance of propinquity in shaping our lives. Our proximity to a particular stimulus provides an opportunity for interaction and the level and quality of exposure results in a specific

Implications of Attraction Research

I

Hate

I

Strong Dislike

I

I

Mild Neutral Disliking

33

I

Mild Liking

I

Strong Liking

Figure 1: The attraction response dimension

I

Love

34

Byme and Neuman

affective response. A classic example of this fact involves the selection of a mate. It is unlikely that most people believe that their choice of a mate is influenced, to a major extent, by their physical proximity to that other individual prior to acquaintanceship. It may, therefore, be surprising to find that 34% of the first 5000 matriage licenses issued in Philadelphia, in 1931, were issued to men and women who lived within five blocks of each other (Bossard, 1932). In that investigator's words, "cupid may have wings, but apparently they are not adapted for long flights" (Bossard, 1932, p. 222). Both physical proximity (the physical distance between individuals) and functional proximity (the ease with which contact can take place) have been shown to facilitate friendship formation. For example, Caplow and Foreman (1950) found that close physical proximity between apartments, in a married student housing unit, predicted the likelihood of friendship formation. In a similar field study, Festinger, Schachter, and Back (1950) found that individuals who lived on the ground floor of an apartment building were more likely to have friends on the second floor if their apartments were located next to a stairway than was true for other ground floor residents. Students sitting in close proximity (side-by-side seating) to each other in a classroom are more likely to become friends than any other pairs of students in the room (Byrne & Buehler, 1955; Segal, 1974), and people are more likely to become friendly with their immediate next-door neighbors, especially when driveways are sideby-side, than with other neighbors (Whyte, 1956).

Repeated Exposure. Physical and functional proximity not only provide an opportunity for initial contact but also an opportunity for repeated exposure. In a series of experiments, Zajonc (1968) demonstrated that repeated exposure to a stimulus can result in a more positive (or less negative) evaluation of that stimulus. This tendency to feel better about familiar stimuli probably has survival value, in an evolutionary sense. Any novel person or object is potentially dangerous and hence threatening, whereas familiar objects are predictably safe because they caused no harm previously. Regardless of the underlying mechanism, however, research consistently indicates a preference for the familiar. Clearly, close physical and functional proximity facilitate repeated exposure and, hence, increasingly positive evaluations. People who come in contact with each other on multiple occasions are more likely to interact and more likely to become friendly -- assuming that the interactions are of a positive nature. Regardless of the nature of the interaction, a relationship is most likely to begin with an environmentally-guidedopportunity.

Implications of Attraction Research

35

Just as propinquity can influence one's evaluation of others and the selection of a mate, it can influence one's evaluation and selection of a career. Bandura (1969) stresses the importance and the ubiquity of vicarious learning that occurs when we observe what others do and the consequences of their acts. In choosing a career path, it is probably not the case that the average young person sits down and browses through more than 47,000 different occupations listed in me Dictionary of Occupational Titles. In place of this kind of systematic exposure to relevant information, most of us are more likely to respond to such unsystematic influences as our parents, school counselors, accidental exposure to specific occupations, and the job choices of our friends (Conger & Petersen, 1984). For example, even when controlling for social-class, sons of physicians, lawyers, and scientists are more likely to enter these occupations than young men whose fathers are not involved in these occupations (Mortimer, 1976; Werts, 1966; Werts, 1968). Young girls are more likely to express positive attitudes toward working both within and outside the home if their mothers have successfully balanced these roles (Baruch, 1972). Stagner (198 1) describes a situation that is quite possibly representative of the way most people wind up pursuing a particular career. In attempting to trace the development of Industrial-Organizational Psychology, Stagner (1981) solicited autobiographical material from 13 past presidents of Division 14 of APA (Division of Industrial and Organizational Psychology). The autobiographies supplied information about the participants' childhoods, graduate training, and early work experiences. Stagner reported that "one of the most striking observations from these autobiographical essays is the extent to which chance determined a person's entry into the field" @. 497). In fact, one of the respondents, Ray Katzell said, "If I had to name a single most important factor that shaped my career, I would have to say Circumstance. Or call it, if you prefer, Opportunity, Luck, Chance or God's Will" @. 497). When you consider the enormous impact that careers have in shaping our lives, including the way we perceive ourselves and how we are perceived by others, it is unsettling to believe that the choice of a career could be the result of a random process. Are our identity, prestige, purpose, and status merely "roles" of the dice? The revelation that 13 past presidents of Division 14 just "happened" upon their occupations may be startling but, on the basis of a great deal of past psychological research, it is scarcely a cause for surprise.

Byme and Newnan

36

Attitude Similarity. As discussed previously, perceived attitude similarity is one factor that can lead to interpersonal attraction. This effect is not restricted to relationships with another individual. Good and Good (1974) investigated the effect of perceived attitude similarity on attraction to a social organization. In this study, 65 college students received surveys supposedly filled out by members of a college fraternity. Subjects evaluated the fraternity on several measures and indicated their desire to belong to that group. Just as in the interpersonal attraction studies outlined earlier, attraction to the fraternity was a function of perceived attitude similarity to the group. In a similar vein, Good, Good, and Golden (1974) hypothesized that perceived similarity between job candidates and employees in a company would result in attraction to a company. The authors tested this hypothesis by presenting subjects with a set of heterogeneous attitudes supposedly obtained from employees in a business organization. As predicted, attraction to the company was a function of perceived attitude similarity to the employees. If attitude similarity and propinquity play such a significant role in both interpersonal and organizational attraction, ou would expect them to be equally important within the context o personnel selection, as they are ubiquitous to this process. Your faith in this assumption would not go unrewarded.

Y

PERSONNEL SELECTION: THE FIRST DATE When the representative of an organization first becomes aware of a job applicant, the decision to follow this up with face-toface contact resembles the decision to go on a first date with someone described by a friend, relative, or co-worker.

Applications, Resumes, and Letters of Recommendation In the early stages of the selection process, job applicants provide the prospective employer with a heterogeneous array of information, hoping to demonstrate their value to the organization. Although the intent is to provide job-relevant information, application blanks, resumes, and letters of recommendation provide many kinds of information which may not be directly related to future success on the job.

Stereotypical Assumptions about Names and Occupations. To begin, applications and resumes contain the applicant's name and address. Does the applicant have a "foreign sounding" name? Is this a minor-

Implications of Attraction Research

37

ity candidate? Is the candidate male or female? In response to preexisting biases, a personnel officer may respond in either a positive or negative way to each informational segment. For example, within the context of a public opinion poll, administered before an actual Canadian parliamentary election, Kamin (1958) asked voters to choose among fictitious candidates for a fictitious office. Results indicated that within each of two ethnic groups polled (either an English group in the city of Kingston or a French group in the city of Cornwall), respondents were significantly less likely to vote for a candidate whose name suggested an ethnic affiliation opposite to that of the respondent. Similarly, Byrne and Pueschel (1974) collected data on 3,600 candidates in 500 central committee elections in the state of California between 1948 and 1970. These researchers found that candidates with Scandinavian surnames had a 24 percent advantage over candidates with other ethnic names. Conversely, Spanish, Jewish, and Italian surnames resulted in an 11, 14, and 29 percent disadvantage, respectively. In another study, Lawson and Roeder (1986) had subjects rate women's first names on a 7-point semantic differential scale. Further, the investigators observed that positive and negative biases toward particular names could often be traced to television programs on which the names were used. For example, One Day at a Time was a popular television program at the time. The character of Ann, on the program, was portrayed as an iron-willed, women's liberation type, and this perception was attributed to the name Ann by some subjects in the study. In a similar vein, Paunonen, Jackson, and Oberman (1987) found that people possess preconceptions about ideal personality traits associated with certain occupations. Subjects who are perceived as possessing these particular attributes are more likely to be hired for these positions. Given our penchant for stereotyping in response to names, persons, and occupations, who do you suppose would have a better chance at landing a job as a construction worker: Clarence or Butch? And which of these two job applicants would be perceived as the better accountant?

Gender. If a "trivial" attribute such as one's name can affect the personnel selection process, what of a major stimulus characteristic such as gender? Some research has indicated that females often receive lower evaluations than their male counterparts (Cohen & Bunker, 1975; Dipboye, Fromkin, & Wiback, 1975; Ferris & Gilmore, 1977). This differential response to male and female applicants is particularly true in the case of personnel officers who are high in authoritarianism. A study by Simas and McCarrey (1979)

Byrne and Neuman

38

revealed that "high authoritarian" personnel officers of both sexes rated male applicants more favorably than female applicants. Recent research seems to indicate that gender related effects are most likely to occur with respect to jobs that are traditionally dominated by one sex over the other. That is, if jobs are typically dominated b males, then females will receive lower evaluations; conversely, i an occupation is generally dominated by females, then males are likely to receive lower evaluations (Cash, Gillen, & Bums, 1977; Cohen & Bunker, 1975; Heilman, 1980; Shaw, 1972). The perception of whether a job is "masculine" or "feminine" tends to be based on preconceptions associated with the job title and task composition. Cleveland and Smith (1989) had subjects rate various job titles and job descriptions in terms of probable job incumbents. The job of Warehouse Manager was clearly viewed as masculine while Health Care Services Coordinator was viewed as feminine. Additionally, subjects rated various tasks as being either masculine or feminine. Both job title and job content significantly affected perceptions of typical incumbents, the gender type of the job, and expectations about the probable success of the applicant. The chapter in this volume by Kelley and Streeter also addresses gender issues.

Y

Employment History. Application blanks also solicit information about employment and salary history. Obviously, personnel officers are looking for stability in a potential employee as well as signs of professional growth within and across jobs. Gaps in employment history and unimpressive promotions or salary increases are red flags to personnel officers. But there can be a more subtle factor in operation -- what is your present salary? Because women continue to earn less than their male counterparts, Mitchell and Henning (1987) decided that it would be important to look at the potential effects of this disparity in the hiring process. Undergraduate business students rated resumes from both male and female applicants who had either a high or low present salary. Compared to high present-salary applicants, low present-salary applicants were offered a lower starting salary for all jobs and were perceived as being less qualified for some jobs. Perceived Similarity. As previously discussed, perceived similarity is strongly associated with interpersonal attraction. Employment applications and resumes provide personnel officers with information relevant to the degree of similarity between themselves and the applicant. Some of the information that a job candidate includes may relate to personal interests, religious affiliations, hobbies, marital

Implications of Attraction Research

39

status, political preferences, number of children, and military history. Rand and Wexely (1975) found that personnel officers perceive job candidates who are biographically similar to themselves as being better suited to the job, more intelligent, and better liked. Aside from the content of supporting documents, there is the question of how the documents are prepared. Are they hand-written or typed? Are they visually striking in their appearance, or do they resemble a prescription prepared by a busy physician on his or her way to the golf course? Research has not examined such effects, but our theoretical model clearly predicts that evaluations of the applicant would be influenced by the quality of these materials.

Letters of Recommendation. Can you be "damned by faint praise?" Most letters of recommendation are written on behalf of the most gifted, honest, creative, intelligent, forthright, and dependable human beings to have ever graced this good earth. Against this background, it is possible that the slightest criticism will prove quite damaging to the applicant. Paradoxically, there is a possibility that because of the lack of variation in letters of recommendation (Muchinsky, 1979) a small amount of criticism may actually help. Knouse (1983) found that a small amount of negative information in a letter may improve the candidate's chances because the writer is viewed as being more honest in his or her evaluation.

Inte wiews Those job candidates who survive the evaluation process involving their resumes and application forms typically constitute a pool eligible for further scrutiny in a face-to-face interview, where additional and usually irrelevant attraction related variables can operate.

Foreign Accents. Stimulus characteristics which are not obvious in written communication may be extremely obvious within the context of a person-to-person meeting. For example, a candidate may not have a "foreign-sounding" name but a strong foreign accent. Aside from the potential difficulty in being able to communicate, foreign accents can elicit stereotypical images and pre-existing biases. In one study, Kalin and Rayko (1978) found that candidates with foreign accents received lower ratings for high status jobs and higher ratings for low status jobs. Of potential importance, but as yet unstudied, are the effects of regional accents within a society such as (for the United States) New England, Southern, or New York speech styles.

40

Byme and Neuman

Race, Discrimination in hiring, on the basis of race, has been the focus of a great deal of civil rights legislation. There is, however, comparatively little empirical research in this area and what does exist is rather inconclusive. In one study where fictitious resumes were mailed to actual employers, McIntyre, Moberg, and Posner (1980) found that black applicants received preferential treatment over white applicants. However, Newman and Krzystofiak (1979) found that when managers are aware of their role in a research study they are more likely to treat all applicants equally; in contrast, when unobtrusive measures are used, personnel managers are more likely to base decisions on the race of the job applicant. There also is an indication that an interaction exists between the job and the candidate in that black applicants receive lower evaluations for "white" jobs than equally qualified white applicants. Conversely, blacks receive higher evaluations when applying for "black" jobs (Schmitt & Hill, 1972; Terpstra & Larsen, 1980). Perceived Similarity. Just as with resumes and job applications, similarity variables operate in the interview process. A preponderance of both verbal and nonverbal stimuli contribute to perceptions of either similarity or dissimilarity with predictable results. In experimental research, Baskett (1973) found job applicants perceived as similar to the interviewer were viewed as more competent than dissimilar applicants and received recommendations for higher starting salaries. Similarly, Peters and Terborg (1975) found that similarity resulted in a greater perception of liking for the applicant, higher salary recommendations, and greater likelihood that the interview would culminate in a decision to hire the candidate. In a situation where students were supposedly being hired for positions as research assistants in a major university, Griffitt and Jackson (1970) found that attitude similarity correlated significantly with hiring recommendations ( r = S 3 ) and with salary recommendations ( r = .41). These investigators also found that attitude similarity related to perceptions of reliability, attractiveness, cooperativeness, goodness, and flexibility. Within the context of an actual job interview situation, Orpen (1984) had interviewers complete attitude questionnaires as they believed the interviewee would fill them out. Later, these questionnaires were compared with attitude questionnaires the interviewers themselves had completed. The outcome measure of interest was whether or not the applicant was actually hired. Orpen found that perceived similarity was strongly related to both attraction and hiring decisions.

Implications of Attraction Research

41

Agective Congruence. In addition to the effects of attitude similarity, Alliger and Williams (in press) point to the operation of what they have termed "affective congruence" (similarity of affect between the interviewer and interviewee). In a recent study, they assessed interviews conducted by 10 trained interviewers on 120 applicants for a university based early management potential program. Their results indicated that affective congruence, as opposed to interviewer or interviewee affect alone, was significantly associated with the decision to hire as well as with psychometric ratings of the applicants' management potential. It seems that both misery and joy love congruent company. Physical Appearance. Attitudes are not always readily apparent and emotions can be rather complex. There are, however, many other attraction relevant variables which are much less subtle in their operation -- physical attractiveness is one such variable. An examination of our culture's movies, television, advertising, and literature reveals that physical attractiveness is viewed as a highly desirable attribute. Because an endless array of health and fitness products and services are available to improve many aspects of appearance, few are "doomed" to an aesthetically unappealing future. Unattractive features, which cannot be sweated off or toned up, can be tucked, lifted, reduced, enlarged, reshaped, or removed by means of cosmetic surgery. Research has found that physically attractive people are liked better than unattractive individuals (Byrne, Ervin, & Lamberth, 1970; Byrne, London, & Reeves, 1968; Perrin, 1921; Walster, Aronson, Abrahams, & Rottman, 1966) and are associated with higher levels of social status (Thylor, 1956). There are also differential expectations about personali ty characteristics. Specifically, physically attractive individuals, as opposed to unattractive ones, are expected to be perceptive, confident, assertive, happy, active, amenable, humorous, and outspoken. They are also viewed as complex rather than simple, and flexible as opposed to rigid (Hatfield & Sprecher, 1986; Miller, 1970). "Not only are physically attractive persons assumed to possess more socially desirable personalities than those of lesser attractiveness, but it is presumed that their lives will be happier and more successful" (Dion, Berscheid, & Walster, 1972, p. 290). Physical attractiveness research also has been conducted in the area of organizational behavior. It was hypothesized that perceptions of goodness associated with beauty would influence personnel selection. Dipboye, Fromkin, and Wiback (1975) found that male college recruiters and business students both preferred physically attractive

42

Byme and Neuman

candidates to unattractive candidates and made their hiring decisions accordingly. In a follow-up study, Dipboye, Arvey, and Terpstra (1977), found that regardless of interviewer sex, highly qualified applicants were selected over unqualified applicants but, again, attractive candidates were preferred over unattractive candidates. Attractive candidates were also more likely to get higher starting salaries than unattractive candidates. Additionally, physical attractiveness has been shown to result in higher evaluations of job related qualifications (Cash, Gillen, & Burns, 1977). In this volume the chapter by Stone, Stone, and Dipboye discusses unattractiveness as a stigma, and the Kelley and Streeter chapter reviews research on gender in relation to this topic. Gillen (1981) has noted that there may be two types of "goodness" associated with physical attractiveness: a general goodness and a sex-typed goodness. Physically attractive females may be viewed as more feminine while physically attractive males are perceived as more masculine. Along these lines, it is possible that physical attractiveness may be more relevant to "sex-typed" jobs. Cash et al. (1977), tested this hypothesis by having subjects review resumes of male and female applicants for jobs previously rated as stereotypically masculine (automobile salesperson or wholesale hardware shipping and receiving clerk), stereotypically feminine (telephone operator or office receptionist), or neuter (motel desk clerk or photographic darkroom assistant). Results of this study indicated that males applying for masculine positions were rated as more qualified, were expected to be more successful, and received stronger hiring recommendations than equally qualified females. Conversely, females received significantly higher ratings than males with respect to perceived qualifications and expectations of success as well as hiring recommendations for stereotypically female jobs. Cohen and Bunker (1975) also found that interviewers were biased against male applicants applying for traditionally female positions just as they were for females applying for traditionally male jobs.

Cosmetic Use. For the reasons just stated, factors that affect the perception of masculinity and femininity can influence personnel decisions. Cox and Glick (1986) found that cosmetic use enhances female stereotypes and that women using cosmetics, as opposed to women depicted without cosmetics, received lower evaluations of expected performance in a secretarial position. Presumably, this provoked the sexist stereotype of the attractive secretary who was not hired as a result of her typing ability.

Implications of Attraction Research

43

Dress. Forsythe, Drake, and Cox (1985) investigated the effect of female applicants' dress on selection decisions. Personnel administrators viewed videotaped interviews of female applicants in either masculine or feminine attire. In this particular situation, women applying for management positions were more likely to be hired if they were dressed in masculine, as opposed to feminine, clothing. In similar research, Davis (1987) found that those dressed in masculine attire are viewed as more successful in their occupations than persons wearing feminine attire. Davis found this to be true in both traditionally masculine and feminine occupations. Obesity. At the beginning of this section, we alluded to a link between physical fitness and the perception of physical attractiveness. Obese individuals are not perceived as attractive. There is an obvious bias in our culture favoring slim as opposed to stout. In fact, Venes, Krupka, and Gerard (1982) found that students ranked embezzlers, cocaine users, shoplifters, and blind persons ahead of obese persons as a potential spouse. Within an organizational context, Larkin and Pines (1979) found that obese individuals are not considered good workers, and Rothblum, Miller, and Garbutt (1988) found that existing stereotypes of obese individuals included having poor personal hygiene, being lazy, and lacking discipline. Fortunately, obese individuals were also perceived as having a good sense of humor and of being friendly -- a decided advantage for those spending time on an unemployment line. Research seems to indicate that it is primarily the impact of obesity on physical attractiveness that generates negative affect and not obesity itself. When physical attractiveness is held constant, the effects of obesity are mitigated (Rothblum et al., 1988). Mrbal and Nonverbal Communication. Some other important attraction variables that have been studied i n an organizational setting include verbal cues, gestures, and propinquity. Because the ostensible purpose of an interview is to obtain information, what an applicant says to an interviewer should have a significant impact on that interviewer's evaluation of the applicant. This, in fact, does appear to be the case. In a study employing college campus recruiters, Hollandsworth, Kazelskis, Stevens, and Dressel (1979) found that the judged qualifications of an applicant were most affected by the appropriateness of verbal content. There is, however, more to the story. Hollandsworth et al. found that fluency of speech was second in importance, after speech content, in evaluating applicants. In actual interviews for positions at a large amusement park, Parsons and Liden (1984) found that verbal articulation accounted for 66% of

Byme and Newnan

44

the variance in interviewers' ratings of applicants' qualifications, and McGovern and Howard (1978) found that voice modulation and speech hesitations (urn's and ah's) affect the evaluation process. In face-to-faceinteractions, much that remains unspoken does not necessarily remain unsaid. A wide variety of nonverbal communications pervade the employment interview. One such factor has to do with functional propinquity and has been referred to as immediacy -"the degree of directness and intensity of interaction between two entities, such as two people" (Mehrabian, 1967, p. 325). Interactions between interviewer and interviewee marked by greater amounts of eye contact, smiling, closer physical proximit direct body orientation, and gesturing can convey a perception o increased availability. In a laboratory study conducted by Imada and Hakel (1977), job applicants who engaged in behaviors categorized as promoting immediacy were judged to be warmer and more enthusiastic. In addition, when compared to subjects who demonstrated low immediacy behaviors, the subjects in the "immediate" condition were better liked by the interviewer, perceived as more qualified, and judged to have a greater likelihood of success on the job. In fact, 86% of the subjects in the immediate condition were recommended for the job whereas only 19% of the subjects in the non-immediate condition were so evaluated. Amalfitano and Kalt (1977) have also found that direct eye contact results in rating applicants as more alert, assertive, dependable, confident, responsible, and possessing greater initiative. Altogether, eye contact, smiling, and gesturing have been shown repeatedly to result in more favorable interviewer evaluations (Amalfitano & Kalt, 1977; Imada & Hakel, 1977; Washburn & Hakel, 1973; Young & Beier, 1977). Hopefully we have demonstrated that we were not engaging in hyperbole at the beginning of this section when we described the personnel selection process as a first date. During both the initial attraction period and the selection phase, affect plays a central role in both types of interaction. Attraction variables are very much in evidence at every level of the evaluation process.

?

ORGANIZATIONAL ENTRY AND BEYOND: FOR BETTER OR WORSE? The roller-coaster of emotion that is present in the early stages of an interpersonal relationship is very much in evidence at the early stages of an organizational relationship. Having negotiated the hurdles present during the early stages of an interpersonal relationship, strong attraction may result in marriage. Having

Implications of Attraction Research

45

negotiated the analogous hurdles in the early stages of the organizational relationship, strong attraction also leads to marriage -"we now pronounce you employee and employer. 'I

Organizational Entry In the beginning, marriages and jobs are bathed in the rosy glow of positive emotions and expectations.

The Honeymoon. The beginning of an interpersonal relationship is usually an exciting period. It is generally marked by a substantial amount of positive affect which flows from a seemingly endless supply of smiles, hand-holding, longing gazes, verbal praises, and physical closeness. To some degree this is true for those newly hired in an organizational setting as well. Immediate superiors are likely to praise every occurrence of successful behavior and check to see that the new employee is comfortable and happy in his or her new environment. Colleagues are typically cordial and supportive and a good deal of attention is forthcoming from individuals charged with the responsibility of training the new employee. Unfortunately, "all good things must come to an end." Over a period of time (frequently, all too brief), the personal attention and positive reinforcement wane, and what was novel becomes routine. At this point, individuals must begin to deal with the more mundane, but extremely important, day-to-day aspects of the relationship. The Honeymoon Is Over. During the recruiting period (the courting stage), many promises are made or implied. Individuals enter into relationships with certain expectations which may, or may not, be fulfilled. In some cases, we are intentionally deceived by others. At other times, the deception may be unintentional or self-induced. For example, in interpersonal interactions we are motivated to say and do the things that are most appropriate in each situation and project the most favorable image possible (Baumeister, 1982; Goffman, 1959, 1963, 1967; Riess, Rosenfeld, Melburg, & Tedeschi, 1981). Analogous to this, organizations have traditionally engaged in a similar form of self-presentation, or impression management. In an attempt to attract a large pool of qualified applicants, organizations have typically engaged in "selling" themselves to potential employees by communicating information that is primarily favorable and flattering to themselves. In these instances, expectations are formed on the basis of information which was either intentionally distorted or withheld. At the same time, applicants are busily engaged in their own deceptions. Needless to say, this does not characterize a free

46

Bym and Neuman

and honest exchange of information. Apart from managing information meant for public consumption, human beings are quite adept at distorting information without conscious effort or external justification. Research indicates that in the absence of information to the contrary, human beings exhibit a general tendency to assume that others believe as they do and maintain similar attitudes (Byrne & Blaylock, 1963; Hensley & Duval, 1976; Levinger & Breedlove, 1966; Newcomb, 1978; Ross,. Greene, & House, 1977; Zuckerman, Mann, & Bernieri, 1982). Additionally, individuals are inclined to overestimate the probability of future success while underestimating the probability of future failure (Einhorn & Hogarth, 1978; Neuman, 1990). "Very few of us harbor dreams of a future consisting of a dull job, a drab home, an unhappy spouse, and unappreciative kids" (Byrne & Kelley, 1981, p. 515). To the extent that our expectations are overly optimistic or our attitudes and beliefs are incongruent with organizational realities, the stage is set for disappointment and dissatisfaction. This, in turn, can translate into substantial hardships for both individuals and organizations. Research has demonstrated an association between dissatisfaction and voluntary worker turnover (Mobley, Griffeth, Hand, & Meglino, 1979; Muchinsky & Tuttle, 1979; Porter & Steers, 1973) and absenteeism (Muchinsky, 1977; Porter & St'eers, 1973). Both of these behaviors can result in substantial financial loss to organizations (Mirvis & Lawler, 1977). On a more personal level, a 15-year longitudinal study of the aging process found that the single greatest predictor of longevity was work satisfaction (MIT, 1973) -- the ultimate dependent measure!

Satisfaction and Perj4ormance Though satisfaction is a worthwhile goal for its own sake and though job satisfaction is associated with less turnover, another question is obviously important: are satisfied employees more productive than dissatisfied ones?

Affect and Productivity. A good deal of empirical research has tended to contradict the popular notion that "a happy worker is a productive worker" (Iaffaldano & Muchinsky, 1985). It is clear that the relationship between satisfaction and worker productivity is not a simple one. Satisfaction in the absence of motivation or ability will not result in high performance. Rather than satisfaction leading to productivity, there has been more empirical support for the hypothesis that productivity leads to satisfaction (Locke, 1965, 1966, 1967; Porter & Lawler, 1968). Although the view of satisfaction as an

Implications of Attraction Research

47

antecedent to performance has been somewhat tarnished, the question is far from settled. In addition to determining evaluative responses, the elicitation of affect mediates cognitive processes and behavioral responses (Isen, 1984).

Aflect and Cognition. The creative process has been described as one in which diverse associations are formed into new combinations (Mednick, 1962) and problem solving results in solutions that are unusual and useful (Murray, 1959). Within these parameters, recent research has uncovered an association between positive affect and creativity. Persons in whom positive affect has been induced, by receiving a gift, a bag of candy, a compliment, or as a result of viewing a few minutes of a comedy film, tend to categorize stimuli more inclusively than subjects in a control condition (Isen & Daubman, 1984), give more unusual and diverse first-associates to neutral words (Isen, Johnson, Mertz, & Robinson, 1985), and demonstrate improved performance on tasks designed to assess creativity -- such as Duncker's (1945) candle task and M. T. Mednick, S.A. Mednick, and E.V. Mednick's (1964) Remote Associates Test (Greene & Noice, 1988; Isen, Daubman, & Nowicki, 1987). Positive affect also serves as a retrieval cue for positive material in memory (Isen, Shalker, Clark, & Karp, 1978; Laird, Wagener, Halal, & Szegda, 1982; Nasby & Yando, 1982; Teasdale & Fogarty, 1979). Conversely, subjects in whom negative affect has been induced tend to view ambiguous stimuli as less pleasant (Forest, Clark, Mills, & Isen, 1979; Isen & Shalker, 1982) and demonstrate lower tachistoscopic thresholds for failure-related words than subjects in a control condition (Postman & Brown, 1952). Emotional states also play a role in mediating "rational" decision making processes. Kahneman and Tversky (1983) demonstrated that preferences for specific decision alternatives change as a result of the manner in which alternatives are framed -- human beings demonstrate a general tendency toward risk-aversive behavior when considering potential gains and risk-seeking behavior when presented with a potential loss. This framing effect was demonstrated experimentally by Huber, Neale, and Northcraft (1987) within the context of a simulated personnel selection task. Subjects were presented with application material for 20 job applicants and asked to review this information. One group of subjects was instructed to select applicants they would accept for an interview while another group of subjects was asked to indicate which applicants they would reject for an interview. When cost salience of interviewing was high (subjects were explicitly advised that interviewing candidates was a very expensive process) subjects in the "gain" condition (those using

48

Byme and Neuman

an acceptance strategy) selected significantly fewer applicants to be interviewed than those in the "loss" condition (using a rejection strategy). From a purely "rational" decision-making perspective, the context in which the decision was being made should have had no bearing on the outcome of the selection process because all subjects were reviewing the same "objective" information. With respect to affect and risk-seeking, losses seem worse to people who are feeling happy -- they show a more negative subjective utility for losses than controls (Isen, Nygren, & Ashby, 1988).

Afect and Behavior. A considerable body of evidence suggests that positive affect is associated with helping behavior (Aderman, 1972; Berkowitz, 1987; Isen, 1970; Isen & Levin, 1972; Isen, Clark, & Schwartz, 1976). Generally, people are more willing to engage in helping behavior when they are in a positive, as opposed to negative, mood state. At the same time, people who are feeling happy prefer to remain in that state and are disinclined to engage in behaviors that will jeopardize this desirable condition. As previously mentioned, positive affect is strongly associated with interpersonal attraction. It is also associated with an increased willingness to initiate conversations with others (Batson, Coke, Chard, Smith, & Taliaferro, 1979; Isen, 1970) and an increased level of receptiveness to persuasive communication (Galizio & Hendrick, 1972; Janis, Kaye, & Kirschner, 1965). The relationship between positive and negative affect is not symmetrical; that is, the effects associated with negative affect are not always opposite to those associated with positive affect. For example, in a study by George (1989), positive affect was significantly and negatively associated with worker absenteeism while no significant relationship was demonstrated with respect to negative affect. In much of the research on emotional states, the effects associated with positive affect are fairly consistent and predictable while those associated with negative affect are rather inconsistent and unpredictable (Isen, 1984). On these grounds alone, positive affect has our vote. To summarize, positive affect is associated with increased levels of creativity, optimism, helping behavior, liking for others, and decreased levels of absenteeism. It impacts decision-making, risktaking, interpersonal communications, receptivity to persuasion, and approach-avoidance behavior. Satisfaction-pe~ormanceRevisited. We propose that affect plays a more important role in worker productivity than has been documented to date. The weak relationship between satisfaction and performance may be based on methodological problems associated with the

Implications of Attraction Research

49

measurement of satisfaction (refer to footnote 3) and the evaluation of performance. If worker productivity is primarily measured in terms of products manufactured, lines of computer code generated, management tasks completed, or services rendered, our assessment will be incomplete. What about inventions not invented, options not considered, chances never taken, assistance never rendered, and extra effort not invested? Whether measuring productivity in terms of quantity or quality, we are constrained to measure what is and not what might have been. The question may not be whether positive affect leads to performance but, rather, when does it lead to increased performance, under what circumstances, and in what manner. There are times when positive affect may prove detrimental -- being risk-averse is not a virtue in all situations nor is being more receptive to persuasive communications (assuming that you are the target of the persuasion attempt). The skill, knowledge, and ability required for successful completion of a particular task varies considerably from job to job. Consequently, the degree to which positive affect facilitates or inhibits performance will vary accordingly. We now turn our attention to the relationships among attraction variables, resultant affect, and worker productivity.

Interpersonal Attraction and Wrker Pe formance In a work setting, does it matter whether an individual likes, dislikes, or is indifferent to his or her fellow workers?

Attitude Similarity. Krivonos, Byrne, and Friedrich (1976) hypothesized that an educational setting which fosters an affectively positive atmosphere should facilitate performance on various intellectual tasks. Because attitude similarity is strongly associated with positive affect, it was hypothesized that subjects paired with similar others would perform significantly better on various cognitive tasks than subjects paired with dissimilar others. Results indicated that subjects paired with similar others felt affectively more positive and performed significantly better on tasks that required analysis, synthesis, integration, and recall of textual material as well as an evaluative task that required judgment and decision making. Analogous to this, high interpersonal attraction has been shown to result in increased levels of performance on a conditioning task (Sapolsky, 1960), learning Spanish equivalents of English words (Lott & Lott, 1966), and simple mechanical tasks (Nelson & Meadow, 1971). Attitude similarity also impacts worker performance by mediating group dynamics. Inter-member liking, resulting from attitude

50

Bym and Newnan

similarity, leads to a more cohesive work group. Depending upon the task, cohesiveness can lead to higher levels of performance. Within the context of a longitudinal field experiment, Terborg, Castore, and DeNinno (1976) were unable to confirm a relationship between attitude similarity and performance but they did find that, over time, attitude similarity leads to group cohesiveness. Other research has demonstrated that group cohesiveness is associated with increased levels of innovation (West & Farr, 1989), increased quality of communication among group members and lower levels of absenteeism and turnover (Lott & Lott, 1966; Stogdill, 1972). Additionally, because group membership is highly valued and communications are unrestrained, within cohesive work groups, individual members are more likely to respond to group norms. Therefore, cohesive groups are likely to perform better than noncohesive groups when performance standards a r e high (Berkowitz, 1954; Schachter, Ellertson, McBride, & Gregory, 1951). Another outcome associated with interpersonal attraction is that of cooperation and participation, as opposed to conflict and competition. In an experiment examining productivity as a function of intergroup cooperation/competition and intra-group cooperation/competition, Workie (1974) found that cooperation was significantly more productive than competition. Analogous to this, in a meta-analysis reviewing 122 studies comparing the effectiveness of cooperation and competition, Johnson, Maruyama, Johnson, Nelson, and Skon (1981) found that cooperation is superior to competition in promoting achievement and productivity. Perceived similarity has also been found to be associated with worker motivation. Within the context of equity theory (Adams, 1963), individuals seek to maximize equity in relationship to a comparison other. Whether being overpayed or underpayed, individuals are motivated to restore equity. Griffeth, Vecchio, and Logan (1989) investigated the relationship between interpersonal attraction (attitude similarity-dissimilaritybetween subjects and a comparison other) and three levels of equity (underpayment, overpayment, and equitable payment). Results indicated a significant interaction between level of equity and interpersonal attraction on several measures. Subjects in the high attraction condition (similar attitudes), indicated that they would work significantly more hours if given the opportunity, but this effect was only noted for subjects in the underpayment or equitable payment conditions. Those in the low attraction condition (dissimilar attitudes) would only agree to additional participation if they were in the overpayment condition. High attractiveness also accounted for more variance across equity conditions for

Implications of Attraction Research

51

subjective and objective measures of performance quality. Because attitude similarity can determine evaluative responses, it should come as no surprise that it can impact the performance appraisal process. Pulakos and Wexley (1983) found that perceived similarity between managers and their subordinates led each to give higher performance ratings to the other. Turban and Jones (1988) found that superiors who perceived subordinates as being similar to themselves recommended significantly higher merit increases than those they perceived as being dissimilar. According to Landy and Farr (1980), judgmental measurements of performance are in wide use -- we suspect that attitude similarity is in wide use as well when appraisals are made.

Propinquity. Both physical and functional propinquity play a role in worker productivity. The size of a work group, and its physical setting, will dictate the nature, quality, and frequency of worker interaction. Employees working together in a small and open work area are more likely to form a cohesive unit than workers separated by partitions, offices, or great distances. Even the way furniture is arranged in a room can serve to facilitate or inhibit productivity. As an example, arranging work stations in such a way as to reduce interaction (unnecessary socializing on the job) can actually decrease, rather than increase, productivity (Baskin & Aronoff, 1980). Other research demonstrates that placing a desk or table between individuals results in a less friendly and more formal atmosphere (McCaskey, 1979). Propinquity can also involve inappropriately close (or distant) interpersonal space. Muchinsky (1990) gives an example of the subtle, but significant, impact of cultural differences on preferred interpersonal distances in describing an interaction between two diplomats. One diplomat was from a northern European country and the other was from a country in the Middle East. As Hall (1959) has reported, people from northern Europe prefer greater physical distance when interacting with others than individuals from the Middle East. Throughout their conversation, the diplomat from the Middle East would attempt to move closer while the European diplomat worked to regain what he considered to be an appropriate physical distance. After returning to their respective embassies, the diplomat from Northern Europe described the representative from the Middle East as "incessantly pushy," while the Middle Eastern diplomat felt that he had been dealing with someone "cold and standoffish" (Muchinsky, 1990). Neither of the diplomats thought the meeting went particularly well. The effects of functional propinquity are clearly evidenced during

52

Byrne and Neuman

interactions between superiors and subordinates. To the extent that a superior maintains a functional distance from the subordinate by asserting his or her status and power, the interaction is marked by a high degree of formality and guarded communications. This is not conducive to an honest and open exchange of ideas.

Overt Stimulus Characteristics. We have identified numerous ways in which a wide variety of stimulus characteristics, such as gender, age, physical attractiveness, verbal and nonverbal communications affect interpersonal processes. These variables are always present in the day-to-day operation of business and impact the process in more ways than we can imagine, let alone recount. We will, therefore, confine our discussion to just a few examples. Leadership can be defined as 'la process whereby an individual influences the group toward the attainment of desired group or organizational goals" (Mitchell & Larson, 1987, p. 434). Although leaders should be chosen for their ability to lead groups in the right direction, very often leaders are chosen on the basis of physical attributes and not leadership ability. For example, regardless of their sex, individuals are more likely to emerge as group ieaders if they possess masculine, as opposed to feminine or androgynous, gender role characteristics (Goktepe & Schneier, 1989) and are physically attractive (Stogdill, 1974). Additionally, Americans like to "look up" to their leaders. Trml individuals are more likely to obtain positions of leadership and authority; for example, Americans have overwhelmingly preferred the taller of the two major predidential candidates in each election since 1900 (Feldman, 1975). To the extent that leader emergence is based on the aforementioned physical attributes, as opposed to leadership ability, group performance may be adversely affected. Environmental stimuli also can affect worker productivity. Beyond the obvious considerations, such as sufficient space, efficient office and equipment layouts, appropriate safety considerations, and properly designed and maintained equipment, there are more import considerations; such as, who gets the office with the window and who gets the office with the carpeting? Is your desk made of metal or mahogany? In fact, do you even have a desk or an office? In a recent field experiment, Greenberg (1988) demonstrated that office space constitutes an important job reward. In this study, life insurance underwriters were moved to new offices because their own offices were being refurbished. Some of these individuals were moved to offices of higher-status persons (spacious and well decorated), others were moved to offices of lower-status workers (more cramped and stark in appearance), and still others were moved to

Implications of Attraction Research

53

offices of equal-status persons. Finally, a control group of underwriters continued working in their own offices. Results revealed that subjects moved into higher-status offices demonstrated significantly increased levels of productivity, and subjects moved into lower status offices demonstrated significantly reduced levels of productivity. Performance levels for the control group and equalstatus group remained unchanged during the study. At the conclusion of this study, all subjects were returned to their original offices and their performance returned to pre-experimental levels. Assuming that you are fortunate enough to choose the right career, be selected by the right organization, land the right job, occupy the right office, work for the right boss and with the right kind of people, you might view your prospects in rather optimistic terms. Let us further suppose that these conditions remain constant over an extended period of time. At this point, you might be prepared to predict a storybook ending; however, you're not out of the woods just yet.

ORGANIZATIONAL EXIT: SEPARATION AND DIVORCE I cannot love as I have loved, And yet I know not why; It is the one great woe of life To feel all feeling die. [Philip James Bailey: Festus]

Habituation, Evaluation, and Acffect Robert Bums' observation that the best laid plans of mice and men often go astray is certainly operative within the context of any relationship, especially one of a long-term nature. Relationships can fail for many reasons and certainly poor planning is one of them. But there are other causative factors that have less to do with wisdom and more to with the physiology of the organism and the dynamics of the situation. Although repeated exposure to a stimulus tends to result in a more positive, or less negative, evaluative response to that stimulus (Zajonc, 1968), one can get too much of a good thing -- familiarity can breed contempt.

Habituation. The capacity of a stimulus to elicit positive affect can diminish over time. This may result in mere indifference toward that

54

Byme and Newan

stimulus (boredom) or, in the extreme, a highly negative evaluative response (hate). In long-term romantic relationships, this may manifest itself in the form of declining interest in one's sexual partner (Griffitt, 1981). The brisk rentals of sexually explicit videotapes, and numerous advertisements for marital aids, support the observation that people are looking for ways to heighten (or revitalize) their sexual experiences. The diminished capacity of a familiar stimulus to result in sexual arousal is not restricted to human beings. Habituation to sexual stimuli also has been demonstrated in laboratory studies involving animals (Clemens, 1967; Krames, Costanzo, & Carr, 1967). The impact of habituation on both interpersonal and organizational relationships can be substantial. For example, Hill, Rubin, and Peplau (1976) found that for both males and females, the most frequently cited cause of a failed relationship was "boredom.I' The effects of habituation have been noted in an organizational context, as well. Workers involved in highly specialized or routine jobs become bored and dissatisfied (Dunham, 1979). In an effort to combat this problem , organizations have instituted job enlargement programs to add variety and job enrichment programs to add responsibility and autonomy (Lawler, 1969). The research evidence on the effectiveness of these programs is mixed. Although some research suggests that these programs can result in increased levels of job satisfaction and reduced levels of turnover (Ford, 1969; Paul, Robertson, & Herzberg, 1969), other research suggests that these gains may be short-lived (Maher & Overbagh, 1971). It has been suggested that these short-term gains may be due, in part, to a Hawthorne effect (Muchinsky, 1990, p. 454). Possibly, workers are only responding, for a period of time, to the novelty of the situation and the attention they receive from individuals running the program. In short, this is performance in response to a novel stimulus -- sound familiar? Landy (1978), in noting that an individual's level of satisfaction will change over time (even if all aspects of the job remain constant) provides a physiological explanation for the habituation effect. He proposes that an "opponent process" works against both highly negative or positive emotions in such a way as to maintain a state of equilibrium in the central nervous system. According to this theory, the nervous system produces a response in opposition to that of the emotional stimulus. Each time this opponent process is activated it becomes stronger, which explains the reduction in affect over time. Fortunately, in attempting to explain habituation, we are not constrained to a static environment. There are a sufficient number of environmental explanations for the variability in worker satisfaction.

Implications of Attraction Research

55

Evaluation. As indicated in the previous section pertaining to organizational entry, the continual support and reinforcement that is present upon entry into an organization dissipates over time and the ratio of positive to negative interactions may become decidedly more negative in nature. In a study conducted by Birchler, Weiss, and Vincent (1975), comparing distressed and nondistressed married couples, observations were made when the couples interacted with each other and also when they interacted with strangers. In the case of distressed couples, there was a tendency to agree and express approval when interacting with strangers. However, when interacting with each other, distressed couples were more likely to engage in negative forms of communication -- marked by complaints and criticism (Levinger & Senn, 1967). In the workplace, complaints, criticism, and coercive communications may occur in the course of daily interpersonal interaction or they may manifest themselves within more formal contexts. Structurally, large organizations are departmentalized and this, in turn, leads to natural conflict. For example, in the clothing industry, the sales and marketing departments may dream of unlimited inventories containing a diverse assortment of stock -- clothing to suit every size and taste, available on demand. For the people in production, this constitutes their worst nightmare. Production dreams of a single product manufactured to a single specification and the finance department's fantasy involves the prospect of zero overhead. There are, to be sure, diverse interests to be satisfied in any organization. If these interests are not integrated properly, conflict will be the rule and not the exception. The formal structure of business also involves hierarchies, promotions, salary increases, merit pay, bonuses, and performance appraisal systems. These items are most often associated with employee dissatisfaction or, at best, indifference (Herzberg, Mausner, & Snyderman, 1959). Regardless of how it occurs, over a period of time, the level of rewards an employee derives from employmen t may decline while costs attributed to continued employment may increase. This can result in serious consequences to all concerned. Fanell and Rusbult (1981) noted that employees who exited an organization, as opposed to those who remained, experienced a greater decline in the level of rewards and an increase in the level of costs.

Inter ersonal Attraction and Emp oyee-Organizational Linkages

P

Fortunately for the institutions of marriage and business, many relationships not only endure but actually flourish. However, in

56

Byme and Newnan

those instances where negative affect predominates, employee-organizational linkages may become severely strained and, eventually, broken.

Satisfaction and Withdrawal. As previously discussed, attraction is a linear function of the amount of positive and negative affect elicited. Given a decrease in positive affect, due to habituation, or an increase in negative affect, in response to various environmental stimuli, the attraction response is affected accordingly. Initial evaluations of love or strong liking may degenerate to mere indifference, mild dislike, strong dislike, or hate. As the attraction response deteriorates, so does the linkage between the organization and the employee. Up to this point, we have only considered worker absenteeism and voluntary turnover in terms of their costs to an organization. There are instances where this is not necessarily the case. Dissatisfied workers who remain on the job can engage in activities that are very costly to an organization. For example, employees who are dissatisfied about compensation have been known to engage in sabotage, damaging important production equipment (Stagner & Rosen, 1965). Or, at the very least, dissatisfied workers may engage in a pattern of neglect, passively allowing the situation to deteriorate (Farrell, 1983). Under these circumstances, turnover can be desirable for both employee and employer. What causes a dissatisfied employee to engage in sabotage or neglect, as opposed to organizational exit? Or, in more general terms, what causes a dissatisfied person to leave any relationship? Within the context of romantic relationships, Rusbult, Zembrodt, and Gunn (1982) investigated the relationship among three variables and the decision to engage in four behaviors: exit (leaving the relationship), voice (active, constructive, behavior), loyulty (passive support for the relationship), and neglect (passive, destructive behavior). The three predictor variables involved: (1) prior satisfaction with the relationship; (2) the amount of resources an individual has invested in the relationship, and; (3) the presence and quality of alternatives to the relationship. Prior levels of high satisfaction with the relationship, and high levels of investment, were associated with voice and loyalty and the decreased likelihood of neglect and exit. Conversely, lower levels of satisfaction and investment, as well as the presence of attractive alternatives, was associated with an increased likelihood of exit. These results have been supported within the context of romantic relationships (Rusbult, Johnson, & Morrow, 1986) and organizational relationships (Farrell & Rusbult, 1981; Rusbult & Farrell, 1983; Rusbult, Farrell, Rogers, & Mainous, 1988). Within the attraction paradigm, this can be conceptualized as the

Implications of Attraction Research

57

number and intensity of reciprocal rewards exchanged within a relationship as compared to the number and intensity of reciprocal punishments (Byrne & Blaylock, 1963). This ratio is then compared to a ratio of the rewards and punishments associated with leaving the organization.

CONCLUDING COMMENTS Human beings with their attitudes, judgments, beliefs, expectations, values, and fantasies are the organization. Organizations have no life apart from people and process. It is human decision-making that is centralized or decentralized, not some inanimate organizational structure. It is people w h o are departmentalized, not buildings, plants, and divisions. It is the work that people accomplish that must be integrated, not merely systems and functions. And so, for good or ill, attitude similarity, physical attractiveness, gender, race, firm handshakes, foreign accents, tallness, shortness, offices with windows, and large mahogany desks influence our attitudes, color our judgment, and affect our behavior. They can be blatantly obvious or extremely subtle in operation. Regardless of their intensity, they pervade every aspect of our daily lives. We began this chapter by suggesting that people are generally disinclined to admit the influence of affect in the dollars-and-cents world of business while quite willing to acknowledge its influence in other domains. We hope that we have accomplished our two objectives: (1) demonstrated that the evolution of romantic relationships parallels that of organizational relationships and (2) provided a considerable amount of empirical evidence to support our contention that affect plays an important role in both business and pleasure. We trust that after you have had an opportunity to digest all of the information provided, and carefully consider all of the evidence presented, you will be in a position to make an informed judgment about the merits of our case. More likely, by the time you reached the second page of this chapter, you either liked it or you didn't. Of course, we don't know that for sure, it's just a feeling.

NOTES 1. Rosenbaum has suggested that rather than similarity leading to attraction it is actually dissimilarity which leads to repulsion. For a discussion of this "repulsion hypothesis," and a presentation of disconfirming data, the reader is directed to Byme, Clore, & Smeaton (1986), Rosenbaum (1986a; 1986b), and Smeaton, Byme, & Mumen (1989).

Byme and Neuman

58

2. Contrary to this traditional approach of "selling" the organization to potential job candidates, many organizations now employ realistic recruiting procedures which attempt to convey information which more accurately reflects the conditions employees are likely to encouter. Refer to Wanous (1980) for a thorough treatment of this topic.

3. George (1989) points out that much of the theorizing and research related to job satisfaction has focused on cognition rather than affect. To the extent that these variables may operate independently of each other (Zajonc, 1980, 1984), the observed relationship between satisfaction and performance will be weakened.

REFERENCES Adams, J. S. (1963). Toward an understanding of inequity. Journal of Abnormal and Social Psychology, 67, 422-436. Aderman, D. (1972). Elation, depression and helping behavior. Journal of Personality and Social Psychology, 24, 91-101. Alliger, G. M., & Williams, K. J. (in press). Affective congruence and the employ men t interview. Advances in Information Processing. Amalfitano, J. G.? & Kalt, N. C. (1977). Effects of eye contact on the evaluation of job applicants. Journal of Employment Counseling, 14, 46-48. Bandura, A. (1969). Principles of behavior modijication. New York: Holt , Baruch, G. K. (1972). Maternal influences upon college women's attitudes toward women and work. Developmental Psychology, 6, 32-37. Baskett, G. D. (1973). Interview decisions as determined by competency and attitude similarity. Journal of Applied Psychology, 57, 343-345. Baskin, 0. W., & Aronoff, C. E. (1980). Interpersonal communication in organizations. Glenview, Illinois: Scott, Forseman. Batson, C. D., Coke, J. S . , Chard, F., Smith, D., & 'Pdliaferro, A. (1979). Generality of the "Glow of goodwill": Effects of mood on helping and information acquisition. Social Psychology Quarterly, 42, 176-179. Baumeister, R . F. (1982). A self-presentational view of social phenomena. Psychological Bulletin, 91, 3-26. Becker, F. D., Gield, B., Foggatt, C. C. (1983). Seating position and impression formation in an office setting. Journal of Environmental Psychology, 3, 253-261.

Implications of Attraction Research

59

Berkowitz, L. (1954). Groups standards, cohesiveness, and productivity. Human Relations, 7, 509-519. Berkowitz, L. (1987). Mood, self-awareness, and willingness to help. Journal of Personality and Social Psychology, 52, 721729. Birchler, G. R.., Weiss, R. L., & Vincent, J. P. (1975). Multimethod analysis of social reinforcement exchange between maritally distressed and nondistressed spouse and stranger dyads. Journal of Personality and Social Psychology, 31, 349-360. Bossard, J. H. S. (1932). Residential propinquity as a factor in marriage selection. American Journal of Sociology, 38, 2 19224. Byrne, D. (1961). Interpersonal attraction and attitude similarity. Journal of Abnormal and Social Psychology, 62, 7 13-715. Byrne, D. (1971). The attraction paradigm. New York: Academic Press. Byrne, D., & Blaylock, B. (1963). Similarity and assumed similarity of attitudes between husbands and wives. Journal of Abnormal and Social Psychology, 67, 636-640. Byrne, D.? & Buehler, J. A. (1955). A note on the influence of propinquity upon acquaintanceships. Journal of Abnormal and Social Psychology, 51, 147-148. Byrne, D., & Clore, G. L. (1970). A reinforcement model of evaluative responses. Personality: An International Journal, 1 , 103-128. Byrne, D., & Clore, G. L., & Smeaton, G. (1986). The attraction hypothesis: Do similar attitudes affect anything? Journal of Personality and Social Psychology, 51, 1167-1170. Byrne, D., London, O., & Reeves, K. (1968). The effects of physical attractiveness, sex, and attitude similarity on interpersonal attraction. Journal of Personality, 36, 259-271. Byrne, D., & Nelson, D. (1965). Attraction as a linear function of proportion of positive reinforcements. Journal of Personality and Social Psychology, 1 , 659-663. Byrne, D., Ervin, C. R., & Lamberth, J. (1970). Continuity between the experimental study of attraction and real life computer dating .-Journal of Pehonality and Social Psychology, 16, 157-165. Byrne, D., & Kelley, K. (1981). An introduction to personality (3rd ed.). New Jersey: Prentice Hall. Byrne, G. C., & Pueschel, J. K. (1974). But who should I vote for for county coroner? Journal of Politics, 36, 778-784.

60

Byrne and Neuman

Caplow, T., & Foreman, R. (1950). Neighborhood interaction in a homogeneous community. American Sociological Review, 15, 357-366. Cash, T. F.,. Gillen, B., & Burns, D. S. (1977). Sexism and "beautyism" in personnel consultant decision making. Journal of Applied Psychology, 62, 301-310. Clemens, L. G. (1967). Effects of stimulus female variation on sexual performance of the male deermouse. Peromyscus Maniculatus. Proceedings, 75th Annual Convention of the APA, 119120. Cleveland, J. N., & Smith, L. A. (1989). The effects of job title and task composition on job and incumbent perceptions. Journal of Applied Social Psychology, 19, 744-757. Clore, G. L:,& Byrne, D. (1974). A reinforcement-affect model of attraction. In T. L. Huston (Ed.), Foundations of interpersonal attraction @p. 143-170). New York: Academic Press. Clore, G. L., & Gormly, J. B. (1969, November). Attraction and physiological arousal in response to agreements and disagreements. Paper presented at the meeting of the Psychonomic Society, St. Louis, MO. Cohen, S. L., & Bunker, K. A. (1975). Subtle effects on sex role stereotypes on recruiters' hiring decisions. Journal of Applied Psychology, 60, 566-572. Conger, J. J., & Petersen, A. C. (1984). Adolescence and youth: Psychological development in a changing world (3rd ed.). New York: Harper & Row. Cox, C. L., & Glick, W. H. (1986). Resume evaluations and cosmetics use: When more is not better. Sex Roles, 14, 51-58. Davis, L. L. (1987). Effect of sex, inferred sex-role and occupational sex-linkage on perceptions of occupational success. Perceptual and Motor Skills, 64, 887-898. Dion, K. K., Berscheid, E., & Walster, E. (1972). What is beautiful is good. Journal of Personality and Social Psychology, 24, 285-290. Dipboye, R. L., Arvey, R. D., & Terpstra, D. E. (1977). Sex and physical attractiveness, of raters and applicants as determinants of resume evaluations. Journal of Applied Psychology, 62, 288-294. Dipboye, R. L., Fromkin, H. L., & Wiback, K. (1975). Relative importance of applicant sex, attractiveness, and scholastic standing in evaluation of job applicant resumes. Journal of Applied Psychology, 60, 39-43. Duncker, K . (1945). On problem solving. Psychological Monographs, 58 (5, Whole no. 270).

Implications of Attraction Research

61

Dunham, R. B. (1979). Job design and redesign. In S. Kerr (Ed.), Organizational behavior. Columbus, OH: Grid. Einhorn, H. J., & Hogarth, R. M. (1978). Confidence in judgment: Persistence of the illusion of validity. Psychological Review, 85, 395-416. Farrell, D. (1983). Exit, voice, loyalty, and neglect as responses to job dissatisfaction: A multidimensional scaling study. Academy of Management Journal, 26, 596-606. Farrell, D., & Rusbult, C. E. (1981). Exchange variables as predictors of job satisfaction, job commitment, and turnover: The impact of rewards, costs, alternatives, and investments. Organizational Behavior and Human Perjormance, 27, 78-95. Feldman, S. D. (1975). The presentation of shortness in everyday life -- height and heightism in American society: Toward a sociology of stature. In S. D. Feldman & G. W. Theilbar (Eds.), Life styles: Diversity in American society. Boston: Little, Brown. Fems, G. R., & Gilmore, D. C. (1977). Effects of mode of information presentation, sex of applicant, and sex of interviewer on simulated interview decisions. Psychological Reports, 40, 566. Festinger, L., Schachter, S., & Back, K. (1950). Social pressures in informal groups: A study of a housing community. New York: Harper. Ford, R. N. (1969). Motivation through the work itself. New York: American Management Association. Forest, D., Clark, M. S., Mills, J., & Isen, A. M. (1979). Helping as a function of feeling state and nature of the helping behavior. Motivation and Emotion, 3, 161- 169. Forsythe, S . , Drake, M. F., & Cox, C. E. (1985). Influence of applicant's dress on interviewer's selection decisions. Journal of Applied Psychology, 70, 374-378. Galizio, M., & Hendrick, C. (1972). Effect of musical accompaniment on attitude: The guitar as a prop for persuasion. Journal of Applied Social Psychology, 2 , 350-359. George, J. M. (1989). Mood and absence. Journal of Applied Psychology, 74, 317-324. Gillen, B. (1981). Physical attractiveness: A determinant of two types of goodness. Personality and Social Psychology Bulletin, 7, 277-281. Goffman, E. (1959). The presentation of self in everyday life. Harmondsworth: Penguin. Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood Cliff, New Jersey: Prentice Hall.

62

Byme and Neuman

Goffman, E. (1967). Interaction ritual: Essays onface-to-face behavior. Garden City, New York: Doubleday. Goktepe, J. R., & Schneier, C. E. (1989). Role of sex, gender roles, and attraction in predicting emergent leaders. Journal of Applied Psychology, 74, 165-167. Good, L. R., & Good, K. C. (1974). Similarity of attitudes and attraction to a social organization. Psychological Reports, 34, 1071-1073. Good, L. R., Good, K. C., & Golden, S. B. (1974). Attitude similarity and attraction to a company. Psychology, 11, 52-55. Gouaux, C. (1971). Induced affective states and interpersonal attraction. Journal of Personality and Social Psychology, 20, 37-43. Greenberg, J. (1988). Equity and workplace status: A field experiment. Journal of Applied Psychology, 73, 606-613. Greene, T. R., & Noice, H. (1988). Influence of positive affect upon creative thinking and problem solving in children. Psychological Reports, 63, 895-898. Griffeth, R. W., Vecchio, R. P., & Logan, Jr., J. W. (1989). Equity theory and interpersonal attraction. Journal of Applied Psychology, 74, 394-401. Griffitt, W. (1970). Environmental effects on interpersonal affective behavior: Ambient effective temperature and attraction. Journal oJ”Persona1ityand Social Psychology, 15, 240-244. Griffitt, W. (1981). Sexual intimacy in aging marital partners. In J. Marsh & S. Kiesler (Eds.), Aging: Stability and change in the family @p. 301-315). New York: Academic Press. Griffitt, W., & Jackson, T. (1970). The influence of ability and nonability information on personnel selection decisions. Psychological Reports, 27, 959-962. Hall, E. T. (1959). The silent language. Garden City, NY Doubleday. Hatfield, E., & Sprecher, S . (1986). Mirror, mirror.. . n e importance of look in everyday life. Albany: SUNY Press. Hensley, V.? & Duval, S. (1976). Some perceptual determinants of perceived similarity, liking, and correctness. Journal of Personality and Social Psychology, 34, 159-168. Heilman, M. E. (1980). The impact of situational factors on personnel decisions concerning women: Varying the sex composition of the applicant pool. Organizational Behavior and Human Peformance, 26, 386-396. Herzberg, E., Mausner, B., & Snyderman, B. (1959). The motivation to work. New York: Wiley.

Implications of Attraction Research

63

Hill, C. T.., Rubin, Z., & Peplau, L. A., (1976). Breakups before mamage: The end of 103 affairs. Journal of Social Issues, 32, 147-168. Hollandsworth, J. G., Kazelskis, A., Stevens, J., & Dressel, M. E. (1979). Relative contributions of verbal, articulative, and nonverbal communication to employment decisions in the job interview setting. Personnel Psychology, 32, 359-367. Huber, V. L., Neale, M. A., & Northcraft, G. B. (1987). Decision bias and personnel selection strategies. Organizational Behavior and Human Decision Processes, 40, 136-147. Iaffaldano, M. T., & Muchinsky, P. M. (1985). Job satisfaction and job performance: A meta-analysis. Psychological Bulletin, 97, 251-273. Imada, A. S . & Hakel, M. D. (1977). Influence of nonverbal communication and rater proximity on impressions and decisions in simulated employment interviews. Journal of Applied Psychology, 62, 295-300. Isen, A. M. (1970). Success, failure, attention and reactions to others: The warm glow of success. Journal of Personality and Social Psychology, 15, 294-301. Isen, A. M. (1984). Toward understanding the role of affect in cognition. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition (Vol. 3., pp. 179-236). Hillsdale, NJ: Lawrence Erlbaum Associates. Isen, A. M., & Daubman, K. A. (1984). The influence of affect on categorization. Journal of Personality and Social Psychology, 47, 1206-1217. Isen, A. M., & Levin, P. E. (1972). The effect of feeling good on helping: Cookies and kindness. Journal of Personality and Social Psychology, 21, 384-388. Isen, A. M., & Shalker, T. E. (1982). The influence of mood state on evaluation of positive, neutral, and negative stimuli: When you "accentuate the positive, 'I do you "eliminate the negative"? Social Psychology Quarterly, 4.5, 58-63. Isen, A. M., Clark, M., & Schwartz, M. F. (1976). Duration of the effect of good mood on helping: "Footprints on the sands of time." Journal of Personality and Social Psychology, 34, 385393. Isen, A. M., Daubman, K. A., & Nowicki, G. P. (1987). Positive affect facilitates creative problem solving. Journal of Rrsonality and Social Psychology, 52, 1122-1131.

64

Byme and Neuman

Isen, A. M., Johnson, M. M. S.,.Mertz, E., & Robinson, G. F. (1985). The influence of positive affect on the unusualness of word associations. Journal of Personality and Social Psychology, 48, 1-14. Isen, A. M., Nygren, T. E., & Ashby, F. G. (1988). Influence of positive affect on the subjective utility of gains and losses: It is just not worth the risk. Journal of Personality and Social Psychology, 55, 710-717. Isen, A. M., Shalker, T., Clark, M., & Karp, L. (1978). Affect, accessibility of material in memory and behavior: A cognitive loop? Journal of Personality and Social Psychology, 36, 1-12. Janis, I. L., Kaye, D., & Kirschner, P. (1965). Facilitating effects of "eating while reading" on responsiveness to persuasive communications. Journal of Personality and Social Psychology, 11, 181-186. Johnson, D. W., Maruyama, G., Johnson, R., Nelson, D..,& Skon, I,. (1 98 1). Effects of cooperative, competitive, and individualistic goal structures on achievement: A meta-analysis. Psychological Bulletin, 89, 47-62. Kahneman, D., & Tversky, A. (1983). Choices, values, and frames. American Psychologist, 39, 34 1-350. Kalin, R., & Rayko, D. S . (1978). Discrimination in evaluative judgments against foreign-accented job candidates. Psychological Reports, 43, 1203-1209. Kamin, L. J. (1958). Ethnic and party affiliations of candidates as determinants of voting. Canadian Journal of Psychology, 12, 205-2 12. Knouse, S . B. (1983). The letter of recommendation: Specificity and favorability of information. Personnel Psychology, 36, 331341. Krames, L., Costanzo, D. J., & Carr, W. J. (1967). Responses of rats to odors from novel versus original sex partners. Proceedings, 75th Annual Convention of the APA, 117-118. Krivonos, P. D., Byrne, D., & Friedrich, G. W. (1976). The effect of attitude similarity on task performance. Journal of Applied Social Psychology, 6 , 307-313. Laird, J. D., Wagener, J. J., Halal, M . , & Szegda, M. (1982). Remembering what you feel: The effects of emotion on memory. Journal of Personality and Social Psychology, 42, 646-657. Landy, F. J. (1978). An opponent process theory of job satisfaction. Journal of Applied Psychology, 63, 533-547. Landy, F. J., & Fan, J. L. (1980). Performance rating. Psychological Bulletin, 87, 72-107.

Implications of Attraction Research

65

Larkin, J. C., & Pines, H. A. (1979). No fat persons need apply. Sociology of Mrk and Occupations, 6, 312-327. Lawler, E. E. (1969). Job design and employee motivation. Personnel psycho lo^, 22,426-435. Lawson, E. D., & Roeder, L. M. (1986). Women's full first names, short names, and affectionate names: A semantic differential analysis. Names, 34,175-184. Levinger, G., & Breedlove, J. (1966). Interpersonal attraction and agreement: A study of marriage partners. Journal of Personalit y and Social Psychology, 3, 367-372. Levinger, G., & Senn, D. J. (1967). Disclosure of feelings in marriage. Merrill-Palmer Quarterly, 13, 237-249, Locke, E. A. (1965). The relationship of task success to task liking and satisfaction. Journal of Applied Psychology, 48, 379-385. Locke, E. A. (1966). Relationship of task success to task liking: A replication. Psychological Reports, 18, 552-554. Locke, E. A. (1967). Relationship of success and expectation to affect on goal-seeking tasks. Journal of Personality and Social Psychology, 7,125-134. Lott, A. J.! & Lott, B. E. (1966). Group cohesiveness and individual learning. Journal of Educational Psychology, 57,61-73. Maher, J. R., & Overbagh, W. B. (1971). Better inspection performance through job enrichment. In J. R. Maher (Ed.), New perspectives in j o b enrichment. New York: Van Nostrand Reinhold. McCaskey, M. 3. (1979, November-December). The hidden messages managers send. Harvard Business Review, 57,135-158. McGovern, T. V., & Howard, E. A. (1978). Interviewer evaluations of interviewee nonverbal behavior. Journal of Vocational Behavior, 13, 163-171. McIntyre, S., Moberg, D. J., & Posner, B. Z. (1980). Preferential treatment in preselection decisions according to sex and race. Academy of Management Journal, 23, 738-749. Meadow, B. L. (1971). m e efects of attitude similarity-dissimilarity and enjoyment on the perception of time. Unpublished master's thesis, Purdue University, West Lafayette, IN. Mednick, S . A. (1962). The associative basis of creative process. Psychological Review, 69, 220-232. Mednick, M. T., Mednick, S. A., & Mednick, E. V. (1964). Incubation of creative performance and specific associative priming. Journal of Abnormal and Social Psychology, 69, 84-88. Mehrabian, A. (1967). Orientation behaviors and nonverbal attitude communication. Journal of Communication, 17, 324-332.

66

Byme and Neuman

Miller, A. G. (1970). Role of physical attractiveness in impression formation. Psychonomic Science, 19, 24 1-243. Mitchell, T. R.,.& Larson, Jr., J. R. (1987). People in organizations: An introduction to organizational behavior(3rd ed.). New York: McGraw-Hill. Mirvis, P. H., & Lawler, E. E. (1977). Measuring the financial impact of employee attitudes. Journal of Applied Psychology, 62, 1-8. MIT (1973). Ubrk in America: Report of a special task force to the secretary of health, education, and welfare. Cambridge, MA: The MIT Press. Mitchell, T., & Henning, J, (1987, March). Eflects ofpresent salary on resume evaluations: Sex discrimination? Paper presented at the Annual Meeting of the Southeastern Psychological Association, Atlanta, GA. Mobley, W. H., Griffeth, R. W., Hand, H. H., & Meglino, B. M. (1979). Review and conceptual analysis of the employee turnover process. Psychological Bulletin, 86, 493-522. Mortimer, J. T. (1976). Social class, work, and the family: Some implications of the father's occupation for familial relationships and sons' career decisions. Journal of Marriage and the Fmily, 38, 241-256. Muchinsky, P. M. (1977). Employee absenteeism: A review of the literature. Journal of Bcutionul Behavior, 10, 3 16-340. Muchinsky, P. M. (1979). The use of reference reports in personnel selection: A review and evaluation. Journal of Occupational Psychology, 52, 287-297. Muchinsky, P. M. (1990). Psychology applied to work: An introduction to industrial and organizational psychology (3rd ed.). Pacific Grove, CA: Brooks/Cole. Muchinsky, P. M., & Tuttle, M. L. (1979). Employee turnover: An empirical and methodological assessment. Journal of Mcationa1 Behavior, 14, 43-77. Murray, H. A. (1959). Vicissitudes of creativity. In H. H. Anderson (Ed.), Creativity and its cultivation. New York: Harper & Row. Nasby, W., & Yando, R. (1982). Selective encoding and retrieval of affectively valent information. Journal of Personality and Social Psychology, 43, 1244-1255. Nelson, D., & Meadow, B. L. (1971, September). Attitude similarity, interpersonal attraction, actual success and the evaluative perception of that success. Paper presented at the meeting of the American Psychological Association, Washington, D. C.

Implications of Attraction Research

67

Neuman, J. H. (1990). Realistic job previews: The role of decision framing, job offer attractiveness, and message elaboration in job offer acceptance (Doctoral dissertation, University at Albany, State University of New York, 1990). Dissertation Abstracts International, 51, 2099B. Newcomb, T. M. (1978). The acquaintance process: Looking mainly backward. Journal of Personality and Social Psychology, 36, 1075-1083. Newman, J., & Krzystofiak, F. (1979). Self-reports versus unobtrusive measures: Balancing method variance and ethical concerns in employment discrimination research. Journal of Applied psycho lo^, 64, 82-85. Orpen, C. (1984). Attitude similarity, attraction, and decisionmaking in the employment interview. Journal of Psychology, I1 7, 111-120. Parsons, P. C., & Liden, R. C. (1984). Interviewer perceptions of applicant qualifications: A multivariate field study of demographic characteristics and nonverbal cues. Journal of Applied Psychology, 69, 557-568. Paul, W. J., Robertson, K. B., & Herzberg, F. (1969). Job enrichment pays off. Harvard Business Review, 41, 61-78. Paunonen, S . V., Jackson, D. N., & Oberman, S. M. (1987). Personnel selection decisions: Effects of applicant personality and the letter of reference. Organizational Behavior and Human Decision Processes, 40, 96-1 14. Perrin, F. A . C. (1921). Physical attractiveness and repulsion. Journal of Experimental Psychology, 4 , 203-217. Peters, L. R., & Terborg, J. R. (1975). The effects of temporal placement of unfavorable information and of attitude similarity on personnel selection decisions. Organizational Behavior and Human Performance, 13, 279-293. Porter, L. W., & Lawler, E. E. 111 (1968). Managerial attitudes and performance. Homewood, IL: Irwin. Porter, L. W., & Steers, R. M. (1973). Organizational, work, and personal factors in employee turnover and absenteeism. Psychological Bulletin, 80, 151-176, Postman, L., & Brown, D. R. (1952). Perceptual consequences of success and failure. Journal of Abnormal and Social Psychology, 47, 213-221. Pulakos, L. H . , & Wexley, K. N. (1983). The relationship among perceptual similarity, sex, and performance ratings in manager-subordinate dyads. Academy of Management Journal, 26, 129-139.

Byrne and Neuman

68

Rand, T. M., & Wexley, K. M. (1975). Demonstration of the effect, "similar to me, in simulated employment interviews. Psychological Reports, 36, 535-544. Riess, M., Rosenfeld, P., Melburg, V. & Tedeschi, J. T. (1981). Self-serving attributions: Biased private perceptions and distorted public descriptions. Journal of Personality and Social Psychology, 41, 224-23 1. Rosenbaum, M. E. (1986a). The repulsion hypothesis: On the nondevelopment of relationships. Journal of Personality and Social Psychology, 51, 1 156-1 166. Rosenbaum, M. E. (1986b). Comment on a proposed two-stage theory of relationship formation: First repulsion; then, attraction. Journal of Personality and Social Psychology, 51, 11711172. Ross, L., Green, D., & House, P. (1977). The "false consensus effect": An egocentric bias in social perception and attribution process. Journal of Experimental Social Psychology, 13, 279301. Rothblum, E. D., Miller, C. T., & Garbutt, B. (1988). Stereotypes of obese female job applicants. International Jouml of Eating Disorders, 7, 277-283. Rusbult, C. E., & Farrell, D. (1983). A longitudinal test of the investment model: The impact on job satisfaction, job commitment, and turnover of variations in rewards, costs, alternatives, and investments. Journal of Applied Psychology, 68, 429-438. Rusbult, C. E., Farrell, D, Rogers, G. & Mainous, A. G. (1988). Impact of exchange variables on exit, voice, loyalty, and neglect: An integrative model of responses to declining job satisfaction. Academy of Management Journal, 31, 599-627. Rusbult, C. E., Johnson, D. J., & Morrow, G. D. (1986). Predicting satisfaction and commitment in adult romantic involvements: An assessment of the generalizability of the investment model. Social Psychology Quarterly, 49, 8 1-89. Rusbult, C. E., Zembrodt, I. M., & Gunn, (1982). Exit, voice, loyalty, and neglect: Responses to dissatisfaction in romantic involvements. Journal of Personality and Social Psychology, 43, 1230-1242. Sapolsky, A. (1960). Effects of interpersonal relationship upon verbal conditioning. Journal of Abnormal and Social Psychology, 60, 241-246. Schachter, S., Ellertson, N., McBride, D., & Gregory, D. (1951). An experimental study of cohesiveness and productivity. Human Relations, 4 , 229-238. I'

Implications of Attraction Research

69

Schmitt, N., & Hill, T. E. (1977). Sex and race composition of assessment center groups as a determinant of peer and assessor ratings. Journal of Applied Psychology, 62, 261-264. Segal, M. W. (1974). Alphabet and attraction: An unobtrusive measure of the effect of propinquity in a field setting. Journal of Personality and Social Psychology, 30, 654-657. Shaw, E. A. (1972). Differential impact of negative stress-typing in employee selection. Personnel Psychology, 25, 333-338. Simas, K., & McCarrey, M. (1979). Impact of recruiter authoritarianism and applicant sex on evaluation and selection decisions in a recruitment interview analogue study. Journal of Applied Psychology, 64, 483-49 1. Smeaton, G., Byrne, D., & Murnen, S. K. (1989). The repulsion hypothesis revisited: Similarity irrelevance or dissimilarity bias? Journal of Personality and Social Psychology, 56, 54-59. Stagner, R. (1981). Training and experiences of some distinguished industrial psychologists. American Psychologist, 36, 497-505. Stagner, R., & Rosen, H. (1965). Psychology of union-management relations. Belmont, CA: Wadsworth. Staw, B. M. (1980). Rationality and justification in organizational life. In B. Staw & L. Cummings (Eds.), Research in Organizational Behavior (Vol. 2, pp. 45-80). Greenwich, CT: JAI Press. Stogdill, R. M. (1972). Group productivity, drive, and cohesiveness. Organizational Behavior and Human Pei$ormance, 8, 26-43. Stogdill, R. M. (1974). Handbook of leadership. New York: Free Press. Taylor, M. J. (1956). Some objective criteria of social class membership. Unpublished manuscript. Teasdale, J. D., & Fogarty, S. J. (1979). Differential effects of induced mood on the recall of pleasant and unpleasant events from episodic memory. Journal of Abnormal Psychology, 88, 248-257. Terborg, J. R., Castore, C., & DeNinno, J. A. (1976). A longitudinal field investigation of the impact of group composition on group performance and cohesion. Journal of Personality and Social Psychology, 34, 782-790. Terpstra, D. E . , & Larsen, J. M. (1980). A note on job type and applicant race as determinants of hiring decisions. Journal of Occupational Psychology, 53, 117-119. Turban, D. B., & Jones, A. P. (1988). Supervisor- subordinate similarity: Types, effects, and mechanisms. Journal of Applied Psychology, 228-234.

70

Byrne and Neuman

Velten, E. (1967). l?~?induction of elation and depression through the reading of structured sets of mood-statements. Unpublished doctoral dissertation, University of Southern California. Venes, A. M., & Krupka, L. R., & Gerard, R. J. (1982). Overweightiobese patients: An overview. The Practitioner, 226, 1102-1109. Wanous, J. P. (1980). Organizational entry: Recruitment, selection, and socialization of newcomers. Reading, MA: AddisonWesley Publishing Company. Walster, E., Aronson, V., Abrahams, D., & Rottman, L. (1966). Importance of physical attractiveness in dating behavior. Journal of Personality and Social Psychology, 4, 508-516. Washburn, P. V.,, & Hakel, M. D. (1973). Visual cues and verbal content as influences on impressions formed after simulated employment interviews. Journal of Applied Psychology, 58, 137-141. Werts, C. E. (1966). Social class and initial career choice of college freshmen. Sociology of Education, 39, 74-85. Werts, C. E. (1968). Paternal influence on career choice. Journal of Counseling Psychology, 15, 48-52. West, M. A., & Farr, J. L. (1989). Innovation at work: Psychological perspectives. Social Behavior, 4, 15-30. Workie, A. (1974). The relative productivity of cooperation and competition. Journal of Social Psychology, 92, 225-230. Whyte, W. W. (1956). The organization man. New York: Simon and Schuster. Young, D. M., & Beier, E. G. (1977). The role of applicant nonverbal communication in the employment interview. Journal of Employment Counseling, 14, 154-165. Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology Monographs Supplement, 9,1-27. Zajonc, R . B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35, 151- 175. Zajonc, R. B. (1984). On the primacy of affect. American Psychologist, 39, 117-123. Zuckerman, M., Mann, R. W., & Bernieri, F. J. (1982). Determinates of consensus estimates: Attribution, salience, and representativeness. Journal of Personality and Social Psychology, 42, 839-852.

Issues, Theory, and Research in Industrial1 Organizational Psychology - K. Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

71

CHAPTER 3 TAKING THINGS A BIT TOO FAR: SOME PROBLEMS WITH EMERGENT INSTITUTIONAL THEORY

Richard H. Hall Institutional theory has rapidly emerged to a place of prominence within organizational theory (hereinafter OT). The purpose of this paper is to examine and critically evaluate the major characteristics of institutional theory (hereinafter IT). IT will first be located within OT. Then, a brief historical review will be presented. Some major problems of IT will then be considered. The paper will conclude with some very hopeful signs that IT can in fact make major contributions to OT.

IT AND OT IT was hardly a dot on the horizon a short time ago. In 1987, Bedeian reviewed OT for the industrial and organizational psychology audience. He focused on effectiveness studies, the organizational environment interface, organizational learning, and organizational decline. These were major topics of the time. There is no mention of IT at all, even though one of the major IT papers (DiMaggio & Powell, 1983) is mentioned in passing. Bedeian (1987) did make a most prescient observation, however, when he commented:

...the popularity of research themes has either grown or declined in direct proportion to their market-ability rather than the degree to which basic issues have been resolved and critical phenomena understood. One consequence of this tendency to focus on 'hot' topics is what some see as an extreme iconoclasm

Hall

72

resulting in an unhealthy fragmentation of the field.. ..instead of building on previous findings, a vogue seems to exist in some quarters that encourages a preoccupation with "fad" and "fashion. .this would seem to explain why a measure of OT research seems to be "literature driven" rather than problem driven. One possible consequence is that much organization research is non-cumulative and non-communicable across quarters. (p. 23) 'I..

IT is now hot. If Bedeian were to write his article in 1990-91 I am sure that IT would have a prominent position. IT has come rushing to the fore. This rush was aided by an American Sociological Association grant to Lynne Zucker. The grant helped sponsor a conference at UCLA in 1985 which resulted in a major book on IT (Zucker, 1988). IT places a lot of emphasis on symbolism. When the December 1987 issue of Administrative Science Quarterly arrived (in 1988), the lead article (Scott, 1987) was overview of IT. This was a clear symbol of IT's arrival as a theoretical perspective to be reckoned with. IT advocates would claim that their approach is in fact problem driven. At the same time, the rapid emergence of IT contains the danger that it could go in the direction of "fad" or "fashion." It would be easy to see some of Aldrich's (1988) dreaded "paradigm warriors" championing IT's cause. Fortunately, this danger appears to be being averted. The applications of IT have tended to be temperate. As a preliminary example of this temperateness, in an examination of the financial reporting practices of the Fortune 200 firms, Mezias (1990) reports that "institutional variables" explain more of the variance in such practices than do approaches from applied economics. This is not an argument that the economics factors were not operating, but rather that the understanding of their operation is enhanced by the use of IT. The application of theories in combination has been typical of the use of IT in OT. This aspect of IT will be developed more fully at a later point.

WHAT IS IT AND WHERE DID IT COME FROM? Scott (1987) identifies four strands of IT. The first strand is identified with the work of Phillip Selznick and his students and stems from Selznick's classic TVA and the Grass Roots (1949). The focus is on the wavs in which organizational practices come to be

Institutional Theory

73

instilled with values. The fourth strand is almost classically sociological and is concerned with "spheres of life," such as family, religion, and the state, and the value systems and actions developed between these spheres. Scott's second and third strands are the most crucial for contemporary OT and will be the focus here. For the sake of simplicity, these will be labeled (1) the DiMaggio and Powell strand and (2) the Meyer and Rowan and Zucker strand. At one time the labels might have been East Coast and West Coast respectively, but geographical mobility and the dispersion of second-generation IT advocates have blurred these regional identities. In the sections that follow, the key features of these forms of IT will be identified.

DiMaggio and Powell Paul DiMaggio and Walter Powell's strand of IT can perhaps best be appreciated in terms of the ways in which it seeks to explain why organizations take the forms that they do. Much of the research here has been carried out in not-for-profit organizations with rather indeterminant technologies. DiMaggio and Powell (1983) argue that "institutional isomorphism" is now the dominant reason why such organizations assume the forms that they have. According to DiMaggio and Powell, Weber's original (1952, 1968) analysis of the driving force behind the move toward rationalization and bureaucratization was based on a capitalist market economy, with bureaucratization an "iron cage" in which humanity was bound since the process of bureaucratization was irreversible. DiMaggio and Powell (1983) believe that major social changes have altered this situation to such a large extent that an alternative explanation is needed. Their analysis is based around the assumption that organizations exist in "fields" of other, similar organizations. They define an organizational field as follows:

By organizational field, we mean those organizations that, in the aggregate, constitute a recognized area of institutional life: key suppliers, resource and product consumers, regulatory agencies and other organizations that produce similar services and products. The virtue of this u n i t of analysis is that i t directs our attention not simply to competing firms, as does the population approach of Hannan and Freeman (1977), or to networks of organizations that actually interact, as does the interorganizational network approach of Laumann et al. (1978), but to the totality of relevant actors. (p. 148)

Hall

74

According to this perspective, organizations are increasingly homogeneous within fields. Thus, public universities acquire a sameness, as do department stores, airlines, professional football teams, motor vehicle bureaus, and so on. DiMaggio and Powell (1983) cite three reasons for this isomorphism among organizations in a field. First, there are coercive forces from the environment, such as government regulations and cultural expectations which can impose standardization on organizations. Government regulations, for example, force restaurants (hopefully) to maintain minimum health standards. As Meyer and Rowan (1977) have suggested, organizations take forms that are institutionalized and legitimated by the state. DiMaggio and Powell also note that organizations mimic or model each other. This occurs as organizations face uncertainty and look for answers to their uncertainty in the ways in which other organizations in their field have faced similar uncertainties. Rowan (1982) argues, for example, that public schools add and subtract administrative positions in order to come into isomorphism with prevailing norms, values, and technical lore in their institutional environment. DiMaggio and Powell argue that large organizations tend to use a rather small number of consulting firms which "like Johnny Appleseeds, spread a few organizational models throughout the land" (p. 152). A very concrete example, noted by DiMaggio and Powell is the fact that Japan consciously modelled its courts, postal system, military, banking, and art education programs after Western models in the late nineteenth century. As DiMaggio and Powell (1983) note: American corporations are now returning the compliment by implementing (their perceptions of) Japanese models to cope with thorny productivity and personnel problems in their own firms. The rapid proliferation of quality circles and quality-of-work issues in American firms is, at least in part, an attempt to model Japanese and European successes. @. 151) A third source of institutional isomorphism comes from normative pressures as the workforce, and especially management, becomes more professionalized. Both professional training and the growth and elaboration of professional networks within organizational fields leads to a situation in which the managerial personnel in organizations in the same field are barely indistinguishable from one another. As people participate in trade and professional associations, their ideas

Institutional Theory

75

tend to homogenize. The institutional perspective thus views organizational design not as a rational process based around goals, but rather one of both external and internal pressures which lead organizations in a field to resemble each other over time. In this perspective strategic choices or attempts at member control would be viewed as coming from the institutional order in which an organization is embedded. DiMaggio (1988) clearly identifies limitations on the scope of IT. There are five conditions in which IT is most appropriately used. First, "institutional theory bears on those aspects of organizational life that are so exteriorized and intersubjective that no actor is likely to question them" @. 6). Instances here would be the age-grading of schools or an already negotiated contract. The second appropriate condition is in certain kinds of fields that are highly institutionalized and have weak technical bases, or that play a trivial role in the allocation of resources. Examples of the first type here would be banks, but not computer software designers. The second type here attracts little political attention and would be exemplified by funeral parlors, but not long-distance trucking companies. Thirdly, IT is appropriate in situations that are already well institutionalized (this sounds tautological, as will be specified later). DiMaggio uses the well-documented evidence regarding the diffusion of innovations in which the technical and political explanations of the development of an innovation have less predictive power regarding diffusion, once the innovation has been accepted. The fourth condition is the explanation of long term change, particularly where interests are stable. IT thus is useful in situations that are not politically charged. Finally, IT is useful in explaining variations across nation-states. Thus Crozier's (1964) French bureaucracies are the result of characteristics of the French state and its history. The differences of French bureaucracies from other bureaucracies can thus be explained in institutional terms. In his 1988 paper, DiMaggio specifically urges that OT must bring the issues of agency and interest into situations in which IT is not appropriate. This argument for limitation of scope is most welcome, but not particularly heeded by the other contributors to Zucker's (1988) anthology.

Meyer and Rowan and Zucker The second major strand of IT is most clearly identified with

Hall

76

the work of John Meyer and Brian Rowan and Lynne Zucker. It is worth noting that this strand has a home institution, Stanford University, as does the DiMaggio and Powell version at Yale University. As noted earlier, these genealogical distinctions have become blurred as geographical mobility has taken place and new generations of scholars grapple with IT issues. In summarizing this strand of IT, I will borrow heavily from Scott's (1987) discussion of it. Scott, also a Stanford scholar, is himself closely identified with IT (Meyer & Scott, 1983). Scott (1987) notes:

The argument is that social order is based fundamentally on a shared social reality which, in turn, is a human construction, being created in social interaction. It is recognized that man or woman as a biological organism confronts few limits or constraints in the form of instinctaul patterns, yet constraints develop in the form of a social order. Berger and Luckman (1967) argued that this order "is a human product, or, more precisely, an ongoing human production. It is produced by man in the course of.. .. on-going externalization.. ..Social order exists only as a product of human activity" (p. 52). Social order comes into being as individuals take action, interpret that action, and share with others their interpretation. These interpretations of "typifications"are attempts to classify.. ..behavior into categories that will enable.. .. actors to respond to it in a similar fashion. The process by which actions become repeated over time and are assigned similar meanings by self and others is defined as institu tionalization. "Institutionalization occurs whenever there is a reciprocal typification of habitualized actions by types of actors" (Berger & Luckman, 1967, p. 54). Both actions and actors are typed; actions come to be associated with certain classes of actors; e.g., supervisors give orders, workers follow them. (p. 495) Scott goes on to note that this strand of IT places a great deal of emphasis on the shared history of organizations. Shared history, of course, is an ongoing process so that reality is constantly being produced. This type of formulation is prominent in the early (1977) works of both Zucker and Meyer and Rowan. For example, Zucker (1977)

Institutional Theory

77

notes that institutionalized acts are both objective and exterior. They are objective in the sense that they are interpreted as having a reality separate from the actor involved. They are exterior in that they are in fact acts. Meyer and Rowan's (1977) approach is slightly different. They focus on the ways in which "social processes, obligations, or actualities come to take on a rulelike status in social thought and action" @. 341). This is the core premise around which Meyer and Rowan's well known ideas of "rationalized institutional elements" and "rational myths" are based. According to Scott (1987): The common feature in all of these definitions is that institutionalization is viewed as the social process by which individuals come to accept a shared definition of social reality -- a conception whose validity is seen as independent of the actor's own views or actions but is taken for granted as defining the "way things are" and/or "way things are to be done." @. 496) One application of this approach was Tolbert and Zucker's (1983) study of the diffusion of civil service reforms in municipal governments from 1880-1935. Early adopters were found to be acting in rational self interest, since city characteristics predicted adoption. Later adopters were characterized as responding to the established legitimacy of such practices. There are two important aspects of the Tolbert and Zucker research. First, they used the IT explanation in conjunction with another explanation--they did not test explanations against each other. The second aspect is that this research comes awfully close to a conceptualization that is like that of t h e DiMaggio and Powell strand in the sense that the organizational forms are based on mimicry. Despite this apparent convergence of the two strands of IT, Zucker (1988) maintains that there are important differences. The major difference is the degree of stability of institutionalized forms. According to Zucker, the DiMaggio and Powell version suggests that institutionalized forms are stable, whereas her approach, with its emphasis on ongoing institutionalization, has stability as its major problem. Also, DiMaggio and Powell examine organizations within institutional fields, while Zucker attempts to incorporate the entire social system into her formulation.

Hall

78

Combining the Strands Zucker has correctly identified differences between the major strands of IT. At the same time there is a strong conceptual core to IT. Scott (1987) captures this when he notes: The institutional features of environments are receiving increasing attention, in ways I have tried to document, as important determinants of the structure and functioning of organizations. Until the introduction of the institutional conceptions, organizations were viewed primarily as production systems and/or exchange systems, and their structures were viewed as being shaped largely by their technologies, their transactions, or the power-dependency relations growing out of such interdependencies. Environments were conceived of as task environments: as stocks of resources, sources of information, or loci of competitors and exchange partners. While such views are not wrong, they are clearly incomplete. Institutional theorists have directed attention to the importance of symbolic aspects of organizations and their environments. They reflect and advance a growing awareness that no organization is just a technical system and that many organizations are not primarily technical systems. All social systems--hence, all organizations--exist in an institutional environment that defines and delimits social reality. And just as with technical environments, institutional environments are multiple, enormously diverse, and variable over time. To

neglect their presence and power is to ignore signiflcant causal factors shaping organizational structures and practices: to overlook these variables is to misspecifi our causal models. (italics added; pp. 507508)

Scott goes on to conclude that institutional arguments ought not to be formulated in opposition to rational or efficiency arguments but that they are better seen as complementing and contextualizing them. Fortunately, this appears to be taking place in a number of pieces of research, as will be documented later in this paper. First, though, some problems with IT will be considered.

79

Institutional Theory

SOME PROBLEMS WITH IT The first problem is potential tautological reasoning. This form of reasoning was a major contributor to the demise of functional theory within sociology. A tautology is circular reasoning in which variables are defined in terms of each other, thus making causes and effects obscure and difficult to assess" (Turner & Maryanski, 1979, p. 124; see also Turner, 1979). This problem appears to creep into DiMaggio's (1988) analysis when he notes: 'I

Put simply, the argument of this section is that institutionalization is a product of the political efforts of actors to accomplish their ends and that the success of an institutionalization project and the form that the resulting institution takes depend on the relative power of actors who support, oppose, or otherwise strive to influence it. I refer to the politics of institutionalization as structural because they follow an internal logic of contradiction, such that the success of an institutionalization process creates new sets of legitimated actors who, in the course of pursuing distinct interests, tend to delegitimate and deinstitutionalize aspects of the organizational form to which they owe their own autonomy and legitimacy. Central to this line of argument is an apparent paradox rooted in the two senses in which the term institutionalization is used. Institutionalization as an outcurne places organizational structures and practices beyond the reach of interest and politics. By contrast, institutionalization as aprocess is profoundly political and reflects the relative power of organized interests and the actors who mobilize around them. (italics in original; P- 18) If this is not tautological reasoning, it is uncomfortably close to it. This quotation also contains the seeds of a problem which will be noted later--the tendency to bring all organizational phenomena under the institutional label. This problem also plagued the functionalists. The second problem is that institutional theory has paid almost no attention to what is institutionalized and what is not. This can be indirectly seen in the weirdly ironic case of Talcott Parsons. Parsons wrote fairly extensively about organizations and even specified the institutional level of analysis (Parsons, 1960). This is not even

Hall

80

mentioned in Scott's (1987), DiMaggio's (1988) or Zucker's (1988) works. This is not a call to go back to Parsons, but rather empirical evidence that not everything that says institutional is institutionalized. DiMaggio and Powell (forthcoming) have clearly recognized the Parsonian theoretical legacy and its contribution to contemporary IT, but the major earlier interpretation did not. This is a critical problem. There is a tendency to apply institutional theory in an ex post facto manner. This can be done in an almost mystical manner. Ideas and practices come and go for no reason other than institutionalization. It would appear that in reality, some performance criteria are applied in assessing the success of a practice. DiMaggio (1988) has made a move in the right direction here, with his emphasis on the importance in interest and agent in IT, Nonetheless, the basic question remains unanswered--why are some practices institutionalized and others not? I will later argue that the adoption of structures or practices is much more than institutionalized whim. The third problem is essentially ontological. The prime interest rate and changes in it are very real to borrowing and lending organizations; the number of 18-year-olds in a given year is very real to college administrators; and the number of 21-year-olds is very real for brewers. I am in the very preliminary stages of analyzing data from over 500 small businesses. When the owners of these businesses were asked what was the major disadvantage in running their own businesses, the most common response was: "the buck stops here. These men and women saw their problems as very real and not as legends or myths. Institutional theory can be very useful at this point. Individual and collective organizational myths and legends do develop about the meanings of these realities. This point has been very well demonstrated. The danger, however, is in making the reality that was the source of the myth or legend into the myth itself. Perrow (1985) made this point in a slightly different manner in his review of Meyer and Scott's (1983) book. Perrow suggests that IT can go "overboard with myths and symbols." Perrow (1985) notes: 'I

What happens .......is that the reaction of some community groups and the federal government to the discrepancy between goals and actual behavior in the schools is treated with disdain, as yet another meaningless bit of symbolic behavior. In actual fact, what seems to have happened is that the federal government was pushed by civil rights groups and parts of the

Institutional Theory

federal bureaucracy to pursue quality education and especially social equity in schools. The federal government mandated more and more programs to raise standards and, more specifically, to help the poor, minorities, and the handicapped. It sent bundles of money along with the mandates. The failure of the schools to help the disadvantaged was exposed, and reforms were encouraged. The schools took the money--it was not much, for only 10% of their budgets come from federal monies--but complained bitterly about the intrusion and the paper work. They hoped to continue with mere sorting and certification. But neither Meyer nor Scott is willing to see group interests and the misuse of resources as the issue; instead, a cultural interpretation in terms of empty symbols is advanced. Meyer and Scott assert that myths explain the growth of rules and other intrusions; that in itself may be a myth, if the rules are attempts to control subversions of the mandates and funds. It is quite possible that the schools simply lied about the way they spent the earmarked funds, and the federal government responded with more restrictions. The complexity of regulation of hospitals arises not because of a rational myth but because hospitals keep trying to find a way to get around a regulation; this only means that another regulation has to be issued to close off that loophole. The nursing home industry, almost totally financed by the government, is buried under forms because it is so corrupt that forms at least provide grounds for legal actions. Scott argues that nursing homes fit the model of institutionalized organizations; the nursing home "turns its back on its technical core in order to concentrate on conformity to its institutional environment'' (p. 125). That is not what dozens of investigations show. Actually the industry hides its corruption as it pretends to conform; there is a large difference in interpretation here. This infatuation with cultural myths and symbols to the neglect of power and group interest leads to some distressing examples of blaming the victims and of cynicism about the motives of reformers, especially in the Meyer chapters. (pp. 153-154)

81

82

Hall

Perrow's criticism of Meyer in particular should not be applied to all institutional theorists. DiMaggio and Powell (1983) were careful to hypothesize that institutional.effects were more likely in situations of indeterminant technology and ambiguous goals. DiMaggio's most recent (1988) paper pays direct attention to the political process. Nonetheless, the distressing tendency to try to turn everything into myth and legend does exist. This tendency is enhanced by the fact that Meyer's influential writings have themselves become institutionalized. The fourth problem with IT is seemingly similar to the one just discussed--seeing everything as myth or legend. This particular problem involves the application of IT itself to situations in which the supporting evidence is weak or non-existent. The problem is that IT can be given as the explanation for all unexplained variance. This situation exists when other theoretical explanations have been tested and shown to explain the usual limited amount of variance. Researchers then can say that the remainder of the variance is due to institutional factors. They also can simply assert that institutiondization has taken place. This problem can be illustrated by Boeker's (1989) research on firms in the semiconductor industry. Boeker examined the power of research and development, sales and marketing, and manufacturing and production subunits over time in a set of Silicon Valley firms. Boeker found that environmental and technological conditions at the time of founding and characteristics of the entrepreneur (such as occupational specialty) were critical in establishing the "functional importance" of the subunits. He then argues that inertial and institutional constraints operate to maintain original subunit power even in the face of environmental and technological shifts and thus the ..."current importance of functions in the organization can be viewed in part as a product of earlier-established patterns of functional importance that have become institutionalized" (Boeker, 1989, p. 407). Unfortunately, there is not a shred of evidence in regard to institutionalization. It is simply assumed to have taken place. The more parsimonious explanation would be that powerful subunits maintained their power because they retained possession of key resources. This explanation was not considered. We cannot be sure if Boeker's institutional explanation is right or wrong or somewhere in between. Finally, IT has suffered from inconsistent operationalization of some major concepts. For example, the key idea offield has suffered from being operationalized in a variety of ways. Galaskiewicz and Wasserman (1989) consider interorganizational networks as the equivalent of a field, but this is not how DiMaggio and Powell (1983) specified the term. Other analysts have implied that indus-

Institutional neory

83

tries or populations are the equivalent of fields. Inconsistent operationalization will slow the development of IT. When viewed as a totality, these problems suggest that IT has the potential to go the route of psychoanalytic theory in that everything can be explained, but nothing is explained. Fortunately, there are strong signs of hope.

SOME SIGNS OF HOPE Boeker's (1989) research, although flawed by the absence of data to support the institutional claims, contains an approach that appears to be crucial for the maturation and growth of organizational theory. Theoretical explanations are used in combination and theories are not tested against one another. Boeker considers environmental, technological, and strategic contingencies, along with ecological considerations and the institutional explanation. In the paragraphs that follow I will provide a sampling of this approach to theory development. Lmgley (1989) examined the manner in which formal analysis is used in organizations. In addition to the rational purposes served, such as information, communication, and direction and control, such analysis was also used for symbolic purposes. More importantly for the purposes here, Langley concludes that these processes .must be viewed as being closely intertwined rather than as mutually incompatible" @. 626). Hamilton and Biggart (1988) examined the growth and structure of organizations in the Far East. They found that market and cultural factors were the best explanations of organizational growth and that authority patterns and legitimation (Institutional) strategies best explained organizational structure. Further support for the use of multiple theories can be found in the work of Mezias (1990) which was noted earlier, Pfeffer and Davis-Blake (1987), and Singh, Tucker, and House (1986). Even the staunchest advocates of population ecological approaches, Michael Hannan and John Freeman (Freeman & Hannan, 1989; Hannan & Freeman, 1989) are bringing institutional explanations into their work. An example of this form of theory construction can be found in the work of Galaskiewicz and Wasserman (1989). They examined patterns of charitable giving among a set of organizations in the Minneapolis-St. Paul metropolitan area. They found that mimetic processes operated in decision-making situations characterized by uncertainty. They suggest: 'I..

Hall

84

.....we believe that our findings add to the literature on strategic decision making in several ways. First, our results strongly suggest that so-called institutional processes are critical in explaining organizational behavior, as suggested by DiMaggio and Powell (1983). Our research suggests that when faced with uncertainty, decision makers will mimic the behavior of other actors in their environment. If clear criteria do not exist, decision makers will try what others have done and have found to work. Second, social networks are important in determining which actors decision makers will imitate. There are several possible options that decision makers can pursue under conditions of uncertainty; there are several models they can adapt. We have argued and shown that decision makers will mimic the behavior of those in their networks, those whom they know and trust. ('p. 472-473) By implication, Galaskiewicz and Wasserman are saying that under more certain conditions, factors other than institutional ones will be operative. It is this type of reasoning that appears to be the hope for OT. In 1985 Fligstein concluded:

...each school of thought has tended to view its theory as a total causal explanation of organizational phenomena. This suggests that one of the central tasks in organizational theory is to reorient the field in such a way as to view competing theories as contributing to an understanding of organizational phenomena. (p. 377) If recent major publications are any indication, Fligstein's plea is being answered. OT is developing as IT is being incorporated along with other theoretical perspectives. IT, when appropriately employed, has become a powerful tool for the organizational theorist. The danger is that IT can be taken too far, as some of its advocates have done in the past.

NOTE 1 . I am grateful for Julio Sanchez' comments on an earlier draft of this paper.

Institutional Theory

85

REFERENCES Aldrich, H. E. (1988). Paradigm warriors: Donaldson versus the critics of organization theory. Organization Studies, 9, 18-25. Bedian, A. G. (1987) Organizational theory: Current controversies, issues, and directions. In G. L. Cooper and I. T. Roberts (Eds.), International review of industrial and organizational psychology (pp. 1-33). New York: John Wiley and Sons. Berger, P. L., & Luckman, T. (1967). The social construction of reality. New York: Doubleday. Boeker, W. (1989). The development and institutionalization of subunit power in in organizations. Administrative Science Quarterly, 34, 388-410. Crozier, M. (1964). The bureaucratic phenomenon. Chicago: University of Chicago Press. DiMaggio, P. J. (1988). Interest and agency in institutional theory. In L.G. Zucker (Ed.), Institutional patterns and organizations: Culture and environment. Cambridge, MA: Ballinger. DiMaggio, P. J., & Powell, W. W. (1983). The Iron Cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48, 147160. DiMaggio. P. J., & Powell, W. W. (forthcoming) The new institutionalism in organizational analysis. Chicago: University of Chicago Press. Fligstein, N. (1985). The spread of the multidivisional form among large firms. American Sociological Review, 50, 377-39 1. Freeman, J., & Hannan, M. T. (1989). Setting the record straight on organizational ecology: Rebuttal to young. American Journal of Sociology, 95, 425-438. Galaskiewicz, J., & Wasserman, S. (1989). Mimetic processes within an interorganizational field: An empirical test. Administrative Science Quarterly, 34, 454-479. Hamilton, G. G., & Biggart, N. W. (1988). Market, culture, and authority: A comparative analysis of management and organization in the Far East. American Journal of Sociology, 94, S52-S94. Hannan, M. T., & Freeman, J. P. (1977). The population ecology of organizations. American Journal of Sociology, 82, 929-64. Hannan, M. T., & Freeman, J. P. (1989). Organizational ecology. Cambridge, MA: Harvard University Press. Langley, A. (1989). In search of rationality: The purposes behind the use of formal analysis i n organizations. Administrative Science Quarterly, 34, 598-631.

86

Hall

Laumann, E. O., Galaskiewicz, J., & Marsden, P. (1978). Community structure as interorganizational linkage. Annual Review of Sociology, 4, 455-484. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83, 340-363. Meyer, J. W., & Scott, W. R. (1983). Organizational environments: Ritual and rationality. Beverly Hills, CA: Sage. Mezias, S. J. (1990). An institutional model of organizational practice: Financial reporting at the Fortune 200. Administrative Science Quarterly, 35,431-457. Parsons, T. (1960). Structure and process in modern society. New York: The Free Press. Perrow, C. (1985). Review essay: Overboard with myth and symbols. American Journal of Sociology, 91, 151-155. Pfeffer, J., & Davis-Blaker, A. (1987). The effect of the proportion of women on salaries: The case of college administrators. Administrative Science Quarterly, 32, 1-24. Rowan, B. (1982). Organizational structure and the institutional environment: The case of public schools. Administrative Science Quarterly, 27, 259-279. Scott, W. R. (1987). The adolescence of institutional theory. Administrative Science Quarterly, 32, 493-5 11. Selznick, P. (1949). TVA and the grass roots. Berkeley, CA: University of California Press. Singh, J. V., Tucker, D. J., & House, R. J. (1986). Organizational legitimacy and the liability of newness. Administrative Science Quarterly, 31, 44-58. Tolbert, P. S., & Zucker, L. G. (1983). Institutional sources of change in the formal structure of organizations: The diffusion of Civil Service reforms, 1880-1935. Administrative Science Quarterly, 28, 22-39. Turner, J. H. (1979). The structure of social theory, Revised Edition. Homewood, IL: Dorsey. Turner, J. H., & Maryanski, A. (1979). Functionalism. Menlo Park,CA: BenjaminKumming. Weber, M. (1952). The Protestant Ethic and the spirit of capitalism. New York: Scribner. Weber, M. (1968). Economy and society: An outline of interpretive sociology(Vo1. 1-3). New York: Bedminster. Zucker, L. G. (1977). The role of institutionalization in cultural persistence. American Sociological Review, 42, 726-743.

Institutional Theory

87

Zucker, L. G. (1988). Where do institutional patterns come from? Organizations as actors in social systems. In L. G. Zucker (Ed.), Institutional patterns and organizations: Culture and environment. Cambridge, MA: Ballinger.

This Page Intentionally Left Blank

Part 11: TRADITIONAL AREAS OF RESEARCH IN ORGANIZATIONS

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial1 Organizational Psychology - K. Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

91

CHAPTER 4 ANOTHER LOOK AT PERSONALITY AND MANAGERIAL PmENTIAL: APPLICATION OF THE FIVE-FACTOR MODEL

John R. Lillibridge and &vin J. Williams Managerial talent is a critical organizational resource. Although managerial leadership has long been a subject of investigation in organizational psychology, interest in identifying and developing managerial talent is increasing as the nature of both work and the workplace continue to undergo dramatic changes (Cleveland & Thorton, 1990; Howard & Bray, 1988; Offermann & Gowing, 1990; Thorton & Cleveland, 1990). Compared to their contemporaries, managers of the future may need a broader range of skills to function effectively in rapidly changing work environments. An increase in the skills and behavioral repertoire required for success as a manager will make the already formidable task of predicting managerial effectiveness even more difficult. Accordingly, new conceptual models of managerial competence may be needed to improve prediction in this area (Williams & Lillibridge, 1990). The goal of this chapter is to expand recent models of managerial competence to include personality or dispositional variables that are related to critical managerial skills. In the first half of the chapter, we outline a proactive model of managerial competence and suggest a role for the "Big Five" personality domains (Goldberg, 1990) in the prediction of managerial talent. In the second half of the chapter, we present evidence that these personality domains are related to early behavioral assessment of managerial competence. Predicting managerial competence presents a number of unique concerns for industrial/organizational psychologists. First, operationalizing managerial competence is difficult. Different measures, including level in hierarchy, rate of promotion, department profitability, subordinate satisfaction, supervisor ratings, and turnover rates,

92

Lillibridge and Williams

have been used as indicators of effectiveness, but no clear consensus regarding the best or most appropriate measure exists (Luthans, Hodgetts, & Rosenkrantz, 1988). Second, the time span between predictor assessment and criterion evaluation is quite large. Effective managers typically rise through the ranks of an organization and traverse many organizational boundaries (Howard & Bray, 1988). Thus, predicting managerial success often involves trying to forecast competence many years into the future. In order to predict accurately managerial competence one needs to identify temperament, skills, and abilities that are not only related to critical managerial behaviors but also facilitate boundary passages or boost performance in several kinds of positions. Recently, we presented a model of managerial potential which attempts to identify such characteristics (Williams & Lillibridge, 1990).

THE WILLIAMS AND LILLIBRIDGE MODEL There are two major features of the Williams and Lillibridge (1990) model. First, we argued for a proactive perspective of managerial effectiveness. According to this perspective, research should attempt to identify individual differences that explain (1) why individuals select or are selected into different environments, (2) how certain individuals are able, often unintentionally, to elicit certain responses from others, and (3) how individuals intentionally create or alter certain environments to fit their interests or capabilities. These three processes are referred to as selection, evocation, and manipulation, respectively (Buss, 1987). These processes may be responsible for the realization of managerial talent: Managerial positions may be sought based on one's motives or goals; certain personal characteristics may attract the attention of significant others (e.g., mentors); and, finally, effective managers may openly act on part of their environment to progress toward their goals (Williams & Lillibridge, 1990, pp. 71-72). Researchers in the managerial leadership area have been interested primarily in selection processes. However, evocation and manipulation processes may be more important in determining long term success. Clawson, Kotter, Faux, and McArthur (1985), for example, found that successful managers were ones that took responsibility for their own development by actively seeking mentors and new assignments. It needs to be stressed that manipulation processes

Five Factor Model

93

are not defined in negative or malevolent terms, but rather as attempts to influence or change one's environment, especially the attitudes and behaviors of others (Buss, 1987). Defined in these terms, the ability to manipulate one's environment is crucial to managerial success. The manipulation of others is especially important, even necessary, for obtaining resources, establishing alliances, gaining promotions, and working effectively with others. The second feature of the Williams and Lillibridge (1990) model is that it identifies specific behavioral functions that help individuals respond to and shape their environments. According to Ford (1987), attempts to influence one's environments are dependent upon five behavior functions: directive, regulatory, control, arousal, and transactional. Descriptions of these functions are outlined in n b l e 1. Effective managers are likely to have skills or abilities in each of these domains. That is, they should have motives or aspirations to lead or manage (directive functions), the ability to monitor progress toward one's goals (regulatory functions), complex problem solving and planning skills (control functions), the ability to selectively vary the rate and intensity of relevant behaviors (arousal functions), and the ability to collect and disseminate information (transactional functions). Williams and Lillibridge (1990) reviewed the empirical literature on managerial effectiveness and grouped predictors by behavior functions. Examples of predictors known to influence managerial success and predictors that may potentially predict managerial success are listed in n b l e 1. A large number of predictors were found that served directive, control, and transactional functions. Few predictors, however, were uncovered for arousal and regulatory functions. Two conclusions can be drawn from our earlier review. First, future studies need to incorporate as predictors of managerial talent individual difference variables that facilitate boundary passage and advancement in organizations. Second, particular attention should be paid to identifying predictors of arousal and regulatory processes. In order to identify such predictors, researchers may have to explore new predictor domains, or at least adopt different perspectives on familiar predictor domains. One area likely to provide information on causes of variation in behavior intensity and regulation across settings and time is research in personality. Personality dimensions have been identified as causes of long-term behavioral outcomes (Caspi, 1987) and thus may be relevant for predicting eventual managerial success. Personality variables may also be related to harbingers of successful managerial careers. Howard and Bray (1988), for example, found evidence that the level of one's interpersonal skills at the beginning of one's career

Lillibridge and Williams

94

nble 1 Ford's (19871 behavior functions. Behavior Function

Role

Known or Potential Predictors

Directive

Specify intended states of individuals.

Motivation to manage (K) Need for Power (K) Values (K)

Regulatory

Specify criteria among options and evaluate progression toward intended states.

Self-efficacy (P) Self-monitoring (P)

Control

Cognitive processes (e.g., planning, reasoning) that combine current perceptual information with one's knowledge base and behavioral repertoire to guide behavior.

Verbal Ability (K) Intelligence (K) n c i t Knowledge (K)

Arousal

Vary the amount, rate, and intensity of behavior.

Vigor/Surgency (P) Empathy (P)

Allows for information flow and exchange.

Communication skills (K) Interpersonal skills (K)

Transactional

Note: Adapted from Williams and Lillibridge (1990). K = known or established predictor; P = potential or possible predictor.

Five Factor Model

95

was predictive of success in management 20 years later. Personality traits such as agreeableness, empathy, and extraversion are likely to predict interpersonal skills and thus may be used as predictors of managerial competence. From a proactive perspective, individual differences in temperament and disposition are also linked to selection, evocation, and manipulation processes. Individuals select social environments that are congruent with their personalities (Emmons, Diener, & Larson, 1986), and select mates with personalities similar to theirs (Buss, 1987). Certain personality types are likely to elicit specific perceptions, reactions, and behavior from observers (Buss, 1987). Shy individuals, for example, are less likely than outgoing or sociable individuals to elicit social invitations from others. Manipulation processes, perhaps most crucial to managerial success, are also linked to personality factors, in that personality influences the tactics used to manipulate one's environment. Buss, Gomes, Higgins, and Lauterbach (1987), for example, found that individuals high on agreeableness were less likely to use coercion as a manipulation tactic.

PERSONALITY AND MANAGERIAL COMPETENCE Within the managerial leadership area, there is a long research tradition examining the relationship between personality traits and managerial effectiveness (Stogdill, 1974; Yukl, 1989). Much of the this research, however, lacks a conceptual or integrative framework. Researchers have concentrated on specific traits related to managerial effectiveness, with little regard for underlying personality structure. Two exceptions to this trend are research on need for power (e.g., McClelland & Burnham, 1976; Miner, 1978) and Howard and Bray's (1988) assessment center research. Empirical tests of Miner's (1978) role-motivation theory and McClelland's leadership motive pattern (McClelland & Boyatzis, 1982; McClelland & Burnham, 1976) demonstrate that individual differences in the motivation to manage, as reflected by high strivings for power and dominance coupled with self-control, are positively related to managerial effectiveness. Although this research has a strong theoretical focus, one limitation is that it adopts a narrow view of personality structure. Howard and Bray's (1988) research represents a more comprehensive attempt to relate personality differences to managerial success. Managers attending an assessment center early in their careers were measured on 37 different personality and motivation characteristics. These measures were used to predict level of

Lillibridge and Williams

96

progress in the company 20 years later. Factor analyses were conducted on the 37 measures in an attempt to identify an underlying personality structure. Six factors emerged from this analysis: SelfEsteem, Leadership Motivation, Positiveness, Impulsivity, Affability, and Ambition. Impulsivity and Ambition showed the strongest relationships with managerial progress, with correlations ranging from .25 to .37. A major contribution of Howard and Bray's (1988) research is the identification of broad personality domains related to managerial success. One limitation, however, is that their factor structure does not correspond to a coherent trait structure that has emerged in recent personality research, i.e., the "Big Five" robust domains of personality.

ROBUST DIMENSIONS OF PERSONALITY THE BIG FIVE Personality researchers have established a five-factor model of trait ratings which has substantial convergent and discriminant validity (Conley, 1985; Digman & Inouye, 1986; Digman & "hkemotoChock, 1981; Goldberg, 1990; McCrae, Costa, & Busch, 1986). These dimensions are usually labeled Surgency (or Extraversion), Agreeableness, Conscientiousness (Impulse Control), Neuroticism vs. Emotional Stability, and Intellectual Interests. These non-overlapping trait domains show significant relationships to relevant behavior patterns, both concurrently and longitudinally, and have been found in studies utilizing self-report inventories (McCrae et al., 1986), ratings by peers (Digman & Takemoto-Chock, 1981), and ratings by trained observers (Digman & Inouye, 1986). Defining terms for the five factors are presented in Bble 2. The core of the surgency domain reflects the enjoyment of others' company, termed sociability (McCrae & Costa, 1987), coupled with high activity levels and dominance (Goldberg, 1990). Agreeableness is defined as trustful and sympathetic relationships with others and is the opposite of antagonism. The conscientiousness domain reflects the extent to which individuals are well organized, directed, careful, and self-disciplined (McCrae & Costa, 1987). Neuroticism reflects the extent to which people experience negative affect and emotional distress; the opposite pole of this domain is emotional stability. There is some debate about the meaning of Intellect, the fifth factor. "Culture," "Openness to Experience," and "Intellect" have all been used to describe this factor, with Digman and Inouye (1986) suggesting "Inquiring Intellect. This fifth factor contains both intelligence I'

Five Factor Model

97

nble 2 The Big Five Personality Domains Domain

Defining Terms

Surgency

enthusiastic, spirited, extroverted, expressive, playful, gregarious, sociable, active, energetic, dominant, assertive, ambitious, courageous

Agreeableness

cooperative, helpful, amiable, cordial, friendly, empathic, understanding, considerate, courteous, generous, affectionate, easygoing, honest

Conscientious- organized, concise, efficient, self-disciplined, ness precise, cautions, punctual, deliberate, decisive, predictable, economical, logical Neuroticism

Intellect

defensive, fretful, insecure, emotionally instable, temperamental, excitable, envious, nervous, anxious, fearful, gullible, intrusive Contemplative, intellectual, insightful, complex, perceptive, bright, smart, curious, inquisitive, creative, innovative, sophisticated

Note: Terms gathered from synonym clusters provided by Goldberg (1990). The negative pole of Factor 4 (Neuroticism) has been used instead of the positive pole (Emotional Stability) because most measures of this factor assess the negative pole.

98

Lillibridge and Williams

notions as well as broad interests and openness to cultural experiences. Applying the five-factor model to research on managerial potential may aid researchers in three ways. First, it provides a meaningful and efficient way of organizing results. Grouping specific traits related to managerial effectiveness by personality domain would increase the understanding of why specific traits predict, as well as how different traits might relate to one another. For example, eleven of the thirteen traits identified by Yukl (1989) as characteristic of successful leaders can be placed in one of the robust domains in 'lhble 2. Five fall within the surgency domain: ambitious and achievement-oriented, assertive, dominant, energetic, and self-confident. N o fall within the agreeableness domain: adaptable to situations and cooperative. Four fall within the conscientiousness domain: decisive, dependable, persistent, and willing to assume responsibility. Only "tolerant of stress" and "alert to social environment" are not clearly identified with one of the domains, although perhaps they can be seen as part of the neuroticism and intellect domains, respectively. By categorizing traits in this manner it can be seen that the surgency and conscientiousness domains have been most strongly linked to managerial effectiveness. The second advantage of the five-factor model is that it provides the opportunity to link behavior and personality taxonomies. YuM (1989), for example, has grouped eleven critical managerial behaviors into four categories: building relationships (i.e, . networking, supporting, managing conflict and team building), influencing people (i.e., motivating, and recognizing and rewarding), making decisions (problem solving, planning and organizing, consulting and delegating), and giving-seeking information (informing, clarifying, monitoring). The relations between these broad behavior domains and the robust personality domains should be investigated. For example, managers high on traits in the agreeableness domain may be effective at building relationships, while those high on traits in the conscientiousness domain may be good decision makers. The third advantage of the five-factor model is that it can guide new research aimed at identifying critical processes involved in boundary passage and hierarchical progression in organizations. Specifically, it may help guide research on evocation and manipulation processes (Buss et al., 1987). One potentially interesting area of research would be to determine whether others' perceptions of a manager's standing on the robust personality domains influences interactions with the manager in such a way to facilitate or hinder early career advancement. Perceptions of specific leader traits (e.g., intelligence, dominance) influences leadership emergence and ability

Five Factor Model

99

of leaders to influence others (Lord, DeVader, & Alliger, 1986). Research should examine whether similar effects occur for perceptions of the five robust domains, and how such perceptions compare to perceptions of specific traits. The previous discussion suggests a number of possibilities for future research, not all of which can be addressed in a single study. As a first step, we will apply the five-factor model to the assessment of managerial potential. As such, this study represents an initial attempt at measuring the five-factor model in a managerial assessment setting.

EXPLORING THE RELATION BETWEEN PERSONALITY AND MANAGERIAL TALENT: A PRELIMINARY INVESTIGATION The relation between one's standing on the five robust personality domains and managerial talent was investigated using data obtained from a managerial identification and development program. The assessment of managerial capability in this program was similar to an assessment center. Applicants completed a battery of personality and attitude surveys, performed role-play simulation exercises, and responded to questions in a structured interview. Selection into the program was based primarily on ratings from the structured interview. For the present analysis, role-play ratings and interview ratings were used as dependent measures and scores on the personality scales were used as predictors. The role-plays and structured interview ratings can be seen as providing behavioral assessment of managerial potential. Obviously, longitudinal data related to managerial success is needed to test many of the suggestions discussed earlier, but because the program is still in its infancy these data are not yet available. Behavioral assessment, however, has been found to be a valid predictor of subsequent performance (Borman, 1982), and the initial assessment data are useful for testing the relation between personality factors and managerial potential or talent, as measured by performance on content-valid tasks. Personality was measured by inventories commonly used in assessment center research (Howard & Bray, 1988). This raised a potential concern because most studies examining the Big Five factor structure have used self and other ratings of trait adjectives (e.g., Goldberg, 1990). One objective of the present study, therefore, was to examine whether common personality inventories alone or in combination are able to capture the Big Five. Only one personality

100

Lillibridge and Williams

inventory -- the NEO (Costa & McCrae, 1985) -- has been designed with the intent of measuring the five robust domains, and being relatively young, it does not have a long history of validation or prediction associated with its use. The main concern in using different personality inventories is whether there is a sufficiently comprehensive pool of items for the five-factor structure to be identified. Factors will not appear if they are underrepresented or not represented at all in the item pool. Likewise, only a subgrouping of a specific trait domain may appear from factor analyses of a limited set of items. Depending on the mix of measures employed in a given study, a factor in the resulting structure may represent the full content of a major domain, a subgrouping of traits from that domain, or the domain may be unrepresented. Thus, the present study had two objectives: (1) to explore whether the Big Five personality domains could be measured adequately with common personality inventories, and (2) to relate observed personality domains to the behavior assessment of managerial potential.

Setting and Subjects An early managerial identification program at a major technological university in the Northeast U.S. served as the setting for this study. The intent of this program was to identify undergraduate science and engineering majors with the aptitude to become successful managers, and provide them with training in such areas as decision making, interpersonal relations, and managerial leadership. Funding for the program was provided by major corporations in the U. S. Subjects were selected into the program on the basis of a number of behavioral factors identified by Stutzman (1985). A structured interview was used to assess applicants' standing on these behavioral dimensions. Trained interviewers provided ratings of specific behavior dimensions and an overall rating of managerial potential. These ratings were used by the Director of the program in making the final selection decision. Applicants to the early managerial identification program over a two-year period comprised the sample for this study. A total of 210 applicants (163 males and 47 females) from two separate years were included (91 from the first year and 119 from the second). The ratio of males to females reflected that of the University. Sex of applicant was not systematically related to the predictor or criterion variables and thus is omitted from the following analyses. All subjects were science or engineering majors entering their junior or senior year at the University.

Five Factor Model

101

Measures Personality Inventories. The full version of the Edwards Personal Preference Scale (EPPS, Edwards, 1959) and t h e GuilfordZimmerman Temperament Survey (GZTS, Guilford & Zimmerman, 1949) were used in this study. Subjects completed these personality inventories prior to the interview and role-playing sessions. Raters did not view subjects' responses or scores on these personality measures. Thus, knowledge of one's standing on personality dimensions was not considered in the rating or selection process. Structured Interview. The structured interview consisted of 21 questions addressing different dimensions of managerial behavior identified by Stutzman (1985): Salesmanship, Inner Work Standards, Reliance on Self, Energy, Technical Competence, Achievement, Leadership, and Need for Advancement. Answers were objectively scored using a behaviorally anchored rating scale. Interviewers summed these scores to obtain ratings for each applicant on the different interview dimensions. They also provided an overall managerial potential score (OMP) for each applicant. OMP was a subjective evaluation based on the dimensions' scores and the interviewer's overall impressions of the candidate. This overall score was used as a dependent measure in the following analyses. The interviews were conducted by trained interviewers who had attended an intensive, four-week rater training program. Adequate in ter-rater reliability was confirmed during the training program. Interviewers were paid for their services. Role-play Simulations. In addition to the structured interview, applicants were required to take part in a series of role-play exercises. These exercises were designed to measure three dimensions of managerial potential: Problem Solving, Interpersonal Skills, and Behavioral Flexibility (i.e., the ability to modify one's own behavior to reach a goal). Interviewers rated the applicants on 5 subscales for each dimension on a 5-point scales anchored at poor (l), average (3), and outstanding (5). The subscale ratings were averaged to yield on overall rating of problem solving ability, interpersonal skills, and behavioral flexibility.

Results Correlations Between Personality Scales and Managerial Potential. Zero-order correlation coefficients were computed between ratings of overall managerial potential (OMP) and applicants' scores on the

102

Lillibridge and Williams

GZTS and EPPS scales. In general, weak correlations between OMP and the personality scales were observed. A number of these correlations, however, were statistically significant. OMP was significantly related (p < .05) to four scales from the GZTS: General Activity ( r = . 15), Ascendance (r= .20),Objectivity ( r = . 17), and Personal Relations (r= .15). Higher ratings were given to individuals who could be described, respectively, as high in enthusiasm and energy, eager to voice their opinions, not suspicious of others, and tolerant of others. OMP was significantly related to three scales from the EPPS: Achievement ( r = . 16), Autonomy ( r = - . 16), and Abasement (r= -.16). Higher ratings were given to individuals who could be described as high in achievement orientation, low on independence, and low on feelings of guilt. A number of the personality scales were also significantly related to the selection decision, although again these correlations were weak. Those selected into the program were more likely to score high on General Activity, Ascendance, Sociability, and Achievement (all biserial rs between .15 and .18), and to score low on Autonomy (r=-.18). The ability to predict managerial talent from the separate scales of the GZTS and EPPS was relatively weak. This is not surprising because any one scale may not capture broader personality domains related to managerial potential. However, combination of these scales falling within the same domain may show stronger predictive power. In the present case, for example, Personal Relations and Objectivity, which measure respectively the extent to which people are tolerant of others and not hypersensitive, can be seen as part of the Agreeableness domain. Combining these measures, along with any others that reflect the domain of interest, may improve prediction.

Second-order Factor Analysis. To determine whether the Big Five domains were represented in the GZTS and EPPS, a second-order factor analysis was performed on the 25 subscale scores from the two personality inventories. Principal components analysis with varimax rotation yielded five interpretable factors, as determined by scree analysis. These factors and their factor loadings are presented in llble 3. It should be noted that three additional factors emerged with eigenvalues greater than 1.O, but were not retained because either they had only one item loading on them (Factor 7 - Masculinity, and Factor 8 - Achievement) or were not interpretable (Factor 6). The factors in Tkble 3 seem to provide evidence that four of the five robust personality domains were captured. Scale definitions for the scales loading on each factor are presented in l’hble 4.

103

Five Factor Model

Table 3 Factor Analvsis of GZTS and EPPS Scales. Variable Factor Factor Factor 1 2 3 Ascendance (G) Sociability (G) Dominance (E) Abasement (E) Activity (G)

Factor 4

Factor 5

.84 .73 .71 -.57 .54

Friendliness (G) Objectivity (G) Personal Re]. (G) Aggression (E)

.75 .69 .67 -.60

Change (E) Order (E) Exhibition (E) Endurance (E)

-.73 .60

-.38

-.54 .36

Affiliation (E) Nurturance (E) Succorance (E)

.75 .70 .53

Thoughtfulness (G) Intraception (E)

.82 .78

Heterosexuality (E)

-.40

Emotion Stab. (G)

.49

Autonomy (E) Eigenvalue % variance

.50 -.32

3.39 13.6

3.33 13.3

-.43 2.43 9.7

1.71 6.8

1.64 6.5

Note: Only loadings > .35 are shown. Eigenvalues for Factors 6, 7, and 8 were 1.3, 1.2, and 1.O, respectively.

Lillibridge and Williams

104

Table 4 Defining Terms for GZTS-EPPS Combined Factors. Scale

Scale Definition

FACTOR I (SURGENCY) Sociability

friendship, gregariousness, social activity

Activity

energetic, rapid pace, enthusiastic

Ascendance

spealung skills, persuasion, eager to voice one's own opinion

Dominance

to lead and direct others

Abasement (-)

not to feel guilty, not to accept blame

FACTOR I1 [AGREEABLENESS) Friendliness

tolerate hostile action, not belligerent

Personal Relations

cooperative, tolerant of people, faith in social institutions, not suspicious

Objectivity

thick-skinned, not hypersensitive

Aggressive (-)

not to become angry or criticize others

Five Factor Model

Table 4, continued FACTOR 111 (CONSCIENTIOUSNESS) Order

to have things organized, orderly

Restraint

serious minded, not impulsive, not carefree

Change (-)

not to do new things, not to experience change

Exhibition (-)

not to be the center of attention

FACTOR IV lAffiliation) Nurturance

to show affection for others

Affiliation

to be loyal to friends, do things with others

Succorance

to get help and encouragement from others

FACTOR V (INTELLECT) Intraception

to analyze one's motives and feelings, to analyze the behavior of others

Thoughtfulness

reflective, meditative, mental interests

105

106

Lillibridge and Williams

Table 5 Results of Multiple Regression Analvsis Predicting Overall Manaperial Potential Ratings from Personality Factor Scores.

-r

Variable

h

B

Surgency

,189

.195

2.79"

.23**

Agreeableness

.141

.122

1.73'

.17*

Conscientiousness .137

.080

1.13

.08

Intellect

-.052

-.028

-.39

-.08

Affiliation

-.273

-.113

-1.55

-.17*

t

Notes: R=.293; R2=.086, Adjusted R2=062; F(5, 193) = 3.62, p < .005; = p < .lo, * = p < .05, ** = p < .01. +

Five Factor Model

107

Factor 1 was interpreted as Surgency. The scales loading on this factor -- Ascendance, Sociability, Dominance, General Activity, and (negative) Abasement -- measure a number of traits from the Surgency domain (Goldberg, 1990), including gregariousness, expressiveness, energy level, talkativeness, and assertion. Factor 2 was interpreted as agreeableness and was comprised of Friendliness, Personal Relations, Objectivity, and (negative) Aggression. In terms of defining traits from the agreeableness domain (Goldberg, 1990), individuals scoring high on Factor 2 would display high levels of cooperation, amiability and generosity, and low levels of belligerence, overcriticalness and distrust. Factor 3 was interpreted as Conscientiousness, and was comprised of Restraint, Order, (negative) Change, and to a lesser extent, (negative) Exhibition. Individuals scoring high on Factor 3 would display such domain-relevant traits as persistence, predictability, organization, and low recklessness. Factor 4 did not correspond to any of the Big Five domains. Succorance, Affiliation, and Nurturance loaded the highest on this factor, and indicate high levels of help from others, loyalty to others, and high helping behavior. Thus, this factor was interpreted as Affiliation. Finally, Factor 5 was interpreted as Intellect, the fifth of the Big Five factors. The two scales loading highest on this factor, Intraception and Thoughtfulness, measure the extent to which individuals are analytical, introspective, and meditative. To test concerns regarding the stability of these factors, the average intercorrelation coefficient was computed for the scales loading on each factor, using r to Z transformations. A moderate to high intercorrelation among the scales loading on a given factor indicates that the factor should be stable across studies. The average intercorrelation coefficients were .39, .35, .30,.30, and .39, for Factors 1 to 5 , respectively. These correlations provide evidence that these factors are fairly stable.

Big Five Factors and Managerial Potential. Factor scores were computed for the factors represented in Table 3, and were correlated with OMP ratings. These correlations are presented in the last column in Table 5. Three of the five factors were significantly related to OMP ratings. Surgency and Agreeableness were positively related to OMP, r= .23 and .17, respectively, p < .05, while Affiliation was negatively related to OMP, r = - . 17, p < .05. Multiple regression analysis was conducted to determine the extent to which these factors predicted OMP. The results of this analysis are also presented in Table 5 . The overall regression equation was significant, and the five factors accounted for 8.6% of the variance in

108

Lillibridge and Williams

OMP. In the final regression equation, only the beta weight for Surgency was significant at t h e p < .01 level. The regression weight for Agreeableness was significant at thep C .10 level. A stepwise multiple regression anal sis was conducted regressing OMP on 9 dimension ratings rom the structured interview and the 5 factor scores to examine whether applicant personality contributed to the prediction of overall ratings over and above the effects of interview dimension ratings. Five of the nine interview dimension ratings (Salesmanship, Inner Work Standards, Leadership, Technical Competence, and Need for Advancement) entered the equation in steps 1 - 5, accounting for 82.5% of the variance i n OMP. Surgency entered t h e equation at Step 6 , significantly @< .02) incrementing R2 by .6%. Thus, controlling for performance on the behavioral dimensions of the structured interview, individual differences in Surgency significantly influenced raters' judgments of managerial potential. Finally, differences between those selected into the program and those not selected were examined. Those selected into the program had higher Surgency scores (M=58.8)than those not selected (M=53.2), t(200) = -3.46, p < .001. They also had lower (M=24.5) scores on Factor 4 (Affiliation) than those not selected (M=26.2), t(200)=2.31, p < .05. Overall, the factor scores representing the big five domains slightly improved prediction of managerial potential over the use of individual scales. Although the increase in explained variance was not great, using the factor scores rather than the individual scales clarified how underlying personality structure relates to managerial potential. These results suggest that the subjective component of managerial potential ratings was influenced significantly by the applicant's standing on the Surgency domain.

r

Relation between Personality Factors and Simulation Exercises. It could be argued that the effects of personality on OMP ratings in general, and the effects of Surgency in particular, reflect rater bias rather than actual differences in managerial talent. That is, sociable or extraverted individuals may make a more favorable impression on raters and hence receive higher ratings. Fortunately, the role play simulations used during the first year of the program provided an opportunity to assess "on-the-job" behavior in response to realistic organizational demands. Although still prone to rater distortion or bias, role play simulations provide more direct assessment of managerial competence (Cleveland & Thorton, 1990). In this case, role playing exercises were designed to elicit problem solving ability, interpersonal skills, and behavioral demonstration or flexibility.

Five Fuctor Model

109

Table 6 Correlations Between Personality Factor Scores and Role-Play Ratings (Program Year 11. Role-Play Exercise Rating Behavior Flexibility

Problem Solving

Interpersonal Skills

Surgency

.26*

.26*

.ll

Agreeableness

.23*

.18

.22*

Conscientiousness

.22*

-15

-.02

Intellect

.17

.14

.24'

Affiliation

-.08

-.08

.01

Factor

m:N = 81; * = p < .05,** = p < .01.

110

Lillibridge and Williams

Factor scores on the five factors were correlated with ratings of the three simulation dimensions. These correlations are presented in Table 6. All of the factors except Affiliation were significantly correlated with at least one simulation rating. Individuals scoring high on the Surgency domain received higher ratings on behavioral flexibility and problem solving skills. The correlation between Surgency and behavioral flexibility can be explained by a positive relation between energy, spontaneity, and animation and an individual's behavioral repertoire. The explanation for the positive relation between Surgency and problem solving is not as apparent, but may be due to the combination of activity, ascendance, and dominance traits possessed by individuals, which allows them to address problems aggressively. Individuals scoring high on Agreeableness received higher ratings on interpersonal skills and behavioral flexibility. Cooperation, amiability, and empathy are characteristics of the agreeableness domain that should facilitate interpersonal interactions; to the extent that the scales loading on Factor 2 measure these traits, individuals scoring high on this factor should be adept at managing interpersonal problems and conflicts. Conscientiousness did not show the expected positive relation with problem solving. Instead, individuals scoring high on Conscientiousness received higher ratings on behavioral demonstration. Intellect was positively correlated with interpersonal skills. Individuals scoring high on this dimension are, among other things, likely to analyze motives, feelings, and behavior of self and others. This analytic disposition may improve understanding of appropriate or constructive behavior in different interpersonal situations. Multiple regression analyses were conducted regressing the role play ratings on the five factor scores. The five factors accounted for 17.2% of variance in behavior flexibility, F(5, 75)=3.18, p < .02. Significant regression coefficients were found for Surgency and Conscientiousness. For problem solving, the five factors accounted for 12.3% of variance, F(5,75)=2.10, p < .08,with a significant regression coefficient found for Surgency. For interpersonal skills, the five factors accounted for 10.9% of variance, F(5,75) = 1.83, n.s. Overall, the factor scores reflecting the Big Five personality domains accounted for substantially higher variance in role-play ratings than ratings from the structured interview. The is encouraging because the role-play simulations reflected critical managerial behaviors to a greater extent than did the structured interview. Once again, Surgency consistently affected behavioral assessment ratings of managerial talent.

Five Factor Model

111

Implications The results of this preliminary study suggest that the fivefactor model may be useful in predicting managerial potential. The robust personality domains accounted for more variance in the behavioral role-play ratings than did the individual subscdes, and the relation between personality and assessment ratings was more clearly interpretable using the five-factor model. It appears that one's standing on the surgency domain, as identified by the present factor analysis, has the strongest influence on assessments of managerial potential. This finding is consistent with previous research demonstrating that such traits as dominance, self-confidence, energy level, assertiveness, and extraversion are related to managerial effectiveness and leader emergence (Lord et al., 1986; Yukl, 1989). One interpretation of the effects of surgency is that individuals high on this domain elicit positive reactions from others because of selfpresentation skills or the fact that their behavior matches prototypical representations of effective managers (Lord et al., 1986). Alternatively, as the role-play ratings suggest, critical managerial skills and behaviors may be related to the traits comprising the Surgency domain. Despite the consistent effects for surgency, the overall effects of personality on assessed potential were relatively weak. In fact, the strength of the observed effects is similar to that found in previous studies not using the five-factor model (Howard & Bray, 1988; Stogdill, 1974). Thus, the present results do not, in themselves, suggest that the five-factor model improves upon prediction of managerial potential. However, the weak effects could be due to the way in which the five-factor model was measured, rather than to the inadequacy of the model. Indeed, despite evidence that portions of four of the five domains were captured by the factor analysis, we urge caution in combining common personality inventories to identify the five factor model. It is doubtful that the GZTS and EPPS scales provide a comprehensive set of traits to represent the full content of each domain. Neuroticism did not emerge at all from our analyses, and some of the other domains may have been underrepresented. The finding that conscientiousness and intellect were not strongly related to the criterion measures may be due the fact that they were represented by only three and two scales, respectively. It may be the case that rather than identifying the core of these domains, only subgroupings of each domain were covered. Prediction of managerial potential from subgroupings of a domain will not be as strong as prediction from the full content of each domain. Other researchers have also failed to capture fully the five

Lillibridge and Williams

112

factors using questionnaire statements from personality inventories (Zuckerman, Kuhlman, & Camac, 1988). Thus, future studies should consider using personality inventories, such as the NEO-PI (Costa & McCrae, 1985),' designed to measure the five factors or adjective inventories which have shown the most success in capturing the five-factor model (Goldberg, 1990). When other personality inventories have to be used, such as to provide longitudinal or comparison data, attempts should be made to augment these inventories with additional measures related to the five domains.

CONCLUSION We started this chapter by presenting a proactive model of managerial competence and suggesting that the five-factor model of personality may be useful for organizing results and guiding research on managerial potential. The results of the preliminary study are encouraging in that they suggest that surgency, at least, is related to assessments of managerial potential. Further research, however, is needed before the utility of the advocated approach can be assessed. Specific attention should be given to methods of measuring the five robust domains. If stable measures of these domains can be obtained, the five-factor model has considerable potential in research examining (1) the relationship between personality and behavior taxonomies, and (2) selection, evocation, and manipulation processes in organizations.

NOTES

'

The NEO-PI was developed by factor analyzing responses from the 16PF (Cattell, Eber, & Tatsuoka, 1970), the GZTS, and the Eysenck and Eysenck (1975) Personality Inventory. This analysis yielded three of the factors: Neuroticism, Surgency (labeled Extraversion), and Intellect (labeled Openness). They subsequently added items to measure the Agreeableness and Conscientiousness domains, and were thus able to capture fully the five factors.

REFERENCES Borman, W. (1982). Validity of behavioral assessment for predicting military recruiter performance. Journal of Applied Psychology, 67, 3-9. Buss, D.M. (1987). Selection, evocation, and manipulation. Journal of Personality and Social Psychology, 53, 1214-1221. Buss, D.M., Gomes, M., Higgins, D.S., & Lauterbach, K. (1987). Tactics of manipulation. Journal of Personality and Social PsyChOlOgY, 52, 1219-1229.

Five Factor Model

113

Caspi, A. (1987). Personality in the life course. Journal of Personality and Social Psychology, 53, 1203-1213. Cattell, R.B., Eber, H.., & 'Itatsuoka, M. (1970). Handbook for the Sixteen Personality Questionnaire. Champaign, IL: Institute for Personality and Ability Testing. Clawson, J., Kotter, J., Faux, V., & McArthur, C. (1985). Selfassessment and career development (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall. Cleveland, J.N. & Thorton, G.C. I11 (1990). Use of simulations in management development: Reciprocity between science and practice. In K.R. Murphy & F.E. Saal ( a s . ) , Psychology in organizations: Integrating science and practice (pp. 111- 132). Hillsdale, NJ: Erlbaum. Conley, J.J. (1985). Longitudinal stability of personality traits: A multitrait-multimethod-multioccasionanalysis. Journal of Personality and Social Psychology, 49, 1266-1282. Costa, P.T., Jr., & McCrae, R.R. (1985). The NEO Personality Inventory manual. Odessa, FL: Psychological Assessment Resources. Digman, J.M., & Inouye, J. (1986). Further specification of the five robust factors of personality. Journal of Personality and Social Psychology, 50, 116-123. Digman, J.M., & 'Ihkemoto-Chock, N.K. (1981). Factors in the natural language of personality: Re-analysis and comparison of six major studies. Multivariate Behavioral Research, 16, 149-170. Edwards, A.L. (1959). Edwards Personality Preference Schedule. New York: Psychological Corporation. Emmons, R.A., Diener, E., & Larsen, R.J. (1986). Choice and avoidance of everyday situations and affect congruence: Two models of reciprocal interactionism. Journal of Applied PSyCholOgy, 51, 815-826. Eysenck, H.J. & Eysenck, S.B. (1975). Manual f o r the Eysenck Personality Inventory. San Diego, CA: Educational and Industrial Testing Service. Ford, D.H. (1987). Humans as self-constructing living systems: A developmental perspective on behavior and personality. Hillsdale, NJ: Lawrence Erlbaum Associates. Goldberg, L.R. (1990). An alternative "description of personality": The Big-Five factor structure. Journal of Personality and Social Psychology, 59, 1216-1229. Guilford, J.P., & Zimmerman, W.S. (1949). The Guilford-Zimmerman Temperament Survey. Beverly Hills, CA: Sheridan Supply.

Lillibridge and Williams

114

Howard, A., & Bray, D.W. (1988). Managerial lives in transition: Advancing age and changing times. New York Guilford. Lord, R.G., DeVader, C.L., & Alliger, G.M. (1986). A metaanalysis of the relation between personality traits and leadership perceptions: An application of validity generalization procedures. Journal of Applied Psychology, 71,402-410. Luthans, F., Hodgetts, R.M., & Rosenkrantz, S . (1988). Real managers. Cambridge, MA: Ballinger. McClelland, D.C. & Boyatzis, R.E. (1982). Leadership motive pattern and longterm success in management. Journal of Applied PsychologyJ 67, 737-743. McClelland, D.C., & Burnham, D.H. (1976). Power is the great motivator. Harvard Business Review, March-April, 159-166. McCrae, R.R., & Costa, P.T. (1987). Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology, 52, 81-90. McCrae, R.R., Costa, P.T., & Busch, C.M. (1986). Evaluating comprehensiveness in personality systems: The California QSet and the five-factor model. Journal of Personality, 5 4 , 430-446. Miner, J.B. (1978). Twenty years of research on role-motivation theory of managerial effectiveness. Personnel Psychology, 31, 739-760. Offermann, L.R. & Gowing, M.K.(Eds.) (1990). Special issue: Organizational Psychology. American Psychologist, 45, 95273.

Stogdill, R.M. (1974). Handbook of leadership: A survey of the literature. New York: Free Press. Stutzman, T. M. (1985, May). A model for early management identification. In H. G. Gueutal (Chair), Managerial selection in technically trained individuals. Symposium presented at the meeting of the Eastern Academy of Management, Albany, NY. Thorton, G.C. I11 & Cleveland, J.N. (1990). Developing managerial talent through simulation. American Psychologist, 45, 190-199. Williams, K.J.., & Lillibridge, J.R. (1990). The identification of managerial talent: A proactive view. In K.R. Murphy and EE. Saal (Eds.), Psychology in organizations: Integrating science and practice (pp. 69-94). Hillsdale, NJ: Erlbaum. Yukl, G.A. (1989). Leadership in organizations (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall.

Five Factor Model

115

Zuckerman, M., Kuhlman, D.M., & Camac, C. (1988). What Iies beyond N and E? Factor analyses of scales believed to measure basic dimensions of personality. Journal of Personality and Social Psychology, 54, 96-107.

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial1 Organizational Psychology - K . Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

117

CHAPTER 5 TEAM BUILDING AND ITS INFLUENCE ON TEAM EFFECTIVENESS: AN EXAMINATION OF CONCEPTUAL AND EMPIRICAL DEVELOPMENTS Scott I. Tannenbaum, Rebecca L. Beard, and Eduardo Salas American organizations appear to be using work teams with increasing frequency (Hackman, 1989). As U.S. businesses begin to rely more on the effectiveness of teams to attain organizational goals, and as the number of self-managing and semi-autonomous work teams, quality circle teams, task forces, and other forms of project and functional teams used in the U.S. increase, the need to enhance team effectiveness increases as well. Team effectiveness is different from individual effectiveness (e.g., Yetton & Bottger, 1982). There is ample evidence that work group problems can impede team performance (DeMeuse & Liebowitz, 1981). Therefore, it is imperative that we continue to improve our understanding of team functioning and team performance and continue to examine the usefulness of systematic interventions such as team building. Work teams can hinder or facilitate organizational effectiveness and can be a source of satisfaction or frustration for individuals. Thus, it is not surprising that a considerable amount of attention has been directed towards understanding how teams work and how to improve their effectiveness. One of the most prevalent and promising interventions designed to improve team functioning and effectiveness is team building (Beer, 1976; French & Bell, 1984). One

~ n ~ n bBeard, a ~ and ~ ,Salas

118

review indicated that team building was the dominant intervention used in 40% of all organizational development (OD) programs conducted through 1975 (Porras & Berg, 1978). This chapter examines team building and its influence on team effectiveness. Previous reviewers have noted a lack of definitional clarity in this area (DeMeuse & Liebowitz, 1981), so first we provide an explanation of what we mean by "teams" and "team building". Next, we describe an input-throughput-output model of team effectiveness to provide a context for examining team building interventions. Third, we highlight three previous reviews that examined team building research through 1980. Finally, we review the empirical research on team building published during the ~O'S,comparing it to the findings of the previous research and suggesting avenues for future research.

TEAMS AND TEAM BUILDING: DEFINITIONS

Teams There is a substantial body of research on groups. However, while all teams are groups, not all groups can be considered teams. For a group to qualify as a team, its members must rely on each other and share a common goal. For our purposes, a team is defined as a distinguishable set of two or more people who interact dynamically, interdependently, and adaptively toward a common and valued goal/objective/mission, and who each have some specific roles or functions to perform (Salas, Dickinson, Converse, & Tannenbaum, in press). Thus, a group of people attending a meeting together would not necessarily qualify as a team. Some form of task dependency exists among team members, some coordination is required, and some structure exists. Furthermore, our definition recognizes that a team is not a static entity but instead, its members interact dynamically and adaptively to meet its goals. Teams evolve over time (Gersick, 1988; Glickman, et al., 1987; Morgan, Glickman, Woodard, Blaiwes, & Salas, 1986). Roles and norms evolve, team members develop new skills and attitudes, tasks are modified, communication patterns unfold, goals are revised, personnel may change and hopefully, progress occurs. Change is inevitable. Although there are differences between a team and a group we believe that group research, selectively examined, offers some insights into understanding teams as well. We use the terms "group" and "team" interchangeably in this chapter although our focus is on teams throughout.

Team Building

119

Team Building Team building, sometimes called team development or group development, refers to interventions designed to improve the effectiveness of a work group (Woodman & Sherwood, 1980a). Most team building interventions are based upon an action research model of data collection, feedback, and action planning (Beckhard, 1969; Beer, 1976). While there are a variety of approaches to team building, most focus on improving team operations or processes. Group processes can be improved by removing barriers, clarifying roles, improving interpersonal relations, establishing agreed upon goals, or through other targeted intervention strategies. Team building is based on the assumption that a team's active involvement in planning change is more likely to result in favorable consequences than imposing change and that the p p l e closest to the task situation can solve their own problems if provided with an appropriate structure and process for doing so. In general, the underlying assumptions and values concerning team building are consistent with the humanistic philosophies of management (See Liebowitz & DeMeuse, 1982 for more detail on relevant assumptions). While team building interventions share a conceptual similarity they can vary with regard to their focus. Beer (1976) divided team building interventions into four basic approaches based on their primary focus: goal-setting, interpersonal, role, and managerial grid. While combinations and hybrids of these approaches are common, Beer's breakdown provides a conceptual framework for examining different types of team building interventions. In subsequent work, Beer concentrated on the first three approaches (1980). Dyer (1987) and Buller (1986) describe a more general problem solving approach to team building. Beer's first three approaches and Dyer and Buller's problem solving approach are briefly summarized below.

Goal-Setting Approach. This approach involves group members in the process of developing individual and group goals with the help of a consultant. The goals can be strategic in nature or may be more targeted such as specifying desired productivity or sales levels. Goals may also be set to change the internal environment and process. Interpersonal Approach. The interpersonal approach focuses on improving interpersonal relations in the group. It is based on the underlying assumption that an interpersonally competent group can function more effectively as a team (Argyris, 1962). This approach

120

Tannenbaum, Beard, and Salas

aims to improve group trust and encourage mutual supportiveness and non-evaluative communications within the team.

Role Approach. This approach involves discussion and negotiation among team members regarding each of their respective roles. A role refers to a set of behaviors which a person feels obligated to perform and which other people expect that person to perform (Katz & Kahn, 1978). This approach is based on the assumption that a team is basically a series of overlapping roles and that much of a team's behaviors can be understood in terms of individuals' perceptions of their roles. Role conflict and role ambiguity have been shown to result in anxiety, dissatisfaction, and potential turnover (House & Rizzo, 1972). Thus, team building interventions based on the role approach attempt to clarify roles and reduce conflict and ambiguity. Problem-Solving Approach. This is a more general approach to team building in which team members identify major problems, generate relevant information, engage in problem solving and action planning, and implement and evaluate action plans (Buller & Bell, 1986). This approach is based on the assumption that work teams become more effective by solving their major problems together. The differentiating feature of this approach is its focus on task-oriented problem solving although as part of this approach interpersonal, goal, and role "problems" may be identified and addressed. Prior to examining the research on team building we will first discuss the factors that can influence team effectiveness, including the role that team building can play.

FACTORS THAT INFLUENCE TEAM EFFECTIVENESS Why do some teams perform well and others poorly? Why does a particular team perform well for a while and then exhibit decrements in performance? Several researchers have presented useful theoretical models of team or group performance to help address these questions (e.g., Hackman, 1983; Nieva, Fleishman, & Rieck, 1978; Steiner, 1972). Each model has a slightly different emphasis, however they share a great deal in common ( S e e Salas et a]., i n press; Goodman, 1986 for an overview of existing frameworks). In Figure 1 we present an integrative model based on previous team research and theory. It is a variant of a model presented in Salas et al. (in press) and lhnenbaum, Dickinson, Salas,

ORGANIZATIONALAND SITUATIONAL CHARACTERISTICS Reward S ~ l r m s

ManPgrmcnl Control

OrganllstlomI Climslr

Iolrrgmup Rcbllom

Resource S a d l y

Lrrcl Of s1-

ComptlUon

EnvlmnmCnlsl U l l C e ~ l y

INPUT

hr

OUTPUT

THROUGHPUT

TEAMCHANGES CIURACTERISTICS -TrkTw -Task Complexlty

.N m No- New R o b .New C O m m u n l u U o m

STRUCTURE -Workkignmcnt ~Ttnm No- Commudcsllon Struc1urC

t t >< INDMDUAL CHARACTERISTICS

- Tprk K S A s .G e n e d Abllltlc. - Motbellon .Allltudc. - PIMnsllly

.Mcntnl Models

TI%% PROCESSES

I

1 TWLMPERFORMANCE

CHARACTERISTICS

.Porer Dl+trtbution

.Member Homogcnclly

.Team Resournr - CLlmnlr -Team Cobrrl*emrs

I

.

Feedback

TEAMINTERVENTIONS

'

Figure I : Team effectiveness model.

I

I

t

E. cy

9

z

122

Tannenbaum, Beard, and Salas

& Converse (1990), modified to incorporate variables relevant to our discussion of team building. The framework shown in Figure 1 is an input, throughout, output model. The major categories of variables are task characteristics, work structure, individual characteristics, team characteristics, team processes, team interventions, team changes, team performance, and individual changes. Specific variables are noted as bullets within each box. These variables are meant to be representative of the broader categories and are not an exhaustive list. Team inputs include individual and team characteristics (including its physical and financial resources), the characteristics of the task(s) on which the team is working, and the manner in which the work is structured. The throughputs are the manner in which the team interacts over a period of time. They are the processes through which the team communicates, coordinates, makes decisions, and converts its inputs into outputs. The primary outputs are the quality and quantity of the products produced and the services delivered, as reflected in the team's performance. Other outputs are changes in the team (e.g., new norms) and changes in individuals (e.g., improved skills), which in turn may influence subsequent team performance. Interventions, such as training and team building, may be targeted to improve team processes, enhance individual or team characteristics, or modify the work structure or task. All of these variables must be considered within the existing and developing organizational environment. Each major element of the model is discussed in more detail below.

Organizational and Situational Context Work teams operate within a particular environmental context (Hackman, 1983; Nieva et al., 1978). Although groups are often viewed as the context variable for individual behaviors, the organizational environment should aIso be considered as the context variable for group behavior (Gladstein, 1984). For example, one organizational variable that could influence a team's behavior is the organization's reward system (Hackman, 1983; Steiner, 1972). The existence and nature of group and individual based rewards can influence the level of cooperation or competition between team members, which in turn may affect team effectiveness (Lawler, 1981; Pritchard, Jones, Roth, Stuebing, & Ekeberg, 1988). The nature of intergroup or lateral relations within the company may also influence team effectiveness (Brett & Rognes, 1986). Cooperation between departments, functions, and teams may be

Team Building

123

necessary for successful team performance. Cross-unit relations is but one indication of organizational culture. A supportive organizational culture has been associated with higher performing teams (Sundstrom, Perkins, Futrell, & Hoffman, 1990).

Task Characteristics The influence of the team task on team performance has been well documented (Gladstein, 1984; Steiner, 1972; Goodman, 1986). Task complexity has been shown to be related to task performance (Herold, 1978) and a meta-analysis of previous team performance research revealed that task difficulty accounts for a significant amount of variance in team performance (Xmnenbaum et al., 1990). Tmsk organization and task type can also influence team effectiveness (McGrath, 1984; Kabanoff & O'Brien, 1979a). In sum, it should be remembered that all teams are not alike; task variations differentiate teams and influence performance.

W r k Structure A team can approach a given task in several ways. There can be differences in the way work is assigned to team members, for example whether each team member assumes a single role or exchanges roles as needed. Another element of work structure is the formal communication structure of the group. The nature of the communication structure, that is, who is allowed or required to speak with whom, can also influence the way work is performed (Naylor & Dickinson, 1969). Furthermore, teams establish norms regarding the way work is to be performed. Team norms can have a strong influence on team processes and subsequent performance (Hackman, 1987). As the model suggests, the nature of the task and the makeup of the individual team members will influence the way in which a team (or team leader) structures its work.

Individual Characteristics It has been well established that all things being equal, teams with better individual task proficiency, abilities, and skills will perform better (Gladstein, 1984; Kabanoff & O'Brien, 1979b; Steiner, 1972; Tmnnenbaum et al., 1990). Naturally, all things are not always equal; the sports world is full of examples of teams with lessor abilities outperforming "superior" rivals. However, on the average, holding task characteristics, work structure, and team processes constant, teams with greater individual abilities will out-

124

Tannenbaum, Beard, and Salas

perform teams with lessor individual abilities. Individual team members also bring various personalities and mental models to the team. Personality variables, such as sociability, adjustment, and likability may be related to team performance (Driskell, Hogan, & Salas, 1987). Driskell et al. argue that different tasks should be performed better by teams made up of individuals with certain personality traits. Mental models, or the symbolic representations that individuals have for various concepts, may also affect team performance (Cannon-Bowers, Salas, & Converse, 1990). Team members' mental models of their team's tasks and operations could determine individual behavior and team effectiveness.

Team Characteristics Several characteristics of the team can be important for understanding its effectiveness. The makeup of a team, including its homogeneity, has been shown to be related to team performance Gunderson & Ryman, 1967). A critical team characteristic is team cohesiveness. Team cohesiveness reflects a team's feeling of belongingness and sense of teamness. The importance of team cohesiveness has been discussed in the clinical psychology literature ( S e e Kellerman, 1981) and empirical work has shown its relationship to team performance (Dailey, 1980; 'Ihnnenbaum, et al., 1990).

Team Processes Team processes are the intragroup and intergroup actions that transform resources into a product (Gladstein, 1984). It refers to the way team members interact, including the way in which they communicate, make decisions, resolve conflicts, solve problems, and coordinate their actions. Communication processes have been shown to be related to team performance (Tushman, 1979; Katz, 1982), as have coordination processes (Kabanoff & O'Brien, 1979a, b) and decision making processes (See Guzzo, 1986 for a review of decision making and group effectiveness). The way in which a team manages its interactions with other units and non-team members is another key team process. "Boundary spanning" can influence team performance if cross-unit interdependencies exist (Adams, 1976; Gladstein, 1984; Katz & Kahn, 1978). In fact, some organizational development (OD) interventions are specifically targeted to improve the interactions between teams (Beer, 1976; French & Bell, 1984). There is a general assumption in almost all models of group

Team Building

125

performance that process is directly related to group effectiveness (e.g., Gladstein, 1984). For example, Beer (1980) suggested that if the groups throughout an organization are effective in their interpersonal relations, communications, problem-solving , goal-setting and decision making, then overall organizational effectiveness would also be high. However, some research has questioned whether that connection is always warranted. For example, Porras and Wilkins (1980) described an OD intervention where organizational performance increased following deterioration in process variables. Gladstein's (1984) research showed a connection between process and effectiveness when self-reports were used but no connection when objective performance measures were used. She suggests that the relationship seen in the self-report measures may reflect participant's implicit theories regarding appropriate team processes. Another possible explanation is that in a dynamic team context the relationship between process and performance outcomes may be lagged; it may take time before process changes lead to performance changes. Regardless, some questions remain regarding the nature of the relationship between process and effectiveness. This has important implications for team building, as most team building interventions attempt to change team processes and assume that they will lead to improved team effectiveness.

Out uts: Team Changes, Team Per$ormance, an Individual Changes

B

The primary output in our model is team performance. This refers to the quantity and quality of products and services, as well as time, errors, costs, and overall productivity. I n some situations, team performance can be measured as an average of individual outputs but in other cases aggregating individual measures would be misleading. While team performance is the primary output in our model it is not the only one. We consider changes in the team (e.g., new roles and processes, or greater/lessor cohesiveness) as outputs as well. Similarly, changes i n individuals (e.g., improved/decreased skills, attitudes, or motivation) are also considered outputs. In fact, Hackman's (1976) review of group research showed that groups can have a considerable effect on individual team members over time. Theoretically, these team and individual changes can ultimately influence a team's performance as well. Finally, as depicted in the model, a team's performance can serve as feedback to maintain or revise individual characteristics (e.g., train people, hire new members), work structure (e.g., assign

126

Tannenbawn, Beard, and Salas

tasks differently), or other team inputs and processes. Hackman (1987) suggested that teams evaluate their performance as they work, and that these evaluations affect team processes, which in turn influence subsequent performance. This cyclical framework is consistent with our view of a team as a dynamic entity, changing continually over time.

Team Interventions: naining and Bum Building There are many possible interventions to improve team effectiveness including job redesign, group incentive programs, training, and team building. We limit our discussion to training and team building. Although both team training and team building are interventions designed to enhance team functioning and improve team effectiveness, they are not the same. Training is a systematic effort to facilitate the development of job related knowledge, skills, and attitudes (K,S,A's). The specific knowledge, skills, and attitudes to be developed are determined and learning objectives are established prior to the start of the training. The training content is then derived and structured to convey the critical K,S,As and to meet the learning objectives (See Swezey & Salas, in press, for a detailed description of team training development). In contrast, team building is more of a "process intervention" (Beer, 1980). A process intervention is a set of activities which is aimed at helping individuals and groups examine and act upon their behavior and relationships (Schein, 1969). Although the general framework of a team building intervention may be established up front, the content of the intervention will in part be determined by the discussions among team members. Interestingly, although team training and team building interventions often focus on similar concerns (e.g., communication, decisionmaking, coordination) the means of approaching the concerns will differ. Based on the framework depicted in our model we consider the implications of these interventions for improving team performance; first focusing on training and then on the four approaches to team building described earlier,

naining. Access to necessary training has been shown to differentiate high performance teams from others (Sundstrom, et al., 1990). Training interventions can focus on the development of individual skills or on the development of team skills. As noted earlier, individual task proficiency is related to team performance so training that enhances individual characteristics (e.g., task K,S,A's,

Team Building

127

motivation) should contribute to team performance. However, team tasks require more than simply well-trained individuals. According to Steiner (1972) team performance is not only a function of the potential of team members to perform their assigned subtasks, but it is also a function of the ability of team members to coordinate their work flow and communicate effectively with one another. Although there has been little empirical research in this area, there is some evidence that training "team skills", such as coordination and communication, should result in improved team performance (Lassiter, Vaughn, Smaltz, Morgan, & Salas, 1990; McRae, 1966; Parsons, 1981). Some team training interventions would affect team performance by enhancing the team process skills, while others would concentrate on enhancing team characteristics by teaching interpersonal skills and changing team climate.

Goal-Setting Approach to Team Building. This approach to team building could enhance team effectiveness in several ways. First, establishing agreed upon output levels may clarify differences regarding what the team's goals should be and may help elucidate what resources are needed to accomplish the goals. Second, individual characteristics can be changed; more specifically, participative goal-setting can heighten team member's motivation and commitment (Latham & Yukl, 1975). Third, the team may establish goals to revise the nature of their task, work structure, or tern processes. Interpersonal Approach to Team Building. The interpersonal approach to team building could enhance team effectiveness in two ways. First, team characteristics can be changed as the result of an interpersonal team building intervention. Team climate could improve with the team exhibiting greater levels of trust, cooperation, and cohesiveness. Second, improvements in interpersonal relations among team members may lead to improvements in team processes. For example, as a result of the intervention, communication or conflict resolution processes may be enhanced. In sum, this approach is based on improving the way teammates feel about each other and the way they interact to enhance team effectiveness. Role Approach to Team Building. In terms of our model, the role approach could improve team effectiveness in several ways. First, by negotiating and revising member's roles, the team's work structure could be improved. Second, individual characteristics can be changed; for example, a team member's perceived role ambiguity could be reduced. Third, team characteristics could be changed as improved role clarity may enhance overall team cohesiveness. Final-

128

Tannenbaum, Beard, and Salas

ly, team processes could be enhanced as the removal of role disagreements can result in improved coordination and communication. In sum, this approach is based on clarifying team members roles and responsibilities to enhance team effectiveness.

Problem-Solving Approach to Team Building. As a more general approach to team building, this approach could improve team effectiveness by influencing almost any component of our model. First, team processes are directly affected as the team adopts a systematic problem-solving process. Then, based on the particular problems that are identified during the diagnostic stage and the agreed upon solutions that are implemented, virtually any element in the model could be affected. This could even include organizational characteristics as the work team might recommend changes in the reward system or other organizational level factors.

TEAM BUILDING: PREVIOUS REVIEWS Several researchers have reviewed the team building literature from the early 1960's through 1980, including Woodman and Sherwood (1980a), DeMeuse and Liebowitz (1981), and Buller (1986). These reviews provide a baseline with which to examine the research conducted in the 1980's. Woodman and Sherwood (1980a) summarized the research evidence on the utility of team development for increasing organizational effectiveness. They noted that team development is designed to improve the effectiveness of people with interdependent jobs, where effectiveness refers to managing problems and accomplishing group goals. They used the terms team development and team building interchangeably, stating that team building is designed to improve team operation by developing problem-solving procedures and skills and increasing role clarity, thereby enhancing organizational effectiveness. Their perspective assumes that team building facilitates team processes which i n turn influences team performance. Woodman and Sherwood delineated the difference between team development and T-group training, excluding the latter from their review. They identified 30 team development studies conducted between 1964 and 1978, and categorized them according to the major types of team building interventions specified by Beer (1976). They reported that team building interventions were used more frequently with management teams than with lower level work groups, and were more common with established work groups than as a means of developing newly formed teams. They expressed some concerns with the research designs and

Team Building

129

outcome measures used in many of the studies. All but one of the studies they reviewed were field studies, so it is not surprising that the formation of true control groups and the utilization of experimental designs was limited. The most common design used in the 30 studies was a pre-post, non-equivalent control group design, which when properly implemented, can be a useful design in field settings (Cook & Campbell, 1979). However, many of the studies reviewed relied on designs with poor internal validity. They noted that studies that used the goal-setting, problem-solving model of team building tended to use designs with better internd validity. Only one third of the research studies used objective performance data (e.g. , productivity, cost, absenteeism, turnover, quality). Most relied on affective outcome measures of satisfaction, climate, or attitudes or subjective assessments of effectiveness. They concluded that team building is related to affective responses but that its effect on performance is unsubstantiated. DeMeuse and Liebowitz (198 1) conducted a similar review. They examined research that pertained to team building interventions, focusing on 36 studies conducted from 1962 to 1980. Despite focusing on a similar research domain, out of a pool of 66 studies, DeMeuse and Liebowitz's review only shared 14 studies in common with the Woodman and Sherwood review. They note that this is a reflection of the vagueness of typical definitions of team building. They define team building as a long-term, data-based intervention in which intact work groups learn to increase their skills for effective teamwork by utilizing a structured agenda, usually with the aid of a behavioral-science consultant. They excluded lab studies, studies with multiple OD interventions, and T-group studies from consideration. They intended to exclude any studies that did not meet a twelve month time standard but were forced to drop that requirement because of the limited number of longitudinal studies. As with Woodman & Sherwood (1980a), DeMeuse and Liebowitz (1981) expressed concern with the quality of the research. They noted that very few researchers predicted specific relationships between independent variables and dependent variables prior to conducting the study. Many used a shotgun approach, collecting a variety of outcome measures and assessing if anything changed. Twenty of the thirty-six studies used pre-experimental designs, designs that Cook and Campbell (1979) would describe as those that do not permit reasonable causal inferences. The remaining sixteen were considered quasi-experimental designs, many of which were pre-post, non-equivalent control group designs. They also expressed some concern regarding the operationalization of team effectiveness. They noted that some researchers

130

Tannenbaum, Beard, and Salas

operationalized team effectiveness in terms of increased intragroup cohesiveness, reduced group-member absenteeism, and improved interpersonal relations. Based on our model, changes in these measures would indicate that team characteristics and team processes were effected but would not suggest that team pe?$omzance or effectiveness had truly been enhanced. All but one of the studies measured attitudinal/perceptual change as the outcome measure, fourteen examined behavioral changes, and nine considered organizational results. Almost half relied exclusively on perceptual dependent variables. Eighty percent of the results were positive and seventeen percent were mixed. Only one study reported negative results. While the general trend is overwhelmingly positive, DeMeuse and Liebowitz also expressed extreme caution in interpreting the results due to the generally low level of methodological rigor seen in the existing research. Work by Terpstra (1981) supports their call for caution. He found an inverse relationship between the degree of methodological rigor of evaluation employed and the degree of reported outcome success of OD interventions. DeMeuse and Liebowitz argued that increased rigor is needed and noted that Cook and Campbell (1979) offer some "refreshing" approaches for conducting quality research in field settings. Buller (1986) expressed concern about the ambiguity of previous team building models. He proposed a team-building-problemsolving approach to team development which he defined as a planned series of meetings facilitated by a third party consultant, with a group of people having common organizational relationships and goals that is designed specifically to improve the team's task accomplishment by developing problem solving procedures and skills and then solving the team's major problems. His approach focuses on improving team processes but also suggests the development of problem solving skills. He suggested that within this framework, team building activities can focus on task or interpersonal skills and may include goal-setting and role clarification. In essence, his apiroach subsumes any 6f the team building models suggested by Beer (1976). Buller identified fifty-two studies conducted through 1980 and focused on the nine studies that employed objective individual, group, or organizational performance criteria. Of those, five used quasi-experimental or experimental designs that provided some control over extraneous variables. Of those five, only one did not include other activities (e.g., training) that have been found to produce performance improvement by themselves. As with the other reviews, he concludes that the relationship between team building

Team Building

131

and performance is inconclusive. Business trends suggest that the use of work teams is increasing. Most theoretical models indicate that team performance is a complex phenomenon but agree that the way in which a team interacts is central to its effectiveness. Team building interventions are a prevalent means of attempting to improve team operations. Unfortunately, while the need to understand and evaluate team development increases, the research up through 1980 has been largely inconclusive. DeMeuse and Liebowitz stated that the eighties will be a critical period for the field. In the next section we review the research conducted in the 1980's and examine if the next generation of team building studies has addressed the concerns expressed at the end of the seventies.

TEAM BUILDING RESEARCH SINCE 1980 We conducted a literature search of the team building research published in the 1980s. A computer search was performed using the Psychologicul Abstracts, ERIC, ABI-Inform and NTIS databases. This was supplemented by a hand search of relevant journals and a cross-referencing of each article's reference list, We excluded organizational development efforts that did not include team building per se. We also eliminated studies that included only author's impressions of effectiveness (e.g., Cohen & Ross, 1982; Murrell & Valsan, 1985), as well as those studies in which artificial teams worked on hypothetical problems (e.g., Gear, Marsh, & Sergent, 1985) or used groups that did not fit our definition of a team. We did not exclude studies with multiple interventions if the primary intervention was team building. When provided by the authors, we include separate information regarding the effectiveness of other interventions (e.g., Fiedler, Bell, Chemers, & Patrick, 1984). The search process yielded seventeen empirical team building studies that met our criteria and were reported in the 1980s. Three of the studies were also reported in previous reviews. Collapsing across the current and previous reviews (i.e., Buller, 1986; DeMeuse & Liebowitz, 1981; Nicholas, 1982; Woodman & Sherwood, 1980a), 67 different empirical team building studies have been reported. Fourteen studies were reported in the 1960's, 36 in the 1970's, and 17 in the 1980's. 'bble 1 shows the seventeen studies conducted in the 1980's. The table includes the type of team(s) being studied, the number of teams and overall sample size, the research design employed, the team building approach(es) used, and a summarization of the results. To allow for comparisons over time we used a categorization schema similar to those used in previous reviews.

132

Tannenbawn, Beard, and Salas

Table 1: Summary of 1980's Team BuildingStudies(continued) OUTCOME VARIABLES ~

#OF SAMPLE AUTHOR

TEAM

TEAMS

SIZE

Eden (1985)

CommandTeams. Logistics Units, Israeli Defense Forces

18

147

Eden (1986)

Command Teams, Combat Corps, Israeli Defense Forces

16

Approx 160

Fiedler, Bell, Work Teams, Chemers. B UndergroundMining Patrick (1984) Company

Fleming B Fleming (1983)

Halstead et al. (1986)

DESIGN

APPROACH

lnii React

AttiV Percep

+

mainly 0

Behav Org Change Change

Experimental Pretest-Posttest True Control Group Randomization

Goal-Setting. Interpersonal. Role

Quasiexperimental Pretest-Posttest Non-equivalent Contrd Groups

Goal-Setting. Interpersonal. Role

Quasiexperimental Pretest-Posttest Non-equivalent Control Group

ProMem Sdving

+

+

+c

+

Structured Management Training

+

+

+

+

t

Multidisciplinary Educational Teams, Elementary 8 High School Levels

5

Approx 35

Pre-experimental Pretest-Posttest

Goal-Setting. Problem Solving

Multidisciplinary Healthcare Teams

3

Approx 21

Qwsiexperimental Pretest-Posttest Non-equivalent Control Group

Problem Solving

mainly

+

+

+

+ CL

w

w

c

Table 1: Summary of 1980's Team Building Studies(continued)

w OUTCOME VARIABLES

AUTHOR

TEAM

#OF SAMPLE TEAMS SEE

DESIGN

APPRQACH

lnit AnW ~eact P m p

+

Hughes. Cadet Squadrons, Rosenbach, B USAF Academy clover (1983)

2

136

QuasiexperimentaI Pretest-PMest Non-equivalent Control Group

GdSening. Interpersonal. Rde

Uaplan. Lombar- Top Level do, 8 Mazique Management Team. (19W PuMic Agency

1

17

Preexperimental PretestPosttest

Other

+

97

Preexperimental P m e s t Only

Other

+

Margerison. Mc- Aircrews. Cann 8 Davies Australian Airlines

i

(1988)

(see also Margerison. Davies, B Mccann, 1987)

Mitchell (1986)

Management Teams (4) MBA Field Project Teams (13)

17

Approx 76

Quasiexperimental Pretest-Posttest Non-equivalent Control Group

Interpersonal

0

Other

+d

Behnv OQ Clmngo Change

t

P

Table 1: Summary of 1980's Team Building Studies(continued) ~

~~

OUTCOME VARIABLES #OF SAMPLE

AUTHOR

Morrison a Sturges (19W

Paul 8 Gross (1981)

TEAM

Top Management Team, large State Gwemment Departrnent

TEAMS SIZE

1

13

Communication and Electric D i i o n . City of San Diego

Food Service Managersand Staff

116

1436

(1QW

Woodman B Shewood

EngineeringStudent Survey Teams

67

192

Porras B Wilkins

(19Mb)

a

DESIGN

APPROACH

Pre-experimental Patched Design Pretest-Posttest No Control Group

Role. Problem Solving

Qwsiexperimental Some Premeasures Posttest Non-equivalent Control Groups

Role. ProMem Sobinge

Quasiexperimental Time Series Non-equivalent Control Group

Problem Solving. Interpersonale

Experimental Posttest Only Random Assignment True Control Group

Goal-Setting, Problem Sdving

lnit Read

+

AHi Beluv Oq Parccp Change Change

mixed

+

+

Comparison was between teambuilding and pal selling versus goal setting on&, a no-rrearmenr condition was not used. Goal Selling considered dislincl from teambuilding by author. Strike at mine prohibited collecrion ofproductiriry data - however, absenteeism declined. Significant change pretest lo posttest. No significanl dflerences between "other" and interpersonal approaches. Teambuilding conducted as pan of a larger orgonizotional deivlopmenr effon including individual developmenl and Tstem development.

'

mixed

0

t

136

Tannenbaum, Beard, and Salas

We coded each study's research design as either pre-experimental, quasi-experimental, or experimental ( S e e Cook & Campbell, 1979). Supplemental information regarding the use of pre- and posttests, control groups, and randomization is also included. We categorized each team building intervention according to Beer's (1976, 1980) breakdown (i.e., goal, role, and interpersonal). Consistent with Beer's later work we did not consider the managerial grid as a team building intervention. Regardless, no empirical studies were found that used the managerial grid as a team development intervention. We also included a problem-solving category (Buller, 1986; Dyer, 1987) and an "other'' category. If more than one approach was used, each approach is noted. Study outcomes were categorized similarly to DeMeuse and Liebowitz (1981). They based their schema on Kirkpatrick's (1976) training evaluation typology; it includes initial reactions, attitudinaUperceptua1changes, behavioral changes, and organizational changes. Initial reactions refers to evaluative remarks concerning the usefulness of the intervention made at the conclusion of the off-site session(s). AttitudinaUperceptual changes involve the subjective assessment of various intergroup conditions (e.g., climate, trust, openness, satisfaction) by participants. Behavioral changes includes team member behaviors such as those collected as critical incidents (e.g., specific helping behaviors). Organizational changes refers to the impact that the intervention has on the organization as a whole, for example its effect on organizational productivity or turnover, and not merely within team changes. Each study's outcomes were categorized as positive (+), negative (-), no change (0),or mixed.

Comparisons With Previous Research How do the studies of the 80's compare with those from the previous reviews? The primary focus is still on intact, established work groups as opposed to newly formed teams. To our knowledge there has been no research that has examined the relationship between the stage of a team's development and the usefulness of various team development interventions. In addition, most of the research is still being conducted with white collar teams although Buller's work with mine production crews provided a fascinating look at team building within a blue collar environment. Buller and Bell (1986) reported a general skepticism of modern management techniques within their blue collar production teams. Although task differences are reputed to be an important factor in understanding team performance and processes (McGrath, 1984) there have been

Team Building

137

no systematic efforts to consider the implications of task differences for team development interventions.

Research Designs A common complaint of previous team building research, and of OD interventions in general, is the poor quality of the research designs. Perhaps the biggest concern with previous research is the use of inadequate methodological controls. Over half the studies reported by DeMeuse & Liebowitz used pre-experimental designs that make it extremely difficult to attribute causality. In addition, no studies prior to 1980 used a true experimental design (DeMeuse & Liebowitz, 1981; Woodman & Sherwood, 1980a). Even when quasi-experimental designs were used, concerns remained about the similarity of the teams experiencing the team building interventions and any non-equivalent control groups (Hughes, Rosenbach, & Clover, 1983). Much of the pre-1980 research was based on fairly small sample sizes. Over 40% of the studies that reported the number of teams included only one team in their study (DeMeuse & Liebowitz, 1981). Lastly, concerns have been voiced about the types of outcome measures that have been used in previous research. This last concern will be addressed in a later section. There appears to be some improvement in the research methodologies employed in the 1980's. Eleven of the seventeen studies used a quasi-experimental design and two (Eden, 1985; Woodman & Sherwood, 1980b) employed true experimental designs. Interestingly, the two experimental studies were conducted with military and student teams, two environments where researchers often have greater control over participants than in private sector studies. One study that used a pre-experimental design incorporated several elements to improve its interpretability (Morrison & Sturges, 1980). They used a "patched-up" design (See Campbell & Stanley, 1966) with multiple data collections and dependent variables. DeMeuse & Liebowitz considered this to be a quasi-experimental design. However, based on Cook and Campbell's (1979) categorization we did not feel it qualified as such, and instead we consider it to be a "better" pre-experimental design than most. The most prevalent design used in the 80's was the pre-test, post-test, non-equivalent control group design. Some researchers carefully matched non-equivalent control groups and some randomly assigned teams to conditions when it was impossible to randomly assign individual participants (e.g., Hughes et al., 1983). However,

138

Tannenbawn, Beard, and Salas

non-equivalence of control groups continued to be a concern. One study (Fiedler et al., 1984) used industry averages as controls. This approach has been used in previous OD studies (e.g., Beckhard, 1966) although it has since been criticized by Hughes et al. (1983). In addition, the number of teams examined within a given study remained fairly small although the incidence of single team studies has decreased. In particular we applaud Eden (1985, 1986) for his work comparing results across different conditions of methodological rigor. He conducted one of only two published team building field experiments and intended to conduct a second, but organizational constraints reduced it to a quasi-experimental study. In one study (1985) he explicitly compared results across three different types of evaluation designs. This study supports the work of Terpstra (1981) in suggesting an inverse relationship between rigor and effectiveness. However, in a subsequent study, Eden's well-conducted quasi-experiment (1986) revealed positive results for the team building intervention despite a fairly high level of rigor. We cannot be certain that pre-experimental evaluations are not being conducted at a similar rate as before. However, the published team development research, particularly in psychological journals, relies much more heavily on quasi-experimental studies than in the 60's and 70's. On average, the methodological rigor of team building studies improved during the 80's. However, small sample sizes still abound and questions regarding internal validity within some quasi-experimental studies makes it difficult to interprete some of the findings. Mitchell's (1986) work with MBA field project teams is a good example of a study with some redeeming methodological features, but with some lingering threats to internal validity. He used a control group and randomly assigned teams to conditions, two excellent features in comparison to older team building studies. He developed a team building intervention based on alignment theory whereby individuals disclose their internalized frameworks. He compared the alignment approach with the conventional interpersonal approach to team building. Unfortunately, despite obviously favoring the alignment approach, he personally delivered both treatments, Furthermore, the interpersonal condition appeared less intensive than similar interventions in previous research (it was only 2.5 hours in length), Lastly, large differences in pre-tests scores made regression to the mean a potential problem and suggests that the alignment team had more room for improvement. In sum, although he incorporated several useful quasi-experimental features into his study, other

Team Building

139

threats to internal validity make it difficult to interpret his results unambiguously.

Nature of the Team Building Interventions According to Woodman and Sherwood (1980a) two-thirds of the previous team building studies used a goal-setting approach, either alone or in conjunction with other approaches. About forty percent of the studies used an interpersonal approach, and less than twenty percent used a role approach. Thirty percent used more than one approach to team building Of the seventeen studies conducted in the go's, 47% used an interpersonal approach (solo or in conjunction with others), 35% used a role approach, and 29% used a goal approach. Unlike Woodman and Sherwood, we included the problem solving approach as a distinct category. Problem solving interventions tend to focus on several of the traditional team building issues of goal, role, and interpersonal; 53% of the studies conducted in the 80's used a problem solving approach. Over seventy-five percent of the studies used either a multi-faceted problem solving approach or combined two or more of the approaches specified by Beer. "Other" approaches included Mitchell's alignment approach and the use of a management simulation as a diagnostic stimuli (Kaplan, Lombardo, & Mazique, 1985). It appears that the interpersonal approach is used with about the same frequency as before and the use of role clarification interventions has increased slightly. The use of goal-setting has been reduced, although this probably reflects its natural use as part of the more generic problem-solving approach. While Dyer (1987) noted a shift in emphasis away from focusing strictly on social interactions, towards a greater attention on team behaviors and tasks, the interpersonal approach to team development is continuing to be used and studied as part of an acceptable intervention strategy. Unfortunately, most of the studies do not include much detail regarding the intervention itself, perhaps due to journal page limitations. In general, the most interesting trend regarding the type of interventions used pertains to the use of multiple team building approaches. Over 75% of the studies conducted in the 80's used interventions that target more than one obstacle to team development. This may reflect an increased awareness of the complexity of team performance or may simply be a shot-gun approach on the part of the intervention developers. A key question that needs to be addressed is, besides personal preference, why choose one approach over another? Unfortunately, details regarding pre-intervention

140

Tannenbaum, Beard, and Salas

diagnosis and problem analysis are rarely included in the articles. The linking of specific team concerns with appropriate interventions is a logical necessity. The problem-solving approach, at least conceptually, is designed to ensure that linkage.

Outcome Measures Almost all the studies reviewed by DeMeuse and Liebowitz examined participants'attitudes/perceptions. In addition, 36 % considered behavioral variables, and 22 % considered organizational changes. Similarly, in the studies we examined, almost all examined participant's attitudes/perceptions. However, the measurement of behavioral variables appeared to increase (65 %) and organizational changes were assessed slightly more than before (29%). There are two other ways of categorizing outcomes measures that have relevance for understanding the effectiveness of team development interventions. The first distinction is whether the outcome measure was provided via self-report. The second distinction is whether the outcome measure focused on team processes or on team performance. Both distinctions are worth considering further. Only one third of the studies in Woodman and Sherwood's review included objective or non-self report outcome measures, while 53% of the studies in the eighties did so. There is some indication that results can differ according to which type of measure is used. For example, Eden (1985) reported positive results based on self-assessment but non-significant results with other measures. Similar findings were reported by Momson and Sturges (1980). Self-report data may reflect participant satisfaction with the team building or participants' implicit theories regarding appropriate team processes rather than actually measuring the effectiveness of the team building. Therefore, it is important that whenever possible, researchers attempt to include some non-self report indicators in their team building studies. It appears that some improvements were made in this regard during the 80's. Most of the research on team building has focused on processrelated outcomes or what Hackman (1983) considers intermediate criteria. Referring back to Figure 1, most of the research has examined team changes or individual changes as outputs or indications of the effectiveness of the team building intervention. For example, in his study of military teams, Eden (1986) examined involvement, conflict handling, cohesion, and clarity as dependent variables of interest. In contrast, far fewer studies included team performance measures as the dependent variable of interest.

Team Building

141

Approximately 80% of the studies conducted in the eighties measured team or individual changes. Only 40% assessed team performance - either via self-report or through some more objective method. When only objective measures are considered the number shrinks even further. Buller (1986) reported that only 17% of the studies he examined included objective measures of either individual, group, or organizational performance. This reliance on process measures as indications of the effectiveness of team building interventions is troubling. As noted earlier, there is not necessarily a connection between improved team processes and improved team performance. Yet most organizational interventions are implemented with an implicit (if not an explicit) expectation that it will yield improved performance or it will be discontinued. Subsequent research studies should attempt to assess changes in team performance in addition to examining changes in team processes and operations. For example, Buller & Bell (1986) collected data regarding changes in process (e.g., task performance strategies), behaviors (e.g., used proper slushing techniques) as well as changes in performance (e.g., quantity and quality of ore produced). Process measures are critical because they can help explain why an intervention works or fails. However, changes in process do not necessarily mean that a team intervention has been successful (Campbell & Dunnette, 1968).

Efectiveness of Team Building Although the quality of team building research improved in the

~O'S,methodological concerns still made it difficult to interpret the results of some studies unequivocally. However, within the methodological limits noted above, some general comments can be made regarding the effectiveness of team building interventions. As with the earlier team development research, most of the studies in the 80's yielded predominantly positive findings. However, the results were not uniformly positive, particularly with regard to behavioral changes. Despite the unwritten but generally acknowledged bias towards publishing positive findings, four of the ten studies that assessed behavioral change reported mixed or non-significant results. Improvements in team member perceptions or attitudes were evidenced more consistently, although some studies reported mixed or flat results (e.g., Eden, 1985; Momson & Sturges, 1980) and one study reported a decrement in job satisfaction and commitment (Porras & Wilkins, 1980). When we consider only the six studies that used the strongest

142

Tannenbaum, Beard, and Salas

research designs, a similar pattern emerges. Two studies used experimental designs (Eden, 1985; Woodman & Sherwood, 1980b) and four used stronger quasi-experimental designs (Buller, 1988; Buller & Bell, 1986; Eden, 1986; Hughes et al., 1983). Five of these six studies examined perceptuaVattitudinal change; four exhibited positive results. Four of the six studies examined behavioral change with only one showing positive results; the other three reported no behavioral improvements or mixed results. It makes sense that team building interventions would show a stronger effect on attitudes and perceptions than on performance. As seen in our model, team performance is a function of many factors, some of which are beyond the scope of most team building interventions. A team building program could enhance team cohesiveness, clarify member roles, and improve member satisfaction, yet the team still may not improve its performance due to a lack of resources, poor intergroup relations, technical problems, or adverse conditions in the environment. The further removed the dependent variable is from the immediate control of the team (i.e.! the more distal the measure), the less likely it is that the team building intervention will demonstrate improvements. Thus, a study that measures team communications is more likely to show improvement than one that measures team performance, since communications are within the direct control of the team. Team performance improvements are less likely to be demonstrated since team performance is partly in control of the team and partly externally determined.

The Role of Diagnosis. Team development activities can target several areas of concern. To date, the accumulated research still provides insufficient evidence to suggest that one team building approach works more frequently than another; each approach appears to work in some instances. However, unidimensional interventions (e.g., role clarification) should be most likely to affect corresponding dependent varibles (e.g., role ambiguity) but should be less likely to affect more global measures of effectiveness such as team performance. Kaplan (1979) and Hughes et al. (1983) suggested that improved process outcomes, such as enhanced openness in communications, may only contribute to improved performance in work groups that have been experiencing conflict. This suggests that diagnostic accuracy is critical to the success of any team building intervention. Most of the studies in the 80's used more than one team building approach, attempting to improve team processes and performance through some combination of role clarification, interpersonal

Team Building

143

enhancement, goal-setting, and problem-solving. Naturally, if the increased breadth of intervention exhibited in these studies was determined via an accurate diagnosis, then the probability of enhancing team effectiveness should be greater. To the extent that a team building intervention addresses several true team deficiencies, then performance improvement should be more likely. Conceptually, this is the logic behind the problem-solving approach to team building, where the team conducts a self-diagnosis and then focuses on the specific problems/concerns that it faces. Unfortunately, it is impossible to determine the thoroughness of the diagnostic efforts in most studies. Kaplan et al., (1985) described the use of a management simulation as a diagnostic team building intervention. This approach appeared to provide some useful diagnostic information regarding team operations. However, Kaplan et al. (1985) suggested that a team simulation should only be used and interpreted in conjunction with other forms of diagnosis. Further research on team diagnostic techniques is clearly needed.

Multiple Interventions: Team Building Plus., . Team building is not always the appropriate intervention to improve poor team performance. In some cases, the underlying reason(s) for poor performance may be skill deficiencies of team members (suggesting the need for training), motivational problems (which may suggest the need for changes in the reward system), resource problems, or task problems (perhaps calling for job redesign). In some cases, team building may fail to enhance performance because it is an insufficient intervention strategy for addressing the team's problems. A few studies have examined the effectiveness of team building in conjunction with another intervention. Buller (1988) and Buller and Bell (1986) examined the combined effects of team building and a goal-setting intervention (not a goal-setting approach to team building but a I'true" goal-setting program). They provided only limited evidence that the two interventions together enhanced team effectiveness. Porras and Wilkins (1980) and Paul and Gross (1981) described team building interventions that were part of larger OD efforts that included training and other forms of individual and system development. Paul and Gross (198 1) reported improvements in morale and productivity as a result of a combined team building, counseling, and training effort. Porras and Wilkins (1980) reported decrements in attitudes and team processes but improvements in unit performance in their study of a combined OD/team building intervention. The research regarding the combined use of team building and

144

Tannenbaum, Beard, and Salas

other interventions is mixed. It is possible that, when properly diagnosed, team building may serve as an "enabler" to enhance the efficacy of other interventions. Perhaps team building may prepare teams for other changes. However, at this point that is merely speculative. Future research should systematically examine the potential interactions between team building and other interventions.

Process to Pe@ormance: Time Lag, An assumption of most team building interventions is that improving team processes leads to improved team performance. However, it is still unclear under which conditions this assumption is valid. In addition, a related question is, how long does it take for improved processes to translate into improved performance? It is possible that in some studies (e.g., Woodman & Sherwood, 1980b), the time lag between the intervention and the collection of team performance data was too short to allow process improvements to translate into performance improvements. In a simple laboratory task, process improvements may be accompanied by an immediate improvement in performance. However, in complex organizational contexts, it may take some time before improved performance occurs, Thus, process and performance measures collected at the same time may not reveal how improvements in process over a period of time leads to better performance. Buller's research provides some support for this contention. Buller and Bell (1986) reported changes in team members' behaviors 3 to 6 months after a team building intervention but the teams showed no improvements in productivity. However, when Buller (1988) examined data from the same teams 15 months after the intervention, he found improvements in productivity and in profitability per shift. It is possible that three to six months was too short a time period for improved team relations and processes to result in improved performance, but that fifteen months was sufficient. It would be inappropriate to conclude that performance lags process by a specific and constant period of time, It is more likely the case that a variety of factors can influence the relationship between process and performance. In particular, we would speculate that the nature of the task (e.g., degree of interdependence, task complexity) would play a key role. In addition, it is likely that performance feedback influences team processes in a cyclical fashion as well (Hackman, 1987). Nonetheless, there is clearly a need for longitudinal research that examines the relationship between changes in team processes and performance. Future researchers should consider measuring post-intervention processes and perform-

Team Building

145

ance in a time series framework to allow for a closer examination of the relationship over time.

Regression or Fade-out. Another research issue addressed during the 80's was the question of regression or fade-out. Concerns have been voiced about the on-going effectiveness of off-site team development interventions (Beer, 1980). Similar concerns have been voiced in the training literature (cf., Marx, 1982). Boss (1983) examined the usefulness of a post-intervention strategy for maintaining team building effectiveness. After a typical interpersonal team building intervention, Boss had each team leader hold a separate role negotiation meeting with each of his or her subordinates, and then conduct private, one-on-one meetings on a regular basis. He compared these teams with those who underwent team building but were not given a follow-up strategy. Both sets of teams reported positive results immediately after the team building intervention. However, teams without a follow-up strategy showed significant decrements three months after the intervention, while the experimental teams showed improvements up to three years after the original intervention. His research suggests that, as with training, what happens after the intervention can be just as important as the nature of the intervention itself and that team building should be thought of as an on-going intervention and not a one-shot deal.

Summary of Team Building Research in the 80's We offer the following summary based on our analysis of the research conducted in the 1980's:

* The quantity of research on team building has decreased. However, the quality of research has improved. The first two experimental studies of team building were published and most studies used some form of quasi-experimental design. However, methodological concerns remain. * The research still focuses, for the most part, on intact, whitecollar teams. Little attention has been directed at newly formed or blue-collar teams. * Fewer studies used a unidimensional approach to team building. More studies used multiple approaches or a broader problem-solving framework. However, there is insufficient evidence to suggest that one approach or combination of approaches is most useful. It is likely that targeted diagnosis is a key.

146

Tannenbaum, Beard, and Salas

* More studies included behavioral measures than in the past and more studies included objective measures or at least measures other than self report. This is a positive development. Unfortunately, only 40% of the studies examined team performance. * There is further evidence that the assumed connection between process and performance is not always warranted. Improvements in process may only yield improved performance if process problems existed. Furthermore, there may be a time delay before process improvements translate into performance enhancement. * In general, team building interventions were fairly effective. Team building appears to have a positive effect on perceptions and attitudes. However, the results are more equivocal with regard to behavioral outcomes. At this point, it is still unreasonable to claim that team building alone can enhance organizational effectiveness, although it may be useful as part of a larger improvement strategy. It is important that decision makers have realistic expectations regarding the usefulness of team building interventions. * Activities after the actual team building intervention may play an important part in determining the long-term effectiveness of the intervention. It may be a good idea to establish a post-intervention strategy as part of the development of a team building intervention. Research Needs/Suggestions We have identified the following research needs/suggestions based on the existing research on team development. Some are suggestions regarding research design while others are issues or topics that we believe merit further attention:

* There have been improvements in research quality. While it is often impossible to conduct true field experiments, we encourage the use of at least a pre-post, non-equivalent control group design in future studies. In addition, in light of the differences exhibited between self-report and more objective measures, we encourage the collection of at least some non self-report measures. * The nature of the relationship between process and performance deserves further attention. To assess this properly calls for a longitudinal research strategy, perhaps utilizing a form of time-series analysis. Complex causal modeling may be needed to test for reciprocal relationships between process and performance over time. Alternatively, longitudinal qualitative research could help shed some light on this issue as well. The collection of rich, qualitative data before and after interventions, supplemented by periodic objective

Team Building

147

performance measures could be a fruitful research paradigm (cf., Van Maanen, 1975). * Two variables that could moderate the effectiveness of team building interventions are task type and stage of team development. To our knowledge no research has examined the effects of these variables. * It appears clear that accurate diagnosis is a key to team development, but typically little information is provided regarding how the diagnosis was conducted. Additional research should be directed towards validly diagnosing team inputs, processes, and outputs. * We would like to see research that examines the potential interaction of team building and other interventions. Since there is general agreement that team performance is a complex phenomenon it seems unlikely that any one intervention will consistently account for large amounts of variance. Systematic efforts to assess the impact of combined interventions may be useful. * Finally, it is important that negative findings be published, as well as positive findings. A great deal can be learned from examining why an intervention failed as well as why it succeeded.

REFERENCES Adams, J.S. (1976). The structure and dynamics of behavior in organizational boundary roles. In M.D. Dunnette (Ed.). Handbook of industrial and organizational psychology (pp. 1175-1200). Rand McNally: Chicago. Argyris, C. (1962). Interpersonal competence and organizational behavior. Homewood, Ill: Irwin Beckhard, R. (1966). An organization improvement program in a decentralized organization. Journal of Applied Behavioral Science, 2 , 3-25. Beckhard, R. (1969). Organizational development: Strategies and models. Reading, MA: Addison-Wesley. Beer, M. (1976). The technology of organization development. In M.D. Dunnette (Ed.). Handbook of industrial and organizational psychology (pp. 937-994). Rand McNally: Chicago. Beer, M. (1980). Organization change and development: A systems view. Glenview, Ill: Scott, Foresman & Co. Boss, R.W. (1983). Team building and the problem of regression: The personal management interview as an intervention. Journal of Applied Behavioral Science, 19, 67-83. Boss, R.W. & McConkie, M.L. (1981). The destructive impact of a positive team-building intervention. Group and Organization Studies, 6, 45-56.

148

Tannenbaum, Beard, and Salas

Brett, J.M & Rognes, J.K. (1986). Intergroup relations in organizations. In P.S. Goodman (Ed.). Designing effective work groups (pp. 202-236). San Francisco: Jossey-Bass. Buller, P.F. (1986). The team building-task performance relation: Some conceptual and methodological refinements. Group and Organization Studies, 11, 147-168. Buller, P.F. (1988). Long term performance effects of goal-setting and team building interventions in an underground silver mine. Organization Development Journal, 6, 82-93. Buller, P.F. & Bell, C.H. (1986). Effects of team building and goalsetting on productivity: A field experiment. Academy of Management Journal, 29, 305-328. Campbell, J.P. & Dunnette, M.D. (1968). Effectiveness of T-group experience in managerial training and development. Psychological Bulletin, 70, 73-10. Campbell, D.T. & Stanley, J. C. (1966). Experimental and quasiexperimental designs for research. Chicago: Rand McNall. Cannon-Bowers, J.A., Salas, E:, & Converse, S.A. (1990). Cognitive psychology and team traming: Training shared mental models of complex systems. Human Factors Society Bulletin, 33(12), 1-4. Cohen, M.H. & Ross, M.E. (1982). Team building: A strategy for unit cohesiveness. The Journal of Nursing Administration, 12, 29-34. Cook, T.D. & Campbell, D.T. (1979). Quasi-experimentation: Design and analysis issues forfield studies. Chicago: Rand McNally. Dailey, R.C. (1980). A path-analysis of R & D team coordination and performance. Decision Sciences, 21, 357-369. DeMeuse, K.P. & Liebowitz, S.J. (1981). An empirical analysis of team-building research. Group and Organization Studies, 6, 357-378. Driskell, J.E., Hogan, R., & Salas, E. (1987). Personality and group performance. Review of Personality and Social Psychology, 9, 9 1-112. Dyer, W. G . (1987). Team building: Issues and alternatives (2nd ed.). Reading, MA: Addison-Wesley. Eden, D. (1985). Team development: A true field experiment at three levels of rigor. Journal of Applied Psychology, 70, 9410. Eden, D. (1 986). Team development: Quasi-experimental confirmation among combat companies. Group and Organization Studies, 11, 133-146.

Team Building

149

Fiedler, F., Bell, C.H., Chemers, M . M . , & Patrick, D. (1984). Increasing mine productivity and safety through management training and organization development: A comparative study. Basic and Applied Social Psychology, 5 , 1-18. Fleming, D.C. & Fleming, E.R. (1983). Consultation with multidisciplinary teams: A program of development and improvement of team functioning. Journal of School Psychology, 21, 367376. French, W.L. & Bell, C.H. (1984). Organization development:

Behavioral science interventionsfor organization improvement (3rd ed.). Englewood Cliffs, NJ: Prentice-Hall.

Gear, T.E., Marsh, N.R., & Sergent, P. (1985). Semi-automated feedback and team behavior. Human Relations, 38, 707-721. Gersick, C.J.G. (1988). Time and transition in work teams: Towards a new model of group development. Academy of Management Review, 31, 9-41. Gladstein, D. (1984). Groups in context: A model of task group effectiveness. Administrative Science Quarterly, 29, 499-5 17. Glickman, A.S., Zimmer, S . , Montero, R.C., Guerette, P. J., Campbell, W.S., Morgan, B.B. & Salas, E. (1987). The evolution of teamwork skills: An empirical assessment with implications f o r training (Technical Report 87-016). Orlando, FL: Naval Training Systems Center. Goodman, P.S. (1986). (Ed.). Designing efSective work groups. San Francisco: Jossey-Bass. Gunderson, E.K. & Ryman, D. (1967). Group homogeneity, compatibility, and accomplishment. (Report No. NNNRU-67-16). San Diego, CA: Navy Medical Neuropsychiatric Research Unit. Guzzo, R.A. (1986). Group decision making and group effectiveness in organizations. In P.S. Goodman (Ed.). Designing eflective work groups (pp. 34-71). San Francisco: Jossey-Bass. Hackman, J.R. (1976). Group influences on individuals. In M.D. Dunnette (Ed.), Handbook of industrial and organizational psychology (pp. 1455-1526). Chicago: Rand McNally. Hackman, J.R. (1983). A normative model of work team effectiveness (Technical Report No. 2). New Haven, CT: Yale University. Hackman, J.R. (1987). The design of work teams. In J. Lorsch (Ed.), Handbook of organizational behavior (pp. 315-342). New York: Prentice-Hall. Hackman, J.R. (1989). Groups that work (and those that don't). San Francisco: Jossey-Bass.

150

Tannenbawn, Beard, and Salas

Halstead, L.S., Rintala, D.H., Kanellos, M., Griffin, B., Higgins, L., Rheinecker, N., Whiteside, W., & Healy, J.E. (1986). The innovative rehabilitation team: An experiment in team building. Archives of Physical Medicine and Rehabilitation, 67, 357-361. Herold, D.M. (1978). Improving the performance effectiveness of groups through a task-contingent selection of intervention strategies. Academy of Management Review, 3, 3 15-325. House, R.J. & Rizzo, J.R. (1972). Role conflict and ambiguity as critical variables in a model of organizational behavior. Organizational Behavior and Human Performance, 7 , 467505. Hughes, R.L., Rosenbach, W.E., & Clover, W.H. (1983). Team development in an intact, ongoing work group: A quasi-field experiment. Group and Organization Studies, 8 , 161186. Kabanoff, B. & O'Brien, G.E. (1979a). The effects of task type and cooperation upon group products and performance. Organizational Behavior and Human Peformance, 23, 163-181. Kabanoff, B. & O'Brien, G.E. (1979b). Cooperation structure and the relationships of leader and member ability to group performance. Journal of Applied Psychology, 64, 526-532. Kaplan, R.E. (1979). The conspicuous absence of evidence that process consultation enhances task performance. Journal of Applied Behavioral Science, 15, 346-360. Kaplan, R.E., Lombardo, M.M., & Mazique, M.S. (1985). A mirror for managers: Using simulation to develop management teams. Journal of Applied Behavioral Science, 21, 241253. Katz, R. (1982). The effects of group longevity on project communication and performance. Administrative Science Quarterly, 27, 81- 104. Katz, D. & Kahn, R.L. (1978). The socialpsychology of organizations (2nd ed.). New York: Wiley. Kellerman, H. (1981). The deep structures of group cohesion @. 221). In H. Kellerman, (Ed.). Group cohesion: liraeoretical and clinical perspectives. New York: Grune & Stratton. Kirkpatrick, D.L. (1976). Evaluation of training. In R.L. Craig (Ed.). Training and development handbook, (2nd ed.). New York: McGraw-Hill.

Team Building

151

Lassiter, D.L., Vaughn, J.S., Smaltz, V.E., Morgan, B.B., & Salas, E. (1990). A comparison of two types of training interventions on team communication performance. Paper presented at the meeting of the Human Factors Society, Orlando, FL. Latham, G.P. & Yukl, G.A. (1975). A review of research on the application of goal-setting in organizations. Academy of Management Journal, 18, 824-845. Lawler, E.E. (198 1). Pay and organization development. Reading, Mass: Addison Wesley. Liebowitz, S.J. & DeMeuse, K.P. (1982). The application of team building. Human Relations, 35, 1-18. Margerison, C., Davies, R. & McCann, D. (1987). High flying management development. Training and Development Journal, 41, 38-41. Margerison, C., McCann, D. & Davies, R. (1988). Air-crew team management development. Journal of Management Development, 7, 41-54. Marx, R.D. (1982). Relapse prevention for managerial training: A model for maintenance of behavior change. Academy of Management Review, 7, 433-441. McGrath, J.E. (1984). Groups: interaction and performance. Englewood Cliffs, NJ: Prentice Hall. McRae, A. V. (1966), Interaction content and team effectiveness (HumRRO-TR-66-10, AD-637 31 1). Alexandria, VA: George Washington University, Human Resources Research Office. Mitchell, R. (1986). Team building by disclosure of internal frames of reference. Journal of Applied Behavioral Science, 22, 1528. Morgan, B.B., Glickman, A.S., Woodard, E.A., Blaiwes, A.S., & Salas, E. (1986). Measurement of Team Behaviors in a Navy Environment. (Technical Report No. NTSC TR-86-014). Orlando, FL: Naval Training Systems Center. Morrison, P. & Sturges, J. (1980). Evaluation of organization development in a large state government organization. Group and Organization Studies, 5 , 48-64. Murrell, K.L. & Valsan, E.H. (1985). A team-building workshop as an OD intervention i n Egypt. Leadership and Organization Development Journal, 6, 11-16. Naylor, J.C. & Dickinson, T.L. (1969). a s k structure, work structure, and team performance. Journal of Applied Psychology, 53, 167-177.

152

Tannenbaum, Beard, and Salas

Nicholas, J.M. (1982). The comparative impact of organization development inter ventions on hard criteria measures. Academy of Management Review, 7, 531-542. Nieva, V.F., Fleishman, E.A., & Rieck, A. (1978). Team dimen-

sions: Their identity, their measurement and their relationships

(Contract No. DAHC 19-78-C-0001). Washington, D.C.: Advanced Research Resources Organization. Parsons, K.L. (1981). The effects of load sharing system training upon team performance (Doctoral dissertation, Michigan State University, 1980). Dissertation Abstracts International, 42 (6A), 2876-2877. Paul, C.F. & Gross, A.C. (1981). Increasing productivity and morale in a municipality: Effects of organization development. Journal of Applied Behavioral Science, 17, 59-78. Porras, J.I. & Berg, P.O. (1978). Impact of organization development. Academy of Management Review, 3, 249-266. Porras, J.I. & Wilkins, A. (1980). Organization development in a large system: An empirical assessment. Journal of Applied Behavioral Science, 16, 506-534. Pritchard, R.D., Jones, S . , Roth, P., Stuebing, K., & Ekeberg, S. (1988). Effects of group feedback, goal-setting, and incentives on organizational productivity. Journal of Applied Psychology, 73, 337-358. Salas, E . , Dickinson, T.L., Converse, S.A., & Tannenbaum, S.I. (in press). Toward an understanding of team performance and training. In R.W. Swezey & E. Salas (Eds.) Teams: Their training and performance. Norwood, NJ: Ablex. Schein, E.H. (1969). Process consultation: Its role in organization development . Reading , MA : Addison-Wesley. Steiner, I.D. (1972). Group process and productivity. Orlando: Academic Press. Sundstrom, E., Perkins, M., George, J., Futrell, D. & Hoffman, D. A. ( 1990). Wrk-teum context, development, and efectiveness in a manufacturing organization. Presented at the Fifth Annual Conference of the Society for Industrial and Organizational Psychology, April, Miami. Swezey, R.W. & Salas, E. (in press). Guidelines for use in team training development. In R.W. Swezey & E. Salas (Eds.), Teams: Their training and Performance. Norwood, NJ: Ablex. Tannenbaum, S.I., Dickinson, T.L., Salas, E. & Converse, S.A. (1990). A metu-analysis of teum performance and team training. Unpublished Manuscript, State University of New York at Albany.

Team Building

153

Terpstra, D.E. (198 1). Relationship between methodological rigor and reported outcomes in OD evaluation research. Journal of Applied Psychology, 66, 54 1-543. Tushman, M.L. (1979). Work characteristics and subunit communication structure: A contingency analysis. Administrative Science Quarterly, 24, 82-97. Van Maanen, J. (1975). Police socialization: A longitudinal examination of job attitudes in an urban police department. Administrative Science Quarterly, 20, 207-228. Woodman, R.W. & Sherwood, J.J. (1980a). The role of team development in organizational effectiveness: A critical review. Psychological Bulletin, 88, 166-186. Woodman, R.W. & Sherwood, J.J. (1980b). Effects of team development intervention: A field experiment. Journal of Applied Behavioral Science, 16, 2 11-227. Yetton, P.W. & Bottger, P.C. (1982). Individual versus group problem solving: An empirical test of a best-member strategy. Organizational Behavior and Human Pe formance, 29, 307321.

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial1 Organizational Psychology - K. Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

155

CHAPTER 6 PERCEIVED SELF-COMPETENCE AND ORGANIZATIONAL BEHAVIOR Kevin J. Williams and John R. Lillibridge

INTRODUCTION Current cognitive theories of motivation and action stress the importance of self-referent thoughts, beliefs, and expectations in determining human behavior (Bandura, 1986; Carver & Scheier, 1981; Locke & Latham, 1990; Rotter, 1966; Weiner, 1986). A prominent component of these theories is the perceived contingency between one's abilities and various outcomes (Markus, Cross, & Wurf, 1990). In general, these theories hypothesize that individuals will be highly motivated and perform better on tasks in which they perceive themselves to be competent. Empirical evidence has established links between perceived self-competence and motivation and performance across a wide variety of settings, including academic (Dweck & Leggett, 1988; McCombs, 1986; Schunk, 1984), athletic (Feltz, 1982; Lee, 1982; Weinberg, Gould, &Jackson, 1979), interpersonal (Holahan & Holahan, 1987; Huber & Neale, 1984), and work settings (Barling & Beattie, 1983; Gist, 1987). Within the field of organizational psychology, researchers have recently expressed considerable interest in the effects of self-perceived competence. Self-competence and related constructs, such as self-efficacy and self-esteem, have become more increasingly prominent in organizational research and theory (e.g., Brief & Aldag, 1981; Brockner, 1988; Gist, 1987; Lee & Gillen, 1989; Locke & Latham, 1990; Wagner & Morse, 1975; Williams, Williams, & Ryer, 1990). For example, a computer search of Psychological Abstracts from 1983 to 1990 identified 48 studies examining selfefficacy and work behavior (757 citations were found for self-efficacy), compared to no studies for the previous eight-year period (71 citations were found for self-efficacy during this period). Given this

156

Williams and Lillibridge

surge in research activity, it is useful to consider what is known about the antecedents and consequences of self-perceived competence. Surprisingly, few attempts have been made to integrate research in this area (see Gist, 1987, however, for a recent attempt). Progress has been hindered in part by the fact that perceived selfcompetence has been operationalized in different, often quite disparate, ways by researchers. In addition, most researchers have examined only a narrow range of behavior related to self-competence, making it difficult to integrate or generalize findings across studies. The intent of this chapter is to review research on the antecedents and consequences of perceived self-competence in organizations and to provide a framework for interpreting previous findings and guiding future research. We will also consider recent theoretical developments related to perceived self-competence.

CONCEPTUALIZATION OF PERCEIVED SELF-COMPETENCE Before reviewing research on perceived self-competence, it is necessary to define what we mean by self-competence, and place it within the broader self-evaluation system. In addition, it is also important to distinguish between perceived self-competence and related, but distinct, constructs such as self-esteem and outcome expectancy. Perceived self-competence is part of a global self-evaluation system, which typically has been divided into three components: perceived competence or efficacy, personal control, and social acceptance (Ford, 1985; Litt, 1988; McCombs, 1986). Perceived self-competence is defined as an individual's subjective evaluation of task-related ability, a self-appraisal of what one can do in a specific situation (Ford, 1985; McCombs, 1986; McEnrue, 1984). As such, it represents the psychological component of competence and is related to other components of competence such as one's capabilities for producing goal-directed activity and the effectiveness of one's behavior (Ford, 1985). Perceived self-competence can be considered a performance expectancy; persons who perceive themselves as competent in a certain area are likely to expect that they can achieve a high level of performance, or that greater effort on their part will result in greater performance. From this perspective, selfcompetence encompasses the construct of self-efficacy (Bandura, 1977, 1986). Perceptions of self-efficacy relate to the belief in one's capability to mobilize the motivation, cognitive resources, and behavioral repertoire necessary to control events in one's life (Wood

Perceived Self- Competence

157

& Bandura, 1989). Perceptions of self-efficacy and competence relate to a sense that "I can do it" and "I am effective" (Markus et al., 1990; Williams & Lillibridge, 1990). Two important aspects of perceived self-competenceare that it is situation-specific and malleable. Perceived self-competence is a task-specific construct related to one's assessment of level of expertise on a specific task or in a specific setting. Accordingly, it is influenced by the joint effects of situational and dispositional factors. Past accomplishments exert considerable influence on competence beliefs, but these beliefs can be changed by present and future experiences and by interventions aimed at changing one's self-belief structure. Perceived personal control, the second component of the selfevaluation system, refers to the belief in one's own ability to alter events and regulate one's environment significantly (Burger, 1989; Greenberger & Strasser, 1986). Beliefs about the ability to control the behavior of others towards oneself are especially important in determining perceived personal control (Litt, 1988). In general, situations are controllable when one perceives a strong covariation between action and various outcomes. In addition, perceived control, like perceived self-competence, is situationally specific and changeable. Perceived control and perceived self-competence may mutually influence one another (McCombs, 1986): perceived control over one's environment may increase one's sense of competence, and vice versa. Despite the overlap between perceived control and perceived self-competence, however, the two constructs should be treated as distinct. Control beliefs specifically relate to expectations about the possibility that one's actions bring about desired outcomes (i.e., an outcome expectancy), while competency beliefs relate to expectations about the extent to which one possesses the potential means to attain effectively the desired level of performance (i.e., a behavioral expectation). Expectancies about how one's performance will affect outcomes (i.e., "are desired outcomes controllable") are differentiated from self-appraisals of what one can accomplish (i.e., "can I achieve the necessary level of performance?"; McCombs, 1986). In addition, different patterns of self-perceptions of control and competence may exist within people and these patterns may have important implications for organizational behavior (Ford, 1985; Wood & Bandura, 1989). Individuals who have self-perceptions of high competence and personal control perceive themselves as effective, controlling agents (Ford, 1985). However, individuals who perceive themselves as competent in an area, yet do not have perceptions of personal control are more likely to think of themselves as potentially

158

Williams and Lillibridge

effective but powerless in the specific situation because control resides in others (Ford, 1985). As a result, perceptions that the environment is controllable may lead people to exercise their personal efficacy strongly, while perceptions that the environment is uncontrollable may lead people to exercise their personal efficacy weakly (Bandura, 1986; Wood & Bandura, 1989). Social acceptance, the third component of the self-evaluation system, refers to the evaluation of self by others. Reflected appraisals from others reduce uncertainty about one's performance or specific role behaviors (Williams et al., 1990), and also provide a means of testing or verifying self-perceptions. Approval or acceptance by others in most circumstances boosts self-evaluations. Self-competence beliefs can also be distinguished from selfesteem and outcome expectancy. Like self-competence, interest in self-esteem has recently increased in the work motivation area (Brockner, 1988; Pierce, Gardner, Cummings, & Dunham, 1989). Self-esteem is related to the self-evaluation process and to our definition of perceived self-competence, but differs in that it reflects global rather than specific self-evaluation. It is perhaps best to think of self-esteem as an individual's attitude concerning approval or disapproval of himself or herself (Pierce et al., 1989). It indicates the extent to which persons believe themselves to be generally effective, reflecting a personal judgment of worthiness. Self-esteem is partially dependent on perceptions of self-competence, and, in fact, research examining self-esteem at work parallels research on selfcompetence at work. Pierce et al. (1989) have studied organizationbased self-esteem, which they define as a belief that one is important, meaningful, effectual, and worthwhile within the organization. They view self-efficacy perceptions across a variety of organizational tasks as contributing to organization-based self-esteem. As such, Pierce et al. I s (1989) construct of organization-based self-esteem may be considered a more global or superordinate measure of perceived-self competence. Accordingly, recent work on the correlates of organization-based self-esteem has been included in the literature review that follows. A word of caution is needed, however. Although self-esteem may be highly dependent on perceptions of self-competence, it is possible for individuals to feel competent in an area, but still not like themselves (Brockner, 1988). Thus, selfesteem and self-competence should not be treated as interchangeable constructs. Outcome expectancy refers to the perceived likelihood of a desired outcome occurring (Carver & Scheier, 1981), and is thus close in meaning to perceived personal control. Motivation to act is likely to be higher if individuals perceive that the desired level of perform-

Perceived Self- Competence

159

ance produces positive outcomes. For example, individuals are more likely to diet, exercise or adopt other health-related changes in lifestyle if they see that health-risk is reduced as a result of their behavior. Perceived self-competence is one determinant of outcome expectancy, but so are environmental impediments (e.g., time constraints) and prior experience or knowledge of behavioral outcomes (Carver & Scheier, 1981). As mentioned earlier, perceived selfcompetence represents a behavioral expectancy (i. e., effort-performance) and its effects on behavior may be independent of outcome expectancy (Barling & Beattie, 1983). To summarize, perceived self-competence refers to self-appraisal of one's performance capabilities in a specific situation, and together with perceptions of personal control and social acceptance comprises the self-evaluation system. Although perceived self-competence contributes to one's self-esteem, and interacts with outcome expectancies to influence behavior (Carver & Scheier, 1981), it should be kept distinct from these constructs. In the following sections we turn our attention to identifying the critical outcomes and antecedents of self-competence.

ORGANIZATIONAL OUTCOMES OF PERCEIVED SELF-COMPETENCE In this section we summarize the empirical literature on perceived self-competence and organizational behavior. Our review focuses on major outcome variables -- performance, job attitudes, and psychological well-being -- and mediating or intervening mechanisms responsible for the effects of perceived self-competence on these outcomes. The variables identified as intervening and outcome variables are depicted in Figure 1. It should be noted that much of the research on self-efficacy in organizational settings has been correlational in nature; only a few studies have employed longitudinal or experimental designs capable of providing evidence of causality. Much of the causal evidence that does exist comes from laboratory studies. Evidence of a causal relationship between variables is depicted in Figure 1 by filled arrows.

Job Perj4omance Self-efficacy and self-competencebeliefs have been related to job performance in a variety of settings. In a study of sales representatives, Barling and Beattie (1983) found that self-efficacy beliefs were correlated with objective measures of sales performance (i.e., poli-

Williams and Lillibridge

160

Perceived Self -Competence

MEDIATORS

. . . . .

Persistence Attributions Goal Difficulty Performance Strategies Coping Strategies

Job Attitudes commitment satisfaction

Figure I : Consequences of perceived self-competence.

Perceived Sey- Competence

161

cies sold, calls per week, commissions). Outcome expectancies, on the other hand, were not related to sales performance, leading Barling and Beattie (1983) to conclude that self-efficacy beliefs were more important in predicting sales performance than were beliefs about rewards for success. The finding that self-efficacy predicts sales performance has been replicated by Lee and Gillen (1989). 'bylor, Locke, Lee and Gist (1984) found that beliefs about personal efficacy in research significantly predicted research productivity in a sample of university professors, and Huber and Neale (1984) found that negotiators' self-beliefs about their interpersonal skills were positively related to negotiation outcome. Perceptions of personal efficacy and competence also help individuals sustain performance in difficult or demanding task environments. Kahn and Long (f988) found that the performance of clerical workers low in perceived self-efficacy was impaired during periods of high stress to a greater extent than those high in self-efficacy. In a laboratory simulation, Freedman and Phillips (1985) found that feelings of self-competence enabled individuals to overcome performance constraints and perform at the same level as individuals not encountering such constraints. Evidence that self-perceived competence and performance are causally linked is provided by a series of simulation studies examining complex decision making (Wood & Bandura, 1989; Wood, Bandura, & Bailey, 1990). In these studies, college students were asked to role-play managers making multiple strategy decisions in a complex environment. A strong belief in one's managerial and problem-solving capabilities was consistently related to efficient analytic thinking and higher quality decisions. Path analysis results revealed that self-efficacy beliefs directly influenced performance (Wood & Bandura, 1989). Self-efficacy had direct effects on decision-making performance, as well as indirect effects through increased personal goal difficulty. These results support earlier laboratory research demonstrating that self-efficacy predicts task performance after controlling for ability and previous performance (Locke, Frederick, Lee, & Bobko, 1984). Wood and Bandura's (1989) simulation studies also indicate the reciprocal nature of the relationship between self-competence and performance. Pre-existing beliefs in personal efficacy increased initial task performance, which in turn further elevated perceptions of personal efficacy. As we will discuss in greater detail later, performance and perceived self-competence are involved in an ongoing, self-reinforcing feedback loop.

162

Williams and Lillibridge

Job Attitudes High levels of organizationalcommitmenthave been found among workers high in perceived competence (Mathieu & Zajac, 1990; Moms & Sherman, 1981). In their meta-analysis of the antecedents of organizational commitment, Mathieu and Zajac (1990) reported a mean correlation of .63 between attitudinal measures of organizational commitment and self-competence. They suggested that the self-belief system links an individual to the organization; individuals high i n perceived self-competence may be more committed to the organization than those low in perceived selfcompetence because they believe their performance will result in desired organizational rewards. Although researchers, including Mathieu and Zajac (1990), have assumed self-competence is an antecedent to organizational commitment, causality has not been directly addressed. It is possible that high levels of organizational commitment increase feelings of efficacy. Social identity theory ("hjfel, 1982), for example, would predict that strong identification with a group or organization would raise levels of self-esteem, and perhaps self-efficacy (Britt, 1991). Future research is needed to address the exact nature of the relation between perceived selfcompetence and organizational commitment. Feelings of competence have been identified as the proximal cause of intrinsic task motivation and task satisfaction (Deci & Ryan, 1985; McCombs, 1986; Thomas & Velthouse, 1990). Accordingly, perceived work-competence should be related to job satisfaction in workers. Surprisingly, we uncovered few studies that directly assessed the relation between self-competence and job satisfaction. Bhagat and Allie (1989) found that perceived competence was positivel related to satisfaction with supervisors, coworkers, and work itsell! Williams et al. (1990) found a weak, but statistically significant, positive correlation between job satisfaction and perceived selfcompetence in a sample of practicing school psychologists. In laboratory experiments, however, a stronger relationship between perceived self-competence and task satisfaction has been found (Freedman & Phillips, 1985).

Psychological Wll-being People's beliefs in their capabilities are apt to affect their psychological and emotional adjustment to work situations. Research in social and health psychology reveals that perceived competence is a crucial determinant of emotional well-being and mental health

Perceived Self- Competence

163

(Warr, 1987), and an extensive body of literature exists which examines the relation between self-efficacy and the ability to cope with stressful situations (Bandura, 1986). Perceived inefficacy has been related to arousal, preoccupancy with negative thoughts and personal deficiencies, poor coping strategies, and subjective distress in threatening situations (Bandura, 1989; Feltz, 1982; Folkman & Lazarus, 1980). In addition, efficacy perceptions have been causally linked to such physiological manifestations of anxiety as catecholamine secretion (Bandura, Thylor, Williams, Mefford, & Rarchas, 1985) and immunological system functioning (Wiedenfeld, O'Leary, Bandura, Brown, Levien, & Raska, 1990), suggesting that individuals low in perceived self-efficacy to cope with stressful events are more susceptible to stress-related diseases. Similar conclusions can be drawn from organizational research, where feelings of mastery and high levels of work selfefficacy have been associated with low levels of job stress and anxiety (Kahn & Long, 1988; Karasek & Theorell, 1990). In a recent study (Bhagat & Allie, 1989) perceived competence was negatively related to emotional exhaustion and feelings of depersonalization at work. In addition, perceived competence moderated the relationship between life stress and job stress. Thus, it appears that feelings of competence at work may reduce psychological strain and job stress. These were correlational studies, however, and thus further research is needed that examines the causal relationship between competence and job stress.

Intervening Mechanisms Although some of the research reviewed above demonstrated direct effects of perceived self-competence on outcomes (e.g., Wood & Bandura, 1989), most theorists have identified intervening processes responsible for the effects of perceived competence. The effects of efficacy beliefs can be seen as influencing performance mainly through motivational processes; that is, they influence the initiation, direction, intensity, and persistence of behavior. Eflort Expenditure and Persistence. A major proposition of Bandura' s self-efficacy theory is that perceptions of personal efficacy will be causally related to effort. According to Bandura (1980), people with a strong sense of self-efficacy will pursue goals with vigor and persistence. Indeed, laboratory results have shown that greater effort and greater persistence occurs when individuals have high as opposed to low levels of perceived self-competence (Bandura & Cervone, 1983, 1986). Bandura and Cervone (1983), for example,

164

Williams and Lillibridge

found that subjects responded with greater effort after failure on a laboratory task when they had high rather than low feelings of selfcompetence. Similarly, Sandelands, Brockner, and Glynn (1988) found that subjects high in self-esteem persisted on difficult tasks longer than subjects low in self-esteem, especially when persistence was identified as a wise strategy. Although experimental evidence from the laboratory supports the relation between perceived self-competence and effort and persistence, few researchers have directly examined this relation in organizational settings. Gibson and Dembo (1984) found that teachers high in instructional self-efficacy spent more time and effort helping failing students than did those low in instructional self-efficacy. Presumably, teachers high in instructional self-efficacy believed that they could, through persistence, raise students' grades. Further evidence that perceptions of work ability and effort may be linked is provided by Williams, Suls, Alliger, Learner, and Wan (in press). Experience sampling methodology (Larsen & Csikszentmihalyi, 1983) was used in this study to measure employed mothers' mood states and perceptions of ability and effort as they performed work and family tasks throughout the day. Perceived ability was significantly related to self-reported effort and mood in both work and family settings. These studies provide initial field support for the laboratory findings that perceived self-competence and effort are positively related.

Attributions for Perfomance. The attributions one makes for his or her performance may also mediate the effects of self-competence on work outcomes. Of crucial concern may be the attributions one makes following failure. Individuals low in perceived self-competence may attribute failure to stable internal causes, i.e., to their lack of ability. As a result, these individuals are likely to reduce their effort in the future or, if possible, withdraw from the performance situation. Individuals high in self-competence, on the other hand, are likely to adopt a mastery orientation to difficult situations, and attribute initial failure to effort rather than ability (Dweck & Leggett, 1988). Attributing performance to lack of effort rather than lack of ability is likely to increase task persistence and ultimately one's chances of success. In general, attributional styles that underutilize stable, internal attributions for success and overutilize such attributions for failure will reduce assessments of self-competence (Thomas & Velthouse, 1990). Goal Setting. Self-efficacy may also increase performance through its effects on personal goal setting. Individuals perform better when

Perceived Self-Competence

165

they are committed to specific, challenging goals than nonspecific or easy goals (Locke et al., 1981; Locke & Latham, 1990) and research suggests that p p l e high in self-efficacy set more difficult goals for themselves and are more committed to these goals than those low in self-efficacy (Bandura & Cervone, 1983; Lee, 1988; Locke et al., 1984; Wood & Bandura, 1989). Britt (1991), for instance, found a strong correlation (r= .61) between perceived self-competence and difficulty of self-set goals for individuals working on an novel laboratory task. Locke et al. (1984) examined the joint effects of ability, self-efficacy, goals, and task strategies on task performance. Selfefficacy was found to affect goal level, task performance, and commitment to self-set goals. Subjects with a strong sense of personal efficacy set higher goals, were more committed to these goals, and performed better than subjects with a weak sense of personal efficacy. In addition, self-efficacy significantly predicted future performance even when past performance and ability were controlled, with the level of prediction better for increasing levels of goal difficulty. Wood and Bandura's (1989) research on complex managerial decision making also suggests that self-efficacy influences goal setting. In addition to direct effects on performance, they found that self-efficacy indirectly influenced decision-making performance through increases in difficulty of personal goals.

Coping Strategies. Perceptions of personal competence not only affect one's ability to perform a task but also are likely to affect ability to cope with task demands. Two common strategies for dealing with environmental demands are problem-focused and emotion-focused coping (Folkman & Lazarus, 1980; Latack, 1986); the former is characterized by direct attempts to solve problems and coordinate activities, while the latter is characterized by a greater concern with controlling feelings and self-perceptions. Problemfocused coping has been found to be more effective than emotionfocused coping in dealing with a variety of work demands (Parkes, 1990), and research exists to suggest that individuals high in perceived competence are more likely to adopt problem-focused coping strategies. People who judge themselves unable to cope with environmental demands tend to become more self-diagnostic rather than task-diagnostic (Wood & Bandura, 1989). Self-diagnosis tends to focus on personal deficiencies and diverts cognitive resources from the task at hand. Accordingly, low efficacious people should be more likely to adopt dysfunctional emotion-focused coping strategies. Stumpf, Brief, and Hartman (1987) found support for this hypothesis: feelings of competence and mastery reduced emotionfocused coping for applicants preparing for job interviews. Feelings

Williams and Lillibridge

166

of competence may lower the need to regulate one's emotions and increase the likelihood of adopting problem-focused coping strategies (Stumpf et al., 1987; Wood & Bandura, 1989).

Summary Strong support exists for the hypothesis that perceived selfcompetence is related to positive work outcomes. Competence beliefs are positively related to job performance and attitudes, and negatively related to job stress. However, evidence of causal relationships among these variables in organizational settings is scarce. Data from simulation and laboratory studies are highly suggestive of the hypothesis that self-competence influences work performance through its effects on effort, personal goals, and type of coping strategy. Future research is needed that firmly establishes these links in organizational settings. In addition, research is also needed that examines causal relationships between self-competence and job attitudes.

ANTECEDENTS AND DEVELOPMENT OF PERCEIVED COMPETENCE IN ORGANIZATIONS In this section we address the question of how perceptions of self-competence at work can be strengthened. This question has received little research attention relative to research examining correlates and outcomes of self-efficacy. We agree with Ford (1985), however, who argues personal agency and competence must be viewed from a developmental perspective: self-referent thoughts and beliefs reflect an evolving theory of the self as we confront new challenges and obstacles. It is important for organizational psychologists to consider work and job characteristics that may enhance or decrease one's belief in personal efficacy. Individuals develop skills and competence through interactions with their environments. Accordingly, the social environment contains multiple sources of information regarding one's capabilities. The most extensive analysis of sources of self-competence information is Bandura's (1986) work on sources of self-efficacy. Bandura (1977, 1986) has posited that individuals develop feelings of efficacy through enactive and vicarious experiences of success, and through persuasory opinions of others and self-interpretation of arousal. Beliefs in personal efficacy are increased when one experiences success on relevant tasks (enactive mastery), observes successful performance by similar others (modeling), receives positive efficacy

Perceived SeEf-Competence

167

appraisals from credible observers (social persuasion), and construes somatic information as natural reactions to performance situations or evidence of capability rather than personal inadequacies. The first three of Bandura's sources (i.e., enactive mastery, modeling, and social persuasion) can be controlled directly by organizations, and work environments that provide these sources of efficacy information may be capable of raising workers' perceptions of competence. Figure 2 outlines potential ways in which organizations can provide workers with efficacy information.

Empowerment Enactive mastery is considered the most effective and powerful source of self-efficacy beliefs (Bandura, 1986; Stumpf et al., 1987). Thus, the most effective organizational interventions in terms of raising workers' feelings of self-competence may be those that allow workers to learn new skills and master challenging assignments. One way in which organizations can do this is through empowering workers. Although it has been defined in various ways by researchers, empowerment is an increasingly popular construct used by theorists to explain motivational processes and organizational effectiveness (Conger & Kanungo, 1988; Thomas & Velthouse, 1990). Conger and Kanungo's (1988) definition of empowerment as a motivational process in workers is particularly useful for the purpose of this discussion. They explicitly define empowerment in selfefficacy terms, stating that the empowerment process enhances workers' self-efficacy beliefs. Accordingly, efforts to empower workers can be seen as increasing workers self-competencebeliefs; that is, interventions that empower workers increase their perceptions that they can perform competently on the job (Thomas & Velthouse, 1990). Many current organizational development and managerial principles can be seen as strategies for empowering workers. Job enrichment, goal setting, and participative decision making may positively influence motivation through their effects on feelings of competence.

Job Enrichment An underlying objective of psychological approaches to job enrichment is to increase intrinsic motivation in workers. Although different techniques have been used to accomplish this objective, all involve raising workers' self-perceptions of mastery or competence. The notions of enactive mastery and perceived competence are prominent, although often overlooked, concepts in Hackman and

Williams and Lillibridge

168

Antecedents

goal setting job enrichment €eedback systems Competence 1

I

Leader Style and Expectations 1

I

r

I

1

I I

Organizational Controllability

Figure 2: Antecedents of perceived self-competence.

Perceived Self-Competence

169

Oldham's (1980) job characteristics model of job design. Their model postulates that workers obtain intrinsic rewards when they learn (knowledge of results) that they personally (experienced responsibility) have performed well (my italics) on a task they care about (experienced meaningfulness; Hackman, 1983). The focus of much of the research on the JCM has been on the experienced psychological states of meaningfulness, responsibility, and knowledge of results. Less attention has been paid to how competent workers feel when their jobs have been redesigned according to the model's principles. It is clear from the above quote, however, that perceptions of self-competence are necessary for intrinsic rewards to be obtained. Providing workers with skill variety and autonomy, for example, will increase intrinsic motivation to the extent that workers experience a sense of personal accomplishment and competence. A recent approach to job design that is consistent with the notion that enriching one's job increases feelings of efficacy is Karasek and Theorell's (1990) decision latitude model. According to this model, decision latitude, operationalized as high personal control and skill utilization at work, relates to psychological well-being and job performance. The amount of decision latitude experienced by workers interacts with the job's psychological demands (i.e. , workload, task complexity, and required concentration levels) to determine healthy or unhealthy outcomes. Jobs that are high i n both psychological demand and decision latitude foster learning of new skills, or what Karasek and Theorell (1990) refer to as "active learning." Longterm development of feelings of mastery is stimulated as active learning accumulates. Coincidentally, feelings of mastery inhibit anxiety perceptions, thus facilitating effective coping and likelihood of further learning. Karasek and Theorell's (1990) model would predict that workers in high demand-high discretion jobs should have higher levels of perceived self-competence than workers in other jobs. To date, no direct tests of this hypothesis have been conducted; future research is needed to see whether or not job discretion and job demands interact to increase feelings of competence.

Goal Setting Realistic challenging goals also provide the opportunity for enactive mastery. The motivating potential of specific, challenging goals has been widely established (Locke & Latham, 1990), and much of the current research on goal theory focuses on the cognitive processes responsible for the goal-performance relationship. Empirical support is growing for the hypothesis that goal setting has its main effects on motivation through an individual's sense of efficacy

170

Williams and Lillibridge

or feeling that he or she can function at a specific level of performance (Evans, 1986; Locke & Latham, 1990). Lee (1988) examined the effects of goal setting and a bonus system on performance in a sample of telephone solicitors. Performance goals were established based on previous performance levels and bonuses given to workers who exceeded these goals. Self-efficacy levels increased over the course of the study as assigned goals were met, suggesting that goal accomplishment results in increased self-efficacy. An assigned goal apparently affects one's confidence in his or her ability to meet the goal, with difficult goals increasing one's confidence to a greater extent than easy goals (Locke et al., 1984; Locke & Latham, 1990; Wood & Bandura, 1990).

Pygmalion Effect Communicating high performance expectations to workers is also likely to raise competency beliefs. The Pygmalion effect refers to situations in which holding high expectations of others evokes behavior that confirms these expectations. Education researchers have found evidence of such self-fulling prophecies in the classroom (Rosenthal & Jacobson, 1968), where increasing teachers' expectations of children's learning potential has boosted student performance. In work settings, raising managers' expectations of subordinates has resulted in subsequent increases in subordinate performance (Crawford, Thomas, & Fink, 1980; Eden, 1990). The Pygmalion effect may operate in a number of ways. First, supervisors holding high expectations of subordinates are likely to convey such expectations in persuasive attempts to increase performance, as predicted by Bandura. By expressing high expectations, supervisors signal confidence in workers, which is likely to increase workers' self-confidence. Second, as Eden (1988) has suggested, supervisors holding high expectations of subordinates may set difficult goals which motivate and raise the performance levels of subordinates.

Participative Decision Making Organizations may also raise efficacy beliefs by allowing workers greater participation and involvement in organizational decision making. Potential outcomes of shared decision making include a better understanding of jobs and how they can be performed, more creative ideas, and support for higher goals (Locke & Schweiger, 1979). All of these factors may increase workers' confidence that they can effect change. In support of this hypothesis, Steel and Lloyd (1988) found that participation in quality circles (one form of

Perceived SeljX'ompetence

171

shared decision making) increased workers perceptions of task competence, independent of performance increases. One issue that has generated much debate in the goal theory literature is whether participation in the goal setting process boosts productivity (Locke & Schweiger, 1978; Miller & Monge, 1986). The current conventional wisdom appears to be that participation does not have any independent effect on performance; it is useful only to the extent that more difficult goals are set through the participation process (Locke & Latham, 1990). We suggest, however, that the joint effects of difficult goals and participation may be similar to those brought about by Karasek and Theorell's (1990) active learning jobs. Difficult goals may create psychological demand, which coupled with the ability to make decisions about what strategies to use will increase learning and feelings of enactive mastery. Feelings of mastery, in turn, are likely to have a positive impact on future performance. The long-term effects of participation in goal setting have not been adequately addressed by goal theory researchers.

Pe?$ormance Feedback Performance feedback is a valuable individual resource that influences worker motivation, performance, and job attitudes (Ashford & Cummings, 1983; Ilgen, Fisher, & Taylor, 1979; Locke, Cartledge, & Koeppel, 1968), and thus can be expected to influence perceptions of personal efficacy. Feedback is especially crucial for the development of competence because it is essential for personal control of one's environment (Warr, 1987). In addition, performance feedback is an integral component of enactive and vicarious mastery processes and forms the basis for persuasory opinions of efficacy (Bandura, 1986). Frequent feedback about the effectiveness of one's actions increases self-beliefs that one can execute the desired behavior. Efficacy feedback can be obtained from many sources: reflected appraisals from significant others, comparisons with self-standards or one's own past performance (i.e., self-generated feedback), and comparisons with the performance of others (i.e., social comparison; Herold & Parsons, 1985; Schwalbe, Gecas, & Baxter, 1986; Williams et al., 1990). Self-perceived competence at work is likely to be high to the extent feedback from these sources is perceived to be frequent and positive (Steel, Mento, Davis, & Wilson, 1989). Reflected appraisals are persuasory opinions regarding efficacy (Bandura, 1986) and thus can be expected to have direct effects on perceived self-competence. Gist (1987), however, suggested that self-

172

Williams and Lillibridge

generated feedback may be especially useful for increasing selfefficacy perceptions because it is equivalent to self-guided enactive mastery. That is, self-monitoring processes provide workers with continuous feedback regarding goal progression and attainment. Williams et al. (1990) directly examined the relation between sources of job feedback and self-perceptions of competence among school psychologists. Respondents in this study rated their competence in several job areas (assessment, intervention, counseling, and consultation) and indicated the extent to which they obtained positive and negative feedback from administrators, coworkers, students, parents, and self-assessments. Both quality and type of feedback predicted self-competence: self-perceptions of competence were high when positive reflected appraisals from others were frequent and low when negative feedback from self and coworkers was frequent. Stepwise regression analysis revealed that the best predictors of perceived self-competence were infrequent negative self-feedback and frequent positive feedback others (i.e., parents, teachers, and administrators). Unfortunately, this study was correlational in nature and thus causal effects cannot be ascertained. The availability of different types of feedback may increase perceptions of selfcompetence, but the opposite causal sequence may also be true: competent individuals may receive more feedback. Nonetheless, this study has a number of implications for raising workers' self-efficacy or competence perceptions. First, it is consistent with other studies indicating a strong relation between performance feedback and selfefficacy. Second, it supports Gist's (1987) suggestion that selffeedback is important for self-competenceperceptions, in addition to feedback from others. Many occupations are similar to that of school psychologist, in which supervisory feedback is not readily available. Designing self-appraisal and accountability programs may be especially useful in raising self-efficacy in such occupations (Williams & Williams, in press). Along these lines, Lane and Herriot (1990) found that managers' self-ratings of performance predicted subsequent performance. They suggested that positive self-ratings raise self-efficacy beliefs, which, in turn, increase performance.

Organizational Controllability Perceived self-competence may also be affected by organizational constraints on individual behavior and personal control (Wood & Bandura, 1989). As we mentioned earlier, individuals may be more likely to exercise their efficacy when they perceive their environment as changeable and controllable. Indeed, Wood and Bandura

Perceived SeEf-Competence

173

(1989) found that subjects displayed higher levels of self-efficacy, set higher goals, and made more effective decisions when they felt that they could control organizational events and outcomes than when did not feel outcomes were under their control. Wood and Bandura (1989) suggest that organizational controllability is best increased by (1) increasing the opportunity for individual discretion and action, (2) relating even small changes by workers to major outcomes, such as profit or productivity, and (3) increasing the predictability of organizational events and work behaviors.

Leadership Finally, leadership styles can be expected to affect followers' perceptions of self-competence (Gist, 1987). Inspirational or charismatic leaders, for example, have been found to increase personal goals and feelings of competence in followers (House & Baetz, 1979; Thomas & Velthouse, 1990). By presenting a strong idealized vision of future goals and successes, these leaders are likely to mobilize followers to action and increase behavioral and outcome expectancies (Conger & Kanungo, 1987). Similary, leaders can increase self-competence by stressing the importance of individual and organizational learning (Bennis & Nanus, 1985; YuM, 1989). Leaders who provide coaching and social support, promote skillsdevelopment seminars, reward innovation, and tolerate or accept mistakes as part of the learning process are likely to have more motivated and self-efficacious followers.

Summary A number of interventions have been indentified that are capable of increasing perceived self-competence. What has not been adequately addressed, however, is how these interventions change selfperceptions of ability. Markus et al. (1990) have suggested that a cognitive representation of competence is necessary before an individual can express and display competence. This cognitive representation includes not only an understanding of task-relevant abilities, but also knowledge of strategies for employing these abilities. From this perspective, job enrichment, goal setting, pygmalion effects, feedback, and charismatic or leadership styles can all be seen as strengthening one's understanding of how they can use their skills and abilities to contribute to success in organizations. Furthermore, these interventions may enable individuals to simulate anticipated performance, thereby bridging the gap between current performance and desired outcomes (Markus et al., 1990).

Williams and Lillibridge

174

RECENT DEVELOPMENTS Much of the research discussed up to this point has focused on self-competence in a specific setting. In contrast, recent research has begun to look at self-perceptions of competence across different settings or domains. In this section we will briefly discuss research on how self-competence beliefs help individuals navigate through their social environments and how different domain-specific perceptions of competence relate to one another.

A Proactive View of Self-competence Researchers have tended to focus on coping or reactive processes in relation to self-competence and self-efficacy. That is, they have mainly examined how individuals behave in existing environments, and how experiences shape individuals. The emphasis on coping is clearly reflected in Bandura's (1982) statement that judgments of efficacy determine "how much effort people will expend and how long they will persist in the face of obstacles or aversive experiences" (p. 123, italics added). This interest in the effects of selfefficacy on coping perhaps reflects the initial concerns of clinical psychologists. Organizational psychologists, however, should also be concerned with proactive behavior, or how individuals select and manipulate their environments to suit their preferences and capabilities (Buss, 1987; Williams & Lillibridge, 1990). A proactive view of self-competence argues that individuals who are high in selfcompetence are better able to mold, create, and choose their environments than those who are low in self-competence. Proactive effects of perceived self-competence at work are most apparent in research on work socialization and feedback-seeking behavior.

Wrk Socialization. Jones (1986) developed and tested a theoretical model of the role of self-efficacy in the work socialization process. He argued that newcomers low rather than high in self-efficacy would be influenced by an organization's socialization tactics, because they are likely to readily accept definitions of situations offered by others. Persons with high levels of self-efficacy, on the other hand, are more likely provide their own interpretations of work situations. Furthermore, Jones (1986) hypothesized that newcomers high in self-efficacy would "take a proactive stance toward performance in order to demonstrate their abilities" @. 267). Results of the study supported this hypothesis and indicated that newcomers high in work self-efficacy were more likely to adopt an innovative role orientation than newcomers low in work self-efficacy. High efficacy

Perceived Self-Competence

175

newcomers were also more likely to attempt to alter work procedures to fit their interests and capabilities. Those low in self-efficacy conformed to definitions offered by others. This study suggests that a high sense of personal efficacy may help individuals mold their environments to suit their preferences and strengths, thereby increasing their chances of success. A self-perpetuating cycle is likely to result whereby high efficacy produces favorable work environments and high chances for success, which reinforce feelings of competence.

Feedback Seeking Behavior. Given that feedback has been identified as a necessary component for the development of competence (Ashford & Cummings, 1983; Warr, 1987), one would expect that all individuals would desire and seek competence information (White, 1959). The degree of feedback seeking engaged in by an individual, however, may be a function of dispositional factors (Ashford & Cummings, 1983; Feinzig, 199l), including perceptions of self-competence. Research in the self-perception literature suggests that performance feedback is seen as threatening to people low in perceived self-competence, while it is ego-enhancing to those high in self-competence (Carver, Antoni, & Scheier, 1985). Thus, it seems logical to hypothesize that persons high in self-competence would engage in more feedback-seeking behaviors than those low in self-competence. Although no studies have examined the role of perceived self-competence in feedback seeking behavior at work, Miller and Jablin (1991) recently presented a model of informationseeking tactics and identified a number of dispositional and situational factors influencing these tactics. They suggested that persons. low in self-esteem are concerned about the risks and costs of obtaining performance information (e.g, poor reflected or self-appraisals), and thus are likely to engage in minimal search for information. In addition, they are likely to use indirect or disguised tactics which may shield them from public scrutiny or embarrassment. Persons high in self-esteem, on the other hand, may be less concerned with the social costs of information seeking and thus more likely use direct and sustained tactics. These effects can also be predicted for perceived self-competence and task-specific, feedback-seeking behavior. If supported by future research, this model would also demonstrate how individuals who feel competent are able to sustain proactively and further develop such feelings,

Williams and Lillibridge

176

nble 1 Intercorrelation$ Between Domain-Specific ComFetency Ratings and Feedback Sources in Universitv Athletes. Competency Domain ~

~

Competency Domain

Social

~~~~

Athletic

Academic

Social

1.o

Athletic

-26'

1.0

Academic

.29'

.29*

1.o

Reflected appraisal

.26*

.24*

.13

Self appraisal

.17

,33*

.30*

Social comparison

.08

.05

.03

Feedback Source

~

Note: * = p

<

.05.

~~

Perceived Sev- Competence

177

Relationships Between Domain-Specijic Perceptions of Competence Another question of interest is the extent to which perceived selfcompetence in one domain relates to perceived competence in another domain. It would be useful to know whether feelings of competence in one area spill over to affect perceptions in another and whether the same or different factors influence perceived competence in different areas. We recently addressed this question in a sample of student-athletes in a large university. Members of intercollegiate athletic teams were asked to rate perceived self-competence in three different domains -- athletics, social relationships, and academics -- and to indicate the extent to which they relied on reflected appraisals from others (e.g., coaches, friends, professors), self-appraisals, or social comparison information when evaluating their ability in each area. Intercorrelations among the selfcompetence ratings and between the feedback sources and competency ratings are presented in Table 1. As can be seen, self-competence in one area was only weakly related to self-competence in another domain. Not surprisingly, ratings were significantly related across domains, but the strength of the correlations are relatively weak. It is interesting to note that different feedback sources were related to self-competency ratings in the different domains. Reflected appraisals were related to competency in social relationships and self-appraisals were related to academic competency ratings. Both self and reflected appraisals were correlated with perceived competence in athletics. These findings have a number of implications for future research. First, they indicate that people have relatively distinct views of their capabilities in different domains. Second, they suggest that different factors may be related to competency beliefs in different domains. Such information may be especially useful in examining the interface between work and nonwork domains, where beliefs about one's competence in different roles may affect psychological adjustment to multiple role occupancy (cf., Williams et al., in press).

GENERAL CONCLUSION The research reviewed in this chapter suggests that the increased interest among organizational psychologists in the role of perceived self-competence is warranted. A host of antecedent and outcome conditions have been identified, enabling us to develop a preliminary or nascent model of self-competence in the work place. It is our hope that this model will be shaped and refined by future research.

Williams and Lillibridge

178

Also, by adopting a proactive perspective, one is able to see how individuals high in perceived self-competence are able to sustain and reinforce their feelings of competence by positioning themselves in their environments so as to ensure success (or make success more probable) and obtain crucial feedback regarding their actions. Organizations can benefit by designing work conditions so as to assist individuals in developing this self-perpetuating competence cycle.

REFERENCES Ashford, S.J., & Cummings, L.L. (1983). Feedback as an individual resource: Personal strategies of creating information. Organizational Behavior and Human Pelgfonnance, 32, 370398. Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological Review, 84, 191-215. Bandura, A. (1980). Gauging the relationship between self-efficacy judgment and action. Cognitive Therapy and Research, 4 , 263-268. Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37, 122-147. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44, 1175-1184. Bandura, A., & Cervone, D. (1983). Self-evaluative and self-efficacy mechanisms governing the motivatiorral effects of goal systems. Journal of Personality and Social Psychology, 45, 1017-1028. Bandura, A., & Cervone, D. (1986). Differential engagement of self-reactive influences in cognitive motivation. Organizational Behavior and Human Decision Processes, 38, 92- 113. Bandura, A., Taylor, C.B., Williams, S.L., Mefford, I.N. ,. & Rarchas, J.D. (1985). Catecholamine secretion as a function of perceived coping self-efficacy. Journal of Consulting and Clinical Psychology, 53, 406-4 14, Barling, J. & Beattie, R. (1983). Self-efficacy beliefs and sales performance. Journal of Organizational Behavior Management, 5 , 41-51. Bennis, W.G., & Nanus, B. (1985). Leaders: The strategies for taking charge. New York: Harper & Row.

Perceived Self-Competence

179

Bhagat, R.S. & Allie, S.M. (1989). Organizational stress, personal life stress, and symptoms of strain: An examination of the moderating role of sense of competence. Journal of Mcationa1 Behavior, 35, 231-253. Brief, A.P., & Aldag, R.J. (1981). The "self" in work organizations: A conceptual review. Academy of Management Review, 6, 75-88. Britt, M.A. (1991). l%e role of social identity theory in self-eflcacy, goal setting, and peflormance. Unpublished doctoral dissertation, The University at Albany, State University of New York, Albany, NY. Brockner, J. (1988). Self-esteem at work: Research, theory, and practice. Lexington, MA: Lexington Books. Burger, J.M. (1989). Negative reactions to increases in perceived personal control. Journal of Personality and Social Psycholoa,56, 246-256. Buss, D. (1987). Selection, evocation, and manipulation. Journal of Personality and Social Psychology, 53, 1214-1221. Carver, C.S., Antoni, M., & Scheier, M.F. (1985). Self-consciousness and self-assessment. Journal of Personality and Social Psychology, 48, 117-124. Carver, C. S., & Scheier, M. F. (198 1). Attention and self-regulation: A control-theory approach to human behavior. New York: Springer-Verlag. Conger, J.A., & Kanungo, R. (1987). Toward a behavioral theory of chansmatic leadership in organizational settings. Academy of Management Review, 12, 637-647. Conger, J.A., & Kanungo, R. (1988). The empowerment process: integrating theory and practice. Academy of Management Review, 13, 471-479. Crawford, K.S., Thomas, E.D., & Fink, J.J. (1980). Pygmalion at sea: Improving the work effectiveness of low performers. Journal of Applied Behavioral Science, 16, 482-505. Deci, E.L., & Ryan, R.M. (1985). Intrinsic motivation and selfdetermination in human behavior. New York: Plenum. Dweck, C.S:, & Leggett, E.L. (1988). A social-cognition approach to motivation and personality. Psychological Review, 95, 256273. Eden, D. (1988). Pygmalion, goal setting, and expectancy: Compatible ways to boost productivity. Academy of Management Review, 13, 639-652. Eden, D. (1990). Pygmalion without interpersonal contrast effects: Whole groups gain from raising manager expectations. Journal of Applied Psychology, 75, 394-398.

180

Williams and Lillibridge

Evans, M.G. (1986). Organizational behavior: the central role of motivation. Journal of Management, 12, 585-592. Feinzig, S . (199 1). Individual differences in the preference for feedback: Conceptualization and operationalization. Unpublished manuscript, University at Albany, State University of New York. Feltz, D. L. (1982). Path analysis of the causal elements in Bandura's theory and an anxiety-based model of avoidance behavior. Journal of Personality and Social Psychology, 42, 764-781. Folkman, S . , & Lazarus, R.S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-239. Freedman, S.M., & Phillips, J.S. (1985). The effects of situational performance constraints on intrinsic motivation and satisfaction: The role of perceived competence and self-determination. Organizational Behavior and Human Decision Processes, 35, 397-416. Ford, M.E. (1985). The concept of competence: Themes and variations. In H.A. Marlowe, Jr. & R.B. Weinberg ( a s . ) , Competence Development (pp. 3-49). Springfield, IL: C.C. Thomas. Gibson, S . & Dembo, M.H. (1984). Teacher efficacy: A construct validation. Journal of Educational Psychology, 76, 569-582. Gist, M.E. (1987). Self-efficacy: implications for organizational behavior and human resource management. Academy of Management Review, 12, 472-485. Greenberger, D.B., & Strasser, S . (1986). Development and application of a model of personal control in organizations. Academy of Management Review, 1 1 , 164-177. Hackman, J.R. (1983). Designing work for individuals and groups. In J.R. Hackman, E.E. Lawler 111, & L.W. Porter (Eds.), Perspectives on behavior in organizations (pp. 242-257). New York: McGraw-Hill. Hackman, J.R. & Oldham, G.R. (1980). W r k redesign. Reading, MA: Addison-Wesley. Herold, D.M., & Parsons, C.K. (1985). Assessing the feedback environment in work organizations: Development of the job feedback survey. Journal of Applied Psychology, 70, 290-305. Holahan, C.K., & Holahan, C.J. (1987). Life stress, hassles, and self-efficacy in aging: A replication and extension. Journal of Applied Social Psychology, 17, 574-592. House, R.J., & Baetz, M.L. (1979). Leadership: Some empirical generalizations and new research directions. Research in Organizational Behavior, I , 399-42 1.

Perceived Self-Competence

181

Huber, V.L., & Neale, M.A. (1984). Effects of self and competitor goals on performance in an interdependent bargaining task. Journal of Applied Psychology, 72, 197-203. Ilgen, D.R., Fisher, C.D.,& lhylor, M.S. (1979). Consequences of individual feedback on behavior in organizations. Journal of Applied Psychology, 64, 359-371. Jones, G.R. (1986). Socialization tactics, self-efficacy, and newcomers' adjustments to organizations. Academy of Management Journal, 29, 262-279. Kahn, S.E., & Long, B. C. (1988). Work-related stress, self- efficacy, and well-being of female clerical workers. Counseling Psychology Quarterly, 1 , 145-153. Karasek, R. & Theorell, T. (1990). Healthy work. New York: Basic Books. Lane, J. ,. & Herriot, P. (1990). Self-ratings, supervisor ratings, positions and performance. Journal of Occupational Psychology, 63, 77-88. Larsen, R.J., & Csikszentmihalyi, M. (1983). The experience sampling method. In H.T. Reis (Ed.), Naturalistic approaches to studying social interaction (pp. 41-56). San Francisco, CA: Jossey-Bass. Latack, J. C. (1986). Coping with job stress: Measures and future directions for scale development. Journal of Applied Psychology, 71, 377-385. Lee, C. (1982). Self-efficacy as a predictor of performance in competitive gymnastics. Journal of Sport Psychology, 4, 405409. Lee, C. (1988). The effects of goal setting and monetary incentives on self-efficacy and performance. Journal of Business and PsyChOlOgy, 2, 366-372. Lee, C., & Gillen, D.J. (1989). Relationship of Type A behavior pattern and self-efficacy perceptions on sales performance. Journal of Organizational Behavior, 10, 75-8 1. Locke, E.A., Cartledge, N., & Koeppel, J. (1968). Motivational effects of knowledge of results. Psychological Bulletin , 70, 474-485. Locke, E.A, Frederick, E., Lee, C., & Bobko, P. (1984). Effect of self-efficacy, goals, and task strategies on task performance. Journal of Applied Psychology, 69, 24 1-251. Lake, E.A. & Latham, G.P. (1990). Work motivation and satisfaction: Light at the end of the tunnel. Psychological Science, I , 240-246.

182

Williams and Lillibridge

Locke, E.A., Shaw, K.N., Saari, L.M., & Latham, G.P. (1981). Goal setting and task performance: 1969-1980. Psychological Bulletin, 90,125-152. Locke, E.A. & Schweiger, D.M. (1979). Participation in decisionmaking: One more look. In B.M. Staw (Ed.) Research in organizational behavior (Vol. 1). Greenwich, CT: JAI. Markus, H., Cross, S . , & Wurf, E. (1990). The role of the selfsystem in competence, In J. Kolligan, Jr., & R. Sternberg (Eds.), Competence considered: Perceptions of competence and incompetence across the lifespan. New Haven, CT: Yale University Press. Mathieu, J.E., & Zajac, D.M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108, 171- 194. McCombs, B.L. (1986). The role of the self-system in self-regulated learning. Contemporary Educational Psychology, 11, 314332. McEnrue, M.P. (1984). Perceived competence as a moderator of the relationship between role clarity and job performance: A test of two hypotheses. Organizational Behavior and Human Pe~ormance,34, 379-386. Miller, K.I., & Monge, P.R. (1986). Participation, satisfaction, and productivity: A meta-analytic review. Academy of Management Journal, 29, 727-753. Miller, V.D. & Jablin, F.M. (1991). Information seeking during organizational entry: Influences, tactics, and a model of the process. Academy of Management Review, 16, 92-120. Morris, J.H., & Sherman, J.D. (1981). Generalizability of an organizational commitment model. Academy of Management Journal, 24, 5 12-526. Parkes, K.R. (1990). Coping, negative affectivity, and the work environment: Additive and interactive predictors of mental health. Journal of Applied Psychology, 75, 399-409. Pierce, J.L.? Gardner, D., Cummings, L., & Dunham, R. (1989). Organization-based self-esteem: Construct definition, measurement and validation. Academy of Management Journal, 32, 622-648. Rosenthal, R., & Jacobson, L. (1968). Pygmalion in the classroom: Teacher expectation and pupil’s intellectual development. New York: Holt, Rinehart, & Winston. Rotter, J. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs 80 (Whole No. 609).

Perceived Self-Competence

183

Sandelands, L.E., Brockner, J., & Glynn, M.A. (1988). If at first you don't succeed, try, try again: Effects of persistence-performance contingencies, ego involvement, and self-esteem on task persistence. Journal of Applied Psychology, 73, 208-216. Schunk, D.H. (1984). Self-efficacy perspective on achievement behavior. Educational Psychologist, 19, 48-58. Schwalbe, M.L., Gecas, V., & Baxter, R. (1986). The effects of occupational conditions and individual characteristics on the importance of self-esteem sources in the workplace. Basic and Applied Social Psychology, 7, 63-84. Steel, R.P., Mento, A.J., Davis, C.L., & Wilson, B.R. (1989). Psychometric properties of a measure of sense of competence. Educational and Psychological Measurement, 49, 433-446. Stumpf, S.A., Brief, A.P., & Hartman, K. (1987). Self-efficacy expectations and coping with career-related events. Journal of Qcational Behavior, 31, 91-108. Tajfel, H. (Ed.) (1982). Social identity theory and intergroup relations. London: Cambridge University Press. Xtylor, M.S., Locke, E.A., Lee, C., & Gist, M. (1984). Type A behavior and faculty research productivity: What are the mechanisms? Organizational Behavior and Human Pevormance, 34, 402-418. Thomas, K.W., & Velthouse, B.A. (1990). Cognitive elements of empowerment: An "interpretive" model of intrinsic task motivation. Academy of Management Review, 15, 666-681. Wagner, F.R., & Morse, J.J. (1975). A measure of individual sense of competence. Psychological Review, 36, 45 1-459. Warr, P.B. (1987). Work, unemployment, and mental health. Oxford: Claredon Press. Weinberg, R.S., Gould, D. & Jackson, A. (1979). Expectations and performance: An empirical test of Bandura's self-efficacy. Journal of Sport Psychology, 1 , 320-33 1. Weiner, (1986). An attributional theory of emotion and motivation. New York: Springer-Verlag. White, R.W. (1959). Motivation reconsidered: The concept of competence. Psychological Review, 66, 297-333. Wiedenfeld, S.A., O'Leary, A., Bandura, A., Brown, S . , Levien, S . , & Raska, K. (1990). Impact of perceived self-efficacy in coping with stressors on components of the immune system. Journal of Personality and Social Psychology, 59, 1082-1094. Williams, K.J., & Lillibridge, J.R. (1990). The identification of managerial talent: A proactive view. In K. Murphy & F. Saal (Eds.) , Psychology in organizations: lntegrating science and practice @p. 69-94). Hillsdale, NJ: Erlbaum.

184

Williams and Lillibridge

Williams, K. J., Suls, J., Alliger, G.M., Learner, S .M., & Wan, C.K. (in press). Multiple role juggling and daily mood states in working mothers: An experience sampling study. Journal of Applied Psychology. Williams, K.J., & Williams, G.M.(in press). Applications of social psychology to school employee evaluation and appraisal. In F. Medway & T. Cafferty (Eds.), School psychology: A social psychological perspective. Hillsdale, NJ: Erlbaum. Williams, K.J.,Williams, G.M., & Ryer, J.A. (1990). The relation between performance feedback and job attitudes among school psychologists. School Psychology Review, 19, 550-563. Wood, R.! & Bandura, A. (1989). Social cognitive theory of organizational management. Academy of Management Review, 14, 361-384. Wood, R., Bandura, A.? & Bailey, T. (1990). Mechanisms governing performance in complex decision-making environments. Organizational Behavior and Human Decision Processes, 46, 181-201. Yukl, G . A . (1989). Leadership in organizations (2nd edition). Englewood Cliffs, NJ: Prentice Hall.

Issues, Theory, and Research in Industrial1 Organizational Psychology - K . Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

185

CHAPTER 7 CURRENT TRENDS IN MANAGERIAL TRAINING

Bernadette M. Racicot and Rosalie J. Hall "May your managers live in interesting times." Whether this is curse or blessing is open to debate. Less debatable, however, is the likelihood that managers will face many new challenges over the next twenty years. Some of those challenges are unknowable in the present. Others can be forecast now. The ability of managers (and therefore their organizations as a whole) to deal with known and unknown challenges will be strongly influenced by the quality of today's managerial training and development. In this chapter we discuss the current state and future direction of managerial training in general, then focus more specifically on four environmental challenges that must be addressed: (a) the increasing internationalization of companies; (b) rapid technological advances that increase the need for managers not only to keep current with policy and day-to-day operations, but to acquire new technical knowledge; (c) demographic changes in the composition of the workforce (increasing numbers of minorities and women); and (d) increased pressure to provide financial justification for all organizational expenditures, including human resource spending. There are two basic Human Resource Management (HRM) approaches to improving the quality of organizational employees: selection and training. Organizations may choose to emphasize one of these strategies over the other, but most typically use both. Even in those organizations that rely heavily on selection as a strategy for maintaining the quality of managers, training may still play an important, ongoing role as the organization itself grows, adopts new technologies, hires employees with more diverse demographic characteristics, etc., thereby changing the job demands for managers already holding their positions. A plethora of techniques falls under the loose heading of managerial training and development, ranging from formal training

Racicot and Hall

186

in closely specified skills such as decision making, to education and general topics in management, to informal, on-the-job experiences designed to meet long-term development needs (Keys & Wolfe, 1988; Wexley & Baldwin, 1986a). Yet, there are commonalities in the recommended overall approach used to ensure that effective training and development experiences are provided to managers. Experts in the design of organizational training programs are in agreement on the basic tasks involved in their creation (Goldstein, 1986; Wexley & Latham, 1981). The three basic tasks to be performed are: (a) identify training needs and write appropriate objectives; (b) create and implement training programs; and (c) evaluate the effectiveness of the training effort. In the following sections, we discuss these three tasks in more detail. Before addressing these issues, however, we feel it is important to provide a conceptual definition of the term "manager."

LEADERSHIP VERSUS MANAGEMENT: AN OPEN SYSTEMS VIEWPOINT Although managerial training programs are commonplace in most organizations, the types of training implemented vary in both scope and nature. The effectiveness of the programs is questionable. Empirical theory based research on training is scarce. One major problem in this area seems to be in developing criteria of effectiveness. This should not be surprising, because the concept of "manager" is difficult to define. When one asks the question "What is a manager?" a variety of responses is received. Still more confusing is the concept of "leader. Is a manager necessarily a leader? If managers must be leaders in order to be effective (Yukl, 1989), then what is an effective leader? This concept may be even more difficult to define than that of manager. Leadership by its nature is more abstract and has almost a mystical quality to it. At least some progress has been made with respect to identifying managerial behaviors critical to success although they are sometimes difficult to operationalize. An operational definition of leadership remains somewhat elusive; the chapter by Lillibridge and Williams in this volume also discusses leadership. Katz and Kahn (1978) identify leadership as the influential increment beyond the organization Is routines and the mechanical compliance they imply. This definition implies that there is a difference between management and leadership, but there are also some similarities. While managers may rely solely on power ascribed to them by their position, others may also utilize sources of power 'I

Managerial Daining

187

which reflect the aspect of "influential increment" discussed above. In contrast to managers who rely only on the use of coercive, reward, and position power (French & Raven, 1960), those who also utilize expert (power based on knowledge) and referent (power derived from respect and admiration) power have more control over subordinates (Katz & Kahn, 1978). In addition, the latter two sources of power are available to all organizational employees. Since they are not role related, referent and expert power can be used to improve employee performance and organizational effectiveness by all organizational members. In answer to the question: "Does management equal leadership?" the answer must be no. Leaders may be managers, peers, or subordinates. In situations where managers rely exclusively on role prescriptions and the use of legitimate power, they would not be considered leaders by the definition of leadership described above. On the other hand, managers who are also capable of exercising personal influence (expert and referent power) would be considered leaders. In the above section, only one aspect of leadership and management was discussed, power. Although the differential use of power is an important factor in distinguishing leadership and management, more insight could be gained by examining behaviors and functions that differentiate managers and leaders. Again, turning to Katz and Kahn (1978) and their systems perspective, we find a useful framework for examining the concept of leadership/management. The key difference between this model and other theories of leadership is the inclusion of external environmental influences on organizational functioning. The external perspective recognizes the importance of obtaining information about the environment surrounding the organization. This information gathering function becomes more important at higher levels of management. In other words, upper levels of management are involved in boundary spanning activities (Tushman & Scanlan, 1981). Boundary spanning individuals (BSI's) seek out important information which is likely to affect the internal functioning of their organization, interpret this information and transform it into usable data to be disseminated to relevant organizational members for use in decision making. In contrast, middle managers are concerned mostly with the internal workings of the organization. At this level, managers are involved in activities aimed at achieving organizational goals. Specifically, the middle manager functions as somewhat of a "liaison" person between subordinates and superiors. On the one side this individual is responsible for clarifying organizational goals and motivating employees to achieve organizational objectives. On the other side, the middle

188

Racicot and Hall

manager is responsible to upper levels of management for meeting objectives determined to be important by the upper echelon. At the lower level of management, the administrator is concerned with enforcing organizational rules and policies. Although this may seem, at first glance, to be a straightforward function, the challenge of the first level supervisor is in the application of rules and policies in an equitable and consistent manner (Katz & Kahn, 1978). In addition to application of rules, the entry level manager is also involved with some interpretation of organizational policies. Since policies are rarely, if ever, perfectly clear, supervisors are responsible for the judicious use of sanctions and rewards within the context that they were intended.

Levels of Management, Power, and Leadership In line with the discussion of power and its relationship to the concept of leader, it seems appropriate to now consider how leadership, management, and power can be united. All three levels of management described by Katz and Kahn require leadership to some extent. Whether or not a manager is a leader depends on two things: (1) the possession of power needed to be a leader (expert and referent power) and (2) how effectively the individual uses available power to fulfill the requirements of the job. According to the open-systems framework successful upperlevel managers will rely more on referent power. The nature of their organizational role requires them not only to communicate information to key members within the organization but to gather information from and influence members outside their organization. In an interesting application of open systems theory to leadership, Day and Lord (1988) discuss how organizations that are successful at co-opting their environment are more effective than those that are not. Furthermore, they attribute some of this success to a good leader who interacts successfully with that environment, A study conducted by Hirsch (1975) and cited in Day and Lord (1988) may serve to clarify this point. Hirsch (1975) compared the relative effectiveness of the pharmaceutical and recording industries. A careful analysis revealed some key differences between these two industries that would account for their differential effectiveness. The pharmaceutical industry, much more effective than the recording industry, was more able to co-opt their environment. In sum, successful upper levels of managers rely more on referent power than other types of power. However, it should be noted that expert power may also be quite useful in communicating with outside members.

Managerial Training

189

In contrast, successful middle-level managers have as one of their major tasks clarifying the means by which organizational goals will be achieved. In order to perform effectively they must be able to communicate with their employees and provide them with the necessary information to be used in performing their jobs. This requires a more internal perspective. However, as noted earlier, this requires more than interaction solely with subordinates. Middlelevel managers may often find themselves floating in a sea of information. They gather information from upper levels of management, from subordinates, and often from other middle managers, and at times, from individuals outside the organization. Essentially, they are involved in both vertical and lateral communication. This requires them to be boundary spanners of a different sort. In the phrasing of Tushman and Scanlan (1981) middle managers could be considered "internal stars" as opposed to upper level managers who are "external stars. " While external stars communicate with others outside the organization, internal stars focus on internal sources of information. Accurate information gathering requires that middlemanagers develop a rapport and understanding with other departments requiring use of both referent and expert power. In order to achieve this end, it is important that they understand the functioning of those departments as well as the language used by those individuals. At the other end, disseminating this information to subordinates requires the ability to "translate" and filter information needed for them to perform their jobs. In essence, one task of the middlemanager is to protect their employees from information overload. It is no wonder that the job of middle-manager is considered to be one of the most difficult. Although application of rules and procedures, the major function of first-line managers, may not, at first glance, seem like leadership, the effective use of policy does require leadership skills. me for example, the first-line supervisor who relies exclusively on reward and punishment to control subordinates. This type of control, although perhaps effective, has potential negative consequences. Employees managed in this fashion may put in as much effort needed to gain rewards and avoid negative sanctions. They are extrinsically motivated to work but not intrinsically motivated to perform. A first-line supervisor who is able to combine legitimate with referent and expert power may be better able to instill in their subordinates a sense of competence and encourage self-motivation. Employees working to gain the respect and admiration of their supervisor may be expected to put forth that extra effort needed for creative and innovative solutions to problems that may also lead to greater effectiveness (for a comprehensive review of the effect of extrinsic

190

Racicot and Hall

reward on intrinsic motivation, see Deci & Ryan, 1985). In summary, although managers are not necessarily leaders, more effective managers would seem to be those that possess the ability to use more than legitimate power in their dealings with organizational members. Although leadership and "managership" (Ackerman, 1985) may be separate concepts, more effective organizations may be those that employ managers who combine the necessary managerial skills with leadership ability. These individuals exercise the "influential increment" described by Katz and Kahn (1978). The above discussion has focused almost exclusively on leadership from an open-systems perspective. In a later section, models of training that have been developed form current theories of leadership along with their effectiveness and deficiencies will be discussed. Before addressing specific training models, it is important to consider the need for managerial training.

Selecting and Daining Managers Are leaders/managers made or are they born? If one subscribes to a trait perspective, then managerial training would be somewhat superfluous. The emphasis would be on identifying traits necessary for success and selecting individuals who possess these traits. On the other hand, a purely situational or behavioral perspective would suggest that, under the right conditions, anyone could be an effective leadedmanager. Managerial effectiveness could be achieved by the use of a set of appropriate behaviors. Research on the trait approach to predicting effective leadership has not proved very fruitful. However, the emphasis on use of referent power for influencing persons both inside and outside the organization suggests that personality may be an important factor in predicting successful managerial behavior (Ackerman, 1985). Several personality characteristics have been suggested to be important. According to Ackerman (1985) the most widely accepted traits are honesty, loyalty, and integrity. Intelligence has also been included as an important characteristic (Gordon, 1988). According to Gordon (1988), for training to be effective, individuals to be trained must possess both the ability and motivation necessary to benefit from training. Recent research on self-monitoring (Garland & Beard, 1979; Ripkey, 1990) suggests that this may be an important trait for future investigations in examining managerial effectiveness. Since high self-monitors are more sensitive to external cues, they may also be more effective leaders. These individuals may be more sensitive to needs of employees and to changes in the environment that would

Managerial Training

191

warrant action. In addition, leadership theories have also focused on traits of both leaders and followers for predicting leadership effectiveness. With respect to behavior, several taxonomies have been proposed for defining the role of manager. We have chosen to focus on Yukl's (1989) taxonomy because it is comprehensive and incorporates other perspectives. According to Yukl (1989) it is difficult for an individual to be an effective manager without being a leader. In line with this reasoning, it should not be surprising that many classifications of managerial behavior include elements found in many leadership theories. Yukl (1989) identified four broad categories of managerial behavior: (1) building and maintaining relationships; (2) giving and seeking information; (3) making decisions; and (4) influencing people. Within each of these categories, are specific behaviors associated with managerial success. Consistent with an opensystems framework, Yukl's taxonomy includes both internally and externally oriented managerial behaviors. For example, he emphasizes the importance of recognizing and rewarding, motivating, supporting, and clarifying which would be considered internal activities. Externally oriented behaviors include networking and monitoring. In summary, both trait and behavioral perspectives may offer important insights into understanding managerial effectiveness. With respect to traits, there is some consensus that intelligence and charisma are important factors. More research is needed to establish the role of traits in predicting managerial success. Many of the behaviors described in Yukl's model have been identified by others. In addition, they are evident in many leadership theories that have been applied to managerial training and development. Assuming that both selection and training are important activities for improving managerial effectiveness, the specific skills needing improvement must be identified before developing and implementing a training program. We now turn our discussion to one of the most important activities in which training personnel is involved, needs assessment. Following our discussion of needs analysis, leadership models and their application to managerial training will be discussed.

Identifiing Managerial Daining Needs A diversity of tasks is performed under the title "manager." The lack of a level and function-sensitive, comprehensive classification system of managerial behaviors with accompanying theories of managerial effectiveness hinders a fully scientific approach to implementing training and development programs. Leadership theo-

192

Racicot and Hall

ries provide the best guides, but they may focus on narrow aspects of managerial behavior, ignoring behaviors that are necessary for success but do not involve "leadership" per se. Other guides to understanding managerial development needs include the study of effective executives in order to determine the experiences that influenced their career progress (e.g., McCall, Lombardo, & Momson, 1988). This approach has the advantage of being firmly grounded in the realities of organizational practice, but suffers from its reliance on self-report and lack of control groups. Yet a third approach is to provide managers with skills training on an as-needed basis ("if it ain't broke, don't think about it"), waiting for conditions to make it painfully obvious who needs what kind of training. While this approach eliminates the ambiguity associated with forecasting training needs, it may also exact a high price in terms of lost opportunities and unnecessary stress and strain for organizational employees. In practice, all three of these approaches may determine the direction of managerial training in organizations. As will be demonstrated, in many organizations there is a sizable gap between recommended training practices and actual training practices. In some cases, the gap may result from special issues that arise when providing managerial training. These need to be acknowledged and studied further. Finally, we argue strongly for the application of these recommended practices not only to formal training and development programs, but to the many informal, onthe-job training efforts. The development of any training program begins with a thorough needs assessment (Goldstein, 1986; McGehee & Thayer, 1961; Wexley & Latham, 1981), which is discussed in greater detail in the chapter by Caska, Kelley, and Christensen in this volume. qpically, the needs assessment includes analysis of the job/task to be trained (job or operations analysis), assessment of the abilities of current and future employees with respect to the task (person analysis), and identification of any organizational factors (e.g., politics, financial considerations, etc.) that might be relevant to the success of the training program (organization analysis). From needs assessment comes the information to develop training objectives. These objectives in turn guide the development of the content of the training program and provide a basis for constructing an evaluation of the effectiveness of the training effort. From this somewhat idealized viewpoint, training program development follows a rational and logical progression, thus maximizing the chances that employees will be trained in the skills they need while the organization avoids the costs of training on irrelevant skills. A look at the needs assessment procedures actually followed in organizations shows a more frag-

Managerial Training

193

mented picture (e.g., Moore & Dutton, 1978; Schneier, Guthrie, & Olian, 1988). Many organizations do not perform needs assessment for managerial level jobs. Saari, Johnson, McLaughlin, and Zimmerle (1988) report that only 27% of a sample consisting of 611 companies randomly selected from Dun’s Market Identzjiers (company size 1,000 and up) conduct needs assessments for managerial jobs. And, of those companies that do perform needs assessments, the procedure is more typically done for first-level supervisory jobs than for higher level managerial jobs. Because the lack of needs assessment is so pervasive, we suggest that researchers and practitioners alike could benefit from looking at this issue from a descriptive rather than a prescriptive perspective. Rather than assume that needs assessments are not performed because of ignorance or laziness, the forces that motivate against managerial needs assessment may be profitably considered. For example, needs assessment procedures may appear cumbersome and slow-moving to managers pressured to deal with uncertain and rapidly changing issues and environments. Also, managers may have both the desire and the power to exercise a high degree of control over their own training and development. Training and development needs of managers may be difficult to describe concretely, compared to those of employees with some easily observable skill. Some managerial training needs may even be potentially organizationally embarrassing and thus for political reasons not likely to be documented in written form as required in a traditional needs assessment. The approaches that we describe below may help remove some of these barriers to doing needs assessment. Many potential sources of information may be consulted for the needs assessment, ranging from written and oral statements of organizational goals and objectives, manpower inventories, various indicators of overall organizational climate and effectiveness (e.g., grievances filed, turnover and absenteeism data, efficiency indices, customer complaints, exit interviews), literature review, the accumulation of requests for training and analysis of operational problems, to task analyses, work samples, and performance evaluations (Goldstein, 1986; Miner & Miner, 1977; Moore & Dutton, 1978; Wexley & Latham, 1981). The process of collecting and integrating the needs assessment information has been discussed by a number of authors. For example, Schneier et al. (1988) describe a comprehensive framework to use in implementing a KSA (knowledge, skills, and abilities) based needs assessment. These aspects of doing needs assessment, although requiring effort, political sensitivity, and technical skill, have not shown alot of change over the past 20 years. The rest of this section addresses other managerial needs assessment

194

Racicot and Hall

questions that are less resolved. For example, what do surveys of managerial needs suggest are key deficiencies? How can companies customize this information to meet the specific needs of their own employees? Are managerial employees capable of doing their own [self-] assessment? What do we know about the use of assessment centers to identify current performance problems or blocks to future career advancement?

Determining Generic Managerial Training Needs. Several researchers have argued that managers have generic needs that cut across organization type. Leadership theories provide an expanding source of theoretically based and empirically validated information about necessary managerial skills and abilities. For example, Crouch and Yetton (1987), building upon the Vroom-Yetton model of decisionmaking, looked at reasons why managers often express reluctance towards using group decision-making as a means of resolving conflict. They found that managers must possess a minimum critical level of slull in legitimizing conflict in order for the group decision method to influence positively subordinate performance. Thus, organizations planning to adopt a training program using the VroomYetton approach also need to determine whether their managers have the requisite conflict legitimization skills, and plan supplementary training accordingly. Other, empirically based studies also provide information that may aid the determination of managerial training needs. Lower level managers' needed skills include communications, planning, organizing, team building, motivation, employee training, production control, performance evaluation, and building their own influence (Alpander, 1986). Typically, the higher the management level, the greater self-perceived need for training in conceptual as opposed to concrete technical skills (Bernick, Kindley, & Pettit, 1984). Some perceived training needs (i.e., basic supervisory and interpersonal skills) appear to generalize across lower and middle managerial levels (Ford & Noe, 1987). We are not advocating relying solely on such generic descriptions of needs. Doing so may lead to problems such as: (1) increased expenses from training managers in skills they already possess or will not use and (2) poor transfer of training due to failure to identify important organization-specificcontextual factors that should be incorporated in the design of the training programs. However, given these caveats, knowledge of the key training needs of managers across organizations and levels may shorten and focus the needs assessment period. An example of how a set of generic categories can provide the basis for a needs assessment tailored to the organization is Yukl's

Managerial Training

195

(Yukl, 1981; Yukl & Nemeroff, 1979; Yukl & Van Fleet, 1982) work with the Managerial Behavior Survey (MBS). The MBS questionnaire assesses managers in terms of a taxonomy of nineteen behavioral categories (e.g., inspiration, information dissemination, criticism-discipline) relevant to holding leadership roles. Organizational members who know the manager fill out the MBS twice - first describing the manager's actual behavior, then indicating the behavior of an ideal manager. When a negative discrepancy occurs, it indicates a potential need for training. Thus, theoretically based and observationally derived classifications of managerial behavior may hasten the development of needs assessment instruments, and may help ensure their comprehensiveness. In some cases, the resulting needs assessment information will facilitate the transition from needs assessment to development of training content because extensive training materials have already been developed and tested in other organizations. For example, Komaki's recent development of the Operant Supervisory 'hxonomy and Index (Komaki, 1986; Komaki, Zlotnick, & Jensen, 1986) for describing supervisory behavior could yield information that would be easy to incorporate into a behaviorally oriented training program in managerial interpersonal skills.

Self-assessment of Needs. Several researchers have suggested that it might be possible, even valuable, for managers to assess their own training needs (e.g., McEnery & McEnery, 1987). Oppenheimer (1982) provides a description of a company's program of self-assessment of task and person analysis by middle and senior level managers across 150 jobs. Incumbents responded to a needs assessment survey by rating 21 skill and knowledge areas (e.g., teamwork, goal setting) generated from a survey of management literature, assessment center information, and subject matter expert suggestions. For each skill and knowledge area, respondents to the survey indicated how helpful training in that area would be, prioritized the five most helpful areas, indicated their personal training interests, and were free to include additional suggested training areas. The five top training priorities were similar across respondents, suggesting agreement in the assessment of training needs. The resulting set of training programs was at least moderately successful (several sites did not participate fully). Ford and Noe (1987) studied self-assessment survey responses from 506 middle and lower level managers. The surveys consisted of a list of 80 managerial skill items (basic management skills, interpersonal skills, administrative skills, and quality control skills) generated from managerial job analyses, subject matter experts, and objectives from intact training programs.

196

Racicot and Hall

Overall, the results suggested that managers showed slight differences in their self-assessment of needed skills that were consistent with predictions based on their level and function, but that many managerial training needs appeared to generalize across level and function. Further research is needed to determine the accuracy of such selfassessments, for example, by comparing them with other sources of needs assessment information. Self-assessment of training needs is not without its potential hazards. Ford and Noe (1987) showed that a manager's favorable attitude toward the utility of training (as assessed using a five-item scale) may be associated with a higher reported need for training in specific skills. In their study, managers with less favorable attitudes toward training were less likely to request training in quality control skills. Kaplan, Drath, and Kofodimos (1987) note barriers to selfdevelopment for executive level employees, including insulation from personal criticism, hectic work schedules that lead to devaluation of introspection because it does not address immediate problems, a need to be and appear exceptionally competent, and fear of experimenting with their already successful approach to management. These barriers are likely not as strong for lower and middle level managers because of the lower power and responsibility of their positions, but may impede self-assessment to some extent for this group. Although Graham and Mihal (1986) advocate selfassessment, they suggest that following self-assessment procedures thoughtlessly may be problematic. For example, they describe a self-assessment effort where managerial employees indicated that they needed computer training and indeed received it -- in spite of the fact that they had no access to computer equipment! The research literature also addresses the issue of self-evaluation of ability and performance in general. Most of this work is cast in a performance appraisal context, however, it seems likely that some of the same processes also may hold in a training context. For example, Mabe and West (1982) used a meta-analytic technique to review 55 studies reporting comparisons of self- and other-assessed evaluations of specific abilities or skilled performance. Overall, their results suggested a rather low correspondence between self and other evaluations (mean unweighted r = .29; weighting by sample size increased the correlation to .3 1). However, they demonstrate that attention to the conditions of measurement can substantially increase the correspondence between self and other assessments. Specifically, self-assessment is more accurate when: (1) raters are instructed that their self-assessment will be compared to a criterion, (2) raters have previous experience with self-evaluation, (3) the rating instructions include assurances of confidentiality, and (4) the

Managerial Training

197

instructions emphasize comparing own performance or ability with others, rather than on an absolute scale. Mabe and West estimated that when all four of these conditions are present, the true correlation between self- and other-assessments may be around .60. (For additional relevant research, see also the work of Fahr and his colleagues: Fahr & Dobbins, 1989; Fahr & Werbel, 1986; Fahr, Werbel, & Bedeian, 1988). In sum, self-assessment of training needs may give managers more control over, and therefore make them more receptive to, training efforts (McEnery & McEnery, 1987). Such assessment approaches require careful planning to ensure their validity and relevance to organizational objectives. Certainly, self-assessments should be supplemented with needs assessment information from other sources.

Needs Assessment through Simulation. Thorn ton and Cleveland (1990) suggest that the use of simulations (e.g., business games, role playing exercises) for management training and development purposes is widespread, and may play multiple roles in the training process, including functioning as a needs assessment technique, a means of skill development, and a means to collect performance criteria against which to evaluate the success of other training endeavors. Techniques used in these simulations include one-on-one interview simulations, leaderless group discussions, in-basket techniques, complex decision-making simulations, and large scale behavioral simulations (e.g., CCL's Looking Glass). Conclusions. Thus, in spite of pressures upon organizations to avoid needs assessment for managerial training, training programs are likely to be of higher quality and more strongly accepted if the needs assessment is performed. Needs assessment aids the development of training objectives that will determine choices about training program content and method of delivery, as well as serve as a basis for determining evaluation criteria. And, the benefits of the data collected in needs assessment may not be restricted to training and development efforts. Schneier, Guthrie, & Olian (1988) argue that needs assessment may provide a valuable data base for other human resource planning and development efforts (e.g., recruiting, selection, compensation, career development). A mere sampling of needs assessment research and thinking was discussed in this section. Other valuable and recent trends that were not covered include: (1) renewed emphasis upon tying needs assessment and training to strategic organizational objectives (e.g., Bolt, 1985; Fisher, 1989) and (2) the increasing use of sophisticated, computerized systems for tracking the training and development needs of employees.

Racicot and Hall

198

CURRENT APPROACHES TO MANAGERIAL TRAINING Once training needs are identified, the task at hand is to choose and implement a training program. Many of the current approaches to managerial training are based on leadership theories. In this section, five theories of leadership, their application, and effectiveness will be discussed. The five theories are: (1) Path-Goal theory (House & Mitchell, 1974); (2) Fiedler's Contingency theory (Fiedler, 1964; Fiedler, 1969); (3) Vroom-Yetton Contingency theory (Vroom & Yetton, 1973); (4) Leader-Member Exchange theory (Dansereau, Graen, & Haga, 1975); (5) Situational Leadership theory (Hersey, Blanchard, & Natemeyer, 1979).

Path-goal Theory Path-goal theory, a contingency theory of leadership, integrates three components of organizational behavior, leadership, motivation, and job satisfaction (House & Mitchell, 1974). It focuses on leader behaviors but also takes follower characteristics into account. The basic tenet of path-goal theory is that a successful leader is one who motivates subordinates to perform well by defining goals and removing obstacles to goal attainment. Subordinates who perceive a clear path to desired rewards through goal attainment will be more satisfied with the leader. The relationship between leader behavior and subordinate satisfaction is moderated by two general types of contingency variables, subordinate characteristics and the environment. House and Mitchell (1974) suggest two personal characteristics that may interact with leader behavior to affect subordinate satisfaction: (1) locus of control and (2) subordinates' perceived task competence. Individuals with an internal locus of control (who believe that events occur because of their own behavior) are more satisfied with the participative method of leadership while those with an external locus of control (who believe that events occur due to chance) are more likely to accept a directive leader. In addition, individuals with a strong sense of their own ability will find directive leadership less acceptable. With respect to the environment, House and Mitchell suggest three variables that moderate the relationship between leader behavior and subordinate satisfaction: (1) task complexity; (2) task content; and (3) opportunity to attain valued rewards. The more routine the task, the less the perceived need for directive leadership. In addition, directive leadership will be less acceptable on dissatisfying or unpleasant tasks. Finally, directive leadership will be more

Managerial Training

199

acceptable under circumstances where such leader behaviors are necessary for individuals to attain goals which will result in desired rewards. One of the major strengths of path-goal theory is that it is comprehensive, incorporating leadership, motivation, and job satisfaction. According to Neider and Schriesheim (1988) it is the most promising theoretical perspective available for understanding leadership. In addition, it is a framework which can be applied in practical settings for the purpose of enhancing manager effectiveness. Neider and Schriesheim (1988) present a three-stage model which highlights organizational and managerial functions necessary for leadership effectiveness. The three stages, precursor conditions, maintenance functions, and reassessment and monitoring, deal with different but interdependent processes of managing. The precursor phase includes such activities as job analysis, compensation analysis, employee selection, and orientation training. Maintenance functions include career pathing, goal setting, contingent reward and punishment, and "individualised consideration. " The reassessment and monitoring stage highlights the importance of performance feedback, problem solving, and reassessment of subordinate needs. This model emphasizes the complexity and dynamic nature of the managerial job. In line with an open systems perspective, it can also be seen that managerial effectiveness is influenced by factors that are largely out of the manager's direct control. This is particularly evidenced by the inclusion of selection in the precursor stage. The manager's behavior will depend, at least somewhat, on the types of employees that can be recruited into the organization. As a result, flexible managerial behavior will be required depending on characteristics of subordinates and the situation. Further evidence for the importance of managerial adaptability are provided in both the maintenance and reassessment phases of the model. The individualized nature of path-goal theory presents both implications and challenges for the training of managers. Training programs which emphasize the awareness of employee needs and the importance of flexible behavior appear necessary for improving managerial effectiveness. Although path-goal theory has a great deal of potential for managerial training, it also has some shortcomings. Because of its complexity and the difficulty in operationalizing key constructs, only partial tests of the model have been conducted. Most research has focused on job satisfaction as the major outcome variable. Although the results with job satisfaction have resulted in support for the model, more research is needed to provide support for predictions made by the theory (Fulk & Wendler, 1982).

200

Racicot and Hall

Fiedler 's Contingency Theory of Leadership Fiedler's theory of leadership is a trait-based contingency theory. According to this theory, effective leaders are those matched to compatible situations. Individuals who describe their least preferred co-worker (LPC) in relatively negative terms, on the LPC measure are classified as task-oriented (low LPC). Individuals who describe their least preferred co-worker in relatively positive terms are classified as relationship-oriented (high LPC). Since LPC is considered to be a stable trait, they key to effective leadership is in creating situations which are compatible with leadership style (Fiedler, 1969). The three major situational variables identified by the theory are: (1) leader position power; (2) leader-member relations; and (3) task structure. Position power can be equated with legitimate power, or power granted by the organization. Leader-member relations can be considered similar to referent power. Finally, task structure is the extent to which the task(s) to be performed are well-defined. Through their research Fiedler and his colleagues concluded that low LPC (task-oriented) leaders were most effective in either very favorable or very unfavorable situations. Favorable situations are those in which the leader has good leader-member relations, a great deal of position power and a highly structured task. Unfavorable situations are those characterized by poor leader-member relations, low position power and low task structure. On the other hand, high LPC leaders were found to be more effective in situations of mixed favorableness. Stated another way, low LPC leaders are most effective when they have either a great deal of influence or very little influence over their subordinates while high LPC leaders are most effective in situations where they have moderate influence (Fiedler, 1969). The Leader-Match training program (Fiedler & Mahar, 1979) is an application of Fiedler's contingency theory. The training program focuses on assessing leadership style through completion of the LPC scale, diagnosis of the leadership environment, and training individuals in engineering the job to better match leadership style. Research on Fiedler's contingency theory of leadership has produced mixed results. Using meta-analysis, Strube and Garcia (1981) reported substantial support for the model. On the other hand, Peters, Hartke, & Pohlmann (1985) reported mixed results finding relatively strong support for the model in laboratory studies but not in field studies. They concluded that the theory is incomplete and suggested further theory development emphasizing the need for research into the investigation of moderating factors. A field

Managerial Daining

201

study with ROTC cadets suggests support for the effectiveness of the Leader Match training program (Fiedler & Mahar, 1979). They reported that individuals who received Leader Match training performed better than those that did not. Csoka and Bons (1978) also provided support for the validity of the Leader Match program reporting that performance ratings were higher for individuals trained in Leader Match than those who were not. Finally, Hosking and Schriesheim (1978), in their review of the leader-match training program, offered several criticisms of the model including questionable reliability of the LPC scale, poorly justified measurement of situational favorability, a narrow view of the leadership situation, and lack of empirical support for the model.

Vroom-Yetton Contingency Theory The Vroom-Yetton theory (Vroom & Yetton, 1973) is a behavioral contingency theory which is prescriptive in nature. The theory's major focus is improving decision-making effectiveness of leaders. Through examination of situational characteristics, appropriate decision-making styles can be identified. More specifically, the leader must first classify the nature of the problem by applying seven decision rules intended to protect the quality of the decision and acceptance of the decision by subordinates. After applying these seven rules, the leader is left with a "feasible set" of alternatives for solving the problem. The three rules to protect the quality of decisions emphasize the extent to which the leader possesses adequate information to solve the problem, the commitment of subordinates to pursue organizational goals, and the structure of the problem. Rules to protect the acceptance of the decision accentuate the importance of decision acceptance for effective execution of the decision, degree of conflict between subordinates, the relative importance of quality versus acceptance of the decision, and commitment of subordinates to pursue organizational goals in situations where acceptance is essential. Depending on the answers to questions, based on these decision rules, the leaders chooses a decision making style from the "feasible set. The leadership styles vary in amount of subordinate participation allowed in the decision making process. The five styles identified by the theory range from completely autocratic, where the leader makes the decision on his or her own (AI) to a group decision process where the leader is willing to abide by the decision reached by the group (GII;Vroom & Jago, 1978). Tests of the model suggest that it is more useful for protecting acceptance of decision than quality (Field, 1982; Vroom & Jago, 1978). This model is considered to be somewhat narrow in scope, 'I

202

Racicot and Hall

dealing exclusively with decision making which is only one of many activities performed by a leader, Furthermore, it may be impractical in situations where a decision must be made quickly. In spite of its shortcomings, the model has its positive attributes. The emphasis on behavioral flexibility has been cited by others as an importance characteristic of a good manager. In addition, it may be a good training model for assisting managers in gaining insight into their personal decision processes and could be used as an aid in teaching managers to increase flexibility in decision making.

Vertical Dyad Linkage Theory The vertical dyad linkage (VDL) model (Dansereau, Graen, & Haga, 1975) is also a contingency theory of leadership. The major tenet of the model is that leaders and followers mutually influence each other's behavior. As such, the theory focuses on the dynamic interchange between leader and follower dyads. According to the model, leaders categorize subordinates as members of either the ingroup or out-group based on three characteristics of subordinates: (1) competence and skill; (2) trustworthiness; and (3) willingness to accept additional responsibility. Subordinates perceived as having greater competence, being more trustworthy and more willing to take on responsibility are classified as in-group members and are awarded more consideration and support from the leader. These individuals are also assigned tasks which are more challenging and rewarding than those assigned to out-group members. Out-group members, those perceived as less competent, trustworthy, and willing to take on extra responsibility, receive less support from the leader and are relegated to more routine work. Research on the VDL model has been generally supportive. Vecchio and Gobdel (1984) reported that bank tellers characterized as in-group members had higher job satisfaction and performance than out-group members. It should be noted that the predictions of the model held up better with "soft" measures than with "hard" measures. Specifically, significant differences between in- and outgroup members were found with respect to supervisor performance ratings and self-rated performance but not with magnitude or frequency of errors or propensity to quit. A field study conducted by Liden and Graen (1980) also provided support for VDL theory. The results of their study indicated that supervisors differentiate among their subordinates. Furthermore, individuals who reported higher quality exchange relationships with their supervisor performed different tasks than those with a lower quality relationship. Specifically, the former group spent more time on tasks that were outside the

Managerial Training

203

realm of activities contained in their job description. Research on the leader-member exchange (LMX) model (derived from VDL theory) may be particularly relevant for managerial training. In a field study, Scandura and Graen (1984) investigated the moderating effect of the quality of initial leader-member exchange (LMX) on the effectiveness of managerial training. The managers in the LMX treatment group received twelve hours of seminar training consisting of use of the LMX model, active listening skills, exchanging mutual expectations, exchanging resources, and practice in using these skills. Results indicated that the LMX training was more advantageous for initial low-LMX members than for the high-LMX members in terms of job satisfaction, productivity and quality of leader member relationships. This program may be particularly useful for assisting managers in recognizing their differential reactions to subordinates and for repairing "damaged" relationships. From what we know about the impact of cognitive biases such as first impression error and halo error, it is entirely possible that an individual who started out "on the wrong foot" with a supervisor may have a difficult time moving from the out-group to the in-group. LMX training may have potential for improving performance of such individuals through better leader member exchanges.

Situational Leadership Theory Situational leadership theory (Hersey et al., 1979) is a contingency model which prescribes the type of power which will be most effective given the maturity level of subordinates. Hersey et al. (1979) delineate seven sources of power and suggest four different styles of leader behavior: (1) delegating; (2) participating; (3) selling; and (4) telling. These styles differ on two dimensions, relationship behavior and task behavior. Behaviors high on task behavior and low on relationship behavior are referred to as telling (Sl). High task and high relationship behaviors are characteristic of selling behavior (S2). Participating is characterized by high relationship and low task behaviors ( S 3 ) . Finally, leadership style low on both task and relationship behaviors is referred to as delegating (S4). 'Risk behavior is defined as the degree of direction provided by the leader to subordinates. Relationship behavior refers to the degree of support provided by the leader, characterized by two-way communication and active listening. Four levels of maturity are included in the theory. Maturity is described i n terms of subordinate ability and motivation to take responsibility for self-direction. With respect to power, at lower levels of maturity, the leader should be most effec-

204

Racicot and Hall

tive when using those sources provided by the organization. On the other hand, at higher levels of maturity, the theory prescribes the use of power earned by the leader. Research pertaining to the effectiveness of situational leadership theory is scarce (Vecchio, 1987). In an empirical test of the theory, with high school teachers and principals, Vecchio (1987) reported that the predictions of the theory held only for low-maturity individuals. That is, telling behavior (low relationship and high task behavior) was an effective management style for low-maturity teachers. Vecchio suggested that further research and refinement of the theory, particularly with respect to high-maturity subordinate predictions, is needed before conclusions on the utility of the theory can be drawn. In summary, five models of leadership that have been applied to managerial training were discussed. These models were chosen because they either explicitly or implicitly incorporate both behaviors and traits as being important in determining managerial/leadership effectiveness. Although research results on the adequacy of these theories for predicting effectiveness has been somewhat mixed, we feel that further research and refinement may prove useful in understanding managerial success. Furthermore, we agree with Neider and Schriesheim (1988) that path-goal theory may be quite useful as an application to management training and development programs. With respect to the model presented by Neider and Schriesheim, we feel that it is particularly relevant to understanding and predicting managerial effectiveness for several reasons: (1) It is consistent with an open-systems framework. Specifically, the inclusion of such elements as selection and compensation analysis suggests that managers be aware of the external environment and its impact on the organization. The type of employees that an organization is able to attract and maintain may affect the type of leadership needed by the organization; (2) It incorporates behaviors and uses of power that are consistent with the "influential increment'' described by Katz and Kahn (1978). In addition to the use of contingent reward and punishment, Neider and Schriesheim emphasize "individualised consideration" and career pathing which suggest the need for establishing relationships based on referent power; (3) The model incorporates both traits and behaviors that should be important to managerial effectiveness. Although path-goal theory emphasizes the role of traits of subordinates, the inclusion of flexible behaviors, particularly "individualised consideration" suggests that leaders with certain personal characteristics may be more able to adapt their behavior to different types of situations and subordinates. We feel that this may be a fruitful area for future

Managerial Training

205

research. Specifically, research that focuses on theoretically meaningful traits that are associated with effective management behavior may help practitioners in selecting individuals that are most likely to benefit from managerial training.

OTHER MANAGERIAL TRAINING TECHNIQUES In this section, two additional types of programs that have been used in managerial training will be discussed. Although many others exist, we have chosen to focus on assessment centers and human relations training for two reasons: (1) Assessment centers are comprehensive and incorporate many of the other methods used in managerial training such as business games and role playing; (2) Both assessment centers and human relations training focus on skills that are consistent with our definition of leadership and our premise that more effective managers are also leaders.

Assessment Centers and Management Development The assessment center is a standardized technique used for evaluating management potential (Wexley & Latham, 1981). Individuals are evaluated on a number of dimensions (determined through job analysis) which have been deemed necessary for success on the job. Although assessment centers have traditionally been considered a tool for selection and promotion of managers, they have great potential for use in management development (Byham, 1971). Assessment centers incorporate several types of activities, such as business games, discussion sessions, and role playing, utilized in common management development programs. Through a variety of exercises (e.g. leaderless group discussions; in-baskets) and inventories (e.g. cognitive ability tests; personality questionnaires) managers can gain feedback on strengths and weaknesses and improve skills in such areas as human relations, behavior flexibility, decision making, communications as well as other areas considered important for managerial success (Cascio, 1987). The emphasis on skill areas such as decision making, human relations, communication, and behavior flexibility indicates that one purpose of assessment, whether used for selection or training, is to evaluate candidates on leadership ability. The types of exercises and skills training emphasized will vary somewhat according to managerial level and position. In general, the use of assessment centers as an aid to managerial development is more appropriate for middle and upper levels of management rather than lower levels. The detailed information gathered in assessment centers pro-

206

Racicot and Hall

vide a strong basis for identifying training needs. Most assessment center reports include the major strengths and weaknesses of participants as well as developmental recommendations for improvement of skills. These data can then be used to guide training efforts. Byham (1971) suggests several ways that assessment centers can be useful for management development. First, the feedback provided to participants during the exercises can stimulate self-development. Second, assessment and formal training based on the assessment can be combined into one program in a cost-effective manner. Finally, assessors, typically line managers, benefit from both the formal assessor training and from participation in the center. Through their experience as an assessor, they can improve their own managerial skills. Although the use of assessment centers for managerial training appears promising, more research on the utility of assessment centers for use in management development is needed.

Daining in Interpersonal Skills Training supervisors and managers in order to improve human relations is one of the most universally acknowledged managerial training needs. Many of the leadership theories discussed earlier have specific training programs associated with them (e.g., Leader Match). A meta-analysis of the effectiveness of various training areas (e.g., problem solving, general management) suggested that human relations training was clearly effective based on subjective learning criteria, subjective behavior criteria, and objective results criteria (Burke & Day, 1986). Training in interpersonal skills may be provided through multiple methods (e.g.? lecture, discussion). This section focuses on training using a behavior modeling approach, on the basis that it was identified by Burke and Day (1986) as one of the most effective techniques. Behavior modeling approaches include not just passive listening, but the opportunity to observe, apply, and get feedback on the new skills to be learned. In contrast to many other training techniques, behavior modeling training: (a) is theoretically grounded; (b) has a substantive empirical research base; and (c) is described in sufficient detail in various written sources to allow trainers to design their own programs (e.g., Decker & Nathan, 1985; Goldstein & Sorcher, 1974). Reviewing the recent literature on behavior modeling suggests that the technique is going through a phase of fine tuning, including learning about the possible undesired effects of behavior modeling. Mayer and Russell (1987) pointed out deficiencies in our knowledge about behavior modeling on four counts: validity, "cost effectiveness" (comparative effectiveness of behavior modeling with other

Managerial Paining

207

techniques), process variation in the way behavior modeling training is conducted, and whether or not it produces long term changes in trainee attitudes and values. For example, the variability in provision of behavior modeling training may be great: models may be live or videotaped; models may demonstrate inappropriate behaviors as well as those to be learned; the trainees' own managers may participate in the program; the opportunity for practice may vary widely. More knowledge is needed about the effects of these variations. Additionally, some research has shown unexpected results may occur from behavior modeling training. A simple imitative learning model does not completely predict the results of behavior modeling training -- it is likely that cognitive structures such as schema guide the interpretation of behavior demonstrated by the behavioral model and thus the resulting behaviors demonstrated by trainees (Manz & Sims, 1986). Further refinement of our knowledge about behavior modeling training could focus on the contributions of individual difference variables. Gist, Schwoerer, and Rosen's (1989) suggestion that behavior modeling is appropriate for persons with low self-efficacy needs further study. Additionally, we know little about the effects of covert (versus overt) practice of skills in behavior modeling training. Finally, we need additional study of the factors that facilitate training transfer after behavior modeling training.

The Development o JIFainin Programs

-- Maximizing the I flects ! of$ raining

Many components can influence t h e success of a training program, including the choice of instructional techniques, as well as employee attitudinal and motivational factors that affect learning. We discuss these, then look at the effects of three current trends (internationalization, technology, and diversity) on the content and methods used i n training. We do not claim to cover all possible techniques, instead we try to focus on those we believe are newer, tie into the trends we discuss, are more likely to focus on managers as the training population of interest, or have more empirical support.

Attitudinal and Motivational Factors Increasing attention has recently been devoted to cognitive, perceptual, and motivational aspects of training and the trainee that influence the success of a training program. Self-efficacy (a person's beliefs about his or her ability or competence to complete a

208

Racicot and Hall

given task) appears to be both a predictor of learning and an important outcome of training as recent work by Gist suggests (Gist, 1987; Gist, 1989; Gist, Schwoerer, & Rosen, 1989). Hill and Elias (1990) demonstrate the importance of both organizational factors and individual characteristics in determining mid-career managers' self-efficacy (and therefore persistence) to retrain. A number of training methods (e.g., self-management training, behavior modeling training, goal setting training) have explicit ties to psychological theories of learning and motivation, and thus incorporate social cognitive and motivational factors in the training approach. As such, evaluations of these methods present an opportunity to evaluate the effects of manipulating motivational factors in training. For example Gist, Bavetta, and Stevens (1990) compared the relative effectiveness of two training methods (self-management versus goal setting) for increasing the transfer of negotiation skills training. Self-management training (e.g., Frayne & Latham, 1987; Kanfer, 1980; Latham & Frayne, 1989) typically consists of the following strategic and motivational components: (1) assessing problems and then setting hard, specific goals to achieve; (2) identifying environmental barriers to performance; (3) developing strategies to overcome barriers; and (4) self-administration of rewards/punishments to encourage goal commitment. It is likely that self-management training has its effect in part through increases in trainee self-efficacy (Frayne & Latham, 1987; Latham & Frayne, 1989). The goal setting training that Gist et al. (1990) provided gave trainees information about how to set goals for both performance outcomes and interim practice, different types of goals, and the rationale for linking goal setting to effective performance. Selfmanagement training and goal setting training were compared in terms of their effectiveness in increasing overall performance, skill generalization, and effort expenditure on a transfer task (following the final training session by about one week). Emphasis on goal setting led to significantly higher effort on the transfer task. In contrast, self-management training led to significantly higher overall performance and generalization on the transfer task. However, an earlier study of the maintenance of skills learned in a time management course showed transfer was higher in two goal setting conditions than in a self-management condition when transfer was assessed two months after training (Wexley & Baldwin, 1986b). Direct comparison of the results of the Gist et al. and the Wexley and Baldwin studies is made more difficult by the differences in the way in which goal setting training and self-management training were presented. From a practitioners' standpoint, both goal setting approaches and self-management training are effective, the question

Managerial ll-aining

209

becomes under which task conditions is each preferable, and, which micro-components of the training (e.g., setting performance goals, setting intermediate goals, self-observation, reinforcement) determine the effectiveness of transfer. Gist et al. (1990) suggest that when sheer persistence (effort) is the primary determinant of performance on the task, the training method should emphasize goal setting. When the task is relatively complex, thus requiring more flexibility in behavior to achieve top performance, self-management training may be more effective. Theoretical and empirical work by Noe (Noe, 1986; Noe & Schmitt, 1986) suggests that trainee attitudes are also an important factor affecting the success of training. For example, a trainee's (pre-training) job involvement is positively related to the learning demonstrated at the end of training. Job involvement also appears to be positively related to career planning, which in turn is positively related to the use of trained behaviors on the job (transfer) of newly learned skills (e.g., Baldwin & Ford, 1988; Noe, 1986).

CURRENT TRENDS IN MANAGERIAL TRAINING

Daining the Multinational Manager A relatively new are of interest in management development focuses on the training of managers for overseas assignments. In this section, we focus on three areas of interest: (1) the importance of training individuals for foreign assignments; (2) types of training techniques currently in use; and (3) the effectiveness of current cultural training programs. The importance of providing specialized training to managers given overseas assignments can be seen in a recent review. Consider the following statistics cited by Black and Mendenhall (1990): (1) Sixteen to forty percent of managers sent to work in foreign countries return prematurely; (2) Of those managers who complete assignments, approximately fifty percent are relatively ineffective performers; (3) The cost of one failure has been estimated to be between $50,000 and $150,000; and (4) The direct costs of expatriate failures to U.S. companies is estimated to be over two billion dollars per year. Given these data, and the fact that internationalization of business continues to grow, it is likely that companies will be looking for ways to ensure more successful expatriate assignments. Several types of programs for training managers given overseas assignments are available ranging from documentary materials on the new culture, like those used by Hilton International (Shames, 1986) to more experiential exercises which include case studies,

2 10

Racicot and Hall

cultural assimilators, role-plays, simulations, and field experiences. While documentary programs provide written information about the culture and its people, experiential methods involve more active learning allowing individuals to practice new behaviors (Shames, 1986). Because of the value of active learning, these methods are likely to be more effective (Bogorya, 1985). Perhaps more important than the specific technique used is the content included in the training program. The basic goal of most programs is to decrease misunderstandings resulting from conflicting cultural values. According to Shames (1986) the characteristics and skills most commonly referred to as important for success on expatriate assignments include the following: (1) awareness of personal values and beliefs; (2) the ability to understand the intrinsic logic of other cultures; (3) the ability to see others' perspective and put yourself in their place; (4)the ability to communicate verbally and non-verbally with people from other cultures; ( 5 ) the ability to change expectations and alter strategies as needed, the ability to be flexible; (6) awareness of the impact of culture shock on yourself and your adjustment to it; (7) the ability to establish supportive relationships with different people; and (8) the ability to manage diversity to the company's advantage. Another issue to be considered in the training of cross-national managers is the type of position to be filled. According to Clement (1988) top-level managers and department heads will have a greater need for cultural training than will individuals sent to deal with technical problems. In the former case, the assignment will tend to be long-term and require more interaction with foreign nationals. In the latter case, technical expertise will probably outweigh cultural skills. Thus, type of training required may be contingent on the specific situation. In addition to training concerns, selection of managers for overseas assignment is important. Again, for short term assignments of a technical nature, selection of individuals based on expertise will be more appropriate than for assignments which require more interpersonal interaction (Fisher, 1989). As with any type of training program, a careful needs analysis should be conducted before training and selection strategies are considered. One of the major obstacles to the implementation of crosscultural training programs appears to be the belief of top-management that it is either not necessary or not effective (Black & Mendenhall, 1990). Two recent articles shed some light on this concern. Earley (1987) compared the effects of documentary training to an interpersonal program which incorporated sensitivity training and field experience. The interpersonal program included a role-playing

Managerial Training

21 1

exercise, a simulation, and a field experience in a local Asian community. Participants were assigned to one of four conditions: (1) no training; (2) documentary training only; (3) interpersonal training only; or (4) documentary and interpersonal training combined. All of the participants were managers assigned to work in Korea and attended the training program before departure. Results indicated that those individuals who received training (either documentary or interpersonal) were better performers. In addition, they perceived fewer adjustment problems, and reported a more international outlook. Differences between interpersonal and documentary training methods were obtained only with respect to preferences which suggested a more positive reaction toward interpersonal training. In a recent review of the literature, Black and Mendenhall (1990) suggested that cross-cultural training was positively associated with feelings of well-being, self-confidence, relationship skills, and development of accurate perceptions of foreign cultures. In addition, positive relationships between training and cultural adjustment and performance were reported. Although many of the studies reviewed relied on self-report data, the results suggest that future investigations of the effectiveness of cross-cultural training would be worth while. In general, there appears to be a need for training programs which prepare managers for expatriate assignments. The behaviors discussed earlier may provide some guidance on where to direct efforts in terms of training content. The research on effectiveness, although scarce, suggests that even simple documentary programs can improve skills necessary for dealing with overseas assignments. However, a careful needs analysis will give better insight into the specific type of training needed and how better selection of candidates could increase the success rate of expatriate assignments.

Rapid Technological Advances Managers must deal with two general issues as they adjust to rapid changes in technology. First, they are likely to make increasing use of computerized and automated aids in their own day-to-day work. Secondly, they must develop new skills and a higher level of "technological literacy" as they are faced with supervising a more technologically sophisticated workforce. For example, managers may need to change their supervisory style in order to oversee workers who are telecommuting (using automation that allows them to work from a remote site). Unfortunately, we do not yet have a solid body of information about how training can and should

212

Racicot and Hall

address these needs. Managers play several key roles, depending upon their levellauthority within the organization. They may be faced with the decision of whether or not to recommend the adoption of a new technology. They may be involved in the design of its implementation, may work directly with employees who will be using the technology, or may use it themselves. Unfortunately, the decision to adopt a new technology may be based on the salesmanship of a vendor as much as any purely rational process, or even informed consideration of the organization's needs (Nutt, 1984). Some research on technology implementation suggests that managers typically overestimate ease of adoption effort and time to bring technology up to full capacity (e.g., Klein, Hall & Lalliberte, 1990). This is in part due to the unique nature of technology. This point is crucial because the manager's actions in the implementation process may well play a key role in ultimate success of the technology within the organization. Recent technology literature suggests that cognitions play an increasingly important role in the decisions that workers make while performing their job. Specifically, employees who work in heavily automated organizations often base their decisions about how to proceed on a "mental model" of the machine. Often this mental model does not adequately capture the functioning of the automation -- disasters may be an outcome of the decoupling of the mental model from reality (Weick, 1990).

Changes in the Composition of the Wrkforce The number of women and minorities (e.g., blacks, Hispanics) in the American workplace has been steadily increasing, resulting in increases in "nontraditional" candidates for supervisory and managerial positions (Fullerton, 1985; Geber, 1990; U. s. Department of Labor, 1984). There are strong external pressures, both in the form of legal requirements guiding hiring and promotion practices, and in the form of social norms, to ensure fair treatment and adequate representation of minority employees at all levels of the organization. Additionally, managers that previously supervised a very homogeneous work group must now learn to manage a more diverse population. We wish we could report that these transitions have occurred naturally and easily. Statistics tell us otherwise; minorities and women are still underrepresented in managerial positions (especially at top levels; Morrison & VonGlinow, 1990; U. S. Department of Labor, 1984). Surveys of racism and sexism in the workplace still show that a sizable percentage of American workers report

Managerial Training

213

holding stereotypes and racist and sexist attitudes (Fernandez, 1988). Although organizational interventions to address these problems must be varied in form (Morrison & VonGlinow, 1990), training and development efforts are likely to play an important role (DiTomaso, Thompson, & Blake, 1988; Fernandez, 1988). Supervisors and managers must deal with stereotyping, racism, and sexism from several perspectives. First, managers must have the interpersonal and human relations skills necessary to manage an increasingly diverse work force. In addition, non-minority managers need to gain an understanding of issues their minority counterparts must deal with. One approach to this is the presentation of race and gender awareness workshops (see Dixon & Stewart, 1988 for a description of three workshops conducted at AT&T). Second, programs designed to provide managerial training and development may need retailoring in order to achieve their objectives with female and minority trainees. In some cases, this retailoring will involve supplementary training (e.g., training in deficient skills such as assertiveness, training in managing affective reactions to racism or sexism at work). In other cases, it will simply involve making minor changes to current programs (e.g., including black models in behavior modeling videos, indicating whether some decision-making approaches are more or less likely to be successful depending upon the manager's gender). Training and development approaches that stress experience and informal learning are often seen as particularly important for women and minorities. Mentoring is a frequently suggested intervention for both women and minorities (Blake, 1988; DiTomaso, Thompson, & Blake, 1988; Fernandez, 1988). Companies employing mentoring programs need to be aware of special issues that may arise when the mentee is female or minority. For example, Ford (1988) questions how successful mentoring programs are, by showing that mentoring did not result in tangible outcomes such as higher pay or greater chance of promotion in a sample of minority administrators. As reviewed by Kelley and Streeter in this volume, women may face a number of barriers that make it difficult for them to establish cross-gender relationships, including limited contact with potential mentors, tokenism, stereotypic attitudes towards women on the part of potential mentors, gender differences in socialization practices, norms regarding cross-gender relationships, and gender differences in the acquisition and use of power (Noe, 1988). Finally, we must agree with Morrison and VonGlinow (1990) that much additional research is needed on women's and minority issues in training (women are better represented in the literature than minorities). The research needs range from macro studies of organi-

214

Racicot and Hall

zational practices and overall assessments of effectiveness, to micro studies of individual differences that may affect the success of training. For example, how extensive and successful have company efforts to use training to remedy minority and gender issues been? Can and should private organizations invest in programs to encourage minority students to complete college with majors in mathematics, engineering, computer science, business or economics -- given that college degrees are commonly prerequisites to all but lower level managerial positions (Spilerman, 1988)? To what extent and how should training programs be altered to meet the needs of special groups (for example, is low self-efficacy a factor that might confound training for women and minorities)?

A Trend toward Increasing Accountability This section discusses evaluation with special attention devoted to the issue of describing the outcomes of training in financial terms. Evaluation occurs when instructional activities are selected, adopted, valued, and modified to improve training decisions by systematically collecting descriptive and judgment-related data (Goldstein, 1986). Thus evaluation can yield information about whether a training program has any effect or not, compare the relative values of different training approaches, and in general aid in the decision of whether the effects of a given training program are large enough to justify the time, effort, and other resources spent. Evaluation efforts run the range from qualitative, formative evaluations that provide information to be used in an iterative fashion to fine-tune the training program to summative evaluations of the effects of training on the entire organization. Kirkpatrick (1959a, 1959b, 1960a, 1960b) proposed that measures of training effectiveness can be grouped into four categories or steps: reactions (employees' liking and emotional reactions to the training program), learning (the factual and procedural information learned in the program), behavior (application of the training material on the job), and results (organizational outcomes such as production increases, reduction of turnover, etc., desired to result from the program). Each of these criteria provides unique information about the success of the training program. There is evidence that the four types of measures show relatively low, and in some cases, no relationships with each other (Alliger & Janak, 1989). Behavior measures are important because they address the question of whether or not learning is transferred back into the job setting; results measures address the question of whether or not employees' increased knowledge, skills and abiIities benefit the organization.

Managerial Training

215

Although evaluators often assume that behavior and results measures are always superior to reactions or learning measures, the appropriateness of the criteria chosen depends upon the original training objectives (Alliger & Janak, 1989). A number of authors have noted the paucity of (outcome) evaluation of management training programs. In terms of hard criteria, reactions measures are more frequently collected than learning, behavior, or results measures (e.g., Goldstein, 1986; Saari et al., 1988). The problem is not lack of knowledge about how to evaluate, evaluation techniques and related issues are well covered in the training literature (for example, see Goldstein, 1986; Stone, 1982; Wexley & Latham, 1981). For those who argue that true experimental evaluation designs (requiring control group, random assignment, etc.) are difficult to implement in field settings, quasi-experimental approaches can still yield valuable effectiveness information although they may not eliminate all threats to interpretation (Cook & Campbell, 1979). Certainly they are an improvement upon no evaluation. Again, perhaps we need to increase our understanding of the context in which training and development decisions are made, by trying to understand this continued reluctance to undertake evaluation. Below we suggest several reasons why evaluation of managerial training programs in particular might pose problems, Results from managerial training may be more probabilistic than for non-managerial training. In some cases the knowledge and skills presented in training programs are being acquired because they might potentially be needed, i n unknown situations. This makes planned evaluation attempts uncontrollable because future situations, as well as the extent to which knowledge and skills are learned and used, determine the value of the training. How and when to collect measures of behavior and results becomes more uncertain. If this argument holds, then organizations that boast of positive evaluation findings for all training endeavors may in fact be indicating that their training strategy is too conservative. We could make an analogy with a gambler: in order to win some you have to lose some too. Not only are criteria a problem, but it may be more difficult to define the elements of the training program to be evaluated if it involves on-the-job components that may vary substantially from trainee to trainee (e.g., mentoring, job rotation, developmental assignments). Effectively, the number of "training programs" to be evaluated is equivalent to the number of trainees. Thus the evaluation efforts must focus on the more global, organization-wide effects likely to be associated with a diverse intervention. Power and political issues may be particularly strong factors in the evaluation of managerial training. The failure of a managerial

216

Racicot and Hall

training program as indicated by lackluster evaluation results may be perceived by managerial trainees as indicating a personal failure because they chose to invest time and effort in an ineffective training program. (The potential to find negative results in the evaluation of a managerial training is likely to be blocked because managers may wield considerably more political power than the persons who will conduct the evaluation.) In addition to evidence of training effectiveness as provided by the evaluation of reactions, learning, behavior, and results, organizations are increasingly requesting evidence that training is cost effective. As companies become more cost-conscious, they want to avoid the "error" of training persons who later don't develop into good managers. They also may be faced with the need to balance the benefits of specific managerial training programs against other training needs. In spite of this, Carnevale and Schulz (1990) report that financial accounting for training efforts is scant. Of a group of organizations polled by the American Society for Training and Development that perform evaluations, only 20 percent determined the financial benefits of the training programs to the organization. Interest in tying training results to dollars appears to be growing, however, as evidenced by the special supplementary section on accounting for training that appeared in a recent edition of the Training and Development Journal (Carnevale & Schulz, 1990). A variety of techniques can potentially be used to determine the financial benefits of managerial training ranging from simple to complex. For example, Carnevale and Schulz (1990) suggest a standardized accounting model, termed "consensus accounting, that makes it easier to monitor and control training related costs. This information can then be used for a variety of purposes including comparing costs across departments and organizations, making decisions about spending cuts, and collecting accurate cost information to use as an input to more sophisticated analyses such as the utility analyses described below. Another accounting approach (Schneider, 1990; Technical & Skills Training News, 1991) requires subject matter experts to first identify important supervisory competencies, weight them by their relative importance, then collect managers' estimates of the current competency levels of each supervisor they oversee. These current performance levels are compared to ratings of the level of competency needed to perform the job adequately. The discrepancies represent the Performance Improvement Potential (PIP) of each supervisor. A supervisor who performs at the 100% level on all necessary competencies is viewed as producing all of the value of the cost of their appointment (salary plus benefits plus overhead). To the extent that there is potential for improvement, the I'

Managerial Training

217

total value of the supervisor's appointment is pro-rated by the discrepancy between full competency and the rating of actual competency, the difference between this value and the 100%value is the dollar value of bringing the supervisory up to full competency level. Utility analysis techniques can be used to translate statistical estimates of the effect sizes of training interventions (e.g., correlations, differences between group means) into terms that express the financial benefit to the organization. To this end, various utility formulae have been developed (e.g., Schmidt, Hunter, & Pearlman, 1982). Utility calculations depend upon a number of terms that vary somewhat depending upon the model used but may include: the dollar value of a standard deviation change in dollar performance; the number of employees trained; the duration of training's effect on performance; the cost of providing training (per trainee); and various market conditions. Currently, a debate is raging about how best to estimate the value of a standard deviation in performance in dollar terms (see Bobko, Karren, & Kerkar, 1987; Burke & Frederick, 1984; Greer & Cascio, 1987; Raju, Burke, & Normand, 1990). Mathieu and Leonard (1987) demonstrate the application of utility analysis to the evaluation of a behavior modeling program aimed at improving the supervisory skills of bank employees. They based their analysis upon the Schmidt, Hunter, & Pearlman (1982) utility formula, with modifications based upon work by Boudreau (Boudreau, 1983; Boudreau & Berger, 1985) in order to include the effects of training over time, recurrent versus non-recurrent costs, and various economic factors such as discount and tax rates. The utility calculations suggested significant net utility to the organization for training a single group of sixty-two supervisors. In sum, the methods for evaluating training programs exist, and we need to use them. The techniques for doing utility analysis are still undergoing development. One of the needs in this area is not only formula development but information about the current use of such approaches in organizations -- what are the successes, unmet needs, and pitfalls? Additionally, the study of training and development can benefit from more empirically based "big picture" research such as the meta-analysis of managerial training program studies by Burke & Day (1986). They reviewed 70 studies of managerial training using an empirically based summary technique (meta-analysis). Their purpose was to appraise the effectiveness of the content (e.g., human relations, decision-making) and training methods (e.g., lecture, behavioral modeling) used, against several different types of evaluation criteria (e.g., subjective learning, objective results). Much as selection research attempts to build a base of information about the

Racicot and Hall

218

usefulness of various tests as predictors of performance, efforts like this to pull together our knowledge about the efficacy and strength of training programs is badly needed. For example, effect size estimates such as those made by Burke & Day (1986) might be useful in making utility projections for training programs to be developed. Another example of training research on a large scale is a study by Russell, Terborg, and Powers (1985) that investigated the relationship between training and organizational performance at the organizational level. In a sample of sixty-two retail stores (belonging to the same organization) they showed that two measures of store performance (sales volume per employee and store image) were significantly and positively related to the percentage of sale employees trained in that store. Additionally, a subjective measure of training emphasis (as perceived by sales personnel) was correlated with managers' and merchandising staffs' perceptions of store image. The training variables were more strongly related to store performance than another set of variables that assessed perceived supervisory and merchandising support (of the sales persons). Studies following a similar design, but focused on the overall organizational effects of training supervisory and managerial personnel, are badly needed,

SUMMARY In this chapter, we have suggested that managerial training is an important activity, and further, that it will become an even more important activity as organizations continue to face new challenges and demands. Research on current models of leadership as well as research on assessment centers and interpersonal skills training provide some support for the effectiveness of managerial training. Due to increasing demands originating from outside organizational boundaries (e.g., globalizing economy, changing demographics of the workforce, technological advances, increasing foreign competition), it appears that an open-systems approach to selecting and training managers is particularly appropriate. Increasing amounts of environmental uncertainty and change suggest that boundary spanning roles could become more important for managers of the future. At the same time, managers trained to protect the core of the organization from this uncertainty will also be important. Managers must be aware of how external events affect their organization and their subordinates. Furthermore, they must know how to deal with the effects of these changes. This may require more flexibility in thinking and behavior. Therefore, we argue that managerial training programs that take a broad perspective and that focus on both the

Managerial Daining

2 19

internal and external environment may prove useful in improving managerial performance.

REFERENCES Ackerman, L. (1985). Leadership vs. managership. Leadership and Organization Development Journal, 6, 17-19. Alliger, G. M., & Janak, E. A. (1989). Kirkpatrick's levels of training criteria: Thirty years later. Personnel Psychology, 42, 331-342. Alpander, G . G . (1986). Supervisory training programmes in major U. S. corporations. The Journal of Management Development, 5(5), 3-22. Baldwin, T., & Ford, J. K. (1988). Transfer of training: A review and directions for future research. Personnel Psychology, 41, 63- 105. Bernick, E. L., Kindley, R., & Pettit, K. K. (1984). The structure of training courses and the effects of hierarchy. Public Personnel Management Journal, 13, 109-119. Black, J. S., & Mendenhall, M. (1990). Cross-cultural training effectiveness: A review and a theoretical framework for future research. Academy of Management Review, 15, 113-136. Blake, H. D., DiTomaso, N., & Thompson, D. E. (1988). Action steps for corporations and graduate schools of business. In D. E. Thompson, & N. DiTomaso (Eds.), Ensuring minority success in corporate management. New York: Plenum Press. pp. 347-362. Bobko, P., Karren, R., & Kerkar, S. P. (1987). Systematic research needs for understanding supervisory-based estimates of SDy in utility analysis. Organizational Behavior and Human Decision Processes, 40, 69-95. Bogorya, Y. (1985). Intercultural training for managers involved in international business. Journal of Management Development, 4, 17-25. Bolt, J. F. (1985). Tailor executive development to strategy. Harvard Business Review, Nov-Dee, 168-176. Boudreau, J. (1983). Economic considerations in estimating the utility of human resource productivity programs. Personnel PSyChology, 36, 551-576. Boudreau, J. W., & Berger, C. J. (1985). Decision theoretic utility analysis applied to employee separations and acquisitions. Journal of Applied Psychology, 72, 387-392.

220

Racicot and Hall

Burke, M. J. & Day, R. R. (1986). A cumulative study of the effectiveness of managerial training. Journal of Applied Psychology, 71, 232-245. Burke, M. J:, & Frederick, J. T. (1984). l b o modified procedures for estimating standard deviations in utility analysis. Journal of Applied Psychology, 69, 482-489. Byham, W. C. (1971). The assessment-center as an aid in management development. naining & Development Journal, 25, 1022. Carnevale, A. P., & Schulz, E. R. (1990). Return on investment: Accounting for training. Supplement to the Training and Development Journal, 44, Sl-S32. Cascio, W. F. (1987). Applied psychology in personnel management (3rd ed.). Englewood Cliffs, NJ: Prentice-Hall. Clement, R. W. (1988). Management development in the 1980s: A field in transition. Journal of Management Development, 7, 45-55. Cook, T. D,, & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issues for field settings. Boston, MA: Houghton Mifflin Co. Crouch, A., & Yetton, P. (1987). Manager behavior, leadership style, and subordinate performance. Organizational Behavior and Human Decision Processes, 39, 384-396. Csoka, L. S., & Bons, P. M. (1978). Manipulating the situation to fit the Leader's style: Two validation studies of LEADER MATCH. Journal of Applied Psychology, 63, 295-300. Dansereau, F., Graen, G., & Haga, B. A. (1975). A vertical dyad linkage approach to leadership within formal organizations: A longitudinal investigation of the role making process. Organizational Behavior and Human Performance, 13, 46-78. Day, D. V., & Lord, R. G. (1988). Executive leadership and organizational performance: Suggestions for a new theory and methodology. Journal of Management, 14, 453-464. Deci, E. L., & Ryan, R. M. (1985). Zntrinsic motivation and self determination in human behavior. New York: Plenum Press. Decker, P. J., & Nathan, B. R. (1985). Behavior modeling training: Principles and applications. New York: Praeger. DiTomaso, N., Thompson, D. E., & Blake, D. H. (1988). Corporate perspectives on the advancement of minority managers. In D. E. Thompson and N. DiTomaso (Eds.), Ensuring minority success in corporate management. New York: Plenum Press. pp. 119-136.

Managerial Training

22 1

Dixon, E. P:,. & Stewart, R. (1988). Managing diversity in a competitive market. In D. E. Thompson and N. DiTomaso (Eds.), Ensuring minority success in corporate management. New York: Plenum Press. pp. 269-277. Earley, P. C. (1987). Intercultural training for managers: A comparison of documentary and interpersonal methods. Academy of Management Journal, 3U, 685-698. Fahr, J. L, & Dobbins, G. H. (1989). Effects of comparative performance information on the accuracy of self-ratings and agreement between self- and supervisor ratings. Journal of Applied Psychology, 74, 606-610. Fahr, J. L..,& Werbel, J. D. (1986). Effects of purpose of the apprasal and expectation of validation on self-appraisal leniency. Journal of Applied Psychology, 71, 527-529. Fahr, J. L., Werbel, J. D., & Bedeian, A. G. (1988). An empirical investigation of self-appraisal-based performance evaluation. Personnel Psychology, 41, 141-156. Fernandez, J. P. (1988). Racism and sexism in corporate America: Still not color- or gender-blind in the 1980s. In D. E. Thompson and N. DiTomaso (Eds.), Ensuring minority success in corporate management. New York: Plenum Press. pp. 71100. Fiedler, F. E. (1964). A contingency model of leadership effectiveness. In L. Berkowitz (Ed.), Advances in experimental social psychology, 1, New York: Academic Press. Fiedler, F. E. (1969). Style or circumstance: The leadership enigma. In Natemeyer, W. E., & Gilberg, J. S . (1989) Classics of organizational behavior (2nd ed.). Danville IL: The Interstate Printers and Publishers, Inc. Fiedler, F, E., & Mahar, L. (1979). A field experiment validating contingency model leadership training. Journal of Applied Psychology, 64, 247-254. Field, R. H. G. (1982). A test of the Vroom-Yetton normative model of leadership. Journal of Applied Psychology, 67, 523532. Fisher, C. D. (1989). Current and recurrent challenges in HRM. Special issue: Yearly review of management. Journal of Management, 15, 157-180. Ford, D. L. Jr. (1988). Minority and nonminority MBA progress in business. In D. E. Thompson, & N. DiTomaso (Eds.), Ensuring minority success in corporate management. New York: Plenum Press. pp. 57-69.

222

Racicot and Hall

Ford, J. K., & Noe, R. A, (1987). Self-assessed training needs: The effects of attitudes toward training, managerial level, and function. Personnel Psychology, 40, 39-53. Frayne, C., & Latham, G. P. (1987). Application of social learning theory to employee self-management of attendance. Journal of Applied Psychology, 72, 387-392. French, J. R. P., & Raven, B. H. (1960). The bases of social power, In Cartwright, D. & Zander, A. (Ed). Group dynamics: research and theory (2nd ed.). New York: Row, Petersen, 607-623. Fulk, J., & Wendler, E. R. (1982). Dimensionality of leader-subordinate interactions: A path-goal investigation. Organizational Behavior and Human Perjiormance, 30, 241-264. Fullerton, H. N., Jr. (1985). The 1995 labor force: BLS' latest projections. Monthly Labor Review, 11 7 (Nov.), 17-25. Garland, H., & Beard, J. F. (1979). Relationship between selfmonitoring and leader emergence across two task situations. Journal of Applied Psychology, 64, 72-76. Geber, B. (1990). Managing diversity. Training, July, 23-30. Gist, M. E. (1987). Self-efficacy: Implications for organizational behavior and human resource management. Academy of Management Review, 12, 472-485. Gist, M. E. (1989). The influence of training method on self-efficacy and idea generation among managers. Personnel Psychology, 42, 787-805. Gist, M. E., Bavetta, A. G., & Stevens, C. K. (1990). The effectiveness of self-management versus goal setting in facilitating training transfer. Best Papers Proceedings: Academy of Management, 1990, 50th Annual Meeting of the Academy of Management, San Francisco, CA. Gist, M. E., Schwoerer, C., & Rosen, B. (1989). Effects of alternative training methods on self-efficacv and Derformance in computer satware training. Journal Gf Apphed Psychology, 74. 884-891. Goldstein, I. L. (1986). Training in organizations: Needs assessment, development, and evaluation. Monterey, CA: Brooks/Cole. Goldstein, I. L., & Sorcher, M. (1974). Changing supervisory behavior. New York: Pergamon. Gordon, J. (1988). Doug Bray: You've got to pick your winners. Training, 25, 79-81. Graham, J. L., & Mihal, W. L (1986). Guidelines for designing management development programmes. Journal of Management Development, 40, 38-42.

Managerial Training

223

Greer, 0. L., & Cascio, W. F. (1987). Is cost accounting the answer? Comparison of two behaviorally based methods for estimating the standard deviation of job performance dollars with cost-accounting-based approach. Journal of Applied Psychology, 72, 588-595. Hersey, P., Blanchard, K. H., & Natemeyer, W. E. (1979). Situational leadership and power. In Natemeyer, W, E . , & Gilberg, J. S . (1989) Classics of organizational behavior (2nd ed.). Danville IL: The Interstate Printers and Publishers, Inc. Hill, L. A. & Elias, J. (1990). Retraining midcareer managers: Selfefficacy in learning and the organizational legacy. Boston: Harvard Business School, Division of Research. Hirsch, P. M. (1975). Organizational effectiveness and the institutional environment. Administrative Science Quarterly, 20, 327-344. Hosking, D., & Schriesheim, C. A. (1978). Review of "improving leadership effectiveness. " Administrative Science Quarterly, 23, 496-505. House, R. J., & Mitchell, T. R. (1974). Path goal theory of leadership. In Natemeyer, W. E., & Gilberg, J. s. (1989) Classics of organizational behavior (2nd 4.). Danville IL: The Interstate Printers and Publishers, Inc. Kanfer, F. H. (1980). Self-management methods. In F. H. Kanfer & A. P. Goldstein (Eds.), Helping people change: A textbook of methods. (2nd ed.). N Y Pergamon Press. pp. 334-389. Kaplan, R. E., Drath, W. H., & Kofodimos, J. R. (1987). High hurdles: The challenge of executive self-development. Academy of Management Executive, 1, 195-205. Katz, D. & Kahn, R. L. (1978). The social psychology of organizations (2nd ed.). New York: John Wiley & Sons. Keys, B., & Wolfe, J. (1988). Management education and development: Current issues and emerging trends. Journal of Management, 14, 205-229. Kirkpatrick, D. L. (1959a). Techniques for evaluating training programs. Journal of A S m , 13, 3-9. Kirkpatrick, D. L. (1959b). Techniques for evaluating training programs: Part 2 -- Learning. Journal of ASm, 13, 21-26. Kirkpatrick, D. L. (1960a). Techniques for evaluating training programs: Part 3 -- Behavior. Journal of ASm, 14, 13-18. Kirkpatrick, D. L. (1960b). Techniques for evaluating training programs: Part 2 -- Results. Journal of ASm, 14, 28-32.

224

Racicot and Hall

Klein, K. J.., Hall, R. J., & Lalliberte, M. (1990). Training in the organizational consequences of technological change: A case study of computer assisted design and drafting. In U. E. Gattiker & L. Larwood (Eds.), Technology and end-user training. New York: Walter DeGruyter. Komaki, J. L. (1986). Toward effective supervision: An operant analysis and comparison of managers at work. Journal of Applied Psychology, 71, 270-279. Komaki, J. L., Zlotnick, S., & Jensen, M. (1986). Development of an operant-based taxonomy an observational index of supervisory behavior. Journal of Applied Psychology, 71, 260-269. Latham, G. P., & Frayne, C. (1989). Self-management training for increasing job attendance: A follow-up and a replication. Journal of Applied Psychology, 74, 41 1-416. Liden, R. C., & Graen, G. (1980). Generalizability of the vertical dyad linkage model of leadership, Academy of Management Journal, 23, 451-465. Mabe, P. A.? 111, & West, S. B. (1982). Validity of self-evaluation of ability: A review and meta-analysis. Journal of Applied P~chology,67, 280-296. Manz, C. C., & Sims, H. P., Jr. (1986). Beyond imitation: Complex behavioral and affective linkages resulting from exposure to leadership training models. Journal of Applied Psychology, 71, 571-578. Mathieu, J. E:, & Leonard, R. L. (1987). Applying utility concepts to a training program in supervisory skills: A time-based approach. Academy of Management Journal, 30, 3 16-335. Mayer, S. J., & Russell, J. S. (1987). Behavior modeling training in organizations: Concerns and conclusions. Journal of Management, 13, 2 1-40. McCall, M. W., Jr., Lombardo, M. M., & Morrison, A. M. (1988). The lessons of experience: How successjid executives develop on thejob. Lexington, MA: Lexington Books. McEnery, J., & McEnery, J. M. (1987). Self-rating in management training needs assessment: A neglected opportunity? Journal of Occupational Psychology, 60, 49-60. McGehee, W., & Thayer, P. W. (1961). Training in business and industry. NY John Wiley & Sons. Miner, J. B., & Miner, M. G. (1977). Personnel and industrial relations (3rd Ed.). NY Macmillan. Moore, M. I., & Dutton, P. (1978). Training needs analysis: Review and critique. Academy of Management Review, 3 , 532-545.

Managerial Training

225

Morrison, A. M., & VonGlinow, M. A . (1990). Women and minorities in management. American Psychologist, 45, 200208. Neider, L. L., & Schriesheim, C. A. (1988). Making leadership effective: A three stage model. Special issue: Developing leadership excellence. Journal of Management Development, 7, 10-20. Noe, R. A. (1986). Trainees' attributes and attitudes: Neglected influences on training effectiveness. Academy of Management Journal, 11, 736-749. Noe, R. A. (1988). Women and mentoring: A review and research agenda. Academy of Management Review, 13, 65-78. Noe, R. A., & Schmitt, N. (1986). The influence of trainee attitudes on training effectiveness: Test of a model. Personnel Psychology, 39, 497-523, Nutt, P. A. (1984). Types of organizational decision process. Administrative Science Quarterly, 29, 4 14-450. Oppenheimer, R. J. (1982). An alternative approach to assessing management development needs. Training and Development Journal, 36, 72-76. Peters, L. H., Hartke, D. D., & Pohlmann, J. T. (1985). Fiedler's contingency theory or leadership: An application of the metaanalysis procedures of Schmidt and Hunter. Psychological Bulletin, 97, 274-285. Raju, N. S,, Burke, M. J., & Normand, J. (1990). A new approach for utility analysis. Journal of Applied Psychology, 75, 3-12. Ripkey, D. R. (1990). Self-monitoring: A predictor of managerial success? Unpublished Master's Thesis, Rensselaer Polytechnic Institute. Russell, J. S . , Terborg, J. R., & Powers, M. L. (1985). Organizational performance and organizational level training and support. Personnel Psychology, 38, 849-863. Saari, L. M., Johnson, T. R., McLaughlin, S. D.?& Zimmerle, D. M. (1988). A survey of management training and education practices in U. S . companies. Personnel Psychology, 41, 731743. Scandura, T. A., & Graen, G. B. (1984). Moderating effects of initial leader-member exchange status on the effects of a leadership intervention. Journal of Applied Psychology, 69, 428436. Schmidt, F. L., Hunter, J. E., & Pearlman, K. (1982). Assessing the economic impact of personnel programs on workplace productivity. Personnel Psychology, 35, 333-347.

226

Racicot and Hall

Schneider, H. (1990). Return on training investment: Hard measures for soft subjects. Performance and Instruction, 29, 2835. Schneier, C. E., Guthrie, J. P., & Olian, J. D. (1988). A practical approachto conducting and using the training needs assessment. Public Personnel Management, 17, 191-205. Shames, G. (1986). Training for the multicultural workplace. Cornell Hotel & Restaurant Administration Quarterly, 26, 2531. Spilerman, S. (1988). Sources of minority underrepresentation in corporate employment. In D. E. Thompson and N. DiTomaso (Eds.), Ensuring minority success in corporate management. New York: Plenum Press. pp. 25-36. Stone, E. F. (1982). Research design: Issues in studies assessing the effects of management education. In R. D. Freedman, C. L. Cooper, and S. A. Stumpf (Eds.), Management education @p. 87-132). John Wiley & Sons. Strube, M. J., & Garcia, J. E. (1981). A meta-analytic investigation of Fiedler's contingency model of leadership effectiveness. Psychological Bulletin, 90,307-321. Technical & Skills Training News. (1991). The dollar value of training technical managers. Winter, 3-7. Thornton, G. C., 111, & Cleveland, J. N. (1990). Developing managerial talent through simulation. American Psychologist, 45, 190-199. Tushman, M. L., & Scanlan, T. J. (1981). Boundary spanning individuals: Their role in information transfer and their antecedents. Academy of Management Journal, 24, 289-305. U. S. Department of Labor. 20 Facts on Ubmen Ubrkers. Washington, D. C.: U.S. Government Printing Office. Vecchio, R. P. (1987). Situational leadership theory: An examination of a prescriptive theory. Journal of Applied Psychology, 72, 444-451. Vecchio, R. P., & Gobdel, B. C. (1984). The vertical dyad linkage model of leadership: Problems and prospects. Organizational Behavior and Human Performance, 34, 5-20. Vroom, V. H . , & Jago, A. G. (1978). On the validity of the Vroom-Yetton model. Journal of Applied Psychology, 63, 151- 162. Vroom, V. H., & Yetton, P. W. (1973). Leadership and decisionmaking. Pittsburgh, PA: University of Pittsburgh Press.

Managerial Training

227

Weick, K. E. (1990). Technology as equivoque: Sensemaking in new technologies. In P. F. Goodman, L. S. Sproull, & Associates (Eds.), Technology and organizations. San Francisco: Josey Bass. pp. 1-44. Wexley, K. N., & Baldwin, T. T. (1986a). Management development. In J. G. Hunt & J. D. Blair (Eds.), 1986 Early Review of Management of the Journal of Management, 12, 277-294. Wexley, K. N., & Baldwin, T. (1986b). Post-training strategies for facilitating positive transfer: A n empirical exploration. Academy of Management Journal, 29, 503-520. Wexley, K. N., & Latham, G. P. (1981). Developing and training human resources in organizations. Glenview, IL: Scott, Foresman . Yukl, G. A . (1981). Leadership in organizations. Englewood Cliffs, NJ : Prentice-Hall. Yukl, G. A, (1989). Leadership in organizations (2nd ed.). Englewood Cliffs, NJ: Prentice Hall. Yukl, G. A., & Nemeroff, W. F. (1979). Identification and measurement of specific categories of leadership behavior: A progress report. In J. G. Hunt, & L. L. Larson (Eds.), Crosscurrents in leadership. Carbondale, IL: Southern Illinois University Press. Yukl, G. A., & Van Fleet, D. D. (1982). Cross-situational, multimethod research on military leader effectiveness. Organizational Behavior and Human Pet$ormance, 30, 87-108.

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial/ Organizational Psychology - K . Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

229

CHAPTER 8 ORGANIZATIONAL NEEDS ASSESSMENT: TECHNOLOGY AND USE Barbara A. Caska, Kathryn Kelley, and Edward W Christensen INTRODUCTION This chapter presents an analysis of techniques to gather information during an organizational needs assessment. Needs assessment is part of an organizational diagnosis, carried out to determine how to increase an organization's effectiveness. Qpically, an organizational diagnosis will focus on aspects of behavior (Lawler, Nadler, & Cammann, 1980). Since it is aimed at increasing organizational effectiveness, the diagnosis is oriented towards action. Information will be used by clients requesting the diagnosis to determine some course of action. This distinguishes assessment research from the more traditional types of behavioral research. It is not enough to determine simply what is happening in the organization; this research results in specific recommendations for changes. As a type of behavioral research, organizational diagnoses are conducted according to principles of scientific investigation. At the same time, however, methods employed and resulting data should be acceptable to those persons requesting the diagnosis, and should be appropriate for their organization, To an extent, then, the methods of research used in a diagnosis will depend on the unique features of that particular organization. Perhaps because research design can be unique to each situation, limited research has been conducted to determine which techniques can be most effectively used to gather data during organizational diagnoses. Characteristics of each technique are discussed broadly, but these characteristics are usually not, however, priori-

230

Caska, kidley, and Christensen

tized in desirability. A simple conclusion that technique choice will depend on the particular client and organization is typical. While research design will be influenced by the situation, technique selection should also be influenced by a more stable set of factors: the qualities or characteristics inherent in different techniques. It is possible to evaluate technique selection and use according to presence of those qualities determined to be consistently relevant to organizational diagnoses. This chapter presents an attempt to develop a classification of technique characteristics. The classification was developed according to considerations for both scientific justification and the action orientation of a diagnosis. A variety of techniques used during organizational diagnoses will be evaluated according to this classification. Literature suggests the use of two or more methods to gather data for a needs assessment (Lawler et al., 1980; McKillip, 1987; Webb, Campbell, Schwartz, & Sechrest, 1966). Complementary combinations of techniques, however, are not typically indicated. A second purpose of this chapter is to examine which techniques might be effectively used together in an organizational diagnosis. This will be determined by examining the relevant characteristics of the various methods, after considering general requirements for any diagnosis. The chapter begins by defining an organizational needs assessment, and explaining its role in a diagnosis. Qualities of an effective assessment will be discussed. Next, considerations in deciding which techniques to employ during data collection will be examined. These considerations include both unique organizational factors, and the somewhat less situational-dependent technique characteristics. Techniques will be evaluated according their characteristics. Criteria for defining certain characteristics as relevant considerations in the design of the diagnosis will be explained. Following this, suggestions for complimentary combinations of techniques, according to these characteristics, will be provided. Finally, four examples of technique selection and use during organizational needs assessments will be presented.

THE ROLE OF NEEDS ASSESSMENT IN ORGANIZATIONAL DIAGNOSIS Consultants are called into organizations to recommend solutions to problems, suggest ways to handle new challenges, or to develop programs to improve functioning. A consultant conducts an organizational diagnosis to provide the client with such information:

Organizational Needs Assessment

23 1

behavioral science is used to assess an organization's current state and to discover routes to improvement (Harrison, 1987). A diagnosis allows the consultant to compare the current state of the organization to a preferred state, and to determine ways to narrow the gap between the two. This procedure involves, specifically, the collection, interpretation and use of data from the organization to carry out the purposes of the organizational diagnosis. Figure 1 outline the process of diagnosis carried out by the consultant (Block, 1981; Harrison, 1987; Lawler et al., 1980). During organizational diagnosis, the initial clientlconsultant meeting is referred to as scouting. At this time the clients will present problems to the consultant. Problems brought up at this time are usually mere symptoms of the real problem (Block, 1981). A needs assessment will provide a more comprehensive view of what is actually occumng in the organization. During these initial meetings consultants begin to collect information about it. The contracting phase will follow scouting. During this phase the client and consultant will work out agreements on the objectives of the analysis and types of feedback expected. The organizational needs assessment can then be conducted. As indicated in Figure 1, the assessment includes designing the study, selecting techniques to use during data collection, actual data collection and data analysis. The organizational needs assessment begins with the design of the study. Specifications on dimensions to be studied and the scope of the assessment will aid the consultant in deciding which data gathering techniques to use. Other factors influencing the choice of data gathering methods include organizational characteristics, client feedback expectations, models of organizational functioning employed by the consultant, and limitations imposed by the client or situation. (These factors will be more fully discussed later). After data gathering techniques have been selected, data collection can begin. As information is collected, the techniques used may be altered to broaden or narrow the focus of the investigation or to deal with unanticipated factors. Informal, quick analysis of information being gathered will allow the consultant to determine whether changes in the techniques are required. This is indicated in Figure 1 by a broken line linking analysis to choice of technique. After all needed data are collected, final data analysis is carried out by the consultant(s). Following the organizational needs assessment, the consultant will provide feedback to the clients. Recommendations for action are made at this time. Final decisions on actions to be carried out may or may not include discussion with the consultant.

232

Caska, Kdley, and Christensen

SCOUTING

I

1 CONTRACTING 1 ORGANIZATIONAL NEEDS ASSESSMENT Study Design Choice of Techniques Data Collection Analysis

I FEEDBACK I I RECOMMENDATIONS I

I DECISIONS ON ACTIONS I Figure 1 . The sequence of procedures followed during an organizational diagnosis.

Organizational Needs Assessment

233

ORGANIZATIONAL NEEDS ASSESSMENT AND SCIENTIFIC RESEARCH As a type of behavioral research, an organizational needs assessment should be conducted according to methods of science. The assessment should be guided by theory. Models of organizational functioning are used to select variables to be measure, specify useful data gathering techniques and anticipate results or relationships found variables. Data gathering instruments should be theoretically justified, valid and reliable. In addition, the study should be conducted with some degree of skepticism: a healthy attitude toward any scientific inquiry (Lawler et al., 1980). Though an organizational needs assessment is a type of research, it differs from traditional studies of human behavior in a number of ways. One difference is the initial broad scope of the investigation. According to a holistic perspective, the organization is viewed as a complex, dynamic social system, operating within its environment and interacting with other systems (Lawler et al., 1980). Variables that are considered to be relevant for understanding, predicting, or controlling the particular behavior of interest are selected from the complete functioning organization. These variables will be assessed i n the context of the overall organizational system. Organizational needs assessments differ from other types of behavioral research in their purpose. They are conducted to identify or solve problems, reshape the organization, determine consequences of actions taken, and to generally contribute toward the development of effectiveness. As Block (1981) points out, needs assessments are not done simply to understand the situation. Rather, they are part of a diagnosis conducted to mobilize action. Client considerations will greatly influence the way in which the assessment is carried out. This is another unique feature of assessment research. The consultant cannot act simply as a scientist doing independent research. Instead, the consultant should be concerned with the client's attitude toward assessment proceedings. If there is a negative attitude toward the research, the consultant may not receive the support and cooperation needed to complete the study successfully. Client-imposed limitations may influence the scope of the study. The consultant should focus on assessing factors that are under the control of the client. According to Block (1981), clients are those individuals who attend initial assessment planning meetings, set up objectives, receive feedback, and make decisions on

234

Caska, Kklley, and Christensen

actions to take, Clients mentioned up to this point have been defined as such. Typically, they are high level managers in an organization. It is clear that their input, limitations and expectations will influence the nature of the assessment. However, another type of client may also influence the success of the study: lower level managers and other organizational members who directly provide data, or access to data. These individuals can determine the effectiveness of various techniques used during data collection. They can either cooperate, providing accurate, useful data, or refuse to cooperate, disrupting the entire study. The attitudes of these organizational members toward the assessment, therefore, are an important consideration. Such attitudes can differ vastly from attitudes of the contracting client. The consultant should consider these potential differences when designing the needs assessment. The approval of both types of clients is needed for an accurate, useful assessment. Another distinguishing feature of organizationalneeds assessment research is frequent reliance on qualitative, rather than quantitative data. Traditionally, quantitative, or "hard, data have been considered to be the only truly useful data in scientific research. Evidence does not support this conclusion. As pointed out by Rossi & Freeman (1985), qualitative data offer a unique utility to the assessment. Attitudes and feelings can be revealed with qualitative data. Actual accounts of events and sequences of events can be studied. Such data, which may be described as rich, cannot be collected through the use of strictly quantitative methods (Lawler et al., 1980). The consultant relies on his or her own expertise, perceptions, or even intuition when assessing an organization. As explained by Helmer (1983), the explicit use of expert judgment is not incompatible with scientific objectivity. Such judgments should, rather, be incorporated in the structure of the investigation. They can, for example, be used to check the accuracy of data that has been formally gathered. A final difference between traditional scientific research and the organizational needs assessment is the audience receiving findings. Results of an assessment should be presented to the client in terms that can be easily and clearly understood. Findings should be conveyed in language that will be understood by the client. Scientific terms may have to be eliminated. These results may additionally be have to be persuasive, aimed at convincing the contracting client to implement suggested actions. Such considerations will influence data presentation. 'I

Organizational Needs Assessment

235

COMPONENTS OF AN EFFECTIVE NEEDS ASSESSMENT The quality of the needs assessment depends on both the usefulness of results to the client, and on the accuracy of the description of the organization. In addition to the requirements for utility for the client (just discussed), a assessment should also feature the use of appropriate methods and reliance on sound theoretical models.

meoretical Basis Organizational models are used by consultantsto specify why particular pieces of information should be chosen for assessment and how they should relate to each other (Lawler et al., 1980). Models of organizational functioning are developed through an integration of various areas of behavioral science. The consultant will rely on experience and expertise when selecting those models to guide the assessment. Unfortunately, research has not yet indicated general superiority of one particular model over others, even in dealing with specific problems (Harrison, 1987; Lawler et al., 1980). Existing research on models is largely unintegrated, or not scientifically examined in a variety of situations. Therefore, the particular model or models used by a consultant must be evaluated for accuracy according to predictive ability and power, in comparison to other models. Confidence can be placed in a model that is capable of accurate predictions, and that is consistent with information revealed by an organizational assessment. The most powerful models are those that are capable of making the most accurate predictions, and explaining the domain of interest using the fewest number of variables (Lawler et al., 1980). It will be assumed that the consultant has the experience and expertise to choose and apply valid, powerful models when selecting information to assess.

Appropriate Techniques Instruments or techniques used to gather data should be capable of providing a valid and useful diagnosis. Lawler et al. (1980) discuss four essential qualities of adequate data gathering instruments. The instrument should be sensitive, or able to produce variable results when used in a range of different situations. It should be able to differentiate among objects that have been assumed to be different. Secondly, the instrument should be reliable, producing measurements that remain constant under conditions that have been

Caska, filley, and Christensen

236

assumed to be stable. The instruments must display both convergent and discriminant validity. Convergent validity refers to the extent to which different instruments, designed to measure the same construct, produce similar results when used in the same situation. Presence or absence of convergent validity has been typically determined by evaluating correlations among measurements provided by different techniques in the same situation. Discriminant validity refers to the extent to which instruments designed to measure different constructs produce varying results. Evaluating correlations among measurements of different concepts taken in the same set of situations can indicate discriminant validity (the possibility that such differences in measures could be due simply to differences in the method should be considered additionally). At least two techniques are required to test for convergent and discriminant validity in an assessment. Techniques displaying these qualities can be considered valid, and, within the context of accurate models, will enable the consultant to describe accurately the functioning organization. The selection and use of data gathering techniques are central activities in the organizational diagnosis. The methods used to gather data will influence results of the organizational assessment. Success of the diagnosis depends upon choosing adequate techniques that are relevant to the particular situation, and will provide results that are useful and acceptable to the client.

INFLUENCES ON TECHNIQUE SELECTION Initial Assessment Phases The organizational needs assessment begins with a choice of methods and procedures to be used to gather information. Various unique organizational and situational factors will influence that choice. As indicated i n Figure 1, the actual needs assessment is carried out after the scouting and contracting phases of the organizational diagnosis been completed. During those two phases decisions were made and information revealed that will influence the design of the assessment. The consultant begins to collect preliminary information when he or she initially enters the organization. This occurs prior to the actual assessment. Initially, the client will present a particular problem or situation to the consultant. As indicated previously, although the presented problem is probably just an indication or symptom of the real problem, it gives the consultant some idea of what is occurring in the organization. By the end of contracting the client will have indicated approximate amounts of money and time

Organizational Nee& Assessment

237

available to complete the assessment. Joint client/consultant decisions will have been made on the goals of the study, the scope of the analysis, the participation required from various organizational members, and the type of feedback to be provided (Block, 1981). Dimensions for study will have been broadly determined. The consultant will also gather organizational information in a less obvious manner during initial phases of the diagnosis. The attitude and motivation of the clients will influence assessment decisions made by consultant. Casual contact with other organizational members can give some indication of their attitudes toward the study, and cooperation that can be expected. In addition, the position of the client in the organization will indicate the amount of control he or she will have over various parts of the organization, and the degree of freedom to commit various areas to study. Various other types of information about the organization may be considered before deciding how to carry out the assessment. Harrison (1987) points out that success of data gathering methods can be influenced by organization size, complexity, purpose, technological type, work force composition, degree of bureaucratization, and the stage in its life cycle. In addition, the consultant may consider the organization's strategy for coping with the environment it functions within, and that environment's predictability. The organization's definition of effectiveness should be considered, also. It can be seen that a great deal of information gathered prior to the actual beginning of the needs assessment will influence the consultant's choice of data gathering methods. Such information is unique to that particular situation. It will be considered by the consultant according to his or her own expert knowledge and use of models of organizational functioning. It would be difficult to examine systematically the influence of such information on choice of data collection techniques; it consists of virtually limitless, unique, situation dependent pieces of data, that have been selectively gathered and interpreted by the consultant according to his or her own prior experiences and knowledge. Such information is, in addition, only part of the considerations that will determine the final choice of data collection methods. Qualities, or characteristics, of the various techniques should also be taken into account.

Need For n o Or More Techniques At least two different data gathering techniques should be used during an assessment to ensure an accurate diagnosis. As pointed out by Butler & Butler (1987), a needs assessment requires an integration of research methods, to avoid false confidence in information

238

Caska, Kelley, and Christensen

revealed by one method. This concept is not unique to assessment research. Triangulation, the notion that measurement of a construct requires the use of more than one instrument, has been used in social science research for over 30 years (McKillip, 1987). The reasoning behind triangulation is that techniques used to gather data are subject to several sources of error variance. Specific potential errors in various instruments will be addressed later. For now, it is important to note that no one method is error free. Webb et al. (1966) note that more confidence can be placed in a proposition that has been investigated through multiple measures, since each individual measure will have unique sources of error. By using two or more methods, it is possible to compensate for errors and drawbacks associated with one particular method. It is necessary to use two or more data gathering techniques to ensure that techniques being used are valid. As explained earlier in this paper, methods should display both convergent and divergent validity. Tests for such validity require comparisons of data collected by at least two methods. Another argument for the use of two or more techniques to gather data comes from the simple fact that different methods will produce different types of data. For example, some techniques will reveal data that can be classified as generally quantitative; others, more qualitative. As pointed out by Lawler et al. (1980),multiple methods should be used in an adaptive and mutually confirmatory way to check on the different types of information revealed by different techniques. In this way, it is possible to cross-check the meaning of information and to avoid misinterpretation of the data. It is essential that the meaning of data gathered during an organizational needs assessment is accurately interpreted. These data will be used by the consultant to provide feedback and suggestions for intervention in the organization. An ineffective or inaccurate diagnosis will result if data are not correctly gathered and interpreted. If a client uses inaccurate recommendations and feedback in an attempt to solve a problem or implement a change, the situation could worsen. The organization will not become more effective, and could even encounter greater problems. Such a result could be disastrous and very costly to an organization already facing problems, In addition, an inaccurate diagnosis would be a great waste of both the client's and consultant's time.

EVALUATION OF TECHNIQUES Each method of data collection has unique qualities. Some qualities may be considered generally desirable or undesirable. Other

Organizational Needs Assessment

239

qualities may or may not be required, depending on the particular organizational situation. It is now possible to analyze some of the more commonly used data gathering methods. A review of literature on technique characteristics and their relative importance has provided a system to analyze the quality of data gathering methods, and suggestions for effectively combining techniques in organizational assessments. n b l e 1 is an outline of those dimensions considered to be of greatest significance in technique choice. These dimensions were selected according to concerns for scientific justification and appropriateness to the organizational situation and client. The dimensions are categorized as characteristics that are typically desired in a technique, characteristics that should be considered as situational dependent, and those that should be present in at least one of the techniques used during any assessment. Table 1 first lists generally desired technique qualities. Low financial cost is almost always desired. These costs include the obvious material and consultant fees. They also include losses resulting from requirements for members being removed from their work tasks to participate in data collection. Methods that cost less will usually be simpler to conduct and interpret. This means that data can be quickly collected and analyzed. Clients will likely prefer the assessment and diagnosis to be completed quickly. Face validity is also generally desirable. Organizational members providing data, and clients receiving results, must be convinced that the method is capable of providing important, accurate results. Actually to provide accurate results, the methods must be valid and reliable. Next, Thble 1 lists characteristics that should be considered as situationally dependent. Technique choice should involve consideration of the level of analysis required and member participation. Depending on the type of technique used and the size of the sample, data can be collected on an individual, group, divisional or organizational level. Techniques should be able to provide levels of analysis that are appropriate to the particular organizational situation, The same is true of required member participation. Data collection can range from that requiring little member participation, or even realization that it is occurring, to complete dependence upon member cooperation. It is necessary to choose techniques that require an appropriate amount or type of participation in that organizational situation. Three additional dimensions are listed in n b l e 1 as characteristics that should be present in at least one technique used during an assessment. These qualities are not necessarily desirable in all techniques used during one particular assessment. However, it is impor-

Caska, a l l e y , and Christensen

240

Table 1 Characteristics to Be Considered

Techniaue Choice

Generally Desirable;

* * *

*

Low cost -Simplicity -Rapid completion Face validity Statistical validity Reliability

Situationally Dependent:

* Level of analysis

*

Required member participation

Required Of At Least One Technique:

* * *

Flexibility Capability of providing qualitative data Nonreactiviiy

tant to consider the presence of such qualities when choosing combinations of techniques. At least one method should be flexible, or adaptable. As the assessment progresses, unanticipated areas of focus may prove important. Initially designed techniques may not be providing all data that are required. Adaptable techniques give the consultant freedom to adjust to such situations. At least one method should provide qualitative data. This will allow the consultant to make certain that quantitative results are in line with attitudes, feelings, and general experiences of members of the organization. Lawler et al. (1980) suggest that at a minimum, data gathering measures should be checked against the experiences of evaluators and people inside the organization. In addition, at least one data gathering technique should be nonreactive. According to Webb et al. (1966), reactive measurements, such as those obtained through the use of questionnaires or interviews, will provide results that are subject to respondent biases. Organizational members may provide contaminated data simply because they are aware that they are providing data. Members may

Organizational Needs Assessment

24 1

also consciously or unconsciously play a certain self-selected role during data collection, saying or doing what they believe the consultant is interested in. In addition, reactive effects can occur if members fall into certain response sets. An assessment based solely on data containing respondent errors will likely be inaccurate. Data gathering techniques vary in their tendency to provide reactive results. Those that involve collection of data without direct respondent participation are considered to nonreactive (Webb et al., 1966). Archival data analysis and observation are examples of nonreactive techniques. It is desirable to use at least such method during an assessment to check respondent biases.

TECHNIQUES AND THEIR CHARACTERISTICS Techniques that may be used to gather data during organizational needs assessments will be now evaluated. Each technique will be briefly described. Presence of characteristics, or the extent to which each technique is capable of providing essential qualities in the assessment will be discussed.

Obsenation Data can be gathered by observing organizational members. This technique allows the researcher to focus on actual behavior, as it is occurring. The method is, therefore, distinguished from those that collect information on past behavior. Observation allows the consultant to preserve more easily the complexity of the situation being studied, and to view it from a holistic framework. Observations can be unstructured, semi-structured, or structured (Lawler et al., 1980). Completely unstructured observation is not typically used, except to get a general feeling for what is occurring in the organization at the beginning of the diagnosis. Semi-structured observation involves some prior specification of behavior to focus on, and specified methods of data recording. Structured observation requires exact specification of what is to be observed and how data are to be recorded. Observation techniques are expensive (McKillip, 1987). A great deal of time is needed to use this method. Ideally, observations will be made by two or more persons, to guarantee some reliability. Each observer should be trained to observe and record data accurately. I n addition, physical demands and cognitive strain are placed upon the observers. Sampling should occur over periods of time and events, as well as across various organizational members.

242

Cash, &lley, and Christensen

These factors add to the costs involved in using observations to gather data. Observation can have high face validity, if the observer is visible to members providing data (Lawler et al., 1980). Statistical validity may be difficult to assess, since specifics on exactly what to observe and record will, to some extent, be left up to the individual observer. Validity increases as structure increases. The same is true of reliability. In less structured methods, observers have a wider range of choice in deciding exactly what information to collect. This increases the likelihood of error (Hanlon, 1980). The use of two or more observers can somewhat improve reliability. Biases of observers in recording are a potential source of error (Harrison, 1987). Observation allows for analysis at an individual or group level. Hanlon (1980) points out that individual or group behavior can be interpreted in terms of social context when this method is employed. Information of interpersonal relationships may therefore be available. This technique does not require direct participation by organizational members. If the observer is in the work environment, member cooperation is required. Less structured observation is highly flexible and adaptive (Hanlon, 1980). Structured observation will have moderate to low adaptability. Observation can provide qualitative data. This characteristic is more inherent in unstructured or semi-structured types. Data can be more readily quantified in the structured observation. Quantification can be difficult, however, involving time consuming frequency counts of particular events or specific anticipated behaviors (Hanlon, 1980). Reactive error may bias data. Reactivity is not as great a problem with this method as it is with self-report measures, however. Hanlon (1980) concludes that observation is especially suited to gather data when organizational change is the focus of study. Observation is also useful for investigating organizations that seem to be functioning irrationally, as indicated by organizational models being employed and previously gathered data. It is also a useful technique if the organization has been measured or undergone assessment numerous times already, and member cooperation can not be expected.

Archival Measures The use of archival data involves gathering and analyzing organizational documents, reports or other available written materials from the organization's data banks (Lawler et al., 1980). Data will reveal information about past events or behaviors. Whether or

Organizational Needs Assessment

243

not this method can be used will depend upon the client's ability or willingness to grant the consultant access to such organizational information. This method may or may not be expensive to use. It will depend upon how easily the stored information is coded and quantified (Harrison, 1987). Low costs are possible if quantification goes quickly, since no new data need be collected. Time requirements for method use and simplicity of use will also depend upon ease of coding and quantification. Lawler et al. (1980) point out that archival data collection has high face validity. Statistical validity may be a severe problem. Archival data may have been invalidly recorded, intentionally or unintentionally (Marvis, 1980). An additional source of bias may come from the fact that it is unlikely that all relevant data have been recorded, or accessible to the consultant. Data may also be interpreted out of context (Harrison, 1987). Error may also come from the person analyzing the archival data during coding and classification. Because data can be analyzed and quantified in many various ways, reliability can be low. This technique lacks flexibility. It typically provides quantitative, rather than qualitative data. The data are almost completely nonreactive. Archival data collection is useful when the written data sources can be assumed valid. Sources should ideally be in a form that is easily retrieved, coded and quantified.

Questionnaires Questionnaires or surveys may be used to gather organizational information. These are self-administered written selections of questions. There are three common types: standardized fixed response, custom fixed response or custom free response (Lawler et al., 1980). The standardized fixed response type has been previously developed, tested and used. Respondents must choose from a limited number of possible answers for each question. The custom fixed response offers, similarly, a limited choice in answer selection. This questionnaire, however, will be uniquely designed for the particular organizational situation being analyzed. The custom free response is also custom designed. Respondents are not, however, limited to a predetermined set of possible answers to choose from. This type of questionnaire provides open-ended questions for the respondent to answer in any way he or she chooses. The fixed response questionnaires are relatively low in cost; the free response is moderate in price (Lawler et al., 1980). The higher cost associated with the free response method results from the additional time required for coding and analyzing answers. It is

244

Caska, Kelley, and Christensen

simpler to use questionnaires with fixed responses. In addition, it is easier to use a previously designed questionnaire than to design one to fit the particular situation. Standardized fixed response techniques have the advantages of being refined over time, being based on models of organizational functioning, and being pretested and compared with normative data. As Harrison (1987) warns, however, the consultant should avoid choosing a previously designed survey because of ease or familiarity. It may be inappropriate to the diagnostic questions researched. Its use could also detract from the client-consultant relationship. Caution should be taken when a pre-designed questionnaire is selected. Face validity is high for any type of questionnaire. Statistical validity will be high for standardized fixed response types and for properly developed custom fixed response questionnaires (Lawler et al., 1980). Reliability will also be high in those cases. The custom free response type has a unique bias: respondent scoring. This may cause validity to be somewhat lowered. Questionnaires can provide data for levels ranging from the individual through the organizational. The level of analysis will depend upon sampling. Member participation and cooperation is required from those sampled. Once designed, questionnaires lack flexibility. Fixed response types provide quantitative results; free response types can provide qualitative results. Questionnaires provide the reactive results typical of self-report measures. Questionnaires are best used when large amounts of data are required quickly, and in the least costly manner. Harrison (1987) also suggests the use of questionnaires for repeated measures, such as those required to assess organizational change. In addition, questionnaires can be useful if comparing data with norms is required. Care must be taken to ensure that language used in the questionnaire can be understood by the respondent.

Interviews Individual interviews are frequently used to gather information on how an organization functions. An interviewer poses a series of questions to an organizational member and records his or her responses. Lawler et al. (1980) list three types of commonly used interviews: unstructured, structured open-ended and structured closed-ended. The unstructured interview provides the respondent with relatively little guidance; the respondent may speak freely about any topic. This interview type is most useful when the consultant initiall y enters an organization. A structured open-ended interview fea-

Organizational Needs Assessment

245

tures predetermined questions to guide the respondent, but unrestricted answering. In the structured closed-ended interview the questions and type of answers sought are predetermined. Interview cost is high, especially if large numbers of people must be sampled. Interviewing takes much time. With the unstructured and open-ended types additional time and interviewer skill is required to code and analyze data. All interview types feature high face validity. Statistical validity is lowest for the unstructured type, but can range to high for the structured closed-ended type (Lawler et al., 1980). Similar low to potentially high ranges exist across the interview types for reliability. Various interviewer biases are a potential problem with this technique. Self-report biases can also cause error. Interviews readily provide information on an individual level of analysis. Higher levels of analysis require large sampling sizes. Member participation and willingness to provide data will greatly influence the success of this technique. The unstructured interview method features great flexibility and adaptability. Structured openended interviews are moderately flexible. Structured closed-ended types have limited adaptability. Interviews provide the consultant with qualitative data. Person-to-person verbal communication allows the interviewer the opportunity to build trust with the respondent. A willing respondent can provide much information on emotional reactions to the organization, as well as detailed descriptions of his or her perception of the situation. More structured interviews result in data that are more easily quantified, but less rich. Data provided by the interview method will be reactive. Self-reported information will contain respondent biases.

Focus Groups Organizational members may provide organizational data while participating i n focus groups. Groups of 7 to 12 members are assembled to discuss predetermined topics. Discussion is facilitated by a moderator. The moderator does not, however, assume the role of a leader, or simply question the group for specific responses (Calder, 1980). Free discussion of selected topics takes place among group participants. Data are collected while a researcher working with the moderator records ideas and topics discussed. Focus groups are relatively inexpensive (McKillip, 1987). It may take only a few hours to gather and analyze data. If a number of focus groups are conducted across the organization, time requirements increase. The exact type of analysis performed on data will also help to determine amount of time required to complete the process. Some skill is required to conduct focus groups. The

246

Caska, Kelley, and Christensen

moderator must be able to acquire the trust of the group and to build rapport with members if free discussion of topics is to occur. The moderator must also be able to facilitate discussion without dominating it. Some skill is required to record data accurately. The two researchers should be able to work well together. High face validity is associated with this technique. Statistical validity, however, will vary. As pointed out by Krueger (1988), validity will depend on whether or not the topic of discussion is suitable for focus group inquiry. Validity will be influenced by the content of the discussion. In addition, for validity to be present, participants must be qualified to discuss the issues at hand. Reliability will depend on the analysis of data. Stewart (1990) explains that reliability may be calculated in two different ways. Analysis requires coding the recorded content of the discussion, according to explicit rules. Words, phrases, sentences, and even gestures must be categorized. One method of determining reliability requires establishing reliability of this unit coding. Secondly, inter-rater reliability for raters involved in data analysis may be calculated. Krueger (1988) suggests that inter-rater reliability may be more useful, since the real interest is in determining themes of the discussion. A group level of analysis is initially provided with this technique. Higher levels of analysis are possible if a series of focus groups are conducted throughout the organization. Member participation is, again, very important. Focus groups feature flexibility. While the moderator offers some guidance in topics to be covered, direction of discussion and specific issues are largely left up to the group. Qualitative data is produced by the focus group. Some quantification, such as frequency counts, is possible after content analysis of data. This technique may provide reactive data; self-report biases are possible sources of error. In addition, public pronouncements are required before the group members, potentially resulting in intentional biasing of data. To minimize reactivity, the moderator should encourage frank and open discussions, in a comfortable, trusting atmosphere.

Force Field Analysis Force field analysis is an innovative data gathering technique, involving the use of mapping (Lauffer, 1982). This method focuses on defining forces that influence organizational functioning and determining how theses forces can be manipulated or directed. The use of this technique requires initially defining the problem. Previous use of other needs assessment technology, such as interviews,

Organizational Needs Assessment

247

focus group discussions, or questionnaires can provide such information. Goals or objectives that are being effected by those forces will also be determined. Typically, small groups will be assembled to participate in force field analysis. Restraining and driving forces affecting objectives are analyzed at this point. The potency, consistency and amenability to change of each force is estimated by group members, then mapped. The predictability of forces may also be considered. It is then possible to estimate the likelihood of success in attaining goals through removing or changing forces blocking specific objectives. The consultant can guide members in setting up specific methods to achieve goals. Limited information exists on the characteristics of force field analysis considered relevant in technique choice. It seems as though this method would be fairly low in cost. Extensive analysis of data is not required by the consultant, and the technique is fairly simple to use. Typically, only a few hours will be required to complete this technique. Force field analysis should have high face validity. It is suspected that content validity and reliability will be fairly low, because small groups are generating data. Data can be analyzed at the group level, and higher levels of analysis can be achieved by conducting a series of force field analyses throughout the organization. Member cooperation and participation are, again, necessary. This technique is moderately flexible. Though group discussions are typically very adaptable, this technique begins with prior definition of problems to be solved, goals and objectives; some flexibility is therefore lost. Qualitative information is also typically provided during small group discussion. However, mapping creates a quantitative representation of data in this case. Force field analysis is a reactive technique. Tt is most effectively used in devising methods for organizational change. Ideally, participating members will have the power to implement suggestions to achieve objectives.

EXAMPLES OF ORGANIZATIONAL NEEDS ASSESSMENTS Four examples of actual organizational needs assessments will now be presented. Information on the assessments was gathered by Edward Christensen (one of the authors of this chapter). Combinations of techniques used in each assessment will be presented. It will be possible to determine in each case if techniques actually being

248

Caska, Kklley, and Christensen

used exemplify characteristics recommended, or fit the particular assessment situation.

Example I : A Bank System Consultants were called into a bank system undergoing a merger. Initially suspected problems revolved around climate and communication, at the organizational level. Smaller bank branches, previously not under the control of the bank system, were adjusting to control from a higher organization, Feelings of isolation from the system, confusion and powerlessness were suspected i n these branches. Climate and communication were chosen for assessment according to the consultant's experience with similar situations, previous work experience in the bank system, and client specifications. Data gathering techniques chosen included structured, openended interviews and two different questionnaires. At the time of this assessment, general organizational information had to some extent been gathered through previous experience with this bank system. Interviews were conducted with managers throughout the system. At the time of the interview, the two questionnaires were completed. Interviews covered the areas of training adequacy, growth, and development. One of the scales was designed to assess the character of the communication system in the bank system. Responses to 11 questions on the communications process in the organization were made on a scale of one to four, ranging from one extreme to another. The second scale was designed to assess organizational climate. Feelings toward the work unit, division and entire system were marked on a continuum for each of five variables. From this data, the consultant was able to assess organizationwide managerial views on the effectiveness of training programs, expectations about growth and development, communications and organizational climate. Level of analysis ranged from the manager's immediate work unit to the entire organizational system. The combination of techniques used in this assessment provided for high face validity. The level of analysis achieved was appropriate for this situation. Total costs were reasonable; though interviews can be expensive, the structured open-ended type used in combination with inexpensive questionnaires provided for only a fairly costly assessment. The techniques used provided a reliable, valid assessment. Questionnaire items were selected for this situation from more generalized, tested scales. Therefore, they are

Organizational Needs Assessment

249

considered to be reasonably valid and reliable. Member participation was a key requirement in this assessment; but this was not a problem. Past experience with the system provided any indication of problems to be expected with required participation. As recommended, both qualitative and quantitative data were gathered. In addition, at least one technique (the interview) was to some degree flexible. The techniques used in the assessment were both reactive, and subject to self-report biases. It is assumed that prior experience with the organization provided some nonreactive information to be considered in addition to data gathered at this time.

Example 2: A Utility The utility assessed in this next example had recently undergone decentralization. Some loss of jobs had occurred in the process. This, combined with recent changes in the industry, had resulted in concern and bitterness among organizational members fearing job loss. In addition, smaller offices resented the control and influence of central headquarters in their operations. Policies and objectives from headquarters seemed confusing, and changed frequently. Headquarters would frequently call local office people away for meetings without concern about effects this would have on already busy local offices. In addition, new top managers in local offices were hired from outside the community. It was felt that these new managers did not understand the small office operating in a particular area. Background data were gathered during initial experiences with organizational members, prior to the actual assessment. The consultant's previous experience with newly decentralized industries aided in choice of areas for analysis. Focus groups were used to gather data for the needs assessment. A total of 49 managers from all levels and districts of the utility participated. Seven focus groups were conducted. Prior to beginning the discussion, confidentiality of responses was assured, and the necessity of cooperation stressed. It was also explained that findings reported would be summaries of data; specific individual or group sources of data would not be available to clients. Topics covered in focus groups included positive and negative forces on the manager in the work place, training needs, the impact of recent changes, and communication. The discussion began with asking group members to individually answer questions on time spent in communication, frequency of types of communication and persons responsible for initiating communication. Averages and

250

Caska, kidley, and Christensen

ranges were quickly tabulated and discussed; additional topics were covered later in discussion. Results of discussions were presented in quantitative form: data was categorized according to frequency of mention for each response. Results of individual communication questions were reported on group and organizational levels. Groups consisted of members from seven distinct districts and levels, with a summary of group data providing organizational level analysis. The focus group technique allowed the assessment to be completed quickly, at a fairly low cost to the client. Assuring anonymity and confidentiality of responses helped to produce valid results, though response bias and problems with group influences were potential sources of error. This technique was appropriate to the required levels of analysis. Members were willing to participate and cooperate. Flexibility was present. Because the focus group was the single technique used to gather assessment data, its flexibility was important. Quantitative data were requested by the client. This technique had the advantage of providing qualitative data that could be easily quantified. Ideas, contexts and feelings could be discussed, yet averages, ranges and frequencies could be tabulated. Data collected during this assessment were potentially reactive. This is a possible problem. Ideally, another less reactive technique would also have been used, to verify data for an organizational needs assessment. This was not the case in the study. However, if only one technique could be used due to client-imposed limitations (e.g., cost, time), the focus group method seems most appropriate for this situation.

Example 3: A Chamber of Commerce This organizational needs assessment featured an in-depth analysis of severe problems facing the organization. Many different techniques were used to provide thorough, valid data. Much of the data collection took place over a weekend retreat. All members of the organization were able to participate in data collection, because the organization was fairly small. Some unique data gathering methods were used during the assessment, along with the more typical group discussions, interviews and questionnaires. Prior to the weekend retreat some data were gathered to determine the nature of problems facing the organization and expectations concerning the assessment. Interviews were conducted with some staff managers, community leaders and politically influential individuals. They were conducted to provide the consultant with

Organizational Needs Assessment

25 1

broad impressions of problems facing the chamber of commerce. In addition, prior to the retreat, questionnaires were sent to key organizational members. Members were questioned on their perceptions of the purpose of the organization, rated on a scale from one (of least importance) to ten (of greatest importance). Questions about expectations for the retreat were also answered. Much of the data were collected in small groups, consisting of members representing varied interests. Techniques used included focused group discussions and "brainstorming" within small groups, force field analysis and tinkertoy representation. The tinkertoy representation involved small groups constructing models of the organization. Each group was asked to show its model and explain it to the full group, Data were gathered with this technique by noting degree of agreement or disagreement of members on the structure created, and which members represented specific interests during construction. The actual created structure provided additional information. It could, for example, be balanced or unbalanced. The force field analysis was used to determine type and intensity of forces driving toward or hindering organizational change, for each specific previously defined objective. Results of small group data collection were presented to the entire group, for discussion and comment, after each small group technique was completed. This organizational assessment required much time from members and was fairly costly. That did not present problems for this particular organization. Those conducting the assessment had skill and experience with a wide variety of techniques, allowing for effective use the many methods employed during the study. Because a great number and variety of techniques were used to define the situation of this organization, it is likely that results were both valid and reliable. In addition, face validity was high. Organizational level data were provided by assessing and combining information from all small groups. Much member participation and cooperation was required. However, because members had willingly decided to attend the retreat to participate i n data collection, there was motivation to take part actively in the process. Flexibility was present in the overall assessment, due to the use of initial interviews and small group techniques. Both qualitative and quantitative data were available for analysis. Techniques used to gather information were, for the most part, reactive. However, because a great number and variety of methods were combined to provide results, errors from individual biases should have been reduced to a great extent.

252

Caska, Kelley, and Christensen

Though a concentrated block of time was required to gather and analyze data needed for this assessment, recommendations for changes were available by the end of the retreat. Since the problems facing the organization were classified as a "crisis," solutions were needed as quickly as possible. The retreat made this possible.

Example 4: An Erecutive Board This needs assessment was conducted to aid in organizational planning. Participants had initially attended a larger, general workshop on planning, conducted by the consultants. These organizational members had requested a workshop specifically for planning in their own organization. The assessment was conducted for that particular purpose. Data were gathered during a focus group discussion. Ten members participated. To begin with, participants individually listed those achievements they felt represented activities of the board. Responses and frequency of responses were quickly tabulated at this point. Next, members were asked to list individually objectives they wanted the board to achieve in the future. Again, a list (referred to as a "wish list") was compiled, with frequencies of mention tabulated. Resources for and inhibitors to achieving objectives were next discussed. Records of discussions were kept, and frequencies again tabulated. From analysis of data gathered at this point, suggestions for future directions could be made by consultants. Areas of focus for improvement included operational functions membership, professionalism and organizational development intervention. As in the case of the utility example, the focus group was the only technique used to gather data during the assessment. Once agam, if only one technique could be used to gather information, the focus group was a good choice. It provided for an inexpensive, quickly completed analysis. Face validity was present, as well as flexibility. Both qualitative and quantitative data could be presented to the clients, as analysis results. Member participation was a key element in the assessment. This was, however, not a problem. These members had originally requested the assessment, and were highIy committed and enthusiastic in providing data. Validity and reliability are potential areas of concern in this assessment. The use of additional data collection methods could ensure the presence of these characteristics. In addition, the data were completely reactive. Bias is a potential problem. Public pronouncement before a group could also lead to biased data. The

Organizational Needs Assessment

253

force field analysis technique could have been useful in this assessment, in addition to the focus group.

CONCLUSIONS These examples are cases of organizational assessments that were considered to be successful in improving organizational effectiveness. Generally, techniques chosen featured those qualities labeled as typically desirable. Dimensions of those qualities considered to be situation dependent were situationally appropriate. Combinations of techniques generally featured one technique that was adaptable or flexible and at least one providing qualitative data, as recommended. There seemed to be very limited use of nonreactive techniques. This means that all data collected were subject to bias error, and consequently lowered validity. There are a number of possible reasons for the reliance on reactive data. Perhaps in some cases nonreactive techniques, such as archival research and observation, would have been unacceptable to the client, or were simply inappropriate to the organizational situation. Perhaps the consultants conducting these assessments were able to determine that the data gathered with the reactive measures were reasonably valid, through their knowledge of similar past experiences and expertise. Or, perhaps the importance and value of nonreactive data is simply not stressed in the actual consulting world. Because reactive measures are more commonly used than nonreactive measures in organizational needs assessments, there is a great possibility of response bias or error in data. Some solutions to the problem of reactive effects may be suggested. When analyzing reactive data from techniques such as questionnaires or focus groups, that have been collected from various sections of the organization, it will be useful to examine differences in overall scores between different groups of members. These differences can be compared to an overall mean. General trend differences between levels of the organization or specific work units indicate that members see the variables of interest in different ways. The consultant can then attempt to find out why such differences exist, and whether they are problematic. General trends can be reasonably assessed even if data are reactive. Reactive measures can have psychological costs for the organization being assessed. This refers to the after-effects of questioning members about their attitudes or views on potential organizational problems. After providing such data, members will expect results

254

Caska, Kelley, and Christensen

and solutions. If action is not taken to remedy problems, members will be annoyed or dissatisfied (to say the least). A potential solution to this problem is to ask members for their suggestions on how to improve the situation, or how to solve the problems. These suggestions can be presented to the clients. Simple, prioritized suggestions will then at least be considered by the higher organizational members. Those providing data will appreciate the opportunity to suggest changes. The opportunity to plan solutions can reduce the psychological costs incurred from responses to reactive data. Another potential problem is the use of only one technique in the examples of the utility and the executive board. Perhaps, again, situational limitations made the use of only one technique possible. In such cases, the focus group technique was the best choice for data gathering, featuring most of those qualities considered necessary in technique use. In conclusion, an accurate needs assessment is necessary if a consultant is to diagnose correctly an organization's characteristics. Part of that accuracy will depend upon the correct use of adequate, appropriate techniques to gather organizational information. This paper has provided an evaluation of techniques according to factors considered most relevant to the success of the assessment. It should be noted that this evaluation is applicable to needs assessments conducted by internal or external consultants at the request of clients within the functioning organization, as part of diagnosis. As pointed out by Harrison (1987), this type of assessment differs from those conducted for the purposes of planning, monitoring and assessing costs or impacts of social programs, in the fields of health, education, and welfare. The organizational needs assessment examines operations, processes and structures, rather than concentrating on whether or not projects are carried out according to plans. The organizational needs assessment does not focus on the impact of projects on some target population, as other needs assessments do. Different types of techniques are used in these different types of assessments, and results are provided to different clients. The organizational needs assessment also differs from training analysis. Training analysis is conducted with more narrowed focus, for one specific purpose. Unique techniques, such as task analysis, may be required in such analyses. For these reasons, the evaluation of techniques presented in this paper and recommendations for technique use may not apply to other needs assessment situations.

Organizational Needs Assessment

255

REFERENCES Block, P. (1981). Flawless consulting: A guide to getting your expertise used. San Diego, CA: University Associates, Inc. Butler, L. M., & Butler, R. 0. (1987). Guidelines. In D. E. Johnson, L. R. Meiller, L. C. Miller, & G. F. Summers (Eds.), Needs assessment: Theory and methods (pp.271-284). Iowa: Iowa State University Press. Calder, B. J. (1980). Focus group interviews and qualitative research in organizations. In E. E. Lawler, 111, D. A. Nadler, & C. Cammann (Eds.), Organizational assessment:

Perspectives on the measurement of organizational behavior and the quality of work life (pp. 399-417). New York: John

Wiley & Sons. Hanlon, M. E. (1980). Observational methods in organizational assessment. In E. E. Lawler, 111, D. A. Nadler, & C. Cammann (Eds.), Organizational assessment: Perspectives on the measurement of organizational behavior and the quality of work life (pp. 349-371). New York: John Wiley & Sons. Harrison, M. I. (1987). Diagnosing organizations: Methods, models, andprocesses. Newbury Park,CA: Sage. Helmer, 0. (1983). Looking forward: A guide tofutures research. Beverly Hills, CA: Sage. Kaplan, A. (1964). The conduct of inquiry: Methodology f o r behavioral science. San Francisco: Chandler. Krueger, R. A. (1988) Focus groups: Apractitioner’s guide for applied research. Newbury Park,CA: Sage. Lauffer, A . (1982). Assessment tools f o r practitioners, managers and trainers. Newbury Park, CA: Sage. Lawler, E. E., Nadler, D. A., & Cammann, C. (Eds.) (1980).

Organizational assessment: Perspectives on the measurement

of organizational behavior and the ~ u a l ofi ~work lge. New York: John Wiley & Sons. Marvis, P. H. (1980). Assessing physical evidence, documents, and records i n organizations. In E. E. Lawler, 111, D.A. Nadler, & C. Cammann (Eds.), Organizational assessment:

Perspectives on the measurement of organizational behavior and the quality of work life (pp. 418-443). New York: John Wiley

& Sons. McKillip, J. (1987). Needs analysis: Toolsfor the human services and education. Newbury Park, CA: Sage.

256

Caska, Kelley, and Christensen

Nadler, D. A. (1980). Role of models in organizational assessment. In E. E. Lawler, 111, D. A. Nadler, & C. Cammann (Eds.), Organizational assessment: Perspectives on the measurement of organizational behavior and the quality of work life (pp. 119-131). New York: John Wiley & Sons. Rossi, P. H. & Freeman, H. E. (1985). Evaluation: A systematic approach. Beverly Hills, CA: Sage. Stewart, L. W. & Shamdasani, P. M. (1990). Focus groups: Theory andpractice. Newbury Park, CA: Sage. Van d e Ven , A. H. & Ferry, D. L. (1980). Measuring and assessing organizations. New York: John Wiley & Sons. Webb, E. J., Campbell, D. T., Schwartz, R. D., & Sechrest, L. (1966). Unobtrusive measures: Nonreactive research in the social sciences. Chicago: Rand McNally & Co.

Part 111: ISSUES OF CONCERN TO THE INDIVIDUAL IN ORGANIZATIONS

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial1 Organizational Psychology - K . Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. AlL rights reserved.

259

CHAPTER 9 HEALTH IN THE WORKPLACE

Donna Musialowski Ashcraft In the past 15 years there has been an increasing interest in the impact of health on cost containment in the workplace. Illness and accidents cost industry billions of dollars each year through absenteeism, reduced productivity, health care, and insurance costs. Four hundred million work days are lost each year in the United States because of absenteeism, an average of 5.1 days of absenteeism per employee (Rhodes & Steers, 1990). It has been estimated that each day of absenteeism, including salary, lost productivity and fringe benefits, costs the employer $66 (Hinrichs, 1980; Rhodes & Steers, 1990). Total estimates on how much absenteeism alone costs industry vary, with a conservative estimate at 8.5 billion dollars (Rhodes & Steers, 1990). Since this estimate was based on minimum wage, a more likely estimate is that absenteeism costs American business 26.4 billion dollars each year (Rhodes & Steers, 1990). Other reviews of the literature suggest yearly losses as high as 30 40 billion dollars (Ballagh, Maxwell & Perea, 1987). Although there are other causes of absenteeism, the primary contributing factors are illness and injury. In addition to these costs, absenteeism, illness, and injury are taxing to businesses because of insurance premiums, disability benefits and additional employee training due to excessive job turnover caused by premature death or sickness. Businesses in the United States paid 27.4 billion dollars for worker's compensation in 1987 alone (Nelson, 1990). This figure is 2.7 billion dollars more than the previous year and three times as large as the figure of ten years ago. Payments for medical and hospital expenses were 9.9 billion dollars which was approximately 36% of the 1987 total. General Motors spends more money on employee health insurance than on steel. In 1980 Chrysler spent $220 dollars for employee health benefits for each new car it sold that year. Given the high price of health (or the lack thereof) it is not surprising that businesses and industries are interested i n containing these overhead costs by

260

Ashcraft

promoting health. In fact, very large companies can expect to mirror the rate of illness in the general population. Thus, for example, businesses can expect one out of four employees to develop cancer; one out of two to develop heart disease (Brennan, 1982). This chapter reviews some of the factors that affect health in the workplace and suggests that health promoting programs could prevent some illnesses and ultimately lead to a decrease in the price of injuries and illness, not to mention human suffering. The chapter is divided into five sections. (1) The first section discusses stress. It is well known that stress and illness are correlated (e.g., Fava, Perini, Santonastaso & Fornasa, 1989; Magni et al., 1989; Tudiver, 1989). Thus it is important to review the sources of stress in the workplace and ways industry tries to control these stressors. It is suggested that decreasing stress in the workplace can cut overhead costs by reducing absenteeism and subsequent health care. (2) The second section deals with the work environment. There is an increasing amount of literature which suggests that chemicals that people work with and produce can have damaging effects on a person's health, and thus preventive techniques can cut down on industrial costs by decreasing accidents and by decreasing the amount of diseases, primarily cancers, contracted.

(3) The third section deals with the impact of changes in technology on the worker's health and subsequently employer's costs. Such technological changes as the increased use of video display terminals are discussed.

(4)The fourth section explores individual behavior of workers and how such factors as alcohol and drug abuse, smoking, eating and exercise habits affect overhead costs. By promoting good health in employees via health promotion programs and employee assistance programs, companies can save money by decreasing the amount of absenteeism and loss of productivity due to diseases such as cardiovascular disease and alcoholism. (5) The final section discusses the aging workforce and what types of effects can be expected from it.

26 1

Health

STRESS

The Nature of Stress Stress can be defined as physiological, psychological, and behavioral reactions which individuals experience in situations that threaten to overwhelm their ability to cope (Baron, 1983). Stress, however, has been conceptualized in many ways (Cox, 1978). Some people conceptualize stress as a person's response to negative stimuli (e.g., Selye, cited i n Cox, 1978); others describe it as being the shmulus itself (e.g., Symonds, cited in Cox, 1978); and still others describe stress in terms of a person's ability to deal with stimuli, whether positive or negative (e.g., Cox & Mackay, cited in Cox, 1978). Our definition utilizes this third conceptualization. In any case, there are many responses to stress including physiological (e.g. , illness), cognitive, emotional, and behavioral reactions (Taylor, 1986). Similarly, sources of stress are varied, and even among the sources there are variations as to the amount of stress associated with them. These sources of stress will be discussed next. First, to illustrate the variation in the amount of stress associated with different stimuli, a summary of the types of work that are most stressful will be reviewed. Then, various sources of stress, at work, and outside of work, will be discussed. It should be noted, however, that not all people react to high levels of stress in the same way. Indeed, some people seem to be immune to the effects of stress. These people have been termed as having a "hardy" personality (Kobasa, 1982). Stress-related illness can also be eliminated or reduced by various coping strategies and by having a large social support network (e.g., Cohen & Wills, 1985; Thylor, 1986).

Tjpes of Mbrk f i a t Are Most Stress-1 Some types of work seem to be more stressful than others. In a 1988 Newsweek article (Miller et al., 1988) that received widespread attention, the 10 most stressful jobs according to t h e American Institute of Stress were listed. They include inner city high school teacher, police officer, miner, air traffic controller, medical intern, stock broker, journalist, customer complaint clerk, department store worker, waitress, and secretary. The key components involved in the high amount of stress in these jobs seem to be responsibility and lack of control. The majority of the occupations that made up the 10 most stressful list carry large amounts of responsibility but little power, which is consistent with the health psychology literature.

262

Ashcraft

Health psychology literature suggests that uncontrollable events are perceived as more stressful than controllable ones. Feelings that they can predict, modify, or terminate an aversive event or feel they have access to someone who can influence i t will lead to experiencing it as less stressful, even without action. Feelings of control reduce the subjective experience of stress and influence the individual's biochemical reactivity to stress (Taylor, 1986). As will be discussed below, workers in the new "electronic workplace" are also at risk for high levels of stress, especially video display terminal operators (Brod, 1984).

Sources of Stress Sources of stress in the workplace are many and varied. Stress has been correlated with working more than 75 hours a week; job dissatisfaction; boring or repetitive work; or work that is either too easy or too difficult relative to an employee's level of skill (Cox, 1978). Most of these factors can be considered under the realm of quality of work life. The poorer the quality, the higher the level of stress on the job and this, subsequently, can lead to lower levels of company loyalty and higher levels of employee turnover. Examples of this correlation between loyalty and these expenditures include the fact that the turnover rate at AT&T, since the break up of Bell, has more than tripled and the fact that Dupont's early-retirement plan has attracted twice as many takers as was originally anticipated (Werther, 1988). Furthermore, additional overhead costs are incurred by places of employment with high levels of stress because that stress causes illness, and illness increases absenteeism, disability, and insurance rates. (This is more fully discussed below.) In fact, Cox (1978) suggests that 37 million work days are lost each year in Britain due to physical symptoms associated with stress, such as headaches. More serious health effects can also result, however, including ulcers and cardiovascular problems. Other factors that may cause stress in the workplace include a poor work environment. Poor lighting, high levels of noise, uncomfortable temperatures, and unbearably close proximity to others all create higher levels of stress. Other sources of stress that impact on the workplace need not come from the work environment itself. Personal problems, including family problems, deaths, births, etc., can all carry over into the work place and affect performance (e.g., Crouter, 1984). Life change, both positive and negative, also produces stress. Holmes and Rahe (1967), for example, have suggested that such positive factors as marriage, vacation, and Christmas, as well as

Health

263

such negative factors as trouble with the boss, death of a spouse, and divorce, can all produce stress which may contribute to stress-related illness. While employers can not control stress outside of the workplace, they can help by offering maternity and paternity leave; daycare; flexible schedules; leave for death of a close family member; and by helping a newly relocated employee's spouse find a job in hidher field.

Consequences of Stress As mentioned previously, there seems to be a strong link between stress and illness. Over half of all visits to general practitioners are related to stress (Burdick, 1983). As the amount of life stress increases, the likelihood of experiencing illness increases. Among other diseases research shows that as life stress increased, incidence of physical symptoms, onset of myocardial infarction, insomnia, skin conditions such as psoriasis, and dysfunctional uterine bleeding increased (Chernovsky, 1989; Fava et al.? 1989; Magni et al., 1989; Matthews, Scheier, Brunson & Carducci, 1989; Tudiver, 1989). These physical diseases are affected by the amount of stress a person experiences because the stress seems to affect the immune system. More specifically, uncontrollable stressors affect the immune system negatively. In a review article by Maier and Laudenslager (1985), it was stated that rats exposed to uncontrollable stressors had slower T cell production than rats exposed to no stressors. Since T cells are the cells that identify foreign bodies, like viruses, in the body and destroy them, it follows that this suppression of the T cells, caused by stress, is at least one of the reasons why stress and illness are correlated. Stress may not only negatively affect the immune system but interfere with health by decreasing health seeking behaviors. Bausell and Damrosch (1989) conducted a study which suggested that stress may interfere with health promotion behaviors such as exercise and eating a healthy diet. Physical diseases are not the only type to be affected by stress. Mental illnesses are also influenced by this factor. Rahe (1989) for example, found that as life stress increased, depression rates increased. R. Day (1989) found a positive association between stress and the onset of schizophrenia. Strober (1989) found that as stressful life events increased "instigation, maintenance and reinstatement of bulimic eating increased in anorexic patients" (p.245). Just as physical illness due to stress is a concern of employers, so too can mental illness result from stress. As Ballagh et al. (1987) suggest,

264

Ashcrajl

mental health impairments increase concerns about safety, attendance and efficiency. Because stress is correlated with illness, it also indirectly leads to higher overhead costs for employers because of absenteeism and lost productivity due to both physical and mental illness, as mentioned previously. Indeed, a study conducted by Educational Research Inc. (1980) found a positive correlation between stress and absenteeism. Thus, stress reduction programs are of interest to business and industry.

Stress Management Recent research suggests that stress alone may not be the only reason for increases in the rates of illness. How a person copes with stress may also be very important. Thus, two people may be under the same amount of stress and and one will get sick while the second does not, primarily because of their coping strategies (e.g., Taylor, 1986). Similarly, it should be noted that the effect of coping can be negative, as when someone turns to drugs or alcohol in order to relieve stress. It is thus important that employers keep stress in the workplace at a cost-effective minimum and provide stress management programs for their employees. Such considerations will ultimately lead to happier and healthier employees. This job satisfaction may then progress into loyalty and the benefits thereof; and the health of the employees will benefit not only themselves, but also the employer in reduced absenteeism, worker's compensation, and health insurance costs. One way to reduce stress in the workplace is to increase employee satisfaction, to create a positive quality of worklife. Various methods have been successful in improving quality of worklife, thereby decreasing costs, absenteeism and turnover. The autonomous work group is one such method, in which a group of workers acts as a unit to complete an entire job. By giving employees more responsiblity and a variety of tasks, the autonomous work group can increase worker satisfaction (Cummings & Molloy, 1977). Another method is job restructuring, where jobs are redesigned to benefit the employee. An example of this is job enrichment which allows a worker to complete a job by engaging i n a sequence of tasks, rather than only one or two. Cummings and Molloy (1977), as well as other researchers, have found that job enrichment increases worker satisfaction and decreases turnover, thus supporting the thesis that decreasing stress by increasing worker satisfaction can save employers money. Likewise, participative management can reduce stress and stress-

Health

265

related illness by increasing the control workers have on decisions which will affect their work lives. Employees become involved in the decision process and take more of an interest in the task, This strategy has been found to increase production and positive attitudes about the workplace, and to decrease absenteeism (Bragg & Andrews, cited in Cummings & Molloy, 1977; Chaney & Teel, 1972). Also, as noted previously, considerations like flexible working hours, day care, and maternity and paternity leave can decrease the stresses involved in employees' personal lives, thereby increasing production and decreasing turnover both due to reductions in dissatisfaction and stress-related illness. Flexible working hours, for example, have been positively correlated with workers' morale and productivity (Wade, cited in Cummings & Molloy, 1977). One additional way to decrease stress and stress-related illness includes employers offering stress management programs. Stress management programs are educational programs designed to teach coping strategies, relaxation techniques, and skills, cognitive and otherwise, which can be used to deal with stressful situations. Such programs have been successful in reducing stress-related illnesses. Shapiro, Schwartz, Ferguson, Redmond, & Weiss (1977), for example, found that a stress management program helped individuals control chronically high blood pressure.

WORK ENVIRONMENT

Toxics in the Wrkplace Environmental factors can have negative effects on health. Epstein (1978) has suggested, for example, that 70% - 90% of all cancers are caused by environmental factors. Large numbers of these carcinogens are chemicals that employees work with and/or produce in industry. Being exposed to high levels of these cancerproducing agents places these workers at risk for these diseases. In fact, Cullen, Cherniack, and Rosenstock (1990a, 1990b) suggest that 125,000 deaths occur annually due to occupational disease. Actual incidence rates of occupational diseases are much higher. Approximately 500,000 workers are victims of occupational diseases each year (Cullen et al., 1990a). One of the most common types of occupational diseases is asbestosis which results from prolonged exposure to asbestos. Rom and Upton (1990) suggest that two-thirds to three-quarters of workers exposed to asbestos show evidence of parenchymal fibrosis 20 years later. It is estimated that ten percent of asbestos insulators have died from malignant mesothelioma.

266

Ashcraft

Other common types of occupational diseases are noise-induced hearing loss; various lung diseases such as bronchitis and asthma; and carpal tunnel syndrome -- a musculoskeletal disorder of the hand which can result from prolonged use of vibrating hand tools. Still other types of disorders can result from prolonged exposure to chemicals that some employees frequently work with. Neurological disorder can result from working with lead and arsenic; working with methylenedianiline (an epoxy hardener) and dimethylformamide (a solvent) can lead to liver or gastrointestinal disorders. Lead, carbon disulfide, as well as some pesticides and some metals, decrease sperm count and are linked to other endocrine and reproductive disorders. In fact, research has found that over 300 compounds have carcinogenic influences upon lab animals (Cullen et al., 1990b). Besides the human suffering that these studies imply, other negative repercussions are created. Absenteeism, the need for worker's compensation, 2nd employee turnover will increase; productivity and morale will decrease because of these occupational hazards. All of these effects will increase employer overhead costs. One of the most documented examples of monetary costs caused by an occupational disease is black lung, a respiratory disease contracted by coal miners. In 1987, 1.5 billion dollars was paid to 412,000 benficiaries via the Black Lung Program. This figure was 2.3% lower than the previous year, primarily due to deaths attributable to black lung. In most cases employers can only expect an increase in the amount of money paid out in compensation claims. Employers spent more than 38 billion dollars i n 1987 to provide worker's compensation for victims of this disease (Nelson, 1990).

Prevention Given the costs, both monetary and human, many manufacturers and industries are taking preventive measures against occupational illnesses. As Robins and Klitzman (1988) suggest, the majority of occupational illnesses are due to chemical exposures in synthetic conditions. Therefore, they are preventable by changing those same conditions. Industries and the government have improved the safety of machinery and equipment and this has successfully decreased fatalities and injuries in the workplace (Vojtecky, 1988). However, the employees must also be educated about the dangers within their workplace and how to use equipment safely. Robins and Klitzman (1988) found that 60% of employees changed their work practices after such an educational program; and two years later, 42% of the

Health

267

employees continued these safer practices. It should be noted though, that when implementing such a program it is important to consider cultural factors, plant characteristics (e.g., size), and other possible influences on the success of the program. In sum, many industrial employees work under potentially dangerous conditions. These conditions, however, can be altered to be made safer and employees can be educated to practice safer behaviors in the workplace. These preventive programs can decrease overhead costs by cutting absenteeism, worker's compensation payments, and employee turnover, among other expenditures.

THE IMPACT OF CHANGES IN TECHNOLOGY

Den& and Innovations in Technology Affecting the W r k Environment In addition to the toxics that employees work with, the equipment they work with can negatively affect their health. The latest trends in technology, of course, revolve around equipment powered by electricity, and which may generate harmful magnetic fields. With the onset of the computer revolution, for example, increasing numbers of employers are utilizing computers to bolster productivity. Statistics that illustrate this trend, for example, include the fact that the 724,000 personal computers sold by 24 companies in 1986 rose to over a million sold by one company, Apple, in 1983 (Brod, 1984). The use of video display terminals (VDT's) increased with the computerization. Such equipment has multiple uses; offices use them to check on customer data, as well as to engage in word processing. Warehouses and stores use them to keep track of inventory and, i n some cases, cash customers out with them. They are also used in research projects and data analysis, and to FAX information to other locations. Other technological changes include computerized robotics to increase productivity in industry and reduce the number of people needed to run a factory. As factories become more automated, the worker's job will change from manual tasks to those involving the programming, overseeing, and maintenance of the sophisticated machine systems (Riggio, 1990). Changes in communication will also occur that will allow larger numbers of employees to work at home and communicate with the office via computer communication networks. The work environment will also change i n that i t will become more comfortable. Advanced environmental monitoring and control

268

Ashcrafl

equipment will create a physical environment that is comfortable in terms of temperature, humidity, lighting, and pollution (Riggio, 1990). As noted previously, reducing stressors such as uncomfortable temperatures and humidity can increase productivity and decrease stress-related illness. More minor changes include technology for employee comfort such as automatic drip coffee makers and flourescent light bulbs. These changes in technology may affect employees' health because of the electric and magnetic fields generated, as will be discussed later. Working materials are different today as well. Houses are often built, not with plywood, but with particle board soaked in formaldehyde. Automobiles are increasingly made from plastics, as opposed to metals. These synthetic materials have toxic effects just as the toxics discussed previously do. Manufacturing them and working with them may place employees at risk for occupational diseases.

Health Considerations and Eflects of Technological Changes in the Wbrkplace This new technology has been suggested to create additional stress which Brod (1984) labels as technostress. He suggests that the computerization of the workplace creates tension revolving around the possibility of a worker losing hidher job because the computer is more cost efficient than the employee. Likewise, becoming computer literate may be a harrowing experience. Who, reading this chapter, has not had a nerve-wracking experience in which the computer program would not run? There are feelings in these situations that the computer is almost human and is being deliberately mean, to which we want to respond with violence! Brod (1984) also suggests that computers are negatively affecting our relationships by making us less sensitive to others, thereby creating even more tension. They can also create stress by enabling workers to be more and more productive which may set new standards for the employee to meet. In any case, these new technological advances can create additional stress in the workplace. In fact, Brod notes a 1981 NIOSH study which indicated that computer use among clerical workers was associated with higher stress levels than any other occupational group. And, as noted previously, stress can produce illness. Video Screens. These new technological advances can also create illness simply by using them. Recent research indicates that VDTs may be harmful to the user's health. (For a review of this debate see

Health

269

Brodeur, 1989.) Reported effects of VDT's include eye problems such as pain, burning, itching, blurring, and temporary near-sightedness; musculoskeletal symptoms such as soreness in the back and shoulders; headaches; and mood disturbances such as irritability. These symptoms are most likely due to problems with the work stations such as overcrowding and poor placement of the keyboard and/or monitor (Brodeur, 1989; Cullen et al., 199Ob). However, there has also been some research which suggests that VDT's may cause more serious health problems and that these problems may be due to the electromagnetic fields they produce. More severe health effects such as rashes, cataracts, spontaneous abortions, and birth defects have implicated VDT's as a cause but the research is inconclusive. Opthamalogists, for example, have been concerned that electromagnetic radiation from the video screens hastens the development of cataracts but measurements of X-ray, ultraviolet and infrared radiation they produce fall below health standards. However, it has been suggested that some of these health problems may be due to an altering of individual cells in the body by the magnetic field produced by the VDT (Stevens, 1989). For example, research suggests that electromagnetic fields can interfere with the functioning of DNA and R N A , which control the development of the cell, thus creating birth disorders, and possibly cancers.

Other Technologies. The results of these studies can also apply to other types of equipment that people work with such as electric typewriters. Toasters, televisions, shavers, hair dryers, electric blankets, waterbed heaters, flourescent lights, as well as power lines, all set up magnetic fields and thus could possibly be harmful to their users and those who come in contact with them. In fact, some research indicates that children who live or attend school near electrical distribution lines are more likely to develop cancer than children who do not live near such distribution lines (Stevens, 1989). Not all research supports this thesis, however. Likewise, consideration should be given that many of the new technologies developed, are made in such a way that it is convenient to use for one gender but not the other. Computer keyboards, for example, may be comfortable for females to use, but males may have a more difficult time using them because of larger hands. Similarly, some of the protective gear developed to guard against occupational disease are designed for men. Protective masks and shields which guard against mists and dusts are often too big for women, leaving gaps between them and the face allowing dangerous substances to be breathed in. Some dusts are toxic, others merely

270

Ashcrafi

uncomfortable to breathe. Gloves also may be too large and thus clumsy to work with. Similarly, as more women are entering traditionally masculine domains, the machines and tools that men have worked with may have grips and knobs that are too large for women to work with because of their different size and body build (Mackay & Bishop, 1984). This influx of women into masculine fields and of men into feminine fields, and other technological changes mentioned, may be harmful to workers, both through the use of inadequate protective gear and by inducing stress.

Implications of Technological Changes for the Health of Wrkers The implications of these technological changes include the possibility are that workers' health is endangered and because of this, business and industry will need to change equipment and methods of working. The health risks surrounding working with equipment that produces magnetic fields is probably not as severe as the health risk of smoking or working with such toxics as asbestos, because the field decreases exponentially with distance. (However, it should be noted that the fields associated with a VDT seem to be more dangerous than that caused by, for example, a radio because the pulsing of the image on the VDT produces a pulsing field; Marha, cited in Stevens, 1989). Nevertheless, precautions need to be taken. The government should implement an occupational exposure standard, just as it has done with toxics. No such standard exists. Likewise, businesses should protect their workers by creating safer work stations. This, i n the long run, may save businesses money by decreasing health care costs. Work stations, for example, could be designed so that workers are an adequate distance from the VDT. Computers and VDT's are inadequately shielded on the side and back, only protecting users sitting in front of the terminal. Because of this, workstations should not be placed too close together and especially should not be placed back to back or a user will be exposed to stronger fields from adjoining computers. New VDT's, which shield against magnetic fields, are also being manufactured and used in Scandinavian countries (Brodeur, 1989). These or other similar models being designed, tested, and manufactured, could replace older VDT's. Work station design should also be made more comfortable so that muscle and eye strain can be decreased. Some computer stations, for example, are now being manufactured so that the user could look down on the VDT instead of upward or ahead. Also, as noted previously, stress, in this case technostress, can

Health

271

be reduced by job redefinition and training (Brod, 1984). In sum, the technological advances we have recently experienced has increased our productivity and efficiency. However, these same advances may be affecting the health of workers. In the future, business and industry will have to protect workers by changing work stations and using safer equipment. Not only is this beneficial for employees, but it is also beneficial for employers in that it can reduce health care costs.

THE INDIVIDUAL Alcohol and Drug Abuse Just as other types of illnesses result in absenteeism and lost productivity so does alcoholism and drug abuse. It has been estimated that alcoholism costs American employers $26 billion each year (Ballagh et al., 1987), as does drug abuse (Harwood, Napolitano, Kristiansen & Collins, 1984). Alcoholism and drug abuse can result in an increase in the number of accidents that can occur in the workplace, subsequently increasing insurance premiums, worker' s compensation payments, and health care costs, not to the mention human costs involved. Rothman (1988) states that 40% of industrial fatalities and 47% of industrial accidents are alcohol related. Similarly, alcoholism and drug abuse increases absenteeism and reduces productivity. Quayle (1983) states that an alcoholic employee, for example, has an absenteeism rate 16 times higher than a nonalcoholic employee and that alcoholics or drug addicts work at only two-thirds their potential. Another cost relating to drug abuse is random drug testing. Rothman (1988) reports that 5% of Fortune 500 firms engaged in drug testing programs in 1982, with the percentage increasing to 25% in 1985, and closer to 50% in 1990. Drug testing is expensive, with low cost tests running $500 for 100 persons tested. However, these low cost tests have a 25% false positive rate (Rothman, 1988) and so multiple tests may be needed or more expensive tests may be selected. It should also be noted though that while these tests are being used i n order to protect employees from possible accidents while under the use of drugs and to also maintain productivity, the tests themselves interrupt productivity and may spurn resentment resulting in higher amounts of turnover. Rothman (1988) points out that abuse of alcohol, cigarettes, and food can be as dangerous as drug abuse and thus suggests alternatives to drug testing which will be discussed later.

272

Ashcra3

Smoking Other individual behaviors, besides drug and alcohol abuse, affect both employee health and employer overhead costs. Smoking, another of these behaviors, increases a person's risk of cardiovascular disease. Mahler (1990) suggests that 30% of all cancer deaths are attributable to tobacco and even greater numbers of people die from smoking-related cardiovascular disease. In fact, 11.3% of the total cost of illness is produced by smoking-related illnesses. An extra $204 is added to a smoker's annual insurance premium for health care (Kristein, 1982) and this increases insurance premiums paid by employers. In fact, Kristein (1982) states that a smoking employee costs an employer between $274 and $287 in excess insurance costs. Likewise, smokers are absent 33-45% more than nonsmokers, which amounts to approximately two to three days' excess absenteeism per year. More generally, Brennan (1982) and Kristein (1982) estimate that 81 million workdays are lost each year in American business and industry due to smoking. Productivity is also affected in that employees who smoke take time to smoke at the workplace. It is estimated that an employer loses one minute each working hour for each smoking employee. Inefficiency and errors may also increase due to eye irritation from smoke, lowered attentiveness, and decreased cognitive capacity and functioning (Kristein, 1982). Brennan (1982) estimates that the loss of productivity due to smoking and its effects at 19 billion dollars per year, although later statistics indicate losses of 30 - 43 billion dollars (Mahler, 1990). Additional revenues must, too, be spent on extra maintenance and air filtering. Other costs to employers such as fire insurance premiums are also increased. Likewise, the health of nonsmoking employees is negatively affected by passive smoking (i.e., breathing in the smoke of others), thereby increasing absenteeism and worker's compensation. Mahler (1990) states that 3,800 passive smokers die each year of lung cancer and that the medical costs of smoking add $100 per year to the taxes and insurance premiums of nonsmokers. Furthermore, smoking can have a synergistic effect with toxics in the workplace. Thus, employees who smoke and work with toxics have an even higher chance of developing an occupational disease than employees who do not smoke and who work under the same conditions. This increased risk may be due to a reduction in the latency period of various cancers (Archer, Wagoner, & Lunden, 1973). In fact, exposure to environmental smoke at the worksite is a key occupational health issue, according to the 1986 U.S.Surgeon General's Reports (Gottlieb & Nelson, 1990). One example of the

Health

273

synergistic effect of toxics and smoking can be found by studying uranium miners. Archer et al. (1973) and Kristein (1982) state that uranium miners who smoke have a risk for lung cancer that is six times as great as the risk for nonsmoking uranium miners. Kristein also notes that working with asbestos and smoking increases the mortality rate 11 times over that of non-smoking asbestos workers. Although gold is not a toxic, it can become one when mining it is combined with smoking. African gold miners who smoke have a 50% chance of developing chronic bronchitis as opposed to an 8% chance for nonsmoking gold miners. Therefore, smoking, both alone and in conjunction with working with toxics, creates human costs in terms of the quality and value of the employees' lives and in terms of overhead costs for employers.

Poor Eating and Ejrercise Habits It is well known that eating a nutritious diet and exercising regularly is positively correlated with good health. Unfortunately, an estimated 30% of the U. S. adult population is at moderate or high risk for coronary heart disease, primarily due to high blood cholesterol levels (Glanz & Mullis, 1988). Approximately 20% of American adults are more than 20% overweight, and therefore are also at risk for cardiovascular problems. Exercising and eating nutritious foods can decrease this risk. A proper diet can also decrease one's chances of developing various cancers, since approximately 35% of all cancers are estimated to be linked to nutrition (Glanz & Mullis, 1988). Though less well known, maintaining the good health of employees can save a company millions of dollars in various ways. Premature death of employees costs American business 19 billion dollars each year (Martin, 1978). Likewise, three billion dollars is lost every year in employee turnover due to chronic illness. Xerox Corporation, for example, spends approximately $600,000each year to replace executives who leave the company because of cardiovascular disease (Donoghue, 1977). Cardiovascular disease is just one example of preventable diseases which cost employers money in absenteeism, lost productivity and other expenses. Thus, more and more companies are turning to health promotion programs to reduce overhead costs and decrease human suffering.

Health Promotion Programs Health promotion programs seek to prevent illness (and thus reduce costs) by educating employees about health; encouraging

274

Ashcraft

employees to attend exercise programs which the employer may subsidize or design; and by having employees engage in health appraisal programs. Many of these programs offer smoking control activities. In a 1985 national survey of health promotion activities 36% of organizations offered such activities (Gottlieb & Nelson, 1990). The most common activity was the implementation of a formal smoking policy. However, many places of employment provided information on the health effects of smoking and self-help information. Others conducted special events or classes or workshops for smoking cessation. Such programs are useful in that if a person quits smoking the risk of myocardial infarction reduces and dissipates largely after two to three years (Rosenberg, Palmer, & Shapiro, 1990). Thus, these programs reduce the costs associated with smoking mentioned previously. Smoking cessation promotion in the workplace does seem to be successful in saving management money. Campbell Soup Company, for example, received a 20-50% return on their investment in a smoking cessation program for each year after an employee quit smoking (Fielding, 1982). Another key component is nutrition education programs and exercise programs. The health promotion program at Northern Natural Gas Company, for example, gave employees the opportunity to participate in an aerobics program (Martin, 1978). Those who participated in the program lost significantly fewer days from work due to illness than those who did not participate in the program. Similarly, some studies have found increases in productivity when employees were involved in an exercise program (e.g., Laporte, 1966). Other health education programs are also available. AT&T, for example, provides a program on breast cancer awareness which has increased breast examinations among employees (Parkinson, Denniston, Baugh, Dunn, & Schwartz, 1982). Thus, such programs can decrease human suffering and save employers money in employee turnover, worker's compensation and health care costs such as health insurance. Other employers are opting for Acquired Immune Deficiency Syndrome (AIDS) prevention programs. Backer (1988) suggests that AIDS will cost employers 55 billion dollars in 1991. According to a survey conducted by Fortune magazine, it is among the top three concerns of American employers, especially in high incidence areas, since these employers will face increased insurance premiums. Thus, many companies have developed AIDS prevention programs, including Wells Fargo Bank, Pacific Northwest Bell, and Levi Strauss.

Health

275

While health promotion programs, especially comprehensive ones, can be expensive, they do save companies money in the long run. New York Telephone, for example, spent 2.84 million dollars on 80,000 employees for health promotion in 1980 and realized a two to one return on their investment, including savings in absenteeism and treatment (Plyman & Perkins, 1983). Novelli and Ziska (1982) estimate that New York Telephone saved 27 million dollars in 1981 through its health promotion programs. These programs also provide other benefits such as employee loyalty and an increased pool of applicants. Some skeptics of health promotion programs have suggested that participants in such programs are healthier than nonparticipants, thus accounting for the results. Zavela, Davis, Cottrell, and Smith (1988) examined this question, and found that people who intended to participate in health promotion programs had lifestyles similar to those who did not intend to participate in such programs. It can be concluded, then, that health promotion programs are beneficial to employees. It should also be noted though that there are many factors that can influence the success of these programs (Green, 1988; Haskell & Blair, 1980). Employee assistance programs (EAP's) are an offshoot of the health promotion programs. They are programs designed to treat employees' alcoholism and drug addiction. Unfortunately there are few studies on the effectiveness of EAP'S on cost containment in the workplace and of the programs themselves. The few studies that do exist, however, indicate positive results. Xerox Corporation and Control Data Corporation are two such employers who have implemented EAP'S with positive results (C. C. Day, 1983; Wright, 1982). Likewise, Quayle (1983) believes that an employer can expect a two to twenty dollar return on each dollar spent on these programs. It should be noted, however, that the success of such programs vary depending upon the content of the programs. Some employers have implemented random drug testing in an effort to reduce costs associated with drug abuse. While random drug testing cannot be considered an EAP, some employers may consider it as such. Rothman (1988), therefore, suggests alternatives to drug testing. He suggests that both alcohol and drug abuse be treated equally in that employees may resent being tested for drugs when they know of co-workers who have alcohol problems who are not being tested for their abuse of a different type of drug (i.e., alcohol). Related to this is his suggestion that EAP's be established as an alternative to drug testing and that all employees should be trained in identifying symptoms of drug and alcohol abuse. This latter suggestion would enable

Ashcrajl

276

all employees, in addition to management, to help create a drug-free workplace. It should further be noted that EAP's are reactive solutions to alcohol and drug abuse. Preventive techniques such as education may be more beneficial (Nathan, 1983).

THE AGING WORKFORCE Aging and the Wrl;Cforceof the Future It is well known that the United States population is aging. In 1900 only 4 % of the U. S. population was over the age of 65. By 1985 this percentage had increased to 11.7% and it is predicted that by the year 2000 it will further increase to 13%; by 2010, 14% (Hooyman & Kiyak, 1988; Mahler, 1990). This increase in the elderly population is especially dramatic for people age 85 or older. The number of people in this population increased 165% between 1960 and 1982 and is expected to increase five-fold by the middle of the next century. It is estimated that they will then form 5% of the general population and 24% of the population age 65 and older. It should be noted that although these statistics are generated from the United States population, this aging trend is a world-wide phenomenon. This aging of the population will affect the workforce. Hooyman & Kiyak (1988) note that a man born in 1900 would live approximately 46 years and work for 32 of those years. A man born in 1981, however, can expect to live to be 70 and work for 38 years. Statistics are comparable for women in that they spend less time bearing and raising children today, in comparison to previous years, thus giving them more time to spend in the workforce. Furthermore, 75% of the labor force prefers to continue working part-time after retirement, primarily at positions which traditionally have been filled by younger workers (e.g., fast food places). Reflecting these trends is the estimate that the median age of the workforce is expected to reach 37.3 years in 1995 as compared to 34.8 years in 1982 (Hooyman & Kiyak, 1988).

m e Effects of Aging on W r k Employers are also concerned that this aging of the workforce will have negative effects upon job performance and labor costs.

Likewise, they are concerned about possible shortages of entry level workers with the increased age of the workforce. Related concerns include the obsolescence, burnout, and physiological decline of older

Health

277

workers (Robinson, Coberly, & Paul, 1985). Because of these concerns research has been conducted on age differences in cognition, reasoning, decision-making, memory, spatial abilities, perceptual-motion, cognitive speed, and sensory factors (Salthouse, cited in Robinson et al., 1985). While lab studies do show some decline in these cognitive functions with age, the effects are absent from job performance. However, productivity can decline with age in some occupations and there may be some minor declines in physical capacity, reaction time, and speed of performance. On the other hand, though, tests of creativity, flexibility, and facility of information processing, as well as absenteeism, accident rates, and turnover compare older to younger workers either equally or favorably (Robinson et al., 1985). Likewise, it should be noted that some of the declines in the performance of older workers may be due to cohort, educational attainment, lack of training and environmental factors and so adjustments in the workplace can decrease these age differences in productivity.

Health Considerations and the Effects of Aging in the Wrkplace As the population ages, additional resources will be needed to support the increasing demand for medical expenditures that will occur simply because of age. As people age, they require more medical services. In 1980 dollars, personal medical services cost more than three times as much for those over 65 ($1,606) than those under 65 ($491; Mahler, 1990). This expense will carry over into the workplace because, as noted, as the population ages so too will the workforce. Thus, employers, as well as taxpayers, will pay additional revenues to support the aging population and workforce. One source of this type of expense is in social security disability insurance and supplemental security income. The percentage of people receiving these types of benefits increase with the age of the employee. For example, in 1986, 12.3% of females receiving these benefits were aged 18 - 29; but 25.8% were aged 60 - 64. Likewise, for men the comparable statistics were 13.3% of males aged 18 - 29 who received social security disability and supplemental security income and 26.1 % of males aged 60 - 64 (McCoy & Weems, 1989). This means that twice the percentage of people aged 60 - 64 receive disability insurance than people aged 18 - 29. In fact, social security financing, in general, will increase substantially (Ballantyne, 1989).

278

Ashcrafi

nenh in the Ubrkplace to Accommodate the Aging Population These statistics and effects do not imply that the aging population should be eliminated from the workforce, by for example, having a mandatory retirement age. Indeed, this is a discriminato practice and employers would be depriving their businesses o years of knowledge and experience from present employees if such a thing were suggested. Likewise, the current cohort of older people believes in hard work, company loyalty, and occupational stability, characteristics most employers would find favorable. However, organizations should prepare for this inevitable phenomenon. Medical experts and their supporters in business suggest lobbying Congress to increase spending for research on many illnesses that make life difficult for the elderly, like Alzheimer's disease (Freudenheim, 1990). This research can help defray the high cost of health care by devising better treatments and/or cures for various diseases that afflict the aging population. Not only can such additional research reduce human suffering, but it can reduce the cost of the health care burden for the taxpayers and can reduce such overhead costs for employers as disability and health insurance. Likewise, improved treatments can reduce employee turnover, in that, as the older population gains more productive years, they may wish to remain in the workforce for longer periods of time. Similarly, structural changes in the workplace may need to be made to accommodate the physical deterioration of the aging population. Other changes in the work setting that may occur in the future are changes in social security and pension restrictions against earnings. Because most retired people prefer to work part-time, and because employers are concerned about shortages of workers to fill entry level jobs, eliminating these restrictions would benefit both parties. Likewise, tax credits or other tax incentives could be given to employers who hire older workers. Older workers could return to the workplace i n a part-time basis as consultants. Indeed, some European countries have stages of retirement whereby the employee who is retiring gradually reduces hidher workload to a part-time schedule. Similarly, job retraining, flexible hours and allowing work to be done at home, would be helpful to older workers (Hooyman & Kiyak, 1988). It should be noted that all of these possible trends can help to maintain the health of the elderly population in that their working gives them an opportunity to interact with others and form social support systems that are so important in coping mechanisms that help to defray the impacts of stress on health.

7

Health

279

REFERENCES Archer, V. E., Wagoner, J. K., & Lunden, F. E. (1973). Uranium mining and cigarette smoking effects on man. Journal of Occupational Medicine, 15, 204- 2 11. Backer, T. E. (1988). Managing AIDS at work: Psychology's role. American Psychologist, 43, 983-987. Ballagh, J. H., Maxwell, E. B. & Perea, K. A. (1987). Absenteeism in the workplace. Chicago: Commerce Clearinghouse, Inc. Ballantyne, H. C. (1989). Social security financing in North America. Social Security Bulletin, 52, 2- 13. Baron, R. A. (1983). Behavior in organizations: Understanding and managing the human side of work. Boston: Allyn & Bacon, Inc. Bausell, R. B. & Damrosch, S . P. (1989). Stress as a potential (Ed.) Stressful life barrier to health. In T. W. Miller events. Madison, Connecticutt: International Universities Press, Inc. Brennan, A. J. J. (1982). Health promotion: What's in it for business and industry? Health Education Quarterly, 9, 9-19. Brod, C. (1984). Technostress: The human cost of the computer revolution. Reading, MA. : Addison-Wesley Publishing Company. Brodeur, P. (1989, June). Annals of radiation: The hazards of electromagnetic radiation fields, I11 - Video display terminals. The New Yorker, pp. 39-68. Burdick, G. R. (1983). Wellness: A wholistic approach to wellbeing. Public Management, 65, 13-16. Chaney, F. & Teel, K. (1972). Participative management: A practical experience. Personnel, 49, 8- 19. Chernovsky, Z. Z. (1989). Life stress measures and reported frequency of sleep disorders. In T. W. Miller (Ed.) Stressful life events. Madison, Connecticutt: International Universities Press, Inc. Cohen, S. & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 3 10-357. Cox, T. (1978). Stress. Baltimore, MD.: University Park Press. Crouter, A. C. (1984). Spillover from family to work: The neglected side of work-family interface. Human Relations, 37, 425-442. Cullen, M. R:, Cherniack, M. G., & Rosenstock, L. (1990a). Occupational medicine, Part I. New England Journal of Medicine, 322, 594-601.

280

Ashcrafl

Cullen, M. R., Cherniack, M. G., & Rosenstock, L. (1990b). Occupational medicine, Part 11. New England Journal of Medicine, 322, 675-683. Cummings, T. G. & Molloy, E. S . (1977). Improving productivity and the quality of work life. New York: Praeger Publishers. Day, C. C. Jr. (1983). Pushing pills out of the office. Modern Ofice Procedures, 28, 70-76. Day, R. (1989). Life events and schizophrenia: The triggering hypothesis. I n T. W. Miller (Ed.) Stressful life events. Madison, Connecticutt: International Universities Press, Inc. Donoghue, S. (1977). The correlation between physical fitness, absenteeism, and work performance. Canadian Journal of Public Health, 68, 201-203. Educational Research Service, Inc. (1980). Employee absenteeism: A summary of research. Arlington, VA: Author. Epstein, S . S . (1978). The politics of cancer. San Fransisco: Sierra Club Books. Fava, G. A., Perini, G. I . , Santonastaso, P. & Fornasa, C. C. (1989). Life events and psychological distress in dermatologic disorders: Psoriasis, chronic uticaria, and fungal infections. In T. W. Miller (Ed.) Stressfil life events. Madison, Connecticutt: International Universities Press, Inc. Fielding, J. E. (1982). Effectiveness of employee health improvements. Journal of Occupational Medicine, 24, 907-916. Freudenheim, M. (1990, March 13). Business and health: Research outlays to aid the aging. New York Times, p. D2. Glanz,K. & Mullis, R. M. (1988). Environmental interventions to promote healthy eating: A review of models, programs, and evidence. Health Education Quarterly, 15, 395-4 15. Gottlieb, N. H. & Nelson, A. (1990). A systematic effort to reduce smoking at the worksite. Health Education Quarterly, 17, 99118. Green, K. L. (1988). Issues of control and responsibiltiy in worker's health. Health Education Quarterly, 15, 473-486. Harwood, H. J., Napolitano, D. M., Kristiansen, P. L., & Collins, J. J. (1984). Economic costs to society of ulcohol and drug abuse and mental illness: 1980. Research Triangle Park, N.C.: Research Triangle Institute. Haskell, W. L. & Blair, S . N. (1980). The physical activity component of health promotion in occupational settings. Public Health Reports, 95, 109-118. Hinrichs, J.R. (1980). Controlling absenteeism and turnover: Highlights of the literature. Scarsdale, N. Y.: Work in America Institute, Inc.

Health

28 1

Holmes, T. H. & Rahe, R. H. (1967). The social readjustment scale. Journal of Psychosomatic Research, 11, 2 13-218. Hooyman, N. R. & Kiyak, H. A. (1988). Social gerontology: A multidisciplinary perspective. Needham Heights, MA: Allyn & Bacon, Inc. Kobasa, S . (1982). The hardy personality. In G . Sanders & S . Suls (Eds.) Social psychology of health and illness. Hillsdale, N. J. : Lawrence Erlbaum Associates, Publishers. Kristein, M. M. (1982). The economics of health promotion at the worksite. Health Education Quarterly, 9, 27-36. Laporte, W. (1966). The influence of gymnastic pause upon recovery following post office work. Ergonomics, 9, 501-506. Mackay, C. J. & Bishop, C. M. (1984). Occupational health of women at work. Ergonomics, 27, 489-498. Bombardelli, Magni, G . , Cirfini, A., Berto, F., Rizzardo, R., S . , & Miraglia, G. (1989). Life events and myocardial infarction. In T. W. Miller (Ed.) Stressful life events. Madison, Connecticutt: International Universities Press, Inc. Mahler, D. M. (1990). Health resource allocation in the 1990's. Business Horizons, 33, 118-126. health: Maier, S . F. & Laudenslager, M. (1985). Stress and Exploring the links. Psychology Today, 19, 44-49. Martin, J. (1978). Corporate health: A result of employee fitness. Physician Sports Medicine, 6, 135-137. Carducci, Matthews, K. A., Scheier, M. F.., Brunson, B. I., & B. (1989). Why do unpredictable events lead to reports of physical symptoms? In T. W. Miller (Ed.) Stressful life events. Madison, Connecticutt: International Universities Press, Inc. McCoy, J. L. & Weems, K. (1989). Disabled-worker beneficiaries and disabled SSI Recipients: A profile of demographic and program characteristics. Social Security Bulletin, 52, 16-28. B., Miller, A., Springen, K., Gordon, J., Murr, A . , Cohn, Stress on the Drew, L., & Barrett, T. (1988, April 25). job. Newsweek, pp. 40-45. Nathan, P. E. (1983). Failures in prevention: Why we can't prevent the devastating effect of alcoholism and drug abuse. American Psychologist, 38, 459- 467. Nelson, W. J. Jr. (1990). Worker's compensation: Coverage, benefits and costs, 1987. Social Security Bulletin, 53, 2-12. Novelli, W. D. & Ziska, D. (1982). Health promotion in the workplace: An overview. Health Education Quarterly, 9, 2026.

282

Ashcrajl

Parkinson, R. S . , Denniston, R. W., Baugh, T., Dunn, J. P., & Schwartz, T. L. (1982). Breast cancer: Health education in the workplace. Health Education Quarterly, 9, 61-72. Plyman, J. S . & Perkins, C. D. (1983). Fitness monitoring. Public Management, 65, 6-9. Quayle, D. (1983). American productivity: The devastating effect of alcoholism and drug abuse. American Psychologist, 38, 454-458. Rahe, R. H. (1989). Recent life change stress and psychological depression. In T. W. Miller (Ed.) Stressful life events. Madison, Connecticutt: International Universities Press, Inc. Rhodes, S. R. & Steers, R. M. (1990). A systematic approach to diagnosing employee absenteeism. Faculty research working paper series #79- 11. Syracuse, N. Y. : Syracuse Universit Y. Riggio, R. E. ( 1990). Introduction to industrial/organizational psychology. Glenview, Illinois: Scott, Foresman and Company. Robins, T. G. & Klitzman, S. (1988). Hazard communication in a large U. S. manufacturing firm: The ecology of health education in the workplace. Health Education Quarterly, 15, 451472. Robinson, P. K., Coberly, S . &Paul, C. E. (1985). Work and retirement. In R. H. Binstock & E. Shanas (Eds.) Handbook of aging and the social sciences (2nd ed.). New York: Van Nostrand Reinhold Company. Rom, W. N. & Upton, A. (1990). Asbestos-related diseases. New England Journal of Medicine, 322, 129, Rosenberg, L., Palmer, J. R., & Shapiro, S . (1990). Decline in the risk of myocardial infarction among women who stop smoking. New England Journal of Medicine, 322, 213-217. Rothman, M. (1988). Random drug testing in the workplace: Implications for human resource management. Business Horizons, 31, 23-27. Shapiro, A. P., Schwartz, G. E., Ferguson, D. E. E., Redmond, D. P., & Weiss, S . M. (1977). Behavioral methods in the treatment of hypertension: A review of their clinical status. Annals of Internal Medicine, 86, 626-636. Stevens, W. K. (1989, July 11). Scientists debate health hazards of electromagnetic fields. The New York Times, pp. C1, C10. Strober, M. (1989). Stressful life events associated with bulimia in anorexia nervosa: Empirical findings and theoretical speculations. In T. W. Miller (Ed.) Stressful life events. Madison, Connecticutt: International Universities Press, Inc.

Health

283

Taylor, S . (1986). Health psychology. New York: Random House, Inc. Tudiver, F. (1989). Dysfunctional uterine bleeding and prior life stress. In T. W. Miller (Ed.) Stressjhl life events. Madison, Connecticutt: International Universities Press, Inc. Vojtecky, M. A . (1988). Education for job safety and health. Health Education Quarterly, 15, 289-298. Werther, W. B. Jr. (1988). Loyalty at work. Business Horizons, 31, 28-35. Wright, C . C. (1982). Cost containment through health promotion programs. Journal of Occupational Medicine, 24, 965-968. Zavela, K. J., Davis, L. G., Cottrell, R. R.,. & Smith, W. E. (1988). Do only the healthy intend to participate in worksite health promotion? Health Education Quarterly, 15, 259-267.

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial1 Organizational Psychology - K. Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

285

CHAPTER 10 THE ROLES OF GENDER IN ORGANIZATIONS Kathryn Kelley and Deborah Streeter There has been an increasing concern about issues relating to gender within organizations during recent years, and likewise a corresponding growth of research in this area. This chapter will discuss a number of issues related to gender in organizations by presenting an overview of many of these issues, and its coverage will thus necessarily be broad. We begin by examining patterns of work, especially shifts in labor force participation, and the related factor of occupational sex segregation and its effects. The origins of career choice, the impacts of denigration and role congruence due to gender, and directions and dynamics in career development also receive scrutiny. Explanations for inequalities in the workplace, such as gender differences in compensation, focus on three major levels: the individual, structural, and societal-cultural theories. Observed as well as perceived gender differences in performance, attitudes, and perceptions receive attention both among managers and workers. Other issues of concern are also examined: work and family, health and stress, and sexual harassment.

TRENDS IN WORK PATTERNS Labor Force Participation The proportion of women and men in the labor force has changed dramatically, with women's participation showing the larger amount of growth. As of January, 1990, approximately 57.6% of women in the United States age 16 and over were either employed or are seeking employment (U.S. Department of Labor, 1990); the

Kelley and Streeter

286

Table 1 Labor Force Participation Rates of Women Ages 25-34 Country

1970

1988

Canada Japan France Germany Netherlands Sweden United Kingdom

41.2% 46.8 52.2 47.6 23.9 60.7 43.3

74.9% 54.5 74.5 61.5 55.4 89.4 66.0

Source: Based on data from the U.S. Department of Labor, 1990 corresponding percentage for men is 76.6%. It is projected that by the year 2000, 66 million American women will be in the workforce (Fullerton, 1987). The recent trends of early retirement and declining labor participation by men has caused these changes in workforce characteristics to be even more dramatic (Herz, 1988; Reskin & Hartmann, 1986). Participation rates differ between subgroups of women watthews & Rodin, 1989). Approximately 65.3% of never married women are employed, while 76.0% of divorced women participate in the labor force. The largest increase in labor force participation has been among women with young children; the rate for this group has risen to 54%, from 30% in 1970. These increases in the participation rates for women began rising shortly after World War I1 (Hayghe, 1990). While single women usually worked outside the home, this employment was most often terminated after marriage. During the war, married women's participation increased from 14% to 22 % . Women tended to leave the labor force after the war, but, within a few years, their participation rate equaled that at the height of the war. From 1950 to 1965, approximately 400,000 married women joined the workforce each year. After 1965, this number rose to 700,000. This trend has occurred in the United States as well as in other industrialized countries (U.S. Department of Labor, 1990). As illustrated in Table 1, Canada, Japan, and Sweden, among others, have seen dramatic increases in the labor force participation rate of women, and particularly women in their primary childbearing years.

Gender In Organizations

287

These changes have resulted from numerous factors, such as changing attitudes, family structure, and financial as well as individual needs. Career women are no longer viewed suspiciously; reflecting the trends noted above, it has become the norm for women to be employed outside the home. Women have become better educated and are more likely to remain in the workforce after marriage and childbirth, and are less likely to define their roles solely in terms of "mother" or "wife". Due to increased reproductive control, couples are able to plan when they will start a family, allowing parents more freedom to establish careers. Also, the rise in single-parent households and changes in economic realities often mean that some women have to work, simply to support their family. The character of womens' participation in the labor force has changed somewhat: they have taken two-thirds of the new jobs created after 1970, in decades commonly described as the start of the information age. The numbers of women have reached critical masses in professions including business and computer science; women have begun starting new businesses at twice the rate of their male counterparts (Paternoster, 1990). Some professions continue to employ predominantly men, slowly realizing advantages to employing women. Among international negotiators, for example, the use of women in these roles has proved to the United States' advantage in Japan, where career women are rare (Farnsworth, 1988).

Occupational Sex Segregation While women work for a variety of reasons, they are concentrated in relatively few occupations, typically those with very small percentages of men. This phenomenon is known as occupational sex segregation. Approximately 48 % of women worked in occupations that were over 80% female in 1980; 71 % of men were in occupations that were 80% male (Reskin & Hartmann, 1986). In order for sex segregation to be eliminated, about 60-70% of workers would have to change jobs (Bielby & Baron, 1986). For some occupations, the sex-type has switched, but the job has remained segregated; this is the case with the position of bank teller, which was once predominantly male, but is now considered a "female" position. Ironically, some occupations that are sex-integrated are segregated; men and women may have the same occupational title, but it is rare that they both work in the same organization (Bielby & Baron, 1986) or perform the same jobs, as shown in a study of radio programming (Lont, 1990). This organizational segregation tends to be masked when aggregated data are used to assess segregation, leading to conservative estimates of the degree of occupational sex segregation.

288

Ulley and Streeter

Consequences of Sex Segregation For the Individual. The consequences of sex segregation can extend beyond the workplace. These outcomes are usually more detrimental to women, who typically work in occupations lower in both pay and status, and less desirable on a variety of dimensions. The fact that traditionally female jobs are more poorly compensated has certain implications: the economic costs of entering such a job are higher, for example (Eccles, 1987). As a result women have less money to support themselves and their families. There are economic costs in the long run as well; sex segregation means women are less liekly to work in companies and sectors that provide pension coverage (Reskin & Hartmann, 1986). Also, because of their lower wages, women receive fewer social security and retirement benefits. Sex segregation involves a cycle, in which women are less likely to be placed in positions where they have access to training and advancement mechanisms. Thus, their goals for advancement are thwarted; they can remain in female jobs, and have difficulty switching to a male career line. The lack of opportunity to demonstrate their skills may lead them to become disenchanted with their work, thereby decreasing their motivation and productivity and reducing their chances for advancement. Being restricted to certain low status, low pay jobs has implications for women's personal lives as well. Their lower pay may result in their having less decision-making power within the family (Reskin & Hartmann, 1986). Their career may be secondary to their husband's, and may be the first to suffer when certain decisions are made, such as relocating, which is rarely done to enhance wives' job prospects (Gutek & Burley, 1988; Spitze, 1984). Being effectively barred from a large number of occupations also may limit women's possibility of self-fulfillment, satisfaction, and may have an effect on self-esteem (Eccles, 1987). In addition, maternal employment, especially in traditionally male jobs, may have beneficial effects on children, such as providing role models and serving to contradict societal gender role stereotypes; to the extent that women are segregated in only certain jobs, the positive social outcomes of maternal employment are limited (Eccles, 1987). For the Organization. While the effects on individuals are somewhat more obvious, there are a number of ways in which sex segregation impacts the organization as well. By engaging in segregation, the organization exposes itself to possible legal consequences (Noe, 1988). Court cases concerning discrimination can be lengthy, costly, and harmful to the organization from a human relations perspec-

Gender In Organizations

289

tive. Firms may be hurting themselves by not identifying and selecting qualified women; the potential loss of talent due to sex segregation and gender role stereotypes is great (Hunt & Michael, 1983; Ragins, 1989). Ironically, employers' reservations about hiring women for some jobs could lead them to be faced with a labor shortage for male jobs, even though there are competent, qualified female applicants (Reskin & Hartmann, 1986). I n addition, the lack of cross-gender interaction within organizations serves to maintain the status quo, by not providing role models of successful inter-gender interaction (Ragins, 1989). If occupations and firms were to some extent desegregated, men and women could become accustomed to working together, leading to more acceptance of men and women in nontraditional fields.

DYNAMICS OF CAREER CHOICE Socialization and Individual Factors Occupational sex segregation appears to be due to individual choices and influences, as well as restrictions imposed by organizations and personnel selection procedures. Individual factors come from a variety of sources, including socialization of gender roles (Eccles, 1987). Children learn that certain behaviors are appropriate for one gender, and that people have specific beliefs about the general abilities of each gender. Parents and teachers often believe that girls have better verbal skills than boys, and that boys are better at math; these beliefs lead to differences in expectations and support, which may eventually evolve into actual differences (Sandberg, Ehrhardt, Mellins, Ince, & Meyer-Bahlburg, 1987). While a son and daughter may each start out with ability in math, the son is encouraged to take math courses, while the daughter may not be (Yee & Eccles, 1988). Subsequently, the daughter may lack confidence in her ability, and not cultivate it. Other activities, such as play and household chores, are defined as gender-linked, and lead to differential skill development and expectations (Astin, 1984). Boys' competitiveness is often rewarded, but girls are encouraged to be more cooperative and sensitive to the feelings of others. Lever (1978) noted that boys were more likely to play outdoors and go farther from home, encouraging a sense of independence; girls, on the other hand, tend to stay indoors more, where physical play is restricted. Girls are more likely to be expected to care for younger siblings; boys are more likely to work outside of the house, mowing the lawn or helping to fix things. Shifts in the career interests of young women have occurred, sug-

290

kklley and Streeter

gesting that parental involvement in gender-role socialization has some limitations. For example, mothers of female college students express more interest, confidence, and support for home-oriented activities, whereas their daughters equally value home/and job-related commitments (Stake & Rogers, 1989). Yet, parental reward for their daughters' achievement striving increases interest in occupational competition and success (Paludi & Fankell - Hauser, 1986). The acceptance of gender-role definitions may lead to different expectancies of success for various tasks. Traditional stereotypes portray females as less competent and able than males, and the extent to which these stereotypes are believed to be self-descriptive influence individuals' educational and occupational decisions. In fact, males, in general, report higher expectations of success and perceived ability (Vollmer, 1986). Women's career plans and occupational values reflect their gender-role socialization. Relative to men, women plan to sequence education, career, marriage, and procreation in ways that will impede occupational achievement (Steel, Abeles, & Card, 1982). And compared to men, women rate more highly the importance of helping and considering other people in their own occupations, whether help delivery to clients or consideration of their own family's needs is involved (Bridges, 1989). The feminine sex role has a strong impact on career aspirations that can partially explain the routes that women can take in the work setting.

Effects of Education Formal education impacts future occupational choice. In the past, girls have been explicitly discouraged from entering traditionally male fields (Betz, 1989). While that may be the exception now, women experience disadvantages in career selection. Several diagnostic instruments are used to help students assess their abilities and make adequate career decisions. However, there are problems with these interest inventories, many of which are based on Holland's (1973) occupational model. This approach attempts to match a person's personality type with a list of jobs; the widely used Strong-Campbell Vocational Interest inventory is based on this mode1. Unfortunately, this model serves to encourage sex segregation because it is based on existing labor market structures, and the matching of the individual with gender-typed jobs (Moore & Ollenburger, 1986). Jobs are categorized as reflecting masculine and feminine personality traits. In fact, some occupations are subdivided into "male" and "female" lists; for example, male registered nurses fall into a category different from female registered nurses, with the

Gender In Organizations

29 1

female nurse position reflecting a more nurturant, caregiving purpose, and the male nurse position reflecting a more scientific purpose. Teachers and counselors may espouse a less directive style than in the past, and allow students to choose their careers without making gender an issue. Freeman (1979) refers to this as a null educational environment. However, this type of environment does not exist in practice; the attempt to portray education as gender-neutral ignores the external socialization environments of children. At the same time gender is being downplayed i n schools, females are being rewarded for engaging in appropriate behaviors learned in the home and from books, television, and other media. This lack of encouragement from teachers and guidance counselors in support of girls' pursuit of traditionally male occupations is one of the factors encouraging sex segregation to continue. If school counselors advised appropriate training for female students who displayed interest and ability in science and abstract reasoning, findings such as the earlier adoption of computerization by female than male patent examiners (Vernon-Gerstenfeld, 1989) might not be regarded as anomolous. Among college students, for example, men express greater confidence, interest, and acceptance of computer use, and this gender difference translates into more frequent enrollment in computer courses and more time spent using computers among men than women (Arch & Cummins, 1989; Miura, 1987).

Role Models and Social Support The availability of achievement-oriented role models differs for males and females; the lack of female role models in traditionally male fields has been noted as a barrier to women's development in these fields. In a study of 180 college women, Hackett, Esposito, & O'Halloran (1989) found that those who had higher educational goals and who expected their career to be very important in their lives report having been exposed to strong female teacher role models and to supportive adult males. The authors concluded that role model influence predicted career-related aspirations and choices (Hackett et. al, 1989). The importance of female teachers and professors was supported, and it was found that there were some negative effects of important male role models on women's career and educational decisions. A comparison of women in traditional and nontraditional college majors by Fitzpatrick and Silverman (1989) found that nontraditional women reported receiving more parental support for their career choice. College professors and same-sex friends were also found to have strong positive influences. Thus, it appears that the

&llq and Streeter

292

presence of female role models and a strong social support network may provide women with the encouragement and confidence they need to enter and succeed in traditionally male-dominated fields. To the extent that these influences are missing, women are at a disadvantage in entering these fields, and may be more likely to enter traditionally feminine fields. The prospect of maintaining the status quo is especially bleak for advanced training in the sciences and engineering, where only onefourth of those working toward doctorates are female. Most female students of science concentrate in the biological, social, and behavioral sciences rather than math or physics (Finkbeiner, 1987).

ORGANIZATIONAL EFFECTS ON SEX SEGREGATION The structure of employment opportunities affects the career choice of women (Astin, 1984). Economic conditions, unemployment rates, and other external factors influence whether women perceive jobs to be available. To some extent, both these constraints apply to both genders. However, restrictions also exist within the particular organization that the individual aspires to enter.

Gender Denigration Evidence of gender denigration and discrimination in hiring exists; in some research women have received lower evaluations than men on employment selection measures and in interviews (Arvey, 1979; Cann, Siegfried, & Pearce, 1981). However, Graves and Powell (1988) found no evidence that applicant's gender affected recruiters' evaluations; interpersonal attraction toward the applicant played a strong role. Elliott's (1981) results showed that female applicants for bank clerk positions received more positive ratings than males on several interpersonal dimensions such as manner and appearance, but no gender differences occurred in overall ratings. A partial reconciliation of these findings can be found in data showing that both applicants (von Baeyer, Sherk, & Zanna, 1981) and interviewers (Binning, Goldstein, Garcia, & Scattaregia, 1988) shift their behavior to reflect the gender-role composition of the dyad. Also Heilman & Martell (1986) discuss the limitations on situations in which gender bias occurs in hiring decisions. When raters have information about the successful performance of women in the specific job being filled, gender bias is reduced. Individuals making hiring decisions have been exposed to the same socialization forces that applicants have experienced, Gender stereotypes portray men as instrumental and competent, and women

Gender In Organizations

293

as less able and more emotional. A recent meta-analysis reported that the effects of gender on hiring recommendations tend to be small; only about 4% of the variance is explained by applicant's and rater's gender (Eagly, 1987; Olian, Schwab, & Haberfeld, 1988). However, these small effects can influence actual hiring; when research subjects are asked to rank job applicants that have been similarly rated, males are ranked first far more often than females (Dipboye, Arvey, & Terpstra, 1977). When there is only one position to be filled, this puts women at a disadvantage, and limits their access to jobs. Arvey (1986) has reviewed gender bias in job evaluations, and found that stereotyping affects ratings in discriminatory ways; the influence of this bias thus extends beyond the hiring decision.

Gender Role Congruence If gender denigration were the only process operating, women would have trouble getting any job; this is obviously not the case, because women dominate some occupations. Perhaps more directly related to occupational sex segregation is the tendency to match the gender of job applicants to the gender-type of the job, known as gender role congruence. Research has found support for this influence on selection (Arvey, 1979; Cohen & Bunker, 1975; Hodgins & Kalin, 1985). This process of selection partially explains why women attempting to enter a traditionally female job have the advantage over male applicants; preferential treatment for female applicants has been found in the selection for traditionally female positions including retail sales clerk (Arvey, Miller, Gould, & Burch, 1987) and school psychologists (Miller & Routh, 1985). While selection based on gender-role congruence may benefit women seeking traditional jobs, women trying to obtain nontraditional jobs are at a disadvantage. Male applicants are preferred; their gender matches the job's gender-type. Women may find that the job they want is not the job they are being considered for; Tucker (1985) related the story of a woman M.B.A. who was repeatedly asked to take typing tests for a clerical position. This process is not usually so overt, and often the people doing the hiring are not aware that they are exhibiting a preference. School administrators who participated in a selection research project stated that gender was irrelevant to the job of school psychologist, but showed an overwhelming preference for female applicants (Miller & Routh, 1985). In the context of gender, denigration, role congruent hiring, and difference i n performance expectations illustrate why women may have limited access to many occupations. Women express higher expectations for their own performance in female-dominated

294

Kelley and Streeter

jobs, but men expect to perform well in all fields regardless of gender dominance (Bridges, 1988).

GENDER EFFECTS ON CAREER DEVELOPMENT A number of factors related to gender influence career development and subsequent career attainment. Unfortunately, women often find that their career mobility is blocked, and that they are unable to advance into upper organizational levels.

Lack of Mentors Mentoring relationships affect career development and the attainment of opportunities (Hunt & Michael, 1983). Ragins (1989) identified several benefits of mentoring relationships. Mentors provide advice and training to proteges, as well as information which may not be obtained otherwise (e.g., that certain positions will become vacant). Also, mentors can bolster the self-confidence of proteges, and provide social and power-based support. High-ranking mentors can gain special access for proteges, thereby improving their chances for upward mobility. Kanter (1977) noted that proteges also gain a measure of reflected power from their mentors; proteges may find that they are able to open doors and obtain opportunities on their own, due to their association with a powerful sponsor. Because women face greater organizational and individual barriers to advancement than men, mentoring relationships may be especially important (Ragins, 1989). Mentors may buffer women from discrimination and help them circumvent obstacles. Since women tend to have less experience with organizational politics and have fewer resources for career guidance and information than men, mentors serve as a means to overcome these disadvantages. A review of gender and mentoring by Noe (1988) cited evidence that females who have had mentoring relationships report greater job success, job satisfaction, self-confidence, and use of job skills, Unfortunately, several factors function as barriers to women who want to obtain mentors. Because so few women work in the upper levels of organizations, a very small supply of female mentors exists. Also, because the potential mentors themselves face obstacles to advancement, they may simply not have the time or energy to be mentors (Ragins, 1989). Because male mentors tend to be higher in the organization, they are likely to have more power, which is a desired characteristic in a mentor; thus, potential proteges often prefer male mentors.

Gender In Organizations

295

However, there are numerous problems associated with crossgender mentoring relationships. Mentors tend to adopt proteges that they identify with; the protege reminds the mentor of himself or herself at a younger age. This identification becomes more difficult across genders (Kanter, 1977). In addition, becuase women tend to work in lower organizational positions, they have fewer interactions with powerful individuals and less contact with a potential mentor (Noe, 1988). Perhaps even more problematic is the reluctance of male mentors to take on female proteges, for a variety of reasons. Because of stereotypic and negative attitudes toward women in organizations (Dubno, 1985), they may not consider these women serious candidates for advancement. They may believe that women possess neither the characteristics necessary for managerial success nor adequate career commitment (Hulbert, 1988; Ragins, 1989). Female proteges are often perceived as risky, because the high visibility that results from their gender results in increased attention paid to them. If women subsequently fail, their lack of success is more likely to be noticed and reflect poorly on the mentor. Finally, they may have concerns about potential or perceived sexual involvement; there can be negative repercussions whether or not sexual or romantic relations occur, a situation which threatens the female protege's unstable position within the organization (Ragins, 1989). Overall, women are much less likely to enjoy the benefits to be had in a mentoring relationship, and this reduces their opportunities for advancement (Bhatnagar, 1988).

Career Ladders and Promotions The positions into which women are hired to tend to have relatively short career paths; there are fewer steps on the ladder, and the rungs are closer together (Reskin & Hartmann, 1986). Kanter (1977) points out that this lack of opportunity may have subsequent effects on behavior. Dissatisfaction with the degree of opportunity in a job leads to coping in a number of ways, such as disengagement from the job, a decrease in commitment, or an increase in complaining. This may lead to a self-fulfilling prophecy on the part of these individuals; management placed them in low level positions because of uncertainty or doubt about their future performance, but they are put in a situation where they have little opportunity to demonstrate their abilities. The process of promotion differs for men and women, and, ironically, women seem to receive more promotions than men (Stewart & Gudykunst, 1982; Tsui & Gutek, 1984). However, women do not advance through the organizational hierarchy at the same rate as

Kelley and Streeter

296

men. A study of over 2000 financial institution employees by Stewart and Gudykunst (1982) found a significant correlation of .41 between number of promotions and hierarchical level for men, but no significant correlation for women. When women reach higher organizational levels, the occupational socialization process appears to be similar for men and women, and their attitudes and behaviors converge (Gomez-Mejia, 1983). Thus, while women have a more difficult time advancing, once they attain some status, they are exposed to socialization forces similar to their male counterparts'. To summarize, occupational sex segregation appears to be the result of a combination of individual, organizational, and societal factors. Personal factors influence individuals to pursue one career or another, and gender-role socialization forces may lead men and women to have different views of their abilities. Gender-role stereotypes may also lead to discrimination; women's entry into some jobs is prevented in part because it is believed that they do not have the skills requisite for success. Men and women are also differentially channeled into certain jobs, because their gender matches the gendertype of those jobs. Within a job, different factors influence the career development of each gender; women find it more difficult to obtain mentors, access to organizational resources, and promotion into upper organizational levels. Some portion of women's failure to advance as far or as fast as men in organizations has been attributed to their exclusion from what Farr (1988) calls Good Old Boys Sociability (GOBS)groups. A shared history of class, gender, and masculine play in boyhood contributes to dominance bonding which supports masculine identity, male privilege, and reaffirmation of superiority through inclusion in GOBS. Shared sociability among men has been blamed as an additional factor for women not penetrating the glass ceiling that apparently shields most of upper-level management in Western societies (Morrison, White, & Van Velsor, 1987).

GENDER DIFFERENCES IN COMPENSATION Pay Disparity One result of sex segregation is a disparity in the average earnings of men and women. Overall, women earn approximately 64.7% of what men earn (Sager & Olson, 1986), and this differential increases when race is considered. Each of the forms of sex segregation has a unique contribution to the wage gap. The jobs that women tend to be concentrated in generally involve lower status and lower pay compared to men, and this explains a

Gender In Organizations

297

nble 2 Women’s Percentage of Men’s Earnings in Selected Occupational Categories as Reported by Seager & Olson (1986) Category Professional/Technical

Management/Administration Sales Clerical Craftworkers

Percentage 71.8 60.8 52.0 67.0 66.5

significant portion of the wage difference. Reskin and Hartmann (1986) estimate that occupational sex segregation accounts for 20 to 25% of the gender diffference in earnings. However, Seager and Olson (1986) report that there are differences within occupations as well (see Table 2); therefore occupational segregation alone cannot account for this effect. Cases involving pay differential due to gender discrimination are filed under the Equal Pay Act of 1963 or Title VII of the Civil Rights Act of 1964, as amended in 1965. Cooper & Barret (1984) reviewed 171 federal cases involving such discrimination, and recommended several changes and improvements in policies involved in them. For example, job evaluations should be made more objective; and seniority, merit, and personnel practices can be reviewed periodically for evidence of discrimination (Beuhring, 1989; Greig, Orazem, & Mattila, 1989; Mount & Ellis, 1989). Barrett, Alexander, Amesgart, & Doverspike (1986) cautioned against the inappropriate use of regression analysis to investigate salary discrimination, and recommended alternative procedures such as micro-analysis of the factors actually used to determine salaries. Because discriminatory pay practices have not been fully addressed by laws aimed at equal employment opportunity, comparable worth has been discussed as a remedy for the inequities found in organizations (Taylor, 1989). According to the doctrine of comparable worth, two jobs judged equal in worth should be associated with the same wage even though they may require different duties. Controversy about it has erupted, with arguments centering on its potential costs to the organization, and technical issues like

298

Kelley and Streeter

criticisms of the psychometric properties of job evaluation ratings as measures of job worth. Thylor's (1989) analysis of cases implementing comparable worth describes the financial cost as low. Buttner & Rosen (1987) provided data from an experimental simulation showing that the forces of supply and demand affect salary allocations to gender-typed jobs differently. A predicted labor shortage resulted in salary increases for male jobs, but not for female jobs. Among business school graduates, Dreher, Dougherty, and Whitely (1989) found evidence of anti-female bias in salary, despite the lack of differences between genders both in personal or organizational attributes and the use of tactics for upward influence. "liken together, these studies imply that gender bias exists in salary, and that women's work bears the brunt of this discrimination.

Explanations for the Compensation Gap Sex Segregation. In addition to occupational segregation, sex segregation within occupations, both between and within firms, accounts for a portion of the wage gap between genders (Reskin & Hartmann, 1986). Women are likely to work in smaller organizations, and even though they do the same work as men in other companies, they are likely to be paid less. Reskin and Hartmann cite the example of waiters; male waiters are more likely to work in more expensive restaurants, while small diners and coffee shops tend to employ women. Blau (1977) attributed more of the wage differential within an occupation to between-firm factors, rather than within-firm factors. This is not to say that there are not differences in the earnings of men and women within a particular firm, doing the same job. However, the size of this effect is unclear, and the difference varies from organization to organization. Frequently, even though men and women have the same occupation, differences in their specific jobs contribute to pay differences (Reskin & Hartmann, 1986). In sales jobs, men tend to work in "big ticket" departments, such as furniture or electronics, which typically cost more and produce more revenue, leading to larger commissions. Women, on the other hand, often work in less personally profitable areas, such as clothing or housewares. The relative seniority, experience, rank, and education levels of men and women can account for some disparities within organizations. Men's higher status in the organization gives them an advantage. However, these differences do not explain the entire wage gap. Sigelman, Milward, and Shepard (1982) examined the pay

Gender In Organizations

299

status of 424 full-time university executive, administration, and management personnel, and found an average salary difference of $5,343 between men and women. When rank and responsibility level were controlled for, this gap decreased, but there was still a difference of over $2,000, a difference the authors attributed solely to gender. Some analysts have attributed gender discrimination in wages to an underlying sense of deservingness for female gender-typed jobs and for work performed by women (Hegtvedt, 1989). Comparable worth strategies would correct this situation by substituting a fair system. Major (1989) has proposed that women have been socialized to perceive less entitlement to fair wages, thus tolerating injustice directed toward underpaid women through salary differentials. Jackson (1989) focuses on gender differences in the value of pay on the basis of relative deprivation theory, and reviews studies supporting the notion that women's satisfaction with low wages stems from their socialized expectations, and from a failure to recognize the gap between their desired and actual wages. Among 74 husband and wife management teams, for example, wives expressed significantly greater satisfaction with their lesser advancement and pay (Summers & DeCotiis, 1988). Women compared their salaries to female relatives and significant others as the primary referent, while men used other male managers as their reference point. Fortunately Jackson does not assume that the fault lies with women's job values, but recommends instead the provision of accurate information relevant to stereotypes about women's work as a prelude to their own decisions about what they value in a job.

EXPLANATIONS FOR GENDER-RELATED INEQUALITIES Theorists have proposed several explanations for sex segregation and the inequalities associated with it, though no one theory has provided a comprehensive model. Rather, these outcomes result from a combination of cultural beliefs and social processes, the structure of work and opportunity, discriminatory practices, and individual circumstances and decisions. We shall describe the theories according to their focus on the individual, organizational structure, or society.

300

Kklley and Streeter

Individual Theories: Noneconomic As discussed earlier, socialization does appear to have an influence on career choice. Reskin and Hartmann (1986) regard genderrole socialization as contributing to sex segregation on the basis of sex preferences for occupations identified to be appropriate to that sex, while individuals become disinclined, ignorant of, or pessimistic regarding their interests in most other occupations. This theory does not explain, however, inequalities between men and women in the same occupations, who may have very similar gender-role orientations. Often cited as a reason for gender disparities are the possible intrusions of family and marital responsibilities; women tend to have primary responsibility for child care and running the household. It is also often assumed that a husband's job takes precedence over the wife's, because he usually has a higher salary. While family roles may play an important role in women's work experiences (these will be discussed in a later section), it is not clear whether or how these factors contribute to sex segregation (Reskin & Hartmann, 1986). Bielby and Bielby (1988) examined the hypothesis that women put less effort into their jobs because of family responsibilities, and found that women actually put more effort into work than men, even though they spend almost twice as much time as men on housework. If marriage and family does have an effect on women's employment patterns, it would be expected that never married women would be in work situations similar to men. Examining samples from 12 developed countries, Roos (1983) found that while never married women were more similar to men than ever married women, there were still substantial differences in terms of pay and occupational prestige. Another individual level theory of inequality focuses on employers, rather than employees, and their potential to engage in discriminatory behaviors. Becker (1957) argued that employers have "tastes" for employees, preferring certain types of people in certain jobs. Employers who have a "distaste" for employing women will hire them if they will work for a wage low enough to compensate for the distaste. Also, their negative reaction to employing women does not necessarily have to be on the part of the employers; the basis for discrimination may originate from a perceived negative response from consumers. While little work has been done assessing this theory, Reskin and Hartmann (1986) cite some studies that report a believed client preference, such as for male lawyers (Epstein, 1981).

Gender In Organizations

301

Individual Theories: Economic Economic theory has also been used to explain gender inequalities in the workplace and labor market. Blau and Jusenius (1976) reviewed several economic theories. An initial theory focused on overcrowding: a large supply of women contrasts with a low demand for them as employees, because of their confinement to a limited set of occupations. This confinement is hypothesized to be due to tastes, discussed above. However, Blau and Jusenius concluded that this theory focuses on the consequences of sex segregation, making it inadequate to explain the phenomenon. Proponents of the human capital theory argue that women, for a variety of reasons, do not "invest" in themselves and their potential resources for employers; because of their shorter time spent in the work force, they tend to accumulate less value on the job market (Blau & Jusenius, 1976). It postulates differences in the quality of labor of men and women, because women lack certain desired characteristics. However, human capital theory cannot account for differences between men and women who do have the same qualifications and education, nor can it deal with the fact that several "female" jobs require very high levels of skill and training, such as nursing; it cannot explain why these positions are more acceptable for females. Overall, Blau and Jusenius (1976) conclude that individual level economic theories do not adequately explain sex segregation and the wage gap. The theoretical factors do not contribute persuasive causes of these differences, and segregation on the basis of gender is not necessarily the expected outcome of these models. Also, these models assume perfect competition in the labor market, which does not occur in concert with gender discrimination.

Structural Theories Both economic and noneconomic individual theories have focused almost exclusively on the supply of workers and their characteristics, and these explanations do not adequately account for the great extent of sex segregation. Any segregation equation should also include the demand side, or employers. StructuraZ theories move the focus from the individual employee to the labor market and organizational environment where the person functions. Structural theories have the advantage of combining some of the arguments of the individual theories with structural and demand considerations. The term statistical discrimination has been applied to the effect of individuals' beliefs that gender differences exist in prefer-

302

Kkliey and Streeter

ences, performance, and predictability in regard to certain types of work (Aigner & Cain, 1977). Gardner and Discenza (1988) illustrated the effect of this concept, by observing that raters stereotype types of jobs for which each gender typically applies according to their gender-role expectations. Statistical discrimination is one outcome of gender role congruence discussed earlier. Bielby & Baron (1986) base their discussion on the perceptions of the employers about men and women as groups, not individuals. Bielby and Baron suggest that employers regard men and women as different in terms of their marginal productivity for a given line of work. For example, employers may believe that, in general, women are more likely to quit or have less commitment to work than to their family. Employers may place women in jobs they consider "low cost", e.g., a high degree of turnover in those jobs does not seriously disrupt the organization. The truth or falsehood of this belief is less important than its effect: stereotypes lead to the perception that the individual is typical of his or her group (BIau & Jusenius, 1976). Unfortunately, under this model, small differences between men and women and contrasts in job requirements become amplified into large differences in gender composition; this process facilitates the perception of differences rather than similarities between the genders. It also has the potential to lead to self-fulfilling prophecies, in which women may actually exhibit higher turnover and lower ability because of their placement in jobs without the opportunity to disconfirm the employer's stereotypes. As an example of how statistical discrimination works, Larwood, Szwajkowski, and Rose (1988) apply rational bias theory to the study of situational influences in an organization which cue managers and others to expect that bias would be rewarded or punished. Organizational climate and specific attitudes congruent with it can thus support or discourage discrimination. The economic theory of the dual labor market posits two distinct labor markets. The highly developed, primary market conveys a high potential for upward mobility and higher pay. The less structured, secondary market has many points of entry but short promotion ladders and low wages (Blau & Jusenius, 1976). According to this theory, men and women have different access to these markets, due to access discrimination. Indeed, Bielby and Baron (1986) found that men more often work in large organizations and specialized jobs. However, this model did not explain sex segregation within these "separate" labor markets; men still had the advantage even when they worked in secondary firms.

Gender In Organizations

303

The dual labor market theory developed into the internal labor market theory (Blau & Jusenius, 1976). This model proposes one labor market, containing two categories of jobs: those filled internally and those filled externally. External jobs at the entry level provide general access to the organization. However, statistical discrimination and individual factors influence employment in these positions. Individuals attain internal positions, on the other hand, through promotions and skill acquisition after entering the organization. The type of entry job predetermines advancement, and women typically gain entry through jobs with low advancement potential and somewhat less opportunity to gain the necessary skills and knowledge to overcome this initial deficit. When women do have the necessary skills, they continue to experience statistical discrimination. Men may also encounter these factors: however, given cultural gender stereotypes, these processes work more often in favor of men. Women may have an advantage in gaining entry to traditionally feminine jobs, but they still lag in advancement. In addition to statistical discrimination and employers' personnel practices, a number of institutional factors influence both job entry and advancement, several of which discriminate against women (Roos & Reskin, 1984). Women may not have access to apprenticeship and federal job training programs, which often direct women toward gender-congruent occupations that essentially promote sex segregation. Entrance requirements may limit the number of spaces available for women, such as through veteran's preference or seniority systems. Women also tend to have limited access to recruiting and information networks; recruiters may not consider women interested i n certain jobs, and subsequently not provide recruitment information to them (e.g., for blue collar jobs). Also, workers in professional and managerial positions have more opportunities to develop personal, informal contacts that can benefit job advancement; however, women's employment in service and support positions makes them less likely to come in contact with the people who could help them (Roos & Reskin, 1984). Finally, differential practices associated with processes such as job posting can limit women's mobility; decentralized work environments may lead to jobs being posted only in some areas with favorable sex ratios. Reskin and Hartmann (1986) noted that existing sex segregation may continue through informal factors that exist in the workplace. Exclusion of women from traditionally male jobs and the necessary on-the-job training can lead to their failure. Women in any type of job, but especially male jobs, may experience overt harassment, both of a sexual and nonsexual manner (sexual harassment will be discussed in a separate section). These behaviors can create a hostile

304

kWey and Streeter

working environment, discouraging women from entering or remaining in male sex-typed jobs and supporting sex segregation.

Societal-cultural Theories Cultural gender role beliefs also influence sex segregation in the workplace, mainly by contributing to several of the theories and effects discussed earlier (Reskin & Hartmann, 1986). The division of labor on the basis of gender extends to jobs through a number of societal beliefs. These beliefs depend to some degree on the patriarchical nature of Western society, and some feminists argue that sex segregation maintains the status and power differential between genders. Examples include the beliefs that a woman's place is in the home and that women are not fit to be in positions of power, because they are naturally passive, submissive, and vulnerable. In fact, a series of laws passed in the United States in the 1800's and early 1900's prohibits women from jobs that might jeopardize their sexual purity. Laws prohibited women bartenders, messengers, meter readers, and elevator operators (Reskin & Hartmann, 1986). Beliefs about personality and ability differences between men and women supposedly made some jobs beyond the capabilities of women. As was stated earlier, no single theory adequately explains the phenomenon of sex segregation. Rather, a combination of several of these theories, representing multiple levels of analysis, is the likely answer. An integrative explanation must also take account of actual vs. perceived sex differences in organizational settings, which the next section examines.

PERFORMANCE, ATTITUDES, AND PERCEPTIONS The gender-typing of jobs stems from stereotypes and cultural gender-role beliefs (Reskin & Hartmann, 1986). For example, women, considered more nurturant and care-giving, pursue nursing and teaching; men, perceived as more instrumental and aggressive, seek jobs in business and other powerful occupations. The influence of specific stereotypes vary as well. For example, in the United States dentistry is a male occupation; in other countries, such as Poland and the Soviet Union, dentists tend to be female (Reskin & Hartmann, 1986). In India, domestic household workers are male, while construction workers are often female; this is reversed in the United States.

Gender In Organizations

305

Observed Gender Diflerences Per$ormance. Given the stereotyped differences between men's and women's work, it is valid to ask whether gender differences exist in performance or work attitudes. Overall, little evidence has demonstrated performance differences due to gender. Tsui and Gutek (1984) examined the performance of industrial middle managers, and found that women were rated as favorably or better than their male counterparts. Pulakos, White, Oppler, and Borman (1989) examined ratings by peers and supervisors in a military setting, and found that gender accounted for minimal variance in ratings, without consistent patterns in gender differences. Hall and Hall (1976) performed a laboratory study, in which subjects rated the performance of male and female managers with identical performance; the gender of the ratee did not affect the ratings. While these are only a few of the empirical studies done on gender differences in performance, they illustrate the lack of actual differences. Concerning the infrequent instances of gender differences in performance, Yammarino and Dubinsky (1988) point out an emphasis on gender-related differences because they may be more frequent, interesting, and important than similarities. These authors discuss the conclusions of other researchers that research finding gender differences in performance is not typically replicated, nor can any existing theory predict these differences. Attitudes. Several other variables have been examined with respect to gender differences, including work attitudes and job involvement. Keys (1985) compared male and female Certified Management Accountants on a series of career decision variables. Women reported having achieved the same amount of success as men, even though they had lower salaries and less experience. However, women also rated intelligence and willingness to work as more important than men, suggesting that they feel they have to work harder and be more intelligent in order to achieve the same level of success as men. Keys argued that women may judge their success more on the basis of difficulty of achievement rather than salary. Men and women perceived their work situations in slightly different ways in Keys' (1985) study. Interestingly, gender differences did not occur in ratings of the importance of several personal, family, and work life goals. While women viewed their chances of success somewhat differently from men, both genders rated their careers as equally important to them. A survey of 388 middle-aged adults found positive correlations between satisfaction with life and with work. However, women emphasized occupational status and

306

Kelley and Streeter

men, income, as a means of perceiving control over their lives (Crohan, Antonucci, Adelman, & Coleman, 1989). The significant sources of satisfaction in life and work converge between the sexes, but differences reflect the reality of sex segregation and gender bias in pay. Keys' (1985) conclusions are in part supported by the findings of Bielby and Bielby (1988), who found that women report putting greater effort into their work than men. Education influenced this finding; women who were better educated allocated more effort to their work activities than women in the same job who were less educated. The authors noted that statistical discrimination against men might, ironically, be more profitable than that against women. Yammarino and Dubinsky (1988) examined three different ways of explaining employee behaviors: gender, job, and situation. Gender was found to relate to involvement, with women reporting slightly greater job involvement than men, but these effects were quite small; overall, gender accounted for only about three percent of the difference, and the role of situational variables exceeded that figure. The authors concluded that few gender effects directly affected work responses, and that gender similarities outnumbered presumed differences.

Gender Roles. Bem (1974) and Spence and Helmreich (1978) initiated empirical, psychometric, and theoretical work on gender roles. A person's gender role consists of a description of the degree of masculinity and/or femininity characterizing oneself. The androgynous gender role is composed of a combination of these traits. Correspondence exists between gender role and occupational choice and success. Women in male-dominated jobs typically ascribe the masculine gender role to themselves; a similar relationship exists for men, female-dominated jobs, and femininity (Yount, 1986). Women's masculinity enhances their coping with problems by focusing effort away from emotions onto solutions, especially in maledominated occupations (Long, 1989). The androgynous role, with its subcomponent of masculinity, characterizes those in supervisory leadership positions more often than does the traditional role (Kapalka & Lachenmeyer, 1988). Among engineers of both genders, masculinity alone or together with femininity correlates with higher salary, responsibility, and satisfaction (Jagacinski, 1987). Instrumentality, or the tendency to view oneself as a doer who accomplishes goals, contributes to the major portion of masculinity as a success-related trait. Minor drawbacks to the androgynous role do exist. For example, women may have difficulty performing assertively in a socially

Gender In Organizations

307

approved manner, independently of others' perceiving their assertiveness as aggressiveness, while regarding men' s aggressiveness as appropriate. Training for behavioral assertiveness, particularly in the presence of men, can overcome their lack of participation in mixed-sex discussions (Lewittes & Bem, 1983). A second problem stems from androgynous and masculine women's success in a variety of occupational and domestic tasks: role strain, or perhaps workaholism, can result. But awareness of the effect of this strain and possible solutions may counteract it (Doerfler & Kammer, 1986).

Perceived Gender Direrences Despite the existence of relatively little evidence of gender differences, perceived differences reflect beliefs that there are differences in performance or in the attributed causes of some performance (Deaux & Emswiller, 1974; Garland, Hale, & Burnson, 1982; Reed, 1983). In addition, some types of situations promote perceptual differences.

Sex-inconsistent Jobs and Tasks. Deaux (1976) argued that performance attributions depend on the degree of congruency between the task and actor's gender. Specifically, ability inferences occur when the person succeeds at a gender-consistent task. For a genderinconsistent task, different outcomes lead to different attributions for males and females. For a male task, the success of a male actor is attributed to internal factors, while female actors receive external attributions. On the other hand, success on a feminine task leads to internal attributions for both males and females, due to the stereotypes of male competency and the perception of feminine tasks as easier. In a laboratory study, Deaux and Emswiller (1974) tested these hypotheses using male and female tasks of equal difficulty. Success on the male task was attributed to ability for the males, but more to luck for the females. On the feminine task, ratings for males and females did not differ, suggesting an overall tendency to see men as skillful in both types of tasks. The Token Situation. The composition of a group can have profound effects on interactions between members; Kanter (1977) discussed the theory of proportions in her study of a large industrial organization. Being the only person from a particular class or group of people makes gender-role stereotypes salient. The "femaleness" of a woman in an otherwise all-male group is made salient. Kanter (1977) refers to this individual as a token, and argues that token status results in a number of processes that influence stereotyping.

308

Dlley and Streeter

Specifically, tokens are highly visible, receiving more than their share of awareness from other individuals. Also, token status leads to making contrasts between the token and the members of the dominant group. The token's difference increases the awareness of dominants about their similarities, and they exaggerate the perceived differences between token and group members. Finally, assimilation occurs, in which dominant group members attempt to fit the token into some stereotypic category. According to Kanter (1977), the token situation facilitates stereotyping, because the existence of only one instance of the token status leaves group members unable both to compare the token to someone else of the same category and to conclude that there are distinctions between them. Tokens are often perceived as an instance of a "general" framework, so that they are assumed to represent their entire category (e.g.., "the woman's point of view") instead of being perceived as unique individuals. Processes of social perception lead to a variety of performance pressures on the token. The token woman (or man, minority, etc.) is more easily noticed and remembered, especially when she makes a mistake. Also, the added pressure of representing the entire group results in the actions of a token woman having implications for other women in the organization. Token edipse may occur; the token's status overshadows the person's accomplishments. In her study, Kanter (1977) reported that the token status of a woman almost invariably led to comments on her appearance or other non-ability traits, characteristics that were never discussed about her male coworkers. A number of studies have attempted to investigate Kanter's (1977) theory of the effect of proportions and group gender ratios. Izraeli (1983) studied union committees with varying proportions of men and women. Women considered themselves less influential when they formed a minority i n the group, but not in balanced groups. Also, a larger proportion of people in the skewed groups thought that women's purpose on the committees was to work for "women's issues", such as child care; these issues did not include eliminating wage disparities or differences. In addition, a comparison of scores on a scale of gender stereotypes revealed that women in balanced groups viewed women as more skillful in leadership than men, compared to women in skewed groups. Overall, however, the explanatory value of group proportion was limited, due to a main effect of respondent's gender. Men exhibited a pro-male bias regardless of their group proportion. However, this pro-male bias was stronger in skewed groups, and the author concluded support for Kanter's (1977) theory of proportions. Other researchers have argued that gender stereotypes produce

Gender In Organizations

309

different effects of tokenism. In a study of small group functioning, Crocker and McGraw (1984) varied gender ratios within the group. Token men were twice as likely as token women to be identified as group leader. People in groups with tokens more often indicated that they preferred to work in a group with a different gender ratio, especially when the token was a woman. Yoder and Sinnett (1985), in a field experiment which assigned male and female workers to a token status position, found that token men did not experience the negative consequences of tokenism experienced by women as Kanter (1977) proposed; regardless of their token status, men, not women, more often received a recommendation for promotion. Adelmann (1989) found a curvilinear relationship between the proportion of men in an occupation and their job attitudes. The highest degrees of job satisfaction, interest, and commitment occurred in occupations which were predominantly female or male, suggesting rewards for men in either type of job. Token or solo status has detrimental effects on women, but not on men in studies of policewomen and male nurses (Ott, 1989). Yoder and Sinnett (1985) regard tokenism effects as due to more than the numbers considered separately.

The Managerial Level Attitudes. The influences of both gender inconsistency of a job and token status assume greater importance at the managerial than subordinate level. Often women are the minority in this male sex-typed position (Powell, 1988). Women managers typically face unenviable pressures and expectancies. Dubno (1985) reported that male M.B.A. students held negative attitudes toward women managers in 1975; another sample tested in 1983 showed no identifiable changes. Heilman, Block, Martell, and Simon (1989) reported that descriptions of men in general resembled descriptions of successful managers more closely than did descriptions of women in general. The descriptions of successful women managers resembled those of successful managers, but not as closely as descriptions of successful men managers did. However, Schein, Mueller, and Jacobson (1989) provided data indicating that women MBA students gender-typed the managerial job less strongly than their male peers did; the tendency to assume that this position is appropriately male but not female may be changing slowly. Cadets at the Coast Guard Academy have expressed more tolerance for women in the military following their legally mandated integration into its activities in 1976 (Stevens & Gardner, 1987). With increased exposure to women performing successfully in the same roles as men, attitudes shift toward greater acceptance. A number of discrepancies between descriptions of men

3 10

Kelley and Streeter

and women managers cast women in a negative light. For example, women managers were described as bitter, quarrelsome, and selfish, while men are described as appropriately dominant. (Heilman et al., 1989). Morrison, White, and Van Velsor (1987) pointed out that, even though similarities between men and women managers outnumber differences, women have a more limited range of behavior that they are viewed as appropriately engaging in; they cannot appear too feminine, but at the same time should avoid excessive masculinity--toughness without machismo, for example. Female managers perceived a peer's videotaped disagreement with a male manager as aggressive, but male managers viewed her as appropriately assertive (Mathison, 1986). Thus women's attitudes toward their own gender's initiatives can be less positive and tolerant than men's. Both male and female managers are expected to take risks, but women's increased visibility reduces tolerance for their mistakes. Morrison et al. (1987) conclude that unrealistic expectations of executive women have resulted from stereotypes, and that they face significantly more hurdles than men.

Appearance. Dress and attractiveness have assumed the status of assets perceived as important to attributions about managerial performance. The chapter in this volume by Stone, Stone, and Dipboye discusses general research on this variable. For example, physical attractiveness has inflated ratings of job applicants (Cash, Gillen, & Bums, 1977; Gilmore, Beehr, & Love, 1986). Business undergraduates, however, attributed the successful job performance of managers to negative factors if their faces were attractive--women succeeded because of luck or bias, and men were described as devoting little effort to their jobs. Despite the prevalence of skirted suits as the uniform for women, professionals depicted in ads in business magazines (Saunders & Stead, 1986) suggest that style of dress has not been adopted by the majority of female managers. More commonly, perceptions of both men and women vary as a function of the color and overall style of dress. Wearing the color blue or a jacket can create a more favorable perception of both men and women, compared to wearing red or no jacket (Scherbaum & Shepherd, 1987). Factors due to the situation or individual which make gender-role stereotypes salient may also lead to different attributions for men and women. For example, physical attractiveness (Heilman & Stopeck, 1985) may serve to enhance perceived gender-congruent characteristics and differences in evaluations. In the Heilman and Stopeck (1985) study, subjects made performance evaluations of attractive or unattractive males or

Gender In Organizations

311

females in either a male or a female gender-typed job. Males' attractiveness did not influence their performance ratings. However, attractiveness and job-type strongly affected females' performance ratings. They considered attractiveness beneficial to females in a feminine job, but thought it had a negative effect in a masculine job. Further analysis indicated that attractive females were perceived to be more feminine than unattractive females, and thus less appropriate for male gender-typed jobs. This enhancement of stereotypes did not extend to differentially attractive males; according to the authors, the "what is beautiful is gender-typed" stereotype only applies to women.

Performance. Are there gender differences in leadership and managerial performance? Occasionally differences occur, as in Peters et al. (1984), in which there was a pro-female bias, but more commonly no differences are reported (Goktepe & Schneier, 1988; Izraeli & Izraeli, 1985; Tsui & Gutek, 1984). Powell and Butterfield (1982) noted that male and female leaders received similar evaluations when they engaged in the same behaviors, and ratings depended primarily on the success of the work group. Dobbins and Platz (1986) reviewed research on gender differences in leadership and determined that differences in the ratings of leader effectiveness appeared mainly in ratings of male and female leaders in laboratory but not in field studies, favoring males. Differences occur in settings where the rater has comparatively little information about the leader. This coincides with the conclusion of Brown (1979), who reported that differences occur with student subject samples; research with actual managers tend to produce no differences. Management Style, Because social roles have a powerful effect on the behavior of women and men, questions exist about whether women would behave differently in positions of power (Mednick, 1987). One potential model is Norway, where in 1987 seven of 17 cabinet ministers were women, and family needs receive strong government support (Overholser, 1987). Both university women with job experience (Russell, Rush, & Herd, 1988) and managers of both genders and their secretaries (Statham, 1987) expected women to be task and people-oriented, or in theoretical terms to display both consideration, a feminine quality, and initiating structure, a masculine characteristic (Cann & Siegfried, 1990; Fleishman, Harris, & Burt, 1955; McGregor, 1967). These subjects described women as exhibiting more of both dimensions than men. The expectations for the behavior of women managers may

3 12

Kelley and Streeter

exceed those for men, at least among those with job experience. Among business undergraduates, in contrast with the studies cited above, males perceived women managers as having less knowledge and skill, but greater interpersonal skills; females preferred a male supervisor (Frank, 1988). The influence of women's assumed prowess in interpersonal skills appears in Lipman-Blumen' s (1988) theory of connective leadership. Defining it as leadership that relates the individual to the task and group through collaboration and use of instrumental styles including manipulation, tipman-Blumen (1988) also regards it as altercentric, by bringing others' views to bear and nonegocentrically taking pride in others' accomplishments. This style involves a new image next to that of the powerful leader with a masculine mode, and focuses on consultation, networks, and negotiations instead of competing, taking charge, and individually achieving. Rosener (1990) discusses a similar concept called interactive leadership; the drawback found in research on this construct is its focus on self-report instead of observation.

Attributions. Despite the lack of evidence for gender differences in leadership abilities, different success attributions exist for men and women middle managers, indicating a pro-male bias (Powell & Butterfield, 1982). Garland, Hale, and Burnson (1982) studied the attributions of 110 state human service employees about the success or failure of a female manager. Despite generally favorable attitudes toward women managers, gender differences in attributions resulted. Men made weaker attributions to ability and effort and stronger attributions to luck than did women i n both success conditions. However, this attributional pattern for success and failure did not extend to ratings by chief executives i n organizational settings (Heimovics & Herman, 1988). As Morrison et al. (1987) noted, masculine behavior among women managers may have negative implications, particularly in research with small groups. Watson (1988) had students play the roles of a manager and subordinates in groups with a female manager. In half of the groups, the female manager acted in a dominant, gender-inconsistent manner, and female manager in the other groups acted in a more gender-consistent, considerate manner. Dominant women reported less acceptance of their suggestions, and this sentiment was mirrored by the subordinates. Male subordinate groups rated dominant leaders less influential than considerate leaders. Mixed-gender subordinate groups did not differentiate the two types of leaders, but male subordinates rated the manager more positively than women did. While this particular study had limited power, the results, as Watson (1988) states, are provocative, and research

Gender In Organizations

3 13

should attempt to identify ways women can use toughness without negative effects. Two circumstances found in various studies serve to mitigate negative perceptions of women as managers. First, gender differences occur relatively less frequently in field than laboratory samples, as discussed previously (Dobbins & Platz, 1986). Second, having specific information about a manager's competence limits gender differences or biases (Powell & Butterfield, 1982). Finally, gender per se may not be the important variable; research on emergent leaders by Goktepe and Schneier (1989) shows that individuals with a masculine gender-role orientation were more likely to become group leaders. Cann and Siegfried (1987) reported that the traits desired in a manager depend on who is asked, subordinates or higher level supervisors; supervisors preferred more masculine characteristics than did subordinates. Perhaps more importantly, though, both sets of respondents rated gender-neutral characteristics as more relevant for manager effectiveness. Thus effective leadership did not require masculinity. Further, reported preferences for male managers may not represent gender bias. Because male managers occupy higher positions within the organization than female managers, this preference may reflect a more positive attitude toward a manager assumed to have more power and more ability to provide desired resources (Liden, 1985).

WORK AND FAMILY A significant amount of research has focused on the effects of women's working on their families and on their more traditional roles, paralleling women's increasing labor force participation and changes in attitudes. Zedeck and Mosier (1990) identified several models proposed to describe and explain the interaction between work and family, all of which focus on the individual, rather than the family unit. Spillover theory suggests that what occurs in one situation also occurs in the other, either i n a positive or negative direction. Compensation theory proposes that work or family activities are undertaken as a substitute for what is missing in the other area. The segmentation model considers work and family separate. The conflict model proposes that sacrifices occur in one area in order to achieve satisfaction in the other. Finally, instrumental theory argues that one situation serves as a means to obtain goals in the other setting. Each of these models has received some support but they have not been examined in relation to one another, illustrating one of the problems with this literature. Many unrelated findings come from a

314

Kdley and Streeter

variety of sources and lack organization. In addition, most of the research reflects the nuclear family form, and seems inapplicable or inadequate to explain nontraditional families, such as single-parent households. Zedeck and Mosier (1990) conclude that more research should identify and examine links between work and family situations. Chusmir and Mills (1989) provided an example of a linkage by reporting the conflict resolution styles male and female managers used at home vs. work -- accommodating at home and competitive at work. Further, Galinsky (1988) found that employees' responses to scales of supervisory support and corporate culture regarding family life revealed the degree of conflict they experienced between work and family.

Effects of Multiple Roles Competing hypotheses exist about the effects of multiple roles on women (Pietromonaco, Manis, & Frohardt-Lane, 1986), such as job holder, mother, wife, housekeeper, and so on. Some have argued that women with several roles experience more stress and more negative emotions, while others have adopted the position that a variety of roles and statuses leads to benefits and acts as a buffer against stressors in any single role. A study of 500 women by Pietromonaco et al. (1986) found higher self-esteem among women with more roles, and observed that women with fewer roles expressed less job satisfaction; no differences were reported in satisfaction with marriage or partnership. In this study women's lives involved a high level of stress, but stress did not correlate with the number of roles held. The authors concluded that having multiple roles was psychologically healthful for many women, though they warn against interpreting a causal direction for this relationship. Work roles have taken on increasing importance in the lives of women, and the study of individual differences in orientation toward them and family has just begun (Gilbert, Dancer, Rossman, & Thorn, 1991). Matthews and Rodin (1989) argue that employment may provide several kinds of benefits to women, providing them with resources, social interaction, self-esteem, and a sense of accomplishment. In a sample of blue-collar men and women with and without children, mothers endorsed the greatest number of intrinsic and extrinsic reasons for working, perhaps due to their increased aspirations and responsibility (Hams & Earle, 1986). W t , Padgett, and Baldwin (1989) performed a meta-analysis to estimate the relationship between job and life satisfaction. As proposed by Kavanaugh and Halpern (1977), the correlations observed between job and life satisfaction for women were larger for more recent

Gender In Organizations

3 15

studies, suggesting that work roles have taken on a more central role in women's lives. Comparatively little work has been extended to the multiple roles of men, perhaps reflecting the cultural belief that men's major role is that of worker or breadwinner. Sekaran (1986), in apparent contrast to this trend, found wives' level of stress exceeded their husbands' and that wives indeed reported more multiple roles. Among psychologists, women reported more interference with their family lives by their work than men, and equal amounts of conflict in work produced by their family involvements (Burley & Gutek, 1988). In a sample of 76 dual-earner, professional couples with children at home, women restructured their work more than men to accommodate family needs, but men with low self-efficacy restructured more than men with high levels of this dimension (Brett, Dornbush, & Yogev, 1988). These findings suggest that women's adaptation to multiple-role stress is more generalized than men's. Some researchers have examined t h e effects on men of their wives' working. Parasuraman, Greenhaus, Rabinowitz, Bedeian, and Mossholder (1989) hypothesized that a husband's well-being would be indirectly related to his wife's employment status, through the effects of wife's employment on husband's time commitment to work, work-family conflict, and satisfaction with child-care. Husbands of working women reported lower job satisfaction and lower quality of life than husbands of housewives; however, these differences were small, explaining less than two percent of the variance in the dependent variables. This research suggests that as men start to fulfill some of the responsibilities traditionally handled by women, their well-being decreases slightly, while women benefit from taking on some of the roles traditionally held by men.

Household Responsibilities It is commonly accepted that women have the primary responsibility for running the home and caring for the children; this assumption has a number of implications for women's work, as one of the explanations for women's lower status in the workplace (discussed in previous sections). Research has consistently found that women devote much more time to housework and childcare than do men (Bielby & Bielby, 1988; Levant, Slatterly, & Loiselle, 1987; Nyquist, Slivken, Spence, & Helmreich, 1985). While the total time men spend on household responsibilities has increased slightly in recent years, their contributions are primarily performed on weekends, rather than weekdays when the demands on working women are at their highest (Douthitt, 1989). Ironically, men's household

316

Dlley and Streeter

labor time is unrelated to their wives' employment status; the amount of time women spend working outside the home is unrelated to the amount of time men spend on household tasks (Levant, Slatterly, Loiselle, 1987; Nyquist, Slivken, Spence, & Helmreich, 1985).

Child Care Child care has traditionally been considered a women's issue, which serves to reinforce traditional gender roles within the family (Kamerman, 1985). For example, individuals expressing traditional gender roles endorse beliefs that maternal employment has more negative effects on children than do androgyns (Greenberger, Goldberg, Crawford, & Granger, 1988). Child care can also be considered a parental responsibility, illustrated by the increasing call for parental leave in addition to or instead of maternity leave. Despite this, women are considered primarily responsible for child care, and inadequate child care is frequently given as a reason for women not being in the labor force or for being restricted in their employment opportunities (Kamerman, 1985). Child care arrangements have implications for working women's emotional adjustment (Ross & Mirowsky, 1988); difficulty in obtaining adequate, satisfactory child care has influenced working mothers' depression. Husbands' lack of participation in child care also affects the development of their wives' depression; working women whose husbands assist in child care report less depression. General support from their husbands, and satisfaction with the worker and parental roles, combine to provide a buffer against working mothers' depression (Tiedje & Downey, 1988). Number of children is unrelated to working mothers' well-being, and child care concerns do not affect husbands' well-being (Ross & Mirowsky, 1988). Individual women, employers, and the government have dealt in varying ways with workers' needs for child care. Women today marry and have children years later than a generation ago, attempting to establish their careers before beginning their families (Baber & Monaghan, 1988). Young professional women in traditionally male fields have begun to choose career over family more often than older cohorts of peers (Hulbert, 1988). However, having a strong, intrinsic religious orientation counteracts this trend among undergraduates (Jones & McNamara, 1991). To alleviate conflicts between work and family, some employers have instituted parental leave programs, as well as alternative work schedules including flextime and job sharing (Zedeck & Mosier, 1990). Only a small proportion of organizations have implemented these programs, so

Gender In Organizations

317

that a great many employees do not have access to these benefits. Organizations have typically been cautious about interfering in employees' personal lives, and have had a traditionally slow response to family-related needs. Zedeck and Mosier (1990) point out statistics that organizational policymakers are not directly exposed to the problems, because most have wives who are primarily homemakers. They suggest that organizational decision-makers could increase their degree of sensitivity to the intricacies involved in trying to balance both work and family roles. Governments have also been called upon to respond to the need for child care and develop pro-family national policy or public programs to assist working parents. Unfortunately, the stereotype of the typical, nuclear family still involves two parents, usually with the mother able to choose whether or not to work; policies based on this are inadequate to meet the needs of large numbers of people (Russo & Denmark, 1984). Good quality infant and child care are beyond the means of many people, who cannot afford prolonged absence from the workforce (Zigler & Muenchow, 1983). The United States lags notably behind other industrialized nations in providing assistance for day care. It is the only industrialized nation that does not have a federal family leave policy (O'Carolan, 1987); in fact, the majority of working women do not have access to maternity leave (Zigler & Muenchow, 1983). European counties, especially Scandinavian nations, do far more to assist working parents; for example, Sweden, which has been the most responsive to the child care issue, provides for paid parental leave, public day care and preschool, and the option of part-time work (Sorrentino, 1990). While organizational responses to the need for child care have grown slightly, the costs of these programs are high. Relatively few companies are willing to incur the added expenses associated with child care, especially leave programs (O'Carolan, 1987). Likewise, a great deal of controversy exists within government over what should be done and who should bear the responsibility. Unfortunately, while these issues are being considered, the burden of child care still falls on the parents, usually the mother, and this has implications for both employment characteristics and emotional adjustment. Research in the area of work and family relations has increased as women enter the labor in greater numbers, resulting in several new concerns. A "male" model has often been adopted in workfamily research (Voydanoff, 1988). Findings from studies of men may be incorrectly generalized to women, and much research reflects the belief that the home is a stress-free environment, which is probably not the case for working mothers (Baruch, Biener, & Barnett, 1987). Also, researchers tend not to distinguish between

318

Kelley and Streeter

"dual-earner" families and "dual-career" families. Additionally, in applying and testing the models identified in the beginning of this section, little research has attempted to compare these models or suggest their confluence (Zedeck & Mosier, 1990).

HEALTH, STRESS,AND WORKING WOMEN Interest in the areas of health and stress among working women has developed relatively recently, partly reflecting women's growing share of the workforce, and partly out of a realization that work done with men may not apply to women (Baruch, Biener, & Barnett, 1987; Robison, 1987). Issues of adherence to health care advice, social support (Dubbert, 1988), and health care beliefs (Stem, 1986) deserve more attention in relation to women's behavior (Blechman & Brownell, 1988). Studies on stress and health as applied to each gender tend to be qualitatively different, making comparisons between genders difficult (Haw, 1982). For example, work-related studies of women's stress tend to include home variables or the effects of multiple roles, while studies of men do not; also, the former usually involves less specificjob environment characteristics than research with men (Haw, 1982). Two other chapters in this volume touch on health issues--those by Ashcraft, and Stone, Stone, and Dipboye. Overall, results regarding differences in work-related stress between men and women seem to be contradictory; a meta-analysis by Martocchio and O'Leary (1989) found no gender differences in work stress. Other reviewers have suggested that while both men and women experience stress, the stressors to which they are exposed are different, leading to different manifestations of stress (Jick & Mitz, 1985). It has been proposed that men are more prone to physical illness, while women experience more psychological disturbances. However, though research on women and reactions to stress is limited, both genders exhibit the two types of reactions, and individual personality variables have helped to explain the relationship between environmental stressors and subjective responses (Haw, 1982). The possible sources of stress and subsequent health effects are vast, but work-family conflict and job characteristics have received the most attention. Some home-related variables have already been discussed, such as the ease of arranging child care (Ross & Mirowsky, 1988). The interaction of work and family stressors may result in buffering or additive effects between the settings (Baruch et al., 1987). Haw (1982) cited research on job stress resulting partly from having little control over one's work, in combination with family

Gender In Organizations

319

responsibilities. These types of jobs may lead to an increased risk of heart attacks; this was the case with female clerical workers who had children, as compared to housewives and working women with no children (Reskin & Hartmann, 1986). While much research indicates that employed women are psychologically more healthy than women with few roles, some job conditions act as stressors (Baruch et al., 1987). One stressor is the use of gatekeeping transactions, strategies performed to gain resources needed from coworkers and supervisors, such as power tactics and forcing an issue. Among 78 women professionals, frequent gatekeeping, especially when it was of a negative nature or when most interactions occurred with men, was related to an increased use of drugs and to overeating (Wallston, Hoover-Dempsey, Brissie, & Rozee-Koker, 1989). Traditionally female jobs typically allow the incumbent little control or autonomy, are usually somewhat tedious, and have little opportunity for advancement (Jick & Mitz, 1985). A study by the National Institute of Occupational Safety and Health in 1975 found that, out of 130 occupations, secretaries had the second highest incidence of stress-related diseases. Token women interviewed by Kanter (1977) reported a large degree of stress placed on them by their unique situation. One potential stressor neglected in relation to women has been the effects of unemployment (Baruch et al., 1987; Matthews & Rodin, 1989). Women have been assumed to be somehow unaffected by unemployment, perhaps reflecting the stereotype that employment is a secondary role for most women, and that they work out of choice, not necessity. Baruch et al. (1987) have mentioned that job loss for women has incorrectly been viewed as meaningless or welcome. Ironically, the effects of employment are easier to study among women, precisely because some do choose to remain out of the labor force; fewer men are unemployed, many for health reasons (Matthews & Rodin, 1989). Overall, research on working women's health and stress is increasing, but it is limited and different from studies of men in ways that make comparisons difficult. There is a need for more controlled, balanced research involving both genders, and individual characteristics need to be taken into consideration. Also, more research is needed on the structure of work and its effects on working women (Jick & Mitz, 1985); specific job characteristics and variables have rarely been investigated in studies involving women.

320

mlley and Streeter

SEXUAL HARASSMENT

Perceptions and Consequences of Harassment A topic recieving considerable attention recently is the sexual harassment of employees, attention that is certainly warranted given the breadth and consequences of the problem (Hearn, Sheppard, Tancred-Sheriff, & Burrell, 1990). Reports of women being harassed on the job range from 42 to 90% (Terpstra & Baker, 1987). Men also report being harassed; in one survey of federal employees, 15% of men reported having experienced sexual harassment ('Ihngri, Burt, & Johnson, 1982). Rubin and Borgers (1990) reviewed studies of harassment in university settings and suggested that the incidence of harassment is underreported due to ambiguities in its definition. Harassing behaviors range from verbal comments of a sexual nature to nonverbal sexual attention to rape; according to Title VII of the U.S. Civil Rights Act of 1964 (amended in 1965) prohibiting discrimination in employment, the important point i s that these behaviors are unwelcome, are made part of an individual's terms of employment, or serve to create an intimidating and hostile working environment. Most incidents are repeated, and many persist for several months (Pryor, 1985). Harassing behaviors have generally been divided into two types: quid pro quo harassment and harassment as a continuing work condition (Coles, 1986; Tangri et al., 1982). Quid pro quo harassment involves the exchange of sexual favors for some job outcome or resource; this can be negative, as in the case of exchanging sex to retain a job, or positive, to obtain a promotion. This type of harassment commonly involves female victims, and the most serious forms are more likely to happen with females as targets (Tangri et al., 1982). Harassment as a work condition refers to the "sexualization" of the workplace, and involves making sexuality very salient; this type of harassment arises most often in sex-segregated jobs (Coles, 1986). Sexual harassment has been considered a women's issue simply because the vast majority of victims are women. In a survey of sexual harassment complaints filed with a state agency, Terpstra and Cook (1985) found that, while all groups of women were represented, the "typical" victim was between the ages of 25 and 35, single, and college-educated; these data coincide with the results of other surveys (Coles, 1986). Sexual harassment occurs in all types of jobs, and is not limited to male or female dominated jobs (Reskin & Hartmann, 1986). Despite popular beliefs, several studies have found that the most common source of harassment is coworkers,

Gender In Organizations

32 1

although supervisors do make up a sizable percentage of perpetrators (Coles, 1986; %gri et al., 1982). Male coworkers and supervisors have been found to direct nonverbal and verbal dominance behaviors toward female employees, both of a sexual and nonsexual nature (Radecki & Jennings, 1980). Organizations may benefit from raising awareness among employees about these interactions. Because sexual harassment is often ambiguous, a fair amount of research has been done to identify important variables in the perception and interpretation of these behaviors. Several factors have been found to influence these processes, perhaps the most important of which is gender (Jensen & Gutek, 1982; Powell, 1986; Tangri et al., 1982). Men tend to perceive less harassment than women, and define fewer behaviors as objectionable. In addition, men blame the victim more often than women do. Three additional aspects of harassing situations affect how they are perceived and labeled. One influential variable is the gender of the harasser; female perpetrators are seen as relatively nonharassing (Pryor, 1985). Role inconsistent behavior is perceived to be more harassing than behavior that is not in conflict with roles; for example, sexual behaviors performed by a supervisor are seen as more harassing than the same behaviors performed by a coworker. Second, personal aspects of the situation influence the perception of harassment (Cohen & Gutek, 1985). Harassment is much less likely to be perceived when the behaviors occur between individuals who have had a personal relationship or friendship. Finally, the outcome of the incident in question affects how it is interpretated; negative outcomes are much more likely to lead to decisions that sexual harassment has occurred than positive outcomes (Konrad & Gutek, 1986). Women experience more negative outcomes than men; they are several times more likely to lose their jobs, be forced to quit, or be transferred, among other outcomes. Sexual harassment can lead to an array of consequences for both the individual and the organization, In addition to various job outcomes, victims of harassment may experience negative physical and emotional effects, such as anger, humiliation, fear, and helplessness (Coles, 1986). Many victims also engage in self-blame, though this behavior appears more common among women expressing traditional gender-role orientation (Jensen & Gutek, 1982). Some individuals report negative effects on their performance and attendance (Tangri et al., 1982). However, in a survey of federal employees, only a very small percentage of women submitted to the harasser (about 8%), while almost 25% of male victims said they acquiesced. Of those who complied, about one fifth reported that their situation worsened as a result of it. Most victims do not take formal action

322

Kelley and Streeter

against the harasser, for a number of reasons including fear of retaliation or embarrassment (Tangri et al., 1982). Sexual harassment may have the effect of barring individuals from certain jobs or firms, simply because the situation is unbearable. For the organization, the costs of sexual harassment have the potential to be high. Court cases can be expensive, especially when decided in favor of the plaintiff (Terpstra & Cook, 1985). Some cases have lead to the closure of the firm involved (Coles, 1986). The costs of increased stress and absenteeism, as well as decreased effectiveness and productivity, are more difficult to measure, but are surely not trivial.

Theoretical Explanations Explanations for sexual harassment parallel those provided for other gender inequalities, and represent individual, structural, and societal levels. Individual models interpret sexual harassment as the result of attraction between people, which the harasser has continued despite the fact that the victim has not reciprocated (Tangri et al., 1982). Arguments of this sort have been used to restrict women from some jobs; their presence in the workplace is assumed to invite sexual harassment, because men do not resist the assumed temptation (Reskin & Hartmann, 1986). StructuraUorganizational level explanations posit that harassment occurs because of differences in organizational power, or because of other ways in which the situation is structured (Tangri et al., 1982). Men typically occupy most of the powerful positions in organizations, and harassers abuse that power to obtain sexual gratification. Models involving gender composition of the workplace also fall into this category; some evidence indicates that a skewed gender ratio leads to more incidence of sexual harassment (Gutek & Morasch, 1982). Konrad and Gutek (1986) report that men in gender-integrated workplaces perceive as much harassment as women. Societal/cultural explanations propose that sexual harassment is an extension of male dominance within society. According to this model, men are playing the roles into which they are socialized (l’angri et al., 1982). This would explain the finding of Fain and Anderton (1987) of more harassment in female-dominated jobs than other jobs; even though a minority i n number, males still have dominant status. However, it may also be argued that males have hierarchical power, supporting structural models. Research comparing the three levels of explanations have found that no single theory is adequate to account for all findings concerning sexual harassment; rather some combination of the models is

Gender In Organizations

323

required (Fain & Anderton, 1987; Tangri et a]., 1982). Some harassment may involve romantic interest, and harassment by coworkers and subordinates indicates that structural power is not necessary for the behaviors to occur. Additionally, the fact that men are sometimes the victims suggests that societal and cultural beliefs about women alone cannot account for all findings about harassment. Organizations have an interest in attempting to counteract these behaviors. The message to employees should be that harassment will not be tolerated, through company policies and the quick and proper handling of grievances. Perhaps the most important control mechanism is education; some organizations provide training to sensitize employees about the issue (Reskin & Hartmann, 1986). Terpstra and Cook (1985)view training for supervisory and management personnel as necessary, because they can function as harassers or as keys to solutions which may defuse problems with sexual harassment.

CONCLUSION Interest in gender issues in organizations will continue well into the future. The need exists for information to allow individuals, organizations, and societies to make adequate, intelligent, and necessary decisions with beneficial impact on worker's lives. Much research needs to be done in the areas of work-related health, child care, and work-family interaction. This sensitivity to gender issues may lead to unexpected effects; some research has suggests that preferential selection on the basis of gender leads to lower evaluations by subordinates and detrimental effects on selected women's self-perceptions of ability and performance (Heilman, Simon, & Repper, 1987; Jacobson & Koch, 1977). However, the method of selection is less important than performance; successful performance leads to positive evaluations, irrespective of selection method (Jacobson & Koch, 1977; Seifert & Miller, 1988). Sex discrimination can be decreased by the presence of personal, effective, job-relevant information (Hodgins & Kalin, 1985; Powell & Butterfield, 1982). Organizations need able, competent people of both genders, and can seek to eliminate reliance on gender-role stereotypes. Similarly, by integrating the workplace organizations can try to eliminate gender-based inequalities. Equalization of the gender composition of occupations and firms may help decrease discrimination, sexual harassment, and inequalities in pay, status, and opportunity. This review has demonstrated that individual, structural, and societalcultural factors affect the roles of gender in organizations.

Kelley and Streeter

324

REFERENCES Adelmann, P.K. (1989). Occupational gender mix and men's experience of the work role. Social Behavior, 4, 225-236. Aigner, D., & Cain, G. (1977). Statistical theories of discrimination in labor markets. Industrial and Labor Relations Review, 30, 175-187. Arch, E.C. & Cummins, D.E. (1989). Structured and unstructured exposure to computers: Sex differences in attitude and use among college students. Sex Roles, 20, 245-254. Arvey, R.D. (1979). Unfair discrimination in the employment interview: Legal and psychological aspects. Psychological Bulletin, 86, 736-765. Arvey, R. D. (1986). Sex bias in job evaluation procedures. Personnel Psychology, 39, 315-355. Arvey, R.D., Miller, H.E., Gould, R., & Burch, P. (1987). Interview validity for selecting sales clerks. Personnel Psychology, 40, 1-12. Astin, H . S . (1984). The meaning of work in women's lives: A sociological model of career choice and work behavior. The Counseling Psychologist, 12, 117-126. Baber, K.M., & Monaghan, P. (1988). College women's career and motherhood expectations: New options, old dilemmas. Sex Roles, 19, 189-203. Barrett, G.V., Alexander, R . A . , Amesgart, M.N.., & Doverspike, P. (1986). Frequently encountered problems in the application of regression analysis to the investigation of sex discrimination in salaries. Public Personnel Management, 15(2), 143-157. Baruch, G.K., Biener, L., & Barnett, R.C. (1987). Women and gender research on work and family stress. American Psychologist, 42, 130-136. Becker, G. S . (1957). The economics of discrimination. Chicago: University of Chicago Press. Bem, S.L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162. BenTsvi-Mayer, S . , Hertz-Lazarowitz, R., & Safin, M.P. (1989). Teachers'selections of boys and girls as prominent pupils. Sex Roles, 21, 231-246. Betz, N.E. (1989). Implications of the null environment hypothesis for women's career development and for counseling psychology. The Counseling Psychologist, 17, 136-144. Beuhring, T. (1989). Incorporating employee values in job evaluation. Journal of Social Issues, 45(4), 169-190.

Gender In Organizations

325

Bhatnagar, D. (1988). Professional women in organizations: New paradigms for research and action. Sex Roles, 18, 343-355. Bielby, W.T., & Baron, J.N. (1986). Men and women at work: Sex segregation and statistical discrimination. American Journal of Sociology, 91, 759-799. Bielby, D.D., & Bielby, W.T. (1988). She works hard for the money: Household responsibilities and the allocation of work effort. American Journal of Sociology, 93, 1031-1059. Binning, J.F., Goldstein, M.A., Garcia, M.F., & Scattaregia, J.H. (1988). Effects of preinterview impressions on questioning strategies in same and opposite-sex employment interviews. Journal of Applied Psychology, 73, 30-37. Blau, F.D. (1977). Equal pay in the office. Lexington, MA: Lexington Books. Blau, ED., & Jusenius, C.L. (1976). Ecomonists' approaches to sex segregation in the labor market: An appraisal. In M. Blaxall & B. Reagan (Eds.), %men and the workplace: The implications of occupational segregatio (pp . 18 1- 199). Blechman, E.A. & Brownell, K.D. (Eds.) (1988). Handbook of behavioral medicinefor women. New York: Pergamon Press. Brett, J.M., Dornbush, B.B. & Yogev, S. (1988, August). Restructuring work f o r family: How dual-earner couples with children manage. Paper presented at the meeting of the American Psychological Association, Atlanta, GA. Bridges, J.S. (1988). Sex differences in occupational performance expectations. Psychology of %men Quarterly, 12, 75-90. Bridges, J.S. (1989). Sex differences in occupational values. Sex Roles, 20, 205-21I . Brown, S.M. (1979). Male versus female leaders: A comparison of empirical studies. Sex Roles, 5 , 595-611. Bureau of Labor Statistics (1990). Labor market indicators. Monthly Labor Review, 113(3), 82. Burley, K.A., & Gutek, B.A. (1988, August). Work and family involvement and work-family conflict among psychologists. Paper presented at the meeting of the American Psychological Association, Atlanta, GA. Buttner, E.H, & Rosen, B. (1987). The effects of labor shortages on starting salaries for sex-typed jobs. Sex Roles, 17, 59-71. Cann, A., & Siegfried, W.D. (1987). Sex stereotypes and the leadership role. Sex Roles, 17, 401-408. Cann, A., & Siegfried, W.D. (1990). Gender stereotypes and dimensions of effective leader behavior. Sex Roles, 23, 413420.

326

k2lley and Streeter

Cann, A., & Siegfried, W.D. & Pearce, L. (1981). Forced attention to specific applicant qualifications: Impact on physical attractiveness and sex of applicant biases. Personnel Psychology, 34, 65-75. Cash, T.F., Gillen, B., & Burns, D. (1977). Sexism and beautyism in personnel consultant decision making. Journal of Applied PsyCholO~,62, 301-310. Chusmir, L.H. & Mills, J. (1989). Gender differences in conflict resolution styles of managers: At work and at home. Sex Roles, 20, 149-163. Cohen, A.G., & Gutek, B.A. (1985). Dimensions of perceptions of social-sexual behavior in a work setting. Sex Roles, 13, 317327. Cohen, S.L., & Bunker, K.A. (1975). Subtle effects of sex role stereotypes on recruiters' hiring decisions. Journal of Applied Psychology, 60, 566-572. Coles, F.S. (1986). Forced to quit: Sexual harassment complaints and agency response. Sex Roles, 14, 81-95. Cooper, E.A., & Barret, G.V. (1984). Equal pay and gender implications of court cases for personnel practice. Academy of Management Review, 9, 84-89. Crocker, J., & McGraw, K.M. (1984). What's good for the goose is not good for the gander. American Behavioral Scientist, 27, 357-369. Crohan, S.E., Antonucci, T.C., Adelman, P.K. & Coleman, L.M. (1989). Job characteristics and well-being at midlife. Psychology of Ubrnen Quarterly, 13, 223-235. Deaux, K., (1976). Sex: A perspective on the attribution process. InJ. Harvey, W. Ickes, & R. Kidd (Eds.), New directions in attribution research, Vol. I, Hillsdale, NJ: Lawrence Erlbaum. Deaux, K., & Emswiller, T. (1974). Explanations of successful performance on sex-linked tasks: What is skill for the male is luck for the female. Journal of Personality and Social Psychology, 29, 80-85. Dipboye, R.L., Arvey, R.D., & Terpstra, D.E. (1977). Sex and physical attractiveness of raters and applicants as determinants of resume evaluations. Journal of Applied Psychology, 6 2 , 288-294. Dobbins, B.H., & Platz, S.J. (1986). Sex differences in leadership: How real are they? Academy of Management Review, 1 1 , 118127.

Gender In Organizations

327

Doerfler, M.C.. & Kammer. P.P. (1986). Workaholism. sex. and sex role stereotyping among female professionals. Sex Roles, 14, 551-560. Douthitt, R.A. (1989). The division of labor within the home: Have gender roles changed? Sex Roles, 20, 693-704. Dreher, G.F., Dougherty, T. W., & Whitely, W. (1989). Influence tactics and salary attainment: A gender-specific analysis. Sex Roles, 20, 535-550. Dubbert, P.M. (1988). Women's health issues: Adherence to health care recommendations. The Health Psychologist, 10(1), 4-5. Dubno, P. (1985). Attitudes toward women executives: A longitudinal approach. Academy of Management Journal, 28, 235239. Eagly, A. (1987). Reporting sex differences. American Psychologist, 42. 756-757. EccleK J.S. (1987). Gender roles and women's achievement-related decisions. Psychology of Wmen Quarterly, 11, 135-172. Elliott, A.G.P. (1981). Sex and decision making in the selection interview: A real-life study. Journal of Occupational Psychology, 54, 265-273. Epstein, C.F. (1981). Ubrnen in law. New York: Basic Books. Fain, T.C., & Anderton, D.L. (1987). Sexual harassment: Organizational context and diffuse status. Sex Roles, 17, 291-311. Farnsworth, C.H. (1988, July 4). A difference in societies that can give U.S. an edge in talks. New York Times, p. 8. Farr, K.A. (1988). Dominance bonding through the Good Old Boys Sociability Group. Sex Roles, 18, 259-277. Finkbeiner, A.K. (1987). Demographics or marketforces? Mosaic, 18(1), 10-17. Fitzpatrick, J.L., & Silverman, T. (1989). Women's selection of careers in engineering: Do tradi tional-nontraditional differences still exist? Journal of Mcational Behavior, 34, 266-278. Fleishman, E.A., Harris, E.F., & Burtt, H.E. (1955). Leadership and supervision in industry. Columbus: Ohio State University Press. Frank, E.J. (1988). Business students' perceptions of women in management. Sex Roles, 19, 107-118. Freeman, J. (1979). How to discriminate against women without really trying. In J. Freeman (Ed.), Women: A feminist perspective (2nd ed., pp. 194-208). Palo Alto, CA: Mayfield. Fullerton, H.N, Jr. (1987). Labor force projections: 1986 to 2000. Monthly Labor Review, 110(9), 19-29.

328

Kelley and Streeter

Galinsky, E. (1988, August). The impact of supervisors' attitudes and company culture on worklfamily adjustment. Paper presented at the meeting of the American Psychological Association, Atlanta, GA. Gardner, D.G. & Discenza, R. (1988). Sex effects in evaluating applicant qualifications: A reexamination. Sex Roles, 18, 297-308. Garland, H., Hale, K.F., & Burnson, M. (1982). Attributions for the success and failure of female managers: A replication and extension. Psychology of Wmen Quarterly, 7(2), 155-162. Gilbert, L.A., Dancer, L.S., Rossman, K . M . , & Thorn, B.L. (1991). Assessing perceptions of occupational-family integration. Sex Roles, 24, (1/2), 107-120. Gilmore, D.C., Beehr, T.A., & Love, K.G. (1986). Effects of applicant sex, applicant physical attractiveness, type of rater and type of job on interview decisions. Journal of Occupational Psychology, 59, 103-109, Goktepe, J.R., & Schneier, C.E. (1989). Role of sex, gender roles, and attraction in predicting emergent leaders. Journal of Applied Psychology, 74, 165-167. Gomez-Mejia, L.R. (1983). Sex differences during occupational socialization. Academy of Management Journal, 26(3), 492499. Graves, L.M., & Powell, G.N. (1988). An investigation of sex discrimination in recruiters' evaluations of actual applicants. Journal of Applied Psychology, 73, 20-29. Greenberger, E., Goldberg, W.A., Crawford, T.J. & Granger, J. (1988). Beliefs about the consequences of maternal employment for children. Psychology of Wmen Quarterly, 12(1), 3560. Greig, J.J., Orazem, P.F., & Mattila, J.P. (1989). Measurement error in comparable worth pay analysis: Causes, consequences, and corrections. Journal of Social Issues, 45(4), 135152. Gutek, B.A., & Burley, K. (1988, August). Relocation, family, and the bottom line: Results from the Division 35 survey. Paper presented at the meeting of the American psychological Association. Atlanta, GA. Gutek, B.A., & Morasch, B. (1982). Sex-ratios, sex-role spillover, and sexual harassment of women at work. Journal of Social Issues, 38, 55-74.

Gender In Organizations

329

Hackett, G.! Esposito, D., & O'Halloran, M.S. (1989). The relationship of role model influences to the career salience and educational and career plans of college women. Journal of Wcational Behavior, 35, 164-180. Hall, F.S., & Hall, D.T. (1976). Effects of job incumbents' race and sex on evaluations of managerial performance. Academy of Management Journal, 19, 476-48 1. Harris, C.T.& Earle, J.R. (1986). Gender and work values: Survey findings from a working-class sample. Sex Roles, 15, 487-494. Haw, M.A. (1982). Women, work and stress: A review and agenda for the future. Journal of Health and Social Behavior, 23, 132-144. Hayghe, H.V. (1990). Family members in the work force. Monthly Labor Review, 113(3), 14-19. Hearn, J . , Sheppard, D.L., Tancred-Sheriff, P., & Burrell, G. (Eds.) (1990). The sexuality of organization. Beverly Hills, CA: Sage. Hegtvedt, K.A. (1989). Fairness conceptualizations and comparable worth. Journal of Social Issues, #5(4), 81-98. Heilman, M.E., Block, C.J., Martell, R.F., & Simon, M.C. (1989). Has anything changed? Current characterizations of men, women, and managers. Journal of Applied Psychology, 74, 935-942. Heilman, M.E., & Martell, R.F. (1986). Exposure to successful women: Antidote to sex discrimination in applicant screening decisions? Organizational Behavior and Human Decision Processes, 37, 376-390. Heilman, M.E., Simon, M.C., & Repper, D.P. (1987). Intentionally favored, unintentionally harmed? Impact of sex-based preferential selection on self-perceptions and self-evaluations. Journal of Applied Psychology, 72, 62-68. Heilman, M.E.? & Stopeck, M. H. (1985). Being attractive, advantage or disadvantage? Performance-based evaluations and recommended personnel actions as a function of appearance, sex, and job type. Organizational Behavior and Human Decision Processes, 35, 202-215. Heimovics, R.D.,& Herman, R.D. (1988). Gender and the attribution of chief executive responsibility for successful or unsuccessful organizational outcomes. Sex Roles, 18, 623-636. Herz, D.E. (1988). Employment characteristics of older women, 1987. Monthly Labor Review, 111(9), 3-12.

330

Kelley and Streeter

Hodgins, D.C., & Kalin, R. (1985). Reducing sex bias in judgments of occupational suitability by the provision of sex typed personality information. Canadian Journal of Behavioral Science, 17, 346-358. Holland, J.L. (1973). Making vocational choices: A theory of careers. Englewood Cliffs, NJ: Prentice-Hall. Hulbert, K. (1988, August). Having it all: The impossible dream for young professional women. Paper presented at the meeting of the American Psychological Association, Atlanta, GA. Hunt, D.M., & Michael, C. (1983). Mentorship: A career training and development tool. Academy of Management Review, 8 , 475-485. Izraeli, D.N. (1983). Sex effects or structural effects: An empirical test of Kanter's theory of proportions. Social Forces, 62, 153165. Izraeli, D.N., & Izraeli, D. (1985). Sex effects in evaluating leaders: A replication study. Journal of Applied Psychology, 70, 540-546. Jackson, L.A. (1989). Relative deprivation and the gender wage gap. Journal of Social Issues, 45(4), 117-134. Jacobson, M.B., & Koch, W. (1977). Women as leaders: Performance evaluation as a function of method of leader selection. Organizational Behavior and Human Perjiormance, 20, 149157. Jagacinski, C.M. (1987). Androgyny in a male-dominated field: The relationship of sex-typed traits to performance and satisfaction in engineering. Sex Roles, 17, 529-547. Jensen, I.W., & Gutek, B.A. (1982). Attributions and assignment of responsibility in sexual harassment. Journal of Social Issues, 38, 121-136. Jick, T.D., & Mitz, L.F. (1985). Sex differences in work stress. Academy of Management Review, 10, 408-420. Jones, B.H., & McNamara, K. (1991). Attitudes toward women and their work roles: Effects of intrinsic and extrinsic religious orientations. Sex Roles, 24(1/2), 21-30. Kamerman, S.B. (1985). Child care services: An issue for gender equity and women's solidarity. Child W?lfare,LXZV, 259-271. Kanter, R.M. (1977). Men and women of the corporation. New York: Basic Books. Kapalka, G.M., & Lachenmeyer, J.R. (1988). Sex-role flexibility, locus of control, and occupational status. Sex Roles, 19, 417427.

Gender In Organizations

33 1

Kavanagh, M.J,, & Halpern, M. (1977). The impact of job level and sex differences on the relationship between life and job satisfaction. Academy of Management Journal, 20, 66-73. Keys, D.E. (1985). Gender, sex role, and career decision making of certified management accountants. Sex Roles, 13, 33-47. Konrad, A.M., & Gutek, B.A. (1986). Impact of work experiences on attitudes toward sexual harassment. Administrative Science Quarterly, 31, 422-438. Larwood, L.,. Szwajkowski, E., & Rose, S . (1988). Sex and race discrimination resulting from manager-client relationships: Applying the rational bias theory of managerial discrimination. Sex Roles, 18, 9-29. Levant, R.F., Slatterly, S.C., & Loiselle, J.E. (1987). Fathers' involvement in housework and child care with school-aged daughters. Family Relations, 36, 152-157. Lever, J. (1978). Sex differences in the complexity of children's play and games. American Sociological Review, 43, 471-483. Lewittes, H.J., & Bem, S.L. (1983). Training women to be more assertive in mixed-sex task-oriented discussions. Sex Roles, 9, 58 1-596. Liden, R.C. (1985). Female perceptions of female and male managerial behavior. Sex Roles, 12, 421-432. Lipman-Blumen, J. (1988). Connective leadership: A female perspective f o r an interdependent world. Paper presented at the meeting of the American Psychological Association, Atlanta, GA. Long, B.C. (1989). Sex-role orientation, coping strategies, and selfefficacy of women in traditional and nontraditional occupations. Psychology of Wmen Quarterly, 13, 307-324. Lont, C.M. (1990). The roles assigned to females and males in nonmusic radio programming. Sex Roles, 22(9/ lo), 661-668. Major, B. (1989). Gender differencies in comparisons and entitlement: Implications for comparable worth. Journal of Social Issues, 45(4), 99-116. Martocchio, J.J., & O'Leary, A . M . (1989). Sex differences in occupational stress: A meta-analytic review. Journal of Applied Psychology, 74, 495-501. Mathison, D.C. (1986). Sex differences in the perception of assertiveness among female managers. Journal of Social Psycholog ~ 126, , 599-606. Matthews, K.A., & Rodin, J. (1989). Women's changing work roles. American Psychologist, 44, 1389-1393. McGregor, D. (1967). The professional manager. New York: McGraw-Hill.

332

Kelley and Streeter

Mednick, M.T. (1987, July 16). Women in power are no better than men. New York Times, p. A-26. Miller, D.A., & Routh, D.K. (1985). Where women win: Supervisors of school psychologists prefer female job candidates. Professional Psychology, 16, 42-49. Miura, I.T. (1987). The relationship of computer self-efficacy expectations to computer interest and course enrollment in college. Sex Roles, 16, 303-311. Moore, H.A., & Ollenburger, J.C. (1986). What sex is your parachute? W r k and Occupations, 13, 51 1-531. Morrison, A.M., White, R.R., & Van Velsor, E. (1987, August). Executive women: Substance plus style. Psychology Today, pp. 18-26. (a) Morrison, A.M., White, R.R., & Van Velsor, E. (1987). Breaking the glass ceiling: Can women reach the top of America's largest corporations? Reading, M S : Addison-Wesley. (b) Mount, M.K. & Ellis, R.A. (1989). Sources of bias in job evaluation: A review and critique of research. Journal of Social Issues, 45(4), 153-168. Noe, R.A. (1988). Women and mentoring: A review and research agenda. Academy of Management Review, 13, 65-78. Nyquist, L., Slivken, K.? Spence, J.T., & Helmreich, R.L. (1985). Household responsibilities in middle-class couples: The contribution of demographic and personality variables. Sex Roles, 12, 15-34. O'Carolan, T. (1987). Parenting time: Whose problem is it? Personnel Administrator, 32(8), 58-63. Olian, J.D., Schwab, D.P., & Haberfeld, Y. (1988). The impact of applicant gender compared to qualifications on hiring recommendations. Organizational Behavior and Human Decision Processes, 41, 180-195. Ott, E.M. (1989). Effects of the male-female ratio at work. Psychology of Wmen Quarterly, 13, 41-57. Overholser, G. (1987, June 16). Would women govern differently? New York Times, p. A15. Paludi, M.A., & Fankell-Hauser, J. (1986). An idiographic approach to the study of women's achievement striving. Psychology of Wmen Quarterly, 10, 89-100. Parasuraman, S., Greenhaus, J.H., Rabinowitz, S . , Bedeian, A.G., & Mossholder, K.W. (1989). Work and family variables as mediators of the relationship between wives' employment and husbands' well-being. Academy of Management Journal, 32, 185-201.

Gender In Organizations

333

Paternoster, L. (1990). Women take lead in '90s "megatrends. Business Review, February 12-18. Peters, L.H., O'Connor, E.J., Weekley, J., Pooyan, A,, Frank, B., & Erenkrantz, B. (1984). Sex bias and managerial evaluations: A replication and extension. Journal of Applied Psychology, 69, 349-352. Pietromonaco, P.R., Manis, J., Frohardt-Lane, K. (1986). Psychological consequences of mu1tiple social roles. Psychology of Wmen Quarterly, 10, 373-382. Powell, G.N. (1982). Sex, attributions, and leadership: A brief review. Psychological Reports, 51, 1171- 1174. Powell, G.N. (1986). Effects of sex role identity and sex on definitions of sexual harassment. Sex Roles, 14, 9-19. Powell, G.N. (1987). The effects of sex and gender on recruitment. Academy of Management Review, 12, 731-743. Powell, G.N. (1988). Mmen and men in management. Beverly Hills, CA: Sage. Powell, G.N., & Butterfield, D.A. (1982). Sex, attributions, and leadership: A brief review. Psychological Reports, 51, 11711174. Pryor, J.B. (1985). The lay person's understanding of sexual harassment. Sex Roles, 13, 273-286. Pulakos, E.D., White, L.A., Oppler, S . H . , & Borman, W.C. (1989). Examination of race and sex effects on performance ratings. Journal of Applied Psychology, 74, 770-780. Radecki, C., & Jennings (Walstedt), J. (1980). Sex as a status variable in work settings: Female and male reports of dominance behavior. Journal of Applied Social Psychology, 10, 7185. Ragins, B.R. (1989). Barriers to mentoring: The female manager's dilemma. Human Relations, 42, 1-22. Reed, B.B. (1983). Women leaders in small groups: Social-psychological perspectives and strategies. Social W r k with Groups, 6 , 35-42. Reskin, B.F., & Hartmann, H.I. (1986). Women's work, men's work: Sex segregation on the job. Washington, DC: National Academy Press. Robison, J.T. (1987). Women's health issues: Fact or fiction? m e Health Psychologist, 9(l), 3-4. Roos, P.A. (1983). Marriage and women's occupational attainment in cross-cultural perspective. American Sociological Review, I'

48, 852-864.

334

Kelley and Streeter

Roos, P.A., & Reskin, B.F. (1984). Institutional factors contributing to sex segregation in the workplace. In B.F. Reskin (Ed.), Sex segregation i n the workplace: Trends, explanations, remedies (pp. 235-260). Washington, DC: National Academy Press. Rosener, J. B. (1990). Ways women lead. Harvard Business Review, 68(6), 202. Ross, C.E., & Mirowsky, J. (1988). Child care and emotional adjustment to wives' employment. Journal of Health and Social Behavior, 29, 127-138. Rubin, L.J., & Borgers, S.B. (1990). Sexual harassment in universities during the 1980s. Sex Roles, 23, 397-412. Russell, J.E.A., Rush, M.C., & Herd, A.M. (1988). Sex Roles, 18, 279-287. RUSSO, N.F., & Denmark, EL. (1984). Women, psychology, and public policy. American Psychologist, 39, 1161- 1165. Sandberg, D.E., Ehrhardt, A.A., Mellins, C.A., Ince, S.E., & Meyer-Bahlburg, H.F.L. (1987). The influence of individual and family characteristics upon career aspirations of girls during childhood and adolescence. Sex Roles, 16, 649-668. Saunders, C.S., & Stead, B.A. (1986). Women's adoption of a business uniform: A content analysis of magazine advertisements. Sex Roles, 15, 197-205. Schein, V.E., Mueller, R., & Jacobson, C. (1989). The relationship between sex role stereotypes and requisite management characteristics among college students. Sex Roles, 20, 103-110. Scherbaum, C.J., & Shepherd, D.H. (1987). Dressing for success: Effects of color and layering on perceptions of women in business. Sex Roles, 16, 391-399. Seager, J., & Olson, A. (1986). Ubmen in the world: An international atlas. New York: Simon & Schuster, Inc. Seifert, C., & Miller, C.E. (1988). Subordinates' perceptions of leaders in task-performing dyads: Effects of sex of leader and subordinate, method of leader selection, and performance feedback. Sex Roles, 19, 13-28. Sekaran, U. (1986). Significant differences in quality of life factors and their correlates: A function of differences in career orientation or gender? Sex Roles, 14, 261-279. Sigelman, L., Milward, H.B., & Shepard, J.M. (1982). The salary differential between male and female administrators: Eaual pay for equal work? Academy of Management Journal, 25, 664-671. Sorrentino, C. (1990). The changing family in international perspective. Monthly Labor Review, 113(3), 41-56.

Gender In Organizations

335

Spence, J.T., & Helmreich, R.L. (1978). Masculinity and femininity: Their psychological dimensions, correlates, and antecedents. Austin: University of Texas Press. Spencer, B.A., & Taylor, G.S. (1988). Effects of facial attractiveness and gender on causal attributions of managerial performance. SexRoles, 19, 273-285. Spitze, G. (1984). The effect of family migration on wives' employment. Social Science Quarterly, 65,21-36. Stake, J.E., & Rogers, L.L. (1989). Job and home attitudes of undergraduate women and their mothers. Sex Roles, 20, 445463. Statham, A. (1987). The gender model revisited: Differences in the management styles of men and women. Sex Roles, 16,409429. Steel, L., Abeles, R.P., & Card, J.J. (1982). Sex differences in the patterning of adult roles as a determinant of sex differences in occupational values. Sex Roles, 8, 1009-1024. Stern, P.N. (Ed.). (1986). Wrnen, health, and culture. New York: Hemisphere Publishing. Stevens, G., & Gardner, S . (1987). But can she command a ship? Acceptance of women by peers at the Coast Guard Academy. Sex Roles, 16, 181-188. Stewart, L.P,, & Gudykunst, W.B. (1982). Differential factors influencing the hierarchical level and number of promotions of males and females within an organization. Academy of Management Journal, 25, 586-597. Summers, T.P. & DeCatiis, D. (1988). An investigation of sex differences in job satisfaction. Sex Roles, 18,679-689. 'Fait, M.? Padgett, M.Y., & Baldwin, T.T. (1989). Job and life satisfaction: A reevaluation of the strength of the relationship and gender effects as a function of the date of the study. Journal of Applied Psychology, 74, 502-507. 'Pdngri, S . S . , Burt, M.R., & Johnson, L.B. (1982). Sexual harassment at work: Three explanatory models. Journal of Social Issues, 38, 33-54. 'hylor, S.H. (1989). The case for comparable worth. Journal of Social Issues, 45(4), 23-38. Terpstra, D.E., & Baker, D.D. (1987). A hierarchy of sexual harassment. The Journal of Psychology, 121,599-605. Terpstra, D.E., & Cook, S.E. (1985). Complainant characteristics and reDorted behaviors and conseauences associated with formal 'sexual harassment charges. Personnel Psychology, 38, 559-574.

336

Kklley and Streeter

Tiedje, L.B., & Downey, G. (1988, August). Employed parents' depression: Effects of spouse support and child behavior. Paper presented at the meeting of the American Psychological Association, Atlanta, GA. Tsui, A.S., & Gutek, B.A. (1984). A role set analysis of gender differences in performance, affective relationships, and career success of industrial middle managers. Academy of Management Journal, 27, 619-635. Tucker, S. (1985). Careers of men and women MBAs. ubrk and Occupations, 12, 166-185. U.S. Department of Labor, Bureau of Labor Statistics (1990). Current labor statistics. Monthly Labor Review, 113(3). Vernon-Gerstenfeld, S . (1989). Serendipity? Are there gender differences in the adoption of computers? A case study. Sex Roles, 21, 161-173. Vollmer, F. (1986). Why do men have higher expectancy than women? Sex Roles, 14, 351-362. von Baeyer, C.L., Sherk, D.L., & Zanna, M.P. (1981). Impression management in the job interview: When the female applicant meets the male (chauvinist) interviewer. Personality and Social Psychology Bulletin, 7, 45-5 1. Voydanoff, P. (1988). Women, work and family. Psychology of Mbmen Quarterly, 12, 269-280. Wallston, B.S. Hoover-Dempsey, K.V., Brissie, J.S., & RozeeKoker, P. (1989). Gatekeeping transactions: Women's resource acquisition and mental health in the workplace. Psychology of Wmen Quarterly, 13, 205-222. Watson, C. (1988). When a woman is the boss: Dilemma in taking charge. Group and Organization Studies, 13, 163-181. Wittig, M.A. (1985). Sex-role norms and gender-related attainment values: Their role on attribution of success and failure. Sex Roles, 12, 1-13. Yammarino, F.J., & Dubinsky, A.J. (1988). Employee responses: Gender or job-related differences? Journal of Vocational Behavior, 32, 366-383. Yee, D.K., & Eccles, J.S. (1988). Parent perceptions and attributions for children's math achievement. Sex Roles, 19, 317333. Yoder, J.D., & Sinnett, L.M. (1985). Is it all in the numbers? A case study of tokenism. Psychology of mmen Quarterly, 9, 413-418.

Gender In Organizations

337

Yount, K.R. (1986). A theory of productive activity: The relationships among self-concept, gender, sex role stereotypes, and work-emergent traits. Psychology of Wmen Quarterly, 10,6388. Zedeck, S . , & Mosier, K . L . (1990). Work in the family and employing organization. American Psychologist, 45, 240-25 1. Zigler, E., & Muenchow, S. (1983). Infant day care and infantcare .leaves: A policy vacuum. American Psychologist, 38, 91-94.

This Page Intentionally Left Blank

Issues, Theory, and Research in Industrial1 Organizational Psychology - K . Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

339

CHAPTER 11 THE DYNAMICS OF ETHNIC DIVERSITY IN ORGANIZATIONS: TOWARD INTEGRATIVE MODELS Bernard0 M. Ferdman As civil rights legislation, increased immigration and other demographic shifts have recast the composition of the United States workforce to make it increasingly heterogeneous (Johnston & Packer, 1987; U.S. Dept. of Labor, 1988), many organizations are beginning to consider more intently the question of how best to incorporate and manage ethnic diversity in the workplace (Adler, 1991; Cox, 1991; Kilborn, 1990a, 1990b; Morrison & Von Glinow, 1990; Schmidt, 1988; Thomas, 1990; U.S. Dept. of Labor, 1988). In contending with the process of incorporating increasing numbers of people with diverse ethnic iden tities and cultures, organizations are also faced with dilemmas regarding the proper ways to handle such differences (Ferdman, 1988, 1990; Thomas, 1990). Much like ethnic integration in the larger society, integration in the workplace results in controversy in large part because of disagreements regarding the relevance and even existence of group-based differences as well as dissension regarding issues of collective versus individual rights (Ferdman, 1988, 1989b). Although controversy continues over the use and effects of affirmative action programs (e.g., Blanchard & Crosby, 1989; Crosby & Clayton, 1990; Ferdman, 1989a; Glasser, 1988; Glazer, 1988; Gold, 1990; Harvard Law Review, 1989; Kleiman & Faley, 1988; Levinger, 1987; Nacoste, 1989; Nalbadian, 1989; Schofield, 1986; Williams, 1990) geared towards overcoming previous racial inequities, there has been a shift among management specialists from an emphasis solely on issues of access towards building new, expanded perspectives (e.g., Brown, 1983; Business-Higher Education Forum, 1990; Jones, 1986; Katz, 1989; Katz & Miller, 1988; Pettigrew & Martin, 1987; Thomas, 1990) on

340

Ferdman

diversity in organizations in which affirmative action is only one of a larger set of concerns. These new approaches seek to take into account various aspects of the intergroup dynamics that might be expected in already heterogeneous work environments, including the experience of members of ethnic and cultural minority groups once they have already entered organizations. In line with some of these new approaches, many large corporations have begun to regard ethnic and cultural diversity as a feature that can bring positive benefits to the organization, and therefore to cultivate it actively (Bailey, 1989; Cox, 1991; Cox & Blake, 1991; Kilborn, 1990a; Maraniss, 1990; Thomas, 1990; Tully, 1990). Other organizations continue to focus on creating "color-blind" environments where there is no link between how people are treated and their ethnic backgrounds (e.g:, Kantrowitz, 1988). Dramatically increased numbers of consulting firms and institutions now offer courses, workshops or strategic interventions focused on helping people of diverse backgrounds to work together more effectively, as more and more public and private organizations seek out such services. The integration of the European Economic Community in 1992, increasing international economic interdependence and the globalization of multinational businesses have also contributed to current interest in finding ways of helping people with diverse backgrounds to work together effectively within the same organization. Similarly, the question of diversity and how best to address it has become a burning issue on many U.S. university campuses (e.g., Goode, 1989). Thus, ethnic diversity has become a topic of major concern in organizational life in the United States. These trends have paralleled a more general expansion of interest in the role of ethnicity in American life (Alba, 1988, 1990; Fishman, 1983, 1989; Marger, 1991; Simonson & Walker, 1988; Takaki, 1987). More and more groups identified on cultural and historical grounds claim a right to be recognized as distinct and legitimate entities within the broader United States society, and debates abound as to the proper role of ethnic differences in societal institutions (Ferdman, 1990; Glazer, 1983; Pettigrew, 1988; Triandis, 1988). While the primary emphasis in the past has tended to be on racial distinctions (e.g., Alderfer & Thomas, 1988; Cox & Nkomo, 1990), groups that were previously identified solely in such terms are now often identified using the concepts and language of ethnicity (Jones, 1988, 1990; Phinney, 1990; Sue, 1990). Thus, Black Americans have also become African Americans and their differences as well as similarities to Americans of European descent are noted; Hispanics are differentiated in terms of specific identities such as Puerto Rican, Mexican American, Cuban or Dominican; and

Ethnic Diversity

34 1

many White Americans identify themselves as being of Anglo, Irish, Italian, German or Polish American ethnicity.

ETHNICITY AND ORGANIZATIONAL PSYCHOLOGY This growth in ethnicity as an important phenomenon has not been matched by theoretical or research developments in organizational psychology (Alderfer & Thomas, 1988; Calis, 1986; Cox, 1990; Cox & Nkomo, 1990). Although an extensive literature relevant to the psychology of intergroup relations exists, it remains relatively dispersed and fragmented. For example, social psychologists who focus on social categorization processes have tended to work independently from communication scholars who consider the implications of cultural differences for interpersonal interactions. These groups have been mostly disconnected from personnel and organizational psychologists, who have tended to focus on issues of bias in selection or appraisal. More importantly, the psychological study of intergroup behavior has tended to concentrate on developing concepts that cut across group types and so has devoted little attention to illuminating the ways in which intergroup dynamics vary as a function of the basis for group differentiation (Ferdman, 1987b). For example, we might expect that explaining gender-related interactions will involve different notions and perspectives than explaining interactions based on race, occupational groups, or organizational affiliations. Yet social psychologists who study intergroup behavior have tended to use such social categories interchangeably (Ferdman, 1987b). Although the situation seems to be changing slowly (note for example, Jones, 1988; Phinney, 1990; Sue, 1990), psychology has not yet come to terms with the concepts of ethnicity and ethnic diversity and their relationship to important psychological constructs (Berry, 1986; Bond, 1987). This is as true for organizational psychology (Alderfer & Thomas, 1988) as it is for other areas. Nevertheless, if it is to help meet the emerging needs of organizations in the area of ethnic diversity and to continue to provide useful perspectives for illuminating organizational behavior, organizational psychology must produce and support theoretical frameworks suitable for understanding ethnic dynamics as they actually occur. It must consider and test hypotheses likely to lead to recommendations that will help organizations and their members become more effective. Doing so calls for considering the full range of complexity inherent in the phenomena of interest. Developing a more complete picture of ethnic diversity in organizations entails considering the nature of ethnicity in an intergroup context -- including both between- and

342

Ferdman

within-group variation in degree of identification and adherence to group-level cultural features -- together with the organizational context in which these dynamics are played out. Ethnicity and ethnic diversity present special issues that cut across levels of analysis and traditional disciplinary distinctions. Such boundary-crossing issues have not been sufficiently addressed within organizational psychology (Gioia & Pitre, 1990). For this reason, a focus on ethnicity as a unique phenomenon can lead not only to better action-relevant understanding of a pressing organizational issue but also to new perspectives on questions basic to organizational psychology. By considering the nature of ethnicity and ethnic diversity as they impact on people in organizations, psychologists can develop ways to address better the complexity of most organizational behavior. The primary goal of this chapter is to contribute to an emergent and integrated understanding of the role of ethnicity and the dynamics of ethnic diversity in organizations by reviewing and linking several current empirical and conceptual streams in the literature and elaborating on some of the problems and prospects these present for theoretical integration. In doing so, I hope to point the way for more complex models of behavior in the workplace that can incorporate a fuller range of human variation. In the rest of the chapter, I first explain what is meant by ethnicity and why it requires special attention. I then discuss current approaches useful for understanding the dynamics of ethnic diversity in organizations and discuss the problems these present for developing integrated models, as well as the prospects for doing so. The chapter closes with a discussion of additional factors I believe psychologists will have to consider as we construct more complete models of ethnic diversity in the workplace. Although the research and theory discussed here are primarily based on the situation in the United States, the hope is that those interested in ethnic diversity in other countries will find much that is useful. However, it is important to note that in spite of the emergence of diversity-related concerns in many other nations, most notably Canada, the Soviet Union, and the European Community, there are considerable differences among these societies that make it difficult to generalize immediately to all of them.

CONCEPTS OF ETHNICITY AND ETHNIC DIVERSITY Although psychologists are only recently beginning to pay close attention to ethnicity and its psychological ramifications (e.g., Berry, 1985, 1986; Ferdman, 1990; Phinney, 1990), other social

Ethnic Diversiry

343

scientists have been concerned with this variable for a long time. In particular, anthropologists, historians and sociologists have long debated the nature of ethnicity, of ethnic identification, and of their defining features (e.g., Alba, 1988, 1990; Barth, 1969; Bentley, 1981; Cohen, 1978; Despres, 1975, 1984; Fishman, 1983, 1989; Hirschman, 1983; Isajiw, 1974; Keyes, 1981, McCready, 1983; Petersen, 1982; Royce, 1982; Spickard, 1989; Yancey, Ericksen & Juliani, 1976).

Components of Ethnicity Most treatments of ethnicity define it in terms of both group boundaries and the bases and implications of those boundaries. Ethnicity is marked by consciousness of kind and therefore a sense of groupness among the members of an ethnic group; members of the group share a social identification and recognize themselves and/or are recognized by others as having something in common. Fishman (1989) describes ethnicity as marking the distinctions between us and them and between them and them and so highlights its phenomenological nature. According to Fishman (1989), ethnicity exists only to the extent that collectivities of people actually use it to organize themselves and categorize others and so perceive and experience it as a meaningful basis for classification and interpretation. An ethnic group is distinguished not only by a socially meaningful label, however, but also by its members' common ancestral heritage (Barth, 1969; Buriel, 1987) and by the resultant shared style (Royce, 1982). Scholars of ethnicity have jointly emphasized the boundaries implicit in the notion of ethnic groups and the associated behavior patterns. Thus, an ethnic group has associated with it both a label, demarcating it as a unique social category, as well as distinguishing cultural features -- patterns of behaviors, values and beliefs widely shared by its members. These are linked to a sense of shared ancestry and continuity with the past; ethnic distinctions and their associated cultural differences are based to a large degree on real or perceived historical depth (Fishman, 1989). People do not join ethnic groups; they are born into them, A group of individuals does not simply meet to create a new ethnic group; rather, it is the antecedent existence of the ethnic group that defines the connectedness among the individuals. This sense of having something in common -- of groupness -- is enhanced and maintained through the group's shared culture. Because ethnicity is a collective manifestation of the continuity of culture across generations (Fishman, 1989), it is a source of inter-

344

Ferdman

group differentiation that is profound in both experience and meaning, at both the collective and the individual levels. Heller (1987) provides a useful perspective on what constitutes culture. She points to the way in which the members of a given ethnic group, as they interact, construct shared lenses for interpreting their experiences and their environment. Usually, it is these collective understandings of the world and how it functions that are referred to as culture. However, Heller also emphasizes that culture goes beyond beliefs and values. Culture also includes the standard and typical behaviors that are shared by members of the group. According to this view, culture includes both specific behavioral characteristics typifying a group as well as the underlying views of social reality guiding those behaviors. Triandis (1972) has referred to the latter as a group's subjective culture, which is the group's typical or distinctive manner of understanding the social environment. Similarly, Jones (1983), basing himself on Van denBerghe (1977) and Kroeber and Kluckhohn (1952), defines culture as a configuration of beliefs, values, symbols, and behaviors that are historically-based and that provide a framework for behavior in the future.

Multilevel Perspectives on Ethnicity and Ethnic Diversity Ethnicity and ethnic diversity are particularly complex from a psychological point of view because they require attending simultaneously to individual, group and intergroup dimensions (Alderfer & Thomas, 1988; Ferdman, 1990; Hakuta, Ferdman & Diaz, 1987). Aspects of each of these dimensions are discussed below.

The Individual Level. At the individual level, ethnicity involves both cognitive and affective components. Because it can comprise a significant element of individual's social identity (Babad, Birnbaum & Benne, 1983; Ferdman, 1990; Thjfel gL Turner, 1986), ethnicity is an important basis for people's sense of who they are and who they are not. Individuals can vary in the extent to which they identify with an ethnic group and in how they feel about this identification (Berry, 1984; Edwards & Doucette, 1987; Ferdman, 1990; Herman, 1977; Keefe & Padilla, 1987) and this can change over time (Alderfer & Thomas, 1988). Each person also has his or her own cultural identity, an individual image of the cultural features -- the behaviors, beliefs, values and norms -- that are appropriate to members of the ethnic group(s) to which he or she belongs (Ferdman, 1990). This means that two people can identify equally strongly with a particular ethnicity, yet differ substantially i n what they perceive to be the

Ethnic Diversity

345

attributes most central to this group membership and in the value they attach to these features. Similarly, there is individual variation in views about the role of one's ethnic group in society, its relations with other groups and the value and features of these other groups. Finally, individuals will differ in the degree to which they actually exhibit the cultural features that characterize the group as a whole (e.g., Boekestijn, 1988; Ferdman & Hakuta, 1985). Because in a multi-ethnic society there are varying degrees and types of contact among members of different groups, individuals have many options regarding how to relate not only to other groups but also to their own. Ethnic group members will express their choices in part through the behaviors they demonstrate in different types of situations (e.g., Salamone & Swanson, 1979). Especially in the case of minority group members or immigrants, the extent to which an individual follows the group's typical cultural pattern may be an indication of the degree of that person's psychological assimilation or acculturation (Berry, 1986; Graves, 1967). Jones (1988) points out how such variation may also reflect minority group members' perceptions regarding the instrumentality' of particular behaviors in different contexts, such that an individual may behave in accordance with the group's cultural patterns in some situations but not in others. The distinction between the group and the individual levels is important, in part because a group's acculturation will not be shared to the same degree by every individual member of the group (Berry, 1983, 1986) and conversely, because some individual group members may acculturate more rapidly than the group-as-a-whole.

The Group Level. Adding to the complexity of ethnicity is the nature of ethnic diversity at the group level. At the group level, ethnic groups are characterized by both common features and within-group diversity. While members of ethnic groups generally demonstrate shared cultural features -- leading to between-group differences -- in a multi-ethnic environment a good deal of variation within groups will also be present. These intra-group differences may be due, for example, to historical sub-group differences (Zenner, 1988), to the acculturation of some individuals but not others, or to differences in experience with a host culture. Within a particular social context, such as an organizational setting, a newcomer belonging to an ethnic minority group may behave according to her own culture's norms for that setting, while an old-timer from the same ethnic group may have been socialized to the majority's ways of doing things. Over time, acculturation processes can affect both what the cultural features are at the group level (Taylor & McKirnan, 1984) and whether or not particular individuals demonstrate them in their behavior

346

Ferdman

(Berry, 1986). Thus, in a heterogeneous society, variation within the group can be considered to be an important aspect of ethnicity. Ethnic groups also vary in the features that are viewed as central or core to the group's culture. Elsewhere (Ferdman, 1990), I have described this as cultural identity at a group level, which involves a shared sense of the cultural features that help to define and to characterize the group as well as to distinguish it from other groups. Group cultural identity has to do both with the particular features of the ethnic group and with the significance that is attached to these features in a societal context. For one group, for example, language or dialect may be a distinctive and emblematic (Kochman, 1987) feature while another emphasizes its religious practices and a third, its views on family relationships. Such group-level cultural features are not static and can vary over time, as the group's contacts with the environment influence both these and the group's cultural identity. Groups can also vary in the overall importance that they give to collective cultural identity in the interpretation of individual behavior. While some ethnic groups emphasize the connection of the individual to the group, others tend to see these as very distinct (Markus & Kitayama, 1991; Triandis, 1989). The implication of this is that the construct of culture can be differentially salient across groups.

The Intergroup Level. Ethnicity and ethnic diversity also involve intergroup aspects. Ethnic differences in culture and history, ethnic stratification patterns, the stereotypes and beliefs groups hold about each other, and patterns of intergroup behavior all play a role in influencing specific ethnic groups, their individual members, and the patterns of ethnic diversity as a whole. Salamone and Swanson (1979), in discussing Barth's (1969) theory of ethnic groups, point out that groups develop a sense of being separate and distinct from each other, and thus of their ethnicity, when they occupy a mutual social space. Power and status differentials can be especially important in this regard. For example, minority group status in the sociological sense is based on where an ethnic group falls within a societal hierarchy of group power and prestige (Feagin, 1989; Marger, 1991). One implication of this is that the members of an ethnic group considered to be a sociological minority -- i.e., having less power than the dominant group and whose members are treated unequally and in an inferior manner -- will be more likely than the members of a dominant ethnic group to have salient ethnic identities and be thought about in group terms. In a multiethnic society, the minority group member typically will be identified in group terms, while members of the dominant group will be more likely to see

Ethnic Diversity

347

themselves and to be seen by others in individual terms or as not being part of any definite category (Deschamps, 1982; see also 'Itajfel, 1978). An ethnic group's status in society should also be related to the recognition and legitimacy given to its characteristic cultural patterns. Group-level cultural identity plays an important role in the nature and outcome of intergroup comparisons and thus in the way a group comes to evaluate itself (Ferdman, 1987b; Montero, 1987; Bjfel & Turner, 1986). When a group considers that its cultural features compare favorably with those of other groups, it should come to hold more positive images of itself. If, on the other hand, features central to the group's cultural identity are viewed negatively in the society, the group and its members will probably incorporate a negative component into their self-evaluation ('Ikjfel, 1978). Finally, also at the intergroup level, ethnic diversity tends to be accompanied by controversy over the role of ethnicity in society and over the proper relationship among society's culturally diverse groups. Debate over the various alternatives has persisted throughout U.S.history (Feagin, 1989; Gleason, 1982; Hirschman, 1983) and indeed human history (Fishman, 1989) and is characteristic of plural societies (Babad, Birnbaum & Benne, 1983; Berry, 1984, 1986; Pettigrew, 1988; Schermerhorn, 1970). Ultimately, the choice is a value-laden one. Assimilation perspectives emphasize the dysfunctionality of differences and the maintenance of the dominant culture, and so demand that subordinate groups acculturate. Melting pot views, also referred to as amalgamation theories, maintain that the ideal society takes something from each of its component ethnic groups to create a new culture ultimately shared by all. In contrast, the pluralist position prizes diversity and so holds that it is preferable for the various ethnic groups in a society to co-exist in a kind of "vegetable soup'' (Babad, Birnbaum & Benne, 1983) such that each group maintains its own culture to the extent and in the ways that its members wish to do so. In part, these perspectives can be linked to divergent attitudes regarding the value of ethnicity itself. Fishman (1989) traces two such contrasting ideas historically: In the negative view ethnicity is considered to be disruptive, irrational and peripheral; in the opposite, positive view ethnicity is considered to be a healing and creative force as well as a source of individual and collective joy. In considering ethnic dynamics in organizations, therefore, we must take into account a number of variables that in the past have tended not to be considered together. After discussing two major streams of relevant research and theory in the next section, I suggest

Ferdman

348

ways of linking them with each other and with additional factors that play a role in ethnic diversity.

CURRENT APPROACHES Traditionally, the dynamics of intergroup relations in organizations have been thought about in ways that do not directly address the multi-faceted nature of ethnicity and ethnic diversity. In most analyses of intergroup dynamics ethnicity is included as simply one of many types of group-based differentiations that can impact on organizational behavior. Other analyses address primarily the cultural differences brought about by ethnic diversity without considering factors such as status differences, individual variation in cultural identity, and the impact of the organizational context. In general, the study of intergroup relations has been characterized by a focus on the negative repercussions of differences. Two major streams that have been predominant in the literature are discussed here. While both are important in understanding the dynamics of ethnic diversity, each by itself is incomplete. The first approach, the older and more developed of the two, focuses on the boundaries between ethnic groups and on the effects of ethnic categorization and identification. A major example of this type of work has been the study of intergroup prejudice and discrimination. The second approach, which has recently become more prominent, is concerned with intergroup differences in cultural features and with the resultant interpersonal barriers or conflicts. The study of intercultural communication is the foremost example of this perspective. I now turn to a more detailed exploration of each approach.

The Efsects of Group Membership: Categorization and Labeling Approaches This approach focuses on intergroup differentiation and its implications for attitudes and behavior. It highlights boundary formation, labeling, stereotyping and other individual and collective cognitive and affective processes involving social categories, but it does not usually make major distinctions between different types of social categories. The general assumption from this perspective has been that most types of intergroup differentiation tend to result in similar dynamics. In line with this view, researchers have looked at all kinds of groups in their empirical work, ranging from work groups composed of college students and created in the laboratory, to members of different teams, to ethnically- and racially-based groups.

Ethnic Diversity

349

The predominant work within this approach -- that stemming from the social cognition paradigm (Hamilton, 1981; Sherman, Judd, & Park, 1989) and from social identity theory (Thjfel, 1981, 1982; Thjfel & Turner, 1986) -- assigns a central role to the social categorization process. The question has generally been: how does thinking about another person in terms of his or her group membership influence the way in which information about that person is processed and evaluated? And how does this affect behavior towards that person and subsequent attitudes towards his or her group? (For reviews, see Brewer & Kramer, 1985; Hamilton, Sherman & Ruvolo, 1990; Hamilton & Trolier, 1986; Hewstone & Brown, 1986; Messick & Mackie, 1989; Sherman, Judd, & Park, 1989; njfel, 1981, 1982; Wilder, 1986a, 1986b.) In general, the social cognition approach posits that evocation of a social category will trigger a mental representation consisting of stereotypes and/or prototypes of group members. When this happens, biases may result because information is then processed primarily in terms of the mental category and its associated features and affect (Fiske & Pavelchak, 1986). Social identity theory emphasizes the broader symbolic aspects of group boundaries and their role in maintaining group members' self-esteem, in addition to the cognitive functions of stereotypes. Findings from empirical work in these traditions have been applied to assessing the implications of heterogeneous membership in organizations for intergroup relations and to describing the barriers that exist for creating diversity in the first place (e.g., Braddock & McPartland, 1987; Ilgen & Youtz, 1986; James & Khoo, in press; Pettigrew & Martin, 1987).

Implications of Categorization f o r Understanding and Changing Intergroup Relations. The categorization perspective has been heavily influenced by Sherif's (1966) view of intergroup behavior: "Whenever individuals belonging to one group interact, collectively or individually with another group or its members in terms oftheir group identification, we have an instance of intergroup behavior" (Sherif, 1966, p. 12; author's emphases). Research within this framework considers, first, the determinants of the salience of group boundaries, and second, the implications of such salience for ingroup and out-group perceptions and evaluations. Such research has been guided by an important assumption derived from Sherif's definition of intergroup behavior, It is that intergroup behavior can vary in degree, so that some interactions will be less influenced than others by the interactants' group identifications. This assumption has played a central role in investigations of intergroup contact. A great deal of social psychological research has

Ferdman

350

sought to identify the conditions under which encounters between members of different groups will result in positive interactions and perceptions, and in the reduction of prejudice and bias against the out-group (Ferdman, 1985; Hewstone & Brown, 1986; Katz & Thylor, 1988; Miller & Brewer, 1984). Various formulations of the contact hypothesis (e.g., Allport, 1954; Amir, 1976; Cook, 1984) generally suggest that the most favorable conditions are those that involve equal status for the participants, cooperation in the pursuit of superordinate goals, and institutional norms supporting equality. There is a rich body of literature that documents the profound effects of social categorization. We know that simply creating group boundaries or highlighting existing ones can have strong effects on the perceptions, evaluations, and judgments of both members of ingroups and out-groups. People see members of out-groups as more similar to one another (Allen & Wilder, 1975, 1979; Linville, Fischer, & Salovey, 1989; Linville, Salovey, & Fischer, 1986; Quattrone, 1986; Wilder, 1978a, 1978b, 1981), give them more negative evaluations (Brewer, 1979; Hewstone, Bond & Wan, 1983; Pettigrew, 1979; njfel, 198l), and think about them in more stereotyped terms (Howard & Rothbart, 1980; Rothbart, 1981; Rothbart & St. John, 1985) than in-group members. These findings have led researchers to recommend ways of improving intergroup relations that seek to reduce or weaken the importance or relevance of category boundaries. Worchel (1986), for example, in a discussion of why cooperation works to reduce intergroup conflict, concludes that "any activity that reduces the salience of group boundaries will have an ameliorating effect on intergroup relations" (Worchel, 1986, p. 299). Brewer and Miller (1984, 1988; Miller & Brewer, 1986) adopt this view when they distinguish among three levels of differentiation of out-group members, which they associate with depth of processing and quality of interaction. They refer to these as (1) category-based (2) differentiated, and (3) personalized responding. Moreover, they see personalized interactions, in which the intergroup boundary is virtually eliminated, as the most likely to lead to positive perceptions and interactions. To improve interethnic relations, Brewer and Miller would promote interactions in which ethnic group memberships are not at all evident. Similarly, social identity theory implies that reduction in the salience of group boundaries should lead in-group members to develo more positive perceptions and evaluations of out-group members. Attributional studies i n intergroup contexts (e.g., Duncan, 1976; Mann & Taylor, 1974; Sagar & Schofield, 1980; Stephan, 1977) have supported the notion that behavior is differentially ex-

P

Ethnic Diversity

35 1

plained and evaluated depending on the group memberships of the actor and the perceiver. These studies typically have found that people will make different attributions for the same behavior exhibited by an in-group and an out-group member (for reviews see Brewer & Kramer, 1985; Fletcher & Ward, 1987; Hewstone & Jaspars, 1982; Jaspars & Hewstone, 1982; Pettigrew, 1979; Pettigrew & Martin, 1987). The patterns of attributions are such that negative images of the out-group can be maintained. When desirable behaviors are exhibited, they will be ascribed to stable personality traits or dispositions if by in-group members, but to situational influences or atypical efforts if by out-group members. Conversely, when undesirable behaviors are displayed, out-group members will be seen to have behaved that way for internal reasons, and in-group members for external or situational reasons. Thus, this line of research also suggests that an increase in the salience of group boundaries will result in more negative evaluations of out-group members by the ingroup. Proposals that call for eliminating category boundaries, such as Brewer and Miller's (1984, 1988), are problematic when applied to ethnicity. These suggestions do not take sufficiently into account the importance to individuals of their ethnic identification. Brewer and Miller's approach implicitly assumes that categorizations are usually made more or less salient by the authorities in a situation, i.e., that they are externally imposed. In line with the same assumption, other researchers also have varied categorization by simply attaching labels to stimulus persons without accounting for their source. While some interethnic situations may fit this model, not all social identifications are self-evident. In reality, it is often the interactants in an interethnic encounter who choose whether to make their social identification salient. This is especially the case when ethnic identification is not readily apparent from visual, linguistic or other cues. Regardless of the position advocated by the authorities in a given context, people engaged in an interethnic encounter may be motivated to highlight rather than hide their identities. Under certain conditions this can be a useful strategy for maintaining a positive social identity (Tajfel, 1978; Tajfel & Turner, 1986; Taylor & McKirnan, 1984). Berry (1984, 1986; Sommerlad & Berry, 1970) has investigated what he calls acculturation attitudes, which express the value given by the individual to the maintenance of ethnic identity, and has found that both groups and individuals vary in this regard. In this spirit, and building on social identity theory, Hewstone and Brown (1986) take issue with Brewer and Miller's model of decategorization, and propose instead that positive intergroup relations

Ferdman

352

will be enhanced by mutual intergroup diflerentiation. Hewstone and Brown advocate that in order to improve intergroup relations, people should both see their own group positively and hold positive views of out-groups. Because Hewstone and Brown's view is based primarily on the role they see social identity playing in maintaining individuals' self-esteem, however, it is not tied explicitly to the interethnic situation nor to variation in the self-presentational choices made by individuals. Moreover, Hewstone and Brown (1986) do not consider the implications of group differences in the salience given to ethnicity for these choices or for their model as a whole. Their model nevertheless does point to the active role of individuals in choosing how they are categorized in social settings. Even though perspectives based on social identity theory, such as Hewstone and Brown's, diverge from social cognition approaches on the issue of how feasible it is to get rid of ethnic category boundaries altogether, they ultimately share the view that positive intergroup relations should be based on a de-emphasis of the boundaries, especially to the extent these are tied to differential and ethnocentric evaluations of the groups or of their members.

Organizational Applications. Conceptual frameworks derived from categorization and labeling approaches together with the related empirical findings have been used by a number of authors to explain various aspects of ethnic dynamics i n organizations. In general, such applications have tended to focus on the problems facing members of "marked" groups, such as Blacks or other minorities, and attempt to account for group-based differences in how people are treated in organizations. Pettigrew and Martin (1987) for example, describe the "triple jeopardy'' that African Americans face as they are recruited, enter and seek promotion in organizations. These barriers, which include negative stereotypes, being the only Black in a work group, and attributions of tokenism, have serious implications for the type of evaluations, feedback and support Blacks are likely to receive at work. Indeed, Pettigrew and Martin's (1987) analysis is bolstered by persistent evidence of negative experiences on the part of Black emplo ees and managers in largely White organizations (Alderfer, Alder er, Tucker & Tucker, 1980; Davis & Watson, 1982; Fernandez, 1975, 1982; Greenhaus, Parasuraman & Wormley, 1990; Ilgen & Youtz, 1986; Jones, 1986; Kraiger & Ford, 1985; Morrison & Von Glinow, 1990). Greenhaus, Parasuraman and Wormley (1990), for example, investigated the relationship of race to the organizational experiences, job performance evaluations, and career outcomes of managers in three organizations, and found that, consistent with

Y

Ethnic Diversity

353

Kraiger and Ford's (1985) meta-analysis, approximately 4 percent of the variance in job performance ratings (given predominantly by White managers in their study) could be explained by the ratees' race. On the average, Blacks tended to be systematically rated lower than Whites, and these effects were not mediated by differential organizational experiences. These researchers also found that Blacks, as compared to Whites, were rated by their supervisors as less promotable, were less satisfied with their careers, and had more often reached a career plateau. Research of this type has generally focused on the evaluation or the treatment of members of minority groups by members of the dominant group (e.g., Ilgen & Youtz, 1986) and so has not considered the range of possible interethnic interactions or ethnically-based dynamics. A major assumption in this kind of analysis has been that factors such as race should be irrelevant to treatment or outcomes; thus, any group differences become evidence of the presence of bias, prejudice or discrimination. Approaching the same issues from an ethnic diversity perspective would mean also considering the implications of variation in the number of groups and in their relative status. Similarly, the distinction between Black and White groups would be considered as only one type of ethnic differentiation that could be made, albeit a highly salient one. Alderfer's approach (e.g., Alderfer, 1986), termed intergroup theory and focused primarily at the group and intergroup levels of analysis, varies somewhat from the other approaches discussed so far in that it permits consideration of various types of intergroup patterns, not just minority-majority ones. It also recognizes minority status as being relative and dependent on which social system is being analyzed. For example, Alderfer develops the concept of embedded intergroup relations (Alderfer, 1986; Alderfer & Smith, 1982), which recognizes that any intergroup relationship is affected by the intergroup relationships present in the "supra-system" in which it is embedded. On the basis of his research on race relations in organizations (e.g., Alderfer, Alderfer, Tucker & Tucker, 1980; Alderfer, Tucker, Morgan & Drasgow, 1983) as well as work with other types of groups, Alderfer posits that intergroup relations within a given setting cannot be considered in isolation. For example, we could not explain or predict the racial dynamics in a given department of an organization that had been completely integrated at all levels until we recognize the importance of what is happening outside the department. Inside the department, there are Black and White managers and employees, with all roles containing members of both groups. There are no Blacks present in upper management, however, nor in other key departments of the organization. More-

354

Ferdman

over, even though the department head is a Black woman, she reports to a White junior vice-president, unlike the other, White, department heads, who all report to a White senior vice-president. Even though within the department race seems to be irrelevant to which jobs people have, that this is not the case at upper management levels should have significant implications for understanding the intergroup relations within the "integrated" department of the organization. Intergroup theory also would predict that identity group distinctions, such as ethnicity, will always be relevant in an organizational context. Thus, we cannot assume that they will simply be ignored; rather, we must take them into account in understanding and planning interactions. Nevertheless, Alderfer's approach can be considered part of a general categorization view in that he does not make specific distinctions on the basis of the type of group under consideration. As in social identity theory, the relevance of group distinctions is determined by their meaning in context.

Summary. In sum, the categorization approach focuses in a general way on the implications that labels and group memberships have for intergroup relations. Theory and research within this tradition point to the ways in which highlighting differences in identities can lead to negative intergroup behavior. Thus, the predominant recommendations for improving relations between members of different ethnic groups typically involve finding ways of de-emphasizing interethnic boundaries.

The Role of Group Diflerences: Intercultural Approaches The second major stream of research and theory important for understanding ethnic dynamics in organizations has focused on culture and on between-group cultural differences. This approach, characteristic of the study of intercultural communication, highlights the nature and implications of actual group-based cultural differences and their role in intergroup interactions. There is a vast literature documenting the ways in which ethnic cultures vary in terms of their members' behaviors, values, and beliefs, but there has been less work focusing on the organizational consequences of these differences (Adler, 1983; Boyacigiller & Adler, 199 1). Nevertheless, this area of investigation is a growing one, as it is clear that there are substantial implications of culture for the workplace.

Research on Cultural Diflerences. The five volumes of the Handbook of Cross-CuZturuZ Psychology (Triandis & Brislin, 1980) contain

Ethnic Diversity

355

many reviews of cross-cultural research in various areas of psychology showing cultural diversity on a number of psychological processes. Similarly, journals such as the Journal of Cross-Cultural Psychology and the International Journal of Intercultural Relations regularly publish empirical articles documenting the nature and implications of cultural differences. Kagitcibasi and Berry (1989) recently provided an overview of current research in cross-cultural psychology. One important domain of cultural difference has to do with values and attitudes. For example, Triandis (1975) documented differences in the subjective cultures of American Blacks and Whites. Miller (1984) found that Americans and Hindus differed substantially in how they determined moral responsibility and thus in how they attributed individual behavior. She was able to rule out cognitive, experiential or information differences as explanatory factors and thus provided evidence for the primary importance of shared cultural meanings. Laurent (1983) showed that managers in different countries had widely divergent views of what constitutes an appropriate response i n typical work situations and of the managerial role. Hofstede (1980; Hofstede & Bond, 1984), in an influential largescale international study of managers within the same multinational organization, found notable differences across countries on a number of work-related values. He discovered four major dimensions along which national value systems could be arrayed, terming these power distance, uncertainty avoidance, individualism, and masculinity. Typically, such differences in values are considered to underlie many of the observed differences in behavior (e.g., Stewart, 1976). Adler (1991), Bhagat and McQuaid (1982), Barrett and Bass (1976), Bass and Barrett (1981) and Tannenbaum (1980) provide extensive reviews of literature documenting cross-cultural differences in organizational behavior, including socialization, motivation to work, supervisory attitudes and leadership styles, job satisfaction, and job design. Cultures differ as well in the way their members behave in social interactions. Triandis (1977, 1989) reviewed cross-cultural research on interpersonal relations, and characterized the dimensions along which societies seem to differ (Triandis, 1987). Principal among these is the dimension of individualism/collectivism, along which he has documented ethnic differences in the United States (e.g., Marin & Triandis, 1985; Triandis, McCusker & Hui, 1990; Triandis, Marin, Hui, Lisansky & Ottati, 1984; Triandis, Marin, Lisansky & Betancourt, 1984). Hall (1959, 1966) described in detail the way in which the use of space and time vary as a function of culture. Hanna (1984) discusses the differences in non-verbal behav-

356

Ferdman

ior between American Blacks and Whites, and Poyatos (1983) concentrates his analysis of non-verbal communication on the implications of cultural differences. Similarly, Barnlund and Araki (1985) document differences between Americans and Japanese in how the giving of compliments is managed, and Ramsey and Birk (1983) as well as Barnlund (1975) discuss Japanese-North American differences in communication styles. DeFrank, Matteson, Schweiger and Ivancevich (1985) give evidence for differences in management styles and behaviors between U.S. and Japanese CEO's, and Doktor (1990) describes how culture influences the ways in which U.S. ,. Hong Kong Chinese, Japanese and Korean top executives organize and carry out their work. Hecht and Ribeau (1984; Collier, Ribeau & Hecht, 1986) found that Black, Hispanic and White students in the U.S.had divergent conceptions of what constitutes a satisfying conversation. TingToomey (1986) reports Black-White differences in the handling of communication conflicts. Erickson (1979) showed that Blacks and Whites handled pauses in speech differently, in such a way that more communication breakdowns occurred in intergroup than in intragroup conversations. Finally, Henderson and Argyle (1986) studied the ways in which British, Italian, Japanese and Hong Kong Chinese people varied in their perceptions of the rules of employment relationships. While the literature cited above is by no means exhaustive of research documenting differences in group-level cultural features, it suffices to highlight the ways in which culture is an important influence on individual values, beliefs and behaviors relevant to organizations. Because culture is a major aspect of ethnicity, it is reasonable to expect that it should play a significant role in the dynamics of ethnically diverse organizations. Most research documenting cultural variations has tended to do so, however, by comparing the features of homogeneous groups i n different places. There has been much less work focusing on the nature and impact of cultural differences among members of different ethnic groups working in the same context (Adler, 1983; Adler, Doktor & Redding, 1986; Bhagat & McQuaid, 1982; Cox, Lobe1 & McLeod, 1991; Shaw, 1990). Nevertheless, if we are to shed light on the dynamics of ethnic diversity it is these questions that must be investigated.

Research on Intercultural Contact. Although these issues have not been studied extensively in the context of ethnically diverse workplaces, there is a growing literature that explores the nature and dynamics of intercultural contacts. In general, however, this work tends to concentrate on broadly-defined differences and contacts

Ethnic Diversity

357

taking place across national boundaries. Brislin (1981), for exampIe, summarizes literature on face-to-face contacts between people of different cultural backgrounds, but focuses primarily on encounters involving individuals from different countries. Shaw (1990) presents a model that seeks to explain the interactions of expatriate managers and their host country subordinates. He bases his analysis on cognitive categorization perspectives, tracing the effects that cultural differences can have on how people i n organizations think about their own and others' behavior and thus on both managers' and employees' perceptions of and behavior towards each other. Ferraro (1990) discusses extensively the implications of culture for international business interactions, including the ways in which differences i n language, nonverbal communication behavior and values can affect people who work in more than one country. He does not, however, address within-country ethnic differences, and so makes generalizations about "U.S. culture" as a whole, as well as about the cultures of other countries. Similarly, Adler (1991), in discussing the multicultural workforce and multicultural teams, uses examples and cites research based primarily on between-country comparisons. Even though she briefly mentions domestic multiculturalism, her emphasis is on multinational organizations and international contacts. In these conceptualizations, group-level accounts of cultural features are the primary basis for understanding individual differences. In other words, the differences between people belonging to groups with varying cultures as well as the problems that arise in their interactions are generally traced to the group-level differences. Implications of Cultural Diversity f o r Understanding and Changing Intergroup Relations. As this work nevertheless demonstrates, a focus on culture suggests a view of intergroup contact that is quite different from that of the categorization perspective. Psychologists who concentrate on the intercultural aspects of interpersonal relations (e.g., Adler, 1991; Bochner, 1982; Brislin, 1981; Triandis, 1972) emphasize the role of cultural differences in leading to misunderstandings. In this perspective, it is ignoring the differences in values, beliefs and behaviors that are present in an intercultural situation that can lead to misperceptions and negative views of the outgroup, as well as to strained interactions. To promote positive relations, those who adopt this perspective recommend approaches such as attribution training (e.g., Albert, 1983, 1986; Albert & Adamopoulos, 1976; Triandis, 1984) that teach people of one culture to interpret the behavior of someone from another culture in the same way as would other members of that same culture. The goal of most intercultural training (e.g., Black & Mendenhall, 1990; Brislin, 1986;

358

Ferdman

Brislin, Cushner, Cheme & Yong, 1986, Brislin, Landis & Brandt, 1983; Gudykunst & Hammer, 1983; Hughes-Weiner, 1986) is to improve the process and outcome of interethnic encounters by sensitizing people to the differences between their own and other cultures, and to the ways in which these differences affect perceptions and interactions. From this perspective the problem is not that differences are noted; indeed this is a solution. Rather, it is when differences are negatively valued that is considered to cause obstacles in interactions. Thus, the emphasis is on helping people to learn about the actual differences that exist between groups and to induce them to avoid viewing them negatively. If a White Anglo-Saxon Protestant, for example, can learn to understand how a Hispanic might behave in a particular situation, then their interaction might be facilitated because the WASP would then be able to interpret and evaluate the Hispanic's behavior appropriately. To be sensitive to cultural differences and take them into account in formulating judgments of other people and in deciding how to behave in interactions with them, however, one must know that particular individuals are likely to show them. To know this, one must be acutely aware of the other person's ethnic identification and of its relevance to the current situation. Thus, an intercultural approach implicitly favors the recognition and highlighting of group boundaries. Ignoring ethnic group membership in an intercultural interaction is likely to lead to misunderstandings and negative perceptions, rather than to mutually satisfying interactions. An attributional model also suggests that if ethnic group membership is deemphasized or is non-obvious, perceivers in intercultural situations are likely to explain and evaluate behavior -- and in particular behavior considered to be deviant -- using in-group norms. Indeed, in highly decategorized situations, this tendency is likely to be heightened. In intercultural situations, however, precisely such "deviant" behaviors are more likely to occur. In organizational contexts, for example, majority group members should tend to behave appropriately to the extent that the organizational norms are similar to those of their own culture. Members of an ethnic minority group -- especially newcomers -- are more likely to deviate from these norms, and this deviation can often be traced to cultural differences (Brown, 1983; Davis & Watson, 1982; Ferdman & Cortes, 1991; Hofstede, 1980; Smith & 'kyeb, 1987). In an organizational context in which majority group members work with individual members of relatively unknown groups, a cultural difference may be indistinguishable, practically speaking, from a norm violation, unless the ethnic background of these minority co-workers is made salient. A related organ-

Ethnic Diversity

359

izational implication of considering cultural differences has to do with the criteria used by an organization for evaluating job performance. Because these are likely to be rooted in the dominant ethnic group's culture, highlighting ethnic and cultural differences can help to make evident the degree to which seemingly "poor" performances by members of ethnic minorities often may actually be based on adherence to different assumptions regarding the criteria for success.

Summary. In sum, then, the intercultural approach seeks to discover the ways in which members of different groups vary in values, beliefs and behavior and the consequences of these cultural differences for intergroup interactions. This perspective emphasizes culture as a major construct for understanding both individual behavior and interethnic relationships, and points to the ways in which highlighting and training people to handle cultural differences can lead to more positive intergroup interactions. Thus, the principal recommendations of this approach for improving relations between members of different ethnic groups typically involve finding ways of sensitizing people to interethnic dissimilarities. In contrast to categorization approaches, which suggest diminishing the salience of ethnic boundaries, the intercultural perspective suggests highlighting ethnic boundaries and understanding ethnic differences,

Problems and Prospects for Conceptual Integration Although both of t h e current approaches discussed here are clearly quite relevant for understanding the dynamics of ethnic diversity in organizations, each perspective presents important problems when applied to multi-ethnic contexts. These concerns are related to the complex nature of ethnicity discussed in an earlier section. In an ethnically heterogenous society, in which there is a great deal of within-group diversity as well as between-group differences, neither of these approaches suffices, especially if the goal is the facilitation of interpersonal relationships.

Categorization Approaches. When applied specifically to ethnic distinctions, a major shortcoming of the categorization approach, as mentioned earlier, is that it has not addressed individual, group and situational differences in the importance of ethnic identities and in the source and degree of their salience. Research in this stream has tended to compare, for example, how members of two groups, such as Blacks and Whites, may be differentially evaluated by either ingroup or out-group members. A less-examined question has been that of the effect of differences in the perception of members of a

360

Ferdman

single group as a function of the salience of the target's ethnic identity andlor its importance to the target. Exploration of this type of variable would address more directly the impact of within-group variations in degree of ethnic identification. Ferdman (1987a, 1989a), in an experiment touching on this issue, investigated the effect of type of ethnic categorization information on evaluations of a Hispanic manager's behavior by his Anglo peers, Anglo managers rated a Hispanic manager they had observed in a number of situations, after being presented with introductory information about him that included job-related details combined with individuating facts only, combined with facts highlighting ethnic identity only, or combined with both types of facts. Highlighting information about the Hispanic manager's ethnicity resulted in higher ratings when combined with individuating information, and did not lower ratings when presented alone, suggesting that the effects of boundary salience in interethnic situations are not necessarily negative as predicted by the social cognition and social identity approaches. Ferdman's study also suggests that the effects of ethnic boundaries may depend to a large extent on contextual factors, such as the meaning of ethnicity in an organization or in the context of a specific judgment task. For example, ethnicity may be seen as relevant for some types of comparisons and not others. The interpretations perceivers make regarding this should influence their ultimate judgments and evaluation^.^ As the intercultural approach makes abundantly evident, the categorization perspective generally ignores cultural and other between-group differences. Because of this, it assumes that any variation in response to members of different groups is indicative of some type of bias. Even though it may be true that individuals are in many ways unique and should be evaluated primarily on individual terms, this does not mean that there are not important group-level differences that extend to individuals. Culture is a prime example of this. Relatedly, recommendations based on the de-emphasis of ethnic categories ignore the value that these group differences carry for individuals. People who consider their ethnic and cultural identity as an important and central feature of their selves are unlikely to want to hide them in interethnic situations.

Intercultural Approaches. While intercultural approaches address some of the concerns raised by categorization views, they create other problems. The major one is the confounding of the individual and group levels. Earlier, it was pointed out that ethnic diversity implied within-group variation in the degree to which individuals fit with group-level characterizations. Because cultural features tend to

Ethnic Diversity

361

be assessed and defined at a group level, it is difficult to know the degree to which individuals in a multi-ethnic context are characteristic or typical members of the group. This may be the case especially when cultural generalizations developed in one type of context are applied to another (Ferdman & Cortes, 1991). For example, cultural patterns of behavior observed in the family environment may or may not be displayed in a work situation. The attitudes of Puerto Ricans in San Juan are not the same as those of Puerto Ricans on the mainland U.S. (e.g., Comas-Dim, 1989). Thus, caution is called for in assuming that, just because one understands aspects of the culture of a particular ethnic group, one can therefore clearly understand and more effectively interact with specific individual members of that group. While sometimes this may be a useful working assumption, it can be problematic when group-level generalizations are automatically and uncritically applied to individuals. Ferdman and Cortes (1991), for example, found that Hispanic managers working in a largely Anglo business displayed only some of the cultural features which other researchers had previously attributed to Hispanics as a group. Moreover, the ways in which collective cultural patterns were manifested varied greatly across individuals and situations, such that the connections between individual behavior and these group-level cultural features were not clearly evident if one examined only isolated behavioral incidents. The other side of the individual/group confound is that conclusions about group-level features are often based on comparisons between samples of convenience (Bhagat & McQuaid, 1982; Jones, 1990; Thomas, 1986), leaving unclear the extent to which they generalize to the ethnic categories as a whole and are based on cultural, rather than other, differences (Kagitcibasi & Berry, 1989). Researchers often compare samples of respondents in two or more countries and then assume that any variations found are due to cultural differences. Although some studies have used comparisons between factors to rule out other explanation^,^ most do not do this, relying simply on the presence of intergroup differences as sufficient evidence of the cultural source of the variation. In any case, because of both idiosyncratic and systematic within-group variations in degree of adherence to group-level cultural features, caution is called for in making generalizations about a group's culture from small or unique samples.

Prospects for Integration. Beyond the problems specific to each approach discussed above, there are a number of additional factors that further complicate the picture. Foremost among these is the multifaceted nature of individual identity, such that ethnicity is only

362

Ferdman

one of many aspects of people's sense of who they are. Because researchers from both streams work primarily at a group level as it applies to individuals, they usually look at the impact of one identity at a time. Nevertheless, within individuals, the multiple components of identity will function in conjunction or in interaction with each other (e.g., Babad, Birnbaum, & Benne, 1983; Taylor & Dub& 1986). What this means, for example, is that African American ethnicity will have different implications for Black women than for Black men (e.g., Reid, 1988; W, 1987). Older Chinese can be expected to differ from young Chinese, but these age differences will not be the same among Hispanics, Greeks, or Italians. It will be the composite of an individual's many identities that will most likely be the primary influence in interpersonal relationships and in individual behavior, values and attitudes. Thus, to better understand both the individual and the interpersonal aspects of ethnic dynamics, we need to develop a clearer sense of how ethnicity interacts with other aspects of individual identity, including, for example, work roles. The construct of individual cultural identity, as developed by Ferdman (1990), is one example of an attempt to begin to conceptualize the ways in which group-level cultural features take on psychological reality for individuals. The two approaches -- categorization and intercultural -- that were discussed lead to contradictory recommendations regarding how best to handle interethnic relations. While findings based on the categorization approach suggest the importance of minimizing ethnic boundaries, the intercultural perspective suggests highlighting them. Nevertheless, because both approaches consider significant parts of the picture, it is not possible to rule out one or the other. Each perspective addresses an important aspect of ethnicity and so involves a certain degree of appropriateness and utility. People care about their group memberships, and this leads to marked effects of categorization when explicit or implicit comparisons are made between ingroup and outgroup members. People also care about their culture, which serves as a guide to behavior and provides a way of perceiving the social environment, and this leads to rough spots when meetings occur across cultures. The best prospect for integrating these approaches probably lies in finding ways to consider simultaneously both the individual and the group levels and the relationships between the two. For example, while group-level cultural differences exist, we need to know the degree to which they fit particular individuals. Similarly, the significance of group labels can vary as a function of the importance of ethnicity for individual interactants in particular situations. Thus, attempts to rule out one or the other approach are likely to fail.

Ethnic Diversity

363

The better strategy will be to explore the conditions under which each gives useful guidance for explaining defined aspects of organizational ethnic dynamics.

TOWARD MORE INTEGRATED AND COMPLEX APPROACHES Ultimately, to devise more fully integrated approaches to ethnic dynamics in organizations we will need to go beyond the two perspectives discussed above to consider the complete range of complexity that ethnicity and ethnic diversity present. In this section, I discuss further the need for complexity and then mention a number of factors that should be addressed to begin to develop new ways of understanding ethnicity's impact on organizations.

The Need for Complexity In a recent paper on theory building in the study of organizations, Gioia and Pitre (1990) argue for the need to engage multiple paradigms as we seek to better comprehend organizational phenomena. In trying to understand ethnic dynamics in organizations, it seems especially fundamental to heed their call. Explorations of ethnic dynamics are directly related to issues of social change. Such explorations tend to call into question traditional perspectives on organizational behavior (Alderfer & Thomas, 1988) and often raise the issue of "whose point of view" is being represented and of who is defining social "reality" (e.g., CalPs, 1986). In this line of work, objectivity and subjectivity are very much in question (e.g., Alderfer, 1983, 1986; Alderfer & Thomas, 1988; Berg, 1984; CalAs, 1986; Cox, 1990; Merton, 1972; Rose, 1990). Furthermore, the study of ethnic relations in organizations raises issues regarding the larger societal dynamics within which they take place (Ferdman, 1989b, 1990; Fishman, 1989). For these reasons, it would seem that progress in this area will be most likely to occur within a multiparadigm framework. Thus, an important premise of this chapter is that there is not one correct way to understand the dynamics of ethnic diversity in organizations. If the goal is to permit the development of knowledge on which to base successful interventions, the most effective approach is likely be one that is complex and that uses a variety of lenses so as to create a more complete picture. In organizations, developing such a comprehensive and inclusive picture requires simultaneously taking different types of variables and different levels

364

Ferdman

of analysis into account. As the preceding discussion of current approaches suggests, focusing on only one aspect is less likely to be ultimately fruitful. To some extent, this is true of most phenomena in organizational behavior. Developing a thorough accounting for any type of organizational dynamic involves attending at once to different levels of analysis (Alderfer, 1986), different types of variables (e.g., Katz & Kahn, 1978; Levinson, 1972; Offerman & Gowing, 1990; Robbins, 1988; Schein, 1988), and different analytical perspectives (e.g., Morgan, 1986). In a recent special issue of the American Psychologist devoted to organizational psychology, Offerman and Gowing (1990) introduce the varied set of articles by pointing to the challenges that changes in the workplace and in the workforce generate for industrial and organizational psychologists. They suggest that an important feature of success for I/O psychologists will be the openness they have to interdisciplinary approaches to organizational problems. Ethnicity adds yet another dimension to the usual complexity inherent in studying organizations. Interethnic relations can complicate the picture further. As described earlier, the psychology of ethnicity is such that it incorporates individual-, group-, and intergroup-level aspects. Thus, developing a complete picture means paying attention simultaneously to variables at each of these levels, Below, a number of such factors are discussed. This listing is not meant to be exhaustive; rather, it is intented as an illustration of the type of stretching that will be required in the future,

Ethnic vs. Minority/Majority Perspectives. Typically, research rele vant to ethnic relations has tended to equate ethnicity and minority status. These two variables need to be unconfounded (Berry, 1985). An exclusive focus on minorities tends to ignore important cultural variations among groups with this status, and can confuse effects due to ethnic and cultural identity with those due to relative power and status. It also will not suffice to consider how members of dominant ethnic groups treat members of minority ethnic groups. Perspectives that look at the experience of members of different ethnic minorities in the context of organizations can help to shed more light on how differences are handled and how minority status interacts with ethnic and cultural factors. This type of approach will also permit consideration of intragroup diversity in adherence to group-level cultural features, experiences in adapting to organizational demands, and ways of coping with pressures to assimilate culturally. Many theorists and researchers, in considering the impact of ethnicity on organ-

Ethnic Diversity

365

izational processes, tend to implicitly place ethnic majority group members in charge and so describe the effects of majority perceptions on minority outcomes. More complete perspectives on ethnic dynamics in heterogenous organizations will consider the full range of possible role combinations.

Organizational Factors. There are a number of organizational processes and variables that can be considered in developing more fully integrated perspectives on ethnic dynamics. These include organizational socialization, organizational culture, and the organization's relationship to its environment on issues of ethnicity and ethnic diversity. Because individuals function in organizational roles, we might expect important interactions of these with ethnicity. As people are socialized into organizations, we can expect that they will differentially adapt into their roles as a function of the messages they receive about the value of their ethnic identity and culture. Organizations can vary substantially in this regard (e.g., Cox, 1991; Thomas, 1990). Through their structures or cultures, organizations convey clear social expectations regarding the role of ethnicity in the workplace. Some, through programs such as Affirmative Action, High Potential Advancement Programs, or diversity training, communicate that ethnic differences are to be valued and highlighted. Others implement procedures and structures that seek to ignore or occlude individual ethnicities (e.g., Schofield, 1986). Similarly, organizations can have varying relationships to their external environments on the issues of ethnicity and ethnic diversity. For example, a multinational organization that is subject to legal and societal pressures to diversify its workforce in the United States may engage in outreach programs, affirmative action and intercultural training in that country. In its operations elsewhere, it may not have the same concerns and so can function very differently vis-84s its efforts to consider multiple perspectives and include a variety of ethnicities among its personnel.

The Societal Context. As suggested above, the organization's environment is important to consider. In what kind of society does the organization operate and what are the prevalent attitudes regarding ethnic diversity? These can vary i n both time and place. As U.S. society evolves, for example, we can expect corresponding changes in the ethnic dynamics within organizations. Similarly, we would not expect the same ethnic dynamics to take place in Japan as i n France, given different degrees of societal ethnic diversity, different roles for ethnicity and different approaches for handling differences. Also, whether or not the organization functions in more than one

366

Ferdman

society is an important factor. Adler (1991) discusses the implications of increasing globalization on the degree to which organizations will need to more directly consider the impact of cultural and ethnic variations. Organizations that operate in many countries should, by their nature, tend to incorporate a larger degree of such variations.

The Particular Ethnic Composition. What the particular ethnic mix is in the organization together with its significance in the broader society are aIso important variables. We should expect different dynamics in an organization that is 50 percent Black and 50 percent Jewish than in one that is 50 percent Black and 50 percent Italian. The significance for an organization that 10 percent of its managers are Hispanic will be quite different in Miami, with its large proportion of Hispanics in the population, than in Minneapolis, where there are very few Hispanic residents. Similarly, the historical roles of ethnic group members versus their actual roles should also have an impact on an organization's ethnic dynamics. To the extent that major shifts have occurred in ethnic balance within an organization, this should bear on its intergroup relations. For example, the experience and perceptions of the various ethnic groups will most likely be different in a largely White organization that in the past had a large proportion of Black women in janitorial roles, but now has a sizable number of Black women managers than in an organization that Black women are now entering for the first time at all hierarchical levels,

W u e s Regarding Diversity. Individuals and organizations vary in the value they assign to ethnic and cultural diversity. In part, these differences are related to divergent perspectives and ideologies on what is fairness and how differences should be handled in the context of work (Ferdman, 1988). For example, an emphasis on the merit principle suggests that people should be evaluated solely on the basis of some objective criterion of success or achievement. Acceptance of cultural diversity, however, highlights the ways in which such criteria are developed in the context of specific cultures, thus implicitly valuing some ethnic groups over others. These values can be expected to play a role in how organizations address ethnic differences and in how individuals interact with their co-workers.

CONCLUSION Ethnicity and ethnic diversity present both challenges and opportunities for organizational psychology. As we seek to understand better and influence perhaps the dynamics of ethnically

367

Ethnic Diversity

diverse organizations, we will need to jointly consider and better integrate divergent and even contradictory analytic frameworks and perspectives based on different levels of analysis. Doing this successfully will require adopting paradoxical perspectives (e.g., Ferdman, 1989b; Smith & Berg, 1987), considering new and complex variables, and better reflecting the nature of ethnicity and ethnic diversity. In this chapter, I have argued that a focus on group boundaries or a focus on cultural differences is each insufficient by itself as a framework from which to understand organizational ethnic dynamics, because each approach ignores imporant facets of ethnicity and ethnic diversity. A more complete understanding of ethnic dynamics will be obtained when we can find ways to combine these perspectives as well as others only briefly described above. In attempting to integrate diverse theoretical and research paradigms, we must be careful, however, to remember that integration is not the same as homogenization. In the same way that ethnic diversity can result in pluralism, such that groups maintain their distinctiveness as they participate together in society's institutions, analytic approaches can be informed, modified, and expanded by contact with other views without losing their unique flavor and character. Nevertheless, it is likely that theory and research that are inclusive and broad will provide better and more useful reflections of the complexity of society's ethnic diversity.

NOTES 1. By instrumentality, Jones (1988) refers to the utility of the behavior in obtaining desired outcomes.

2. In contrast to Brewer and Miller's (1984) position, Tajfel and Turner's (1986) social identity theory, however, does not suggest that the salience of social categories such as ethnicity can be easily minimized to such an extent that they become invisible to the perceiver, althouth their impact on behavior in specific encounters may be reduced. In part, this is due to social identity theory's emphasis on the motivational, rather than the purely cognitive, foundations of in-group biases found in intergroup evaluations. According to social identity theory, people are motivated to enhance their self-esteem by evaluating their own group favorably in comparisons with other groups. In many situations, this will also lead people to highlight intergroup boundaries.

3. This type of contextual influence has been demonstrated in work by Gaertner and Dovidio (1986; Dovidio, Mann & Gaertner, 1989). who found that Whites often behaved differentially towards Blacks depending on their perceptions of how relevant race was believed to be in the situation.

Ferdman

368

4. Hofstede (1980), for example, showed that national culture accounted for the variations he found among managers in 40 different countries better than did position in the organization, gender, age or profession.

REFERENCES Adler, N. J. (1983). Cross-cultural management research: The ostrich and the trend. Academy of Management Review, 8 , 22632. Adler, N. J. (199 1). International dimensions of organizational behavior (2nd 4 . ) . Boston, MA: Kent. Adler, N. J., Doktor, R., & Redding, S. G. (1986). From the Atlantic to the Pacific century: Cross-cultural management reviewed. Journal of Management, 12, 295-318. Alba, R. D. (Ed.). (1988). Ethnicity and race in the U.S.A.: Toward the twenty-first century. New York: Routledge. Alba, R. D. (1990). Ethnic identity: Transformation of white Amenca. New Haven: Yale University Press. Albert, R. D. (1983). The intercultural sensitizer or culture assimilator: A cognitive approach. In D. Landis & R. W. Brislin (Eds.), Handbook of intercultural training. %l. 2: Issues in training methodology @p. 186-217). New York: Pergamon . Albert, R. D. (1986). Conceptual framework for the development and evaluation of cross-cultural orientation programs. Zntemational Journal of Intercultural Relations, 10, 197-213. Albert, R. D. & Adamopoulos, J. (1976). An attributional approach to culture learning: The culture assimilator. Topics in Culture Learning, 4, 53-60. Alderfer, C. P. (1983). Intergroup relations and organizations. In J. R. Hackman et al. (Eds.), Perspectives on behavior in organizations (2nd ed.) (pp. 408-16). New York: McGraw-Hill. (Reprinted from Organizational Dynamics, Summer 1980.) Alderfer, C. P. (1986). An intergroup perspective on group dynamics. In J. Lorsch (Ed.), Handbook of organizational behavior @p. 190-222). Englewood Cliffs, NJ: Prentice-Hall. Alderfer, C. P., Alderfer, C., Tucker, L., & Tucker, R. C. (1980). Diagnosing race relations in management. Journal of Applied Behavioral Science, 16, 135-66. Alderfer, C. P. & Smith, K. K. (1982). Studying intergroup relations embedded in organizations. Administrative Science Quarterly, 27, 35-65.

Ethnic Diversity

369

Alderfer, C. P. & Thomas, D. A. (1988). The significance of race and ethnicity for understanding organizational behavior. In C. L. Cooper & I. Robertson (Eds.), International review of industrial and organizational psychology, 1988 (pp. 1-41). Chichester, UK: Wiley & Sons. Alderfer, C. P., Tucker, R. C., Morgan, D. R.? & Drasgow, F. (1983). Black and white cognitions of changing race relations in management. Journal of Occupational Rehaviour, 4, 10536. Allen, V. L. &Wilder, D. A. (1975). Categorization, belief similarity and intergroup discrimination. Journal of Personality and Social Psychology, 32, 971-77. Allen, V. L. & Wilder, D. A. (1979). Group categorization and attribution of belief similarity. Small Group Behavior, 10, 7380. Allport, G. W. (1954). The nature ofprejudice. New York: Doubleday. Amir, Y. (1976). The role of intergroup contact in change of prejudice and ethnic relations. In P. A. Katz (Ed.), Towards the elimination of racism (pp. 245-308). New York: Pergamon. Babad, E. Y.,Birnbaum, M., & Benne, K. D. (1983). The social s e K Group influences on personal identity. Beverly Kills, CA: Sage. Bailey, J. (1989, Nov.). How to be different but equal: A new breed of consultant teaches companies to play fair with all their employees. Savvy Wman, pp. 47,50. Barnlund, D. C. (1975). Public and private self in Japan and the United States: Communicative styles in two cultures. Tokyo: Simul Press. Barnlund, D. C. & Araki, S . (1985). Intercultural encounters: The management of compliments by Japanese and Americans. Journal of Cross-Cultural Psychology, 16, 9-26. Barrett, G. V. & Bass, B. (1976). Cross-cultural issues in industrial and organizational psychology. In M. Dunnette (Ed.), Hundbook of industrial and organizational psychology. Chicago: Rand McNally. Barth, F. (Ed.). (1969). Ethnic groups and boundaries. Boston: Little, Brown. Bass, B. M. & Barrett, G. V. (1981). Cross-cultural issues in industrial and organizational psychology. In People, work, and organizations. Boston: Allyn & Bacon. Bentley, G. C. (1981). Ethnicity and nationality: A bibliographic guide. Seattle: University of Washington Press.

370

Ferdman

Berg, D. N. (1984). Objectivity and prejudice. American Behavioral Scientist, 27, 387-402. Berry, J. W. (1983). Acculturation: A comparative analysis of alternative forms. In R. Samuda & S . Woods (Ed.), Perspectives in immigrant and minority education. New York: University Press of America. Berry, J. W. (1984). Cultural relations in plural societies: Alternatives to segregation and their sociopsychological implications. In N. Miller & M. B. Brewer (Eds.), Groups in contact: The psychology of desegregation (pp. 11-27). Orlando, FL: Academic Press. Berry, J. W. (1985). Cultural psychology and ethnic psychology, a comparative analysis. In I. R. Lagunes & Y. H. Poortinga (Eds.) , From a different perspective: Studies of behavior across cultures (pp. 3-15). Lisse: Swets & Zeitlinger. Berry, J. W. (1986). Multiculturalism and psychology in plural societies. In L. H. Ekstrand (Ed.), Ethnic minorities and immigrants in a cross-cultural perspective (pp. 35-5 1). Berwyn: Swets North America. Bhagat, R. S. & McQuaid, S. J. (1982). Role of subjective culture in organizations: A review and directions for future research. Journal of Applied Psychology, 67, 653-85. Black, J. S. & Mendenhall, M. (1990). Cross-cultural training effectiveness: A review and a theoretical framework for future research. Academy of Management Review, 15, 113-36. Blanchard, F. A. & Crosby, F. J. (Eds.). (1989). Aflrmative action in perspective. New York: Springer-Verlag. Bochner, S. (1982). The social psychology of cross-cultural relations. In S. Bochner (Ed.), Cultures in contact: Studies in cross-cultural interaction @p. 3-44). Oxford: Pergamon Press. Boekestijn, C. (1988). Intercultural migration and the development of personal identity: The dilemma between identity maintenance and cultural adaptation. International Journal of Intercultural Relations, 12, 83-105. Bond, M. H. (Ed.). (1987). The cross-cultural challenge to social psychology. Newbury Park, CA: Sage. Boyacigiller, N. & Adler, N. J. (1991). The parochial dinosaur: Organizational science in a global context. Academy of Management Review, 16, 262-90. Braddock, J. H. & McPartland, J. M. (1987). How minorities continue to be excluded from equal employment opportunities: Research on labor market and institutional barriers. Journal of Social Issues, 43, 5-39.

Ethnic Diversity

37 1

Brewer, M. B. (1979). In-group bias in the minimal group situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307-24. Brewer, M . B. & Kramer, R. M. (1985). The psychology of intergroup attitudes and behavior. Annual Review of Psychology, 36, 219-43. Brewer, M. B. & Miller, N. (1984). Beyond the contact hypothesis: Theoretical perspectives on desegregation. In N. Miller & M. B. Brewer (Eds.), Groups in contact: The psychology of desegregation @p. 28 1-302). Orlando, FL: Academic Press. Brewer, M. B. & Miller, N. (1988). Contact and cooperation: When do they work? In P. A. Katz & D. Taylor (Eds.), Towards the elimination of racism: Profiles in controversy @p. 315-28). New York: Plenum Press. Brislin, R. W . (198 1). Cross-cultural encounters: Face-to-face interaction. New York: Pergamon. Brislin, R. W. (1986). A culture general assimilator: Preparation for various types of sojourns. International Journal of Intercultural Relations, 10, 215-34. Brislin, R. W., Cushner, K., Cherrie, C., & Yong, M. (1986). Intercultural interactions: A practical guide. Beverly Hills, CA: Sage. Brislin, R. W:, Landis, D., & Brandt, M. E. (1983). Conceptualization of intercultural training and behavior. In D. Landis & R. W. Brislin (Eds.), Handbook of intercultural training. W l . I : Issues in theory and design (pp. 1-35). New York: Pergamon. Brown, L. D. (1983). Managing conflict at organizational interfaces. Reading, MA: Addison-Wesley. Buriel, R. (1987). Ethnic labeling and identity among Mexican Americans. In J. S . Phinney & M. J. Rotheram (Eds.), Children's ethnic socialization: Pluralism and development (pp. 134-52). Newbury Park, CA: Sage. Business-Higher Education Forum. (1990). Three realities: Minority life in the United States. Washington, D.C.: American Council on Education. Calis, M. B. (1986, Aug.). Anlother silent voice: Organization theory and research on Hispanic populations. In T. Cox, Jr. (Chair), Research on race efsecrs: What have we learned and where do we go from here?. Symposium presented at the Academy of Management Annual Meeting, Washington, D. C. Cohen, R. (1978). Ethnicity: Problem and focus in anthropology. Annual Review of Anthropology, 7, 379-403.

372

Ferdman

Collier, M. J., Ribeau, S . A., & Hecht, M. L. (1986). Intracultural communication rules and outcomes within three domestic cultures. International Journal of Intercultural Relations, 10, 439-57. Comas-Dfaz, L. (1989). Puerto Rican women's cross-cultural transitions: Developmental and clinical implications. In C. T. Garcia Coll & M. L. Mattei (Eds.), The psychosocial development of Puerto Rican women (pp. 166-99). New York: Praeger. Cook, S . W. (1984). Cooperative interaction in multiethnic contexts. In N. Miller & M. B. Brewer (Eds.), Groups in contact; m e psychology of desegregation (pp. 155-85). Orlando, FL: Academic Press. Cox, T., Jr. (1990). Problems with organizational research on race and ethnicity issues, Journal of Applied Behavioral Science, 26, 5-23. Cox, T., Jr. (1991). The multicultural organization. Academy of Management Executive, 5(2), 34-47. Cox, T., Jr. & Blake, S . (1991, in press). Managing cultural diversity: Implications for organizational competitiveness. Academy of Management Executive. Cox, T., Jr. & Nkomo, S . M. (1990). Invisible men and women: A status report on race as a variable i n organization behavior research. Journal of Organizational Behavior, 11, 4 19-31. Crosby, F. & Clayton, S. (1990). Affirmative action and the issue of expectancies. Journal of Social Issues, 46(2), 61-79. Davis, G. & Watson, G. (1982). Black life in corporate America: Swimming in the mainstream. Garden City, NY Anchor Press. DeFrank, R. S . , Matteson, M. T., Schweiger, D. M., & Ivancevich, J. M. (1985). The impact of culture on the management practices of American and Japanese CEOs. Organizational Dynamics, 13(4), 62-76. Deschamps, J.-C. (1982). Social identity and relations of power between groups. In H. Tajfel (Ed.), Social identity and intergroup relations (pp. 85-98). Cambridge: Cambridge university Press. Despres, L. A. (Ed.). (1975). Ethnicity and resource competition in plural societies. The Hague: Mouton. Despres, L. A. (1984). Ethnicity: What data and theory portend for plural societies. In D. Maybury-Lewis (Ed.), The prospects for plural societies (pp 7-29). Washington DC: American Ethnological Society. Doktor, R. H . (1990). Asian and American CEOs: A comparative study. Organization Dynamics, 18(3), 46-56.

Ethnic Diversity

373

Dovidio, J. F:, Mann, J., & Gaertner, S. L. (1989). Resistance to affirmative action: The implications of aversive racism. In F. A. Blanchard & F. J. Crosby (Eds.), Affirmative action in perspective @p. 83- 102). New York: Springer-Verlag. Duncan, B. L. (1976). Differential social perception and attribution of intergroup violence: Testing the lower limits of stereotyping of blacks. Journal of Personality and Social Psychology, 34, 590-98. Edwards, J. & Doucette, L. (1987). Ethnic salience, identity and symbolic ethnicity. Canadian Ethnic Studies, 19, 52-62. Erickson, F. (1979). Talking down: Some cultural sources of miscommunication in interracial interviews. In A. Wolfgang (Ed.), Nonverbal behavior: Applications and cultural implications. New York: Academic Press. Feagin, J. R. (1989). Racial and ethnic relations (3rd ed.). Englewood Cliffs, NJ: Prentice-Hall. Ferdman, B. M. (1985). Interethnic and intercultural relations: Reexamining the contact hypothesis. Unpublished manuscript, Yale University, New Haven. Ferdman, B. M. (1987a). Evaluating people in intercultural interactions. Doctoral dissertation, Yale University, New Haven. Ferdman, B. M. (1987b, Aug.). Person perception in interethnic situations. Unpublished manuscript, State University of New York at Albany. Ferdman, B. M. (1988, Aug.). Values and fairness in the ethnically diverse workplace. In F. Crosby (Chair), Emancipation, justice and anrrnative action. Symposium conducted at the 2nd International Conference on Social Justice and Societal Problems, University of Leiden, Netherlands. Ferdman, B. M. (1989a). Affirmative action and the challenge of the color-blind perspective. In F. A. Blanchard & F. J. Crosby (Eds.), AfJirmative action in perspective (pp. 169-76). New York: Springer-Verlag. Ferdman, B. M. (1989b, May). The paradox of fairness in the multiethnic workplace: Toward alternative visions. Paper presented at the 15th Congress of the International Society for Intercultural Education, Training and Research, Boston. Ferdman, B. M. (1990). Literacy and cultural identity. Harvard Educational Review, 60, 181-204. Ferdman, B. M. & Cortes, A. C. (1991, in press). Culture and identity among Hispanic managers in an Anglo business. In S. B. Knouse, P. Rosenfeld, & A. Culbertson (Eds.), Hispanics and work. Newbury Park, CA: Sage.

374

Ferdman

Ferdman, B. M. & Hakuta, K. (1985, Aug.). Group and individual bilingualism in an ethnic minority. In K. Hakuta & €3. M. Ferdman (Chairs), Bilingualism: Social psychological reflections. Symposium presented at the American Psychological Association. Los Angeles. Fernandez, J. P. (1975). Black managers in white corporations. New York: Wiley. Fernandez, J. P. (1982). Racism and sexism in corporate life: Changing values in American business. Lexington, MA: Heath. Ferraro, G. P. (1990). The cultural dimension of international business. Englewood Cliffs, NJ: Prentice Hall. Fishman, J. (1983). The rise and fall of the 'ethnic revival' in the USA. Journal of Intercultural Studies, 4 (31, 5-46. Fishman, J. A. (1989). Language and ethnicity in minority sociolinguistic perspective. Philadelphia: Mu1tilingual Matters. Fiske, S . T. & Pavelchak, M. A. (1986). Category-based versus piecemeal-based affective responses: Developments i n schema-triggered affect. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (pp. 167-203). New York: Guilford Press. Fletcher, G. J. 0. & Ward, C. (1987). Attribution theory and processes: A cross-cultural perspective. In M. H. Bond (Ed.), The cross-cultural challenge to social psychology (pp. 230-24 1). Newbury Park, CA: Sage. Gioia, D. A. & Pitre, E. (1990). Multiparadigm perspectives on theory building. Academy of Management Review, 15, 584602. Glasser, I. (1988). Affirmative action and the legacy of racial injustice. In P. A. Katz & D. Taylor (Eds.), Towards the elimination of racism: Proj7les in controversy (pp. 341-57). New York: Plenum Press. Glazer, N. (1983). Ethnic dilemmas: 1964-1982. Cambridge, MA: Harvard University Press. Glazer, N. (1988). The future of preferential affirmative action. In P. A . Katz & D. Taylor (Eds.), Towards the elimination of racism: Profiles in controversy (pp. 329-39). New York: Plenum Press, Gleason, P. (1982). American identity and Americanization. In S . Thernstrom (Ed.), Concepts of ethnicity (Dimensions of ethnicity: A series of selections from the Harvard encyclopedia of American ethnic groups) (pp. 57- 143). Cambridge, MA: Belknap Press of Harvard University Press.

Ethnic Diversity

375

Gold, M. E. (1990). Setback for civil rights? A spring of discontent. ILR Report, 27(2), 3-18. Goode, S . (1989, Oct. 9). By the numbers: The politics of diversity at Berkeley. Insight, pp. 8-15. Graves, T. D. (1967). Psychological acculturation in a tri-ethnic community. Southwestern Journal of Anthropology, 23, 33750. Greenhaus, J. H., Parasuraman, S . , & Wormley, W. M. (1990). Effects of race on organizational experiences, job performance evaluations, and career outcomes. Academy of Management Journal, 33, 64-86. Gudykunst, W. B. & Hammer, M. R. (1983). Basic training design: Approaches to intercultural training. In D. Landis & R. Brislin (Eds.), Handbook of intercultural training. Ml. 1: Issues in theory and design (pp. 118-54). New York: Pergamon. Hakuta, K., Ferdman, B. M., & Diaz, R. M. (1987). Bilingualism and cognitive development: Three perspectives. In S . Rosenberg (Ed.), Advances in applied psycholinguistics. Vol. 2 : Reading, writing and language learning (pp. 284-3 19). New York: Cambridge University Press. Hall, E. T. (1959). The silent language. Garden City, NY Doubleday. Hall, E. T. (1966). The hidden dimension. Garden City, NY: Doubleday. Hamilton, D. L. (Ed.). (1981). Cognitive processes in stereotyping and intergroup relations. Hillsdale, NJ: Lawrence Erlbaum. Hamilton, D. L., Sherman, S . J., & Ruvolo, C. M. (1990). Stereotype-based expectancies: Effects on information processing and social behavior. Journal of Social Issues, 46(2), 35-60. Hamilton, D. L. & Trolier, T. K. (1986). Stereotypes and stereotyping: An overview of the cognitive approach. In J. F. Dovidio & S . L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 127-63). Orlando, FL: Academic Press. Hanna, J. L. (1984). BlacWwhite nonverbal differences, dance, and dissonance: Implications for desegregation. In A. Wolfgang (Ed.), Nonverbal behavior: Perspectives, applications, intercultural insights (pp. 373-409). Lewiston, NY C. J. Hogrefe, Inc. Harvard Law Review. (1989). Rethinking Weber: The business response to affirmative action. Harvard Law Review, 102, 65871. Hecht, M. L. & Ribeau, S . (1984). Ethnic communication: A comparative analysis of satisfying communication. Zntemationa1 Journal of Intercultural Relations, 8, 135-51.

376

Ferdman

Heller, M. (1987). The role of language in the formation of ethnic identity. In J. S. Phinney & M. J. Rotheram (Eds.), Children's ethnic socialization: Pluralism and development (pp. 180200). Newbury Park, CA: Sage. Henderson, M. & Argyle, M. (1986). The informal rules of working relationships. Journal of Occupational Behaviour, 7(4), 25975. Herman, S . N. (1977). Jewish identity: A social psychological perspective. Beverly Hills, CA: Sage. Hewstone, M., Bond, M. H., & Wan, K. C. (1983). Social facts and social attributions: The explanation of intergroup differences in Hong Kong. Social Cognition, 2, 142-57. Hewstone, M. & Brown, R. (1986). Contact is not enough: An intergroup perspective on the 'contact hypothesis. In M. Hewstone & R. Brown (Eds.), Contact and conflict in intergroup encounters @p. 1-44). Oxford: Basil Blackwell. Hewstone, M. & Jaspars, J. (1982). Intergroup relations and attribution processes. In H. Tajfel (Ed.), Social identity and intergroup relations (pp. 99- 133). Cambridge: Cambridge University Press. Hirschman, C. (1983). America's melting pot reconsidered. Annual Review of Sociology, 9, 397-423. Hofstede, G. (1980). Culture's consequences: International di@erences in work-related values. Beverly Hills, CA: Sage. Hofstede, G. & Bond, M. H. (1984). Hofstede's culture dimensions: An independent validation using Rokeach's value survey. Journal of Cross-Cultural Psychology, 15, 417-33. Howard, J. & Rothbart, M. (1980). Social categorization and memory for ingroup and outgroup behavior. Journal of Personality and Social Psychology, 38, 301-10. Hughes-Weiner, G. (1986). The "learning how to learn" approach to cross-cultural orientation. International Journal of Intercultural Relations, 10, 485-505. Ilgen, D. & Youtz, M. A. (1986). Factors affecting the evaluation and development of minorities in organizations. Research in personnel and human resources management, 4, 304-37. Isajiw, W. W. (1974). Definitions of ethnicity. Ethnicity, I , 111-24. James, K. & Khoo, G. (in press). Identity-related influences on the success of minority workers in primarily non-minority organizations. Hispanic Journal of the Behavioral Sciences. Jaspars, J. & Hewstone, M. (1982). Cross-cultural interaction, social attribution and inter-group relations. In S . Bochner (Ed.), Cultures in contact: Studies in cross-cultural interaction (pp. 127-56). Oxford: Pergamon.

Ethnic Diversity

377

Johnston, W. B. & Packer, A. H. (1987). ubrkforce 2000: ubrk and workers for the twenty-first century. Indianapolis, IN: Hudson Institute. Jones, E. W. , Jr. (1986). Black managers: The dream deferred. Harvard Business Review, 64(3), 84-93. Jones, J. M. (1983). The concept of race in social psychology: From color to culture. In L. Wheeler & P. Shaver (Eds.), Review of personality and social psychology (Vol. 4, pp. 11749). Beverly Hills, CA: Sage. Jones, J. M. (1988). Racism in black and white: A bicultural model of reaction and evolution. In P. A. Katz & D. A. Taylor (Eds.), Eliminating racism: Profiles in controversy (pp. 11735). New York: Plenum. Jones, J. M. (1990, Aug.). Psychological models of race: What have they been and what should they be? Invited master lecture at the 98th Annual Convention of the American Psychological Association, Boston. Kagitcibasi, C. & Berry, J. W. (1989). Cross-cultural psychology: Recent research and trends. Annual Review of Psychology, 40, 493-53 1. Kantrowitz, B. (1988, Mar. 7). A tenuous bond from 9 to 5. Newsweek, pp. 24-25. Katz, D. & Kahn, R. L. (1978). 77ze social psychology of organizaNew . York: Wiley. tiom (2nd 4.) Katz, J. H. (1989). The challenge of diversity. In C. Woolbright (Ed.), kluing diversity on campus: A multicultural approach @p. 1-21). Bloomington, IN: Association of College Unions International. Katz, J. H. & Miller, F. A. (1988). Between monoculturalism and multiculturalism: Traps awaiting the organization. OD Practitioner, 20(3), 1-5. Katz, P. A. & Taylor, D. A . (Eds.). (1988). Eliminating racism: ProJles in controversy. New York: Plenum. Keefe, S . E. & Padilla, A. M. (1987). Chicano ethnicity. Albuquerque: University of New Mexico Press. Keyes, C. F. (Ed.). (1981). Ethnic change. Seattle, WA: University of Washington Press. Kilborn, P. T. (1990a, Oct. 4). A company recasts itself to erase decades of bias (Affirmative action: How to make it work. A special report). New York Times, pp. A-1,D-21. Kilborn, P. T. (1990b, July 30). Labor dept. wants to take on job bias in the executive suite. New York Times, pp. A-1,A-10.

378

Ferdman

Kleiman, L. S . & Faley, R. H. (1988). Voluntary affirmative action and preferential treatment: Some legal and research implications. Personnel Psychology, 41, 481-96. Kochman, T. (1987). The ethnic component in Black language and culture. In J. S. Phinney & M. J. Rotheram (Eds.), Children's ethnic socialization: Pluralism and development @p. 2 19-38). Newbury Park, CA: Sage. Kraiger, K. & Ford, J. K. (1985). A meta-analysis of ratee race effects in performance ratings. Journal of Applied Psychology, 70, 56-65. Kroeber, A, L. & Kluckhohn, C. (1952). Culture: A critical review of concepts and definitions. Papers of the Peabody Museum of American Archaeology and Ethnology, Harvard University, 47(whole no. 1). Laurent, A. (1983). The cultural diversity of Western conceptions of management. International Studies of Management and Organization, 8(1/2), 75-96. Levinger, G. (Ed.). (1987). Black employment opportunities: Macro and micro perspectives. Journal of Social Issues, 43(whole no. 1). Levinson, H. (1972). Organizational diagnosis. Cambridge, MA: Harvard University Press. Linville, P. W., Fischer, G. W., & Salovey, P. (1989). Perceived distributions of the characteristics of in-group and out-group members: Empirical evidence and a computer simulation. Journal of Personality and Social Psychology, 57, 165-88. Linville, P. W., Salovey, P. N., & Fischer, G. W. (1986). Stereotypes and perceived distributions of social characteristics: An application to ingroup-outgroup perception. In J. F. Dovidio & S . L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 165-208). Orlando, FL: Academic Press. Mann, J. F. & nylor, D. M. (1974). Attribution of causality: Role of ethnicity and social class. Journal of Social Psychology, 94, 3-13. Maraniss, D. (1990, Mar. 8). Firm makes racial revolution from top down. Whington Post, pp. Al, A-22. Marger, M. N. (1991). Race and ethnic relations: American and global perspectives (2nd 4.).Belmont, CA: Wadsworth. Markus, H. R. & Kitayama, S . (1991). Culture and the self Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-53.

Ethnic Diversity

379

Marin, G. & Triandis, H. C. (1985). Allocentrism as an important characteristic of the behavior of Latin Americans and Hispanics. In R. Guerrero (Ed.), Cross-cultural and national studies in social psychology (pp. 85-104). North Holland: Elsevier. McCready, W. C. (Ed.). (1983). Culture, ethnicity and identity: Current issues in research. New York: Academic Press. Merton, R. K. (1972). Insiders and outsiders: A chapter in the sociology of knowledge. American Journal of Sociology, 78, 947. Messick, D. M. & Mackie, D. M. (1989). Intergroup relations. Annual Review of Psychology, 40, 45-81. Miller, J. G. (1984). Culture and the development of everyday social explanation. Journal of Personality and Social Psychology, 46, 961-78. Miller, N. & Brewer, M. B. (Eds.). (1984). Groups in contact: The psychology of desegregation. Orlando, FL: Academic Press. Miller, N. & Brewer, M. B. (1986). Categorization effects on ingroup and outgroup perception. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 209-30). Orlando, FL: Academic Press. Montero, M. (1987). A travCs del espejo: Una aproximacion tdrica a1 estudio de la conciencia social en AmCrica Latina. In M. Montero (Ed.), Psicologta polittca latinoamericana (pp. 163202). Caracas: Editorial Panapo. Morgan, G. (1986). Images of organization. Beverly Hills, CA: Sage. Morrison, A. M. & Von Glinow, M. A. (1990). Women and minorities in management. American Psychologist, 45, 200-208. Nacoste, R. W. (1989). Affirmative action and self evaluation. In F. A. Blanchard & F. J. Crosby (Eds.), Affirmative action in perspective (pp. 103-10). New York: Springer-Verlag. Nalbadian, J. (1989). The U.S. Supreme Court's "consensus" on affirmative action. Public Administration Review, 49, 38-45. Offerman, L. R. & Gowing, M. K. (1990). Organizations of the future: Changes and challenges. American Psychologist, 45, 95-108. Petersen, W. (1982). Concepts of ethnicity. In S . Thernstrom (Ed.), Concepts of ethnicity (Dimensions of ethnicity: A series of selections from the Harvard encyclopedia of American ethnic groups) (pp. 1-26). Cambridge, MA: Belknap Press of Harvard University Press. Pettigrew, T. (1979). The ultimate attribution error: Extending Allport's cognitive analysis of prejudice. Personality and Social Psychology Bulletin, 5 , 461-76.

380

Ferdman

Pettigrew, T. F. (1988). Integration and pluralism. In P. A. Katz & D. Thylor (Eds.), Towardr the elimination of racism: Profiles in controversy @p. 19-30). New York: Plenum Press. Pettigrew, T. F. & Martin, J. (1987). Shaping the organizational context for Black American inclusion. Journal of Social Issues, 43, 41-78. Phinney, J. S. (1990). Ethnic identity in adolescents and adults: Review of research. Psychological Bulletin, 108, 499-5 14. Poyatos, F. (1983). New perspectives in nonverbal communication:

Studies in cultural anthropology, social psychology, linguistics, literature and semiotics. Oxford: Pergamon Press.

Quattrone, G. A. (1986). On the perception of a group's variability. In S. Worchel & W. Austin (Eds.), Psychology of intergroup relations (2nd ed.) @p. 25-48). Chicago: Nelson-Hall. Ramsey, S. & Birk, J. (1983). Preparation of North Americans for interaction with Japanese: Considerations of language and communication style. In D. Landis & R. W. Brislin (Eds.), Handbook of intercultural training. Vol. 3: Area studies in intercultural training @p. 227-59). New York: Pergamon. Reid, P. T. (1988). Racism and sexism: Comparisons and conflicts. In P. A. Katz & D. Thylor (Eds.), Towards the elimination of racism: Profiles in controversy (pp. 203-2 1). New York: Plenum Press. Robbins, S. P. (1988). Essentials of organizational behavior (2nd ed.). Englewood Cliffs, NJ: Prentice Hall. Rose, P. I. (1990). i%ey and we: Racial and ethnic relations in the United States (4th ed.). New York: McGraw-Hill. Rothbart, M. (1981). Memory processes and social beliefs. In D. L. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup relations (pp. 145-81). Hillsdale, NJ: Lawrence Erlbaum. Rothbart, M. & St. John, 0. P. (1985). Social categorization and behavioral episodes: A cognitive analysis of the effects of intergroup contact. Journal of Social Issues, 41, 81-104. Royce, A. P. (1982). Ethnic identity: Strategies of diversity. Bloomington: Indiana University Press. Sagar, H. A. & Schofield, J. W. (1980). Racial and behavioral cues in black and white children's perceptions of ambiguously aggressive acts. Journal of Personality and Social Psychology, 39, 590-98. Salamone, F. A. & Swanson, C. H. (1979). Identity and ethnicity: Ethnic groups and interactions in a multi-ethnic society. Ethnic Groups, 2, 167-83.

Ethnic Diversity

381

Schein, E. H. (1988). Process consultation. Ml. 1: Its role in organization development (2nd ed.).Reading, MA: Addison-Wesley. Schermerhorn, R. A. (1970). Comparative ethnic relations: A framework f o r theory and research. New York: Random House. Schmidt, P. (1988, Oct. 16). Women and minorities: Is industry ready, New York Times. Schofield, J. W. (1986). Causes and consequences of the colorblind perspective. I n J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 231-53). Orlando, FL: Academic Press. Shaw, J. B. (1990). A cognitive categorization model for the study of intercultural management. Academy of Management Review, 15, 626-45. Sherif, M. (1966). In common predicament: Social psychology of intergroup conflict and cooperation. New York: Houghton Mifflin. Sherman, S. J., Judd, C. M., & Park, B. (1989). Social cognition. Annual Review of Psychology, 40, 281-326. Simonson, R. & Walker, S. (Eds.). (1988). Multicultural literacy: Opening the American mind. Saint Paul: Graywolf Press. Smith, K. K. & Berg, D. N. (1987). Paradoxes of group life:

Understanding conflict, paralysis, and movement in group dynamics. San Francisco: Jossey-Bass.

Smith, P. B. & Tayeb, M. (1987). Organizational structure and processes. In M. H. Bond (Ed.), m e cross-cultural challenge to social psychology @p. 153-64). Newbury Park, CA: Sage. Sommerlad, E. A. & Berry, J. W. (1970). The role of ethnic identification in distinguishing between attitudes towards assimilation and integration of a minority racial group. Human Relations, 23, 23-29. Spickard, P. R. (1989). Mixed blood: Intermarriage and ethnic identity in twentieth-century America. Madison, WI:University of Wisconsin. Stephan, W. (1977). Stereotyping: Role of ingroup-outgroup differences in causal attribution of behavior. Journal of Social Psychology, 101, 255-66. Stewart, E. C. (1976). American cultural pattern: A cross-cultural perspective. Washington DC: SIETAR/Intercultural Press. Sue, S. (1990, Aug.). Ethnicity and culture in psychological research andpructice. Invited master lecture at the 98th Annual Convention of the American Psychological Association, Boston.

382

Ferdmun

Tajfel, H. (1978). The social psychology of minorities. London: Minority Rights Group. njfel, H. (1981). Human groups and social categories. Cambridge: Cambridge University Press. njfel, H. (Ed.). (1982). Social identity and intergroup relations. Cambridge: Cambridge University Press. Tajfel, H. & Turner, J. C. (1986). The social identity theory of intergroup relations. In S. Worchel & W.Austin (Eds.), Psychology of intergroup relations (pp. 7-24). Chicago: Nelson-Hall. Thkaki, R. (Ed.). (1987). From difSerent shores: Perspectives on race and ethnicity in America. New York: Oxford University Press. Tannenbaum, A. S. (1980). Organizational psychology. In H. C. Triandis & R. W. Brislin (Eds.), Handbook of cross-cultural psychology. Ml. 5: Social psychology (pp. 281-334). Boston: Allyn & Bacon. Taylor, D. M. & Dub& L. (1986). Two faces of identity: The "I" and the "we." Journal of Social Issues, 42(2), 81-98. Taylor, D. M. & McKirnan, D. J. (1984). A five-stage model of intergroup relations. British Journal of Social Psychology, 23, 29 1-300. Thomas, D. R. (1986). Culture and ethnicity: Maintaining the distinction. Australian Journal of Psychology, 38, 371-80. Thomas, R. R., Jr. (1990). From affirmative action to affirming diversity. Harvard Business Review, 68(2), 107-17. Ting-Toomey, S. (1986). Conflict communication styles in black and white subjective cultures. In Y. Y. Kim (Ed.), Znterethnic communication: Current research (International and Intercultural Communication Annual, Vol. X) (pp. 75-88). Beverly Hills, CA: Sage. Triandis, H. C. (1972). The analysis of subjective culture. New York: Wiley. Triandis, H. C. (1975). Culture training, cognitive complexity, and interpersonal attitudes. In R. Brislin, S. Bochner, & W. Lonner (Eds.), Cross-cultural perspectives on learning (pp. 3977). Beverly Hills CA: Sage. Triandis, H. C. (1977). Cross-cultural social and personality psychology. Personality and Social Psychology Bulletin, 3, 14358. Triandis, H. C. (1984). A theoretical framework for the more efficient construction of culture assimilators. International Journal of Intercultural Relations, 8, 301-30.

Ethnic Diversity

383

Triandis, H. C. (1987). Cross-cultural contributions to theory in social psychology. In M. H. Bond (Ed.), The cross-cultural challenge to social psychology (pp. 122-40). Newbury Park, CA: Sage. Triandis, H. C. (1988). The future of pluralism revisited. In P. A. Katz & D. Taylor (Eds.), Towards the elimination of racism: Profiles in controversy @p. 3 1-50). New York: Plenum Press. Triandis, H. C. (1989). The self and social behavior in differing cultural contexts. Psychological Review, 96,506-20. Triandis, H. C. & Brislin, R. W. (Eds.). (1980). Handbook of crosscultural psychology. Boston: Allyn & Bacon. Triandis, H. C., Marin, G., Hui, C. H., Lisansky, J., & Ottati, V. (1984). Role perceptions of Hispanic young adults. Journal of Cross-Cultural Psychology, 15, 297-320. Triandis, H. C., Marin, G . , Lisansky, J., & Betancourt, H. (1984). Simpatia as a cultural script of Hispanics. Journal of Personalit y and Social Psychology, 47, 1363-75. Triandis, H. C.,. McCusker, C., & Hui, C. H. (1990). Multimethod probes of individualism and collectivism. Journal of Personality and Social Psychology, 59, 1006-20. Tully, S. (1990, May 21). The hunt for the global manager. Fortune, 122(1l), 140-44. U.S. Department of Labor. (1988). Opportunity 2000: Creative aflrmative action strategies for a changing workforce. Washington, D.C.: U.S. Government Printing Office. Van denBerghe, P. (1977). Race and racism. New York Wiley. Wilder, D. A . (1978a). Homogeneity of jurors: The majority's influence depends upon their perceived independence. Law and Human Behavior, 2, 363-76. Wilder, D. A. (1978b). Perceiving persons as a group: Effects on attributions of causality and beliefs. Social Psychology, I , 1323. Wilder, D. A. (1981). Perceiving persons as a group: Categorization and intergroup relations. In D. L. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup relations (pp. 213-57). Hillsdale, NJ: Lawrence Erlbaum. Wilder, D. A. (1986a). Cognitive factors affecting the success of intergroup contact. In S . Worchel & W. Austin (Eds.), Psychology of intergroup relations (2nd ed.) (pp. 49-66). Chicago: Nelson -Hal1. Wilder, D. A. (1986b). Social categorization: Implications for creation and reduction of intergroup bias. Advances in &penmental Social Psychology, 19, 29 1-355.

384

Ferdman

Williams, R. E. (1990). The Supreme Court's 1989 employment decisions -- civil rights and illusions of Armageddon. ILR Report, 27(2), 19-26. Worchel, S. (1986). The role of cooperation in reducing intergroup conflict. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (2nd ed.) (pp. 288-304). Chicago: NelsonHall. Yancey, W. L., Ericksen, E. P., & Juliani, R. N. (1976). Emergent ethnicity : A review and reformulation. American Sociological Review, 41, 391-403. Zenner, W. P. (1988). Common ethnicity and separate identities: Interaction between Jewish immigrant groups. In Y. Y. Kim & W. B. Gudykunst (Eds.), Cross-cultural adaptation. Newbury Park, CA: Sage.

Issues, Theory, and Research in Industriall Organizational Psychology - K. Kelley (Editor) 0 1992 Elsevier Science Publishers B.V. All rights reserved.

385

CHAPTER 12 STIGMAS IN ORGANIZATIONS: RACE, HANDICAPS, AND PHYSICAL UNATTRACTIVENESS Eugene l? Stone, Dianna L. Stone, and Robert L. Dipboye INTRODUCTION On the Fair Treatment of Individuals Nearly 50 years ago Myrdal (1944) commented on what he considered to be the great American dilemma, i.e., the general endorsement of democratic values by Americans juxtaposed against the widespread belief i n and practice of racial discrimination. Regrettably, racial discrimination is still a problem in the United States as is discrimination against individuals who are stigmatized or marked (cf. Goffman, 1963; Jones, Farina, Hastorf, Markus, Miller, Scott, & French, 1984) by virtue of their status on such dimensions as physical attractiveness, handicap, sexual preference, poverty, socioeconomic status, and religious beliefs. Issues surrounding the fair treatment of stigmatized individuals arise in both society and in the world of work. There are ongoing, often highly heated debates among and between members of such constituencies as legislative bodies, business leaders, union leaders, social philosophers and scientists, and members of the general public about both the appropriate conceptual and operational definitions of fairness. There are also heated debates about the methods that should be used to foster the equitable treatment of individuals in work organizations and various other contexts. Not surprisingly, in almost all debates about fair treatment issues, it is common to find

386

Stone, Stone, and Dipboye

substantial differences in the perspectives taken by individuals who view themselves as being unfairly treated and those who control the resources (e.g:, wealth, occupational status, social status, political power) that might be used to bring about perceptions of fairness. Stated somewhat differently, debates about fairness often center on differences in the allocation of opportunities or outcomes of various types (e.g., jobs, medical care, education, housing, prestige) between those who are members of in-groups and of out-groups. Fairness issues that relate to the world of work have been the focus of much controversy in the past several decades and in all likelihood will continue to command considerable attention among members of various constituencies. One outcome of this controversy has been the promulgation and enactment of legislation aimed at insuring that members of various identifiable groups are treated in an unbiased, equitable fashion in their dealings with work organizations. For instance, 1964 saw the passage of the Civil Rights Act, including Title VII of the act that dealt with issues of unfair discrimination (UD) against blacks, women, and several other protected classes in employment contexts. More recently, the United States Congress passed Public Law 101-336, the Americans with Disabilities Act of 1990. The enactment of this legislation was at least in part motivated by the rationale that substantial numbers of Americans have one or more physical or mental disabilities, UD against disabled people in employment and other areas is a serious and pervasive social problem, such individuals have had no legal rights to redress this discrimination, and thus clear, strong, and enforceable legal standards were needed to deal with UD of this nature. Unfortunately, neither existing laws nor prevailing social norms serve adequately to protect stigmatized individuals from UD in employment contexts. One reason for this is that although laws provide a means of seeking legal remedies for UD that is based upon such variables as race, sex, and physical handicaps, they provide no protection against UD based upon a host of other variables, one of these being physical unattractiveness. Another reason is that litigation is not always an effective method for dealing with UD problems. This is especially true in time periods when conservatives dominate in legislative and judicial bodies (e.g., the Supreme Court). In employment contexts, the standing of job applicants or job incumbents on a host of variables, including sex, race, handicaps, and physical attractiveness, serves as a basis for them being perceived as atypical. Moreover, to the extent that individuals are perceived to differ significantly from the prototype of the ideal job applicant or job incumbent, they may face a host of problems. As is demonstrated below, being perceived as atypical, aberrant, or differ-

Stigmas in Organizations

3 87

ent in one or more respects can lead marked or stigmatized individuals to be unfairly treated by non-stigmatized ("normal") individuals.

Prior Attention to Stigmas by Researchers and Theorists For the most part, industrial and organizational psychologists and researchers in closely related fields (e.g., organizational behavior, human resources management) have not been very active in studying stigmas and their effects. Most of what we know about stigmas and their effects is the result of work by social psychologists, political psychologists (e.g., Sears, 1987), and sociologists. Moreover, when stigmas have been studied by researchers in industrial and organizational psychology and closely related fields, the methods used have generally been highly obtrusive. Unfortunately, research of this nature often suffers from validity problems that stem from subjects managing impressions that they view as socially desirable (Crosby, Bromley, & Saxe, 1980; Stone, 1989; Webb, Campbell, Schwartz, Sechrest, & Grove, 1981). As a result, obtrusive studies that deal with such phenomena as race, sex, physical attractiveness, and handicaps may fail to detect effects that would have been revealed through unobtrusive research strategies. In view of the important role of fairness issues in contemporary society and effects that stigmas may have on the treatment of individuals in organizational contexts, it is vital that scientists and practitioners in industrial and organizational psychology and closely related fields pay considerably more attention to stigmas. Thus, our chapter deals with several issues that relate to stigmas in organizational contexts. Although a wide variety of stigmas might have been considered, we limited our focus to three that have received moderate levels of attention by industrial and organizational psychologists, i.e., race, handicaps, and unattractiveness.

Overview of Issues Considered This chapter deals with several issues connected with stigmas: First, we consider the nature of stigmas and the process through which individuals become stigmatized. This is followed by sections that deal, in turn with the stigmas of race, physical unattractiveness (unattractiveness), and handicaps. Next, we describe strategies that might be used to deal with problems that stigmatized individuals encounter in organizational contexts. Then we consider issues that relate to research on stigmas in organizational contexts. Finally, we

Stone, Stone, and Dipboye

388

present a limited number of conclusions stemming from our consideration of stigma-related theory and research.

STIGMAS AND THE STIGMATIZING PROCESS Definition of Stigma In Goffman's (1963) seminal work on stigmas, a stigma was conceived of as a real or perceived discrepancy between a person's virtual and actual social identities. A person's virtual social identity represents what is expected of the individual in terms of various physical attributes, physical abilities, personality traits, and so forth. It reflects what the person ought to be. Expectations about an individual are a function of the category in which he or she is placed by members of society and the set of attributes that are viewed as natural, ordinary, and normal for members of the same category (Goffman, 1963). It merits emphasis that as used by Goffman and by us the term normal means conforming to expectations about what is usual, typical, or standard. What is considered normal differs from one person, context, and time period to the next. Thus, for example, being unattractive, black, or physically handicapped does not imply abnormalcy in any absolute sense. However, individuals who have such attributes may be viewed as deviating from expectations that particular (e.g., attractive, white, non-handicapped) individuals have about what is customary or typical in a given context. The fact that such expectations often lead to unwarranted, unjust, and dysfunctional consequences for individuals who are perceived to be atypical is something that we view as morally reprehensible. In spite of this, the fact remains that individuals use categorization processes in dealing with complex social and physical environments (cf. Hamilton, 1976, 1981a, 1981b). An individual's actual social identity represents the way that an individual is actually perceived by others or is capable of being perceived. An individual is said to be stigmatized to the extent that the actual social identity is negatively discrepant from his or her virtual identity. Stigmas are the actual (e.g., race, physical handicaps, unattractiveness) or inferred attributes (e.g., mental illness) of the person that serve as the basis for him or her being perceived as atypical, aberrant, or deviant, and thus being discredited by nonstigmatized individuals (i.e., normals). Individuals who have such attributes or might be viewed as having them are said to be discreditable or stigmatizable. However, the mere existence of a flaw does not always result in a person being discredited or stigmatized. Some

Stigmas in Organizations

389

flaws can be concealed from others (e.g., criminal records, prior hospitalization for mental illness). Thus Goffman distinguishes between individuals who have been discredited by others knowing about their flaws (i.e., discredited or stigmatized individuals), and individuals who have managed to hide or conceal their flaws from others (i.e., discreditable or stigmatizable individuals). The process of stigmatization results from a perceiver comparing the actual or inferred attributes of an individual against the perceiver's stereotype or expectation about what the person ought to be: Stigmatization results from the person's actual social identity being negatively discrepant from his or her virtual social identity. In relatively recent work on stigmas, Jones et al. (1984) used slightly different terminological conventions than those of Goffman (1963). They use the term mark to denote a sign or indicator that has the potential to give rise to stigmatization. Individuals who bear such marks are referred to as markable persons. Individuals who perceive marks or intuit their existence are called markers. Finally, individuals who are perceived to bear one or more marks are referred to as marked persons. It should be obvious that while this terminology differs from that offered by Goffman (1963), the underlying concepts used by Jones et al. are quite consistent with those developed by Goffman. Thus, in the material that follows we use these terms interchangeably.

Tjpes of Stigmas and Dimensions of Stigmas Goffman noted that there are three general classes of stigmas: (a) abominations of the body (e.g., physical deformities, unattractive physical features), (b) blemishes of character (e.g. ,. aberrant personality traits, addictions, radical views, addictions, cnminality, unusual passions or behavioral proclivities), and (c) tribal stigmas (e.g:, race, nationality, religious affiliations). As noted earlier, in this chapter we limit our focus to three types of stigmas, i.e., unattractiveness, handicaps, race. The extent to which any given stigma leads to the spoiling of an individual's social identity is a function of a number of variables, including several that relate to the nature of the stigma that he or she is perceived to have. That is, reactions to stigmatized individuals are likely to differ as a function of the nature of given stigmas. Moreover, the nature of stigmas is likely to influence the strategies that stigmatized or stigmatizable individuals will use in dealing with others, especially normals. Jones et al. (1984) identified six, nonindependent dimensions along which marks may vary:

390

Stone, Stone, and Dipboye

Concealability. Some marks are more obvious or apparent than others. Compared to such potentially concealable stigmas as previous criminal conviction, mild forms of neuroses, bizarre sexual proclivities, or atypical religious beliefs, stigmas based upon such attributes as race, gross physical deformity, and severe physical handicap are less capable of being concealed. Thus, the latter types of stigmas are far more likely to cause problems for marked individuals.

Course. The nature of some marks may change over time, whereas others may remain virtually constant. Individuals who have stigmas that are viewed as irreversible (e.g., amputated limbs) or degenerative (e.g., multiple sclerosis) will typically engender more negative reactions from normals than individuals having stigmas that are considered alterable or reversible (e.g. , acne-related skin problems, facial moles or warts, deficient social skills). Disruptiveness. As a result of the nature of some types of marks, there will be strained patterns of interaction, communication, and so forth between stigmatized and non-stigmatized individuals. Strains in interaction will be especially likely when stigmas are disruptive in nature, i.e. , they are visible, dangerous, or aesthetically displeasing. The more disruptive the mark, the greater its tendency to evoke strong, negative reactions in others. Aesthetic qualities. Some types of marks result in the stigmatized person being perceived as repulsive, ugly, and upsetting by normals. All else constant, individuals with such marks as gross physical deformities, severe skin problems, burn-related scarring will evoke more negative affect among normals than individuals with stigmas that are not as aesthetically displeasing.

Origin. The marks that a person has may have resulted from factors over which he or she did or did not have control. All else constant, stigmatized individuals will be treated more harshly by others to the degree that they are seen as being responsible for their marks. Peril. Marks differ in the degree to which they evoke feelings of threat or danger among others. Individuals who are stigmatized by such marks as severe mental illness, convictions for violent crimes (e.g. , murder), and contagious or deadly infections (e.g., leprosy, AIDS), are likely to elicit more negative reactions from others than individuals marked by such stigmas as mild neuroses, baldness, cleft palate, missing limbs, and paraplegia.

Stigmas in Organizations

39 1

Overview of the Factors Associated with the Stigmatizing Process Role of Social Cognition in Stigmatizing Process. In view of the fact that stigmatization results from comparisons between virtual and actual social identities, psychological factors that are connected with this process deserve consideration. Of key importance in this connection is theory and research connected with the field of social cognition (cf. Hamilton, 1976, 1981a; Higgins & Bargh, 1987; Sherman, Judd, & Park, 1989). Researchers and theorists who deal with social cognition issues are concerned with how mental representations (cognitive schemata) influence the way in which information about target persons or events is stored, organized, and processed. Categorization. Categorization, which plays a key role in social cognition, is the process of assigning target individuals or objects to cognitive categories based upon perceptions or beliefs about the targets. Categorization is a frequent phenomenon that helps people to organize and simplify the complexities of their physical and social environments (Brewer & Kramer, 1985; Hamilton, 1976, 1981a). In the process of dealing with others individuals often use one (or a very few) attributes as a basis for assigning stimulus people to categories of various types (e.g., female, black, handicapped, criminal, insane). Moreover, once categorized, subsequent cognitions about the stimulus person are in large part determined by attributes that are perceived to be typical of category members. Inferences about the stimulus person's possession of category-relevant attributes are made even through no first-hand information about the stimulus person's standing visla-vis these attributes is available. Such inferences are especially inappropriate when the stimulus person shares very few of the attributes that are typical of other category members. Another aspect of categorization merits consideration. Once a person has been categorized, the subsequent processing of information about the target is guided by the nature of the category or categories into which he or she was placed. Thus, for example, if an individual is categorized as mentally i l l , subsequent information about him or her (e.g., that the individual shouted at a co-worker, didn't attend a company party) will be processed in light of the category to which the person was previously assigned (cf. Goffman, 1963, p. 15; Higgins & Bargh, 1987; Sherman, Judd, & Park, 1989). As a result, almost all interactions between the target and others will be guided by the categorization of the target and the resultant inferences that are made about his or her behavior. That is the person's interactions with others will be category driven.

392

Stone, Stone, and Dipboye

Stereotypes. Stereotypes refer to interpersonal beliefs and expectancies about social groups or categories that are widely shared, overgeneralized, and generally invalid (Ashmore & Del Boca, 1981; Brewer & Kramer, 1985; Hamilton, 1976, 1981a, 1981b; Jones et al., 1984; Miller & Turnbull, 1986). Moreover, although there are notable exceptions, stereotypes are largely negative in nature (Jones et al., 1984). Parenthetically, it deserves noting that stereotyping is often equated with prejudice. However, Brewer and Kramer (1985) note that recent research on stereotyping supports the position that stereotyping and prejudice are distinguishable from one another both conceptually and empirically. Jones et al. (1984) observe that categorization is often a precursor to stereotyping. That is, in dealing with stimulus persons who have been categorized, individuals are likely to make errors of inference that are based upon stereotypical beliefs (e.g., about assumed correlations between category membership and standing on various attributes). Moreover, memory about category members will be influenced by unfounded beliefs about relationships between category membership and the possession of traits that are typical of category members (i.e., the phenomenon of illusory correlation; Brewer & Kramer, 1985; Hamilton, 1981b). Stereotype-based errors in recollections, judgments, and inferences are especially likely when information about an attribute of a stimulus person is missing but knowledge of his or category membership is available (Brewer & Kramer, 1985; Hamilton, 1981b). For example, stereotypical beliefs might serve as the basis for the unfounded inference that since a stimulus person is black (member of one category) he or she has greater athletic ability and lower cognitive ability than the average individual (another category). In short, category membership combined with extant stereotypes are potent determinants of the impressions formed about others and judgments made about them. In their treatment of social stigmas Jones et al. (1984) argue that marking processes do not always involve the use of elaborate stereotypes. However, stereotyping is frequently used by markers as a means of dealing with the negative affect that they experience in dealing with stigmatized individuals. Thus, they consider stereotyping to be central to the stigmatizing process (cf. Jones et al., 1984, p. 155). Expectancies. Closely related to the notion of stereotypes is that of expectancies. These are anticipatory beliefs that we have about individuals that are based upon such factors as category membership

Stigmas in Organizations

393

and extant stereotypes (Higgins & Bargh, 1987). They are driven by such demographic characteristics of individuals as their age, race, sex, handicaps, and by such other factors as personality traits, previous activities/behaviors, past experiences, etc. (Hamilton, 1976, 1981a; Jones et al., 1984; Miller & Turnbull, 1986). In their consideration of social stigmas Jones et al. (1984) note that people bring cognitive and affective expectancies to relationships that they have with others and that both types of expectancies are greatly influenced by socially-based definitions of deviance and prescriptions for dealing with marked or deviant individuals (cf. Jones et al., 1984, p. 14). These expectancies serve to influence the way stigmatized people are perceived, the goals that normals develop for their relationships with marked individuals, and the plans or strategies that they use in working toward such goals. The effect of expectancies will be especially potent in situations where little or no knowledge is available about relevant attributes of the marked individual. Moreover, even if information about such attributes becomes available at a later point in time, expectancies will serve to override it. For example, based upon erroneous beliefs (expectancies) about the capacities of handicapped persons it is not uncommon to observe instances in which normals speak to individuals who are blind more loudly than they speak to normals or to assume that blind persons have developed special, sixth senses. Expectancies about individuals tend to bias the way they are perceived by others. This bias will be especially strong in cases where little or no objective information is available about the person. However, even when there is information about a person's standing on some dimension, expectancies that are formed about the person may override the information. For example, Jones et al. (1984) note that in the classic case of a stigmatized or marked individual the stigma initiates a drastic inference process that engulfs impressions about the marked individual. Self-fulJillingprophecies are an example of expectancies. Such prophecies or expectancies about others, including those that stem from a perceiver's false definition of a situation, tend to evoke behaviors in others that make the expectancies come true (Jones et al., 1984; Merton, 1948; Miller & Turnbull, 1986). Moreover, expectancies about others lead to behaviors toward them that tend to elicit stereotype-confirming behaviors. In the case of negative expectancies, this often leads to vicious, self-perpetuating cycles (Myrdal, 1944; Rosenthal & Jacobson, 1968). There is considerable empirical support for the effects of expectancies on behavior in social contexts. Moreover, research suggests that the relationship between expectancies and behavior

394

Stone, Stone, and Dipboye

toward a target will be moderated by (a) the goals that the individual has in a situation, and (b) the degree to which target individuals are viewed as modifiable by individuals with whom they are interacting (Miller & Turnbull, 1986).

Memory and Attribution Effects. Research shows that traits or behaviors of target individuals that are relevant to stereotypes about them are more likely to be remembered than those that are inconsistent with or irrelevant to the stereotype (Brewer & Kramer, 1985). Thus, for example, a supervisor would be more likely to recall an instance in which a black worker failed at a task calling for high levels of cognitive ability than the numerous instances in which the worker succeeded at other tasks calling for similar levels of cognitive ability. It is also true, however, that when information about an individual is plentiful, unambiguous, and greatly inconsistent with a given stereotype, that information may override the stereotype when judgments are made about the individual. However, base-rate stereotypes are very difficult to override; that is, they are very resistant to disconfirming information (Brewer & Kramer, 1985). Another consequence of categorization is that the assignment of individuals to categories results in perceptions of accentuated differences between stimuli in different categories and attenuated differences between members of the same category (Brewer & Kramer, 1985). Thus, for example, having characterized a group of new employees as "affirmative action hires" might lead existing workers in an organization to underestimate the extent to which the new hires are qualified to do their jobs. Moreover, having in-group (e.g., being normal) versus out-group (e.g., being marked) status may lead to errors in attributions about individuals with various types of stigmas (Jones et al., 1984; Brewer & Kramer, 1985; Goffman, 1963). Pettigrew (1979) came up with notion of the ultimate attribution error in social relationships: When in-group members exhibit desirable behaviors they attribute their actions to internal causes, but attribute the success of out-group members to external causes. A classic example of this is the attributions that are made when males are asked to explain exceptionally good performance on the part of male or female managers. Motivation to Categorize and Stereotype. Tajfel's (1981, 1982) social identity theory provides an explanation for the motivation to use categorization processes in interpersonal and intergroup contexts: Categorization provides individuals with a mechanism for maintaining a positive social identity. The reason for this is that maintaining

Stigmas in Organizations

395

a positive identity is more likely when we see positive features of members of the category to which we belong and less positive attributes of members of other categories.

Categorization Afects Outcome Allocations. Research clearly shows that not only are perceived attributes and ratings of in-group and outgroup members affected by category membership, but so are decisions about the allocation of positive and negative outcomes (cf. Brewer & Kramer, 1985). Moreover, intergroup competition tends to override equity considerations when allocation decisions are made. Categorization Afects Potential for Contact. Goffman notes (1963) that certain types of stigmas (e.g., character flaws) have the tendency to "spread" to those who have frequent contact with marked individuals. Thus, for example, individuals who are members of a "former mental patient" category will find it difficult to establish and maintain satisfactory social relationships with normals within and outside of work organizations. Normals may feel that association with such ex-patients will lead them to develop mental health problems. Individual Digerences in Prejudice. Although categorization and stereotyping are processes that are used by virtually all individuals, individuals differ with respect to (a) the strength with which stereotypical beliefs are held, (b) the content of their stereotypes, and (c) their tendencies to act upon their stereotype-based beliefs. For instance, there is considerable evidence that there are reliable individual differences in the tendency to behave in prejudicial ways toward members of outgroups (Murphy & Likert, 1938; Adorno, Frenkel-Brunswick, Levinson, & Sanford, 1950; Kogan, 1961). For example, recent research by Bierly (1985) suggests that prejudice toward outgroups is a generalized attitude. Her study revealed moderate to strong relationships between self-report measures of prejudice against blacks, attitudes toward women, attitudes toward old people, and attitudes toward homosexuals.

Consequences of Being Stigmatized As a result of being stigmatized, marked individuals often are deprived of many of the positive outcomes experienced by normals ( e . g . , jobs, educational opportunities, supportive social relationships), and experience a host of negative outcomes (e.g., social isolation, embarrassment, lowered self-esteem). These effects

396

Stone, Stone, and Dipboye

are seen in both work and non-work settings. Goffman (1963) and Jones et al. (1984) provide vivid illustrations of the differences in outcomes of various types experienced by stigmatized and normal individuals. Goffman (1963) noted that stigmatized individuals often feel unsure of how normals will perceive them. Thus, for example, they always feel as if they are being closely scrutinized by others, and are typically more self-conscious about their behaviors in social contexts than are normals. Moreover, marked individuals may fear even minor stigma-related failings, believing that these will call attention to or "amplify" the stigma. As a result, marked individuals may avoid social relationships, undergo self-imposed isolation, and experience depression, anxiety, and bewilderment over their fate. Expecting to be rejected or denigrated by normals, stigmatized individuals may constantly seek evidence of differential treatment by normals. Unfortunately, the resultant behaviors coupled with the anxiety that marked individuals experience in their dealings with normals may be sensed by the normals. Moreover, this may create a sense of discomfort in normals leading them to avoid further contact with marked individuals (Goffman, 1963; Jones et al., 1984) In the context of work organizations, stigmatized individuals may face a host of problems that relate to both (a) access to jobs, and (b) treatment as a job incumbent. With respect to access, stigmatized individuals may be unfairly excluded from jobs through organizational recruitment, and selection practices. For example, a black social psychologist at a major university noted how the recruitment of black faculty at predominantly white academic institutions sets up expectations that the climate of the institution will not be hospitable for blacks, thus reducing the motivation of blacks to accept jobs at such institutions (Sutherland, 1990). Even if they manage to get a job in a work organization that is populated primarily by normals, stigmatized individuals are likely to encounter a number of treatment problems. Among such problems are biases in performance appraisals, lower levels of pay and fringe benefits, dead-end or low-level job assignments, inadequate opportunities for training and skill upgrading, low promotion rates, absence of in-group status, inadequate mentoring and sponsorship opportunities, treatment as tokens, absence of relevant-role models, selfesteem problems, and problems that stem from a variety of selflimiting behaviors (e.g., Ilgen & Youtz, 1986). For instance, Sutherland (1990) noted that once employed by a white academic institution the black faculty member encounters numerous cues that alert him or her to the presence of the harmful effects of racism by Whites (cf. Sutherland, 1990, p. 18).

397

Stigmas in Organizations

Zkeoretical Perspectives mat M Explain Reactions to Stigmatized Individua s

7

Kinder (1986) advanced several potential explanations of why individuals may resist changes in the way that racial minorities are treated in American society. It would appear, however, that the same explanations are generally useful in describing reactions of normal individuals to those who are stigmatized. The explanations or perspectives are as follows:

Frustration-aggression Theory. This perspective argues that the frustrations, stresses, and strains, that normals experience in their lives end up manifesting themselves in aggression toward stigmatized individuals, even though the same individuals may have had no direct responsibility for the experiences of the normals. The psychological mechanism that is relevant here is displacement of aggression. Social Conflict Theory. This view holds that stigmatized individuals pose threats to the group interests or privileges of normal individuals. The normal individuals respond by taking actions to protect what they regard as their legitimate interests and privileges. General Ideological Differences. The notion here is that individuals differ in the extent to which they believe that government should intervene in the affairs of people (e.g., promulgate and enforce laws designed to reduce social inequality). Thus legislation-based approaches to insure the fair treatment of the stigmatized will be resisted by normals on ideological grounds. Cognitions about or "l'beories of Digerences. The position here is that there are innate differences between stigmatized and normal individuals that are unlikely to be altered by such means as education and socialization (e.g., beliefs in the position that differences between the cognitive abilities of racial groups are genetically based). Thus, efforts to reduce differences will be resisted by espousing and fostering "theories" that serve as explanations for observed differences. "

Culture-based Aflrmation of Core Rlues. This view is based upon the position that opposition to attempts to insure fair treatment of stigmatized individuals will be resisted because they violate basic values (e.g., the belief in rugged individualism). Thus, non-stigmatized individuals will fight to protect these values.

398

Stone, Stone, and Dipboye

Research-related Issues In Previous Studies of Stigmas As a result of methodological considerations, legitimate issues can be raised about the internal and external validity of many previous studies of the effects of stigmas in the context of work organizations. A brief consideration of some of the problems follows.

Study Design and Measurement Issues. Most previous stigma-related studies performed by industrial and organizational psychologists and others concerned with work organization-related issues have been of the obtrusive variety, leading to problems of impression management on the part of research subjects (cf. Stone, 1989; Webb et al., 1981). For example, in a number of previous studies dealing with race-related issues research subjects have been placed in the position of rating the performance of black and white workers in contexts where subjects can easily intuit that the researcher is concerned with the effects of ratee race on ratings. In one such study (i.e., Schmidt & Johnson, 1973), for example, black and white trainees in a foreman training program who had just completed a training module on racism were asked to rate their peers on several dimensions. Not surprisingly, the researchers found no race-related effects in the ratings. In another obtrusive study (Newman & Krzystofiak, 1979) subjects in organizations that had OFCC monitored Affirmative Action (AA) programs were presented with pairs of employment applications and accompanying photographs of black and white job applicants and asked to evaluate the applicants. Not unexpectedly, results of the study showed that when the hypothetical applicants had equal job qualifications the black applicant was favored over the white applicant. These and other types of obtrusive studies dealing with such stigmas as race, handicaps, and a host of other stigmas are all of questionable validity. One reason for this is that most of the subjects in such studies probably recognized that it is not socially acceptable to show blatant discrimination against women, blacks, and the handicapped (cf. Crosby et al., 1980). As research reviewed by Crosby et al.( 1980) clearly suggests, biased treatment of out-group members is far more likely to be present in research situations where (a) unobtrusive as opposed to obtrusive methods are used to study bias, and (b) there is low potential for the in-group member to experience negative consequences (e.g., embarrassment, physical harm, disapproval from researchers). Complementary views on this are presented by Brewer and Kramer (1985) who argue in favor of using measures of bias that are less capable of conscious control or socially desirable responding. Several theoretical models might explain differences found

Stigmas in Organizations

399

between obtrusive and unobtrusive studies of behavior toward stigmatized individuals. As Crosby et al. (1980) indicate, one that has particular promise is Kelman's model on attitude or opinion change. On the basis of this perspective it is predicted that in situations where research is overt or there are penalties for unfair treatment of the stigmatized, individuals will manifest compliance-driven, socially desirable behavior. However, when research is unobtrusive and subjects perceive that no negative consequences will ensue for behaving in a prejudicial manner toward stigmatized individuals, prejudiced individuals (e.g., racists, sexists) will manifest internalization-based behavior that results in treating stigmatized individuals in a biased, unfairly discriminatory fashion. One very important implication of this is that even if there really are problems of access and treatment that are based upon individuals having stigmas of various types, obtrusive studies are unlikely to detect them. In view of this, studies that show either no stigma-based effects or show what have been termed trivial effects should not be interpreted as credible evidence that such effects do not, in fact, exist or that they are of trivial practical importance. Unfortunately, the tendency to believe that stigmas have no effects or trivial effects appears all too common in the literature of industrial and organizational psychology, organizational behavior, and closely allied fields. The same tendency does not appear common in other fields of psychology (e.g., social, vocational, rehabilitation, political) or other social sciences (e.g., anthropology, sociology, political science).

The Matter of ESfect Sizes in Research. In a number of previous studies that deal with stigmas, effect sizes of relatively low magnitude have been found. This has led some to argue that biases attributable to such factors as race, unattractiveness, sex, handicap, and other types of stigmas are not of much practical importance. This conclusion, however, seems unwarranted for several reasons, only one of which is that the magnitude of effects detected in any given study varies as a function of the dependent measures that are considered. A clear illustration of this is research by Dipboye, Fromkin, and Wiback (1975) and Dipboye, Arvey, and Terpstra (1977) in which one purpose was to assess the effects of sex on ratings of job applicants and decisions about their hiring. When analyses were performed for the ratings criterion, the independent variable of sex explained only about one percent of the variance in ratings. However, when the criterion was that of who to actually hire, the effect size was substantial. In short, inferences about the practical importance of effects detected in any given study must be

400

Stone, Stone, and Dipboye

conditioned by a consideration of the reliability, validity, and sensitivity of critenon measures used.

RACE AS A STIGMA IN ORGANIZATIONS

The Nature and Prevalence of Racism in Society Nature of Racism. Numerous researchers currently support the position that old-fashioned, good-old-boy, red-neck racism has been supplanted by a newer form of racism, symbolic racism. This form of racism is a combination of traditional forms of racism and a subscription to traditional American values, especially individualism (cf. Kinder, 1986). Symbolic racism has its roots in both earlylearned racial stereotypes and deep-seated feelings of social morality. Kinder (1986) and colleagues (e.g.! Kinder & Sears, 1981) label this modem form of racism as symbolic racism because of their belief that it is based in large measure upon abstract, moralistic racial resentments, complaints about society as a whole, and the endorsement of American values, rather than upon the contemporary racial experiences of individuals. According to McConahay and Hough (1976) symbolic racism is based upon the belief that blacks violate dearly held values or make improper demands for changes in the racial status quo. Moreover, Sniderman and Tetlock (1986) note that those who do research on the symbolic racism construct believe that traditional racism, based upon the tenets of black inferiority, white supremacy, and racial segregation is no longer fashionable in the United States. Incidentally, it deserves noting that Sniderman and Tetlock (1986) do not take issue with the existence, pervasiveness, or harmfulness of racism. For example, they note that much is currently known about racial prejudice including its tenacity and its capacity to present itself in many different forms in various contexts. However, Sniderman and Tetlock (1986) do not agree with several of the arguments and assumptions advanced by those who do research on symbolic racism. Moreover, they raise several important issues about the construct validity of existing, popular measures of symbolic racism. Weigel and Howes (1985) also take issue with the symbolic racism concept. They argue that what Kinder, Sears, McConahay, Hough, Sears, and others have called symbolic racism is nothing more than a generalized tendency to derogate members of outgroups. Moreover, two empirical studies reported by Weigel and Howes (1985) provide empirical support for their view that racism represents a generalized tendency to derogate members of out-

Stigmas in Organizations

40 1

groups. Results of one study, involving data from 92 white residents of a medium-sized city in Massachusetts, showed moderate to strong relationships between measures of what has been called symbolic racism and measures of prejudice against blacks, self-reported political conservatism, and preferences for Reagan over a set of possible Democratic opponents. Results of the second study, using data from 110 white residents of a small, rural New England town revealed moderate to strong correlations between multiple measures of racial prejudice and measures of political conservatism, religiosity, marital traditionalism, authoritarianism, belief in the just-world hypothesis, and need for approval. Weigel and Howes concluded that responses to measures of symbolic racism reflect a generalized tendency (disposition) to disparage members of any outgroups that are perceived as challenging the legitimacy of the existing social order (e.g., women, blacks, homosexuals). At present there are several explanations for contemporary expressions of racism by individuals. Some argue that it reflects a commitment to a uniquely American conception of equality (Sniderman & Tetlock, 1986). Others view the espousal of traditional values as a mechanism for camouflaging more traditional forms of racism (Sniderman & Hagen, 1985). Still others view modern racism as stemming from the combination of traditional forms of racism and a belief in traditional values (Kinder, 1986). Yet others see it as a generalized tendency on the part of individuals to derogate members of outgroups (Weigel & Howes, 1985). Whatever its basis, however, there is little doubt about its existence. There is also little doubt about the durability of major elements of the negative stereotypes of blacks and other minorities that were documented decades ago by such researchers as Katz and Braly (1933), and Karlins, Coffman, and Walters (1969).

Prevalence of Racism. There is some evidence suggesting a decreased tendency on the part of white Americans to endorse statements containing obvious and outrageous racial stereotypes (cf. Weigel & Howes, 1985). However, there is still clear and abundant evidence of various forms of racism in the United States. Among the indicants are a National Opinion Research Corporation study showing that in 1984 25% of Americans surveyed agreed with the stance that white people have a right to keep blacks out of their neighborhoods if they choose to do so, and blacks should respect that right. Moreover, 47% supported the view that the freedom of the homeowner to sell to whom he or she chooses takes precedence over the principle of racial equality (cf. Sniderman & Tetlock, 1986). In addition, data from a 1978 national, Harris survey indicated that

402

Stone, Stone, and Dipboye

about 50% of Americans agreed that blacks tend to have less ambition than whites; one third endorsed the separate views that blacks breed crime, blacks want to live off the handout, and blacks are more violent than whites; and one fourth believed that blacks have less native intelligence than whites (cf. Sniderman & Tetlock, 1986). %ken together, these data provide clear support for the notion that stereotypes about blacks and racist views are alive and well among significant proportions of the population of white Americans. As Kinder (1986) has noted racism remains alive and all too well in the United States. The major shift seems to be in the ways that it is expressed: Although many individuals are increasingly reluctant to endorse views that smack of old-fashioned red-neck racism, they are not seemingly reluctant to support policies and practices in housing, employment, education, and other spheres of activity that are based upon "traditional values" and lead to the differential treatment of whites and non-whites. Put somewhat differently, in today's world many whites are less racist in their words than in their deeds (cf. Deutscher, 1966, 1973). Unfortunately, moreover, it is unlikely that views on race and racial prejudice will change much in the foreseeable future (e.g., Ashmore & Del Boca, 1976; Rokeach, 1973). Racism appears extremely durable because it is linked closely to a host of related beliefs, values, etc. In this regard, Weigel and Howes (1985) argued that to the extent that prejudice and racism are deeply rooted within the social structure, prejudicial sentiments and discriminatory behaviors will be widespread at the level of the individual. Moreover, these sentiments and the behaviors will be legitimized partly by the bigot's sense of being strongly tied to traditional American beliefs and values. In view of this, Weigel and Howes (1985) are not surprised by the seeming resilience of prejudice in the United States.

Unobtrusive Research on Racism As is demonstrated below, many obtrusive studies of the effects of race in organizational contexts have shown no effects, effects of low magnitude, or counter-intuitive effects (e.g., blacks applicants rated more positively than whites). This has led numerous analysts to conclude that race is essentially a non-issue in terms of either access (e.g., hiring) or treatment (e.g. ,.performance evaluations) considerations. However, as noted earlier, such a conclusion seems unwarranted. The validity of the same conclusion becomes even more questionable when one considers the results of a substantial number of unobtrusive studies of racism. As Crosby et al.

Stigmas in Organizations

403

(1980) show, unobtrusive research on racism shows clear effects of race in terms of such criteria as helping, aggression, and various types of nonverbal behavior. With respect to the criterion of receiving help, the Crosby et al. review demonstrated that in unobtrusive studies of racism (a) whites gave more help to other whites than to blacks, (b) the more distant or remote were blacks needing help, the less likely they were to be helped by whites, and (c) the greater the degree to which blacks were seen as subordinate to whites (e.g., behave unassertively), the greater the odds of them being helped by whites. Research reviewed on aggression against blacks revealed that overall, whites aggressed more against blacks than against other whites. In addition, whites exhibited less direct aggression, but more indirect aggression against blacks when (a) there was the potential for blacks to retaliate against whites for aggression, (b) whites could be censured for aggression against blacks, and (c) whites lacked the capacity to aggress in an anonymous manner. Finally, in studies that used nonverbal behaviors as criteria the research showed that (a) the friendliness of whites toward blacks was less than that of their friendliness toward other whites, and (b) the duration of interaction with blacks was lower than that with other whites. Crosby et al. (1980) concluded that prejudice against blacks is still potent among American whites (cf. p. 560). However, the nature of expressing their prejudice entails a shift from blatant or overt discrimination to more subtle or covert forms of racism. Interestingly, the types of effects shown in the unobtrusive studies reviewed by Crosby et al. are generally not found in obtrusive research on racism. What are the implications of this for organizational research on the effects of race? At least one is that many previous obtrusive studies of race i n organizational contexts may have yielded results that have either seriously underestimated race effects or led to false impressions about the direction of such effects (cf. Sears, 1987). More on these issues is offered in connection with the review of organizational research on race that follows.

Organizational Research on the Efsects of Race In previous years, a number of studies have been performed that have dealt primarily or in part with the effects of race on various criteria (e.g., performance ratings, interview outcomes). Moreover, the literature relevant to several such outcomes has been reviewed recently (e.g., Arvey, 1979; Arvey & Campion, 1982; Bartol, Evans, & Stith, 1978; Kraiger & Ford, 1985; Ford, Kraiger, & Schechtman, 1986). Rather than fully describing the content of such reviews, we summarize findings in them that are relevant to issues

404

Stone, Stone, and Dipboye

considered by this chapter, Subsequent to presenting the results of the same reviews we consider the findings of a representative set of studies dealing with the effects of race in organizational contexts. Arvey's (1979) review dealt with factors that may lead to unfair discrimination (UD) in the employment interview, one being interviewee race. Only three studies were found, and only one showed statistically significant race effects. More specifically, a study by Haefner (1977) showed that interviewers' evaluations of interviewees were interactively determined by interviewee race, sex and age. Arvey concluded that there does not appear to be evidence of race effects in interview-based ratings or evaluations. He correctly cautioned, however, that since many of the subjects in the three studies were probably aware of EEOC-related issues, it may have been unlikely for them to provide data indicative of prejudice against blacks (cf. Sears, 1987). Arvey and Campion (1982) reviewed three studies dealing with race effects in the employment interview (McIntyre, Moberg, & Posner, 1980; Newman, 1978; Mullins, 1978). In the case of the criteria associated with all three such studies blacks appeared to fair better than whites. However, we suspect that social desirability biases may have been responsible for a number of the findings i n these studies. Bartol, Evans, and Stith (1978) reviewed literature dealing with the leader behaviors, job performance, and job satisfaction of black versus white leaders. Among their findings were that: (a) In situations where they are i n the minority, black leaders are less likely to initiate interactions with others than are white leaders; (b) Studies of leader potential revealed mixed support for the position that black rates are perceived to have less leadership potential than whites; (c) There were insufficient data on which to base a comparison of the relative performance of black versus white leaders. However, the authors noted that the available evidence revealed a tendency for black leaders to be evaluated in terms of different criteria than white leaders; (d) Race of rater and r a t e influenced ratings of performance; (e) On the basis of limited evidence it appears that black leaders tend to be less satisfied than white leaders, and compared to white leaders, blacks are more likely to indicate that their lower-level needs are not being met and to place greater emphasis on the satisfaction of such needs; and (f) Some research has shown that white subordinates engage in behaviors that hinder the effectiveness of black leaders. Kraiger and Ford (1985) performed a meta-analysis of studies dealing with ratee race effects in performance ratings. Seventy-four of the studies involved white raters (cumulated N = 17,159) and 14

Stigmas in Organizations

405

studies involved black raters (cumulated N = 2,428). Results showed that white raters assigned significantly higher ratings to white ratees than to black ratees. In addition, black raters assigned significantly higher ratings to black ratees than to white ratees. On the basis of a consideration of several potential moderators of effects, the authors reported that (a) race-related effect sizes were much larger in field settings than in laboratory settings, and (b) there was a tendency for race effects to decline (for white raters) as the proportion of blacks in the sample of ratees increased. The authors noted that the latter finding may have resulted from a decreasing tendency of white raters to use race as a cue in formulating ratings. Ford, Kraiger, and Schechtman (1986) studied ratee race effects via a meta-analysis of studies dealing with subjective and objective measures of three criterion classes (job performance, cognitive criteria, and absenteeism). Data were from 53 samples involving cumulated Ns of 10,222 for objective criteria and 9,443 for subjective criteria. Results showed that for these samples, whites received higher ratings in terms of both the objective and subjective criteria. Moreover, black versus white rating differences varied as a function of the type of criterion measure considered. The difference was greatest for objective cognitive criteria. Moreover, for subjective criteria there were no significant differences in black versus white effect sizes. The authors concluded that there are race-based differences in performance, and that these differences vary as a function of the type of criterion considered (e.g., absenteeism, performance quantity, performance quality) and the subjectivity of the measures used to index performance.

Representative Studies Dealing with Race In the past several decades a number of studies have been conducted that deal with the effects of race on various criteria. In virtually all such studies blacks have been compared with whites. A number of such studies are briefly considered here. The review considers studies in terms of three basic types of findings (a) studies showing that whites are better off than blacks, (b) studies showing no effects for race or mixed findings, and (c) studies showing that blacks are better off than whites.

Studies Showing whites Better Of than Blacks. One type of finding in previous research is that whites are better off (e.g., rated more highly, recruited more readily) than blacks with regard to some criterion (e.g., Haefner, 1976; 1977; Huck & Bray, 1976; Mobley, 1982; Parsons & Liden, 1984; Schmitt & Hill, 1977; Hamner, Kim,

406

Stone, Stone, and Dipboye

Baird, & Bigoness, 1974; Andrisani & Nestel, 1976; Cox & Krumboltz, 1958). Among these, several studies have dealt with the effects of race on judgments relevant to access (e.g., recruitment stage ratings, suitability ratings). Haefner (1976) collected data from 286 employers in three cities in Illinois to assess recommendations on hiring hypothetical applicants who differed in terms of race, age, sex, and competence. There were several significant interactions, among them were interactions between (a) race and sex and (b) race and age: Black-females and old-blacks were evaluated least favorably. There was no significant main effect for race, although there was a trend that favored the hiring of whites. In a follow-up, Haefner (1977) interviewed a sample of 588 midwestern, working adults to determine preferences for working with hypothetical coworkers who differed in terms on sex, age, competence, and race (white vs. black). For the most part, subjects had non-managerial, non-professional jobs, and were from two census tracts that had low levels of median annual income. Results showed that a hypothetical white coworker was preferred over a hypothetical black coworker. The race findings are of interest in view of the nature of the study's sample of subjects; perhaps these subjects had lesser desires to provide socially desirable responses than do the college students who are often the subjects in studies dealing with race effects in organizational contexts (e.g., Mullins, 1978). Parsons and Liden (1984) analyzed suitability rating data on 517 black and white applicants for jobs at an amusement park. The interviewers were eight whites who rated the overall quality of applicants on the basis of job application data and eight nonverbal cues (e.g., poise, articulation) emitted in short face-to-face interviews. Results showed that whites were perceived to be more suitable than blacks. Moreover, the data indicated that nonverbal cues had a substantial effect on ratings. Two studies dealt with assessment center ratings (Huck & Bray, 1976; Schmitt & Hill, 1977). Huck and Bray (1976) studied assessment center ratings and the subsequently rated performance of 91 white and 35 black women who attended an assessment center between 1966 and 1971 and were subsequently promoted into one of two managerial jobs in a telephone company. Results of analyses of the assessment center data showed that whites received higher ratings on six of 16 rated variables and on three of four factor score dimensions derived from the original 16 measures. In addition, blacks received less positive performance ratings than whites. Schmitt and Hill (1977) analyzed assessment center-based ratings for 306 nonmanagerial employees of a midwestern utility company. Rates were assessed in groups that differed in terms of their size and racial composition. Results showed that as the proportion of whites in a

Stigmas in Organizations

407

group increased ratings of blacks decreased on several dimensions (forcefulness, oral communication). The authors argued that the magnitude of effects was generally low. Performance evaluations of hypothetical (videotaped) workers or actual employees were the focus of several other studies. Mobley (1982) analyzed performance rating data from 1035 workers in a distribution center to assess potential effects of the race and sex of raters and ratees. Results of ANOVA analyses revealed main effects for rake sex, race, and several interaction effects (e.g., R a t e Race X Ratee Sex X Rater Sex). Overall, the results showed that black ratees were given substantially lower ratings than white ratees by both black and white raters. In addition, black males were rated lowest by all types of raters except white females. Moreover, white females were rated most positively by all types of raters. In spite of these findings the author argued that there was little evidence of UD against blacks. Hamner, Kim, Baird, and Bigoness (1974) had 36 undergraduate business students provide ratings of videotaped stimulus people who differed in terms of race (black vs. white), sex, and performance level. Results revealed that whites gave higher ratings to white than to black ratees. In addition, black ratees gave higher ratings to black than to white ratees. Moreover, the study showed that while there was virtually no relationship between manipulated performance and ratings for the black worker, there was a relatively strong positive relationship between these variables for the white worker. The authors concluded that race may represent a potent source of bias in ratings, even in cases where it is possible to obtain relatively objective data on performance. Cox and Krumboltz (1958) compared peer ratings of airmen for same versus different race rater-ratee pairs. Results indicated that whites gave other whites higher ratings than blacks. A limited number of studies have compared blacks and whites on personality traits or other individual difference variables. For example, Andrisani and Nestel (1976) studied the relationship between locus of control over reinforcement and a number of outcomes among approximately 1483 working adults in the United States. Results of their study revealed that whites had more of an internal locus than blacks and that internality was predictive of such outcomes as occupational attainment, hourly earnings, job satisfaction, perceived financial progress, and change in job satisfaction. The study also showed that compared to blacks, whites had (a) greater occupational attainment, hourly earnings, and perceived financial progress, and (b) lesser unemployment and job satisfaction. The authors speculated that blacks, being at the lowest levels of the

408

Stone, Stone, and Dipboye

occupational hierarchy, were more likely to develop an external orientation than whites. In addition, they argued that "internalexternal expectancies both affect one's behavior toward the environment and are affected by one's environment" (p. 163). They also noted that the externality of those at the lower end of the socioeconomic continuum may reflect the cumulative effects of unfulfilled expectations rather than a lack of initiative.

Studies Showing Mixed Results or Nu Diflerence between whites and Blacks. Ratings of hypothetical job applicants have been the focus of a few studies reporting null findings or mixed findings (Adams, 1978; Greenhaus & Gavin, 1972; McDonald & Hakel, 1985; Newman & Krzystofiak, 1979; Pulakos, White, Oppler, & Borman, 1989; Rand & Wexley, 1975; Schmidt & Johnson, 1973; Wexley & Nemeroff, 1974). Note that in the case of studies showing no differences it is often difficult to determine whether this resulted from (a) there being no real (true) differences or (b) methodological problems that precluded the finding of actual differences (e.g., low statistical power, measures of dependent variables that lacked sensitivity, responses that contain self-presentation biases). Several studies dealt with ratings of applicant suitability. Wexley and Nemeroff (1974) had 120 students review resumes and evaluate the suitability of hypothetical black and white applicants for the job of mechanical engineering technician. The study showed no effect of race on suitability ratings. Rand and Wexley (1975) had 180 white subjects view videotapes of interviews that used black versus white interviewees. Results showed no effect of race on evaluations of the videotaped interviewees. Newman and Krzystofiak (1979) mailed subjects pairs of resumes that differed in terms of the equivalence of applicants' qualifications for a job (equivalent, non-equivalent) and in terms of the race of the applicant (white, black; manipulated using photographs attached to resumes). The stimulus materials were sent to 240 employment managers in four geographic regions. Results showed that when credentials were equivalent the black applicant was rated more highly than the white applicant. When they were nonequivalent the white applicant was rated more positively than the black. The authors correctly noted how their findings may have resulted from self-presentation biases on the part of respondents. It also deserves noting that the study was obtrusive in that (a) respondents were aware that they were in a study comparing the resumes of hypothetical pairs of job applicants and that affirmative action pressures were a consideration, and (b) respondents were all employed by firms that had OFCC-monitored affirmative action programs.

Stigmas in Organizations

409

McDonald and Hake1 (1985) studied interview data from 160 students who interviewed applicants (confederates) who differed in terms of initial suitability (suitable, unsuitable), race (black, white), sex, and responses to questions (confirming or disconfirming of initial, resume-based impressions). The dependent variables were strategy used in questioning the interviewee and ratings five dimensions of suitability. The authors reported that the proportion of negative questions asked interviewees was in part a function of Race X Suitability and Race X Interview Response interactions. They also found that pre-interview ratings of applicant suitability were in part a function of applicant race and a Race X Suitability interaction. Unfortunately, the authors did not report their results in sufficient detail to allow for the determination of the nature of the significant effects. A few studies were concerned with performance ratings of actual workers (during training or on-the-job). Adams (1978) had 406 subordinates of black and white managers in a retailing organization rate their supervisors on a number of dimensions, including the extent to which they showed consideration. Results of multivariate analyses revealed that black managers were rated more highly than whites. In addition, job problems reported by black subordinates rating black managers were lower than those reported by white subordinates under black managers. Schmidt and Johnson (1973) compared ratings of 43 black and 50 white foreman trainees on two dimensions (future success as a manager, and drive and assertiveness). Subjects were trainees in a training program module aimed at reducing racism. Mean ratings on the two dimensions did not differ from one another. However, in view of the nature of the training program and the demands set up by the rating process it seems unlikely that the obtaked ratings would show evidence of racism even if racist sentiments and biases truly existed among the white trainees. Pulakos, White, Oppler, and Borman (1989) studied peer and supervisor performance rating data from 39,537 rater-rate pairs among 19 military occupational specialties in the United States Army. Independent variables in their research were sex and race of raters and ratees. Results of analyses of variance revealed that for peer ratings white ratees were rated more positively than black ratees on technical skill and effort, and personal discipline. However, blacks were rated more favorably than whites on military bearing. For supervisor ratings, whites received more positive ratings on technical skill and job effort, and blacks received more favorable ratings on military bearing. While the authors argued that the rating differences were of generally small magnitude, they noted that racial biases may be less prevalent i n military contexts than in civilian

410

Stone, Stone, and Dipboye

organizations. A limited number of studies dealt with differences in personality between white and black workers. Greenhaus and Gavin (1972), for example, compared effort-reward expectancies for 12 types of positively valent outcomes among 78 black and 186 white nonmanagerial employees of an airline. Results showed that the blacks had greater expectancies than the whites in the case of three types of outcomes (more pay, more say in the job, better benefits). This was in spite of the fact that the whites had higher levels of rated performance than the blacks.

Studies Showing Blacks Better Off than Whites. Interestingly, a number of studies have shown that blacks are better off than whites on measured variables (e.g., Bigoness, 1976; McIntyre, Moberg, & Posner, 1980; Mullins, 1978, 1982; Newman, 1978). However, in view of the results of the earlier described unobtrusive research on race and the numerous social psychological studies dealing with racism and racial stereotypes, the findings of these organizationallyrelevant studies are very puzzling indeed. Several studies have compared performance ratings given to videotaped (hypothetical) workers who differed in terms of their race. Bigoness (1976) had 60 white undergraduate students rate the performance of videotapes having black versus white stimulus people (ratees) who were high versus low performers. Results showed an interaction effect: Ratings of the high performers were equivalent for white versus black ratees; however, low performing blacks were given higher ratings than low performing whites. Schmitt and Lappin (1980) had 73 undergraduate students rate the performance of videotaped workers who differed in terms of performance level, sex, and race (black, white). Analyses of the rating data revealed a main effect for ratee race (blacks rated more positively than whites) and a Rater Race X Ratee Race interaction: The mean level of rated performance for black raters rating black ratees was higher than the mean rating for the other three combinations of rater race and ratee race. The authors noted that race accounted for very little variance in ratings, however. Other studies have compared subjects' reactions to resumes of hypothetical applicants. McIntyre, Moberg, and Posner (1980) sent fictitious resumes to 458 target organizations. The resumes manipulated applicant race (black, white) and sex (for only the white applicants). Actual responses of the firms to the resumes were the dependent variable. Results showed that for one hypothetical applicant there were 78 instances of favoritism toward blacks and 51 instances of favoritism toward whites. For a second hypothetical

Stigmas in Organizations

411

applicant the corresponding numbers were 94 and 52. Overall, the results indicated that with respect to pre-selection responses to resumes, the hypothetical black applicant was better off than the white applicant. The authors concluded that discrimination against blacks may not be as prevalent as it has been presumed to be. However, this conclusion must be tempered by the fact that the authors studied responses to mailed applications, not hiring decisions! Newman (1978) conducted what he characterized as an unobtrusive study of racial discrimination in which he sent bogus resumes of black and white job applicants to 240 companies that had affirmative action programs and observed the resulting responses. Respondents did not know that they were part of a study. Results showed that for larger companies the applications of blacks were viewed more positively than those of whites in terms of a recruitment criterion. Note that while this study suggests favoritism toward blacks, the basis for the favoritism is unclear. Were respondents in the larger firms more sensitive to EEOC guidelines? Were they simply trying to meet their affirmative action goals? In a rather obtrusive study of race effects on applicant suitability judgments, Mullins (1978, 1982) had 176 business students provide ratings of videotaped job applicants who differed terms of race (black, white) and applicant quality (low, high). After rating the applicants, subjects completed a measure of racial prejudice. Interestingly, the results showed that the black applicant was rated more positively than the white applicant. In addition, the greater the level of subject racism the more positively they rated the black applicant. In view of the study's subjects, the year the research was conducted, and the nature of the Applicant Race X Rater Prejudice interaction effect, it's difficult to place much faith in the study's results: Serious questions can be raised about the role that self-presentation biases may have played in this highly obtrusive study of race effects.

Summary Material presented in this section shows several things: First, racism and UD based upon race can take several forms. Nowadays white subjects appear somewhat less likely to endorse views that are blatantly racist. However, they appear quite willing to endorse views that are indicative of more covert forms of racism. Second, racial prejudice against blacks appears pervasive i n America and there's little basis for assuming that it will decrease much in the foreseeable future. Third, although obtrusive studies of racism often show little or no evidence of its existence, unobtrusive studies show a clear picture of racism. Fourth, organizational research on the

412

Stone, Stone, and Dipboye

effects of race shows mixed findings with respect to various criteria. In studies that have been performed in naturalistic settings there are clear effects of race on such criteria as subjective and objective performance ratings, interview-based ratings of suitability, and ratings of leadership or managerial potential; However, laboratory experiments and field experiments that are obtrusive in nature often show either no effects or counter-intuitive effects (i.e., blacks viewed more positively than whites). Fifth, taken as a whole, the evidence considered suggests that blacks will face both access and treatment problems in organizations that have predominantly white members and in organizations in which decisions about the allocation of outcomes of various types are made by white members.

PHYSICAL UNATTRACTIVENESS AS A STIGMA IN ORGANIZATIONS An impressive amount of evidence has accumulated over the last 25 years showing that people are personally attracted to those who conform to conventional standards of physical attractiveness and tend to reject those who do not conform to these norms (Bull & Rumsey, 1988; Hatfield & Sprecher, 1986; Herman, Zanna, & Higgins, 1986). One would hope that this bias would be limited to dating decisions of adolescents, but physical attractiveness biases have been shown to influence more important decisions such as teacher judgments of student intelligence (Clifford & Walster, 1973), voters' preferences for political candidates (Efran & Patterson, 1974), and jurors' judgments in simulated trials (Efran 1974). Physical attractiveness also influences the personnel decisions and evaluations that are crucial to career success. In reviewing this literature, we consider biases that relate to three primary classes of criteria: selection decisions and evaluations, performance appraisal and promotion, and financial rewards. Note that the first criterion category has to do with access issues, while the latter two categories are concerned with treatment considerations.

Personnel Selection Decisions and Evaluations In the attempt to find employment, applicants must pass several hurdles. The most common sequence of events is to first submit an application or rbumt. Screening often occurs at this point as choices are made as to which applicants will be interviewed and which will be rejected. One or more interviews typically follow, culminating in the final selection or rejection of applicants. The research has shown

Stigmas in Organizations

413

that appearance of the applicant is related to selection decisions (Carlson, 1967; Kinicki & Lockwood, 1985; Rynes & Gerhart, 1990; Springbett, 1958). More relevant to our concerns, applicants with a physically unattractive appearance are seen as less qualified than applicants with more attractive appearances. This bias has been consistently shown in laboratory research using student subjects (Cash & Kilcullen, 1985; Dipboye, Arvey & Terpstra, 1977; Heilman & Saruwatari, 1979; Croxton, Van-Rensselaer, Dutton, & Ellis, 1989; Beehr & Gilmore, 1982; Cann, Siegfried & Pearce, 1981). Attractiveness biases also have been found in laboratory and field studies with persons in decision making roles (Carroll, 1966; Cash, Gillen & Bums, 1977; Benson, Severs, Thgenhorst & Loddengaard, 1980; Gilmore, Beehr, & Love, 1986; Raza & Carpenter, 1987; Shahani & Dipboye, 1990; Waters, 1985). Two studies not showing effects of physical attractiveness on selection evaluations (Greenwald, 1981; Jackson, 1983a) used very weak manipulations of the attractiveness variable. In one of the first demonstrations of this bias, Carroll (1966) examined the relationship of attractiveness to the success of graduating business students in finding employment. Handsomeness of the applicant was rated by three judges from photographs. Of 19 characteristics used as predictors in this study (e.g., age, grades, major, work experience), only three were related to success in job-hunting. The strongest of these was appearance. The more handsome the applicant, the more visit offers received ( r = .26), the higher the job-visit ratio ( r = .30),and the higher the applicant scored on a composite criterion of job-hunting success ( r = .27). In a more recent field study, Raza and Carpenter (1987) surveyed eight interviewers and found that interviewer ratings of the applicant's physical attractiveness were significantly related to ratings of whether the applicant would be hired ( r = .30) as well as ratings of the applicant's general employability ( r = .36). The strength of the relationship between perceived physical attractiveness and hiring recommendation was lower than that for judged intelligence ( r = .45), likability (r = .48),and skill ( r = .70). Nevertheless, the relationships found for physical attractiveness were surprisingly strong considering that the researchers combined data from applicants from a wide variety of job types, including managerial, clerical, semi-skilled, and skilled labor. The field research shows that physical attractiveness can enter into actual employment judgments and decisions, but the interpretation of these findings suffers from the causal ambiguity present in all non-experimental, field research. We cannot be sure on the basis of these findings whether physical attractiveness bias represents an

414

Stone, Stone, and Dipboye

independent effect or whether it reflects a confounding with other attributes of applicants (e.g., actual social skill). Experiments are needed to control for possible confounds and to determine whether it is an attractiveness bias that is actually responsible for observed differences in evaluations. A typical laboratory experiment presents subjects with a resume or application and manipulates attractiveness by means of a pre-rated photograph attached to the paper credentials. The most consistent finding in such research has been that attractive applicants are preferred over unattractive applicants (Bieber & Dipboye, 1988; Cash & Kilcullen, 1985; Dipboye, Fromkin, & Wiback, 1975; Dipboye, Arvey, & Terpstra, 1977; Gilmore, Beehr & Love, 1986; Harris, Harris, & Bochner, 1982; Larkin & Pines, 1979; Quereshi & Kay, 1986). For instance, Dipboye, Fromkin, and Wiback (1975) found a bias in favor of attractive applicants on the part of both college recruiters and male college students. The highly attractive applicant was judged as more qualified for a manager-trainee position than the unattractive applicant, and this preference was as strong for male applicants as for female applicants. Moreover, recruiters were as biased against the physically unattractive applicants as the students. A follow-up to this experiment using student subjects replicated the earlier findings (Dipboye, Arvey, & Terpstra, 1977). The above studies were concerned with general facial attractiveness. Other research has been concerned with specific physical features associated with attractiveness. Perhaps the most researched of these is body weight. Several laboratory investigations have shown that obese or overweight applicants are evaluated as less qualified for jobs than slender or normal weight applicants (Hankins, McKinnie, & Bailey, 1979; Larkin & Pines, 1979; Harris, Harris, & Bochner, 1982). These studies also provided evidence that overweight applicants were negatively stereotyped as unassertive, needing supervision, low in industriousness, and low in self-regard. A more realistic demonstration of the bias against overweight applicants was provided by Benson, Severs, Thgenhorst, and Loddengaard (1980). The same cover letter and resume of a female undergraduate was mailed to 70 male public health administrators who were unaware of the bogus nature of the inquiry. In the letter accompanying the resume, the applicant requested career guidance. Weight of the person was manipulated by including a picture that depicted the applicant as either normal in weight or overweight (her clothing was padded). Only 27 percent of the inquiries in the overweight condition received a reply, compared to 57 percent in the normal weight condition. When the nature of the response was examined, only 29 percent of the responses in the overweight condition were optimistic compared

Stigmas in Organizations

415

to 81 percent in the normal weight condition. Although bias against unattractive applicants is a consistent finding, researchers have only begun to explore the factors moderating and mediating the effect. One potential moderator is the age of the applicant. Most of the research discussed so far has manipulated attractiveness of young candidates, typically in their early ~O'S,and a crucial question to examine in future research is whether physical attractiveness has the same effect for older candidates. An experiment by Quereshi and Kay (1986) suggests that attractiveness effects are as strong for older as for younger candidates. MBA students evaluated 18 applicants on their suitability for the job of (a) tax manager of a large corporation in one experiment, (b) postmaster in a second experiment, and (c) vice principal of a high school in a third experiment. The investigators manipulated the physical appearance (above average, average, and below average attractiveness), age (28, 41, 54), and sex of the applicants. Results showed that the attractiveness of the applicant was the only variable that consistently influenced the evaluations across all three positions. Moreover, age of the applicant did not moderate the relationship between attractiveness and evaluations. This is the only study on this issue, however, and much more work is needed in both field and laboratory settings to fully assess the interactions of age and attractiveness. Another issue that remains largely unresearched is whether the unattractive applicant can do anything to reverse the bias by following the popular advice on dressing for success. Bardack and McAndrew (1985) found that dress had limited success in reducing bias against unattractive persons in an experiment in which students evaluated a female applicant for an entry-level management position. The applicant was presented as high, moderate, or low in attractiveness and was either dressed appropriately or inappropriately. Appropriate dress increased the chances of receiving a hiring recommendation more for the attractive applicant (from 82 percent to 100 percent) than for the unattractive applicant (from 68 percent to 76 percent). A third issue is the extent to which the nature of the job moderates the attractiveness bias. Cash, Gillen, and Burns (1977) found that professional personnel consultants favored attractive candidates over unattractive candidates for in-role jobs (men applying for auto salesperson or hardware clerk jobs and women applying for telephone operator or office receptionist jobs), but not for out-of-role jobs (men applying for telephone operator or office receptionist jobs and women applying for auto salesperson or hardware clerk jobs). Attractive applicants also were preferred over unattractive applicants in sex-neutral jobs (motel desk clerk, photographic darkroom assistant). According to Cash et al.,

416

Stone, Stone, and Dipboye

attractiveness intensified sex-typing of the applicants such that attractive men were seen as more masculine and attractive women were viewed as more feminine. Heilman and Saruwatari (1979) carried the reasoning of Cash et al. a step further and hypothesized that a high degree of attractiveness would be a stigma for women applying for traditionally male jobs. Support for this contention was provided in an experiment in which undergraduate college students rated applicants who differed on attractiveness (attractive, unattractive), sex (male, female), and position sought (managerial job, clerical job). Consistent with previous research (Heilman & Saruwatari, 1979), a main effect was found for attractiveness: Attractive males and females were rated as more qualified than unattractive applicants. A three-way interaction revealed that while attractiveness was always an advantage for males, attractiveness was only an advantage for females who applied for the clerical position. As hypothesized, unattractive females were rated as more qualified than attractive females when the job was a managerial position. Despite the intriguing nature of this interaction of sex, job type, and attractiveness, most of the experiments have found that attractive women are evaluated more favorably than unattractive women even for traditionally male positions. Thus, the discrepancy between the Heilman and Saruwatari (1979) findings and other research in this area may reflect mere chance. Another possibility is that the differences in results reflect specific stimulus characteristics of the attractive and unattractive female photographs used in these studies. The few photographs (2 to 6) that were used to manipulate attractiveness in previous research suggests that the "beauty is beastly effect" is limited to the specific types of attractive applicants sampled (e.g., the Marilyn Monroe or Dan Quayle type). Similarly, Hatfield and Sprecher (1986) suggest that studies showing beauty to be a disadvantage may have used pictures in which the applicants had an appearance that was inappropriate in a business context (e.g., long hair, casual dress). To avoid these potential confounds and to provide a sounder basis for generalizing the results, experiments are needed that provide broader and more representative samples of stimulus people. In a recent experiment, Bieber and Dipboye (1988) provided a larger sampling of stimuli than typically used. Thirty pre-rated photographs were sampled to represent each of three levels of male and female attractiveness (high, moderate, low), and statistical procedures were used that treated stimuli as random effects and subjects as fixed effects (Fontenelle, Phillips, & Lane, 1985). Subjects rated highly qualified male and female candidates for male-

Stigmas in Organizations

417

typed and female-typed positions on five point scales (1 = high, 5 = low). Only a main effect of attractiveness was found in which highly attractive candidates were evaluated more favorably (M = 2.89) than moderately attractive candidates (M = 3.09), who in turn were evaluated more favorably than unattractive candidates (M = 3.28). Contrary to the findings of Heilman and Saruwatari (1979) and Cash et al. (1977), being attractive always was an advantage for both the female and male applicant, regardless of the sex-type of the job.

Perjormance Appraisal and Promotion Once an individual is employed by an organization, their career success depends on how they are treated within the organization. Particularly important are the performance appraisals and promotions they receive. and physical attractiveness of the employee can influence both factors. "Performance" can seldom be measured in a completely objective fashion, but instead relies upon perceptions that develop among the various individuals who work with a given employee (ratee). In a complex job, where there are few objective outcomes, physical attractiveness may be one of the few attributes on which evaluators of performance can agree. Rosabeth Moss Kanter (1977) discussed this possibility in her book Men and Ubmen of the Corporation: Managers at Indsco had to have the appropriate "look. The appearance norms at Indsco were such that a visitor could easily intuit them: Managers were invariably white, male, and had a clean-cut, shiny look. The career prospects of secretaries were at least in part a function of their attractiveness and social skills. Moreover, task-related skills seemed to be less critical to job success as secretaries moved up the ranks. A widely publicized Supreme Court decision (Price Waterhouse v. Hopkins, 1988) provides a vivid illustration of bias in a promotion decision against a woman who apparently failed to comply with appearance norms. Ann Hopkins was a candidate for a partnership in a large accounting firm. Despite the fact that she had the best record in securing contracts for the firm in the year that her partnership was considered, her candidacy was deferred largely on the basis of her demeanor. The partners advised her to "walk more femininely, talk more femininely, dress more femininely, wear make-up, have her hair styled, and wear jewelry." The firm was found guilty of discriminating against Ms. Hopkins on the basis of her sex (gender). Consistent with the anecdotal evidence, field and laboratory research has shown a bias against unattractive individuals in performance appraisals and promotion decisions. Ross and Ferris (1981) I'

418

Stone, Stone, and Dipboye

rated the facial attractiveness of male employees in two public accounting firms from photographs. They then regressed salary, performance evaluations, and ratings of the likelihood of making partner on not only the ratings of attractiveness but also weight disparity, tenure, academic training, marital status, rated motivation, attitudes, and height. Weight disparity was strongly related to "relative performance" (i.e., rated performance relative to their cohort). The heavier the employees relative to their height, the more negative the ratings of their relative performance. Both height and facial attractiveness also were positively related to rated likelihood of making partner, but did not predict salary, achieved performance appraisals, or ratings of relative performance. Dickey-Bryant, Lautenschlager, Mendoza, and Abrahams (1986) rated the physical attractiveness of 75 cadets at a military academy from photographs in their yearbook. The more attractive cadets had higher grades and were more likely to be on active duty 12 years later than the less attractive cadets. Although the detected relationships were weak, it is remarkable that any effects were found given the homogeneity of appearance likely to be found in the typical military academy yearbook. These two field studies examined the attractiveness of men. Unfortunately, there are no field studies dealing with the effects of attractiveness on the appraisal and promotion of women, and the few existing laboratory experiments have yielded mixed results. Two experiments conducted by Heilman and Stopeck (1985a, 1985b) suggest that the "beauty is beastly" effect observed in selection decisions applies to performance appraisals and promotions as well. In one of these experiments, Heilman and Stopeck (1985a) had MBA students rate the performance of a male or female employee, who was depicted as either attractive or unattractive, and employed in either a managerial or clerical position. Consistent with their hypotheses, the results showed that attractive women were more likely to be "promoted" than unattractive women when they were in clerical positions, but the reverse was true when they were in managerial positions. No effects were observed for attractiveness when the stimulus person was a man. Heilman and Stopeck (1985b) conducted a second study in which they gave working men and women the work history of an assistant vice president of a midsized corporation who was either attractive or unattractive and either male or female. Attractive males were rated as more capable than unattractive males, but as in the earlier study, the reverse held true for the female stimulus persons. Attractive males were also seen as having more interpersonal integrity than unattractive males, but attractive females were seen as having less

Stigmas in Organizations

419

interpersonal integrity than unattractive females. Also, attractive males were viewed as more masculine than unattractive males and attractive females were viewed as more feminine than unattractive females. Success of an unattractive female in climbing the corporate hierarchy was more often attributed to her ability than it was for the attractive female. An opposite, albeit nonsignificant trend, occurred for the male stimulus person. Different results were found in two other laboratory experiments. Chung and Leung (1988) had male and female Chinese managers from firms in Hong Kong evaluate six hypothetical candidates for promotion. They found that performance was the major factor influencing the decision to promote. The only situation in which physical attractiveness enhanced the likelihood of promotion was when the candidate's performance was mediocre. There was no evidence that unattractive females were preferred over attractive females. Jackson (1983b) presented several vignettes to personnel consultants that described male and female employees who were either attractive or unattractive, and who described themselves in very masculine or feminine terms. The consultants indicated the decisions they would make with regard to promotion, training, job assignments, and a leave of absence for each of three types of jobs (a masculine, a feminine, and a sex-neutral position). No significant effects were found for physical attractiveness, but as in two other previously cited failures to show attractiveness effects (Greenwald, 1981; Jackson, 1983a), the attractiveness manipulation was weak, casting doubt on the validity of the results. The inconsistencies in the above described laboratory results may reflect the same stimulus sampling problems discussed in connection with the research on selection. In other words, the few stimuli used in these studies raises questions as to whether the effects observed in these few studies generalize to more diverse samples of stimuli. Future research should attempt to use larger and more representative samples of target stimuli.

Economic Return of Physical Attractiveness Consistent with evidence that physical attractiveness is related to obtaining employment and promotion, more attractive employees also have been found to enjoy more economic success in their careers. The research so far suggests that attractiveness is more consistently related to economic success for men than for women. Roszell, Kennedy and Grabb (1989) examined the relationship of attractiveness to income attainment for over 1,000 Canadians. Attractive persons earned higher annual salaries than less attractive

420

Stone, Stone, and Dipboye

persons. With each increase in rated attractiveness on a five-point scale of attractiveness, the 1981 annual income of the respondent increased by $1,988. After controlling for respondent sex, the sex composition of the job, and 1979 salary, this figure dropped to $1,046 but was still statistically significant. This relationship was found for men, older employees, and those engaged in male dominated occupations, but not for women, younger employees, and those in female dominated occupations. The authors speculated that because attractiveness is a scarcer resource for older employees, men, and individuals in male-dominated occupations, it is valued more by the employer when it is found in these groups. Two studies at the University of Pittsburgh have provided further evidence that good looks pay. In one study, Good,Olson, and Frieze (1986) used height, weight, and body mass (weight relative to height) as indicators of physical attractiveness. Over 2000 graduates of the MBA program at the University of Pittsburgh between 1973 and 1982 were surveyed. For men, weight but not height was found to predict starting salary and both height and weight predicted the current (1983) salary. However, neither of these variables predicted the starting and current salary of the women in the sample. For each one inch increase in height, salary of the men was $600 higher. Overweight men earned $4,000 less in salary than those of normal weight . In a more recent study, Frieze, Olson, and Russell (in press) asked a group of people with corporate management experience to rate the physical attractiveness of 700 hundred MBA graduates on a five point scale. The starting salaries of male graduates receiving the highest attractiveness rating were approximately $5,000 a year more than those receiving the lowest attractiveness rating. After five years, those receiving the highest rating earned $10,000 more than those receiving the lowest rating. Attractiveness had no impact on starting salaries of women, but was related to later salaries, although not as strongly as for men. For each increment in attractiveness on the five-point scale, women earned $2,000 more in salary five years later.

How Important Are Attractiveness Effects? A review of the research shows that unattractive persons are at a disadvantage with respect to employee selection, performance and promotion appraisals, and salary administration. The consistency with which a bias against unattractive persons has been found indicates that this is an important effect that deserves more attention than it has received so far in organizational research. Questions have been

Stigmas in Organizations

42 1

raised, however, as to whether attractiveness effects are "important. Let us examine three of the possible reasons for considering these effects trivial and the counterarguments. A frequent criticism is that because many of the studies involve college students, the results may not be generalizable to persons in actual decision making roles. However, a consideration of the studies cited here reveals that the bias against unattractive persons also has been found in realistic settings with real decision makers. This evidence suggests that the bias against unattractive males is more than a mere laboratory-based effect. The research also has been criticized for relying on facial attractiveness as indicated in photographs (Guion & Gibson, 1988). Although the research needs to move away from the reliance on photographs, several studies have shown that facial attractiveness as measured i n photographs is related to real-world employment judgments even when there is face-to-face interaction (Carroll, 1966; Ross & Fems, 1981; Shahani & Dipboye, 1990). The small amount of variance typically accounted for by attractiveness relative to more objective factors (e.g., grades, performance) also raises doubts as to whether attractiveness effects are important. For example, attractiveness accounted for only 6% of the variance in the Dipboye, Fromkin, and Wiback (1975) experiment. Effect sizes can be quite misleading, however, and probably should not be used as an indicator of importance. When subjects are asked to choose which one applicant they would hire, the effects of attractiveness appear much more pronounced (Dipboye, Fromkin & Wiback, 1975). Although attractiveness is unlikely to be the most important factor i n employment-related decisions, it may be the deciding factor when decision makers are rendering difficult choices among applicants or employees who possess similar levels of qualifications or performance. 'I

Alternative Sources of Attractiveness Bias The bias against unattractive individuals is likely to reflect cognitive, behavioral, and social factors. Cognitive models such as Heilman's (1984) lack of fit approach assume that unfair discrimination results primarily from biases in the individual rater's processing of information on the ratee. The judge infers attributes from physical appearance and personal characteristics and then evaluates the person on the basis of fit to the perceived requirements of the job. In jobs that are seen as requiring an attractive appearance, an attractive applicant has more of an advantage than in jobs where appearance is not a factor (Beehr & Gilmore, 1982). Attractiveness biases also

422

Stone, Stone, and Dipboye

may be more likely to occur on jobs that are strongly sex-typed (Cash, Gillen, & Bums, 1977). Other important elements of this information processing are the cognitive categories and implicit theories related to attractiveness. One view is that personnel decision makers believe that attractive persons are better on almost all dimensions (e.g.,. dominance, competence, social skill, etc.) and thus evaluate attractive persons as more qualified for almost any job (Hatfield & Sprecher, 1986). An alternative view is that the effects of attractiveness are mediated by sex-typing of the applicant (Cash et al., 1977; Heilman & Suruwatari, 1979). In other words, attractiveness has the effect of intensifying the masculinity of men and the femininity of women. As a consequence, attractiveness is an advantage for men applying for male-typed jobs and for women applying for female-typed jobs. Research can be found to support both these positions, but the previous findings are limited by the small and unrepresentative sampling of stimulus persons. In a study involving a more representative sample, Bieber and Dipboye (1988) examined the trait implications of attractiveness by having students rate 180 photographs (90 male and 90 female) on 17 personality attributes. Principal components analyses performed on ratings of the 90 female and 90 male pictures revealed the same three factors for both samples. One factor was characterized as dominance, another as competence, and a third sociability. The trait implications of attractiveness were found to differ, however, as a function of the stimulus person's sex. For the male pictures, rated physical attractiveness of the pictures was positively related to all three factors, with the strongest relationships found for dominance and sociability. For females a large positive relationship was observed on the sociability factor, a smaller positive relationship was observed on the competence factor, and a negative relationship was found on the dominance factor. Consistent with the Heilman and Saruwatari (1979) view, the findings suggest that attractiveness intensifies the sex-typing of women. Consistent with the "beauty is good" hypothesis of Hatfield and colleagues (Hatfield & Sprecher, 1986), the findings suggest that a variety of desirable traits are attributed to the male stimulus person, independent of the sextyping of the attributes. In addition to examining the trait implications of male and female attractiveness, future research also needs to delve into the specific physical characteristics that are associated with the traits inferred from male and female attractiveness. A potentially important feature to consider is the facial maturity of the stimulus person. Berry and McArthur (1986) found that facial immaturity (e.g:, small and rounded chin, high eyebrows, smooth skin, short nose) is posi-

Stigmas in Organizations

423

tively related to judged attractiveness of women but negatively related to judged attractiveness of men. They also have found that facially immature persons are seen as physically weaker, more naive, warmer, and more submissive. The cognitive approach to interpreting attractiveness bias has been that persons discriminate against attractive others because they stereotype them as deficient in the attributes required in the work. Noncognitive factors also can enter into attractiveness bias. Perhaps the most important of these are the affective reactions of the evaluator to the person evaluated. Physical attractiveness seems capable of evoking a wide range of emotional reactions, ranging from liking or disliking to sexual arousal. Graves and Powell (1988) recently presented evidence that liking for applicants influences judgments of objective qualifications. Raza and Carpenter (1987) used structural equation modeling to support a model in which perceived physical attractiveness influences interviewer judgments of likability, and likability, in turn, influences judgments of the applicant's intelligence and employability. Another source of attractiveness bias are pressures on individuals in decision making roles to conform to organizational and group norms. Organizations have images they wish to project, and those in the position of hiring or appraising performance may feel that an unattractive individual does not fit the appropriate corporate image (Rynes & Gerhart, 1990). In these cases the rater may have no personal bias against the unattractive person and may even be positively inclined toward him or her. Rather than reflecting the rater's personal beliefs, the negative evaluation represents a conformity to both real and imagined expectations of significant others. When there is an opportunity for the evaluator and the evaluatee to interact, differences in supervisory or interviewer evaluations also can reflect self-fulfilling prophecies (Snyder, n n k e , & Berscheid, 1977). Persons who are perceived to be unattractive may act in a socially unskilled manner because of the clumsiness and hesitance with which an evaluator interacts with them. This treatment may evoke behaviors on the part of the unattractive person that fulfill the original expectations of the evaluator. Thus, unattractive employees may be seen as lacking in social skill and may actually act this way because of the way the evaluator interacts with them. A kernel of truth also may be associated with biases against the unattractive individual. Unattractive persons actually may have less skill and ability, perhaps as the result of a lifelong history of being discriminated against. It is possible that appearance is an indicator of how much the applicant really wants the job or desires success. Thus, if a person comes to an interview i n a blue suit with white

424

Stone, Stone, and Dipboye

socks, it might be inferred that he or she does not care about the job enough to dress appropriately. Using this reasoning one might even argue that an individual who really cared would correct all deficiencies in appearance through proper grooming, hairstyling, make-up, dress, diet, exercise, and in extreme cases, cosmetic surgery. In support of the notion that appearance might actually reflect important job-related attributes, a recent large-scale project aimed at assessing managerial talent at Sears found that interviewer ratings of appearance were among the strongest and most consistent predictors of supervisor performance appraisals and training evaluations of salespeople (Bentz, 1985). We cannot discount the possibility, however, that the validities observed in this study simply reflected the fact that both interviewers and those providing the performance ratings were vulnerable to the same attractiveness biases. Nevertheless, there is experimental evidence that physically attractive persons are more socially skilled than less attractive persons (Goldman & Lewis, 1977). Also, the attractiveness of sales persons has been shown to bias customers such that attractive salespersons are more effective in selling than less attractive salespersons (Reingen, Gresham, & Kernan, 1980).

Solutions Although attractiveness may involve real differences in social skill and motivation, attractiveness seems at best a crude and unreliable means of assessing these attributes. Moreover, the experimental research has shown that there is bias against unattractive persons even when they have equivalent levels of performance and qualifications. What are the possible solutions to this problem? The best approach is perhaps to adopt explicit, job-related standards for hiring, promoting, and rewarding employees. If organizations carefully examine the work performed, set specific performance standards and develop behaviorally based measures of performance and potential, then attractiveness should have less of an effect on evaluation. Another possible solution is to train raters to avoid the influence of facial attractiveness and other physical attractiveness factors. Careful attention also should be given to choosing evaluators (e.g., college recruiters) who are not prone to attractiveness biases. Cash and Kilcullen (1985) found that bias in favor of attractive applicants was stronger for raters who rated themselves as masculine or feminine than for those who perceived themselves in more androgynous terms. Snyder, Berscheid, and Matwychuk (1988) found that evaluators who are low on the personality trait of self-monitoring were influenced more by the fit of the applicant's attributes to the

Stigmas in Organizations

425

job and less by appearance than those who were high on self-monitoring.

Conclusions Physical attractiveness as an influence on personnel decisions has been largely ignored by industrial and organizational psychologists and researchers in closely allied disciplines. Yet, we suspect that when difficult, personnel-related decisions have to be made by interviewers, managers, and others in organizations, the physical attractiveness of the individual (e.g., applicant, ratee, trainee) may be the deciding factor. In order to understand its influence, much more research is needed in both laboratory and field settings on the cognitive, affective, social, and behavioral factors contributing to the physical attractiveness bias. Perhaps the findings of such research can eventually convince employers that a well-managed organization, providing a good product or service, can survive and profit regardless of the superficial appearance of its employees.

HANDICAPS AS A STIGMA IN ORGANIZATIONS It has recently been reported that there are over 43,000,000 disabled people in the United States (Americans with Disabilities Act, 1990). Regrettably, these individuals have been confronted with numerous obstacles in their attempts to gain employment, among these being (a) physical arrangements that preclude them from entering workplaces and accessing items needed to do their work, and (b) attitudinal barriers that are based upon stereotypes, invalid expectancies, and other types of perceptual biases. In both cases, the consequence is UD. The employment-related difficulties experienced by the handicapped have serious consequences for individuals, organizations, and society as a whole. For example, Levitan and Thggart (1977) noted that handicapped individuals experience higher rates of unemployment, receive lower levels of compensation, and are more likely to suffer loss of work-status and independence than non-handicapped persons. Specifically, they reported that 40% of the disabled are typically employed compared to at least 75 % of the able-bodied. These authors also reported that on average, disabled workers earn only 70% of the earnings of non-disabled workers. Thus, both relatives of the disabled and society must supplement their reduced earnings. Moreover, they contend that employment problems experienced by the disabled are often compounded by other factors such as age, race, and gender.

426

Stone, Stone, and Dipboye

While there is little doubt that some of the employment-related problems experienced by the handicapped are due to deficiencies in job-related abilities, it is also clear that the handicapped have valuable and useful skills that have been underutilized by employers. The primary reason for this is that employers have long resisted hiring individuals with disabilities. This resistance can be attributed to several factors including employers' firmly held beliefs that disabled employees (a) pose safety risks, (b) increase health-related costs, (c) have higher rates of absenteeism, (d) need more help on the job, and (e) have lower performance levels than non-disabled employees. Interestingly, however, evaluation data from organizations that employ handicapped persons show that these concerns are unfounded. More specifically, a review of 90 evaluation studies by Greenwood and Johnson (1987) revealed that disabled employees (a) consistently show a record of high quality performance, (b) do not have higher absenteeism or turnover rates than non-disabled workers, (c) are not less flexible in job assignments than non-disabled employees, (d) actually have better safety records than non-disabled employees, and (e) do not place greater demands on supervisors than workers without disabilities. Given these and other similar findings, analysts have long argued that many of the employment-related problems experienced by handicapped individuals are the result of biases, stereotypes, and UD. Arguments such as those just considered contributed to support for the passage of the Rehabilitation Act of 1973 and the recently enacted Americans with Disabilities Act of 1990. While the passage of legislation will certainly help overcome some of the access-related UD faced by the disabled, social psychological theory and research (cf. Jones et al., 1984) suggests that disabled employees may encounter recurring treatment-related UD in organizations. Given the importance of the problems faced by disabled individuals in organizations, the purpose of this section of our chapter is to review the theoretical and empirical research on the stigmatizing effect of handicaps in organizations, and to discuss a number of potential strategies that can be used to overcome biases against handicapped persons.

Definition of Handicap Before considering the biasing effects of handicaps in employment contexts it is important to define the term handicap. For the purpose of this chapter we consider a handicap to be any physical or psychological condition that impairs an individual's ability to perform one or more behaviors or activities.

Stigmas in Organizations

427

Explanations for the Stigmatizing Effects of Handicaps Since Goffman's (1963) provocative work on stigmas, a number of interesting theoretical views have been offered to expand Goffman's views and more fully explain why handicapped individuals are denigrated in social interactions (Lerner, 1980; Katz, 1979; Hahn, 1988). Among these explanations are what might be regarded as (a) the attractiveness point of view, (b) the blame-the-victim interpretation, and (c) the strain in social interaction explanation. Each of these viewpoints is described in the following section.

The Attractiveness Point of View. One of the most basic yet infrequently discussed reasons for handicapped individuals being denigrated is that their appearance often deviates from the usual human form or the form others label as attractive in our society. This discrepancy leads handicapped individuals to be seen as less appealing or attractive than non-handicapped persons. Thus, it is also likely that they will be liked less (Byrne, London & Reeves, 1968), and thought to have less desirable characteristics (Benson, Karabenick & Lerner, 1976) than non-handicapped individuals. Consequently, nonhandicapped people may resist working or interacting with handicapped individuals simply because they find them unattractive. In some cases, non-handicapped individuals indicate that the mere sight of a disability (e.g., a facial disfigurement, limb amputation) is repulsive and makes them feel nauseous or queasy (Siller, cited in Hahn, 1988). This should come as no surprise because aversion to physical aberrations is a common response among animals, including humans, and rejection of those who are malformed may be a learned response to insure survival of the fittest or strongest of the species. Despite the obvious effect that appearance of the handicapped may have on others' reactions, little research has directly assessed the effects of physical attractiveness on responses to the disabled. Two studies that did assess the effects of handicaps on social distance, however, provided indirect evidence that attractiveness may be an important factor in explaining reactions to individuals with disabilities (Rumsey, Bull, Gahagan, 1982; Wolfgang & Wolfgang, 1968, cited in Rumsey et al., 1982). The study by Wolfgang and Wolfgang found that type of handicap affects the distance non-handicapped individuals stand from those who are handicapped. More specifically, individuals typically stand close to others who are normal and those with temporary disabilities (e.g.? a broken leg), an intermediate distance from people with physical disabilities (e.g., an amputated leg) and a greater distance from people who are obese. Likewise, the study by Rumsey et al. found that pedestrians stand

428

Stone, Stone, and Dipboye

farther from an individual with a facial disfigurement than a person with no disfigurement. Interestingly, i n both of these studies the level of attractiveness of the handicapped person appeared to influence the physical distance maintained by others.

Blame-the-Victim Explanations. At least two perspectives (i.e., the just world hypothesis, and ambivalence-response amplification theory) maintain that we derogate the disabled and others (e.g., blacks, the poor, the homeless) as a means of defending our belief systems and ourselves against harm. More specifically, the just world hypothesis (Lerner, 1980; Novak & Lerner, 1968) suggests that individuals want to believe in a world that is good and just, a world where people get what they deserve, and good things happen to good people and bad things happen to bad people. The presence of a disabled person threatens such a belief system because the disabled person did nothing to deserve the suffering or punishment. As a result, individuals employ a number of defensive strategies to eliminate the threat to their belief in a just world including (a) reinterpreting the cause of the disability (e.g., inferring that the victim must have done something wrong to deserve the problem) and (b) making negative inferences about the character of the disabled person (e.g., assuming the disabled person is inferior or "had it coming"). As a consequence, the disabled person becomes the target of both blame and denigration. The ambivalence-response amplification explanation (Katz, 1979) also suggests that we derogate the disabled as a means of defending the self from harm. Specifically, this viewpoint suggests that contact with a disabled person arouses ambivalent feelings and such feelings are threatening. For example, when an individual interacts with a disabled person helshe may simultaneously experience feelings of both hostility and sympathy toward the disabled person. These ambivalent feelings threaten the non-disabled person's sense of esteem and his or her potential to maintain feelings of being a humane and reasonable person. According to this theory, the threat gives rise to guilt and individuals reduce the feelings of threat and guilt by engaging in a variety of threat-reducing behaviors, including extremely negative behaviors toward the disabled person. For example, studies by Katz and his colleagues (Katz, Glass, Lucido & Farber, 1977, 1979) have shown that after individuals have been induced to harm target others (e.g., administer loud noxious stimuli) they are more likely to denigrate or lower the worth of handicapped targets than non-handicapped targets. Interestingly, these studies suggest that ambivalence causes individuals to experience more guilt

Stigmas in Organizations

429

from harming disabled than non-disabled individuals and thus to indulge in greater denigration of disabled victims.

Strain in Social Interaction. A final explanation for the denigration of handicapped individuals is that the presence of a disabled person creates strain, uncertainty, and discomfort in social interactions (Kleck, Ono & Hastorf, 1966; Kleck, 1968; Hastorf, Wildfogel & Cassman, 1979; Belgrave & Mills, 1981). This discomfort stems from several sources. First, non-handicapped persons are uncertain about the behaviors that are expected and appropriate in interactions with handicapped individuals. In particular, non-handicapped persons are often faced with conflicting norms about how to treat the handicapped. Conflicts arise between the norm that one should help the handicapped and treat them kindly, and the opposing norm that one should treat the handicapped like anyone else so as not to appear condescending (Kleck et al., 1966.) Feelings of discomfort also emerge because a physical deformity is a novel stimulus and individuals are simultaneously motivated to observe something novel and to avoid staring at a person with a deformity. Thus, conflict occurs between the desire to stare at a novel stimulus, and the fear of violating the social norm against staring (Langer, Fiske, Taylor, & Chanowitz, 1976). A third reason that non-handicapped individuals feel discomfort when interacting with the handicapped is that they fear they might offend handicapped people during normal conversation. Non-handicapped individuals often assume that the handicapped are highly sensitive or emotional about their disabilities. As a result, individuals who are not handicapped are apprehensive about hurting a handicapped person's feelings during social interactions. Interestingly, social psychological studies have found that both nonhandicapped and handicapped individuals experience strain or discomfort when interacting with one another (Belgrave, 1984: Mills, Belgrave & Boyer, 1984).

Research on the Effects of Handicaps In spite of the numerous problems faced by handicapped people in employment contexts, only a few empirical studies in industrial and organizational psychology and the closely related field of personnel and human resources management have specifically focused on the effects of handicaps on personnel decision-making (Krefting & Brief, 1976; Rose & Brief, 1979; C. Stone & Sawatzki, 1980; D. Stone, 1986; E. Stone, Shetzer & Eggleston, 1986; Czajka & DeNisi, 1988; Drehmer & Bordieri, 1986). However, a considerable amount of research in social psychology (cf. Jones et al., 1984) has

430

Stone, Stone, and Dipboye

examined the influence of handicaps or stigmas on social interaction, and some research in rehabilitation psychology has focused on this issue. Thus, social psychological and rehabilitation research provides important indirect evidence on the effects that handicaps may have on personnel-related decisions and evaluations. Therefore, examples of relevant research from industrial and organizational, social, and, rehabilitation psychology are reviewed below. The existing research on reactions to handicapped individuals has focused on three basic questions: (a) Do handicaps have a negative or biasing effect on personnel decisions?, (b) Do factors associated with the handicap (e.g., type, origin, concealability, peril) influence the same decisions?, and (c) What strategies can be used to overcome the biases faced by handicapped individuals in the employment context? Research related to each of these questions is reviewed in the paragraphs that follow. Do Handicaps Have a Negative Impact on Personnel Decisions? As previously indicated, it is commonly thought that handicaps have a negative impact on personnel decision making (e.g., selection and promotion decisions, performance evaluations). However, results of research on this issue are quite mixed. Some studies have found that handicapped job applicants or employees are rated more negatively than those who are not handicapped. For example, a study by Johnson and Heal (1976) found that a job applicant confined to a wheelchair was offered fewer interviews than a non-handicapped applicant. In addition, results of a study by C. Stone and Sawatzki (1980) indicated that a job applicant with a psychiatric disability was rated lower in terms of job suitability than a similar applicant with no disability. Further, a series of interesting social psychological studies by Kleck and his colleagues (Kleck, 1968, 1969; Kleck, Ono & Hastorf, 1966; Snyder, Kleck, Strenta & Mentzer, 1979) revealed that publicly expressed attitudes toward those with disabilities were seemingly favorable, but, nonverbalized behaviors were often more negative. Specifically, results of these studies showed that when normal individuals interacted with an individual who had an amputated left leg (as opposed to a non-handicapped person) they (a) exhibited greater motoric inhibition (Kleck, 1968), (b) terminated the interaction sooner (Kleck et al., 1966), (c) exhibited a greater interpersonal distance (Kleck, 1969), and (d) were more likely to avoid sitting next to the handicapped person when their motives could be concealed (Snyder et al., 1979). Moreover, results of other social psychological studies show that compared to normals, individuals with mental disabilities are viewed as more unusual (Piner & Kahle, 1984), and treated more harshly (e.g., shocked more intensely;

Stigmas in Organizations

43 1

Farina, Holland & Ring, 1966). Likewise, a study by Russel, L e n d , Spicer, Miller, Albrecht, and Rose (1985) found that a physically disabled individual received more negative evaluations than a nondisabled person for comparable performance on a task. Other research, however, has not found differences between ratings of handicapped and non-handicapped individuals. For example, research by Krefting and Brief (1976) found that a paraplegic job applicant was not rated differently in terms of ability, or potential absenteeism than a non-handicapped applicant. Likewise, results of a study by Farina, Felner and Boudreau (1973) indicated that applicants with a history of mental illness were not rated differently than those who had no such history. Note, however, that the findings of this study did reveal that a job applicant who displayed tenseness or anxiety was less likely to be selected than one who was not tense. Results of still other studies have indicated that handicapped individuals are rated or treated more favorably than individuals without handicaps. Specifically, in addition to the findings noted above, the study by Krefting and Brief (1976) found that a paraplegic job applicant was rated higher in terms of motivation, and length of service potential than a similar non-handicapped applicant. Further, a study by Czajka and DeNisi (1988) that was designed to assess the effects of handicaps on performance ratings found that when performance standards were not specified, disabled carpenters were rated more favorably than non-disabled carpenters. However, when performance standards were specified, there were no differences in ratings given to disabled and non-disabled workers. A similar social psychological study by Hastorf, Northcraft and Picciotto (1979) found that a low performing handicapped person received more positive feedback than an able-bodied individual. Moreover, results of a study by Carver, Glass, and Katz (1978) using the bogus pipeline technique (i.e., a technique used to convince participants that the experimenter has access to their actual feelings), indicated that handicapped interviewers were rated more positively than interviewers without handicaps. Finally, a study by Farina, Sherman, and Allen (1968) found that individuals who were confined to a wheelchair were treated less harshly (i.e., shocked for a shorter duration) than normal individuals. Given the inconsistencies in the research findings noted above, it appears that there is no simple answer to the question about whether handicaps influence job-related decisions. Not surprisingly, the research seems to suggest that reactions to handicaps depends on a number of factors including the nature of the handicap, characteristics of the rater (e.g., leniency bias, stereotypes of the disabled,

432

Stone, Stone, and Dipboye

tolerance for ambiguity), characteristics of the handicapped person (e.g., gender, race, attractiveness), and structure of the rating task (e.g., clarity of performance dimensions).

Do Factors Associated with the Handicap Influence Personnel Decision-making? As previously noted, it is often assumed that factors associated with handicaps (e.g., type, concealability, course of condition, disruptiveness, origin, aesthetic qualities, peril) influence reactions to individuals with handicap-related stigmas (Jones et al., 1984). In fact, most of the early research on handicaps focused on this issue (e.g., Richardson, Goodman, Hastorf, & Dornbusch, 1961; Rickard, Triandis, & Patterson; 1963; Tringo, 1970), and attempted to establish hierarchies or rankings of reactions to different types of handicaps. More specifically, in a study using drawings of children having various types of handicaps as stimuli, Richardson et al. (1961) found that children most preferred interacting with children with no handicap, followed by children with physical disabilities, children with a missing left hand, children with facial disfigurements, and children who were obese. In addition, results of a study by Rickard et al. (1963) revealed that subjects showed the most negative reactions to epileptics, ex-prisoners, and former mental patients. The study by Tringo (1970) indicated that individuals have a distinct hierarchy of attitudes toward disabilities. Specifically, attitudes were most positive toward "physical" disabilities (i.e., ulcers, arthritis, asthma, diabetes, heart disease, amputation), less positive toward ''sensory" disabilities (i.e., blindness, deafness, stroke, cancer, old-age), even less positive toward "neurological" disabilities (i.e., epilepsy, cerebral palsy), and least positive toward those such disabilities as mental retardation, alcoholism, and mental illness. Consistent with Tringo's results, subsequent research (Florian, 1978; C. Stone, & Sawatzki, 1980; Bordieri & Drehmer, 1986; D. Stone, 1986; E. Stone, Shetzer, & Eggleston, 1986) has shown that neurological and psychological conditions evoke more negative reactions from others than physical conditions. For example, Florian (1978) found that employers viewed amputees as more employable than former mental patients. In addition, research by C. Stone, & Sawatzki (1980), Bordieri & Drehmer (1986) and E. Stone et al. (1986) indicated that job applicants with mental problems (e.g., depression) were rated less favorably than those with physical problems (e.g., diabetes, paraplegia). Likewise, research by D. Stone (1986) revealed that an applicant with petit mal epilepsy was rated as less suitable for a job than an applicant with arthritis. As is evident from the research presented thus far, a number of characteristics associated with the handicap and the observer seem to affect attitudes

Stigmas in Organizations

433

toward handicapped individuals including (a) the visibility of the handicap, (b) the degree to which the disability leads to uncertainty or the inability to predict a person’s behavior (e.g., mental illness, mental retardation), and (c) the extent to which the disability is perceived to b e caused b y a weakness of character ( e . g . , alcoholism). Although some research has found that the type of handicap influences how others react to the disabled, results of several studies (Rose & Brief, 1979; Drehmer & Bordieri, 1986) have found no relationship between type of handicap and job suitability ratings. For example, Rose and Brief (1979) found that job applicants with epilepsy were not rated differently than an applicant with an amputated foot. Further, Drehmer and Bordieri (1986) found that type of handicap (i.e., emotional disorder, amputation, auditory problems, paraplegia) did not affect reactions to job applicants. However, the same study found that the cause (origin) of the disability was an important determinant of reactions to disabled individuals. Specifically, results revealed that regardless of the type of handicap, applicants who were seen as personally responsible for their problems were evaluated more negatively than those with disabilities caused by factors beyond their control. Furthermore, the study found that the cause (origin) of the handicap influenced hiring recommendations. In particular, applicants whose disabilities were attributed to external causes (i.e.! disease) received higher recommendations than those whose handicaps were attributed to internal causes (i.e., a motorcycle accident). Similarly, research on obesity by Maddox, Back, and Liederman (1968) and acquired immune deficiency syndrome (AIDS) by Triplet and Sugarman (1987) indicated that these two conditions evoke more negative reactions than other physical problems because individuals feel that those who are obese or have AIDS are responsible for their conditions. In particular, a survey by Maddox et al. found that negative reactions to obesity result from the fact that it is attributed to weakness of character rather than physical or metabolic abnormalities. Likewise, Triplet and Sugarman found that AIDS patients were seen as more personally responsible for their illnesses and as a consequence were rated as more undesirable than individuals with other equally contagious diseases (e.g., herpes, hepatitis). It is important to note that these same authors maintain that the perceived responsibility is only one of a number of complex factors that affects reactions to individuals with AIDS. Other important factors include the general uncertainty about the disease and the way it is transmitted, and prejudice against homosexuals (Lee,one of the groups most affected by the disease.) Finally, an interesting study on reactions to cancer and a fictitious disease by Meyerowitz,

434

Stone, Stone, and Dipboye

Williams and Gessner (1987) found that attitudes toward disease were more influenced by controllability (i.e., preventability and treatability) than by the type of disease. Specifically, when a disease was perceived as controllable it was evaluated less negatively than when it is viewed as uncontrollable. Perhaps those diseases that are perceived as uncontrollable cause individuals to feel especially vulnerable or susceptible to death (i.e,, increased peril), and thus, evoke strong, negative reactions. W e n together, the research just reviewed suggests that reactions to handicaps do not simply depend on the type of handicap, but rather a complex set of factors. These factors include but are not limited to the perceived responsibility or origin of the condition, the perceived uncertainty associated with the course or controllability of the condition, and the degree to which the handicap impedes others' abilities to predict the behavior of handicapped persons.

What Strategies Can Be Used to Overcome Biases Faced by Handicapped Individuals? Although most of the research on handicaps has focused on establishing evidence of biases and understanding the factors that influence reactions to handicaps, some research has taken the issue a step further and examined the means of overcoming biases against handicapped persons. A primary motivation for this type of research is the assumption that even though handicapped persons may be able to overcome physical and legal barriers that impede their access to organizations, many psychosocial barriers would remain that could limit their acceptance in organizations. Clearly, this research has important implications for organizational efforts designed to eliminate UD against the handicapped (e.g., affirmative action programs) and increase their assimilation in organizations (e.g., training). As a consequence, results of research on three strategies (i.e., training, increased contact, acknowledgement of the handicap) used to overcome biases against the handicapped are reviewed below. Training. As noted previously, research suggests that lack of knowledge or uncertainty associated with the handicap may be two factors that cause non-handicapped persons to react negatively to the handicapped. To deal with these problems, psychologists have attempted to alter attitudes toward the disabled through various types of training programs designed to provide information about disabled people and their conditions (cf. Schneider & Anderson, 1980). Some of these training programs are purely informational, others use persuasive techniques, and still others are more experiential and involve role playing a handicapped person. Results of research on

Stigmas in Organizations

435

strictly informational and persuasive techniques have typically been quite mixed, and have not conclusively shown that such techniques are effective in altering attitudes (Schneider & Anderson, 1980). However, more affect-oriented approaches, such as role plays designed to increase empathy, have been shown to have a positive and lasting effect on attitudes toward the disabled (Clore & Jeffrey, 1972).

Increased Contact. A second widely used technique to alter attitudes toward the disabled is the "increased contact" approach. Quite simply, this approach increases the social, living, or work-related interactions between handicapped and non-handicapped persons. The assumption behind this strategy is that increased contact with handicapped individuals will increase knowledge about them, and decrease social strain or uncertainty on the part of those who are not disabled. However, research designed to assess the effects of increased contact with the handicapped has not shown consistent results. In some cases, studies have found that level of contact influences both attitudes and actions toward the disabled (Schneider & Anderson, 1980). However, results of other studies have indicated that increased contact has either no effects or negative effects on attitudes toward the disabled (cf. Evans, 1970). Unfortunately, this research suggests that strategies designed to simply expand the number of handicapped individuals in the workplace (e.g., affirmative action) may not lead to increased acceptance of them by the nonhandicapped. Acknowledging Handicaps. As noted previously, one reason that handicapped individuals are treated differently than others is that their presence often creates strain, uncertainty, or discomfort in social interactions. As a result, a number of researchers (Evans, 1970; Hastorf et al., 1979; Belgrave & Mills, 1981; D. Stone, 1986; E. Stone et al., 1986) have argued that handicapped individuals might overcome biases by using interpersonal or impression management tactics designed to reduce others' discomfort and thus facilitate interaction. One such tactic is acknowledging or disclosing information about the handicap in social or work-related contexts. Research on acknowledging one's handicap (Evans, 1970; Hastorf et al., 1979) has typically shown that handicapped individuals who acknowledge or mention their handicap are viewed more favorably than those who do not acknowledge their handicap. However, some recent research on this issue in the personnel selection context (D. Stone, 1986; E. Stone et al., 1986; Tagalakis, Amsel & Fichten, 1988) suggests that the effectiveness of a disclosure tactic may

436

Stone, Sfone,and Dipbuye

depend on a number of factors, including the nature of information disclosed, the timing of the disclosure in the selection process, and the degree to which the disclosure is seen as voluntary. Specifically, results of a study by D. Stone showed that disclosure of a physical handicap (i.e., arthritis) at the beginning of the selection process resulted in more positive suitability ratings of an applicant than nondisclosure. However, disclosure of a neurological problem (i.e., epilepsy) did not have a similar effect on such ratings. In addition, results of a study by 'hgalakis et al. on job interview strategies used by disabled persons found that physically handicapped applicants who did not disclose their disabilities over the phone were evaluated more favorably than those who did. Note, however, that the same persons were less likely to be selected for a job after an on-site interview revealed they were disabled. Further, results a study by E. Stone et al. indicated that the voluntariness of a revelation about a handicap influenced ratings of job applicants. In particular, applicants who voluntarily revealed information about their handicap were perceived as having greater motivation, performance potential, and promotion potential than those who reluctantly revealed the same information. Interestingly, however, voluntariness of the revelation did not affect willingness to hire the applicant. Researchers have identified three other tactics that handicapped individuals might use to overcome biases, i.e:, requesting help, mentioning their disability following some incident involving the disability, and revealing information that handicapped individuals are interested in normal activities (Belgrave & Mills, 1981; Belgrave, 1984). Belgrave and Mills argued that strategies involving either a request for help or the mention of a disability after an incident may be more effective than simply acknowledging a handicap because these strategies provide a more natural context for discussing the handicap and showing others that a handicapped person is emotionally detached from the handicap. Results of studies on the use of these tactics show that mentioning a disability after a request for help or after an incident involving the disability is more effective in increasing social interaction than not mentioning the disability (Belgrave & Mills, 1981; Mills, Belgrave, & Boyer, 1984). Further, Belgrave (1984) maintained that one tactic handicapped individuals might use to overcome biases is to convince others that they are interested in similar things as non-handicapped individuals and are not preoccupied with their handicaps. Results of a study designed to assess the effectiveness of this tactic showed that when a handicapped person was thought to share typical interests with others, the handicapped individual was viewed more positively than when no such interests were shared.

Stigmas in Organizations

437

It appears that a variety of information management or impression management strategies may be useful in reducing the strain experienced by non-handicapped persons when interacting with those who are handicapped. However, more research is needed to assess the effects of these and other strategies on both access- and treatment-related criteria in organizational contexts.

Summary It is clear from the preceding review that handicapped individuals have often been excluded, devalued, and underutilized in organizational settings. Theorists (e.g., Goffman, 1963; Lerner, 1980; Katz, 1979; Hastorf et al., 1979) have argued that this treatment can be attributed to a number of factors including the fact that handicapped individuals (a) are often viewed as less attractive than non-handicapped persons, (b) threaten beliefs in a just world, (c) evoke guilt and ambivalent feelings among non-handicapped persons, and (d) may create strains i n social interactions with those who are not handicapped. Despite the problems faced by handicapped individuals in the workplace, little research in industrial and organizational psychology has focused on these problems, and the results of the same research have been quite mixed. In some cases, handicaps have been found to have a negative impact on personnel decisions, in other cases handicaps appear to have no impact, and in still other cases handicaps have had a positive effect on personnel judgments. It is unclear, however, whether these inconsistencies in research results are due to research-related artifacts (e.g., social desirability response biases, impression management on the part of subjects), factors associated with the handicap, or aspects of the rating process. To address the latter two issues, some research has examined the effects that psychological factors have on reactions to handicapped persons. This same research has found that a complex set of factors influences attitudes toward those with handicaps, including perceived responsibility for the condition, uncertainty about the course of the disability, and inability to predict the behavior of those with disabilities. W e n together, the research reviewed provides some insight into the psychological factors that influence attitudes toward those with disabilities. However, the same research has neglected a number of seemingly important issues. First, little research has focused on the corrosive impact that biases and stereotypes may have on the handicapped individual's feelings of self-worth, status, and independence in organizations (Grealish & Salomone, 1986). As noted previously, handicapped individuals often experience considerable frustration

438

Stone, Stone, and Dipboye

and humiliation in attempting to "blend in" and behave in accord with extant organizational expectations. Therefore, research is needed on the effects that this frustration may have on handicapped employees, their coworkers, their supervisors, and their employing organizations as a whole. Criterion variables that might be considered by such studies include stress, quality of social interactions, self-esteem of stigmatized individuals, and conflict between members of in-groups and out-groups. Second, much of the research on handicaps has attempted to heighten awareness concerning the UD experienced by handicapped individuals in organizations. Such research provides the all important impetus for changing unfairly discriminatory personnel policies and practices (e.g., selection and performance appraisal practices) in the workplace. However, the same research has neglected broader forms of organizational change that may help handicapped individuals bridge the equality gap in organizations. For example, research is needed to assess the effects that job site modifications, changes in bureaucratic systems, job redesign and reshaping of work schedules and assignments may have on handicapped individuals' work-related experiences.

DEALING WITH STIGMAS IN ORGANIZATIONAL CONTEXTS It should be obvious that stigmas that result from race, physical unattractiveness, and various types of handicaps have nontrivial effects on a variety of important access- and treatment-related criteria. In view of this, we briefly consider a number of strategies that might be used to deal with the problems caused by the types of stigmas considered in this chapter.

Changing Physical Aspects of the Organization Organizations can change the physical surroundings of work so as to make it possible for physically handicapped individuals to reach job sites and access equipment and materials needed to do their work. As a result of the passage of the 1990 Americans with Disabilities Act, changes of this nature will probably be common in large organizations, especially those that are direct or indirect suppliers of goods and services to the federal government and its agencies. The same types of changes are less likely in smaller organizations, however. This will limit the range of organizations for which physically handicapped individuals might work.

Stigmas in Organizations

439

Three other points about physical arrangements merit consideration. First, providing handicapped individuals with access to organizations and job sites may have a beneficial side-effect. That is, it will serve to promote interaction between the handicapped and non-handicapped. To the extent that this increased contact changes the views that non-handicapped individuals have of their handicapped co-workers, handicap-related problems should decrease over time. Second, although changes in the physical arrangements of organizations may allow individuals with physical disabilities to hold jobs from which they were previously excluded, there is no guarantee that this alone will lead to them being fairly treated in organizational contexts. In order for UD problems to decrease there will have to be changes in non-handicapped individuals' cognitions about and attitudes toward handicapped individuals. Third, improving physical arrangements for handicapped individuals will not serve to decrease UD problems for individuals who bear other types of stigmas. For example, allowing handicapped individuals access to organizations is unlikely to alter UD against individuals who bear stigmas that are based upon race or physical attractiveness.

Changing Social Aspects of the Organization The elimination of UD against stigmatized individuals hinges on the modification of the social relationships that exist in many work organizations. More specifically, it requires that stigmatized individuals not be relegated to out-group status. As long as individuals who are atypical in some way (e.g., race, handicaps, physical attractiveness) are barred from having in-group status, there will be no lessening of the destructive conflict that typically characterizes relationships between members of in-groups and out-groups (cf. Brewer & Kramer, 1985; Miller & Turnbull, 1986; Sutherland, 1990). Interactions between stigmatized individuals and normals that stress cooperative relationships, the pursuit of superordinate goals, and similar ways of promoting productive contact may do much to reduce negative views about and attitudes toward stigmatized individuals (cf. Weigel & Howes, 1985, p. 134). One benefit of increased interaction between normals and stigmatized individuals is that the greater exposure to stigmatized individuals may provide nonstigmatized individuals with information that is powerful enough to override existing stereotypes (cf. Weigel & Howes, 1985, p. 134). Another approach to dealing with stigma-based UD is to foster social norms in organizations that are promotive of social justice and antagonistic toward bigotry, racism, sexism, and other forms of

440

Stone, Stone, and Dipboye

prejudice. It is especially important that top-level managers in organizations show their support for such views (cf. Weigel & Howes, 1985, p. 134).

Changing Organizational Practices and Processes UD against stigmatized individuals may also be reduced through changes in a number of personnel-related practices, processes, and procedures. As suggested earlier, organizations might change their recruiting practices, selection systems, performance appraisal instruments and processes, placement procedures, promotion processes, and so forth in ways that serve to reduce the extent to which they unfairly discriminate against stigmatized persons. For example, it may be possible to eliminate some biases in the employment interview by increasing the extent to which they use a structured format (cf. Campion, Pursell, & Brown, 1988; Latham, Saari, Pursell, & Campion, 1980).

Changing the Stigmatized Individual Goffman (1963) noted that stigmatized individuals respond to their plight in a number of ways, including (a) correcting the basis of the failing (e.g., plastic surgery to deal with appearance-related flaws), (b) indirectly attempting to correct the stigma (e.g.,. learning to behave or perform similar to normals), and (c) breaking from reality by adopting an "unconventional view" of the stigma. Unfortunately, these strategies appear to be of only limited value for the types of stigmas considered in this chapter. More specifically, with respect to the first strategy proposed, i.e., correcting the basis of the failing, it deserves noting that although clothing, makeup, and other props can be used to modify the appearance of a highly unattractive individual, the basic appearance of the person will remain unchanged. Moreover, plastic surgery is expensive and can deal with only a limited range of flaws. Finally, for individuals who are stigmatized by race or physical handicaps it is generally not possible to alter meaningfully the basis of their "failing." For some types of stigmas it may prove worthwhile for individuals to pursue the second strategy mentioned by Goffman (1963), i.e., attempt to act like normals. At minimum this may provide them with access to roles from which they are typically excluded. Unfortunately, little empirical evidence is available on the effectiveness of this approach in organizational settings. The third approach mentioned by Goffman (1963), i.e., developing an unconventional view of the stigma, would appear to be

Stigmas in Organizations

44 1

of minimal utility to stigmatized individuals in work organizations. The reason for this is that their treatment is more a function of the way that others (normals) perceive them than it is of their self-perceptions.

Legal Strategies Goffman noted that stigmatized individuals often band together, have their own publications, set up action groups to deal with problems, serve leaders of movements, and use lobbyists to fight for their rights. Clearly, such actions have met with some measure of success. For example, personnel practices of many organizations have changed as a result of legislation such as Title VII of the Civil Rights Act of 1964, and will undoubtedly change even more in response the recent passage of the Americans with Disabilities Act (1990). In the interest of correcting problems caused by UD against members of various out-groups (blacks, women, older workers), many organizations have developed and implemented affirmative action (AA) plans. Such plans have met with varying levels of success. Moreover, AA plans have resulted in a number of legal challenges from individuals (i.e., members of majority groups) who perceived such plans to be against their self-interests. On the basis of rulings in several such cases, it appears that the defensibility of AA plans is a function of several factors. Kleiman and Faley (1988) reviewed six court cases in which AA plans were challenged by nonminority plaintiffs. The review led the authors to conclude that such plans are least subject to challenge when they (a) have a remedial intent, (b) are well-conceived and narrowly tailored, (c) are instituted in cases where other, less restrictive strategies are unavailable, (d) have flexible implementation provisions, (e) have supportable goals vis'a-vis labor market conditions and time lines, and (f) have minimal negative consequences for non-minorities.

Summary Overall, it appears that several strategies might be used alone or in combination to deal with UD problems faced by individuals with various types of stigmas. Unfortunately, such problems are unlikely to be solved by the passage of time alone. One reason for this is that recent changes in organizational policies and practices that have served to improve the way stigmatized individuals are treated as job applicants or job incumbents appear to have been

442

Stone, Stone, and Dipboye

motivated more by a desire to comply with UD-related legislation than by voluntary action to promote social justice.

NEEDED RESEARCH ON STIGMAS OF VARIOUS TYPES Paucity of Research on Stigmas in Organizational Contexts Overall there is a paucity of empirical research on the effects of various types of stigmas in organizational contexts. An illustration of this is research on the effects of race. On the basis of a thorough review of studies published in 16 of the top journals dealing with organizational behavior issues, Cox and Nkomo (1990) reported that (a) of 11,804 articles published in such journals between 1964 and 1989, only 201 dealt with issues of race and ethnicity, and that (b) while an average of 11.7 papers per year were published in the 1970s that dealt with race-related issues, an average of only 3.6 per year were published in the 1985-1989 period. It seems highly likely that similar results would obtain if the same journals were searched for studies dealing with either handicaps or attractiveness. Overall, it appears that only a small amount of research on stigma-related issues is being conducted and subsequently published. At present there is no solid evidence on why this is the case. However, Cox (1990) suggests several possible explanations for research on race-related issues. We adapted these to make them applicable to stigmas in general. First, research on stigmas is limited because individuals who do such research are often stigmatized by their colleagues (e.g., viewed as having very narrow interests). Second, funding agencies are reluctant to fund research on stigmas. Third, individuals who do work on stigmas are ostracized by others in their organizations (typically colleges or universities). Fourth, research on stigmas presents methods related issues (e.g., small sample sizes, representativeness of samples of subjects, subject cooperation problems, response artifact problems) that are less common in studies dealing with other topics. Fifth, for various reasons researchers may submit few stigma-related papers to the major journals in industrial and organizational psychology and closely allied fields. Sixth, reviewers at the top journals may not be familiar with stigma issues and may thus provide unfair reviews of papers dealing with such issues. Seventh, and finally, reviewers or editors may not regard stigmas as being of much interest or importance to journal subscribers.

Stigmas in Organizations

443

Needed Research In preceding sections of this chapter we noted a number of specific research needs. In the interest of brevity these are not repeated here. However, several general themes merit restating. First, a great deal more research is needed that deals with the effects of stigmas (physical unattractiveness, race, handicaps) on various access- and treatment-related outcomes. Second, in order to produce findings that are in ternally and externally valid this research will have to avoid a number of problems (e.g., nonrepresentative samples of subjects, research obtrusiveness, demand-related artifacts, impression management) that are common to many previous studies of UD. Third, research needs to consider a wider range of independent variables than it has in the past. For example, most studies that deal with race effects compare only blacks and whites. In view of the large numbers of Mexican-Americans, Asian-Americans, and Native-Americans in the United States, there is a pressing need for research that considers these types of individuals.

Problem Recognition In a number of previous publications in the literature of such fields as industrial and organizational psychology, organizational behavior, personnel and human resources management, the position is taken that the effects of such stigmas as race, handicaps, and unattractiveness are of trivial magnitude and are therefore of little or no moment. However, from the consideration of material in this chapter it should be clear that the empirical evidence is quite inconsistent with this stance. Even if the research evidence were not as clear as it is in demonstrating the destructive consequences of stigmas, it would be incorrect to assume that stigma-related problems were nonexistent or of trivial magnitude on the basis of selected studies that are plagued with methodological problems. If the same assumption were accepted, we would be unlikely to ever solve the numerous stigma-related problems that marked individuals face in organizational settings.

SUMMARY AND CONCLUSIONS As a consequence of a number of factors, increasing attention is being paid to the fair treatment of individuals in society and in organizational contexts. Fair treatment issues are a concern of special importance to individuals who are stigmatized by virtue of their standing on variables such as physical attractiveness, physical

444

Stone, Stone, and Dipboye

and psychological handicaps, and race. Theories and models from social psychology, political psychology, sociology, and other academic disciplines provide a number of useful explanations of stigmatization and its consequences. Unfortunately, stigma-related issues are only infrequently studied by individuals in industrial and organizational psychology and closely allied fields. Moreover, the validity of many stigmarelated studies is subject to question. However, what research there is, including that from social psychology, political psychology, and rehabilitation psychology, suggests that such stigmas as race, physical unattractiveness, and handicaps have nontrivial effects on a variety of criteria, including criteria that relate to the access of individuals to jobs and their treatment as job incumbents. The general nature of these effects is that stigmatized individuals (handicapped persons, blacks, physically unattractive people) are worse off than their non-stigmatized counterparts with respect to both access and treatment considerations. A number of approaches might be taken to deal with these and other stigma-related problems. Among these are (a) changing physical aspects of organizations, social environments, organizational policies and practices, and stigmatized people, and (b) relying upon legal remedies for dealing with UD against stigmatized individuals. A great deal of research is needed to address a myriad of stigma-related issues. This research will need to both use creative strategies and overcome the problems that have plagued many previous studies of stigmas in organizational contexts. Overall, what this chapter suggests is the need for industrial and organizational psychologists and researchers in closely allied fields to pay much more attention to stigmas and stigma-related problems in organizational settings. We need to be much more proactive in studying stigma-related issues and in creatively solving the problems faced by stigmatized individuals. To the extent that we help solve such problems we will contribute measurably to the development of organizational environments in which all individuals are afforded fair treatment.

REFERENCES Adams, E. F. (1978). A multivariate study of perceptions of and attitudes toward minority and majority managers. Journal of Applied Psychology, 63, 277-288. Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. New York: Harper & Row.

Stigmas in Organizations

445

Andrisani, P. J. & Nestel, G. (1976). Internal-external control as contributors to and outcomes of work experience. Journal of Applied Psychology, 61, 156-165. Arvey, R. D. (1979). Unfair discrimination in the employment interview: Legal and psychological aspects. Psychological Bulletin, 86, 736-765. Arvey, R. D., & Campion, J. E. (1982). The employment inteview: A summary and review of recent research. Personnel Psychology, 35, 281-322. Ashmore, R. D., & Del Boca, F. K. (1976). Psychological approaches to understanding intergroup conflicts. In P. A. Katz (Ed.), Towards the elimination of racism ( pp. 73-123). NY:Pergamon . Ashmore, R. D., & Del Boca, F. K. (1981). Conceptual approaches to stereotypes and stereotyping. In D. L. Hamilton (Ed.) Cognitive processes in stereotyping and intergroup behavior @p. 1-35). Hillsdale, NJ: Erlbaum. Bardack, N. R., & McAndrew, F. T. (1985). The influence of physical attractiveness and manner of dress on success in a simulated personnel decision. Journal of Social Psychology, 125, 777-778. Bartol, K. M., Evans, C. I., & Stith, M. T. (1978). Black versus white leaders: A comparative review of the literature. Academy of Management Review, 3, 293-304. Beehr, T. A.? & Gilmore, D. C. (1982). Applicant attractiveness as a perceived job-relevant variable in selection of management trainees. Academy of Management Journal, 25, 607-617. Belgrave, F. (1984). The effectiveness of strategies for increasing social interaction with a physically disabled person. Journal of Applied Social Psychology, 14, 147-161. Belgrave, F. Z., & Mills, J. (1981). Effect upon desire for social interaction with a physically disabled person of mentioning the disability in different contexts. Journal of Applied Social Psychology, 11, 44-57. Benson, P. L., Karabenick, S. A., & Lerner, R. M. (1976). The effects of physical attractiveness, race, and sex on receiving help. Journal of Experiemental Social Psychology, 12, 409415. Benson, P. L., Severs, D., Tagenhorst, J., Loddengaard, N. (1980). The social costs of obesity: A non-reactive field study. Social Behavior and Personality, 8, 91-96.

446

Stone, Stone, and Dipboye

Bentz, V. J. (1985). Research findings from personality assessment of executives. In H. J. Bernardin, & D. A. Bownas (Eds.), Personality assessment in organizations (pp. 82- 145). New York: Praeger. Berry, D. S . , & McArthur, L. Z. (1986). Perceiving character in faces: The impact of age-related craniofacial changes on social perception. Psychological Bulletin, 100, 3-18. Bieber, L. & Dipboye, R. L. (April, 1988). Biases in evaluating managerial job applicants: the efsects of gender and physical attractiveness. Paper presented at the meeting of the Society for Industrial and Organizational Psychology, Dallas, TX. Bierly, M. M. (1985). Prejudice toward contemporary outgroups as a generalized attitude. Journal of Applied Social Psychology, 15, 189-199. Bigoness, W. J. (1976). Effect of applicant's race, sex, and performance, on employer's performance ratings: Some additional findings. Journal of Applied Psychology, 61, 80-84. Bordieri, J. E., & Drehmer, D. E. (1986). Hiring decisions for disabled workers: Looking at the cause. Journal of Applied Social Psychology, 16, 197-208. Brewer, M. B.,. & Kramer, R. M. (1985). The psychology of intergroup attitudes and behavior. Annual Review of Psychology, 36, 219-243. Bull, R., & Rumsey, N. (1988). The social psychology of facial appearance. London: Springer-Verlag. Byrne, D., London, O., & Reeves, I. (1968). The effects of physical attractiveness, sex, and attitude similarity on interpersonal attraction. Journal of Personality, 36, 259-271. Campion, M. A.? Pursell, E. D., & Brown, B. K. (1988). Structured interviewing: Raising the psychometric properties of the employment interview. Personnel Psychology, 41, 25-42. Cann, E., Siegfried, W. D., & Pearce, L. (1981). Forced attention to specific applicant qualification: Impact of physical attractiveness and sex on applicant bias. Personnel Psychology, 34, 65-76. Carlson, R. (1967). Evaluating interview and employment application data. Personnel Psychology, 20, 44 1-460. Carroll, S . J., Jr. (1966). Relationship of various college graduate characteristics to recruiting decisions. Journal of Applied Psychology, 50, 42 1-423. Carver, C. S . , Glass, D. C. & Katz, I. (1978). Favorable evalution of blacks and the handicapped: Positive prejudice, unconscious denial, and social desirability. Journal of Applied Social Psychology, 8, 97-106.

Stigmas in Organizations

447

Cash, T. F.? Gillen,B., & Bums, D. S. (1977). Sexism and 'beautyism' in personnel consultant decision making. Journal of Applied Psychology, 62, 301-310. Cash, T. F., & Kilcullen, R. N. (1985). The aye of the beholder: susceptibility to sexism and beautyism in the evaluation of managerial applicants. Journal of Applied Social Psychology, 15, 591-605. Chung, P., & Leung, K. (1988). Effects of performance information and physical attractiveness on managerial decisions about promotion. Journal of Social Psychology, 128, 791-801. Clifford, M., & Walster, E. (1973). The effect of ph sical attractiveness on teacher evaluation. Sociology of E ucation, 46, 248. Clore, G. L., & Jeffrey, K. M. (1972). Emotional role playing, attitude change, and attraction toward a disabled person. Journal of Personality and Social Psychology, 23, 105-111. Cox, T.,Jr. (1990). Problems with organizational research on race and ethnicity issues. Journal of Applied Behavioral Science, 26, 5-23. Cox, T., Jr., & Nkomo, S. M. (1990). Invisible men and women: A status report on race as a variable in organization behavior research. Journal of Organizational Behavior, 11, 4 19-431. Cox, J. A., & Krumboltz, J. D. (1958). Racial bias in peer ratings of basic airmen. Sociometry, 21, 292-299. Crosby, F., Bromley, S., & Saxe, L. (1980). Recent unobtrusive studies of black and white discrimination and prejudice: A literature review. Psychological Bulletin, 87, 546-563. Croxton, J. S., & Van-Rensselaer, B., Dutton, D. L., & Ellis, J. W. (1989). Mediating effect of prestige on occupational stereotypes. Psychological Reports, 64, 723-732. Czajka, J. M., & DeNisi, A. S. (1988). Effects of emotional disability and clear performance standards on performance ratings. Academy of Management Journal, 31, 394-404. Deutscher, I. (1966). Words and deeds. Social Problems, 13, 235254. Deutscher, I. (1973). what we say/what we do: Sentiments and acts. Glenview, IL: Scott, Foresman. Dickey-Bryant, L., Lautenschlager, G. J., & Mendoza, J. L., & Abrahams, N. (1986). Facial attractiveness and its relation to occupational success. Journal of Applied Psychology, 71, 1619.

d

448

Stone, Stone, and Dipboye

Dipboye, R. L., Fromkin, H. L., & Wiback, K. (1975). Relative importance of applicant sex, attractiveness, and scholastic standing in evaluation of job applicant resumes. Journal of Applied Psychology, 60, 39-43. Dipboye, R. L., Arvey, R. D., & Terpstra, D. E. (1977). Sex and physical attractiveness of raters and applicants as determinants of resume evaluations. Journal of Applied Psychology, 62, 288-294. Drehmer, D. E., & Bordieri, J. E. (1986, August). Cause of disability and the selection evaluations of handicappedjob applicants. Paper presented at the meeting of the Academy of Management, Chicago, IL. Efran, M. G. (1974). The effect of physical appearance on the judgment of guilt, interpersonal attraction, and severity of recommended punishment in a simulated jury task. Joumal of Bperimental Research in Personality, 8, 45-54. Efran, M. G., & Patterson, E. (1974). Voters vote beautiful: The effect of physical appearance on a national debate. Canadian Joumal of Behavioral Science, 6, 352-356. Evans, J. H. (1970). Changing attitudes towards disabled persons: An experimental study. Rehabilitation Counseling Bulletin, 19, 572-579. Farina, A., Felner, R. D., & Boudreau, L. A. (1973). Reactions of workers to male and female mental patient job applicants. Journal of Consulting and Clinical Psychology, 41, 363-372. Farina, A., Holland, C. H., & Ring, K. (1966). Role of stigma and set in interpersonal interaction. Journal of Abnormal Psychology, 71, 421-428. Farina, A., Sherman, M., & Allen, J. G. (1968). Role of physical abnormalities in interpersonal perception and behavior. Journal of Abnormal Psychology, 73, 590-593. Florian, V. (1978). Employer opinions of the disabled person as a worker. Rehabilitation Counseling Bulletin, 22, 38-43. Fontenelle, G. A., Peek-Phillips, A , , & Lane, D. M. (1985). Generalizing across stimuli as well as subjects: A neglected aspect of external validity. Journal of Applied Psychology, 70, 101- 107. Ford, J. K., Kraiger, K., & Schechtman, S. L. (1986). Study of race effects in objective indices and subjective evaluations of performance: A meta-analysis of performance criteria. Psychological Bulletin, 99, 330-337. Frieze, I. H., Olson, J. E., & Russell, J. (in press). Attractiveness and income for men and women in management. Journal of Applied Social Psychology.

Stigmas in Organizations

449

Gilmore, D. C., Beehr, T. A., & Love, K. G. (1986). Effects of applicant sex, applicant physical attractiveness, type of rater and type of job on interview decisions. Journal of Occupational Psychology, 59, 103-109. Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood Cliffs, NJ: Prentice-Hall. Goldman, W., & Lewis, P. (1977). Beautiful is good: Evidence that the physically attractive are more socially skillful. Journal of Experimental Social Psychology, 13, 125-130. Good, D.C., Olson, J. E., and Frieze, I. H. (1986, August). The impact of background factors on the wage diferential between the sexes: A study of MBAs. Paper presented at the Annual Meeting of the Academy of Management, Chicago, IL. Graves, L. M., & Powell, G.N. (1988). An investigation of sex discrimination in recruiters' evaluations of actual applicants. Journal of Applied Psychology, 73, 20-29. Grealish, C. A., & Salomone, P. R. (1986). Devaluing those with disability: Take responsibility, take action. Ncational Guidance Quarterly, 34, 147-150. Greenhaus, J. F., & Gavin, J. F. (1972). The relationship between expectancies and job behavior for white and black employees. Personnel Psychology, 25, 449-455. Greenwald, M. (1981). The effects of physical attractiveness, experience, and social performance on employee decision making in job interviews. Behavioral Counseling Quarterly, I , 275287. Greenwood, R., & Johnson, V. A . (1987). Employer perspectives on workers with disabilities. Journal of Rehabilitation, 53, 3746. Guion, R. M., & Gibson, W. M. (1988). Personnel selection and placement. In M. R. Rosenzweig, & L. W. Porter (Eds.) Annual Review of Psychology, 39, 349-374. Haefner, J. E. (1976). Race, age, sex, and competence as factors in employer selection of the disadvantaged. Journal of Applied Psychology, 62, 199-202. Haefner, J. E. (1977). Sources of discrimination among employees: A survey investigation. Journal of Applied Psychology, 62, 265-270. Hahn, H. (1988). The politics of physical differences: Disability and discrimination. Journal of Social Issues, 44, 39-47. Hamilton, D. L. (1976). Cognitive biases in the perception of social groups. In J. S. Carroll & J. W. Payne (Eds.), Cognition and social behavior (pp. 8 1-93). Hillsdale, NJ: Erlbaum.

450

Stone, Stone, and Dipboye

Hamilton, D. L. (1981a). Cognitive processes in stereotyping and intergroup behavior. Hillsdale, NJ: Erlbaum. Hamilton, D. L. (1981b). Illusory correlation as a basis for stereotyping. In D. L. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup behavior (pp. 115-144). Hillsdale, NJ: Erlbaum. Hamner, W. C., Kim, J. S., Baird, L.,& Bigoness, W. J. (1974). Race and sex as determinants of ratings by potential employers in a simulated work-sampling task. Journal of Applied Psychology, 59, 705-711. Hankins, N. E., McKinnie,W.T., & Bailey, R. C. (1979). Effects of height, physique, and cranial hair on job-related attributes. Psychological Reports, 45, 853-854. Harris, M. B., Harris, R. J., & Bochner, S. (1982). Fat, four-eyed, and female: Stereotypes of obesity, glasses, and gender. Journal of Applied Social Psychology, 12, 503-516. Hastorf, A. H., Northcraft, G. B., & Picciotto, S. R. (1979). Helping the handicapped: How realistic is the performance feedback received by the physically handicapped. Personality and Social Psychology Bulletin, 5, 373-376. Hastorf, A. H., Wildfogel, J., & Cassman, T, (1979). Acknowledgement of handicap as a tactic in social interaction. Journal of Personality and Social Psychology, 37, 1790-1797. Hatfield, E., & Sprecher, S. (1986). Mirror, mirror. . . . New York: State University of New York Press. Heilman, M. E. (1984). Information as a deterrent against sex discrimination: The effects of applicant sex and information type on preliminary decisions. Organizational Behavior and Human Peformance, 33, 174-186. Heilman, M. E. & Suruwatari, L. R. (1979). When beauty is beastly: The effects of appearance and sex on evaluations of job applicants for managerial and nonmanagerial jobs. Organizational Behavior and Human Peformance, 23, 360372. Heilman, M. E., & Stopeck, M. H. (1985a). Being attractive, advantage or disadvantage? Performance based evaluations and recommended personnel actions as a function of appearance, sex, and job type. Organizational Behavior and Human Performance, 35,202-215. Heilman, M . E., & Stopeck, M. H. (1985b). Attractiveness and corporate success: Different causal attributions for males and females. Journal of Applied Psychology, 70, 379-388.

Stigmas in Organizations

45 1

Herman, C. P., Zanna, M. P., & Higgins, E. T. (Eds.). (1986). Physical appearance, stigma, and social behavior: The Ontario symposium, @I. 3. Hillsdale, NJ: Erlbaum. Higgins, E. T., & Bargh, J. A. (1987). Social cognition and social perception. Annual Review of Psychology, 38, 369-425. Huck, J. R., & Bray, D. W. (1976). Management assessment center evaluations and subsequent job performance of white and black females. Personnel Psychology, 29, 13-30. Ilgen, D. R., & Youtz, M. A. (1986). Factors affecting the evaluation and development of minorities in organizations. In K. B. Rowland & G. R. Ferris (Eds.), Research in Personnel and Human Resources Management, 4 @p. 307-337). Greenwich, C T JAI Press. Jackson, L. A. (1983a). The influence of sex, physical attractiveness, sex role, and occupational sex-linkage on perceptions of occupational suitability. Journal of Applied Social Psychology, 13, 31-44. Jackson, L. A. (1983b). Gender, physical attractiveness, and sex role in occupational treatment discrimination: The influence of trait and role assumptions. Journal of Applied Social Psychology, 13, 443-458. Johnson, R . , & Heal, L. (1976). Private employment agency responses to the physically handicapped applicant in a wheelchair. Journal of Applied Rehabilitation Counseling, 7, 12-21. Jones, E. E., Farina, A., Hastorf, A. H., Markus, H., Miller, D. T.! Scott, R. A., & de Sales-French, R. (1984). Social stigma: The psychology of marked relationships. San Francisco, CA: W. H. Freeman & Co. Kanter, R. M. (1977). Men and women of the corporation. New York: Basic Books. Karlins, M.? Coffman, T. L., & Walters, G. (1969). On the fading of social stereotypes: Studies in three generations of college students. Journal of Personality and Social Psychology, 13, 116. Katz, D.! & Braly, K. N. (1933). Verbal stereotypes and racial prejudice. Journal of Abnormal and Social Psychology, 133, 280-290. Katz, I. (1979). Some thoughts about the stigma notion. Personality and Social Psychology Bulletin, 5 , 447-460. Katz, I. (1981). Stigma: A social psychological analysis. Hillsdale, NJ: Lawrence Erlbaum Associates. Katz, I., Glass, D. C., Lucido, D. J:, & Farber, J. (1977). Ambivalence, guilt, and the denigration of a physically handicapped victim. Journal of Personality, 45, 419-429.

452

Stone, Stone, and Dipboye

Katz, I . , Glass, D. C., Lucido, D. J., & Farber, J. (1979). Harmdoing and victim's racial or orthopedic stigma as a determinant of helping behavior. Journal of Personality, 47, 340-364. Kinder, D. R. (1986). The continuing American dilemma: White resistance to racial change 40 years after Myrdal. Journal of Social Issues, 42, 151-171. Kinder, D. R., & Sears, D. 0. (1981). Prejudice and politics: Symbolic racism versus racial threats to the good life. Journal of Personality and Social Psychology, 40, 4 14-431. Kinicki, A. J., & Lockwood, C. A. (1985). The interview process: An examination of factors recruiters use in evaluating job applicants. Journal of Wcational Behavior, 26, 116-125. Kleck, R. (1968). Physical stigma and nonverbal cues emitted in face to face interaction. Human Relations, 21, 19-28. Kleck, R. (1969). Physical stigma and task oriented interactions. Human Relations, 22, 53-60. Kleck, R. E., Ono, H., & Hastorf, A. H. (1966). The effects of physical deviance upon face to face interaction. Human Relations, 19, 425-436. Kleiman, L. S., & Faley, R. H. (1988). Voluntary affirmative action and preferential treatment: Legal and research implications. Personnel Psychology, 41, 48 1-496. Kogan, N. (1961). Attitudes toward old people: The development of a scale and examination of correlates. Journal of Abnormal and Social Psychology, 62, 44-54. Kraiger, K., & Ford, J. K. (1985). A meta-analysis of ratee race effects in performance ratings. Journal of Applied Psychology, 70, 56-65. Krefting, L. A., & Brief, A. P. (1976). The impact of applicant disability on evaluative judgements in the selection process. Academy of Management Journal, 19, 675-680. Langer, E. J., Fiske, S . , Taylor, A. E., & Chanowitz, B. (1976). Stigma, staring and discomfort: A novel stimulus hypothesis. Journal of Experimental Social Psychology, 12, 451-463. Larkin, J. C., & Pines, H. A. (1979). No fat persons need apply. Experimental studies of the overweight stereotype and hiring preference. Sociology of Ubrk and Occupations, 6, 312-327. Latham, G. P., Saari, L. M., Pursell, E. D., & Campion, M. A. (1980). The situation interview. Journal of Applied Psycholog y , 65, 422-427. Levitan, S . A., & Taggart, R . (1977). Employment problems of disabled persons. Monthly Labor Review, 100, 3-13. Lerner, M. J. (1980). The belief in a just world: A fundamental delusion. New York: Plenum Press.

Stigmas in Organizations

453

Maddox, G. L., Back, K. W., & Liederman, V. R. (1968). Overweight as social deviance and disability. Journal of Health and Social Behavior, 9, 298-298. McConahay, J. B., & Hough, J. C. (1976). Symbolic racism. Journal of Social Issues, 32, 23-46. McDonald, T., & Hakel, M. D. (1985). Effects of applicant race, sex, suitability, and answers on interviewer's questioning strategy and ratings. Personnel Psychology, 38, 321-334. McIntyre, S . , Moberg, D. J., & Posner, B. Z. (1980). Preferential treatment in preselection decisions according to sex and race. Academy of Management Journal, 23, 738-749. Merton, R. K. (1948). The self-fulfilling prophecy. Antioch Review, 8, 193-210. Meyerowitz, B. E., Williams, J. G.,, & Gessner, J. (1987). Perceptions of controllability and attitudes toward cancer and cancer patients. Journal of Applied Social Psychology, 17, 471-492. Mills, J., Belgrave, F., & Boyer, K.M. (1984). Reducing avoidance of social interaction with a physically disabled person by mentioning the disability following a request for aid. Journal of Applied Social Psychology, 14, 1- 11. Miller, D. T., & Turnbull, W. (1986). Expectancies and interpersonal processes. Annual Review of Psychology, 37, 233-256. Mobley, W. H. (1982). Supervisor and employee race and sex effects on Derformance amraisal: A field study of adverse impact and generalizabilifi Academy of Management Journal, 25, 598-606. Mullins,.T. W. (1978). Racial attitudes and the selection interview: A factorial experiment. Unpublished doctoral dissertation, University of Houston. Mullins, T. W. (1982). Interviewer decisions as a function of applicant race, applicant quality and interviewer prejudice. Personnel Psychology, 35, 163-124. Murphy, G., & Likert, R. (1938). Public opinion and the individual. New York: Harper & Brothers. Myrdal, G. (1944). An American dilemma: The negro problem and American democracy. New York: Harper. Newman, J. M. (1978). Discrimination in recruitment: An empirical analysis. Industrial and Labor Relations Review, 32, 1523. Newman, J. M. & Krzystofiak, F. (1979). Self-reports versus unobtrusive measures: Balancing method variable and ethical concerns in employment discrimination research. Journal of Applied Psychology, 64, 82-85.

454

Stone, Stone, and Dipboye

Novak, P. W., & Lerner, M. J. (1968). Rejection as a consequence of perceived similarity. Journal of Personality and Social Psychology, 9, 147-152. Parsons, C. K., & Liden, R. C. (1984). Interviewer perceptions of applicant qualifications: A multivariate field stud of demographic characteristics and nonverbal cues. J o u m of Applied Psychology, 69, 557-568. Pettigrew, T. F, (1979). The ultimate attribution error: Extending Allport's cognitive analysis of prejudice. Personality and Social Psychology Bulletin, 5 , 461-476. Piner, K . E., & Kahle, L. (1984). Adapting to the stigmatizing label of mental illness: Forgone but not forgotten. Journal of Personality and Social Psychology, 47, 805-811. Price Waterhouse v Hopkins, 825 F 2d458 (D.C. Cir 1987), cert. granted 108 S . Ct 1106 (1988). Pulakos, E. D., White, L. A., Oppler, S. H., & Borman, W. C., (1989). Examination of race and sex effects on performance ratings. Journal of Applied Psychology, 74, 770-780. Quereshi, M. Y., & Kay, J. P. (1986). Physical attractiveness, age, and sex as determinants of reactions to resumes. Social Behavior and Personality, 14, 103-113. Rand, T. M & Wexley, K. N. (1975). Demonstration of the effect, "'s)imilar to me," in simulated employment interviews. Psychological Reports, 36, 535-544. & Carpenter, B. N. (1987). A model of hiring deciRaza, S. M,., sions i n real employment interviews. Journal of Applied Psychology, 72, 596-603. Reingen, P. H., Gresham, L. G., & Kernan, J. B. (1980). Behavioral consequences of the physical attractiveness stereotype in personal selling. In R. P. Bagozzi, K. L. Bernhardt, P. S . Busch, D. W. Cravens, J. Fittair, & C. A. Scott (Eds.). Marketing in the 80's (pp. 109-1 13). Chicago: American Marketing Association. Richardson, S. A., Goodman, N., Hastorf, A. H., Dornbusch, S. M. (1961). Cultural uniformity in reaction to physical disabilities. American Sociological Review, 26, 241-247. Rickard, T., Triandis, H . , & Patterson, C. (1963). Indices of employer prejudice toward disabled applicants. Journal of Applied Psychology, 47, 52-55. Rokeach, M. (1973). The nature ofhuman values. New York: Free Press. Rose, G.., & Brief, A. (1979). Effects of handicap and job characteristics on selection evaluations. Personnel Psychology, 32, 385-392.

1y

Stigmas in Organizations

455

Rosenthal, R., & Jacobson, L. (1968). Qgmalion in the classroom. New York: Holt, Rinehart & Winston. Ross, J., & Ferris, K. R. (1981). Interpersonal attraction and organizational outcomes: a field examination. Administrative Science Quarterly, 26, 617-632. Roszell, P., Kennedy, D., & Grabb, E. (1989). Physical attractiveness and attainment among Canadians. Journal of Psychology, 123, 547-559. Rumsey, N., Bull, R., & Gahagan, D. (1982). The effect of facial disfigurement on the proxemic behavior of the general public. Journal of Applied Social Psychology, 12, 137-150. Russel, D., Lenel, J. C., Spicer, C., Miller, J., Albrecht, J. & Rose, J. (1985). Evaluating the physically disabled: An attributional analysis. Personality and Social Psychology Bulletin, 11, 23-31. Rynes, S., & Gerhart, B. (1990). Interviewer assessments of applicant 'fit': An exploratory investigation. Personnel Psychology, 43, 13-35. Schmidt, F. L., & Johnson, R. H. (1973). The effect of race on peer ratings in an industrial situation. Journal of Applied Psychology, 57, 237-24 1. Schmitt, N., & Hill, T. E. (1977). Sex and race composition of assessment center groups as a determinant of peer and assessor ratings. Journal of Applied Psychology, 62, 261-264. Schmitt, N. & Lappin, M. (1980). Race and sex as determinants of mean and variance of performance ratings. Journal of Applied Psychology, 65, 428-435. Schneider, C. R., & Anderson, W. (1980). Attitudes toward the stigmatized: Some insights for recent research. Rehabilitation Counseling Bulletin, 23, 299-3 13. Sears, D. 0. (1987). Political psychology. In M. R. Rosenzweig & L. W. Porter (Eds.), Annual Review of Psychology, 38, 229255. Shahani, C., & Dipboye, R. L. (1990, April). Impact of applicant attractiveness, gender, and gender appropriateness of academic major on college admissions. Paper presented at the meeting of the Eastern Psychological Association, Philadephia, PA. Sherman, S. J., Judd, C. M., & Park,B. (1989). Social cognition. In M. R. Rosenzweig & L. W. Porter (Eds.), Annual Review Of Psychology, 40, 281-326. Sniderman, P. M., & Hagen, M. G. (1985). Race and inequality: A study in American values. Chatham, NJ: Chatham House Publishers.

456

Stone, Stone, and Dipboye

Sniderman, P. M., & Tetlock, P. E. (1986). Symbolic racism: Problems of motive attribution in political analysis. Journal of Social Issues, 42, 129-150. Snyder, M., Berscheid, E., & Matwychuk, A. (1988). Orientations toward personnel selection: Differential reliance on appearance and personality. Journal of Personality and Social Psycholog ~ , 5 4972-979. , Snyder, M. L., Kleck, R. E., Strenta, A., & Mentzer, S. J. (1979). Avoidance of the handicapped: An attributional ambiguity analysis. Journal of Personality and Social Psychology, 3 7, 2297-2306. Snyder, M., 'hnke, E. D., & Berscheid, E. (1977). Social perception and interpersonal behavior. Journal of Personality and Social Psychology, 35,656-666. Springbett, B. (1958). Factors affecting the final decision in the employment interview. Canadian Journal of Psychology, 12, 13-22. Stone, C., & Sawatzki, B. (1980). Hiring bias and the disabled interviewee: Effects of manipulated work history and disability information of the disabled job applicant. Journal of Mcational Behavior, 16, 96- 104. Stone, D. L. (1986, August). The efects of nondisclosure ofphysical handicaps on personnel selection decisions. Paper presented at the meeting of the Academy of Management, Chicago, IL. Stone, E. F., Shetzer, L., & Eggleston, S . (1986, August). Efsects

of handicap type, handicap revelation mode, and interviewee on ratings of applicant suitability. Paper presented at the meeting of the Academy of Management, Chicago, IL. Stone, E. F. (1989). Self-presentational biases in organizational research. In R. A. Giacalone & P. Rosenfeld (Eds.), Impression management in the organization (pp. 189-202). Hillsdale, NJ: Erlbaum. Sutherland, M. E. (1990). Black faculty in white academia: The fit is an uneasy one. The Western Journal of Black Studies, 14, 17-23, Tagalakis, V.? Amsel, R., & Fichten, C. S. (1988). Job interview strategies for people with a visible disability. Journal of Applied Social Psychology, 18, 520-532. Xijfel, H. (1981). Human groups and social categories. Cambridge, MA: Cambridge University Press. Tajfel, H. (1982). (Ed.) Social identity and intergroup relations. Cambridge: Cambridge University Press.

Stigma in Organizations

457

Tringo, J. L. (1970). The hierarchy of preference toward disability groups. The Journal of Special Education, 4, 295-306. Triplet, R. G., & Sugarman, D. B. (1987). Reactions to AIDS victims: Ambiguity breeds contempt. Personality and Social Psychology Bulletin, 13, 265-274. Waters, J. (1985). Cosmetics and the job market. In J. Graham & A. Kligman (Eds.), The psychology of cosmetic treatments. New York: Praeger. Webb, E. J., Campbell, D. T., Schwartz, R. D., Sechrest, L., & Grove, J. B. (1981). Nonreactive measures in the social sciences. Boston: Houghton Mifflin. Weigel, R. H . , & Howes, P. W. (1985). Conceptions of racial prejudice: Symbolic racism reconsidered. Journal of Social Issues, 41, 117-138. Wexley, K. N., & Nemeroff, W. F. (1974). The effects of racial prejudice, race of applicant, and biographical similarity on interviewer evaluations of job applicants. Journal of Social and Behavioral Sciences, 20, 66-78.

This Page Intentionally Left Blank

459

Ballagh, J.H., 259, 263, 271, 279 Ballantyne, H.C., 277 A Bandura, A., 35, 58, 155, 156, Ackerman, L., 190, 219 158, 163, 165, 166, 167, 171, Adams, E.F., 408, 409, 444 174, 178 Adams, J.S., 50, 58, 124, 147 Barber, T.X., 16, 25 Adelmann, P.K., 309, 324 Bardack, N.R.,415, 445 Aderman, D., 48, 58 Barling, J., 155, 159, 161, 178 Adler, N.J., 339, 354, 355, 356, Barnlund, D.C., 356, 369 357, 366, 368 Baron, R.A., 261,279 Adorno, T.W., 395, 444 Barrett, G.V., 297, 324, 355, Aigner, D., 302, 324 369 Alba, R.D., 340, 343, 368 Barth, F., 343, 346, 369 Albert, R.D., 357,368 Bartol, K.M., 403, 404, 445 Alderfer, C.P., 340, 341, 344, Baruch, G.K., 35,58, 317, 352, 353, 363, 364, 368, 369 318, 319, 324 Aldrich, H.E., 72, 85 Baskett, G.D., 40, 58 Allen, V.L., 350, 369 Baskin, O.W., 51, 58 Alliger, G.M., 16, 25, 41, 58, Bass, B.M., 355, 369 214, 215, 219 Batson, C.D., 48, 58 Alpander, G.G., 194, 219 Baumeister, R.F., 45, 58 Allport, G.W., 350, 369 Bausell, R.B., 263, 279 Amalfitano, J.G., 44, 58 Becker, ED., 58 American Psychological Becker, G.S., 300, 324 Association, 24, 25 Beckhard, R., 119, 138, 147 Amir, Y., 350, 369 Bedeian, A.G., 71, 85 Andrisani, P.J., 406, 407, 445 Beehr, T.A., 413, 421, 445 Arch, E.C.,291, 324 Beer, M., 117, 119, 124, 125, Archer, V.E., 272, 273, 279 126, 128, 130, 136, 145, 147 Argyris, C., 19, 20, 25, 119, Belgrave, F., 429, 435, 436, 445 147 Bem, S.L., 306, 324 Arvey, R.D., 292, 293, 324, Bennis, W.G., 173, 178 Benson, P.L., 413, 414, 427, 445 403, 404, 445 Ashford, S.J., 171, 175, 178 Bentley, G.C., 343, 369 Ashmore, R.D., 392, 402, 445 Bentvsi-Mayer, S. 324 Astin, H.S., 289, 292, 324 Bentz, V.J., 424, 446 Berg, D.N., 363, 370 B Berger, P.L., 76, 85 Babad, E.Y., 344, 347, 362, 369 Berkowitz, L., 48, 50, 59 Baber, K.M., 316, 324 Bernick, E.L., 194, 219 Backer, T.E., 274, 279 Berry, D.S., 422, 446 Bailey, J., 340, 369 Berry, J. W.? 34 1,342, 344, 345, Baldwin, T., 209, 219 346, 347,351, 364, 370

NAME INDEX

?

460

Betz, N.E., 290, 324 Beuhring, T., 297, 324 Bhagat, R.S., 162, 163, 179, 355, 356, 361, 370 Bhatnagar, D., 295, 325 Bieber, L., 414, 416,422, 446 Bielb W.T., 287, 300, 302, 3 O t 315, 325 Bierly, M.M., 395, 446 Bigoness, W.J., 410, 446 Binning, J.F., 292, 325 Birchler, G.R., 55, 59 Black, J.S., 209, 210, 211, 219, 357, 370 Blake, H.D., 213, 219 Blanchard, F.A., 339, 370 Blau, ED., 298, 301, 302, 303, 325 Blechman, E.A., 318, 325 Block, P., 231, 233, 237, 255 Bobko, P., 217, 219 Bochner, S., 357, 370 Boeker, W., 82, 83, 85 Boekestijn, C., 345, 370 Bogorya, Y., 210, 219 Bolt, J.F., 197, 219 Bond, M.H., 341, 370 Bordieri, J.E., 432, 446 Borman, W., 112 Boss, R.W., 132, 145, 147 Bossard, J.H.S., 34, 59 Boudreau, J., 217, 219 Boyacigiller, N., 354, 370 Braddock, J.H., 349, 370 Brennan, A.J.J., 260, 272, 279 Brett, J.M., 122, 148, 315, 325 Brewer, M.B., 349, 350, 35 1, 367, 371, 391, 392, 394, 395, 398, 439 Bridges, J.S., 290, 294, 325 Brief, A.P., 155, 179 Brislin, R.W., 357, 358, 371 Britt, M.A., 162, 165, 179 Brockner, J., 155, 158, 179

Name Index

Brod, C., 262, 267, 268, 279 Brodeur, P., 269, 270, 279 Brown, L.D., 339, 358, 371 Brown, S.M., 311, 325 Bull, R., 412, 446 Buller, P.F., 120, 128, 130, 131, 132, 136, 141, 142, 143, 144, 148 Burdick, G.R., 263, 279 Bureau of Labor Statistics, 325 Burger, J.M., 157, 179 Buriel, R., 343, 371 Burke, M.J., 206, 217, 220 Burley, K.A., 315, 325 Business-Higher Education Forum, 339, 371 Buss, D., 174, 179 Buss, D.M., 92, 93, 95, 98, 112 Butler, L.M., 237, 255 Buttner, E.H., 298, 325 Byham, W.C., 205, 206, 220 Byrne, D., 30, 31, 32, 34, 37, 41, 46, 57, 59, 427, 446 C Calas, M.B., 341, 363, 371 Calder, B. J., 245, 255 Campbell, J.P., 137, 141, 148 Campion, M.A., 440, 446 Cann, A., 292, 311, 313, 325 Cann, E., 413, 446 Cannon-Bowers, J.A., 124, 148 Caplow, T., 34, 60 Carlson, R., 413, 446 Carnevale, A.P., 216, 220 Carroll, S.J., Jr., 413, 421, 446 Carver, C . S . , 155, 158, 159, 175, 179, 431, 446 Cascio, W.F., 205, 220 Cash, T.F., 38, 42, 60, 310, 326, 413, 414, 415, 417, 422, 424, 447

Name Index

Caspi, A., 93, 112 Cattell, R.B., 112, 113 Caws, 17, 25 Chaney, F., 265, 279 Chernovsky, Z.Z., 263, 279 Chung, P., 419, 447 Chusmir, L.H., 314, 326 Clawson, J., 92, 113 Clemens, L.G., 54, 60 Clement, R.W., 210, 220 Cleveland, J.N., 38, 60, 91, 108, 113 Clifford, M., 412, 447 Clore, G.L., 30, 31, 60, 435, 447 Cohen, A.G., 321, 326 Cohen, M.H., 131, 148 Cohen, R., 343, 371 Cohen. S.. 261. 279 Cohen; S.L., 37, 38, 42, 60, 293. 326 Coles,’ F.S., 320, 321, 322, 326 Collier, M.J., 356, 372 Comas-Dim, L., 361, 372 Conger, J.A., 167, 173, 179 Conger, J.J., 35, 60 Conley, J.J., 96, 113 Cook, S.W., 350, 372 Cook, T.D., 129, 130, 136, 137, 148, 215, 220 Cooper, E.A., 297, 326 Costa, P.T., 100, 112, 113 Cox, C.L., 42, 60 Cox, J.A., 407, 447 Cox, T., Jr., 339, 340, 341, 341, 356, 363, 365, 372, 406, 442, 447 Cox, T., 261, 252, 279 Crawford, K.S., 170, 179 Crocker, 309, 326 Crohan, S.E., 306, 326 Crosby, F., 339, 372, 387, 398, 399, 402, 447 Crouch, A., 194, 220

46 1

Crouter, A.C., 262, 279 Croxton, J.S., 413, 447 Crozier, M., 75, 85 csoka, L.S., 201,220 Cullen, M.R., 265, 266, 269, 279, 280 Cummings, T.G., 264, 265, 280 Czajka, J.M., 429, 431, 447 D Dailey, R.C., 124, 148 Dansereau, F., 198, 202, 220 Davis, G., 352, 358, 372 Davis, L.L., 43, 60 Day, C.C., Jr., 275, 280 Day, D.V., 188, 220 Day, R., 263, 280 Deutscher, I., 402, 447 Deaux, K., 307, 326 Deci, E.L., 162, 179, 190, 220 Decker, P.J., 206, 220 DeFrank, R.S., 356, 372 DeMeuse, K.P., 117, 118, 128, 129, 130, 131, 136, 137, 140, 148 Deschamps, J.-C., 347, 372 Despres, L.A., 343, 372 Dickey-Bryant, L., 418, 447 Digman, J.M., 96, 113 DiMaggio, P.J., 73, 74, 75, 79, 80, 82, 83, 85 Dion, K.K., 41, 60 Dipboye, R.L., 37, 41, 42, 60, 293, 326, 399, 413, 414, 421, 448 DiTomaso, N., 213, 220 Dixon, E.P., 213, 221 Dobbins, B.H., 311, 313, 326 Doerfler, M.C., 307, 327 Doktor, R.H., 356, 372 Donoghue, S., 273, 280 Douthitt, R.A., 315, 327 Dovidio, J.F., 367, 373 Dreher, G.F., 298, 327

462

Drehmer, D.E., 429, 433, 448 Driskell, J.E., 124, 148 Dubbert, P.M., 318,327 Dubin, R., 13, 15, 25 Dubno, P., 295, 309, 327 Duncan, B.L., 350, 373 Duncker, K., 47, 60 Dunham, R.B., 54, 61 Dweck, C.S., 155, 164, 179 Dyer, W.G., 119, 136, 139, 148 E Eagly, A., 293, 327 Earley, P.C., 210, 221 Eccles, J.S., 288, 289, 327 Eden, D., 133, 137, 138, 140, 141, 142, 148, 170, 179 Educational Research Service, Inc., 264, 280 Edwards, A.L., 101, 113 Edwards, J., 344, 373 Efran, M.G., 412,448 Einhorn, H.J., 46, 61 Elliott, A.G.P., 292, 327 Emmons, R.A., 95, 113 Epstein, C.F., 300, 327 Epstein, S.S., 265, 280 Erickson, F., 356, 373 Evans, J.H., 435, 448 Evans, M.G., 170, 180 Eysenck, H.J., 112, 113 F Fahr, J.L., 197, 221 Fain, T.C., 322, 323, 327 Farina, A., 431, 448 Farnsworth, C.H., 276, 327 Farr, K.A., 296, 327 Farrell, D., 55, 56, 61 Fava, G.A., 260, 263, 280 Feagin, F., 346, 347, 373 Ferdman, B.M., 339, 340, 341, 342, 344, 345, 346, 347, 350, 358, 360, 361, 362, 363, 366,

Name Index

367, 373, 374 Fernandez, J.P., 352, 374 Ferraro, G.P., 357,374 Feinzig, S., 175, 180 Feldman, S.D., 52, 61 Feltz, D.L., 155, 163, 180 Femandez, J.P., 213, 221 Ferris, G.R., 37, 61 Festinger, L., 34, 61 Fiedler, F., 131, 133, 138, 149 Fiedler, F.E., 198, 200, 201, 22 1 Field, R.H.G., 201, 221 Fielding, J.E., 274, 280 Finkbeiner, A.K., 292, 327 Fisher, C.D.,197, 210, 221 Fishman, J., 340, 343, 347, 363, 374 Fiske, S.T., 349, 374 Fitzpatrick, J.L., 291, 327 Fleishman, E.A., 311, 327 Fleming, D.C., 133, 149 Fletcher, G.J.O., 351, 374 Fligstein, N., 84, 85 Florian, V., 432, 448 Folkman, S., 163, 165, 180 Fontanelle, G.A., 416, 448 Ford, D.H., 93, 94, 113 Ford, D.L., Jr., 221 Ford, J.K., 194, 195, 196, 213, 222, 403, 405, 448 Ford, M.E., 156, 157, 158, 166, 180 Ford, R.N., 54, 61 Forest, D., 47, 61 Forsythe, S., 43, 61 Frank, E.J., 312, 327 Frayne, C., 208, 222 Freedman, S.M., 161, 162, 180 Freeman, J., 83, 85, 291, 327 French, J.R.P., 187, 222 French, W.L., 117, 124, 149 Freudenheim, M., 278, 280 Frieze, I.H., 420, 448

Name Index

Fulk, J., 199, 222 Fullerton, H.N., Jr., 212, 222, 286, 327

463

Gomez-Mejia, L.R., 296, 328 Good, D.C., 420, 449 Good, L.R., 36, 62 Goode, S . , 340, 375 G Goodman, P.S., 120, 123, 149 Galaskiewicz, J., 82, 83 Gordon, J., 190, 222 Galinsky, E., 314, 328 Gordon, M.E., 3, 14, 26 Galizio, M., 48, 61 Gottlieb, N.H., 272, 274, 280 Gardner. D.G.. 302. 328 Gouaux, C., 30, 62 Graham, J.L., 196, 222 Garland; H., 190, 222, 307, Graves, L.M., 292, 328, 423, 312. 328 449 Gear, T.E., 131, 149 Graves, T.D., 345, 375 Geber, B., 2 12, 222 Grealish, C.A., 437, 449 George, J.M., 48, 58, 61 Green, K.L., 275, 280 Gergen, K.J., 18, 25 Greenberg, J., 52, 62 Gersick, C.J.G., 118, 149 Greenberger, D.B., 157, 180 Gibson, S., 164, 180 Greenberger, E., 316, 328 Gigerenzer, G., 25 Gilbert, L.A., 314, 328 Greene, T.R., 47, 62 Greenhaus, J.F., 408, 410, 449 Gillen, B., 42, 61 Gilmore, D.C., 310, 328, 413, Greenhaus, J.H., 352, 375 Greenwald, M., 413, 419, 449 414, 449 Greenwood, R., 426, 449 Gioia, D.A., 21, 24, 25, 342, Greer, O.L., 217, 223 363, 374 Gist, M.E., 155, 156, 171, 172, Greig, J.J., 297, 328 173, 180, 207, 208, 209, 222 Griffeth, R.W., 50, 62 Griffitt, W., 30, 40,54, 62 Gladstein, D., 122, 123, 124, Gudykunst, W.B., 358, 375 125, 149 Guilford, J.P., 101, 113 Glanz, K., 273, 280 Guion, R.M., 421, 449 Glasser, I., 339, 374 Gunderson, E.K., 124, 149 Glazer, N., 339, 340, 374 Gutek, B.A., 288, 322, 328 Gleason, P., 347, 374 GUZZO,R.A., 124, 149 Glickman, A.S., 118, 149 Goffman, E., 45, 61, 62, 385, H 388, 389, 391, 394, 395, 396, Hackett, G., 291, 329 427, 437, 440 Goktepe, J.R., 52, 62, 311, 313, Hacking, I., 26 Hackman, J.R., 117, 120, 122, 328 123, 125, 126, 140, 144, 149, Gold, M.E., 339, 375 169, 180 Goldberg, L.R., 91, 96, 97, 99, Haefner, J.E., 404, 405, 406, 107, 112, 113 449 Goldman, W., 424, 449 Goldstein, I.L., 186, 192, 193, Hahn, H., 427, 449 Hakuta, K., 344, 375 206,214, 215, 222

464

Name Index

352, 376 Hall, E.T., 51, 62, 355, 375 Hall, F.S., 305, 329 Higgins, E.T., 391, 393, 451 Hall, G.S., 10, 11, 26 Hill, L.A., 208, 223 Halstead, L.S., 133, 150 Hill, C.T., 54, 63 Hamilton, D.L., 349, 375, 388, Hinrichs, J.R., 259,280 391, 391, 393, 449, 450 Hirsch, P.M., 188, 223 Hamilton, G.G., 83, 85 Hirschman, C., 343, 347, 376 Hamner, W.C., 405,407, 450 Hodgins, D.C., 293, 323, 330 Hankins, N.E., 414, 450 Hofstede, G., 355, 358, 368, 376 Hanlon, M.E., 242, 255 Holahan, C.K.,155, 180 Hanna, J.L., 355, 375 Holland, J.L., 290, 330 Hannan, M.T., 73, 83, 85 Hollandsworth, J.G., 43, 63 Harris, C.T., 314, 329 Holmes, T.H., 262, 281 Harris, M.B., 414, 450 Hooyman, N.R., 276, 278, 281 Harrison, M.I., 231,235, 237, Hosking, D., 201, 223 242, 243, 244, 254, 255 House, R.J., 13, 26, 173, 180, Harvard Law Review, 339, 375 198, 223 Harwood, H.J., 271, 280 Howard, A., 9, 11, 26, 91, 92, Haskell, W.L., 275, 280 93, 95, 96, 99, 111, 114 Hastorf, A.H., 429, 431, 435, Howard, J., 350, 376 437, 450 Huber, V.L., 47, 63, 155, 161 Hatfield, E., 41, 62, 412, 416, Huck, J.R., 405, 406, 451 422, 450 Hughes, R.L., 134, 137, 138, Haw, M.A., 318, 329 142, 150 Hayghe, H.V., 286, 329 Hughes-Weiner, G., 358, 376 Hearn, J., 320, 329 Hulbert, K., 295, 316, 330 Hecht, M.L., 356, 375 Hunt, D.M., 289, 294, 330 Hegtvedt, K.A., 299, 329 I Heilman, M.E., 38, 62, 292, 309, 310, 323, 329, 413, 416, 417, Iaffaldano, M.T., 46, 63 418, 421, 422, 450 Ilgen, D., 349, 352, 353, 376 Heimovics, R.D., 312, 329 Ilgen, D.R., 171, 181, 396, 451 Heller, M., 344, 376 Imada, A.S., 44, 63 Helmer, O., 234, 255 Isajiw, W.W., 343, 376 Henderson, M., 356, 376 Tzraeli, D.N., 308, 311, 330 Hensley, V., 46, 62 Isen, A.M., 47, 48, 63, 64 Herman, C.P., 412, 451 Herman, S.N., 344, 376 J Herold, D.M., 123, 150, 171, Jackson, L.A., 299, 330, 413, 180 419, 451 Hersey, P., 198, 203, 223 Jacobson, M.B., 323, 330 Herz, D.E., 286, 329 Jagacinski, C.M., 306, 330 Herzberg, E., 55, 62, 120, 150 James, K., 349, 376 Hewstone, M., 349, 350, 351, Janis, I.L., 48, 64

Name Index

Jaspars, J., 35 1, 376 Jensen, I.W., 321, 330 Jewell, L.N., 7, 26 Jick, T.D., 318, 319, 330 Johnson, D.W., 50, 64 Johnson, R., 430, 45 1 Johnston, W.B., 339, 377 Jones, B.H., 316, 330 Jones, E.E., 385, 389, 392, 393, 394, 396, 426, 429, 432, 451 Jones, E.W., Jr., 339, 352, 367, 377 Jones, J.M., 340, 341, 344, 345, 361. 377 Jones,’G.R., 174, 181 K Kabanoff, B., 123, 124, 150 Kagitcibasi, C., 355, 361, 377 Kahn, S.E., 161, 163, 181 Kahneman, D., 47, 64 Kalin, R., 39, 64 Kamerman, S.B., 316, 330 Kamin, L.J., 37, 64 Kanfer, F.H., 208, 223 Kanter, R.M., 294, 295, 307, 308, 309, 319, 330, 417, 451 Kantrowitz, B., 340, 377 Kapalka, G.M., 306, 330 Kaplan, A., 255 Kaplan, R.E., 134, 139, 142, 143, 150, 196, 223 Karasek, R., 163, 169, 171, 181 Karlins, M., 401, 451 Katz, D., 124, 150, 186, 187, 188, 190, 204, 223, 350, 364, 377 Katz, I., 401, 428, 437, 451, 452 Katz, J.H., 339, 377 Katz, P.A., 377 Katz, R., 120, 124, 150 Kavanagh, M.J., 314, 331

465

Keefe, S.E., 344, 377 Kellerman, H., 124, 150 Keyes, C.F., 343, 377 Keys, B., 186, 223 Keys, D.E., 305, 306, 331 Kilborn, P.T., 339, 340, 377 Kinder, D.R., 397, 400, 401, 402, 452 Kinicki, A.J., 413, 452 Kirkpatrick, D.L., 136, 150, 214, 223 Kleck, R., 429, 430, 452 Kleiman, L.S., 339, 378, 441, 452 Klein, K.J., 212, 224 Knouse, S.B., 39, 64 Kobasas, S., 261, 281 Kochman, T., 346, 378 Kogan, N., 395, 452 Komaki, J.L., 195, 224 Konrad, A.M., 321, 322, 331 Kraiger, K., 352, 353, 378, 403, 404, 452 Krames, L., 54, 64 Krefting, L.A., 429, 431, 452 Kristein, M.M., 272, 273, 281 Krivonos, P.D. 49, 64 Kroeber, A.L., 344, 378 Krueger, R.A., 246,255 Kuhn, T., 24, 26 Kukla, A., 16, 22, 26 L Laird, J.D., 47, 64 Landy, F.J., 51, 54, 64 Lane, J., 172, 181 Langer, E.J., 429, 452 Langley, A., 83, 85 Laporte, W., 274, 281 Larkin, J.C., 54, 65, 414, 452 Larsen, R.J., 164, 181 Lanvood, L., 302, 331 Lassiter, D.L., 127, 151 Latack, J.C., 165, 181

Name Index

466

Marger, M.N., 340, 346, 378 Margerison, C., 134, 151 Marin, G., 355, 379 Markus, H., 155, 157, 173, 182 Markus, H.R., 346,378 Martin, J., 273, 274, 281 Martocchio, J.J., 318, 331 Marvis, P.H., 243, 255 Marx, R.D., 145, 151 Matthews, K.A., 263, 281, 286, 314, 319, 331 Mathieu, J.E., 162, 182, 217, 224 Mathison, D.C., 310, 331 Mayer, S. J., 206, 224 McCall, M.W., Jr., 192, 224 McCaskey, M.B., 51, 65 McClelland, D.C., 95, 114 McCombs, B.L., 155, 156, 157, 162, 182 McConahay, J.B., 400, 453 McCoy, J.L., 277, 281 McCrae, R.R., 96, 114 McCready, W.C., 343, 379 McDonald, T., 408, 409, 453 McEnery, J., 195, 197, 224 McEnrue, M.P., 156, 182 McGehee, W., 192,224 McGovern, T.V., 44, 65 McGrath, J.E., 123, 136, 151 McGregor, D., 3 11, 331 McIntyre, S . , 40, 65, 404, 410, 453 M McKillip, J., 230, 238, 241, Mabe, P.A., 111, 196, 224 245, 255 Mackay, C.J., 270, 281 McRae, A.V., 127, 151 Maddox, G.L., 433,453 Meadow, B.L.,31, 65 Magni, G., 260, 263, 281 Mednick, M.T., 47, 65, 311. 332 Maher, J. R., 54, 65 Mahler, D.M., 272, 276, 277, 281Mednick, S.A.,. 65 Mehrabian, A., 44, 65 Maier, S.F., 263, 281 Merton, R.K., 363, 379, 393, 453 Major, B., 299, 331 Messick, D.M., 349, 379 Mann, J.F., 350, 378 Meyer, J.W., 74, 76, 77, 86 Manz, C.C., 207, 224 Meyerowitz, B.E., 433, 453 Maraniss, D., 340, 378

Latham, G.P., 127, 151, 208, 224, 440, 452 Lauffer, A., 246, 255 Laumann, E.O., 73, 86 Laurent, A,, 355, 378 Lawler, E.E., 54, 65, 122, 151, 229, .230, 231,.233, 234, 235, 238, 240, 241, 242, 243, 244, 245 Lawson, E.D., 37, 65 Lee, C., 155, 161, 165, 170, 181 Lerner, M.J., 427, 428, 437, 452 Levant, R.F., 315, 316, 331 Lever, J., 289, 331 Levinger, D., 46, 55, 65 Levinger, G., 339, 364, 378 Levinson, H., 378 Levitan, S.A., 425, 452 Lewittes, H. J., 307, 331 Liden, R.C., 202, 224, 313, 331 Liebowitz, S.J., 119, 151 Linville, P.W., 350, 378 Lipman-Blumen, J., 312, 331 Locke, E.A., 23, 26, 46, 65, 155, 161, 165, 169, 170, 171, 182 Long, B.C., 287, 306, 331 Lont, C.M., 331 Lord, R.G., 99, 111, 114 Lott, A.J., 49, 50, 65 Luthans, F., 92, 114

'

'

Name Index

Mezias, S.J., 72, 83, 86 Miller, A . , 261, 281 Miller, A.G., 41, 66 Miller, D.A., 293, 332 Miller, D.T., 392, 393, 394, 439,453 Miller, J. G ., 379 Miller, N., 350, 355, 379 Miller, V.D., 171, 175, 182 Mills, J., 429, 436, 453 Miner, J.B., 95, 114, 193, 224 Minis, P.H., 46, 66 Mitchell, R., 134, 138, 151 Mitchell, T.R., 66 MIT, 46, 66 Mitchell, T., 38, 52, 66 Mitroff, I., 16, 18, 21, 26 Miura, I.T., 291, 332 Mobley, W.H., 46, 66, 405, 407, 453 Montero, M., 347, 379 Moore, H . A . , 290, 332 Moore, M.I., 193, 224 Morgan, B.B., 118, 151 Morgan, G . , 364, 379 Morris, J.H., 162, 182 Morrison, A . , 339, 352, 379 Morrison, A.M., 212, 213, 225, 310, 312, 332 Morrison, P., 137, 140, 141, 151 Mortimer, J.T., 35, 66 Mount, M.K., 297, 332 Muchinsky, P.M., 7, 26, 39, 46, 51, 54, 66 Mulaik, S . A . , 23, 26 Mullins, T.W., 404, 406, 410, 411, 453 Munsterberg, H., 10, 26 Murphy, G., 395, 453 Murray, H.A., 47, 66 Murrell, K.L., 131, 151 Myrdal, G., 385, 393, 453

467

N Nacoste, R.W., 339, 379 Nadler, D.A., 256 Nalbadian, J., 339, 379 Nasby, W. 47, 66 Nathan, P.E., 276, 281 Naylor, J.C., 123, 151 Neider, L.L., 199, 204, 225 Nelson, D., 49, 66 Nelson, W. J., Jr., 259, 266, 281 Neuman, J.H., 46, 67 Newcomb, T.M., 46, 67 Newman, J., 40, 67 Newman, J.M., 398, 404, 408, 410, 411, 453 Nicholas, J.M., 131, 152 Nieva, V.F., 120, 122, 152 Nord, W.R., 20, 26 Noe, R . A . , 209; 213, 225, 288, 294, 295, 332 Novak, P.W., 428, 454 Novelli, W.D., 275, 281 Nutt, P.A., 212, 225 Nyquist, L., 315, 316, 332 0

O'Carolan. T.. 317. 332 Offerman,'L.R., 9 i , 114, 364, 379 Olian, J.D., 293, 332 Oppenheimer, R.J., 225 Orpen, C., 40, 67 Ott, E.M., 309, 332 Overholser, G., 311, 332 P Paludi, M . A . , 290, 332 Parasuraman, S . , 315, 332 Parkes, K.R., 165, 182 Parkinson, R.S., 274, 282 Parsons, C.K., 405, 406, 454 Parsons, K.L., 127, 152 Parsons, P. C., 43, 67

Name Index

468

Parsons, T., 79, 86 Paternoster, L., 287, 333 Paul, C.F., 143, 152 Paul, W.J., 54, 67 Paunonen, S.V., 37, 67 Perrin, F.A.C., 41, 67 Perrow, C., 80, 86 Peters, L.H., 200, 225, 311, 333 Peters, L.R., 40, 67 Peters, R.S., 23, 26 Petersen. W.. 343. 379 Pettigrew, T., 339; 340, 347, 351. 352. 379 Pettigrew, T.F., 349, 350, 380, 394, 454 Pfeffer, J., 83, 86 Phinney, J.S., ~340,34 1, 342, 380 Pierce, J.L., 158, 182 Pietromonaco, P.R., 314, 333 Piner, K.E., 430, 454 Plyman, J.S., 275, 282 Polanyi, M., 18, 19, 21, 26 Porras, J.I., 118, 125, 141, 143, 152 Porter, L. W., 46, 67 Postman, L., 47, 67 Powell, G.N., 309, 312, 313, 321, 323, 333 Poyatos, F., 356, 380 Price Waterhouse v Hopkins, 417,454 Pritchard, R.D., 122, 152 Pryor, J.B., 320, 321, 333 Pulakos, E.D., 305, 333, 408, 409, 454 Pulakos, L.H., 51, 67 ,

I

Q

Quattrone, G.A., 350, 380 Quayle, D., 271, 275, 282 Quereshi, M.Y., 414, 415, 454 R Radecki, C., 321, 333

Ragins, B.R., 289, 294, 295, 333 Rahe, R.H., 263, 282 Raju, N.S., 217, 225 Ramsey, S., 356, 380 Rand, T.M., 39, 68, 408, 454 Raza, S.M., 413, 423, 454 Reed, B.B., 307, 333 Reid, P.T.,362, 380 Reingen, P.H., 424, 454 Reskin, B.F., 286, 287, 288, 289, 295, 297, 298, 300, 303, 304, 319, 320, 322, 323, 333 Rhodes, S.R., 259, 282 Richardson, S.A., 432, 454 Riess, M., 45, 68 Riggio, R.E., 267, 268, 282 Rickard, T., 432, 454 Ripkey, D.R., 190, 225 Robbins, S.P., 364, 380 Robins, T.G., 266, 282 Robinson, P.K., 277, 282 Robison, J.T., 318, 333 Rokeach, M., 402,454 Rom, W.N., 265, 282 Romesburg, H.C., 16, 26 Roos, P.A., 300, 303, 333, 334 Rose, G., 429, 433, 454 Rose, P.I., 363, 380 Rosenbaum, M.E., 57, 68 Rosenberg, L., 274, 282 Rosener, 312, 334 RosenthaI, R., 16, 26, 170, 182, 393, 455 Ross, C.E., 316, 318, 334 Ross, J., 417, 421, 455 Ross, L., 46, 68 Rossi, P.H., 234, 256 Roszell, P., 419, 455 Rothbart, M., 350, 380 Rothman, M., 271, 275, 282 Rotter, J., 155, 182 Rothblum, E.D., 43, 68 Rowan, B., 74, 86

Name Index

469

Royce, A.P., 343, 380 Royce, J.R., 27 Rubin, L.J., 320, 334 Rumsey, N., 427, 455 Rusbult, C.E., 56, 68 Russel, D., 431, 455 Russell, J.E.A., 31 1, 334 Russell, J.S., 218, 225 Russo, N.F., 317, 334 Rynes, S . , 413, 423, 455

S S M , L.M., 193, 215, 225

Sagar, H.A., 350, 380 Salamone, F.A., 345, 346, 380 Salas, E., 118, 120, 152 Sampson, E.E., 18, 27 Sandelands, L.E., 164, 183 Sandberg, D.E., 289, 334 Sapolsky, A., 49, 68 Saunders, C.S., 310, 334 Scandura, T.A., 203, 225 Schachter, S . , 50, 68 Schein, E.H., 152, 364, 381 Schein, V.E., 309, 334 Scherbaum, C.J., 310, 334 Schermerhorn, R.A., 347, 381 Schmidt, EL., 2 17, 225, 398, 408, 409, 455 Schmidt, P., 339, 381 Schmitt, N., 40, 69, 405, 406, 410, 455 Schneider, C.R., 434, 435, 455 Schneider, H., 216, 226 Schneier, C.E., 193, 197, 226 Schofield, J.W., 339, 365, 381 Schunk, D.H., 155, 183 Schwalbe, M.L., 171, 183 Scott, W.R., 72, 76, 77, 78, 80, 86 Schultz, D.P., 7, 27 Seager, J., 296, 297, 334 Sears, D.O., 387, 403, 404, 455 Sedlmeier, P., 27

Segal, M.W., 34, 69 Seifert, C., 323, 334 Sekaran, U., 315, 334 Selznick, P., 72, 86 Shahani, C., 413, 421, 455 Shames, G., 209, 210, 226 Shapiro, A.P., 265, 282 Shaw, E.A., 38, 69 Shaw, J.B., 356, 357, 381 Sherif, M., 349, 381 Sherman, S.J., 349, 381, 391, 455 Shore, R., 4, 20, 27 Siegel, L., 7, 27 Sigelman, L., 298, 334 Singh, J.V., 83, 86 Simas, K., 37, 69 Simonson, R., 340, 381 Smeaton, G., 57, 69 Smith, K.K., 381 Smith, P.B., 358, 367, 381 Sniderman, P.M., 400, 401, 402, 455, 456 Snyder, M., 423, 424, 430, 456 Sommerland, E.A., 351, 381 Sorrentino, C., 317, 334 Spence, J.T., 306, 335 Spencer, B.A., 335 Spickard, P.R., 343, 381 Spilerman, S . , 214, 226 Spitze, G., 288, 335 Springbett, B., 413, 456 Stagner, R., 4, 6, 11, 12, 20, 27, 35, 36, 69 Stake, J.E., 290, 335 Statham, A . , 311, 335 Staw, B.M., 69 Steel, L., 290, 335 Steel, R.P., 170, 171, 183 Steiner, I.D., 120, 122, 123, 127, 152 Stephan, W., 350, 381 Stern, P.N., 318, 335 Stevens, G., 309, 335

Name Index

470

Stevens, W.K., 269, 270, 282 Stewart, E.C., 355, 381 Stewart, L.H., 246, 256 Stewart, L.P., 295, 296, 335 Stone, C., 429, 430,432,456 Stone, D.L., 429, 432, 435, 456 Stone, E.F., 215, 226, 387, 398, 429, 432, 435, 456 Strober, M., 263, 282 Strube, M.J., 200, 226 Stumpf, S.A., 165, 166, 167, 183 Stogdill, R.M., 50, 52, 69, 95, 111, 114 Stutzman, T.M., 100, 101, 114 Sue, S., 340, 341, 381 Summers, T.P., 299, 335 Sundstrom, E., 123, 152 Sutherland, M.E., 396, 439, 456 Swezey, R.W., 126, 152

173, 183 Thornton, G.C., 111, 91, 114, 197, 226 Thomas, R.R., Jr., 339, 340, 361, 382 Tiedje, L.B., 316, 336 Ting-Toomey, S . , 356, 382 Tolbert, P.S., 77, 86 Triandis, H.C., 340, 344, 346, 354, 355, 357, 382, 383 Tringo, J.L., 432, 457 Triplet, R.G., 433, 457 Tsui, A.S., 295, 305, 311, 336 Tucker, S . , 293, 336 Tudiver, F., 260, 263, 283 Tully, S . , 340, 383 Turban, D.B., 51, 69 Turner, J.H., 79, 86 Tushman, M.L., 124, 153, 187, 189, 226

T

Xt, M., 314, 335 Thgalakis, V., 435, 456 Tkjfel, H., 162, 183, 344, 347, 349, 350, 351, 367, 382, 394, 45 6 W,R., 340, 362, 382 Tangri, S .S., 320, 321, 322, 323, 335 Tannenbaum, A.S., 355, 382 Tannenbaum, S.I., 120, 123, 152 Thylor, D.M., 345, 351, 362, 382 Thylor, M.J., 41, 69 Thylor, M.S., 161, 183 Thylor, S . , 261, 262, 264, 283 'hylor, S . H . , 297, 298, 335 Teasdale, J.D., 47, 69 Technical & Skills Training News, 216,226 Terborg, J.R., 50, 69 Terpstra, D.E., 40, 69, 130, 138, 153, 320, 323, 335 Thomas, D.R., 382 Thomas, K.W., 162, 164, 167,

U U.S. Department of Labor, 212, 226, 285, 336, 339, 383 V Van denBerghe, P., 344, 383 Van de Ven, A.H., 256 Van Maanen, J., 147, 173 Vecchio, R.P., 202, 204, 226 Velten, E., 30, 70 Venes, A.M., 43, 70 Vernon-Gerstenfeld, S.,291, 336 Vojtecky, M.A., 266, 283 Vollmer, F., 290, 336 von Baeyer, C.L., 292, 336 Voydanoff, P., 317, 336 Vroom, V.H., 198,201, 226 W Wagner, F.R., 155, 183 Wallston, B.S., 319, 336 Walster, E., 41, 70 Wanous, J.P., 58, 70

Name Index

Wapner, S. , 23, 27 Wan; P.B., 163, 171, 175, 183 Washburn, P.V., 44, 70 Waters, J., 413, 457 Watson, C., 312, 336 Webb, E.J., 387, 398, 457 Weber, M., 73, 86 Webb, E.J., 230, 238, 241, 256 Wbster's new collegiate dictionary, 15, 27 Weick, K.E., 212, 227 Weigel, R.H., 400, 401, 402, 439, 440, 457 Weinberg, R.S., 155, 183 Weiner, B., 155, 183 Werther, W.B., Jr., 262, 283 Werts, C.E., 35, 70 West, M.A., 50, 70 Wexley, K.N., 186, 192, 193, 205, 208, 215, 227, 408, 457 White. R.W.. 175. 183 Wiedenfeld, S.A.,' 163, 183 Wilder, D.A., 349, 350, 383 Williams, K.J., 22, 27, 91, 92, 93, 94, 155, 157, 158, 162, 171, 172, 174, 177, 183, 184 Williams, R.E., 339, 394 Winter, S.L., 12, 27 Wittig, M.A., 336 Wood, R., 156, 157, 158, 161, 163, 165, 166, 170, 172, 173, 184

47 1

Woodman, R.W., 119, 128, 129, 131, 137, 139, 142, 144, 153 Worchel, S., 350, 384 Workie, A . , 50, 70 Whyte, W.W., 34, 70 Wright, C.C., 275, 283 Y Yamamoto, T., 23, 27 Yammarino, F.J., 305, 306, 336 Yancey, W.L., 343, 384 Yee, D.K., 289, 336 Yetton, P.W., 117, 153 Yoder, J.D., 309, 336 Young, D.M., 44, 70 Yount, K.R., 306, 337 Yukl, G.A., 95, 98, 111, 114, 173, 184, 186, 191, 195, 227

Z Zajonc, R.B., 29, 30, 34, 53, 58. 70 ZavAa, K.J., 275, 283 Zedeck, S., 313, 314, 316, 317, 318, 337 Zenner, W.P. , 345 , 384 Zigler, E., 317, 337 Zucker, L.G., 72, 75, 76, 77, 80, 86, 87 Zuckerman, M., 46, 70, 112, 115

This Page Intentionally Left Blank

473

Cost containment, 259 See also Absenteeism; Employee turnover; Insurance A premiums; Productivity Absenteeism, 259, 262, 264, 265, Cultural differences, 354-359 266, 267, 271, 272, 273, 277 Cultural identity, 343-344, 346 Accent, 39 Culture, 354-356 Acquired Immune Deficiency Syndrome, 274 D Affect, 29-31, 46-48 Data Aging, 260-278 qualitative, 234 Alcohol abuse, 260, 271, 275quantitative, 234 276 Disability insurance, 259, 262, Appearance, 3 10-31 1 266, 267,271, 272, 274, 277, S e e also Physical appearance; 278 Physical attractiveness Discriminant validity, 236 Archival measures, 242-243 Drug abuse, 260, 271, 275-276 Assessment centers, 95-96, 99, Dual labor market, 302 205-206 Assumptions, of field, 18 E Attitude similarity, 3 1, 36 Education, effects of, 290-291 Attraction, 30-36, 49-53 Employee assistance programs, 260-275 C Employee turnover, 262, 264-267, Career choice, 289-292 273, 274, 277 (beer development, 294-296 Employmentlunemployment ladders and promotions, 295handicaps and, 425 296 Ethnic diversity concepts of, 342-348 mentors and, 294-295 Categorization, 348, 359-363, Ethnicity, 212-214, 341 391, 394, 422 concepts of, 342-348 motivation to categorize, group level of, 345-346 394-395 individual level of, 344-345 Child care, 316-318 intergroup level of, 346-348 Clothing, 43 organizational psychology Cognition, 47-48 and, 341-342 Comparable worth, 297-298 Evocation, 93 Compensation theory, 3 13 Extraversion, 11 1 Connective (altercentric) leadership, 3 12 F Contingency theory of leaderFace validity, 239-240 ship, 200-202 Family, work and, 313-318 Contracting, 23 1-232, 236-237 Feedback, performance, 171-172, Convergent validity, 236 175

SUBJECT INDEX

Subject Index

474

Flexibility, 239-240 Focus group, 245-246 Force field analysis, 246-247

G Gatekeeping transactions, 319 Gender, 37-38, 285-323 See also Gender differences; denigration, 292-293 differences in compensation, 296-299 discrimination, 296-298 health and, 318-319 labor force participation and, 285-287 role congruence, 293-294 roles, 306-307 Gender composition of the workplace, 322 Gender differences attitudes and, 305-306, 309310 compensation and, 296-299 managerial level and, 309-313 perceived, 307-309 performance and, 305 stress and, 318-319 Gender roles, 306-307 Goal setting, 164-165, 169-170 Good Old Boys Sociability (GOBS), 296 H Handicaps admission of, 435-436 bases for discrimination against individuals with, 427-429 defined, 426 effects of type of, 432 organizational research on, 429-434 Health, work and, 318-319 Health Dromotion Drograms. 260. I

"

273-276 Household responsibilities, 315316 Human resource managament, 185 I Individual/group confound, 361 Individual models of sexual harassment, 322 Individual theories, of gender disparity, economic, 301 noneconomic, 300 Injuries in the workplace, 259-260 Institutional isomorphism, 73-75 Institutional level of analysis, 79-80 Institutional theory, 71-84 Instrumental theory, 313 Insurance premiums, 259, 262, 271, 272, 274, 278 Intercultural contact, 360-363 Intergroup theory, 353-354 Internal labor market theory, 303 Interorganizational networks, 82-83 Interpersonal skills training, 206-207 Interviews, 39-44, 244-245 Intrinsic motivation, 162 J Job application, 36-39 Job enrichment, 167-169 Job evaluation ratings, 297-298 Job performance, 159-161 Job satisfaction, 46-49 Job stress, 163, 165 See also Stress Just-world hypothesis handicam and. 428

Subject Index

L Leadership, 173, 186-190, 200-205 connective (altercentric), 3 12 Leadership motive pattern, 95 Letters of recommendation, 39

475

Organizational field, 73-74, 82 Organizational models, 235 Organizational theory, 71-72, 83-84

P Path-goal theory, 198-199 Pay disparity, 296-298 see also salary M Management, 186-190 Performance ratings handicaps and, 430-43 1 Management style, 311-312 physical attractiveness and, Managerial effectiveness, 93-96 Managerial traits, 98 417-418, 424 Manipulation of environments, 93 race and, 404-407, 409-410 Mentors, 294-295 Persistence, task, 164 Mission, of field, 10 Personality, the five-factor Multinational managerial model of, 96-112 Personnel practices, 297 training, 209-211 Multiple roles, 314-315 Personnel selection, 36-44 Physical appearance, 4 1-43 N See also Appearance; Needs assessment Physical attractiveness managerial, 191-197 Physical attractiveness See also Appearance; organizational, 229-247 Nonreactivity, 240 Physical appearance organizational research on, Null education environment, 291 412-425 0 weight and, 43, 414 Obesity, 43 Positivism, 15 Observation, 241-242 Prejudice individual differences in, Occupational illness, 272 prevention of, 266-267, 395 Productivity, 46-47, 259, 265, 269-27 1 266, 271, 272, 273, 274, 277 types of, 265-266, 268-270 Promotion of personnel Occupational sex segregation, 287-289 physical attractiveness and, consequences of, 288-289 417-419 organizational effects on, Propinquity, 32-34 292-294 Pygmalion effect, 170 theories of, 300-305 Q Organizational commitment, 162 Organizational diagnosis, 229, Questionnaires, 243-244 231, 232

476

R Race, 40 organizational research on, 405-4 1 1 Racism nature of, 400-401 old-fashioned or traditional, 400 prevalence of, 40 1-402 symbolic, 400 unobtrusive studies of, 402-403 Rational bias theory, 302 Rationalized institutional elements, 77, 80 Recruiting and selection of personn el handicaps and, 430-431, 433 physica1 attractiveness and, 412-414, 416-417 race and, 406, 408-41 1 Regression analysis, 297 Reinforcement-affect model, 3031 Repeated exposure, 34-35 Research on stigmas effect sizes, 399, 420-421 handicaps, 429-438 measurement issues, 398 obtrusive studies, 387, 398, 408, 41 1 paucity of organizational research, 442 physical attractiveness, 4 12-425 research needs, 422, 437-438, 443 unobtrusive studies, 430 Resumes, 36-39 Retirement, 262 Role models, 291-292

Subject Index

S Salary or earnings See also Pay disparity handicaps and, 425 physical attractiveness and, 419-420 Scientific method, 16 Segmentation theory, 313 Selecting managers, 190-191 Selection, of environments, 93 Self-competence, 155-184 Self-efficacy, 95, 155-184 Self-esteem, 158 Self-evaluation, 156-158 Self-fulfulling prophecies, 393, 423 Self-monitoring, 95 Sex segregation, 298-299 Sexual harassment, 320-323 continuing work condition, 320 perceptions and consequences of, 320-322 quid pro quo, 320 theories of, 322-323 Shared history of organizations, 75-77 Situational leadership theory, 203-205 Smoking, 260, 272-274 Social cognition, 345, 349, 391 Social identity, 343-344 Social interaction and reduction in bias and prejudice, 435, 439 with stigmatized individuals, 435,439 Socialization of gender roles, 289-290 work, 174-175 Societal-culturd theories of gender disparity, 304 of sexual harassment, 322 ~~

~

Subject Index

Spillover theory, 313 Statistical discrimination, 301302 Stereotypes, 392, 423, 437 motivation to stereotype, 394-395 Stigmas consequences of, 395-396 defined, 388 dimensions of, 390 overcoming effects of, 415, 424-425 reactions to individuals with, 397 types of, 389-390 Stigmatization, 389 Stress, 260 S e e also Job stress consequences of, 261, 263, 264 management of, 264-265, 268 nature of, 261 sources of, 262-263, 268 types of work and, 261-262 work and, 318-319 S tructural/organizational theories of sexual harassment, 322 Structural theories, of gender disparity, 301-305 Structure, of field, 4 pedagogical, 6 professional, 9 related to theory, 11 T nutological reasoning, 79 Team building, 119-120, 128-147 effectiveness, 120, 125-128, 140-146 interventions, 139-140, 143145

477

Teams, 118-148 Technological advances, 211-212 See also Technology Technology, 260, 267 Theories of gender-related inequality, 300-305 individual, 300-301 structural, 301-305 societal-cultural, 304 Theories of sexual harassment, 322-323 Theory, inductive vs. deductive, 13 Tokenism, 307-309 Toxics, 260, 265-267 Training effectiveness, 214-218 interpersonal skills, 206-207 managerial, 190-191, 198-219 multinational managerial, 209-2 11 team, 120-128 Triangulation, 238 Turnover, 53-57 U Unfair discrimination laws relating to, 386, 441 organizational norms about, 439

V Validity, 236, 239-240 Vertical dyad linkage theory, 202 Video display terminals, 260, 262, 267-269

w Worker's compensation S e e Disability insurance

This Page Intentionally Left Blank