Misbehaviour and Dysfunctional Attitudes in Organizations

Misbehaviour and Dysfunctional Attitudes in Organizations Edited by

Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky

Misbehaviour and Dysfunctional Attitudes in Organizations

This page intentionally left blank

Misbehaviour and Dysfunctional Attitudes in Organizations Edited by

Abraham Sagie Shmuel Stashevsky and Meni Koslowsky

Selection, editorial matter and Chapter 1 © (the estate of) Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky 2003 Individual chapters (in order) © Kathryne E. Dupré and 2003 Julian Barling; Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar; Shmuel Stashevsky and Jacob Weisberg; Yair Amichai-Hamburger; Thomas J. Kalliath; Lonnie Golden; David G. Schwartz; Bella L. Galperin; Yoav Vardi and Ely Weitz; Lisa M. Penney, Paul E. Spector and Suzy Fox; Deniz S. Ones and Chockalingam Viswesvaran; Muhammed Jamal and Vishwanath V. Baba 2003 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2003 by PALGRAVE MACMILLAN Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010 Companies and representatives throughout the world PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN 0–333–98409–9 This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Misbehaviour and dysfunctional attitudes in organizations / edited by Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky p. cm. Based on papers from the 7th bi-annual meeting of the International Society for the Study of Work and Organizational Values (ISSWOV) in Jerusalem, Israel on June 2000. Includes bibliographical references and index. ISBN 0–333–98409–9 (cloth) 1. Organizational behaviour. 2. Corporate culture. 3. Work ethic. 4. Work environment. I. Sagie, Abraham. II. Stashevsky, Shmuel, 1951– III. Koslowsky, Meni, 1946– HD58.7.M56 2003 302.3′5—dc21 2003043603 10 12

9 11

8 7 10 09

6 08

5 07

4 3 06 05

2 04

1 03

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham and Eastbourne

In memoriam We would like to dedicate our book to Professor Abraham (Rami) Sagie, a professional, a colleague, and a friend, who passed away a few months before the final version of the book was printed.

This page intentionally left blank

Contents

Preface

ix

Notes on the Contributors

xi

1 Introduction: Misbehaviour in Organizations Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky

Part I

1

The Many Faces of Work Misbehaviour

2 Workplace Aggression Kathryne E. Dupré and Julian Barling 3 Comparing Sexual Harassment to Other Forms of Workplace Aggression Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar

13

33

4 Covering-Up Behaviour in Organizations Shmuel Stashevsky and Jacob Weisberg

57

5 Understanding Social Loafing Yair Amichai-Hamburger

79

6 Job Burnout and Dysfunctional Work Attitude Thomas J. Kalliath

103

7 Flexible Work Schedules and Their Impact on Employees Lonnie Golden

122

8 When Bad Email Happens to Good People: A Case of Information Technology Mismanagement David G. Schwartz 9 Can Workplace Deviance Be Constructive? Bella L. Galperin

Part II

138 154

The Dynamics of Organizational Misbehaviour

10 Personal and Positional Antecedents of Organizational Misbehaviour Yoav Vardi and Ely Weitz vii

173

viii Contents

11 Stress, Personality and Counterproductive Work Behaviour Lisa M. Penney, Paul E. Spector and Suzy Fox

194

12 The Big-5 Personality and Counterproductive Behaviours Deniz S. Ones and Chockalingam Viswesvaran

211

13 Type-A Behaviour Pattern: A Canadian Study Muhammed Jamal and Vishwanath V. Baba

250

Author Index

263

Subject Index

272

Preface As with just about any serious enterprise that individuals take upon themselves, it is often easier to see what the final product will look like than appreciate and anticipate the process involved in reaching the desired goal. When the three of us agreed to author an edited book on organizational misbehaviour, it was visualized as a relatively straightforward project that would be completed in a relatively short time-frame. We knew what we wanted: a book that would cover many, if not most, of the misbehaviours that are part of organizational life; it would involve a few phone calls, some secretarial help, and some editing. Even the book’s contributors were known to us: they would be chosen from the presenters at the conference of the International Society for the Study of Work Values in June of 2000. The three editors of the present volume are active in the organization and were involved in organizing the June conference and had discussed the possibility of producing a book before the meetings officially began. It seemed so natural to use the best papers, many of which had focused on various aspects of misbehaviour, as the material that would be included in the book. After making contact with them and inquiring whether the presentation could be expanded into a formal chapter, nearly all agreed to write a chapter for the book. Yet reality often does not conform to expectations. After several planning meetings that focused on the book’s purpose, specific content and potential contributors, it became clear that some important subtopics were missing. We decided to speak to several leading researchers in the field and, after some period of uncertainty, were pleased to get affirmative responses and an agreement to contribute to the book in each of their areas of expertise. This obstacle was a minor one compared with the problems we encountered in the reviewing and editing process. Although the book was to be an edited one and we had pretty much chosen whose contributions would be included, it was also agreed, and each contributor had been informed as such, that all submitted articles would have to undergo an objective review process that would include input from two judges, of which at least one would be an outsider and not an editor. Criteria included the value of the research that was being reported, the quality of the writing, and its appropriateness to the book as a whole. Several contributions were rejected because of ix

x Preface

these criteria and, just as importantly, major revisions were required from others. This took an inordinate amount of time, more than is usually typical of a journal article that has been returned to an author for revision. Thus, a project that was started in the autumn of 2000 and which we expected to be ready by the end of that academic year, or, the latest, by the beginning of the next academic year, 2001–2, did not arrive at the publishing house till the beginning of 2003. Accordingly, the book would be available to the public only after three years after it was first planned had passed. To tell the truth, we were not prepared for this delay and sometimes were not sure when and, even if, the book would be finished. Why had the process taken so long to complete? Unlike a book written by an author or authors, an edited work is dependent on others. This requires coordination and patience that is unnecessary in the usual book. This, however, explains only part of the delay. If allowed to conjecture a little more about the process, it appears to us that some of the contributors were rather surprised, perhaps, taken aback, that a contribution to a book that had been, in their minds, a certainty needed much more work. At the end, each article was modified as requested and we are convinced that the work you are about to read is of a high, professional, quality that will be an important addition to your professional library. ABRAHAM S AGIE S HMUEL STASHEVSKY MENI K OSLOWSKY

Notes on the Contributors Vishwanath V. Baba is Professor of Organizational Behavior and Dean of the Michael G DeGroote School of Business at McMaster University, Hamilton, Canada. Prior to coming to McMaster, he had been editor-inchief of the Canadian Journal of Administrative Sciences. He is currently on the editorial board of the Canadian Journal of Administrative Sciences and the Journal of Organizational Behavior. An engineer by training, he obtained his PhD in organizational behaviour from the University of British Columbia. He is widely published in Canada and abroad. He is currently involved in a programme of research on occupational mental health. Julian Barling is the author of several books, including (with Mike Frone) The Psychology of Workplace Safety, and is the author/editor of well over 100 research articles and 25 book chapters. He is the editor of the American Psychological Association’s Journal of Occupational Health Psychology, and is currently chair of the American Psychological Association’s Task Force on Workplace Violence. In 1995, and again in 1997, he received the annual award for ‘Excellence in Research’ from the School of Business, Queen’s University. In 2002, he was named as the Queen’s Research Professor, and was also elected as a Fellow of the Royal Society of Canada. Lynn Bowes-Sperry is an Assistant Professor of Management at Western New England College in Springfield, Massachusetts. Her research interests include responses to unethical behaviour in the workplace, sexual harassment and other forms of workplace aggression, and organizational justice. She has published several articles in these areas, including recent publications in Human Resource Management Review, Journal of Management, Group and Organization Management, Labor Law Journal, Psychological Reports and Small Group Research. Before commencing her academic career, Lynn was an HR manager. She received her PhD from the University of Connecticut in 1996. Kathryne E. Dupré is a doctoral student in organizational behaviour at Queen’s University. She completed her master’s degree in industrial/ organizational psychology at Saint Mary’s University in Halifax, Nova xi

xii Notes on the Contributors

Scotia. She has published in the Journal of Applied Psychology. Her current research focuses on workplace aggression and violence. Suzy Fox is Assistant Professor at the Institute of Human Resources and Industrial Relations, Loyola University Chicago. She received her PhD in I/O Psychology from the University of South Florida in 1998. Her research areas focus on the roles of emotion and well-being in the workplace, including emotional and behavioural responses to job stress, work constraints, conflict, and perceived injustice. She is also engaged in cross-cultural, international studies of leadership and organizational success among women, work–life conflict, and an exploration of the relationship between subtle/symbolic/modern racism and bullying/ mobbing in the workplace. She is associate editor of Human Relations. She has published articles in the Journal of Organizational Behavior, Organizational Dynamics, Journal of Vocational Behavior, Human Resource Management Review, Journal of Occupational and Organizational Psychology, Journal of Occupational Health Psychology, and the Handbook of Organization Studies. She has presented her work at meetings of the Academy of Management, American Psychological Association, Society for I/O Psychology, International Conference on Emotions in Organizational Life, International Western Academy of Management (in Lima, Peru), the Stetson University 1999 Gender Issues in Business Day, International Conference on the Labor Market in the New Millennium in Gdansk, Poland, and the National Conference on Law and Higher Education. Bella L. Galperin is an assistant professor of international management in the International Business Department at Rollins College, USA. Her research areas include: cross-cultural management, workplace deviance, innovation and entrepreneurship. She has presented her work at a number of professional meetings in the United States, Canada, Mexico, Turkey, the Caribbean, Israel, and Poland. She also served as Division Chair, Program Chair and Academic Reviewer for the International Business Division of the Administrative Sciences Association of Canada. She has worked as a consultant to firms in the telecommunications, pharmaceutical, and clothing industries. Lonnie Golden is Associate Professor of Economics at Penn State University, Abington College. His research primarily focuses on working hours, labour market flexibility and the non-standard work force. He is co-editor of the recent books, Working Time: International Trends, Theory and Policy (Routledge) and Nonstandard Work: The Nature and Challenge

Notes on the Contributors xiii

of Changing Employment Arrangements (Industrial Relations Research Association, Cornell University Press). His current research studies include, ‘Mandatory Overtime Work and Needed Policy Reforms’ and ‘A Behavioral Economic Approach to Understanding Rising Hours of Labor’. His articles have appeared in journals such as the Monthly Labor Review, American Behavioral Scientist, Review of Social Economy, International Journal of Forecasting, American Journal of Economics and Sociology and Journal of Post Keynesian Economics. He serves on the editorial board of the Review of Social Economy. Yair Amichai-Hamburger is a social industrial psychologist. He is a lecturer and researcher at Bar-Ilan University and works as an industrial consultant for a number of large companies. He received his doctorate from Oxford University, England. His research interests include the interaction between human factors and technology, leadership and emotions, organizational conflict, and behaviour in organizations. He is a member of the editorial board of Computers in Human Behavior. Muhammed Jamal is Professor of Management in John Molson School of Business, Concordia University, Montreal. His research and consulting are in the areas of burnout, stress management and cross-cultural management. He has published in many professional journals including Journal of Organizational Behavior, Arab Journal of Administrative Sciences, Organizational Behavior and Human Decision Processes, Personnel Psychology, Stress Medicine, Human Relations, International Journal of Stress Management and Human Resource Management. He has presented over 75 papers in national and international conferences. He has been a member of the editorial board of the Canadian Journal of Administrative Sciences and the Journal of Organizational Behavior. Thomas J. Kalliath has qualifications in psychology and management from the Washington University in St Louis, USA, and Xavier Labor Relations Institute in India. He is the co-editor of the textbook Organizational Psychology in Australia and New Zealand (Oxford University Press), and has co-authored numerous publications in refereed journals in the areas of job burnout, innovations in teaching, and organization development. He has been teaching organizational psychology, consulting and organizational change and related courses at the University of Waikato, Hamilton, New Zealand, since 1995. He is the founderconvener of the New Zealand Organization Development Network, which holds six-monthly conferences to share innovations in organization

xiv Notes on the Contributors

development and change. He has over 25 years of work experiences as a hospital administrator, organizational development consultant, and academic in India, the USA and New Zealand. Meni Koslowsky is Professor in the Department of Psychology at Bar-Ilan University with interests in organizational behaviour and emphasis on withdrawal, power/influence techniques, and research methodology. He has written four books and has more than 100 articles to his credit. He has made numerous presentations at professional meetings. Very active professionally in several international I/O organizations, he is currently the president-elect of the International Society for the Study of Work and Organizational Values. Harsh K. Luthar is an associate professor in the management department at Bryant College. His teaching and research centres on human resource management, business and society, and related areas. His research interests include sexual harassment perceptions, cultural and racial diversity issues, and the role of ethics in business. He has published in Human Resource Management Review, International Journal of Human Resource Management, HRMagazine, Workforce Magazine, Employee Responsibilities and Rights Journal, Journal of Business Ethics, Sex Roles: A Journal of Research and Journal of Individual Employment Rights, among others. He has been interviewed in both the television and print media and cited in a variety of newspapers including the Chronicle of Higher Education. Deniz S. Ones is the Hellervik Professor of Industrial Psychology at the University of Minnesota. She is the author of over 75 articles and over 200 international/national conference papers and published abstracts on topics that include personality at work, integrity testing to performance measurement, influences of social desirability on psychometric properties of non-cognitive measures, and counterproductive behaviours at work. She has received the best dissertation award as well as the early career distinguished scientific contributions award from the Society for Industrial and Organizational Psychology (SIOP). She is a Fellow of both Divisions 5 (Evaluation, Measurement, and Statistics) and 14 (Industrial and Organizational Psychology) of the American Psychological Association. She serves on the editorial boards of Journal of Applied Psychology, Personnel Psychology, Journal of Organizational Behavior, European Journal of Work and Organizational Psychology and Zeitschrift für Personalpsychologie (Journal of Personnel Psychology). She is the current editor-in-chief of the International Journal of Selection and

Notes on the Contributors xv

Assessment. In 2001 and 2002, she edited the two-volume Handbook of Industrial, Work and Organizational Psychology (with Anderson, Sinangil and Viswesvaran) (Sage), a special issue of the journal Human Performance on use of cognitive ability tests (with Viswesvaran), and a special issue of the International Journal of Selection and Assessment on counterproductive work behaviours. Lisa M. Penney received her PhD in the Industrial/Organizational programme from the University of South Florida. She is currently a Research Associate at Personnel Decisions Research Institutes, Inc. in Tampa, Florida, and an adjunct faculty member in the Department of Psychology at the University of South Florida. Her research interests include counterproductive behaviour in organizations, leadership, employee motivation, and organizational development. Abraham Sagie was, until his untimely death in 2003, the Director of the School of Business Administration in Bar-Ilan University, Israel, and the co-author of Participation and Empowerment in Organizations: Modeling, Effectiveness, and Applications (Sage). His research interests includes participative decision-making, work and organizational values, and cross-cultural aspects of leadership, motivation, and behaviour in organizations. He served as a Scientific Chair and Secretary/Treasurer of the International Society for the Study of Work and Organizational Values (ISSWOV). Additionally, he was a member of the editorial board of several journals, including Group and Organization Management, International Journal of Manpower, and the International Journal of Cross Cultural Management, and worked as an organizational consultant to various Israeli firms. David G. Schwartz is a senior lecturer and former head of the Information Technology Division of the Graduate School of Business Administration at Bar-Ilan University, Israel. Since 1998 he has served as editor-in-chief of the Journal of Internet Research (www.emeraldinsight. com/intr.htm). His research has appeared in publications such as IEEE Intelligent Systems, International Journal of Human–Computer Studies, IEEE Transactions on Professional Communications, Kybernetes, and the Journal of Organizational Behavior. His books include Heterogeneous Cooperating Systems (Kluwer) and the edited collection Internet-Based Knowledge Management and Organizational Memory (IGP). He received his PhD from Case Western Reserve University, MBA from McMaster University, and BSc from the University of Toronto, Canada.

xvi Notes on the Contributors

Paul E. Spector is Professor and Director of the Industrial/ Organizational Psychology Program at the University of South Florida. His interests include both content (counterproductive work behaviour, employee well-being, job satisfaction, job stress, and personality) and methodology. He has published in many of the leading journals of the field, including Academy of Management Journal, Journal of Applied Psychology, Journal of Management, and Psychological Bulletin. He is an associate editor for Journal of Occupational Health Psychology, and the point/counterpoint editor for Journal of Organizational Behavior. A 1991 Institute For Scientific Information study listed him as one of the 50 highest impact contemporary researchers (out of over 102,000) in psychology worldwide. Shmuel Stashevsky is the director of the International MBA Program in the Graduate School of Business Administration, Bar-Ilan University, Israel. He is active in both academic and business worlds. His research interests include quality management, participatory programmes, organizational behaviour, internet usage and e-commerce, and business games. He serves as the secretary-treasurer of the International Society for the Study of Work and Organizational Values (ISSWOV). He has professional experience in hi-tech entrepreneurship, management of business units, consulting and training, project management, systems analysis, and software development. Jasmine Tata is an Associate Professor of Management at Loyola University Chicago where she serves on the Advisory Board of the Gannon Center for Women and Leadership. Her research interests lie in the areas of organizational justice, organizational communication, team management, and gender issues on organizations. She has published several articles in these areas, including recent publications in the Journal of Management, International Journal of Intercultural Relations, Journal of Managerial Issues, Psychological Reports, the Journal of Psychology, and Group and Organization Management. In addition, she has presented at meetings of professional and industry associations and conducted executive development seminars on project teams. Before commencing her academic career, she practised architecture and worked on several project teams. Jasmine received her PhD from Syracuse University in 1993. Yoav Vardi has been with the Department of Labor Studies, Faculty of Social Sciences, Tel Aviv University, since 1980 and has also served as an associate professor of management at the School of Business

Notes on the Contributors xvii

Administration, Cleveland State University. He received his PhD in Organizational Behavior from Cornell University, NY School of ILR. His main areas of interest are individual misconduct in organizations and career management. He has published articles in the Academy of Management’s Journal and Review, Organization Science, Journal of Business Ethics, Journal of Applied Psychology, and Journal of Vocational Behavior. He recently co-authored a book entitled Misbehavior in Organizations with Dr Weitz. Chockalingam (Vish) Viswesvaran is an associate professor of psychology and the director of the Industrial/Organizational Psychology programme at Florida International University, Miami. He has published over 80 articles and 170 conference presentations in the areas of personnel selection, performance appraisal, occupational stress, as well as on methodological innovations. He serves on the editorial boards of Personnel Psychology, Journal of Applied Psychology, Educational and Psychological Measurement, and Journal of Clinical Child and Adolescent Psychology. He also serves as the associate editor of the International Journal of Selection and Assessment. He has received the best dissertation award as well as the early career distinguished scientific contributions award from the Society for Industrial and Organizational Psychology (SIOP). He is a fellow of SIOP as well as Divisions 14 (I/O) and 5 (Measurement) of the American Psychological Association. He has recently co-edited a two-volume handbook: Industrial, Work and Organizational Psychology. Jacob Weisberg is the Head of Management Studies at the Graduate School of Business Bar-Ilan University, Israel. His current research interests include labour turnover, compensation and incentive schemes, wage determination, leadership in Russia, appointment of CEOs in the US, and labour market issues in Poland. He has published in Journal of Organizational Behavior, Organization Studies, Human Relations, Journal of Vocational Behavior, International Journal of Manpower, Scandinavian Journal of Management, Economics of Education Review, Industrial Relations – Relations Industrielles, etc. Ely Weitz is a senior lecturer at the Department of Labor Studies, and head of the Training and Development Unit, Faculty of Social Sciences, Tel Aviv University. He received his PhD from Tel Aviv University. His interests include business history, management fashions, organizational misbehaviour, training and employee development, organizational change and design. He has published in Organizations and Organization Science. A book entitled Misbehavior in Organizations was recently co-authored with Yoav Vardi.

This page intentionally left blank

1 Introduction: Misbehaviour in Organizations Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky

The idea to publish a book devoted to misbehaviour and dysfunctional attitudes was not created in the mind of one person only. This idea was born in the minds of several researchers who met each other in an international conference gathered to discuss issues related to the linkage between work values and organizational behaviour. This conference was the 7th bi-annual meeting of the International Society for the Study of Work and Organizational Values (ISSWOV) that took place in Jerusalem, Israel, in June 2000. All these researchers believed that one of the most interesting but neglected questions associated with the conference theme is what kind of organizational behaviour can be expected when normative work values are not a deciding factor. Each of these researchers decided, therefore, to address this issue in the conference. As their cultures, backgrounds, disciplines, and areas of expertise varied, these participants tackled the issue from different perspectives. When the conference was over, it was only natural to choose the best articles that addressed the association between work values, or, more correctly, lack of normative work values, and misbehaviour or dysfunctional attitudes in organizations, to add several chapters written by experts that did not participate in the conference, and to aggregate them into one volume. At the time that the Jerusalem conference convened, Enron, the sixth largest energy company in the world, was considered to be a highly successful firm. Enron had $100 billion of business in 2000, and was listed No. 7 on the Fortune 500. Based on its reported revenues during the late 1990s and 2000, Enron became an admired firm; employees in other firms envied their counterparts in Enron, and the company was considered by many as one of the best places to work. During 2001, however, 1

2 Introduction: Misbehaviour in Organizations

Enron’s stock lost more than 99 per cent of its $60 billion value. At the end of 2001, Enron’s fall was the biggest corporate bankruptcy in American history. Why did such a promising company become a smoking ruin within a year? The answer was revealed on 8 November 2001, when Enron issued a report disclosing that since 1997, the company had overstated earnings by approximately $600 million. The company’s top management and its auditor, Andersen, which was listed among the Big-5 accounting firms, were all involved in this fraud. The fraud was translated into big money; as they felt that their investment in the company was risky, top executives sold their stocks for millions of dollars. According to Newsweek, (21 January 2002), between May 2000 and August 2001, Chairman Ken Lay sold stocks for $37,683,887; former CEO Jeff Skilling sold for $14,480,755, and Unit CEO Lou Pal sold stocks for not less than $62,936,552. Simultaneously, the company prohibited its employees from selling their stocks, until the stock had plummeted from $83 in January to 26 cents in November 2001. As a result, numerous employees lost not only their jobs but also their savings. If one asks whether the Jerusalem discussion on values and misbehaviour was timely, one can simply refer to the Enron case. This was a situation where the company’s top executives preached for one set of values but actually promoted another set. While preaching for values like honesty and loyalty to the company and to the employees, the managers were driven by egotism and greed. The wrongdoing, however, did not stop here. According to another article in Newsweek (11 March 2002), power plays, feuds, abuse, and internal spying were widespread phenomena among Enron’s high-ranking officers. The CEO and other top executives acted to remove not only the external competitors but also potential rivals within the company. Furthermore, they created a ‘sexdrenched, out of control’ corporate culture. Gossip about sex suffused the company; it was widely believed that superiors and subordinates used wanted and unwanted sex relationships to achieve instrumental advantages. As top managers were the only people at the time that knew that Enron was vulnerable, one would have expected them to do something to save the company. It appears, however, that their own pleasure and personal benefits took precedence over the company’s fate. The rise and fall of Enron is a case where misbehaviour and disloyalty to the firm, its employees and values, concentrated primarily in the company’s higher echelon. At the same time, other fraudulent actions by chief officers were found at Enron’s accounting firm, Andersen, at

Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky 3

WorldCom, which was caught inflating its revenues, and at Adelphia, whose founder and two sons were accused of defrauding investors. In all these cases, the companies were destroyed by the misconduct, and the very persons who were responsible for running the firms were the ones who were responsible for their destruction. Not surprisingly, money stolen by a senior executive is considerably greater than the sum that could be stolen by an ordinary employee. Yet, organizational misbehaviour is not the sole province of higher-ranking managers. The literature on this topic clearly indicates that rank-and-file workers commit the lion’s share of workplace crimes. Hence, despite the mass media interest in top executives’ offences, this book covers misbehaviours by employees at all organizational levels. Different researchers use various terms to describe the same phenomena; workplace misbehaviour could be referred to as counterproductive, dysfunctional, deviant, disruptive, antisocial, non-compliant, unconventional, or as wrongdoing. The meaning of all these terms is roughly the same; according to Vardi and Wiener (1996, p. 151), each points to an ‘intentional action by members of organizations that defies and violates (a) shared organizational norms and expectations, and/or (b) core societal values, mores and standards of proper conduct’. Other authors (Fox and Spector, 1999; Griffin, O’Leary-Kelly and Collins, 1998; Robinson and Bennett, 1995) emphasize as well the intention to bring harm to the organization or its members. They do not include behaviours like social loafing and emotional responses like job burnout that are typically unintentional. Nevertheless, the negative effects of these behaviours are substantial. For this reason, we include in the term ‘work misbehaviour’ intentional actions such as work violence as well as recurring or typical behaviours that bring harm to the organization and/or its employees even if they are unintentional like social loafing, unsafe behaviour, or job burnout. This view is similar to Giacalone and Greenberg’s (1997) approach that perceives antisocial work behaviour as ‘any behavior that brings harm, or is intended to bring harm, to an organization, its employees, or stakeholders’ (p. vii). Yet, by limiting accidental negative behaviours to recurring or typical ones, we exclude occasional slip-ups and accidentally caused mischief without evil intent. Ackroyd and Thompson (1999) and Brumback (2001) do not include in their definition of misbehaviour negative activities such as fraud and chicanery, if performed by managers. The majority of the researchers do not share this view; similarly, we include misbehaviours by managers, like fraud, as well as those acted on primarily by non-managers (e.g. withdrawal). The full array of misbehaviours, according to our view, is

4 Introduction: Misbehaviour in Organizations

very wide-ranging; it varies from very severe crimes like murder and assault, through theft, sabotage, and fraud, to withholding efforts at work, absenteeism, accidents, and stress. The prevalence of the various types of work misbehaviour is surprisingly high, as attested by the fact that approximately 70 per cent of employees have engaged in some form of destructive deviant behaviour (Bennett and Robinson, 2000). In the year 2000, work violence, including assaults and suicides, accounted for 16 per cent of all work-related fatal occupational injuries in the USA (Bureau of Labor Statistics, 2001). Violent acts are among the top three causes of workplace fatalities for workers. Furthermore, misbehaviour costs organizations more than most managers dare believe. For example, as Penney, Spector and Fox cite (see Chapter 11 in this volume), the costs to American businesses associated with just one type of misbehaviour, employee theft, have been estimated to be more than $200 billion annually (Govoni, 1992). This book is divided into two parts. Part I discusses some of the dysfunctional behaviours. It was not our aim to present all of the work misbehaviours; such an aim would be far too ambitious. Gruys (1999; see Vardi and Weitz’s chapter in this volume), for example, identified no less than 87 different types of misbehaviours and this number does not include some of the types discussed here. Hence, we limited ourselves to some typical and some interesting but atypical forms of misbehaviour. We start with severe misbehaviours like workplace aggression and sexual harassment, and then continue with some of the softer types including covering-up, social loafing, job burnout, absenteeism, and even a case of information technology mismanagement. This part concludes by looking at the good face of misbehaviour: constructive deviant behaviour. Part II of the book concentrates more on ‘why’ than on ‘what’ and ‘how’. This part, called ‘The Dynamics of Organizational Misbehaviour’, analyzes various individual and organizational antecedents of work misbehaviour. The chapters included in this part present theoretical models, empirical investigations, and practical suggestions for coping with misbehaviour. Let’s consider the book contents in detail. Part I starts with an analysis of one of the most disturbing offences: workplace aggression. In the next chapter (Chapter 2), Dupré and Barling portray different forms of aggressive acts, provide some statistics on the diffusion of violence and aggression in organizations, and identify individual and organizational predictors of aggression. Through their analysis of the literature, the authors list the aversive consequences stemming from workplace aggression and how they impact the victims, other employees, the

Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky 5

organization, and the entire society. As Dupré and Barling show, research has provided reliable answers to many questions on workplace aggression; yet, some annoying questions await further research. Does aggression at work escalate? Does aggression at work generalize to other forms of work misbehaviour? Can organizational policies limit the frequency of workplace aggression? What is the frequency of workplace aggression among young workers? The authors conclude the chapter by providing directions for continued research on these issues. Sexual harassment in the workplace is the topic of Chapter 3 that was written by Bowes-Sperry, Tata and Luthar. In light of the empirical evidence, the authors conclude that the probability of harassment increases when the aggressor is low self-monitor, when he is at the same level as the victim, when the victim is unmarried, and when she is perceived by the aggressor to be provocative, submissive, or poorly integrated into the social network. While comparing sexual harassment with other forms of work aggression, the authors propose questions for continued research, including the following: Are individuals who are targets of sexual harassment also targets of other forms of workplace aggression? Do organizations with high levels of sexual harassment also exhibit high levels of other forms of aggression? Do training programmes for eliminating workplace aggression reduce sexual harassment as well? Empirical answers to these questions may help organizations to cope better with sex harassment and with other forms of workplace aggression. Covering-up, i.e. hiding mistakes and wrongdoings from co-workers and supervisors, is the topic of the fourth chapter, co-authored by Stashevsky and Weisberg. Using a new human capital based multidimensional model, the authors identified characteristics of the person, the job, and the organization, that influence covering-up. Hypotheses derived from the model were empirically tested with a sample of 340 employees. According to the results, more covering-up was reported by the following: female employees as compared with their male counterparts; employees below the age of 30 or above 40 as compared with workers in the ages 30 to 40; people employed in firms engaged in the local-market as compared with employees of export firms; and employees of non-growing firms as compared with those employed in growing firms. Hamburger discusses in Chapter 5 the issue of social loafing, or the decline in the degree of effort invested by a group member when he or she shares a collective task. The literature points to three theoretical explanations of this phenomenon, indicating that low effort is invested if: (1) the member believes that extra effort on his/her part is neither

6 Introduction: Misbehaviour in Organizations

necessary nor meaningful; (2) the member compares himself or herself to some standard or to others and finds that the low effort is sufficient; or (3) the member believes that the valence of his or her personal outcome is low. Based on these explanations, the author reviews empirical findings from the social loafing literature. Also, he discusses personal, group, organizational, and cultural correlates of social loafing, indicating that management may prevent, or, at least, significantly reduce, this form of misbehaviour. The chapter concludes with a list of practical recommendations for management as well as suggestions for continued research. Kalliath (Chapter 6) discusses job burnout and its relationships with dysfunctional work attitudes, such as the employee’s job satisfaction, organizational commitment, and supervisor satisfaction. Unlike the misbehaviour forms analyzed in the former chapters, burnout is a case where the same person is simultaneously the aggressor and the victim. Job burnout is typically considered to be an antecedent of the aforementioned attitudes, and thus, as one’s burnout increases, one’s organizational attitudes are affected. This chapter reports on the possibility of a reverse causality path, i.e. as specific attitudes improve, job burnout is reduced. Moreover, two of the phases of burnout were expected to operate individually; thus, after reduction in emotional exhaustion, sense of depersonalization was expected to decrease. Using three samples consisting of 203 nurses, 145 managers, and 110 laboratory technicians, the hypotheses were supported. One important implication is that management and employees can cope with job burnout by enhancing and varying sources of job satisfaction, organizational commitment, and supervisor satisfaction. The seventh chapter, by Golden, addresses the issue of some passive sorts of misbehaviour like lateness and absenteeism from a new perspective – the use of a practical solution. If one accepts that tardiness and absence are a costly problem in organizations, then flexible work schedules (e.g. flextime programmes allowing the employees to vary their arrival and departure times but requiring them to be present at work in the ‘core’ hours) can be perceived as a practical means to cope with the problem. This arouses a series of questions: Do flexible work schedules actually reduce tardiness and absenteeism? Do they increase positive work outcomes such as individual and organizational productivity? Do they increase the employee well-being, morale, and job satisfaction? Are they used frequently in organizations? Are they equally used with diverse occupations and firm types? The chapter addresses all these issues and concludes that there is a high probability for decreasing

Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky 7

misbehaviour and increasing organizational well-being by using flexible work schedules more extensively and in more sophisticated fashions. Chapter 8 by Schwartz describes a case of computerized information mismanagement that caused a great deal of trouble to mailing-list users. Typically, this is not classified as misbehaviour, as we can detect here a lot of victims without any single aggressor. The ironic title of this chapter, ‘When Bad Email Happens to Good People’, exemplified this situation. The story of the bad email messages, which dominated for several days the mailing lists of 352 innocent participants, sounds like the tale of the sixteenth-century Golem (a robot type of creature) of Prague, which tried to take over the community of its inventor, the famous Rabbi Judah Low. It appears that although information technology mismanagement differs from other types of misbehaviour and no one intends to harm the other, such harm actually happens. Who should be blamed for this harm? And, more important, how can the damage be prevented? The chapter addresses these questions and concludes by making some practical suggestions to management and other information technology users. In the last chapter of Part I, Galperin challenges the common belief that workplace deviance, defined as a behaviour that violates accepted norms, is fundamentally negative. The main argument here is that workplace deviance may play a constructive role. For example, employees who aim to innovate and change work procedures may violate existing norms; ultimately, such a constructive deviant behaviour may improve the organization’s creativity, flexibility, and its competitive advantage. Similarly, employees who engage in discrepant behaviours, such as internal whistle-blowing, may violate the present organizational norms but contribute to the overall well-being of the organization. The problem, however, is that very often the same persons are responsible for both constructive and destructive deviances. However, by preventing all types of deviance, management may decrease its own prospects of competing and succeeding. Part II of the current book does not focus on individual forms of work misbehaviour; rather, by analyzing different forms, its four chapters attempt to understand the internal dynamics of these phenomena. In Chapter 10, which opens Part II, Vardi and Weitz introduce both a theoretical model and an empirical study. The former covers a wide array of antecedents of misbehaviour at the following levels of analysis: individual, job, group, organization, and profession. The latter reports the findings of the relationships between several antecedent variables and misbehaviour. Using 250 employees from various organizations, it

8 Introduction: Misbehaviour in Organizations

was found that the higher that one is on the Type-A personality dimension, the higher the misbehaviour; the higher the professional identity, the lower the misbehaviour; the lower the job satisfaction, the higher the misbehaviour; and, most interestingly, the higher the job autonomy, the higher is misbehaviour. The last finding implies that although job autonomy is universally considered a positive state that promotes the workplace productivity, its influence on misbehaviour is rather negative. More employee autonomy means lower supervisors’ control, which can lead to destructive deviant behaviour. Penney, Spector and Fox, devote the eleventh chapter to the analysis of the relationships among job stress, an environmental factor, personality dimensions, and misbehaviour. In general, the interaction of both the environmental stressor and the employee’s personality traits determines not only whether he or she will exhibit counterproductive behaviour but also what the form and magnitude of this behaviour will be. For example, an environmental stressor such as organizational injustice or interpersonal conflict may act as a facilitator of misbehaviour. However, the specific form and magnitude of the misbehaviour is dependent on one’s personality traits. Anger, for instance, may arouse behaviours like complaining, yelling at people at work, refusing to take on assignments, or even damaging property. In contrast, anxiety may arouse behaviours like staying at home, taking longer breaks than allowed, or procrastinating on important projects. Management that takes this information into account may tap personality dimensions and influence environmental variables. Ones and Viswesvaran based their study (Chapter 12) on a comprehensive meta-analytic review of the literature on misbehaviour. They aggregated the available findings into two sets of correlation matrices: intercorrelations among misbehaviours, and intercorrelations between misbehaviours and personality dimensions. The first set of matrices shows that misbehaviours are highly intercorrelated. This implies that one who tends to use some misbehaviour type (e.g. violence) tends to use other types (e.g. destruction of property, poor attendance, unsafe behaviour, misuse of information, etc.). The second set of matrices indicates that work misbehaviours are moderately related to the personality dimensions known as the Big-5, namely, emotional stability, extraversion, openness to experience, agreeableness, and conscientiousness. As compared to the other dimensions, conscientiousness, however, reveals the highest correlations. Furthermore, its two main facets, achievement orientation and dependability, were found to be highly correlated with various indices of misbehaviour. Finally, factor alpha, the higher order

Abraham Sagie, Shmuel Stashevsky and Meni Koslowsky 9

personality trait that is composed of conscientiousness, agreeableness, and emotional stability, exhibited high correlations with misbehaviour. These findings imply that for better screening and selecting results, management should consider the use of personality indices as predictors of misbehaviour. The last chapter, by Jamal and Baba, is also devoted to an analysis of the relationship between personality and misbehaviour. Yet, rather than the Big-5, the focus here is on the Type-A personality dimension and its components: time pressure and hard-driving. The following behavioural variables, indicating the existence (or absence) of work misbehaviour, were used: job stress, health problems, turnover intention, absenteeism, organizational commitment, and job satisfaction. Two samples, consisting of 175 hospital employees and 110 telecommunication workers, participated in the field study. With the exception of absenteeism, the results support the authors’ hypotheses, showing that measures of the Type-A personality dimension and its two components are relatively good predictors of work misbehaviour. As these measures increased, job stress, health problems, and turnover intention also increased. In addition, an increase in the personality indices was related to a decrease in organizational commitment and job satisfaction. One of the interesting conclusions made by some of the current authors is that misbehaviour is not constant but varies over time. It progresses either within the same misbehaviour form (e.g. from covert to overt aggression or sexual harassment) or from one form of misbehaviour (e.g. aggression) to another form (e.g. sexual harassment). Similarly, it appears that misbehaviour has consequences that are not uniform over time. For example, Dupré and Barling (Chapter 2) show that following workplace aggression, the effects are felt over a wide number of victims beyond that of the direct target of the aggression. As with the directly injured person, indirect victims may also suffer from emotional and psychosomatic problems. Furthermore, misbehaviour may yield not only direct but also indirect negative outcomes to the firm, such as reduction in productivity or impaired reputation. It is not far-fetched to assume that some of the negative consequences, such as an increased fear or insecurity, could even be felt in the broader community. Although this is a pessimistic conclusion, it is only one side of the coin. As seen in many chapters of the book, even if not always possible to prevent misbehaviour, management frequently is able to reduce its frequency. For example, management can influence various organizational and situational variables that are related to misbehaviour. These include organizational culture and goals, organizational justice, group

10 Introduction: Misbehaviour in Organizations

norms, personnel selection, training programmes, work information available to employees, control systems, and flexible work schedules. Also, by rigorously specifying lines of responsibilities it is possible to reduce misbehaviour and errors, even in the case of information technology mismanagement. If indeed misbehaviour is controllable, it may behove management to assume greater (although not absolute) responsibility for its reduction. This is relevant not only to misbehaviours carried out by top managers themselves (like the case mentioned at the beginning of this chapter) but also to misbehaviours carried out by other employees. It is hoped, therefore, that by detecting various forms of misbehaviour, organizational leaders will be able to cope with them more effectively and, ultimately, raise the quality of working life in our society.

References Ackroyd, S. and Thompson, P. (1999) Organizational Misbehaviour. Thousand Oaks, CA: Sage. Bennett, R. J. and Robinson, S. L. (2000) ‘The development of a measure of workplace deviance’, Journal of Applied Psychology, 85, 349–60. Brumback, G. B. (2001) ‘Organizational misbehavior – book review’, Personnel Psychology, 54, 737–41. Bureau of Labor Statistics (2001) ‘How widespread is violence in the workplace?’, http://www.bls.gov/iif/peoplebox.htm #faqd (accessed on 21 October 2002). Fox, S. and Spector, P. E. (1999) ‘A model of work frustration – aggression’, Journal of Organizational Behavior, 20, 915–31. Giacalone, R. A. and Greenberg, J. (1997) Antisocial Behavior in Organizations. Thousand Oaks, CA: Sage. Govoni, S. J. (1992) ‘To catch a thief’, CFO, February, 24–32. Griffin, R. W., O’Leary-Kelly, A. M. and Collins, J. (1998) ‘Dysfunctional work behaviors in organizations’, in C. L. Cooper and D. M. Rousseau (eds), Trends in Organizational Behavior (pp. 65–82). New York: Wiley. Gruys, M. L. (1999) ‘The dimensionality of deviant employee performance in the workplace’, unpublished doctoral dissertation, University of Minnesota, Minneapolis, MN. Robinson, S. L. and Bennett, R. J. (1995) ‘A typology of deviant workplace behaviors: a multidimensional scaling study’, Academy of Management Journal, 38, 555–72. Vardi, Y. and Weiner, Y. (1996) ‘Misbehavior in organizations: a motivational framework’, Organization Science, 7, 151–65.

Part I The Many Faces of Work Misbehaviour

This page intentionally left blank

2 Workplace Aggression Kathryne E. Dupré and Julian Barling

Introduction It will be difficult for many employees and emergency workers to forget the day a former transit employee for Ottawa Carleton Transpo in Canada entered the building in mid-afternoon with a rifle and started shooting, firing at the 150 employees working in the building, ultimately killing four transit workers, seriously wounding two others, and finally killing himself. Likewise, the day on which a man dismissed from the University of Arkansas’ graduate programme shot and killed his faculty supervisor and himself in a building on campus after finding out he had been dropped from the graduate programme will be difficult to remove from memory. And the day after Christmas in 2000 will linger in the memories of employees of Edgewater Technology Inc. in Wakefield, Massachusetts. That day, an employee shot and killed seven co-workers, apparently upset because the Internal Revenue Service wanted to garnish his wages as a result of his failure to pay taxes. The widespread attention being devoted to workplace aggression is probably a consequence of such horrific acts of highly publicized aggression at workplaces. However, these acts account for only a fraction of all incidents of aggression. All individual behaviour aimed at harming others with whom one works, or their organizations, constitutes acts of workplace aggression (Neuman and Baron, 1997a), and a thorough understanding of all aspects of workplace aggression is necessary if we are to understand why such appalling acts of workplace aggression occur, and how they can be prevented. Over the past decade, a large body of research has accumulated on workplace aggression. Addressing the problem of workplace aggression, however, is complicated because this phenomenon has multiple 13

14 Workplace Aggression

sources, targets and causes. For example, workplace aggression occurs in hugely disparate occupational situations, and includes assaults on employees by patients being treated in health care facilities, robbery-related homicides in retail businesses, attacks on teachers by students, attacks resulting from disputes among co-workers, and partner violence that spills over into the workplace. Existing research is plagued by definitional inconsistencies and varied findings. Moreover, given the widespread nature and distinct forms of workplace aggression, it is not surprising that it is difficult to fully understand the extent of aggression in the workplace or the number of victims. Our aim in this chapter is to provide a comprehensive discussion of workplace aggression, focusing on those acts of aggression executed by members of organizations. Thus, we exclude acts of workplace aggression in which the perpetrator has no legitimate relationship to the organization, along with acts of aggression in which the perpetrator has a personal relationship with the victim, but has no relationship with the organization.

Definitions and forms of workplace aggression There is a tremendous range in the acts that constitute workplace aggression (Andersson and Pearson, 1999; Barling, 1996; Greenberg and Barling, 1999; Neuman and Baron, 1998). Efforts to harm others in an organizational context range from subtle and covert actions, through to active confrontations, the destruction of property, and direct physical assaults (Barling, 1996; Baron, 1993; Robinson and Bennett, 1995). Workplace aggression includes direct physical aggression (e.g. punch or shove), direct psychological harm (e.g. verbal insults or ignoring the victim), and indirect harm (e.g. destroying the victim’s property or spreading rumours). In recent years, research has focused on a variety of forms of workplace aggression including bullying (e.g. Einarsen and Skogstad, 1996), incivility (Andersson and Pearson, 1999), organizational retaliatory behavior (Skarlicki and Folger, 1997), emotional abuse (Keashly, 1998), and tyranny (Ashforth, 1994). Jenkins (1996) and LeBlanc and Kelloway (2002) define workplace violence as a form of aggression including physical assaults and threats of assault directed toward employees. Researchers have identified four categories of workplace aggression and most incidents fall into one of these categories (Braverman, 1999; California Occupational Safety and Health Administration, 1995; University of Iowa Injury Prevention Research Center, 2001). In Type-1 workplace aggression, the perpetrator of the violence has no legitimate

Kathryne E. Dupré and Julian Barling 15

relationship to the organization or its employees. Most workplace homicides (85 per cent) fall into this category of occupational aggression. In Type-2 workplace aggression, the perpetrator has a legitimate relationship with the organization and becomes violent while being served by the organization. Perpetrators could include customers, clients, patients, students, and inmates. In Type-3 workplace aggression, the perpetrator is an employee or past employee of the organization who attacks or threatens another employee or previous employee. These kinds of incidents account for approximately 7 per cent of all workplace homicides that are the result of violence, but probably account for a much larger proportion of non-fatal aggression that occurs in the workplace. Type-4 aggression occurs when the perpetrator has a personal relationship with the victim, but does not have a relationship with the organization. Most typical of Type-4 aggression is the situation in which a male arrives at the workplace and aggresses against his estranged spouse or partner.

Incidence of workplace aggression Although our focus in this discussion of workplace aggression will be those acts of aggression engaged in and experienced by members of the organization, statistics on the prevalence of workplace aggression often do not differentiate among various categories of workplace aggression. Moreover, precise statistics on workplace aggression in general are difficult to ascertain because of differences across conceptual and operational definitions, and because little has been done to document and centralize the statistics. No single agency is responsible for collecting all of the data, and organizations are not required to report non-fatal incidents (Andersson and Pearson, 1999). Additionally, not all acts of aggression in every workplace are reported. One study in a hospital context estimated that five incidents of violence occur against employees for each one that is reported (Lion, Snyder and Merrill, 1981), and under-reporting may be even more pronounced where the repercussions for reporting aggression are severe. The evidence and statistics on workplace aggression accumulated through professional agencies and associations shows the numbers are rising, or remaining at high levels. Homicide, the most extreme form of workplace aggression, is one of the leading causes of fatal occupational injuries for all workers in the United States each year (Bureau of Labor Statistics, 2000a, 2000b). It is estimated that in the United States nearly 1,000 workers are murdered, and 1.5 million are assaulted in the

16 Workplace Aggression

workplace each year (Bureau of Labor Statistics, 1995; National Institute for Occupational Safety and Health, 1996). Overall, 1,063 Americans were murdered on the job in 1993, which is 34 per cent more than the annual average for the 1980s. In addition, it appears that millions of others are victims of non-fatal occupational aggression including acts of harassment, intimidation, and physical attacks. American data reveal approximately two million incidents of workplace violence in the United States annually, with an average of 5,500 incidents daily (Bowman and Jude-Zigmond, 1997). In a survey conducted by the Northwestern National Life Insurance Company, 25 per cent of American workers surveyed said that someone in their workplace had been assaulted, threatened with violence, or harassed in the past year. This amounts to an estimated 2.2 million employees who have been directly affected by violence at work (Northwestern National Life, 1993). With about 135 million people working in the United States, two million incidents of workplace violence averages out to about one for every 68 workers and managers (Arbetter, 1994). Although occupational homicide occurs more frequently in the United States than most other countries (e.g. Canada, the UK), other countries are not immune from non-fatal workplace aggression. In a recent study conducted in Nova Scotia, Canada, 80 per cent of respondents reported experiencing some form of violence in their nursing careers: 63 per cent of respondents reported having experienced harsh or insulting language; 25 per cent had been verbally threatened with physical harm; 35 per cent had attempts of physical harm made against them; 24 per cent were sexually harassed in the workplace; and 21 per cent were victims of a physical attack (Nova Scotia Department of Labour, 1998). (As indicated in the percentages, some nurses reported that they had been victims of more than one type of violence.) Moreover, statistics indicate that in some instances, individuals outside of the United States may be more likely to experience certain forms of aggression at work. A survey conducted by the International Labour Organization indicated that close to 4 per cent of Canadian men and 5 per cent of Canadian women reported being assaulted in the last year while at work, compared with 1 per cent of men and 4.2 per cent of women in the United States (International Labour Organization, 1998).

Predictors of workplace aggression Despite the fact that there is a growing body of knowledge on workplace aggression, many of the existing findings are general rather than

Kathryne E. Dupré and Julian Barling 17

specific, making it difficult to comprehend the causes and prevent aggression in specific situations. Greenberg and Barling (1999) showed that it is important to consider the specific source and target of aggression because they found that the factors that predict work-related aggression vary, depending on the target of aggression (i.e. co-workers, subordinates, and supervisors). Specifically, person behaviours (i.e. history of aggression, alcohol consumed) predicted aggression against a co-worker, workplace factors (i.e. justice, surveillance) predicted aggression against a supervisor, and neither person nor workplace factors predicted aggression against a subordinate. Greenberg and Barling proposed that it is possible that employees chose not to aggress against subordinates because subordinates do not have the power to improve negative workplace conditions, and would not necessarily be presumed to be the cause of workplace dissatisfaction. These findings suggest that different predictor variables may play various roles in predicting work-related aggression depending on the target. Therefore, in any examination of workplace aggression, it is important to consider that for any particular type of aggression against specific targets, there may be variation in the antecedents and consequences of this aggression. In the past, research examining the predictors of aggression at work often centred on the development of a profile of an aggressive individual (e.g. Mantell and Albrecht, 1994; Slora, Joy and Terris, 1991), focusing on demographic and psychological correlates of these aggressive behaviors (e.g. Krueger et al., 1994; Perlow and Latham, 1993), to the exclusion of situational determinants of such behaviours (Greenberg and Barling, 1999). Over the past few years researchers have expanded their efforts in determining the causes of workplace aggression and the results indicate clearly that both individual and workplace factors are important in the prediction of workplace aggression. At the same time, research has also shown that there is an interaction between individual and workplace factors when predicting work-related aggression. Much research has demonstrated that understanding human behaviour demands consideration of person-by-situation interactions (e.g. Dekker and Barling, 1998; Fox and Levin, 1994; Mischel, 1968; Skarlicki, Folger and Tesluk, 1999). As other researchers have suggested or implied (e.g. Douglas and Martinko, 2001; Folger and Skarlicki, 1998; Martinko and Zellars, 1998), the incidence of aggression at the workplace is probably the outcome of a complex interaction between individual and situational factors.

18 Workplace Aggression

Individual factors Some people have a predisposition to aggression. In a recent study, Douglas and Martinko (2001) found that individual differences variables (including trait anger, attribution style, negative affectivity, attitudes toward revenge, self-control, and previous exposure to aggressive cultures) accounted for 62 per cent of the variance in workplace aggression. Various personality traits appear to have the potential to influence workplace aggression. For example, certain people are easily aroused to hostility (Neuman and Baron, 1997b), and more likely to react aggressively to certain situations. Hostility biases occur when certain individuals perceive situations as more aggressive than others, and have been identified in aggressive individuals (Dodge et al., 1990). Trait anger is a disposition to experience anger over time and situations (Spielberger, 1996). High trait anger individuals are more likely to respond aggressively to particular situations than are low trait anger individuals because they are more likely to perceive a wider range of situations as anger provoking than do low trait anger individuals. Aggression is a highly stable behaviour (Huesmann et al., 1984), and an individual’s past history of aggression is also important in subsequent acts of aggression (e.g. Riggs and O’Leary, 1989). Aggressiveness at eight years of age has been demonstrated to predict serious antisocial behaviour twenty years later (Huesmann, 1994). One study found that the history of aggression against peers and families as a teenager was a significant antecedent of aggression towards co-workers and subordinates at work (Greenberg and Barling, 1999). In many of the family aggression studies, a history of aggressive or violent behaviour is shown to be most predictive of a recurrence of this behaviour (e.g. Malone, Tyree and O’Leary, 1989; Riggs and O’Leary, 1989). Excessive use of alcohol is related to aggressive behaviour across different contexts. Both Stuart (1992) and Graham (1991) suggest that alcohol abuse is common in individuals who murder at the workplace. Greenberg and Barling (1999) found that alcohol consumption was associated with psychological aggression towards co-workers and subordinates at work. It is likely that alcohol use will increase the likelihood of misinterpreting situations (Cox and Leather, 1994), and impair perceptual processes that would normally inhibit risky behaviour such as aggression (Steele and Josephs, 1990). Evidence clearly indicates that individual factors play a role in workplace aggression. However, we believe that when it comes to understanding the antecedents of acts of workplace aggression carried

Kathryne E. Dupré and Julian Barling 19

out by members of a particular organization, organizational factors are more important. Therefore we focus considerable attention on the workplace factors that are predictive of workplace aggression in the following section.

Workplace factors A number of workplace factors have been associated with work-related aggression. In particular, any interaction that results in the perception of unfair treatment or frustration may elicit unpleasant thoughts and feelings and lead to aggression (e.g. Folger and Baron, 1996; Neuman and Baron, 1998). If employees perceive organizational decisions and managerial actions to be unfair or unjust, they are likely to experience feelings of anger, outrage or resentment (e.g. Skarlicki, Folger and Tesluk, 1999). Where individuals do not receive something to which they feel entitled, a perception of unfair treatment is likely to occur, and they may retaliate against those seen as responsible (Greenberg and Barling, 1999; Neuman and Baron, 1998; Skarlicki, Folger and Tesluk, 1999; Townsend, Phillips and Elkins, 2000), or may be motivated to re-establish a sense of justice (Cropanzano and Folger, 1989). Based on previous literature, three types of justice are considered to be relevant to workplace aggression: distributive justice, which refers to the perceived fairness of outcomes (Moorman, 1991), procedural justice, which refers to the perception that fair procedures were used to derive outcomes (Moorman, 1991), and interpersonal justice, which refers to a perceived demonstration of respect, dignity, sensitivity and courtesy by those who are responsible for executing procedures (Colquitt, 2001; Donovan, Drasgow and Munson, 1998). Some evidence suggests that perceptions of interpersonal injustice will be associated most strongly with workplace aggression, because aggression could be used as a means for employees to ‘fire back’ (Fox and Levin, 1994) at perceived mistreatment by others (Dupré and Barling, 2002). Another factor that has been linked to different types of detrimental behaviours is a lack of control. In the family violence literature, spouses who are not physically or psychologically aggressive towards their partners feel less controlled by their partners than spouses who are physically or psychologically aggressive towards their partners (Ehrensaft et al., 1999). Although much research has been devoted to understanding such control (e.g. DeMore, Fisher and Baron, 1988; Storms and Spector, 1987), far less research has explored how it relates to workplace misbehavior per se. While a certain level of control is appropriate in workplace settings (e.g. supervisory control), over-control may have

20 Workplace Aggression

deleterious effects on employees (Shirom, Melamed and Nir-Dotan, 2000). When employees believe that they can protect themselves from negative outcomes, individual and organizational well-being are enhanced (e.g. Bosma, Stansfield and Marmot, 1998; Spector, 1986). When employees lack a perception of personal control over certain work demands such as work load and work pace, they may experience emotional distress, lowered self-esteem, job dissatisfaction, and lowered job-related motivation (Richer and Vallerand, 1995; Sauter, Murphy and Hurrell, 1990). Research indicates that a feeling of a lack of control tends to make people feel threatened, and that restrictions of control result in attempts to regain control (Bennett, 1998). A study by Stanton and Barnes-Farrell (1996) was conducted to assess the effects of electronic performance monitoring on individuals. The amount of control over performance monitoring varied, and individuals with the ability to delay or prevent electronic performance monitoring indicated higher feelings of personal control and demonstrated superior task performance. When employees perceive they are over-controlled, they may feel the need to restore the balance, and one way of accomplishing this could be to strike back at the over-controlling agent. Greenberg and Barling (1999) found that surveillance methods used to monitor employees, leaving employees with a feeling of a lack of control, was positively associated with aggression against a supervisor. Recently, Dupré and Barling (2002) found that workplace aggression towards supervisors was strongly related to perceptions of over-control by these supervisors. Supervisors’ treatment of employees can go beyond being unfair and over-controlling, and become actively abusive (O’Leary-Kelly, Griffin and Glew, 1996), with important implications for the prediction of workplace aggression. Day and Hamblin (1964) found that punitive supervision produced significant increases in the aggression by subordinates towards supervisors. Abusive supervision will endure within the context of the work relationship until either the employee or supervisor terminates the relationship, the supervisor decides to modify his or her behaviour (Tepper, 2000), or the subordinate takes some action. It is likely that abusive supervision will also be associated with subordinates’ use of aggression towards their supervisors. Tepper (2000) found that subordinates who perceived that their supervisors were abusive were more likely to quit their jobs. For those who remained, there was lower job and life satisfaction, conflict between work and family, and psychological distress. Perceived organizational sanctions play an important role in the prediction of workplace aggression, and explain why employees choose

Kathryne E. Dupré and Julian Barling 21

not to behave aggressively. Dekker and Barling (1998) found that perceptions of organizational sanctions against sexual harassment played a significant role in the prediction of sexual harassment. Males who believed that the organization would take action against the perpetrators were significantly less likely to engage in gender and sexual harassment. Perceived organizational sanctions moderate the effects of negative workplace experiences on workplace aggression. For example, those members of an organization who perceive that they are treated unjustly but that there are organizational sanctions against aggression in their workplace, are less likely to engage in aggression than employees who feel that they are treated unjustly and do not perceive such organizational sanctions against aggression. Dupré and Barling (2002) found that although feeling over-controlled and perceiving injustice predict workplace aggression by employees towards their supervisors, these relationships are minimized when individuals believe that the organization will take action against workplace aggression. When individuals perceived organizational sanctions against aggression, even though control and injustice were still very high, they were much less likely to engage in aggression. There are examples of organizations that have implemented practices and policies against workplace aggression and violence, and found reductions in its occurrence. For example, all employees at both Polaroid and US Postal Services receive training in recognizing and coping with potentially aggressive situations, and the aggression in these firms has decreased since the implementation of this training (Jossi, 1999).

The escalation of workplace aggression As mentioned earlier, more research attention on the topic of workplace aggression has been directed toward the more serious and physical forms of workplace aggression. This particular focus on workplace aggression is probably as a result of the dramatic consequences associated with severe workplace aggression, and due to the fact that obtaining data on other forms of aggression poses significant difficulties (Beale, Cox and Leather, 1996). However, psychological and other forms of seemingly less serious aggression (e.g. ignoring employees, excluding others, verbal attacks, spreading rumours about others, failing to deny false rumours about others) are much more widespread at the workplace than are the more horrific acts of aggression (Braverman, 1999). In one study of 136 men, while two of the men reported using physical aggression at work, 82 per cent, 74 per cent, and 76 per cent admitted to some

22 Workplace Aggression

form of psychological aggression against co-workers, subordinates, and supervisors, respectively (Greenberg and Barling, 1999). Nonetheless, understanding less serious forms of aggression at work is critical because of the ‘spiral of violence’, inasmuch as workplace aggression may serve as the initial step in an upward spiral that culminates in ever-increasing levels of physical aggression (Andersson and Pearson, 1999; Barling, 1996; Murphy and O’Leary, 1989; Neuman and Baron, 1998), thereby predicting more serious forms of aggression. In the family violence domain, Murphy and O’Leary (1989) showed that among couples with no prior experience of physical aggression, psychological aggression predicted the first instance of physical aggression both six and twelve months later. Likewise, deadly assaults at work do not appear to result from one brief encounter or provocation. It is more likely that those involved know each other and have had a series of exchanges that are seen as insulting by at least one individual. Thus, an occurrence that might be a minor annoyance at an earlier stage may later provoke aggression. In the case of the Ottawa Carleton Transpo employee mentioned at the beginning of this chapter, for example, it is reported that the shooter felt, and anecdotal evidence suggests he was, constantly taunted because of his stutter; evidence at the inquest corroborated this. In a study of incarcerated males, Felson and Steadman (1983) revealed that the sequence of events leading up to assault generally begins with an exchange of rude comments and ultimately spirals into a physical attack. Given the prevalence of a wide variety of forms of aggression at work, and the fact that these types of aggression might lead to future more serious aggression, it is important to examine a wide range of acts of workplace aggression in research.

Outcomes of workplace aggression Although a better understanding of the predictors of workplace aggression is merited, so too is an improved understanding of the consequences of workplace aggression. Research indicates that the experience of workplace aggression undoubtedly brings along with it negative consequences (e.g. Budd, Arvey and Lawless, 1996), not only for individuals, but also for organizations and society.

Individual outcomes Barling (1996) discusses the issue of primary and secondary victims of workplace aggression. Primary victims are those who are the target of the aggression, while secondary victims are those who are not the direct

Kathryne E. Dupré and Julian Barling 23

target of aggression, but who are affected by being vicariously exposed to the aggression. For example, witnessing aggression may lead to fear of future violent incidents (Leather et al., 1998; Rogers and Kelloway, 1997). Thus, when considering the consequences of workplace aggression it is important to keep in mind that there may be effects for a wider variety of victims than solely for the direct targets of aggression. Moreover, the outcomes may vary depending on the source of the aggression for employees. LeBlanc and Kelloway (2002) showed that experiencing aggression from co-workers directly predicted the degree of emotional well-being, psychosomatic well-being, and affective commitment, while experiencing aggression from the public indirectly predicted fear of aggression (through the likelihood of future aggression). In each case, experiencing agression resulted in more negative outcome. Fear of crime is a serious and widespread source of anxiety, capable of having a negative impact upon health, and disturbing personal activity (e.g. Budd, Arvey and Lawless, 1996). Evidence from several studies supports the notion that fear is also a direct consequence of workplace aggression. Barling, Rogers and Kelloway (2001) found that both workplace aggression and sexual harassment result in fear of their recurrence in the workplace. Leather et al. (1997) stress the importance of fear as a consequence of workplace aggression that can lead to other negative outcomes. Budd, Arvey and Lawless (1996) found that the experience of violence is strongly associated with worrying about violence, and that fear of workplace violence is strongly associated with other variables. Schat and Kelloway (2001) found that fear of workplace aggression was predicted by violence and control, and that greater levels of this type of fear resulted in poorer emotional well-being. In the study mentioned above in which Barling, Rogers and Kelloway (2001) found that workplace aggression and sexual harassment affected fear of these events, it was also found that fear of these events in turn affected mood. The components of negative mood assessed in this study were anger, anxiety, and sadness. In this same study by Barling, Rogers and Kelloway, negative mood led to other outcome variables such as reduced affective commitment to the organization, and cognitive distractions. Additional previous research has also shown that stressful organizational events detract from an individual’s ability to concentrate fully at work (e.g. Leymann, 1990). Whittington and Wykes (1989) reported that nurses who had been assaulted by patients said that they had difficulties concentrating while working. As a result of a lack of concentration, there may be more unsafe behaviour and a greater likelihood of accidents resulting from exposure to aggression at work.

24 Workplace Aggression

Organizational outcomes Organizational performance and productivity may be affected by workplace aggression (e.g. Barling, 1996). There have been many suggestions that there is a negative relationship between workplace aggression and organizational productivity (e.g. Carcione, 2000; Cohen, 1996; Jossi, 1999; Mello, 1998), presumably because of the negative individual outcomes associated with workplace aggression. Some research suggests that employees who have experienced the effects of workplace aggression may blame the organization, and may retaliate against the organization in ways that are detrimental to overall functioning and productivity (e.g. theft, reduction in organizational citizenship behaviours; Skarlicki, Folger and Tesluk, 1999). Substantial losses in productivity are estimated to occur throughout an organization that experiences workplace aggression, with decreases of up to 80 per cent for up to two weeks immediately after the incident (e.g. McGovern etal., 2000). This has the potential to incur a huge loss upon the organization in the short term, and likely in the longer term as well. However this phenomenon may not be obvious. For example, although productivity may be decreased as a result of bullying, Hoel, Rayner and Cooper (1999) report that people subjected to bullying might be more eager to demonstrate their ability and commitment due to feelings of low self-esteem and organizational status as a result of the bullying. Workplace aggression is associated with higher absenteeism and turnover (e.g. Cohen, 1996). It has been suggested that missed work as a result of workplace aggression averages 3.5 days per incident in the United States, resulting in over $55 million in lost wages annually, not including days covered by sick and annual leave (Bachman, 1994). Additionally, it has been concluded that many organizations experiencing incidents of workplace aggression suffer dramatic increases in employee turnover, which is a costly phenomenon for organizations given the expense of hiring and training replacement employees (Gerhart, 1990; Sager, 1990). Since there is emerging evidence that those who witness aggression will experience many of the same consequences as the actual victims of workplace aggression, the costs of workplace aggression may be exacerbated by the fact that many more individuals are victims of workplace aggression than previously believed. Organizations may be found responsible for these effects as well, and be forced to spend extra money compensating these victims, paying for their health care costs, etc. Employers are supposed to provide a safe workplace and are responsible for acts engaged in by an employee while ‘in the scope of his or her employment’, as well as for negligent hiring, negligent retention, and

Kathryne E. Dupré and Julian Barling 25

negligent failure to warn employees (Fiesta, 1996). The legal system in the United States does not protect employers who should have seen signs of pending aggression and failed to act. Recently, in North Carolina, a court awarded a $7.9 million judgment against an employer when a worker killed two fellow employees (Perry, 2000). The Occupational Safety and Health Administration fined a Chicago psychiatric hospital $5,000 because employees there had suffered numerous injuries at the hands of violent patients (Bureau of National Affairs, 1996). Negative publicity is frequently cited as being a very costly outcome for firms after incidents of workplace aggression (e.g. Bowman and JudeZigmond, 1997; Cohen, 1996; Perry, 2000). Media accounts of the incident, whether accurate or not, and rumours that often follow, may influence the buying decisions of the firm’s customers, as well as decisions made by other valuable organizational stakeholders and shareholders of the firm that has experienced the aggression. Restoring the organization’s reputation following charges of incompetent or irresponsible management, or similar allegations following an incident of workplace aggression, is generally very expensive for any organization (e.g. Carcione, 2000).

Societal outcomes Despite high costs in terms of the outcomes of workplace aggression, our knowledge is still relatively limited. Undoubtedly, there are a number of mechanisms through which workplace aggression affects society. The aggression that individuals experience at work is likely to affect other aspects of their lives, as previous research has demonstrated that violence can affect home and personal life, often greatly straining relationships (e.g. Barling, MacEwen and Nolte, 1993). In a broader sense, learning of acts of workplace aggression through media accounts may create a society more fearful and pessimistic about work in general.

Future research Despite expanding research and awareness on the topic of workplace aggression, significant gaps remain in our knowledge and understanding of its occurrence. In order to reduce the prevalence of this workplace phenomenon more research is needed. Some of the areas that present the greatest opportunity for advancement in this domain are discussed. Does aggression at work escalate? In spite of many suggestions of an escalation of aggression at the workplace, there is negligible research to sustain such a proposition. Although a substantial amount of research

26 Workplace Aggression

alludes to the fact that aggression that is verbal and/or less intense in nature serves as the initial step in an upward spiral that leads to physical and/or more intense forms of negative workplace behaviours (Andersson and Pearson, 1999; Barling, 1996; Neuman and Baron, 1998), to date, no research has tested this important proposition. There are data, however, showing that aggression in romantic relationships escalates over time (Murphy and O’Leary, 1989). Research that specifically examines whether or not an escalation of aggression also occurs in workplace settings will greatly contribute to the understanding of the development of aggression at work, and will have important implications for interventions designed to reduce aggression at work. Does aggression at work generalize to other contexts, and vice versa? It would be important to examine the transference of aggression across contexts. There has been a great deal of discussion in the workplace aggression domain regarding whether aggression tends to be a function of individual factors or situational factors (e.g. Greenberg and Barling, 1999). Some have argued that aggression results from dispositional or personality traits, while others argue that situational factors play a predominant role. However, research in the area of aggression tends to examine this behaviour within a specific context (e.g. supervisor aggression, co-worker aggression, spousal aggression, parental aggression), thus prohibiting a cross-situational examination of the findings that could provide a context within which this question could be answered. Research that specifically examines whether or not individuals tend to behave aggressively across situations will contribute to the understanding of whether acts of aggression result from individual factors, situational characteristics, or an interaction between the two. Can organizational policies limit the frequency of workplace aggression? Despite the fact that there is evidence that policies and practices moderate the effects of sexual harassment (Dekker and Barling, 1998), there are no direct data on workplace aggression. Future research might also focus on the extent to which the implementation of organizational policies limits workplace aggression. What are the social and personal costs of workplace aggression? There is also a need to focus on the measurement of workplace aggression and the outcomes and costs associated with it. To date, only estimations of the prevalence of workplace aggression have been possible. Although it might be difficult to ever ascertain the true extent of workplace aggression, future research should focus on overcoming the difficulties associated with existing measurement tools. This would also allow for conclusions across different studies.

Kathryne E. Dupré and Julian Barling 27

What about workplace aggression among young workers? Although we now have some understanding of workplace aggression among adult employees, research should focus on young workers to determine if the antecedents and consequences of workplace aggression are similar across groups. Preliminary work suggests that there are similarities across different groups. For example, perceptions of injustice are likely to play a role in predicting teenagers’ aggression. Mulvey and Cauffman (2001) discuss the fact that students who trust, feel close to, and are fairly treated by others at their school, are less likely to engage in violent behaviour. Jenkins (1997) found that belief in the fairness and consistent enforcement of school rules, along with a commitment to school, are the most important factors in reducing crime that occurs in schools. A recent study found that perceptions of interpersonal injustice, over-control, and abusive supervision all predicted teenage employees’ aggression towards their supervisors at work (Dupré et al., 2002). This study shows that the same factors that predict workplace aggression by adult workers replicate to younger workers. However, something that may be unique to young employees’ experience of workplace aggression has to do with reasons for working. This same study by Dupré et al. (2002) also examined reasons for working as potential moderators of the relationships between interpersonal injustice, over-control, and abusive supervision and teenage employees’ aggression towards their supervisors at work, and found that financial reasons, and personal fulfilment reasons for working, moderate these relationships. Although reasons for working has not been examined as a potential moderator in similar relationships in studies of adult workplace aggression, such a moderator is potentially especially salient for teenagers because they are more influenced by their work environments than their adult counterparts (Lorence and Mortimer, 1985). The lessons learned during this developmental phase may not be readily amenable to later change, and the consequences of any progression from aggression to violence are of considerable social significance.

Conclusion ‘The prevention of violence affecting workers is one of the most important policy issues facing the workplace today’ (Braverman, 1999, p. 12). Although media coverage of the most severe workplace aggression raises the overall awareness of the hazard, it may inadvertently shift the focus of attention and prevention away from the primary danger. By taking a broad view of the meaning of workplace aggression, employers will be better prepared to prevent it and respond to it.

28 Workplace Aggression

References Andersson, L. M. and Pearson, C. M. (1999) ‘Tit for tat? The spiraling effect of incivility in the workplace’, Academy of Management Review, 24, 452–71. Arbetter, L. (1994) ‘Workplace victims’, Security Management, 38, 15. Ashforth, B. (1994) ‘Petty tyranny in organizations’, Human Relations, 47, 755–77. Bachman, R. (1994) Violence and Theft in the Workplace, Bureau of Justice Statistics, US Dept. of Justice, National Criminal Pub. No. 148199. Washington, DC: US Dept. of Justice. Barling, J. (1996) ‘Workplace violence: its scope and the issues’, in G. VandenBos and E. Q. Bulatao (eds), Violence on the Job: Identifying Risks and Developing Solutions. Washington, DC: American Psychological Association. Barling, J., MacEwen, K. E. and Nolte, M. L. (1993) ‘Homemaker role experiences affect toddler behaviors via maternal well-being and parenting behavior’, Journal of Abnormal Child Psychology, 21, 213–29. Barling, J., Rogers, A. G. and Kelloway, E. K. (2001) ‘Behind closed doors: in-home workers’ experience of sexual harassment and workplace violence’, Journal of Occupational Health Psychology, 6, 255–69. Baron, R. A. (1993) Violence in the Workplace: A Prevention and Management Guide for Businesses. Venture, CA: Pathfinder Publishing. Beale, D., Cox, T. and Leather, P. (1996) ‘Work-related violence – is national reporting good enough?’, Work & Stress, 10, 99–103. Bennett, R. J. (1998) ‘Perceived powerlessness as a cause of employee deviance’, in R. W. Griffin, A. O’Leary-Kelly and J. M. Collins (eds), Dysfunctional Behavior in Organizations: Violent and Deviant Behavior. Stamford, Connecticut: JAI Press Inc. Bosma, H., Stansfeld, S. A. and Marmot, M. G. (1998) ‘Job control, personal characteristics and heart disease’, Journal of Occupational Health Psychology, 3, 402–9. Bowman, J. S. and Jude-Zigmond, C. (1997) ‘State government response to workplace violence’, Public Personnel Management, 26, 289–300. Braverman, M. (1999) Preventing Workplace Violence. London: Sage Publications, Inc. Budd, J. W., Arvey, R. D. and Lawless, P. (1996) ‘Correlates and consequences of workplace violence’, Journal of Occupational Health Psychology, 1, 197–210. Bureau of Labor Statistics (1995) News: National Census of Fatal Occupational Injuries, 1994. Washington, DC: Bureau of Labor Statistics. Bureau of Labor Statistics (2000a) National Census of Fatal Occupational Injuries, 1999. Bureau of Labor Statistics (2000b) Work Injuries and Illnesses by Selected Characteristics, 1999. Washington, DC: US Department of Labor, Bureau of Labor Statistics. Bureau of National Affairs (1996) ‘Psychiatric hospital in Chicago cited by OSHA for workplace violence’, Occupational Safe Health Report, 23, 646. California Occupational Safety and Health Administration (Cal/OSHA) (1995) Cal/OSHA Guidelines for Workplace Security. San Francisco, CA: State of California, Department of Industrial Relations, Division of Occupational Safety and Health. Carcione, S. G. (2000) ‘Workplace violence: what you don’t know could hurt you’, Buildings, 94, 46–8. Cohen, A. (1996) ‘Getting personal’, Sales and Marketing Management, 148, 45. Colquitt, J. A. (2001) ‘On the dimensionality of organizational justice: a construct validation of a measure’, Journal of Applied Psychology, 86, 386–400.

Kathryne E. Dupré and Julian Barling 29 Cox, T. and Leather, P. (1994) ‘The prevention of violence at work: application of a cognitive behavioral theory’, in C. L. Cooper and I. T. Robertson (eds), International Review of Industrial and Organizational Psychology, 9 (pp. 213–45). Chichester, England: John Wiley and Sons Ltd. Cropanzano, R. and Folger, R. (1989) ‘Referent cognitions and task-decision autonomy: beyond equity theory’, Journal of Applied Psychology, 74, 293–9. Day, R. C. and Hamblin, R. L. (1964) ‘Some effects of close and punitive styles of supervision’, The American Journal of Sociology, 69, 499–510. Dekker, I. and Barling, J. (1998) ‘Personal and organizational predictors of workplace sexual harassment of women by men’, Journal of Occupational Health Psychology, 3, 7–18. DeMore, S. W., Fisher, J. D. and Baron, R. M. (1988) ‘The equity-control model as a predictor of vandalism among college students’, Journal of Applied Social Psychology, 18, 80–91. Dodge, K. A., Price, J. M., Bachorowski, J. A. and Newman, J. P. (1990) ‘Hostile attributional biases in severely aggressive adolescents’, Journal of Abnormal Psychology, 99, 385–92. Donovan, M. A., Drasgow, F. and Munson, L. J. (1998) ‘The perceptions of fair interpersonal treatment scale: development and validation of a measure of interpersonal treatment in the workplace’, Journal of Applied Psychology, 83, 683–92. Douglas, S. C. and Martinko, M. J. (2001) ‘Exploring the role of individual differences in the prediction of workplace aggression’, Journal of Applied Psychology, 86, 547–59. Dupré, K. E. and Barling, J. (2002) ‘The roles of control, justice and organizational sanctions in the prediction and prevention of workplace aggression and violence’, unpublished manuscript. Dupré, K. E., Inness, M., Connelly, C., Barling, J. and Hoption, C. (2002) ‘Adolescent antagonism: predicting workplace aggression in part-time teenage employees’, unpublished manuscript. Ehrensaft, M. K., Langhinrichsen-Rohling, J., Heyman, R. E., O’Leary, K. D. and Lawrence, E. (1999) ‘Feeling controlled in marriage: a phenomenon specific to physically aggressive couples?’, Journal of Family Psychology, 13, 20–32. Einarsen, S. and Skogstad, A. (1996) ‘Bullying at work: epidemiological findings in public and private organizations’, European Journal of Work and Organizational Psychology, 5, 185–201. Felson, R. B. and Steadman, H. J. (1983) ‘Situational factors in disputes leading to criminal violence’, Criminology, 21, 59–74. Fiesta, J. (1996) ‘Corporate liability: security and violence’, Nursing Management, 27, 14. Folger, R. and Baron, R. A. (1996) ‘Violence and hostility at work: a model of reactions to perceived injustice’, in G. VandenBos and E. Q. Bulatao (eds). Violence on the Job: Identifying Risks and Developing Solutions. Washington, DC: American Psychological Association. Folger, R. and Skarlicki, D. P. (1998) ‘A popcorn model of workplace violence’, in R. W. Griffin, A. O’Leary-Kelly and J. M. Collins (eds), Dysfunction Behavior in Organizations: Violence and Deviant Behavior (pp. 1–42). Stamford, CT: JAI Press. Fox, J. A. and Levin, J. (1994) ‘Firing back: the growing threat of workplace homicide’, The Annals of the American Academy of Political and Social Science, 534, 16–30.

30 Workplace Aggression Gerhart, B. (1990) ‘Voluntary turnover and alternative job opportunities’, Journal of Applied Psychology, 75, 467–76. Graham, J. P. (1991) ‘Disgruntled employees – ticking time bombs?’, Security Management, 36, 83–5. Greenberg, L. and Barling, J. (1999) ‘Predicting employee aggression against coworkers, subordinates and supervisors: the roles of person behaviors and perceived workplace factors’, Journal of Organizational Behavior, 20, 897–913. Hoel, H., Rayner, C. and Cooper, C. L. (1999) ‘Workplace bullying’, in C. L. Cooper and I. T. Robertson (eds), International Review of Industrial and Organizational Psychology, 14, 195–230. Huesmann, L. R. (ed.) (1994) Aggressive Behavior: Current Perspectives. New York: Plenum. Huesmann, L. R., Eron, L. D., Lefkowitz, M. M. and Walder, L. O. (1984) ‘Stability of aggression over time and generations’, Developmental Psychology, 20, 1120–34. International Labour Organization (1998) ‘Violence on the job – a global problem’ [on-line]. Available: http://www.ilo.org/public/english/235press/pr/ 1998/30.htm. Jenkins, E. L. (1996) ‘Workplace homicide: industries and occupations at high risk’, Occupational Medicine State of the Art Reviews, 11, 219–25. Jenkins, P. (1997) ‘School delinquency and the school social bond’, Journal of Research in Crime and Delinquency, 34, 337–67. Jossi, F. (1999) ‘Defusing workplace violence’, Business and Health, 17, 34–49. Keashley, L. (1998) ‘Emotional abuse in the workplace: conceptual and empirical issues’, Journal of Emotional Abuse, 1, 85–118. Krueger, R. F., Schmutte, P. S., Caspi, A., Moffit, T. E., Campbell, K. and Silva, P. A. (1994) ‘Personality traits are linked to crime among men and women: evidence from a birth cohort’, Journal of Abnormal Psychology, 103, 328–38. Leather, P., Beale, D., Lawrence, C. and Dickson, J. (1997) ‘Effects of exposure to occupational violence and the mediating impact of fear’, Work and Stress, 11, 329–40. Leather, P., Lawrence, C., Beale, D., Cox, T. and Dixon, R. (1998) ‘Exposure to occupational violence and the buffering effects of intra-organizational support’, Work and Stress, 12, 161–78. LeBlanc, M. M. and Kelloway, E. K. (2002) ‘Predictors and outcomes of workplace violence and aggression’, Journal of Applied Psychology, 87, 444–53. Leymann, H. (1990) ‘Mobbing and psychological terror at workplaces’, Violence and Victims, 5,119–26. Lion, J. R., Snyder, W. and Merrill, G. L. (1981) ‘Underreporting of assaults on staff in a state hospital’, Hospital and Community Psychiatry, 32, 497–8. Lorence, J. and Mortimer, J. (1985) ‘Job involvement through the life course: a panel study of three age groups’, American Sociological Review, 50, 618–38. Malone, J., Tyree, A. and O’Leary, K. D. (1989) ‘Generalization and containment: different effects of past aggression for wives and husbands’, Journal of Marriage and the Family, 51, 687–97. Mantel, M. and Albrecht, S. (1994) Ticking Bombs: Defusing Violence in the Workplace. New York: Irwin. Martinko, M. J. and Zellars, K. L. (1998) ‘Toward a theory of workplace violence: a cognitive appraisal perspective’, in R. W. Griffin, A. O’Leary-Kelly and J. M. Collins (eds), Dysfunction Behavior in Organizations: Violence and Deviant Behavior (pp. 1–42). Stamford, CT: JAI Press.

Kathryne E. Dupré and Julian Barling 31 McGovern, P., Kochevar, L., Lohman, W. and Zaidman, B. (2000) ‘The cost of workrelated physical assaults in Minnesota’, Health Services Research, 35, 663–86. Mello, J. P. (1998) ‘Hell in your hallways’, CFO, 14, 53–6. Mischel, W. (1968) Personality and Assessment. New York: Wiley. Moorman, R. H. (1991) ‘Relationship between organizational justice and organizational citizenship behaviors: do fairness perceptions influence employee citizenship?’, Journal of Applied Psychology, 76, 845–55. Mulvey, E. P. and Cauffman, E. (2001) ‘The inherent limits of predicting school violence’, American Psychologist, 56, 797–802. Murphy, C. M. and O’Leary, K. D. (1989) ‘Psychological aggression predicts physical aggression in early marriage’, Journal of Consulting and Clinical Psychology, 57, 579–82. National Institute for Occupational Safety and Health (1996) ‘Violence in the workplace’ [on-line]. Available: http://www.cdc.gov/niosh/violphs.html. Neuman, J. H. and Baron, R. A. (1997a) ‘Aggression in the workplace’, in R. Giacalone and J. Greenberg (eds), Antisocial Behavior in Organizations (pp. 37–67). Thousand Oaks, CA: Sage. Neuman, J. H. and Baron, R. A. (1997b) ‘Type A behavior pattern, self-monitoring, and job satisfaction as predictors of aggression in the workplace’, in G. Chao (Chair), Counterproductive Job Performance and Organizational Dysfunction. Symposium conducted at the meeting of the Society for Industrial and Organizational Psychology, St. Louis, Missouri. Neuman, J. H. and Baron, R. A. (1998) ‘Workplace violence and workplace aggression: evidence concerning specific forms, potential causes, and preferred targets’, Journal of Management, 24, 391–419. Northwestern National Life, Employee Benefits Division (1993) Fear and Violence in the Workplace: A Survey Documenting the Experience of American Workers. Minneapolis, MN: Northwestern National Life. Nova Scotia Department of Labour (1998) Hazard Assessment for the Violence in the Workplace Regulations. NS: Nova Scotia Department of Labour. O’Leary-Kelly, A. M., Griffin, R. W. and Glew, D. J. (1996) ‘Organizationmotivated aggression: a research framework’, Academy of Management Review, 21, 225–54. Perlow, R. and Latham, L. L. (1993) ‘Relationships of client abuse with locus of control and gender: a longitudinal study’, Journal of Applied Psychology, 78, 831–4. Perry, P. M. (2000) ‘Zero tolerance’, Industrial Distribution, 89, 55–60. Richer, S. F. and Vallerand, R. J. (1995) ‘Supervisors’ interactional styles and subordinates’ intrinsic and extrinsic motivation’, Journal of Social Psychology, 135, 707–22. Riggs, D. S. and O’Leary, D. K. (1989) ‘The development of a model of courtship aggression’, in M. A. Pirog-Good and J. E. Stets (eds), Violence in Dating Relationships: Emerging Social Issues (pp. 53–71). New York: Praeger. Robinson, S. L. and Bennett, R. J. (1995) ‘A typology of deviant workplace behaviors: a multidimensional scaling study’, Academy of Management Journal, 38, 555–72. Rogers, K. A. and Kelloway, E. K. (1997) ‘Violence at work: personal and organizational outcomes’, Journal of Occupational Health Psychology, 3, 63–71. Sager, J. K. (1990) ‘How to retain salespeople’, Industrial Marketing Management, 19, 155–66.

32 Workplace Aggression Sauter, S. L., Murphy, L. R. and Hurrell, J. J. (1990) ‘Prevention of work-related psychological disorders: a national strategy proposed by the National Institute for Occupational Safety (NIOSH)’, American Psychologist, 45, 1146–58. Schat, A. C. and Kelloway, E. K. (2000) ‘The effects of perceived control on the outcomes of workplace aggression and violence’, Journal of Occupational Health Psychology, 4, 386–402. Shirom, A., Melamed, S. and Nir-Dotan, M. (2000) ‘The relationships among objective and subjective environmental stress levels and serum uric acid: the moderating effect of perceived control’, Journal of Occupational Health Psychology, 5, 374–85. Skarlicki, D. P. and Folger, R. (1997) ‘Retaliation for perceived unfair treatment: examining the roles of procedural and interactional justice’, Journal of Applied Psychology, 82, 434–43. Skarlicki, D. P., Folger, R. and Tesluk, P. (1999) ‘Personality as a moderator in the relationship between fairness and retaliation’, Academy of Management Journal, 42, 100–8. Slora, K. B., Joy, D. S. and Terris, W. (1991) ‘Personnel selection to control employee violence’, Journal of Psychology and Business, 5, 417–26. Spector, P. E. (1986). ‘Perceived control by employee’s: a meta-analysis of studies concerning autonomy and participation at work’, Human Relations, 39, 1005–16. Spielberger, C. D. (1988) State-Trait Anger Expression Inventory: research edition professional manual. Psychological Assessment Resources Inc., Florida. Spielberger, C. D. (1996) Manual for the State-Trait Anger Expression Inventory (STAXI). Psychological Assessment Resources Inc., Odessa, Florida. Stanton, J. M. and Barnes-Farrell, J. L. (1996) ‘Effects of electronic performance monitoring on personal control, task satisfaction, and task performance’, Journal of Applied Psychology, 81, 738–45. Steele, C. M. and Josephs, R. A. (1990) ‘Alcohol myopia: its prized and dangerous effects’, American Psychologist, 45, 921–33. Storms, P. L. and Spector, P. E. (1987) ‘Relationships of organizations frustration with reported behavioral reactions: the moderating effect of perceived control’, Journal of Occupational Psychology, 60, 227–34. Stuart, P. (1992) ‘Murder on the job (killing of co-workers)’, Personnel Journal, 71, 72–84. Tepper, B. J. (2000) ‘Consequences of abusive supervision’, Academy of Management Journal, 43, 178–90. Townsend, J., Phillips, J. S. and Elkins, T. J. (2000) ‘Employee retaliation: the neglected consequence of poor leader–member exchange relations’, Journal of Occupational Health Psychology, 5, 457–63. University of Iowa Injury Prevention Research Center (2001) Workplace Violence: A Report to the Nation. Retrieved 31 May 2002, from the University of Iowa Injury Prevention Web site: http://www.public-health.uiowa.edu/IPRC/ NATION.PDF Whittington, R. and Wykes, T. (1989) ‘Invisible injury’, Nursing Times, 84, 30–2.

3 Comparing Sexual Harassment to Other Forms of Workplace Aggression Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar

Introduction ‘While animals mostly fight physically, and young children include verbal means in their aggressive repertoire, adult humans harm their enemies in much more subtle and sophisticated ways’ (Bjorkqvist, Osterman and Lagerspetz, 1994, p. 32). Since adult humans spend a large portion of their waking hours at work, it should not be surprising that their ‘subtle and sophisticated’ methods of aggression come to work with them. Workplace aggression or ‘efforts by individuals to harm others with whom they work or the organizations in which they are employed’ (Baron and Neuman, 1996, p. 161) is prevalent within many organizations. Over 50 per cent of the respondents in one study stated that they had experienced workplace abuse such as verbal and physical aggression (Richman et al., 1999). Indeed, workplace aggression has become so widespread that two psychologists initiated a ‘bully-busting’ campaign on the World Wide Web. Their website, www.workdoctor.com, which averages over 40,000 visitors per month, allows workers who have endured abuse from a co-worker or boss to tell their story. For example, one employee complained that his boss works himself into fits of profanity and rage so violent that workers are reduced to tears (Guynn, 1998). In addition to sparking interest in the general public, incidents of workplace aggression have also begun to receive attention from academic researchers (e.g. Allen and Lucero, 1998; Baron and Neuman, 1996; Baron, Neuman and Geddes, 1999; Bjorkqvist, Osterman and Hjelt-Back, 1994; Folger and Baron, 1996; O’Leary-Kelly, Griffin and Glew, 1996; Robinson and O’Leary-Kelly, 1998). However, one form of workplace 33

34 Sexual Harassment

aggression, sexual harassment, has been studied extensively over the past 20 years. The literature suggests that sexual harassment is a widespread phenomenon in the workplace with highly negative consequences for both victims and organizations (e.g. Luthar and Pastille, 2000; O’Leary-Kelly and Bowes-Sperry, 2001; Sbraga and O’Donohue, 2000; Welsh, 1999; 2000). One study found that 44 per cent of working women and 19 per cent of working men experienced some form of unwanted sexual attention while at work (USMSPB, 1995). A reading of EEOC Notice Number 915.002 issued in 1999, and the data compiled by the Office of Research, Information, and Planning from EEOC’s Charge Data System – national database, shows that the number of sexual harassment complaints with such agencies had jumped from 6,883 in 1991 to 15,836 by 2000. The costs incurred by organizations to settle such complaints can be substantial; for example, in 1998 Mitsubishi paid 34 million dollars and Astra paid 10 million dollars to settle sexual harassment lawsuits (see the EEOC website for other settlements of sexual harassment cases: www.eeoc.gov). Although sexual harassment has been classified as a type of workplace aggression (Baron et al., 1999; Lucero et al., 2001; Robinson and Bennett, 1995), few studies have investigated the phenomenon using theories of deviant behaviour or aggression (see O’Leary-Kelly, Paetzold and Griffin, 2000, for an exception). In this chapter, we examine the ways in which sexual harassment differs from and resembles other forms of workplace aggression. Our framework is comprehensive, focusing on aggressors/ harassers, targets of aggression/harassment, and contextual factors associated with aggression/harassment. Specifically we examine questions such as: (1) To what extent do sexual harassers resemble individuals who initiate other forms of workplace aggression? (2) To what extent do targets of sexual harassment resemble targets of other types of workplace aggression? and (3) To what extent do the contextual factors associated with sexual harassment resemble those associated with other forms of workplace aggression?

Sexual harassment as a form of workplace aggression As mentioned previously, the purpose of this chapter is to better delineate the links between sexual harassment and workplace aggression. Perhaps one of the most important distinctions between sexual harassment and other forms of workplace aggression is that sexual harassment is illegal, whereas many other forms of workplace aggression such as failing to deny false rumours, insulting one’s co-workers, and work slowdowns

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 35

(Baron et al., 1999) are not. By 1980, it was well established that quid pro quo sexual harassment (in which targets are threatened with punishment or promised rewards based on their response to initiators’ sexual demands) was a form of sex discrimination, which violated Title VII. In 1980, the Equal Employment Opportunity Commission (EEOC) issued guidelines that addressed not only the quid pro quo aspect of sexual harassment but also hostile environment harassment in which conduct of a sexual nature has the purpose or effect of substantially interfering with an individual’s work performance or creating an intimidating, hostile, or offensive working environment (US Equal Employment Opportunity Commission, 1981). While there is a distinction from the legal perspective, some empirical support exists for classifying sexual harassment as a form of workplace aggression (Baron et al., 1999; Robinson and Bennett, 1995). This is not surprising given that sexual harassment is ‘a stable behavioral construct distinct from but related to legal formulations’ (Gelfand, Fitzgerald and Drasgow, 1995). One study found that sexual harassment and generalized workplace abuse (which includes verbal aggression, disrespectful behaviour, isolation/exclusion, threats/bribes, and physical aggression) were significantly intercorrelated, and that generalized workplace abuse included at least the more subtle forms of sexual harassment (Richman et al., 1999). Furthermore, Baron et al. (1999), who included two items on sexual harassment (physical and verbal) in a forty-item questionnaire they developed to measure the frequency of various forms of workplace aggression, found that the item on verbal sexual harassment was empirically classified as covert aggression (specifically as an expression of hostility) whereas the item on physical sexual harassment was classified as overt aggression. Therefore, sexual harassment resembles other forms of workplace aggression in that both can be expressed as covert or overt behaviour. While it is useful to know that sexually harassing behaviours are exhibited in both overt and covert forms, Baron et al. (1999) did not provide any theoretical analysis focusing specifically on sexual harassment (this is not a criticism of their work – their focus was intentionally on the broad realm of workplace aggression). Therefore, a closer examination of covert and overt forms of sexual harassment may provide valuable insight into sexual harassment as a form of workplace aggression. Sexual coercion (or quid pro quo sexual harassment), which is defined as behaviours that explicitly or implicitly link sexual cooperation to job-related outcomes (Gelfand et al., 1995 ), is by definition overt – both the identity and intent of the harasser are clear when, for

36 Sexual Harassment

example, a supervisor demands that an employee engage in sexual activity to keep his or her job. On the other hand, many forms of hostile environment sexual harassment (unwanted sexual attention and gender harassment) are, at least in part, covert in nature. In the case of unwanted sexual attention or behaviours that are initiated for the purpose of gaining sexual cooperation but are not welcomed by the target nor tied to job-related outcomes (Gelfand et al., 1995), the identity of the harasser is known but his or her intention may be unclear. For example, initiators of such behaviour may (honestly or deceitfully) claim that they did not realize their attention was unwelcome. Thus, their intention may be perceived as romantic rather than harassing. In the case of gender harassment or behaviours that express hostile or insulting attitudes toward targets but are not initiated for the purpose of gaining sexual cooperation (Gelfand et al., 1995) either the identity or intention of the harasser may be disguised. For example, an employee at the Mitsubishi plant in Normal, Illinois (recall that Mitsubishi paid $34 million to settle a class action suit) placed a picture of a woman’s crotch on the safety visor of a female co-worker (Dateline, 1996). When questioned about his behaviour, the employee said he didn’t intend any harm – he thought she would think it was a joke. Regardless of his true intention, he can (without a complete lack of credibility) state that his intentions were not aggressive, thus leading to his behaviour being categorized as covert aggression. Furthermore, the employee who engaged in this act could have easily disguised his identity (once again, making the behaviour covert) by doing it when the target was out of the room. Although gender harassment may also take an overt form if the harasser makes no attempt to hide his or her identity and intention, we have classified it as covert because the possibility of disguising identity and intention exists. Based on the previous discussion, sexual harassment behaviours seem to form a hierarchy from covert (gender harassment), to partly covert (unwanted sexual attention), to overt (sexual coercion). Research on sexual harassment supports the findings of Baron and his colleagues (Baron and Neuman, 1996; Baron et al., 1999) that acts of covert aggression are more likely to occur in the workplace than are acts of overt aggression. Specifically, Fitzgerald et al. (1997) found that gender harassment was experienced more frequently than unwanted sexual attention, which was in turn experienced more frequently than sexual coercion. Therefore, sexual harassment resembles other forms of workplace aggression in that both are more likely to be expressed covertly rather than overtly. As noted by Bjorkqvist, Osterman and Lagerspetz (1994, p. 32),

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 37

‘It seems that adults try to conceal their interpersonal aggression as much as possible in behavior which is ostensibly rational and seems to contain no aggression at all.’ Whereas the categorization of sexual harassment as a form of workplace aggression has been relatively incidental in empirical research (see Lucero et al., 2001, for an exception), O’Leary-Kelly et al. (2000) developed a theoretical model of sexual harassment in which aggression is the central focus. In fact, their definition of sexual harassment as ‘sexual, work-related action taken with the expectation of imposing harm on another person or forcing his/her compliance in order to achieve some valued personal goal’ (p. 373) explicitly establishes sexual harassment as aggression. Furthermore, their model describes the circumstances under which an individual is likely to pursue his or her goals through ‘sexually harassing (versus some other form of) behavior’ (p. 375). They give the example of an employee who wants to date a co-worker – he may initially pursue this goal by using persuasion (a nonaggressive act), but if his attempts at persuasion are unsuccessful, he may progress to verbal coercion (an act of aggression). Essentially, their model attempts to predict the conditions under which an employee will use a particular aggressive behaviour (i.e. sexual harassment) rather than other behaviours to attain their goals.

How similar are sexual harassment and other forms of workplace aggression? What leads an employee to engage in an act of workplace aggression, be it sabotaging a co-worker’s success or sexual harassment? Based on their review of the workplace aggression literature, Allen and Lucero (1998) argue that, in most cases, aggression at work does not occur randomly or spontaneously. Rather, they note that the existence of triggering events is a dominant theme in this body of research. Common triggers of workplace aggression include aversive treatment, modelling, incentive inducements, and the physical environment (O’Leary-Kelly et al., 1996). Such triggers may induce cognitions, emotions, and motives that make a person ‘ready to aggress’ (Folger and Baron, 1996). O’Leary-Kelly et al. (2000) argue that, in addition to being motivated by emotions to initiate aggressive behaviour at work, employees can also exhibit instrumental (or goal-directed) motives such as dominance (gaining power over others or preventing perceived loss of power), impression management (gaining valued identities or preventing perceived loss of status), or retributive justice (re-establishing social order after perceived

38 Sexual Harassment

injustice). For example, the Mitsubishi employee mentioned previously who placed the lewd picture on his co-worker’s visor may have been motivated by impression management (e.g., to establish an identity as ‘one of the guys’) or by retributive justice (e.g. to show her that she was not welcome in a ‘man’s job’). Aggression motives can also be triggered by frustration (i.e. interference with ongoing, goal-directed activity) resulting from aspects of the workplace environment, such as high temperatures, poor lighting, high humidity, high noise levels, and crowding (Cohn and Rotton, 1997). Researchers suggest that whatever the trigger or motive, the translation of aggressive motives into behaviour depends on characteristics of the perpetrator, victim, and situation (e.g. Aquino et al., 1999; Baron et al., 1999; Neuman and Baron, 1998), to which we now turn.

Sexual harassers or simply jerks? In this section, we compare the characteristics of individuals who engage in sexual harassment with those of individuals who engage in other forms of workplace aggression. If there is one thing we know about sexual harassment, it is that men are overwhelmingly more likely than women to be sexual harassers (e.g. USMSPB, 1995). Is this true for other forms of aggression? While research indicates that men are more aggressive than women, gender explains only 5 per cent of the variation in aggression (Hyde, 1984). Furthermore, gender differences are found primarily when aggression is physical rather than psychological (Eagly and Steffen, 1986), and unprovoked rather than provoked (Bettencourt and Miller, 1996). Studies of aggression in the workplace find no effect for gender of the perpetrator (e.g. Bjorkqvist, Osterman and Hjelt-Back 1994; Douglas and Martinko, 2001; Robinson and O’Leary-Kelly, 1998). Given that aggression in the workplace tends to be psychological rather than physical (e.g. Baron and Neuman, 1996), and provoked rather than unprovoked (e.g. Allen and Lucero, 1998; Skarlicki and Folger, 1997), the finding of no gender differences is consistent with the social psychology literature on aggression (Bettencourt and Miller, 1996; Eagly and Steffen, 1986). Furthermore, it appears that men and women both use ‘rational-appearing’ aggression more frequently than ‘social manipulation’ aggression while at work (Bjorkqvist, Osterman and Lagerspetz, 1994). Therefore, it appears that sexual harassment differs from other forms of workplace aggression with regard to gender of the typical perpetrator. Despite gender differences between perpetrators of sexual harassment and other forms of workplace aggression, Lucero et al. (2001, p. 7) note

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 39

‘some of the most compelling research suggesting a link between sexual harassment and aggression is provided by studies of perpetrators’. Although the results of their analysis do not provide a clear typology of ‘perpetrator patterns’, they do indicate that a large number of sexual harassers initiated multiple incidents of harassment and targeted numerous individuals. If sexual harassers do ‘cast their nets far and wide’ (as argued by Lucero et al., 2001), we wonder whether their repertoire extends beyond sexual harassment to include other forms of workplace aggression. In other words, do individuals who choose to sexually harass engage in other forms of workplace aggression such that they could be labelled ‘simply jerks’ or ‘equal opportunity offenders’? The answer to this question has important implications for organizations. For example, if the ‘simply jerks’ perspective is accurate, then organizations might decide to present workplace aggression prevention and awareness training rather than (or in addition to) sexual harassment prevention and awareness training as is the norm at present. Since research on sexual harassment awareness training indicates that men with a high likelihood to sexually harass (LSH) either do not change their attitudes toward sexual harassment after attending such training (Perry, Kulik and Schmidtke, 1998), or develop even worse attitudes on the topic (Robb and Doverspike, 2001), implementing general workplace aggression prevention training may be an effective strategy. It may be that training on preventing general aggression is less threatening to high LSH men because it does not ‘single them out’. Furthermore, since there is evidence of a ‘backlash against laws and organizational policies targeted at men’ (Wayne, Riordan and Thomas, 2001, p. 185), a focus on workplace aggression may be more palatable even to men who have never thought about engaging in sexual harassment. Policies and training focusing exclusively on sexual harassment ‘establish a negative evaluation of actions that are largely male to female, making it likely that the identities of male employees may be invoked and implicated’, thus resulting in psychological reactance (O’Leary-Kelly et al. 2001). There is some support for the ‘simply jerks’ perspective. O’Leary-Kelly et al. (2000, p. 375) argue that some sexual harassers might ‘enact not just sexual harassment but other forms of aggression’. This proposition is based on interpersonal aggression research (e.g. Berkowitz, 1993), which finds that some individuals engage in aggression to vent their own negative emotions. In this sense, sexual harassers are seen as having ‘skill deficits’ with regard to their coping abilities that are likely to ‘affect all social interactions’ (O’Leary-Kelly et al., 2000). While this

40 Sexual Harassment

proposition has not been empirically tested, the research on bullying provides support for the ‘simply jerks’ perspective in a general aggression context. Specifically, Olweus (1979) found that aggressive reaction patterns (for males) were stable – i.e. an individual who responded aggressively in one situation tended to behave similarly in other situations. Much of what we know about harassers comes from research conducted using Pryor’s (1987) Likelihood to Sexually Harass (LSH) scale, which was designed to measure men’s proclivities to engage in quid pro quo sexual harassment. Essentially, this scale asks men to respond to hypothetical scenarios in which they are in a position of power over an attractive woman (i.e., they control an important punishment or reward for her). After imaging themselves as the man in the scenario, subjects are asked to indicate how likely they would be to sexually exploit the woman if they knew there would be no negative consequences associated with their behaviour. Research indicates that the LSH scale is an accurate predictor of sexually harassing behaviour. For example, Dall’Ara and Maass (1999) found that high LSH men were more likely to send pornographic material via email to women than were low LSH men. Pryor (1987) found that men with high scores on LSH touched a woman more frequently and in a more sexual manner than men with low scores on this scale BUT only in certain situations. Specifically, this relationship was found in a situation where touching was somewhat permissible (teaching a woman how to play golf). However, when teaching a woman to play poker (a situation in which touching is not required), the behaviour of high LSH men did not differ from that of low LSH men. This finding is particularly interesting when examined from an interpersonal aggression perspective. The golf situation allows a high LSH man (or a low one for that matter) to disguise his intention with regard to the touching (i.e. he could justify his behaviour by saying that he needed to touch her to physically demonstrate the proper swing). Thus, touching in the golf condition can be considered a form of covert sexual harassment. On the other hand, it would be difficult to use the same type of justification when touching a woman in the poker situation. Since the high LSH men could not disguise their intentions in the poker situation, touching the woman would be considered a form of overt sexual harassment. Therefore, we can interpret Pryor’s (1987) findings as follows – the high LSH men were willing to engage in covert sexual harassment but not overt sexual harassment. As mentioned previously, this finding is consistent with the literature on workplace aggression; covert forms

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 41

of aggression occur more frequently in the workplace than overt forms (e.g. Baron and Neuman, 1996; Bvjorqvist, Osterman and Lagerspetz, 1994). In addition to establishing a link between high LSH and covert sexual harassment, research also suggests a connection between high LSH and other forms of covert workplace aggression (at least toward women). For example, high LSH men are more likely than low LSH men to ask sexist questions of a female during an interview (Rudman and Borgida, 1995), to rate a female’s performance as low (Driscoll, Kelly and Henderson, 1998), to spend less time with a female in a subordinate position (Murphy, Driscoll and Kelly, 1999), and to provide less feedback regarding the performance of a female whom they have been asked to evaluate (Murphy et al., 1999). Behaviours such as those described above are examples of what Bjorkqvist, Osterman and Lagerspetz, (1994) refer to as ‘rational-appearing aggression’, i.e. a form of covert workplace aggression that includes reduced opportunities to express oneself, being interrupted, having one’s work judged in an unjust manner, and being criticized. On the basis of such findings, one could argue that high LSH men have some general hostility towards women as an outgroup (see Pryor and Whalen, 1997, for more on this topic), which is expressed as both sexual and non-sexual aggression. Is this general hostility or aggression also directed at other men? Driscoll et al. (1998) found that high LSH men hold more negative attitudes toward both women and men than low LSH men. However, we have not found any empirical research that investigates the link between high LSH and actual behaviour (rather than merely attitudes) toward men. For example, the studies listed above did not vary the gender of the person to whom the high LSH men were responding, therefore it is possible that high LSH men would have exhibited aggression toward subordinate men as well as subordinate women. Research establishing such a link would provide additional support for the ‘simply jerks’ perspective. Alternatively, the ‘simply jerks’ effect may be moderated by the gender of the person being targeted such that high LSH men are ‘simply jerks’ but only to women (i.e., they initiate various forms of workplace aggression including sexual harassment toward women but do not aggress against other men). In other words, the gender of the potential target may play an important role. Therefore, it is unclear whether high LSH individuals initiate both sexual harassment and other forms of workplace aggression. Are there other personality characteristics that are common to both sexual harassers and those who initiate other forms of workplace

42 Sexual Harassment

aggression? Research indicates that a low level of self-monitoring behaviour (i.e. not paying attention to what is expected in a social situation) is common to both. For example, Dall’Ara and Maass (1999) found that men who are low self-monitors are more likely to engage in sexual harassment than men who are high self-monitors. Driscoll, Kelly and Henderson (1998) found that high LSH men (as compared to low LSH men) exhibit less concern with social desirability (which is conceptually akin to a low level of self-monitoring). Similarly, Neuman and Baron (1997) found that low self-monitors were more likely than high self-monitors to engage in a form of workplace aggression they labelled ‘obstructionism’, which is a type of passive aggression comprised of behaviours such as interfering with the target’s work (as cited in Neuman and Baron, 1998). Therefore, it appears that both sexual harassers and individuals who initiate other forms of workplace aggression exhibit low levels of self-monitoring. The next section of this chapter examines the differences and similarities between targets of sexual harassment and targets of other forms of workplace aggression.

Aggressors in search of targets? We mentioned previously that one of the most noteworthy distinctions between sexual harassment and other forms of workplace aggression is that sexual harassment is illegal; specifically it is a form of sex discrimination (at least in the United States). This is because the majority of sexual harassment is targeted at women and negatively impacts their working conditions (e.g. USMSPB, 1995). However, research on other forms of workplace aggression finds no effect for gender of the target (e.g. Aquino et al., 1999; Bjorkqvist, Osterman and Lagerspetz, 1994). In other words, women are no more likely than men to be targets of general workplace aggression. Therefore, it appears that sexual harassment differs from other forms of workplace aggression with regard to gender of the typical target. Although this gender distinction is important, there may be other characteristics that are common to targets of sexual harassment and other forms of workplace aggression. It seems logical to assume that aggressors, similar to the ‘schoolyard bully’ would pick on weaker individuals, thus maximizing the effect–danger ratio (Bjorkqvist, Osterman and Lagerspetz, 1994), i.e. maximizing the harm done to the target(s) while minimizing negative repercussions for themselves. The ‘danger’ or cost of aggressing toward a less powerful person is much lower – such an individual is much less likely to retaliate or report the aggressive

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 43

behaviour. In research investigating victimization by peers in childhood, Olweus (1993) found that although submissive individuals (such as those described above) are most frequently victimized, a second type of victim, labelled the provocative victim, also exists. Aquino et al. (1999, p. 261) nicely summarized the difference between the two types of victims: ‘Whereas some people become targets of aggression because they behave anxiously or submissively, others are chosen because they frequently violate rules of social interaction or threaten others’ social identities.’ Before proceeding with the analysis of victim characteristics, we want to make clear that we imply no blame to victims and recognize that ‘without a tormentor, there will be no harassment’ (Bjorkqvist, Osterman and Hjelt-Back, 1994).

Provocative victims People often behave aggressively toward the person who causes them to become emotionally aroused and whom they hold responsible for negative outcomes (Berkowitz, 1993). O’Leary-Kelly et al. (1996), who refer to such targets as specific targets, propose that perpetrators aggress against specific targets as a form of retaliation for aversive treatment previously delivered by the target. For example, managers have reported aggressive responses (primarily of a verbal nature) from subordinates after providing them with negative performance feedback (Allen and Lucero, 1998; Geddes and Baron, 1997). In general, it appears that ‘specific targets’ are also provocative victims in the sense that they do something to upset the aggressor, essentially instigating (in the mind of the perpetrator) the aggressive actions that come their way (Olweus, 1993). Empirical research finds that in addition to being found on the playground, provocative victims also exist in the workplace. For example, Aquino et al. (1999, p. 262) hypothesized (and found) that employees who are high in negative affectivity would be the targets of workplace aggression more frequently than those scoring low on this personality trait because they ‘provoke others to respond to them in adverse ways’. Similarly, in a study of arbitration decisions involving subordinate aggression against managers, Allen and Lucero (1998) found that managers were specific targets (i.e. were retaliated against for actions such as criticizing the perpetrator’s job performance) in 90 per cent of the cases investigated. However, this finding may reflect the hierarchical relationship between the aggressor and target; since aggressing against someone at a higher organizational level is riskier, it seems unlikely that an individual would do so unless

44 Sexual Harassment

provoked by that particular individual. Thus, it is not surprising that upward aggression was directed at specific rather than non-specific (managerial) targets. To the extent that sexual harassment is a form of workplace aggression, we would also expect to see provocative victims of sexual harassment. Recall that provocative victims become victims, in part, because they threaten the social identities of those with whom they interact (Aquino et al., 1999; Olweus, 1993). As noted by Tsui and Gutek (1999, p. 53), ‘positive social identity is based to a large extent on a comparison between the in-group (e.g. males) and some relevant out-group (e.g. females)’. According to social identity theory, one’s primary social identity could be based on any number of demographic factors (Tajfel and Turner, 1986); however, gender appears to be used frequently in work situations, especially those in which one gender is in the minority. Konrad, Winter and Gutek (1992) found that women in male-dominated work groups experienced more social isolation and sexist stereotyping, presumably because they were identified as an outgroup. The finding that more sexual harassment occurs in male-dominated occupations and workgroups (e.g. Burgess and Borgida, 1997; Gruber and Bjorn, 1982; Fitzgerald et al., 1997; Lafontaine and Tredeau, 1986) may be explained by social identity theory. For example, if men perceive women in blue-collar positions as taking a job away from a man (i.e. a member of the ‘in-group’ or ‘one of us’), then they may feel that their identity as a man has been threatened. Furthermore, given that many individuals base their identity on the type of work they do, some men (i.e. those with sexist attitudes) may feel a loss of self-worth if ‘even a woman can do their job’. Either way, their identity as a man is jeopardized and must be restored for them to feel good about themselves and their position. In general, positive social identity can be re-established by exaggerating inter-group differences (Tsui and Gutek, 1999). In the case at hand, this could be accomplished by using gender stereotypes as the basis for interacting with female ‘interlopers’. Since gender stereotypes imply that men should be sexually aggressive (Gutek, 1985), sexual harassment is one mechanism that is likely to be used by men (with sexist attitudes) to restore their positive social identity. In summary, the simple fact that a woman is performing ‘a man’s job’ may be enough for some men to feel ‘provoked’, thus rendering women in male-dominated occupations provocative victims or specific targets. (Note that this is not to say that blue-collar women intend to provoke their co-workers or that they are deserving of any negative behaviour targeted at them because of their gender.)

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 45

Dall’Ara and Maass (1999) empirically examined the association between threats to the male social identity and sexual harassment. In their experiment, they manipulated the degree of threat that a (fictitious) woman posed to the male social identity by stating either that (1) she held old-fashioned, traditional ideas about her future role in society, or (2) she expressed egalitarian values supportive of equal opportunities. Clearly, the egalitarian woman poses the greater threat to the male social identity. They hypothesized, and found, that male subjects would be more likely to sexually harass (i.e. send a computer file containing pornographic pictures) a woman who described herself as holding egalitarian rather than traditional sex-role values. As the researchers note, it could be argued that the men sent such pictures to women with egalitarian sex-role attitudes because they thought these women had liberal attitudes toward sex and might enjoy the pictures. However, there is some evidence that this is not the case; the researchers provided a control question in a subsequent study which asked participants to guess how women with each type of value system would respond if sent pornographic material. The results, which indicate that men believed egalitarian women were more likely to feel harassed than traditional women, provide support for a social identity theory of sexual harassment (Dall’Ara and Maass, 1999). Therefore, we propose that sexual harassment resembles other forms of workplace aggression in that both may involve ‘provocative victims’.

Submissive victims Although the evidence indicates that aggressors will often go after those who upset or anger them, there may be high costs associated with such actions. Indeed, in many of the arbitration cases examined by Allen and Lucero (1998), the subordinate who aggressed against a manager was terminated for insubordination. Therefore, displacement of hostility may occur and the employee may aggress against other (non-specific) targets (Berkowitz, 1993; O’Leary-Kelly et al., 1996). Also, there may be times when an employee is not able to attribute aversive treatment to one specific individual. In such cases, employees may become ‘aggressors in search of targets’ in as much as they behave aggressively toward any convenient target (Berkowitz, 1993; O’Leary-Kelly et al., 1996). As mentioned previously, submissive victims tend to be targeted because they are perceived as being less powerful and/or lacking status (Olweus, 1993). Factors such as level within the organizational hierarchy, seniority, and type of employment contract indicate status within the

46 Sexual Harassment

organization. According to the ‘effect–danger ratio’ (Bjorkqvist, Osterman and Lagerspetz, 1994), individuals with higher organizational status are less likely to be targets of workplace aggression because these people present potential ‘danger’ to the aggressor due to the power emanating from their status. On the other hand, lower organizational status has been proposed to act as a vulnerability marker. For example, Aquino et al. (1999) hypothesized that individuals occupying lower hierarchical levels were more likely to be victims of workplace aggression than those at higher levels because they possess less legitimate power and less control over organizational rewards and punishments (and therefore represent less potential for retaliation). Their data did not support this hypothesis. In other words, simply being ‘low on the totem pole’ is not enough to render someone a victim of workplace aggression. However, occupying a lower-level position in conjunction with exhibiting a high level of negative affectivity (i.e. distressing emotions such as hostility, fear, and anxiety) is associated with a higher incidence of workplace victimization. It appears that the relationship between hierarchical level and being the target of sexual harassment is similar to that described above for other forms of workplace victimization. For example, non-supervisory employees were not more likely than supervisory employees to state that they had experienced sexually harassing behaviours (e.g. USMSPB, 1995). Similarly, research indicates that the incidence of sexual harassment is distributed relatively evenly among occupational classes, i.e., professional/technical, clerical, administrative (e.g. USMSPB, 1995); such classes can be used as a proxy for organizational level to some extent. It should be noted that some research has found a relationship between occupational classes and sexual harassment (Terpstra and Cook, 1985). However, in our opinion, supervisory status is a better measure of hierarchical level than occupational class, so we argue that being ‘low on the totem pole’ is not sufficient to render someone the target of sexual harassment. We believe that being targeted has more to do with the hierarchical relationship between the initiator and the target of sexual harassment (as described below). Therefore, we propose that targets of sexual harassment resemble targets of other forms of workplace aggression with regard to their hierarchical level within the organization. Although no consistent link between absolute hierarchical level and workplace victimization (including sexual harassment) has been established, the hierarchical level of a victim relative to that of the initiator has been found to be a predictor of workplace aggression. Specifically, victims of both sexual harassment and other forms of workplace aggression

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 47

tend to be at the same hierarchical level as their harassers/aggressors. For example, one study found that over 75 per cent of employees who experienced sexual harassment reported being harassed by co-workers (USMSPB, 1995). Similarly, studies of workplace aggression find that ‘the majority of aggression occurs among co-workers at the same organizational level’ (Neuman and Baron, 1998, p. 400). Therefore, we propose that targets of sexual harassment resemble targets of other forms of workplace aggression with regard to their hierarchical relationship to the harasser/ aggressor. Aquino et al. (1999) suggest that being poorly integrated into the social network of an organization may be another factor leading to workplace victimization. This may explain the finding that most sexual harassment and other forms of workplace aggression occur between employees at the same level because an employee is more likely to know who the ‘outcasts’ are at his level than at other levels within the hierarchy. Research on workplace aggression indicates that the use of non-traditional employment contracts (e.g. the use of ‘part-timers’) is positively related to increases in both witnessed and experienced aggression at work (Baron and Neuman, 1996). While type of employment contract indicates status within the organization, it is also likely to be associated with integration into social or friendship networks. In other words, individuals with non-traditional employment contracts are less likely to be ‘well-connected’ to others at work. Martindale’s (1991) finding that newer employees are more likely to be sexually harassed seems consistent with the social network perspective, as does Rogers and Henson’s (1997) finding that temporary workers are more likely to be harassed. Therefore, we propose that poor integration into the social network is common to targets of both sexual harassment and other forms of workplace aggression. USMSPB (1988, 1995) studies have consistently found that married women are less likely to report being sexually harassed than unmarried women. Marital status of an employee, therefore, appears strongly related to the experiences of sexual behaviours in the workplace. The question of why married women would experience less sexual harassment can be approached and understood from a variety of perspectives. Luthar and Pastille (2000), perhaps tongue in cheek, have pointed out that sexually harassing a married woman may be inherently riskier (i.e. have a lower effect–danger ratio) due to the possible involvement of an interested third party (the woman’s husband), who may choose to conduct his own informal investigation of the situation! Fain and Anderton (1987) have attempted to explain this finding by suggesting

48 Sexual Harassment

that in our society the sexual pursuit of unmarried women is more socially acceptable in that it can be interpreted as normal male courtship behaviour rather than sexual harassment. Since the initiator’s intention can be disguised when pursuing single women, the behaviour can be classified as covert aggression (Bjorkqvist, Osterman and Lagerspetz, 1994). Given that covert aggression is more likely to occur in the workplace than overt aggression (e.g. Baron and Neuman, 1996), the finding that single women are more likely to be sexually harassed is consistent with findings regarding general workplace aggression. Although we could not find any research examining the link between marital status and general workplace aggression, we propose that no relationship exists. As described above, one explanation for single people being the most frequent targets of sexual harassment is that targeting someone who is ‘available’ incorporates an element of ambiguity into the behaviour (it could be sincere romantic interest); thus it is a type of covert (rather than overt) aggression in as much as intentions can be credibly disguised. However, the ‘romantic interest’ justification seems ridiculous when the behaviour is general workplace aggression. For example, it is difficult to imagine anyone believing an employee who says that he or she sabotaged the work of another employee in an attempt to establish a dating relationship. Therefore, we propose that marital status distinguishes targets of sexual harassment from targets of other forms of workplace aggression. Finally, since the most common response of individuals who are sexually harassed is to ignore the behaviour (Knapp et al., 1997), there is little disincentive for harassers to stop. In terms of the effect/danger ratio discussed earlier, this lack of reporting reduces the danger for harassers, thus making them more likely to pursue this form of workplace aggression. Research has consistently shown that only a small fraction of victims of sexual harassment actually report such incidents. For example, in a study of over 8,000 federal employees, only 6 per cent of those who experienced sexually harassing behaviours filed a grievance or took formal action against the aggressor (USMSPB, 1995). The fear of not being believed, the fear of job loss, and possible retaliation for complaining are among the reasons that some women choose not to report sexually harassing behaviour (Cochran, Frazier and Olson, 1997). It seems unlikely that victims of other forms of workplace aggression would face these same fears. For example, it is hard to imagine an employee refraining from reporting a co-worker who pushes him down the stairs because he fears others will not believe him. Therefore, we propose that the likelihood of reporting objectionable behaviour distinguishes

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 49

targets of sexual harassment from targets of other forms of workplace aggression.

Contextual factors In addition to the characteristics of offenders and victims described in the preceding sections of this chapter, situational factors (Aquino et al., 1999; Elias, 1986) must be considered to gain a comprehensive understanding of the distinctions between sexual harassment and other forms of workplace aggression. As described by Elias (1986), ‘regulatory failures’ refer to the use of inadequate rules and practices for controlling criminal behaviour. Such failures may also occur within an organizational context when policies and practices fail to control various forms of misbehaviour (some of which is illegal and thus ‘criminal’ and some of which is simply deviant or anti-social). One regulatory mechanism that is widely used in organizations to prevent sexual harassment (and to a lesser extent, workplace aggression) is a policy statement. Workplace policies and practices can act as deterrents to sexual harassment and other forms of workplace aggression, to the extent that they increase the ‘danger’ confronting a perpetrator. According to Levy and Paludi (2002, p. 91), an effective policy on sexual harassment ‘requires more than a statement of compliance with the law’, and in fact may require organizations ‘to put together a veritable manual, many pages in length’. In addition to setting an appropriate tone, they note that an effective policy should also express strong disapproval, encourage (not simply allow) reports, and use understandable language (among other things). However, O’Leary-Kelly et al. (2001) argue that many organizations’ policies will not prevent sexual harassment in as much as they ‘allow sexual harassers to renounce accountability for their actions’. One aspect of a sexual harassment policy that is especially relevant to such renouncing of accountability by harassers is the use of ambiguous language. Research on accountability indicates that ambiguous standards are less likely to shape behaviour (Frink and Klimoski, 1998), including sexually harassing behaviour, because harassers can argue that they did not understand the organization’s policy (O’Leary-Kelly et al., 2001). This may explain why ‘clarifying prohibited conduct’ so that employees ‘know what it is they are expected to do or not do’ is critical for an effective policy (Levy and Paludi, 2002, p. 93). Although organizations have also been told to provide examples of inappropriate behaviour in their policies on workplace aggression and

50 Sexual Harassment

violence (e.g. Neuman and Baron, 1998), it seems to us that there is less ‘wiggle room’ for employees who violate these policies than for employees who violate sexual harassment policies. For example, while a sexual harasser could claim that he did not know his sexual advances toward a co-worker were unwelcome (which is a requirement for a behaviour to be classified as illegal sexual harassment), it is hard to imagine an employee claiming that he did not know his yelling at a co-worker was unwelcome. Therefore, we propose that although organizational policy statements are common for both sexual harassment and other forms of workplace aggression, the issues of accountability emanating from these policies differentiate sexual harassment from other forms of workplace aggression. To the extent that organizational policies are written simply to fulfil legal obligations and are not integrated into an organization’s culture, they will do little to prevent sexual harassment or other forms of workplace aggression. In fact, research indicates that the culture or norms of an organization contribute to both general workplace aggression and sexual harassment. For example, Robinson and O’Leary-Kelly (1998) found that aggression is socially contagious in the workplace in the sense that employees who behave aggressively often do so because other employees in their work group behaved aggressively. Similarly, Lavite (1991 – as cited in Pryor, Lavite and Stoller, 1993) found that the presence of a harassing role model increased the amount of sexually harassing behaviour initiated by men with a high likelihood to harass. Watching others misbehave can trigger similar behaviour in the observers for several reasons: observers learn to imitate such behaviours, observers’ inhibitions regarding aggressive behaviour may be reduced, observing such behaviour may stimulate emotional arousal that leads to aggression, and the behaviour may focus observers’ attention on potential targets of aggression (O’Leary-Kelly et al., 2000). Therefore, it appears that both sexual harassers and individuals who initiate other forms of workplace aggression are influenced by contextual factors such as workplace norms.

Conclusion This chapter established a framework for comparing and contrasting sexual harassment and other forms of workplace aggression. A summary of the similarities and differences between the two are presented in Table 3.1. Understanding the similarities and differences between these two phenomena is important for both organizational scholars and practitioners. Organizational scholars can use the analysis presented in this chapter to guide their future research. Our framework has raised several research questions regarding perpetrators. For example, in addition to

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 51 Table 3.1 A summary of the similarities and differences between sexual harassment and other forms of workplace aggression Similarities Forms of behaviour • Existence of two forms – overt and covert • Covert forms are more common than overt forms Perpetrators/initiators • Low self-monitors • At same level within organizational hierarchy as victim/target • Hierarchical level within organization not relevant Victims/targets • Existence of provocative victims and specific targets • Existence of submissive victims and non-specific targets • Hierarchical level within organization not relevant • Poor integration into the organization’s social network Contextual factors • Organizational attempts to control through policy and training efforts • Influence of workplace norms and culture Differences Forms of behaviour • Illegality of behaviour (sexual harassment is illegal; many other forms are not) Perpetrators/initiators • Typical gender (male for sexual harassment, no typical gender for other forms) • Ability to plead misunderstanding of policy (easier for sexual harassment) Victims/targets • Typical gender (female for sexual harassment, no typical gender for other forms) • Typical marital status (single for sexual harassment, no typical status for other forms) • Likelihood of reporting behaviour (less likely for sexual harassment)

predicting sexual harassment, does Pryor’s (1987) Likelihood to Sexually Harass Scale predict other forms of workplace aggression? Are there clear perpetrator patterns? Is perpetrator behaviour over time best described as a ladder (e.g. progressing from covert to overt sexual harassment) or as a bridge (e.g. remaining at the covert level but encompassing sexual harassment and other forms of workplace aggression)? Or do both patterns exist under different conditions? Do perpetrators of sexual harassment and other forms of workplace aggression tend to be well-integrated into social networks at work – or are they more like the ‘loner’ or typical perpetrator or workplace homicide? This chapter has also raised interesting issues regarding targets of sexual harassment and other forms of workplace aggression. Future

52 Sexual Harassment

research questions regarding targets include the following: Are individuals who are targets of sexual harassment also targets of other forms of workplace aggression? One study suggests that this may be the case. A study by Rospenda et al. (2000) found that targets of chronic sexual harassment also tended to be targets of chronic generalized workplace abuse. As an interesting follow-up question, researchers could investigate whether such targets tend to be specific/provocative targets or nonspecific/submissive targets. Finally, this chapter has raised some interesting topics for future research focusing on the contextual factors surrounding incidents of sexual harassment and other forms of workplace aggression. For example, do organizations that exhibit high levels of sexual harassment also exhibit high levels of other forms of workplace aggression? If not, what distinguishes the two? In addition to predicting sexual harassment, does the gender distribution within an occupation, workgroup or organization predict other forms of workplace aggression? Do training programmes that focus on generalized workplace abuse or aggression decrease the amount of sexual harassment occurring within an organization? Organizational practitioners can use the analysis presented in this chapter to guide their effort toward preventing (or at least controlling) sexual harassment and other forms of workplace aggression. Our framework provides support for the prevention strategies offered by Neuman and Baron (1998). Specifically, organizations that want to prevent sexual harassment and other forms of workplace aggression can focus on four issues: (1) personnel selection (i.e. use of background checks and preemployment testing to identify potential aggressors and harassers), (2) clearly articulated and consistently implemented policies against aggression and sexual harassment that spell out definitions of misbehaviours, procedures for reporting such behaviours, and specific sanctions for violations, (3) the maintenance of positive organizational cultures and climates through role modelling of appropriate behaviours by leaders, ensuring fair treatment, and treating employees with trust, respect and dignity, and (4) programmes designed to cope with dysfunctional attitudes and behaviours such as training and employee assistance programmes.

References Allen, R. E. and Lucero, M. A. (1998) ‘Subordinate aggression against managers’, International Journal of Conflict Management, 9, 234–58. Aquino, K., Grover, S. L., Bradfield, M. and Allen, D. G. (1999) ‘The effects of negative affectivity, hierarchical status, and self-determination on workplace victimization’, Academy of Management Journal, 42, 260–72.

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 53 Baron, R. A. and Neuman, J. H. (1996) ‘Workplace violence and workplace aggression: evidence on their relative frequency and potential causes’, Aggressive Behavior, 22, 161–73. Baron, R. A., Neuman, J. H. and Geddes, D. (1999) ‘Social and personal determinants of workplace aggression: evidence for the impact of perceived injustice and the Type A behavior pattern’, Aggressive Behavior, 25, 281–96. Berkowitz, L. (1993) Aggression: Its Causes, Consequences, and Control. New York: McGraw-Hill. Bettencourt, B. A. and Miller, N. (1996) ‘Gender differences in aggression as a function of provocation: a meta-analysis’, Psychological Bulletin, 119, 422–47. Bjorkqvist, K., Osterman, K. and Hjelt-Back, M. (1994) ‘Aggression among university employees’, Aggressive Behavior, 20, 173–84. Bjorkqvist, K., Osterman, K. and Lagerspetz, K. M. J. (1994) ‘Sex differences in covert aggression among adults’, Aggressive Behavior, 20, 27–33. Burgess, D. and Borgida, E. (1997) ‘Sexual harassment: an experimental test of sex-role spillover theory’, Personality and Social Psychology Bulletin, 23, 63–75. Cochran, C. C., Frazier, P. A. and Olson, A. M. (1997) ‘Predictors of responses to unwanted sexual harassment’, Psychology of Women Quarterly, 21, 207–26. Cohn, E. G. and Rotton, J. (1997) ‘Assault as a function of time and temperature: a moderator-variable time-series analysis’, Journal of Personality and Social Psychology, 72, 1322–34. Dall’Ara, E. and Maass, A. (1999) ‘Studying sexual harassment in the laboratory: are egalitarian women at higher risk?’, Sex Roles, 41, 681–704. Dateline (1996) News story about sexual harassment at the Mitsubishi Plant in Normal Illinois. NBC. Douglas, S. C. and Martinko, M. J. (2001) ‘Exploring the role of individual differences in the prediction of workplace aggression’, Journal of Applied Psychology, 86, 547–59. Driscoll, D. M., Kelly, J. R. and Henderson, W. L. (1998) ‘Can perceivers identify likelihood to sexually harass?’, Sex Roles, 38, 557–88. Eagly, A. H. and Steffen, J. (1986) ‘Gender and aggressive behavior: a metaanalytic review of the social psychological literature’, Psychological Bulletin, 100, 309–30. EEOC Notice Number 915.002 (1999) http://www.eeoc.gov/docs/harassment.html#I Elias, R. (1986) The Politics of Victimization: Victims, Victimology, and Human Rights. New York: Oxford Press. Fain, T. C. and Anderton, D. L. (1987) ‘Sexual harassment: organizational context and diffuse status’, Sex Roles, 16, 291–311. Fitzgerald, L. F., Drasgow, F., Hulin, C. L., Gelfand, M. J. and Magley, V. J. (1997) ‘Antecedents and consequences of sexual harassment in organizations: a test of an integrated model’, Journal of Applied Psychology, 82(4), 578–89. Folger, R. and Baron, R. A. (1996) ‘Violence and hostility at work: a model of reactions to perceived injustice’, in G. R. Van den Bos and E. Q. Bulatao (eds), Workplace Violence (pp. 51–85). Washington DC: American Psychological Association. Frink, D. D. and Klimoski, R. J. (1998) ‘Toward a theory of accountability in organizations and human resources management’, in G. R. Ferris (ed.), Research in Personnel and Human Resources Management (pp. 1–51). Greenwich, CT: JAI Press.

54 Sexual Harassment Geddes, D. and Baron, R. A. (1997) ‘Workplace aggression as a consequence of negative performance feedback’, Management Communication Quarterly, 10, 433–55. Gelfand, M. J., Fitzgerald, L. F. and Drasgow, F. (1995) ‘The structure of sexual harassment: a confirmatory analysis across cultures and settings’, Journal of Vocational Behavior, 47, 164–77. Gruber, J. E. and Bjorn, L. (1982) ‘Blue-collar blues: the sexual harassment of women autoworkers’, Work and Occupations, 9, 271–98. Gutek, B. (1985) Sex and the Workplace. San Francisco: Jossey-Bass. Guynn, J. (1998) ‘Mean business: Benicia couple lead fight against workplace bullying’, Contra Costa Times, Knight-Ridder Newspapers, October 7. Hyde, J. S. (1984) ‘How large are gender differences in aggression? A developmental meta-analysis’, Developmental Psychology, 20, 722–36. Knapp, D. E., Faley, R. H., Ekeberg, W. C. and Dubois, C. L. Z. (1997) ‘Determinants of target responses to sexual harassment: a conceptual framework’, Academy of Management Review, 22(3), 687–729. Konrad, A. M., Winter, S. and Gutek, B. A. (1992) ‘Diversity in workgroup sex composition: implications for majority and minority members’, in P. Tolbert and S. B. Bachrach (eds), Research in the Sociology of Organizations, 10 (pp. 115–40). Greenwich, CT: JAI Press. Lafontaine, E. and Tredeau, L. (1986) ‘The frequency, sources, and correlates of sexual harassment among women in traditional male occupations’, Sex Roles, 15, 433–42. Lavite, C. M. (1991) ‘The interaction between situational factors and individual predispositions in the likelihood to sexually harass’, unpublished masters thesis, Illinois State University. Levy, A. C. and Paludi, M. A. (2002) Workplace Sexual Harassment, 2nd edn. New Jersey: Prentice-Hall. Lucero, M. A., Middleton, K., Finch, W. and Valentine, S. (2001) ‘An investigation of sexual harassment perpetrators: a classification of perpetrator patterns’, paper presented at the Academy of Management Meeting, Washington, D.C. Luthar, H. K. and Pastille, C. (2000) ‘Modeling subordinate perceptions of sexual harassment: the role of superior–subordinate social-sexual interaction’, Human Resource Management Review, 10, 211–44. Martindale, M. (1991) Sexual Harassment in the Military: 1988 Report. Arlington, VA: Defense Manpower Data Center. Murphy, J. D., Driscoll, D. M. and Kelly, J. R. (1999) ‘Differences in the nonverbal behavior of men who vary in the likelihood to sexually harass’, Journal of Social Behavior and Personality, 14, 113–29. Neuman, J. H. and Baron, R. A. (1997) ‘Type A behavior pattern, self-monitoring, and job satisfaction as predictors of aggression in the workplace’, in G. Chao (Chair), Counterproductive Job Performance and Organizational Dysfunction. Symposium conducted at the meeting of the Society for Industrial and Organizational Psychology, St. Louis, MO. Neuman, J. H. and Baron, R. A. (1998) ‘Workplace violence and workplace aggression: evidence concerning specific forms, potential causes, and preferred targets’, Journal of Management, 24, 391–420. O’Leary-Kelly, A. M. and Bowes-Sperry, L. (2001) ‘Sexual harassment as unethical behavior: the role of moral intensity’, Human Resource Management Review, 11, 73–92.

Lynn Bowes-Sperry, Jasmine Tata and Harsh K. Luthar 55 O’Leary-Kelly, A. M., Griffin, R. W. and Glew, D. J. (1996) ‘Organizationmotivated aggression: a research framework’, Academy of Management Review, 21, 225–53. O’Leary-Kelly, A. M., Paetzold, R. L. and Griffin, R. W. (2000) ‘Sexual harassment as aggressive behavior: an actor-based perspective’, Academy of Management Review, 25, 372–88. O’Leary-Kelly, A. M., Tiedt, P. and Bowes-Sperry, L. (2001) ‘Answering accountability questions in sexual harassment: insights regarding harassers, targets, and observers’, paper presented at the Southern Management Association, New Orleans. Olweus, D. (1979) ‘Stability of aggressive reaction patterns in males: a review’, Psychological Bulletin, 86, 852–75. Olweus, D. (1993) ‘Victimization by peers: antecedents and long-term outcomes’, in K. H. Rubin and J. B. Asendorf (eds), Social Withdrawal, Inhibition, and Shyness in Childhood (pp. 315–41). Chicago: University of Chicago Press. Perry, E. L., Kulik, C. T. and Schmidtke, J. M. (1998) ‘Individual differences in the effectiveness of sexual harassment awareness training’, Journal of Applied Social Psychology, 28, 698–723. Pryor, J. B. (1987) ‘Sexual harassment proclivities in men’, Sex Roles, 17, 269–90. Pryor, J. B., Lavite, C. M. and Stoller, L. M. (1993) ‘A social psychological analysis of sexual harassment: the person/situation interaction’, Journal of Vocational Behavior, 42, 68–83. Pryor, J. B., and Whalen, N. J. (1997) ‘A typology of Sexual harassment: characteristics of harassers, and social circumstances under which sexual harassment occurs’, in O’Donohue, William (ed.) Sexual Harassment: Theory, Research, and Treatment. Boston: Allyn & Bacon, pp. 129–51. Richman, J. A., Rospenda, K. M., Nawyn, S. J., Flaherty, J. A., Fendrich, M., Drum, M. L. and Johnson, T. P. (1999) ‘Sexual harassment and generalized workplace abuse among university employees: prevalence and mental health correlates’, American Journal of Public Health, 89, 358–64. Robb, L. A. and Doverspike, D. (2001) ‘Self-reported proclivity to harass as a moderator of the effectiveness of sexual harassment-prevention training’, Psychological Reports, 88, 85–8. Robinson, S. L. and Bennet, R. J. (1995) ‘A typology of deviant workplace behaviors: a multidimensional scaling study’, Academy of Management Journal, 38, 555–72. Robinson, S. L. and O’Leary-Kelly, A. M. (1998) ‘Monkey see, monkey do: the influence of work groups on the antisocial behavior of employees’, Academy of Management Journal, 41, 658–72. Rogers, J. K. and Henson, K. D. (1997) ‘‘‘Hey, why don’t you wear a shorter skirt?” Structural vulnerability and the organization of sexual harassment in temporary clerical employment’, Gender and Society, 11, 215–37. Rospenda, K. M., Richman, J. A., Wislar, J. S. and Flaherty, J. A. (2000) ‘Chronicity of sexual harassment and generalized work-place abuse: effects on drinking outcomes’, Addiction, 95, 1805–20. Rudman, L. A. and Borgida, E. (1995) ‘The afterglow of construct accessibility: the behavioral consequences of priming men to view women as sexual objects’, Journal of Experimental Social Psychology, 31, 493–517. Sbraga, T. P. and O’Donohue, W. (2000) ‘Sexual harassment’, Annual Review of Sex Research, 11, 258–86.

56 Sexual Harassment Skarlicki, D. P. and Folger, R. (1997) ‘Retaliation in the workplace: the roles of distributive, procedural, and interactional justice’, Journal of Applied Psychology, 82, 434–43. Tajfel, H. and Turner, J. C. (1986) ‘The social identity theory of intergroup behavior’, in S. Worchel and W. G. Austin (eds), Psychology of Intergroup Relations (pp. 7–24). Chicago: Nelson-Hall. Terpstra, D. E. and Cook, S. E. (1985) ‘Complainant characteristics and reported behaviors and consequences associated with formal sexual harassment charges’, Personnel Psychology, 38, 559–74. Terpstra, D. E. and Baker, D. D. (1988) ‘Outcomes of sexual harassment charges’, Academy of Management Journal, 31, 185–94. Terpstra, D. E. and Baker, D. D. (1992) ‘Outcomes of Federal Court Decisions on sexual harassment’, Academy of Management Journal, 35, 181–90. Tsui, A. S. and Gutek, B. A. (1999) Demographic Differences in Organizations. Maryland: Lexington Books. US Equal Employment Opportunity Commission (1981) ‘Guidelines on discrimination because of sex’, in US Merit System Protection Board (ed.), Sexual Harassment in the Workplace: Is it a Problem? Washington, DC: Government Printing Office. US Merits System Protection Board (USMSPB) (1988) Sexual Harassment in the Federal Workplace. An Update. Washington DC: US Government Printing Office. US Merits System Protection Board (1995) Sexual Harassment in the Federal Workplace. Trends, Progress and Continuing Challenges. Washington DC: US Government Printing Office. Wayne, J. H., Riordan, C. M. and Thomas, K. M. (2001) ‘Is all sexual harassment viewed the same? Mock juror decisions in same- and cross-gender cases?’, Journal of Applied Psychology, 86, 179–87. Welsh, S. (1999) ‘Gender and sexual harassment’, Annual Review of Sociology, 25, 169–90. Welsh, S. (2000) ‘The multidimensional nature of sexual harassment: an empirical analysis of women’s sexual harassment complaints’, Violence against Women, 6(2), 118–41.

4 Covering-Up Behaviour in Organizations Shmuel Stashevsky and Jacob Weisberg*

Introduction Workers’ unethical behaviour may stem from a variety of sources. In a recent nationwide study conducted in the US (McShulskis, 1997), almost half of the workers (48 per cent) reported that they responded to job pressures by acting either unethically or illegally. Moreover, 58 per cent had considered doing so at certain times during their employment. Two of the most common types of unethical behaviour are ‘cutting corners’ on quality control (33 per cent) and covering up incidents (30 per cent). ‘Cutting corners’ has to do with the individual worker and his work performance, while covering-up behaviour involves both the individual and his or her co-workers, managers, and subordinates. Covering-up behaviour is commonly perceived as workers intending to hide from supervisors their own or their co-workers’ behavioural or performancerelated wrongdoings, such as, for example, hiding mistakes on the work site. When an employee observes peers behaving unethically he or she might face a dilemma – whether to disclose the misconduct or to cover up (King and Hermodson, 2000). As such, employees will strive to reduce any unfavourable consequences to themselves and to the firm. A high degree of covering-up behaviour in organizations is an indication of an atmosphere of disloyalty, betrayal, or other unethical aspects. Beu and Buckley (2001) state that the damages from the different types of unethical behaviour are huge, since they inflict costs of billion of dollars on firms in the industry, as well as substantial damage to their * Authors appear in alphabetical order, having contributed equally to this chapter. 57

58 Covering-Up Behaviour in Organizations

image. As such, the question concerning the vulnerability of firms to unethical behaviour and covering-up behaviour is important for management. Lucas (1998) relates the covering-up syndrome to the role of vision in the company, suggesting that within companies lacking in vision, employees merely carry out what is required of them, avoiding difficulties with the help of political behaviour and concealment of errors. In spite of the severe problem of covering-up behaviour, it has not been given sufficient theoretical or empirical attention. In fact, a review of the business ethics literature indicates the lack of a comprehensive theory dealing with the underlying causes of unethical behaviour (Grover and Hui, 1994; Randall and Gibson, 1990). This study investigates covering-up behaviour of employees while suggesting a comprehensive model and testing it empirically. The model includes personal, occupational, and organizational characteristics that relate to and influence the covering-up behaviour in organizations. When an employee observes unethical behaviour by peers, supervisors or subordinates, he or she is likely to face the dilemma of reporting or covering up this behaviour. This type of conflict is resolved by a subjective benefit/cost analysis. King and Hermodson (2000) summarized several factors affecting the decision: the severity of the wrongdoings, the personal role responsibility, the group norms regarding peer disclosure, the perceived threat of co-workers compared with the organization’s retaliation, and the organization’s culture. Similarly, Elangovan and Shapiro (1998) argued that concealing information regarding unethical behaviour is a form of opportunistic betrayal. As with all types of betrayal in the workplace, it is expected to represent a rational behaviour, which weighs the benefits to be gained from the betrayal versus the satisfaction to be gained by remaining trustworthy. The considerations involve an evaluation of the relationship between the employee and the employer (bestowal of trust), and an evaluation of the ethical principles relating to betrayal and loyalty. Further explanation of the covering-up behaviour is provided by the theory of self-interest, which is related to situations of conflict of interests. Based on the self-interest theory (Trevino and Youngblood, 1990), we suggest that people will cover up when the behaviour benefits them. This was applied to the employee–employer relationship via the agency theory (Beu and Buckley, 2001), which contends that employees may cover up wrongdoings from their employer when they have more information than the employer, thus reducing the chances of the covering-up being discovered. In any case, personal benefits

Shmuel Stashevsky and Jacob Weisberg 59

gained for the employee are a prerequisite for such behaviour (Eisenhardt, 1989). The organizational atmosphere among the employees appears to have a significant influence on covering-up behaviour in the work place. Covering up or disclosing unethical behaviour frequently includes feelings of pressure, fear, and threat of the consequences of either decision being used. Fear also stems from work pressures placed on employees and may thus cause them to hide their mistakes in order to escape punishment (Ryan and Oestreich, 1991, pp. 61–2). This concept is reflected in the writings of Deming (1986, pp. 59–60), suggesting that fear leads one to ‘hide’ wrongdoings thereby preventing management from hearing or seeing the inappropriate work plan, poor training, etc. Social and organizational variables are believed to influence the inclination to cover up wrongdoings. For instance, managers as role models can improve the manner in which workers cope with their wrongdoings. If open-minded managers admit their own mistakes and take responsibility for them, it might encourage workers to act in a similar manner (Ryan and Oestreich, 1991, p. 160). Furthermore, covering up or disclosing unethical behaviour is also related to national culture, which may influence the extent of workers’ covering-up and their perception of its morality. For example, Puffer and McCarthy (1995) claim that while American managers regard covering-up as unethical, Russian managers tend to view it as a legitimate behaviour, since they are expected to cope with unrealistic goals set by their bosses. Russian managers do not feel that meeting the goals and deadlines set for them depends entirely on themselves, since in bureaucratic institutions they have low control over production resources, e.g. materials, manpower, and equipment. While this reality may have changed since the transition of Russia to a free market economy, it takes much longer for people’s ethical attitudes and behaviour to change. This concept is supported by an empirical study in which Russian managers rated work events and practices in terms of frequency and ethical value (Ivancevich, DeFrank and Gregory, 1992). ‘Concealing mistakes’ and ‘violations of work rules’ were found to occur quite frequently and not considered as unethical practices.

A multi-dimensional model of covering-up Hitherto it was clarified that covering-up behaviour does not stem from a single factor, but rather from a complex web of groups of factors, which may influence it in different ways.

60 Covering-Up Behaviour in Organizations

Personal characteristics: • • • •

Gender Age Marital status Education

Job/Occupational characteristics: • • • •

Supervisory level Full/part-time job Tenure status Length of service

Covering-up behaviour

Organizational characteristics: • • • • •

Business sector Local/export orientation Firm size Work standardization Firm growth trend

Figure 4.1

A multi-dimension covering-up behaviour model

We offer here a comprehensive model (see Figure 4.1), which aims to explain and predict covering-up behaviour, by studying the concurrent impact of demographic, job/occupational and organizational characteristics. The first group of demographic variables includes factors related to the worker’s personal status, e.g. gender, age, marital status, education. The second group refers to job and occupational characteristics, e.g. full-time versus part-time work, length of service with the employer. The third group includes organizational factors and characteristics, e.g. business sector, company size, work standardization (e.g. ISO 9000), and economic growth trend. We will elaborate on the group of organizational characteristics by including factors that deal not only with the impact of organizational–structural characteristics on covering-up behaviour but also with the impact of firms’ market orientation and the economic growth trend of the firm on covering-up behaviour.

Shmuel Stashevsky and Jacob Weisberg 61

Research hypotheses Within the framework of the model, we propose a set of hypotheses regarding the relationship between each of the factors and the employee’s reported covering-up behaviour. The rationale underlying several of the hypotheses is based on previous research (King and Hermodson, 2000; McDonald and Kan, 1997). Since we include in our model new factors relating to the firms’ market competition and their economic growth we will suggest a rationale as to how they relate to covering-up behaviour. In addition, we will test the relative contribution of each factor in explaining covering-up behaviour.

(a) Personal characteristics Gender: The relationship between gender and ethical dilemmas has repeatedly shown inconclusiveness and mixed results. Dawson (1997) argues that there are significant ethical behaviour differences by gender. In spite of the fact that on some issues males and females demonstrated similar perceptions, on ethical issues a difference was observed: females were less likely to cheat; they were more concerned about ethical issues; they were more inclined to disagree with unethical actions, and in general to be more ethically oriented toward work decisions than males (Beu and Buckley, 2001). As such, ethical behaviour can be partly explained by gender differences. From a labour market perspective, however, unemployment rate among females is constantly higher than for their male counterparts. As such, they feel more vulnerable and tend to display higher covering-up behaviour so as to retain their employment. Based on the above arguments we suggest that women can be expected to display higher covering-up behaviour than men due to their higher vulnerability. Hypothesis 1: Male employees will report lower levels of covering-up behaviour than their female counterparts. Age: The relationship between age and ethical behaviour is inconclusive (Beu and Buckley, 2001). McDonald and Kan (1997) found a significant correlation between age and ethical beliefs, with older employees less likely to express agreement with unethical actions than younger employees. However, although age might have a significant influence on ethical beliefs, the type of relationship (e.g. linear or nonlinear) is not clear. We suggest that ethical behaviour might be affected by the vulnerability of

62 Covering-Up Behaviour in Organizations

employees. Unemployment rates by age groups show a U-shape curve, i.e. younger and older workers have higher unemployment rates than mid-age employees. Since they are at greater risk of being unemployed or dismissed from their jobs, they are likely to exhibit higher levels of covering-up behaviour than mid-age workers. Hypothesis 2: Middle-age employees will report lower covering-up behaviour than their younger or older colleagues. Marital status: Serwinek (1992) investigated the influence of various variables, including the impact of marital status, on ethical decisionmaking. Do married people behave differently from single people when considering ethical dilemmas? Probably the different sense of responsibility of a married employee towards his or her family plays an important role in ethical behaviour as compared with single counterparts. Decisions of married people are expected to be more considerate and responsible. Hypothesis 3: Married employees will report lower covering-up behaviour than co-workers who are single. Education: The relationship between employees of different educational levels and ethical behaviour has been studied widely (Dubinsky and Ingram, 1984; Serwinek, 1992), suggesting that employees with a higher level of education might make decisions concerning ethical issues with greater caution of the possible impact, and to consider more carefully the consequences of their actions. Moreover, as we suggested above, the degree of security at work as perceived by employees plays an important role in conducting ethical or unethical behaviour. Employees with a higher level of education generally feel less vulnerable to dismissals and are more secure in their job, having more confidence in their future employment with the company than those with less education. Hypothesis 4: Employees with more years of schooling will report lower covering-up behaviour than workers with fewer years of schooling.

(b) Job/occupational characteristics Supervisory level: Do higher-ranked employees show higher ethical behaviour than their junior colleagues? We can get some indication

Shmuel Stashevsky and Jacob Weisberg 63

from Zabid and Alsagoff (1993), who demonstrated that employees in higher positions in the hierarchical level were more disapproving of unethical behaviour. In contrast, Newstrom and Ruch (1975) provided evidence that the ethical beliefs of subordinates and those of higher-level managers were similar, and suggested that the similarity could be explained by the fact that higher-level managers are considered as role models by their subordinates. We suggest that since supervisors have more responsibility and greater prospects of promotion than their subordinates, they are generally more involved in and committed to their firm. Therefore, they are expected to set higher ethical standards. As such, they can be expected to display lower levels of covering-up than non-supervisors. Hypothesis 5: Employees holding supervisory positions will report lower covering-up behaviour than employees holding non-supervisory positions. Full/part-time job: In the literature, the distinction between part-time and full-time workers in relation to covering-up behaviour has not been studied. Nevertheless, this distinction seems to be very relevant to covering-up. We suggest that part-time employees are more likely to work elsewhere too, or to be engaged in other roles that do not fit their work roles and are sometimes even in conflict with their duties in the firm. Therefore, they might be less committed to the firm and feel less responsible for the future of the firm as compared with employees with full-time jobs. Hypothesis 6: Workers holding full-time jobs will report lower covering-up behaviour than workers engaged in part-time work. Tenure status: In our review of the literature, we did not find any empirical research that studied the relationship between tenure status and covering-up behaviour. Yet this relationship seems reasonable. We propose that workers with job tenure are more likely to establish a long-term employment relationship as well as a psychological contract with the firm than are non-tenured workers. The former accumulate monetary and non-monetary benefits as they establish this relationship. In contrast, non-tenured employees may have less organizational commitment, as well as less job security. Moreover, their psychological contract with the firm is more fragile. Based on the above, it seems that less commitment leads to more covering-up, and we suggest, therefore, that non-tenured workers can be expected to display higher levels of covering-up behaviour than workers with job tenure.

64 Covering-Up Behaviour in Organizations

Hypothesis 7: Workers with job tenure will report lower covering-up behaviour than non-tenured workers. Length of service: King and Hermodson (2000) suggest that newcomers to an organization are less likely to disclose wrongdoings since they feel they have not had the opportunity to learn how to judge the organizational norms and procedures, as well as group norms and behaviour. On the other hand, they may be more willing to report on wrongdoings of others in order to be in good standing with their superior, which could reward them in the future. However, these behaviours change over the course of one’s working years. Harris (1990) found an inverse relationship between tenure and unethical behaviour. Specifically, he found that the longer the service with the current employer the less the employee will be tolerant of unethical activity. We suggest that employees with longer years of service in their current firm generally feel more secure in their jobs and maintain a more loyal relationship with the company. Employees with longer years of service can be expected to demonstrate lower levels of covering-up behaviour than employees with a short service period. Hypothesis 8: Workers with more years of employment will report lower covering-up behaviour than those with less years of employment.

(c) Firm/organizational characteristics Business sector: Working standards and supervision in manufacturing firms, as compared with non-manufacturing firms (e.g. service organizations), emphasize more control and supervision. Moreover, the manufactured products are tangible and are constantly evaluated by the customers, thus undergoing a type of quality control (Bounds et al., 1994, pp. 49–51). Therefore, covering-up is more likely to be discovered in manufacturing firms as compared with non-manufacturing firms. Hypothesis 9: Workers in manufacturing firms will report less covering-up behaviour than workers employed in non-manufacturing firms. Local market/export orientation: In the literature, we could not find any reference to the relationship between covering-up or unethical behaviour in general, and the destination of the firm’s products: local vs. export markets. Should we expect employees’ covering-up behaviour to be different for exporters compared with those who sell in the local market?

Shmuel Stashevsky and Jacob Weisberg 65

We suggest that companies exporting their products are subject to international competition and world standards, and therefore have to increase the demands from the workers for high-quality products. Reliability and quality of products are therefore more crucial for the survival of firms competing in export markets. Hypothesis 10: Workers in firms with an export orientation will report lower covering-up behaviour than firms and organizations that compete in the local market. Firm size: An additional factor which may influence ethical behaviour is the firm size – the number of workers employed in the firm (McDonald and Kan, 1997). Empirical studies show mixed results concerning the relationship between firm size and ethical behaviour. Dalton and Kesner (1988) concluded that firm size is inversely related to ethical behaviour, meaning that employees in large organizations are more likely to engage in unethical behaviour than employees in small firms. On the other hand, large companies usually have higher levels of internal audit and quality control, which reduces the employees’ incentive to coverup their mistakes. Thus, from this perspective, employees in large organizations are more reluctant to exhibit unethical behaviour and are less likely to coverup than employees in small firms. In spite of the conflicting rationales, the last seems more convincing, and we therefore suggest the following hypothesis: Hypothesis 11: Workers in firms employing a large number of employees will report lower levels of covering-up behaviour than workers in small firms. Work standardization: Competition among firms is increasing and becoming tougher. Due to quality demands the world business has gradually moved to international work standards, which ensure that products will be of a predetermined quality. Still the question of whether firms that have introduced mechanisms controlling work standards will show a positive relationship to ethical behaviour, is open. Employees in firms with structured work standards (e.g. ISO 9000) are subject to a higher level of supervision and ‘built-in’ work control. We suggest the following: Hypothesis 12: Workers in firms with structured work standards will report lower covering-up behaviour than workers in firms with no structured work standards.

66 Covering-Up Behaviour in Organizations

Firm growth trend: Another important factor that may be related to unethical behaviour is the firm’s economic growth trend. It is still an unresolved question in the business ethics literature, whether employees in firms on a growth trend will show higher ethical behaviour. Successful firms exhibiting a trend of economic growth will allocate more resources to improve their products and services, as well as to increase their employees’ job commitment and job satisfaction. Furthermore, the dimensions of wrongdoings in successful firms are probably lower than those in unsuccessful firms. We suggest, therefore, the following: Hypothesis 13: Workers in firms experiencing economic growth will report lower covering-up behaviour than workers in firms on a non-growth experience.

Method The sample consisted of 340 workers, including middle-level managers randomly selected from 118 Israeli organizations. The characteristics of the sample for each of the independent variables in the different categories show the following picture.

(a) Personal characteristics Fifty-seven per cent of the sample were men, and 43 per cent women. This division is representative of the share of the two groups in the Israeli labour market. The mean age of the participants was 33. The marital status of the sample showed that 34 per cent were single, 63 per cent married, and 3 per cent divorced. Respondents’ educational level was as follows: 13 per cent had a secondary or high school education, 16 per cent had some higher education, 49 per cent held a BA degree, 21 per cent held an MA degree, and 1 per cent held a PhD degree.

(b) Job/occupational characteristics Supervisors comprised 40 per cent of the respondents, and workers with no subordinates made up 60 per cent of the sample. The majority of the employees in the sample (91 per cent) held full-time jobs, whereas only 9 per cent held part-time jobs. Most of the respondents (84 per cent) had job tenure. The mean tenure with the current firms was 6.4 years.

Shmuel Stashevsky and Jacob Weisberg 67

(c) Organizational characteristics The firms included in our sample operated in different business sectors: the majority (57 per cent) of the respondents worked for firms providing services, 32 per cent worked for manufacturing firms, and 11 per cent worked for R & D organizations. Respondents who worked for firms that sell their products/services exclusively to the local market were 39 per cent; 31 per cent worked equally for the local market and for the export market; 14 per cent worked for firms that sell mainly to export markets; and only 4 per cent worked for firms that sell exclusively to export markets. The number of employees in the firms was as follows: 26 per cent of the sample worked for companies with up to 99 employees, 29 per cent worked for firms with 100–999 employees, 22 per cent worked for firms with 1,000–5,000 employees, and 23 per cent worked for firms with over 5,000 employees. Firms included in our sample were at different stages of standardizing the work procedures, and consequently were at different stages of obtaining ISO 9000 certification. The majority (59 per cent) of the respondents worked for firms that had the ISO 9000 certification. Thirty-one per cent of the respondents worked for companies that did not have ISO 9000 certification and were not in the process of acquiring it. Ten per cent of the sample worked for firms that were in the process of obtaining ISO 9000 certification, but had not yet achieved it. The majority (55 per cent) of the respondents worked for firms experiencing economic growth in the past year. However, 11 per cent of the respondents worked for companies with declining economic activities, whereas 34 per cent were firms that are in a stable state.

Instruments A structured questionnaire was used to tap the employees’ views concerning themselves and the organizations in which they work. The questionnaire included items relating to personal characteristics, occupational characteristics, organizational characteristics, and covering-up behaviour. Responses were ranked on a Likert scale from 1 – ‘no coveringup behaviour’, to 6 – ‘very high degree of covering-up behaviour’.

Variables Dependent variable: Covering-up was measured by the extent to which the participants believed their co-workers hid mistakes from their managers. Employing a great deal of caution, it was felt that indirectly we

68 Covering-Up Behaviour in Organizations

could learn about the respondents’ own covering-up behaviour from their perception of their co-workers’ conduct. We did not ask directly about the respondent’s own behaviour as his or her self-report would very likely be biased and reveal a low extent of unethical behaviour. Independent variables: The following independent variables were employed in the analysis to test our proposed theoretical model: (a) Personal characteristics: gender, age, marital status, and education. (b) Occupational characteristics: supervisory level, full/part-time employment, tenure status, and length of service. (c) Organizational characteristics: business sector, local/export market orientation, firm size, work standardization (ISO 9000 certification), and firm growth trend.

Data analysis At the first stage, descriptive statistics for the study variables were determined. At the second stage, the hypotheses were tested by independent sample T-tests. At the third stage, stepwise multivariate regression analysis was performed to identify the overall impact of personal, job/occupational and organizational factors on covering-up behaviour.

Results Covering-up behaviour frequencies In order to measure the covering-up behaviour in organizations we first analysed the frequencies of the reported covering-up behaviour. Eighteen per cent of the total sample (340 employees) reported no coveringup behaviour, 26 per cent reported very little covering-up, 25 per cent little covering-up, 24 per cent moderate covering-up, 5 per cent much covering-up, and 2 per cent a lot of covering-up.

Descriptive statistics Table 4.1 shows the means, standard deviations, and Spearman correlations for all the study variables. Spearman correlations were employed because the variables’ scales were ordinal.

Bivariate hypotheses testing For purposes of the bivariate and multivariate analyses, the variables were recoded into binary constructs (see Table 4.2), enabling us to use

Table 4.1

Means, standard deviation, and Spearman correlation coefficients (N = 340)

Variables

Mean

Spearman correlation coefficients (* = P < .05 two-tailed)

Std. dev. 1

Personal characteristics: 1. Gender 2. Age 3. Marital status 4. Education

2

3

4

5

6

7

1.43 1.33 1.66 1.71

.50 .47 .48 .46

−.17* −.08 −.04

Job/occupational characteristics: 5. Supervisory level 1.40 6. Full/part-time job 1.09 7. Tenure status 1.16 8. Length of service 1.41

.49 .29 .37 .49

−.15* .07 .16* .06 .24* −.14* −.16* −.02 −.17* .24* −.21* −.19* .00 −.16* .28* −.06 .14* .24* −.23* .31* −.12* −.30*

Organizational characteristics: 9. Business sector 1.68 10. Local/export orient. 1.61 11. Firm size 1.74 12. Work standardization 1.69 13. Firm growth trend 1.55

.47 .49 .44 .46 .50

.21* .04 −.04 −.25* .00

Dependent variable: 14. Covering-up

2.78

1.24

8

9

10

11

12

13

.09 .13* −.10

−.06 −.05 −.08 .02 −.02

−.17* .16* −.19* .11* .10 −.18* .01 .09 .03 −.01 .11* −.04 −.01 .02 .02 .02 .12* .16* .12* −.09 .14* −.09 −.14* .16* .10 −.02 −.01 −.12* .03 −.08

.18* −.16* −.06

−.04

−.03

.12*

.17* −.11*

−.26* −.17* −.36* −.01 .09

.28* .17* .11* .22* −.08

.08

−.13* −.13* −.14* −.14*

* = P < .05 two-tailed.

69

70 Covering-Up Behaviour in Organizations Table 4.2

Definitions of the independent variables

Independent variables Personal characteristics: Gender

Categories Male (1) = 57 per cent; Female (2) = 43 per cent.

Age

20–29 or 40 + years old (1) = 67 per cent; 30–39 years old (2) = 33 per cent.

Marital status

Single (1) = 34 per cent; Married/divorced (2) = 66 per cent.

Education

Non-academic education (1) = 29 per cent; Academic education (2) = 71 per cent.

Job/occupational characteristics: Supervisory level Workers with no subordinates (1) = 60 per cent; Supervisors (2) = 40 per cent. Full/part-time job

Full-time job (1) = 91 per cent; Part-time job (2) = 9 per cent.

Tenure status

Tenure (1) = 84 per cent; No tenure (2) = 16 per cent.

Length of service

0–4 years (1) = 59 per cent; 5 + years (2) = 41 per cent.

Organizational characteristics: Business sector Manufacturing (1) = 32 per cent; Non-manufacturing (2) = 68 per cent. Local/export market orientation

Local market only (1) = 39 per cent; Export + local (2) = 61 per cent.

Firm size – no. of employees

1–99 employees (1) = 26 per cent; 100 + employees (2) = 74 per cent.

Work standardization (ISO 9000 certification)

Neither certified nor in process (1) = 31 per cent; Certified or in the certification process (2) = 69 per cent.

Firm growth trend

Decline/stable (1) = 45 per cent; Growing (2) = 55 per cent.

Note:

Number in brackets indicates code for the statistical analysis.

T-tests for the bivariate analysis and regression for the multivariate analysis. Testing of the hypotheses (1–13) using bivariate test procedures produced the following results.

(a) Personal characteristics: Two of the four hypotheses relating to the relationship of personal characteristics and covering-up behaviour were confirmed. Hypothesis 1 suggesting that male employees would report a lower level of covering-up behaviour than their female counterparts was confirmed (t(335) = − 2.99, p < .01).

Shmuel Stashevsky and Jacob Weisberg 71 Mean covering-up level 3

2.95

2.91

2.9 2.8 2.7 2.6 2.49

2.5 2.4 2.3 2.2 20–29

30–39

40+

Age group Figure 4.2

Average covering-up behaviour by age groups

Note: The figures on the Y-axis indicate the reported average covering-up level (ranging from 1 to 6).

Hypothesis 2 suggesting that mid-aged employees would report lower covering-up behaviour than their younger or older colleagues was confirmed. The relationship between age and covering-up responses among the employees showed a U-shape (see Figure 4.2). The covering-up behaviour reported by employees aged 30–39 was lower than that reported by younger and older employees. Testing a one-way analysis of variance (ANOVA) for the covering-up behaviour by the three age groups showed significant differences among the means (F(335) = 4.88, p < .01). T-test analysis showed significantly higher covering-up behaviour for the younger 20–29 years versus the 30–39 years of age group (t(259) = 2.98, p < .01). In addition, the oldest age group of 40+ showed higher covering-up behaviour than the 30–39 age group (t(183) = 2.34, p < .05). This confirms the non-linear relationship. Hypothesis 3 suggesting that married employees would report less covering-up behaviour than their single co-workers was not confirmed (t(335) = .83, p > .05).

72 Covering-Up Behaviour in Organizations

Hypothesis 4 suggesting that employees with more years of schooling (employees with academic education) would report lower covering-up behaviour than co-workers with fewer years of schooling (employees with non-academic education) was not confirmed (t(335) = .91, p > .05).

(b) Job/occupational characteristics: Two of the four hypotheses relating to the effect of occupational characteristics on covering-up behaviour were confirmed. Hypothesis 5 suggesting that employees holding supervisory positions would report less covering-up behaviour than employees holding non-supervisory positions was not confirmed (t(334) = .74, p > .05). Hypothesis 6 suggesting that employees engaged in full-time work would report less covering-up behaviour than employees engaged in part-time work was not confirmed (t(335) = − 1.94, p > .05). Hypothesis 7 suggesting that employees with tenure would report less covering-up behaviour than employees with no tenure was confirmed (t(335) = − 3.00, p < .01). Finally, Hypothesis 8, suggesting that employees with more years at their current firm (at least 5 years) would report less covering-up behaviour than workers employed for shorter periods of time with their current firm (less than 5 years) was confirmed (t(335) = 1.97, p < .05).

(c) Organizational characteristics: Only one of the five hypotheses relating to the effect of organizational factors on covering-up behaviour was not confirmed. Hypothesis 9 suggesting that workers in manufacturing firms would report less covering-up behaviour than workers in service and non-commercial companies and organizations was not confirmed (t(335) = − 1.64, nonsignificant). The following four hypotheses were confirmed. Hypothesis 10 suggesting that workers in firms selling to export markets would report less covering-up behaviour than workers in firms and organizations which sell only to the local market was confirmed (t(332) = 2.53, p < .05). Hypothesis 11, suggesting that workers in firms employing a large number of employees (at least 100) would report a lower level of coveringup behaviour than workers in firms with a small number of employees (less than 100), was confirmed (t(334) = 2.37, p < .05). Hypothesis 12, suggesting that workers in firms with structured work standards (ISO 9000 certified or in the certification process) would report less covering-up behaviour than workers in firms with no structured work standards was confirmed (t(321) = 2.36, p < .05).

Shmuel Stashevsky and Jacob Weisberg 73

Hypothesis 13 suggesting that workers in firms experiencing economic growth would report lower covering-up behaviour than workers in firms that are on a decline or in a state of stability was confirmed (t(333) = 2.20, p < .05).

Multivariate hypotheses testing In order to identify the variables that concurrently affect covering-up behaviour, a multivariate regression model was conducted, according to the research model in Figure 4.1. The results of this analysis appear in Table 4.3. In the first step, the personal characteristics were entered in the analysis, and showed the following results: age (β = −.16, p < .01), and gender (β = .13, p < .05) were found to be significant. In the second step, all occupational characteristics were entered but only the employee’s tenure status (β = .15, p < .01) was found to be significant. In the third step, organizational characteristics were entered and showed that two variables are significant: local/export market orientation (β = −.15, p < .01), and firm growth trend (β = −.11, p < .05) The variables that did not show significance were then excluded, producing thereby a parsimonious model displaying the final results. The significant independent variables that affect the covering-up behaviour are:

Table 4.3 Stepwise linear regression in blocks (with covering-up behaviour as the dependent variable) Independent variable

B

Beta

t

Sig.

7.58

<.001

∆R2

Constant

3.34

Personal characteristics: 1. Age 2. Gender

−.42 .33

−.16 .13

−2.97 2.46

.003 .015

.039 .024

.51

.15

2.69

.008

.015

−.38 −.28

−.15 −.11

−2.75 −2.10

.006 .036

.031 .012

Job/occupational characteristics: 3. Tenure status Organizational characteristics: 4. Local/export market orientation 5. Firm growth trend

The model was found to be significant (F = 8.65; Sig. P < .001). Note: The following variables were excluded from the parsimonious model: marital status, education, supervisory level, full/part time, length of service, business sector, firm size, and work standardization (ISO 9000 certification).

74 Covering-Up Behaviour in Organizations

• Age: Younger (aged 20–29) and older (40+) employees showed a • • • •

higher level of reported covering-up behaviour than the middle-aged (30–39) employees. Gender: Women displayed a higher level of reported covering-up behaviour than men. Tenure status: Employees with no tenure showed a higher level of reported covering-up than employees with tenure. Local/export market orientation: Workers in firms whose products are sold only in local markets showed a higher level of reported covering-up behaviour than workers in firms involved in export. Firm growth trend: Workers in firms that did not grow in the last year reported more covering-up behaviour than those in firms experiencing economic growth.

Summary of the hypotheses testing To sumup the results, 8 of the 13 bivariate hypotheses were confirmed and the final parsimonious model showed the multidimensional impact of 5 factors on covering-up behaviour. These results are summarized in Table 4.4. The fact that 3 out of the 8 hypotheses were not confirmed in the multivariate analysis but were confirmed in the T-test analyses is probably due to either weak (although significant) relationships or to co-linearity.

Discussion and conclusion The results of this study clearly demonstrate that employees’ covering-up behaviour is multi-dimensional and is influenced by a complex web of factors. Our interpretation of these findings is based on the labour market principle that workers will act to reduce or avoid ‘pain’ or risk to themselves. First, when controlling for occupational and organizational influences, employees of different demographic characteristics exhibit different degrees of covering-up behaviour. This study found that age and gender have a significant net impact on workers’ covering-up behaviour. Younger (20–29) and older (40+) employees reported more covering-up in their organizations than the mid-aged group (30–40). We would suggest that as covering-up behaviour aims to reduce the likelihood of losing the job, the unequal labour market conditions that are reflected in the probability of unemployment or in prospects of obtaining another job, might account for these results. The unemployment rate

Shmuel Stashevsky and Jacob Weisberg 75 Table 4.4 Bivariate and multivariate hypotheses testing: independent variables with covering-up behaviour Hypotheses Personal characteristics: H1 Males vs. females H2 Middle age vs. younger or older H3 Married vs. single H4 High-educated vs. low-educated

Bivariate analysis

Multivariate analysis

Confirmed (females – higher covering-up) Confirmed (younger or older – higher covering-up) Not confirmed

Confirmed (females – higher covering-up) Confirmed (younger or older – higher covering-up) Not confirmed

Not confirmed

Not confirmed

Job/occupational characteristics: H5 Supervisors vs. Not confirmed non-supervisors H6 Full-time vs. Not confirmed part-time job H7 Tenure vs. no Confirmed (no tenure – tenure higher covering-up) H8 Long service vs. Confirmed (short service – short service higher covering-up) Organizational characteristics: H9 Manufacturing Not confirmed vs. non-manufacturing firms H10 Local market vs. Confirmed (local market – export firms higher covering-up) H11 Big vs. small Confirmed (small firms firms – higher covering-up) H12 Structured vs. Confirmed (non-structured non-structured work standards – higher work standards covering-up) firms H13 Growing vs. Confirmed (non-growing non-growing firms – higher covering-up) firms

Not confirmed Not confirmed Confirmed (no tenure – higher covering-up) Not confirmed

Not confirmed

Confirmed (local market – higher covering-up) Not confirmed Not confirmed

Confirmed (non-growing firms – higher covering-up)

for the middle group is much lower than that of either younger workers, who have not yet built up solid ‘human capital’, or older workers, whose ‘human capital’ is more likely to be less updated with the accelerated use of sophisticated technology and computers. Thus, the mid-aged workers are at much less risk in the labour market, and they are more likely to report less covering-up behaviour in the firm. Similarly, women

76 Covering-Up Behaviour in Organizations

may report more covering-up in the organization since they feel more vulnerable in the labour market than their male counterparts. In regard to occupational characteristics, we found that workers with no tenure report more covering-up behaviour than workers with tenure. This might be due to the fact that employees with no tenure have less job security and might therefore feel lower levels of organizational commitment. The tenure factor showed a significant effect on covering-up behaviour both in the bivariate and the multivariate analyses. However, the length of service showed a significant difference only in the bivariate analysis. This is reasonable because tenured workers are usually full-time workers with a long service period with the firm (as can be clearly seen also from the corresponding correlation coefficients, which are among the highest in Table 4.1). In regard to organizational characteristics, it was found that covering-up behaviour is more prevalent among workers who are employed in organizations that are more local market oriented than among workers employed in organizations involved in export. This might stem from lower levels of strict quality control on the firms’ products. The second organizational characteristic that was found to be significant is the economic growth trend of the firm. This indicates that firms in a declining or stable state of economic growth are more likely to show higher levels of covering-up than successful companies experiencing economic growth. We would like to suggest that employees in declining or stable organizations are more vulnerable to dismissal, and may thus have a greater tendency to hide their mistakes from their superiors. Two additional organizational characteristics were found to be significant in the bivariate analysis but not in the multivariate analysis: firm size and work standardization. Workers in small firms and workers in firms with no structured work standardization are more likely to report higher levels of covering-up behaviour than workers in big firms and workers in firms with structured work standards. These two characteristics were not found to be significant in the multivariate analysis because exporting firms are usually big firms with structured work standardization (as can clearly be seen from the corresponding correlation coefficients in Table 4.1). Covering-up behaviour might also be affected by cultural factors. This study was conducted in Israel, a dynamically developing country with a mix of immigrants coming from different cultures and with strong familial ties. In more conservative societies, familial ties might be less influential, and workers may be more willing to conform to formal regulations, obey managerial decisions, and follow instructions. It would

Shmuel Stashevsky and Jacob Weisberg 77

thus be of interest to replicate the current study in other cultures to see whether similar results are obtained. Workers’ covering-up behaviour was examined here by asking people to report the extent to which they perceive their colleagues are hiding mistakes from superiors. It was felt that this would be an indication of the respondent’s own behaviour. Future studies might do well to include direct questions relating to the respondent’s behaviour in an attempt to construct a comprehensive measure of covering-up behaviour and allow for comparison between the techniques. The current study opens a wide range of opportunities for future research. Other variables, not included in our model, might contribute to a deeper understanding of the phenomenon under investigation. Such variables might include work attitudes, e.g. job satisfaction and organizational commitment, as well as personal traits. It would also be of interest to examine the relationship between covering-up behaviour and the workers’ organizational outcomes, such as performance and work quality levels. Identification of the factors encouraging covering-up behaviour and, as a result, the covering-up organizational atmosphere, may be of clear benefit to a company. However, from our point of view, the major implication of this research is in helping workers to reduce their fear of making mistakes, and thus to decrease the need to cover them up. As Ryan and Oestreich (1991, pp. 3–4) stated: ‘The fear of speaking up can be thought of as a composite of many types of workplace anxieties, which together form a most basic human barrier to improving an organization.’ Allowing workers, for example, to collaborate more with management and participate in decision-making might result in employees being more loyal and committed to the firm and thereby reluctant to coverup wrongdoings. Generally speaking, and stated quite simply, an open relationship between workers and managers will probably decrease covering-up behaviour.

References Beu, D. and Buckley R. M. (2001) ‘The hypothesized relationship between accountability and ethical behavior’, Journal of Business Ethics, 34(1), 57–73. Bounds, G., Yorks, L., Adams, L. and Ranney, G. (1994) Beyond Total Quality Management Toward the Emerging Paradigm. New York: McGraw-Hill. Dalton, D. R. and Kesner, I. F. (1988) ‘On the dynamics of corporate size and illegal activity: an empirical assessment’, Journal of Business Ethics, 7(11), 861–70. Dawson, L. M. (1997) ‘Ethical differences between men and women in the sales profession’, Journal of Business Ethics, 16(11), 1143–52.

78 Covering-Up Behaviour in Organizations Deming, W. E. (1986) Out of the Crisis. Cambridge, Mass: MIT Press. Dubinsky, A. J. and Ingram, T. N. (1984) ‘Correlates of salespeople’s ethical conflict: an exploratory investigation’, Journal of Business Ethics, 3(4), 343–53. Eisenhardt, K. (1989) The Moral Dimension: Towards a New Economics. New York: The Free Press. Elangovan, A. R. and Shapiro, D. L. (1998) ‘Betrayal of trust in organizations’, The Academy of Management Review, 23(3), 547–66. Grant, E. W. and Broom, L. S. (1988) ‘Attitudes toward ethics: a view of the college student’, Journal of Business Ethics, 7(8), 17–19. Grover, S. L. and Hui, C. (1994) ‘The influence of role conflict and self-interest on lying in organizations’, Journal of Business Ethics, 13(4), 295–305. Harris, J. R., (1990) ‘Ethical values of individuals at different levels in the organizational hierarchy of a single firm’, Journal of Business Ethics, 9(9), 741–50. Ivancevich, J. M., DeFrank, R. S. and Gregory, P. R. (1992) ‘The Soviet enterprise director: an important resource before’, The Academy of Management Executive, 6(1), 42–55. King, G. III and Hermodson, A. (2000) ‘Peer reporting of coworker wrongdoing: a qualitative analysis of observer attitudes in the decision to report versus not report unethical behavior’, Journal of Applied Communication Research, 28(4), 309–29. Lucas, J. R. (1998) ‘Anatomy of a vision statement’, Management Review, 87(2), 22–6. McDonald, G. M. and Kan, P. C. (1997) ‘Ethical perceptions of expatriate and local managers in Hong Kong’, Journal of Business Ethics, 16(15), 1605–23. McShulskis, E. (1997) ‘Job stress can prompt unethical behavior’, HRMagazine, 42(7), 22–4. Newstrom, J. W. and Ruch, W. A. (1975) ‘The ethics of management and the management of ethics’, MSU Business Topics, 23(1), 29–37. Puffer, S. M. and McCarthy, D. J. (1995) ‘Finding the common ground in Russian and American business ethics’, California Management Review, 37(2), 29–47. Randall, D. M. and Gibson, A. M. (1990) ‘Methodology in business ethics research: a review and critical assessment’, Journal of Business Ethics, 9, 457–71. Ryan, K. D. and Oestreich, D. K. (1991) Driving Fear out of the Workplace. San Francisco: Jossey-Bass Publishers. Serwinek, P. J. (1992) ‘Demographic and related differences in ethical views among small businesses’, Journal of Business Ethics, 11(7), 555–66. Trevino, L. K. and Youngblood, S. A. (1990) ‘Bad apples in bad barrels: a causal analysis of ethical decision making behavior’, Journal of Applied Psychology, 75, 378–85. Zabid, A. R. M. and Alsagoff, S. K. (1993) ‘Perceived ethical values of Malaysian managers’, Journal of Business Ethics, 12(4), 331–8.

5 Understanding Social Loafing Yair Amichai-Hamburger

Introduction The effectiveness of work in groups has been investigated over several decades. One of the main areas of investigation has been to examine why groups almost inevitably fail to fulfil their potential. One of the first to study this area empirically was Steiner (1972). He argued that when a group of individuals is assembled together to solve a problem or to perform a task, the group activity will produce both losses and benefits; however, the overall result will be negative, namely, that the potential achievement of the group will remain unfulfilled. The literature dealing with group productivity indicators suggests that when people work on a collective task (in which individuals have to pool their individual efforts to reach the desired outcome), they tend to invest less effort as compared to when they perform a coactive task (in which individual outcomes depend solely on his/her own efforts). This decline in the degree of effort invested by the individual when he/ she performs a task in a group is called social loafing (SL) (Latane, Willams and Harkins, 1979). A concept related to that of SL in the economic sphere is that of freeriding (Albanese and Van Fleet, 1985). This refers to individuals who benefit from their membership in a group or a community. Their own contribution to the work is minimal and they do not fulfil their potential, yet they reap the benefits from the work of the other group members (Albanese and Van Fleet, 1985; Weldon and Mustari, 1988). SL is very hard to detect; many, if not most, managers would want to deny or at least minimize its existence in their own establishment. At the same time, however, they would be keen to give the names of any number of neighbouring organizations where its presence is all pervasive! 79

80 Understanding Social Loafing

Why is it that managers fail to notice SL in their own organizations? The main reason is because managers do not usually have the criteria with which to measure the maximum capability of outputs with the actual levels of production. Organizational norms of production are usually satisfactory, but do not necessarily reflect the real ability of the workers. These are only experimental studies that compare levels of group productivity of different sized groups to individual productivity levels. These comparisons show clearly that the SL phenomenon exists. Were managers to carry out careful observations of their workplace, they would see that SL does exist and having identified the problem, may take steps to eliminate it. This chapter aims to provide a greater understanding of the SL phenomenon by assessing its scope, proposing explanations for its existence, building an inclusive list of the variables found to relate to SL, advising how to prevent it, and suggesting ideas for future research.

SL background The SL phenomenon was first demonstrated by Ringelmann (cited in Kravitz and Martin, 1986), who compared the degree of physical effort invested by people jointly engaged in a task of pulling a rope. He found that a pattern emerged, namely, that as the group grew larger, the personal effort of each individual declined. More specifically, Ringelmann found that a person pulling a rope alone exerted on average 65 kg of force. In groups of three, each person exerted on average only 53 kg of force. In groups of eight, the average power exerted by individual group members was on average only 31 kg, less than half of that exerted by an individual performing the same task on his/her own (for an illustration of this phenomenon, see Figure 5.1). Steiner (1972) gave two explanations for the decline in performance: (1) reduced individual motivation and (2) coordination loss. To discover whether the whole phenomenon is really a problem of group coordination or an issue of motivation, Latane et al. (1979) studied the impact of SL with groups of different sizes. They created real groups and pseudo groups, in which participants were told that they were working in groups, although they did not see or hear the others (and, therefore, there was no problem of coordination). In their experiment the group was instructed to shout as loudly as possible or to clap their hands as hard as they could, while the experimenter recorded the amount of noise that was produced. Latane et al. found that as the size of the group increased, the amount of noise produced by an individual group member was reduced. For example, a group of six

Yair Amichai-Hamburger 81

Individual performance

High

Low One person working alone

Small group

Large group

Figure 5.1 Illustration of the impact of group size on group performance

performed only up to 40 per cent of their potential. The SL phenomenon was also found in pseudo groups. These studies showed that the explanation of SL is lack of motivation rather than lack of coordination.

The scope of SL The SL phenomenon has been recorded in many situations involving the joint performance of a physical task, in addition to the example of the rope-pulling (Ringelmann cited in Kravitz and Martin, 1986) and shouting or hand-clapping (Latane et al., 1979). SL has also been recorded during such activities as swimming (Miles and Greenberg, 1993; Williams et al., 1989) and pumping air (Kerr and Bruun, 1981). SL also appears in cognitive tasks: for example, solving mazes (Jackson and Williams, 1985); brainstorming (Harkins and Petty, 1982); electronic brainstorming (Roy, Gauvin and Limayem, 1996); evaluating essays and proposals (Brickner, Harkins and Ostrom, 1986; Petty et al., 1977); songwriting (Jackson and Padgett, 1982). SL was found in data collected in field studies as well as in laboratory studies and among individuals from collectivist cultures and individualistic cultures (Karau and Williams,

82 Understanding Social Loafing

1993). This phenomenon was found to be present regardless of race or age (Karau and Williams, 1993). This may be why Latane et al. (1979) suggested that SL is a disease that has a negative impact on individuals, groups, organizations, and societies.

Teams and SL Many modern organizations try to encourage worker motivation by team building. A team is a group whose members have complementary skills. They are committed to a common purpose or set of performance goals for which they hold themselves mutually accountable (Katzenbach and Smith, 1993). There are several main differences between a group and a team: (1) For groups, performance depends on the work of individual group members; in a team, performance depends both on the individual contribution and on the concerted effort of the team members. (2) In a group, each member is responsible for his/her own performance; in a team, there is mutual accountability. (3) In a group, members may share a common interest in achieving certain goals; in a team, members share a common commitment to a purpose. (4) A group has a manager, whereas a team is by definition a self-managing entity (Harari, 1995). The motivational platform of teams is built on the understanding of the high human motivational needs based on Maslow’s need hierarchy (Maslow, Stephens and Heil, 1998), Locke and Latham’s (1990) goal-setting theory, the job enlargement and job enrichment models (Campion and McClelland, 1993; Luthans and Reif, 1974), and the job characteristics models (Hackman and Oldham, 1976). It would appear, therefore, that an organization wishing to combat SL should change its organizational methods from groups to teams. In teams, members are highly motivated to perform at their best and so a sharp decrease in SL may be expected. However, when we examine the literature on implementing teams in organizations, the empirical picture is not as straightforward as is suggested by team advocators. Mulvey, Veiga and Elsass (1996) examined why, in many cases, participants in team decision-making do not fulfil their potential. They suggest that SL is likely to be one of the main reasons for this. Another disturbing finding is that SL occurs even in very cohesive teams like swimmers (Miles and Greenberg, 1993; Williams et al., 1989). There is empirical evidence showing that: (1) SL is a relevant factor in explaining the reason why teams under-perform; and (2) the elimination of SL is not a simple undertaking, even in cases where a modern organizational management has tried actively to prevent it.

Yair Amichai-Hamburger 83

The findings that even teams are likely to suffer from SL demonstrate powerfully that SL penetrates even modern group structures that were designed especially to elevate motivation to its highest possible level. It is, therefore, a significant challenge to achieve a better understanding of the factors leading to SL and what can be done to minimize them.

Main psychological explanations for SL There are three leading explanations for lack of motivation, which leads to the SL phenomenon: (1) search for meaning; (2) social evaluation; and (3) the collective effort model (CEM).

Search for meaning Motivational theories, such as those of Maslow (1943, 1971) and Alderfer (1972), emphasize that people have a strong need for self-actualization, a feeling that they are fulfilling their potential. Work is one of the primary settings where people may or may not fulfil their potential. When they feel, for example, that they have no impact on the final result, or that they are a small insignificant cog in a very large machine, their personal motivation is likely to fall significantly. The importance attached to workers’ higher needs was further demonstrated by the job characteristics theory. This theory advocates the importance of designing tasks in a way that enhances workers’ motivation (Hackman, 1992; Hackman and Oldham, 1976). Research has shown that people tend to invest less effort when they work in a group because they feel that their personal contribution is irrelevant to the achievement of the final goal. This applies equally when people cannot see a direct or indirect link between their efforts at work and aims and vision of the organization (Kerr, 1983; Kerr and Bruun, 1981). It was also found that in cases when the task had a minimal performance threshold, once this had been reached, there was a steep decline in the degree of effort invested in by the workers. From the workers’ perspective, there was no link between any further effort they may put in and the general success of the organization. They therefore concluded that no extra exertion was necessary. In short, SL implies a decrease in motivation in situations where the group (or team) members believe that any extra effort on their part is neither necessary nor meaningful.

84 Understanding Social Loafing

Social evaluation According to the social evaluation explanation, it is the process of evaluating one’s behaviour, norms, or responsibilities, using some type of social criterion (e.g. comparison to others) that forms the basis of the SL phenomenon. According to the literature, the social evaluation process may follow one of three forms: (1) diffusion of responsibility; (2) fairness judgments; (3) conformity to norms. Diffusion of responsibility: Latane (1981) suggested that a group supervisor serves as a source of social impact, while group members serve as co-targets for that impact. As the group becomes larger the number of co-targets increases, and the pressure on each member diminishes, as does the individual responsibility towards the collective to complete the task. Thus, the number of group members is used as a criterion for assessing one’s responsibility toward the group. This results in a diffusion of responsibility and a decline in individual effort. Fairness judgments: Adams (1965), and Huseman, Hatfield and Miles (1987) suggested that individuals judge the fairness of their own results by comparing the ratio of their outcome and inputs with those of other group members. If the individual feels hard done by, he/she is motivated to restore equity – for example, if he or she resents the fact that other group members are practising SL yet all will be rewarded equally. This perceived injustice is likely to create negative feelings including envy towards the others in the group. Duffy and Shaw (2000) demonstrated that envy of other group members was directly related to a decline in group performance. It seems that envy creates strong negative feelings towards the other group members that are expressed by the practice of SL. The greater the envious comparison, the more detrimental it is to group performance. Tesser and Campbell (1990) suggested that the closer an individual is emotionally to another member of the team, the stronger the feelings of envy are likely to be. Conformity to norms: Cialdini (1993) argued that people in a group (or a team) have a tendency to conform to social norms. Group cohesiveness is highly related to conformity; in a cohesive group, people have strong pressures to conform (Latane and L’Herrou, 1996). One of the results of this conformity is that all members adopt the group’s productivity norms. When the productivity norms are high, people are likely to work hard to achieve them, and when they are low, people will reduce their effort in order to conform with the low norms. SL, according to this explanation, can be a result of group members conforming to what they perceive as low group norms.

Yair Amichai-Hamburger 85

By integrating the conformity to norms and fairness judgment explanations, it could be said that in order to predict the behaviour of an individual, two related factors must be taken into account: (1) the level of cohesiveness of the relevant group and (2) the group norms. When the group is cohesive, individuals are less likely to compare themselves with others and to practise fairness judgments. When the group is cohesive the relevant factor will be the group norms; they may encourage high performance or, if they are rather low, SL would be fostered. Conversely, when the group has loose ties, individual behaviour will be determined by the results of this comparison and adjusted accordingly.

The collective effort model (CEM) Karau and Williams (1993) suggested the CEM as a comprehensive explanation of SL. This model is based on the expectancy theory of Vroom (1964) which argues that the level of individual task motivation is dependent on three basic components: expectancy, instrumentality, and valence. Expectancy is the belief that if one puts in a significant effort, this will be reflected in the performance. Instrumentality is the belief that he/she will be recognized and rewarded according to the performance. Valence is the perception that the rewards received for good performance are valuable and desirable. In other words, individuals work harder when they believe that their contributions are instrumental in bringing about a desired outcome (i.e. being recognized and rewarded). The CEM is a motivational theory in a collective context. It suggests that in a group, individual members need to perceive that their effort will be reflected in the outcome; in addition, that their individual performance will have a significant impact on the group result. The total group performance should be related to group outcomes reward. These outcomes should be valued on the individual level and on the group level. It is therefore obvious that individual motivation in a collective task is much more complex than for a coactive task. In a group context, the individual will exert effort only to the level he/she expects the efforts to be instrumental in achieving valued outcomes. The individual knows that for achieving the task, the outcomes have to be valued by all group members. The fact that he/she as an individual group member may value these outcomes will not be enough. According to the CEM, when working on an individual task as compared with a group task, there is a stronger perceived contingency between the effort and the valued outcomes. This explains why loafing is more pervasive in-group activity as compared with individual activity. According to the CEM, key factors influencing levels of SL are the

86 Understanding Social Loafing

perceived redundancy of individual contributions (Harkins and Petty, 1982) and dispensability of effort (Kerr, 1983). The second factor implies that one who perceives his/her personal contribution to the group as dispensable is more likely to loaf than one who views his/her contribution as indispensable. The CEM may lead to a greater understanding of the compensation effect (Karau and Williams, 1997). Contrary to the belief that people measure their efforts against those of their co-workers and strive to equal them, the compensation effect maintains that people sometimes behave in a non-reciprocal manner and will actually work harder in order to compensate for the poor performance of others. According to the CEM, this will happen when the individual has high motivation to complete the task and believes that his/her effort will help to achieve this goal. The CEM provides a useful understanding as to which outcomes are important to the group as a whole. These may be objective outcomes like payment or subjective outcomes, the value of which is dependent on the perception of the individual – for example: satisfaction, belonging, selfworth, and feeling part of the group spirit (Karau and Williams, 1993). CEM seems to provide the most comprehensive explanation for SL. It may also serve as an ‘umbrella’ theory, incorporating within it the other main explanations for the SL phenomenon. The ‘search for meaning’ explanation may be perceived as part of the valence component in the CEM. However, the search for meaning explanation focuses on intrinsic motivators influencing the individual; in terms of the CEM, this is only part of the understanding of the valence of outcomes for the individual in a group task. The social evaluation explanation may be perceived as part of the expectancy component in the CEM, where the individual assesses him/herself in comparison with the other group members and decides how he/she should act in the group task. While the social evaluation explanation provides a satisfactory understanding as to why a person loafs, it does not explain why, in many cases, individual members of a group with a low performance record do not loaf, but rather perform to their maximum capability. According to the CEM, this may be as a result of the high valence achieved by completing the task (Karau and Williams, 1997).

SL correlates Research carried out over the years has found that many variables are related to the SL phenomenon. Initially research concentrated on the

Yair Amichai-Hamburger 87

Individual characteristics

Situational characteristics

Group characteristics

Cultural and societal norms

Gender

Challenging vs. non-challenging tasks

Group size

Individualistic vs. collectivistic culture

Being unique and better than others

Intellective vs. judgmental tasks

Gender composition

Need for cognition

Task reinforcement and punishment

Group cohesiveness

Approval orientation

Fatigue

Level of communication within the group Intragroup competition Perceived degree of effort invested by fellow members

Social Loafing

Figure 5.2 Range of variables found to relate to the SL phenomenon

variable of group size, but more recent investigations have concentrated on the level of individual characteristics, including personality characteristics, for example the need for cognition as it relates to the SL phenomenon. The main objective of this section is to put these many variables into a framework by dividing the main variables into four different categories: individual, group, task, and cultural characteristics (see Figure 5.2). The individual characteristics category includes: (1) gender; (2) being unique and better than others; (3) need for cognition; (4) approval orientation. The situational characteristics category includes: (1) challenging vs. non-challenging tasks; (2) intellective vs. judgmental tasks; (3) task reinforcement and punishment; and (4) fatigue. The group characteristics category includes: (1) group size; (2) gender composition; (3) group cohesiveness; (4) level of communication within the group; (5) intragroup competition; and (6) perceived degree of effort invested by fellow members. The cultural and societal norms category includes individualistic vs. collectivistic cultures.

88 Understanding Social Loafing

Individual level Gender Although SL is exhibted across both genders, women tend to display less SL than men (Karau and Williams, 1993). In addition, it is suggested that woman reduce their SL more significantly than man when they respect the group highly (Karau and Williams). Kugihara (1999) tested gender differences in SL in Japan. This was examined on performance in a rope-pulling task. The participants had 12 trials: 2 individual and 10 group trials. He found that women engage in SL less than men. In addition, he found that while moving from individual trials to group trials, men’s effort declined suddenly while for women it was more gradual. He explained that this is because women are oriented towards group coordination and human relationships, whereas men are more task achievement oriented. The differences found in the levels of group performance may be the result of the varying male and female expectations as explained by the expectation factor in the CEM.

Being unique and better than others Charbonnier et al., (1998) demonstrated that individuals who are motivated to see themselves as unique and better than others exhibited more SL in a group-learning task. The reason for this seems to be that those individuals perceive themselves as being unable to express their special values in the group. Huguet, Charbonnier and Monteil (1999) found that participants who felt superior to others invested less effort when working on an easy task collectively than working coactively. However, when the task was challenging, such individuals worked harder collectively than coactively. In comparison, people who perceived themselves as average invested the same degree of effort in both types of tasks. According to the CEM, the extent of the effort that people exert in a group task is commensurate with the degree to which they perceive that their contribution is instrumental in achieving outcomes that they value. When an individual member perceives him/herself as unique and superior to other members, he/she is also likely to believe that his/her own input into the group task will not to be instrumental in achieving its successful outcome. In this type of situation, such an individual is likely to loaf.

Need for cognition Cacioppo and Petty (1982) showed that individual differences with regard to the need for cognition influence the tendency to take part and

Yair Amichai-Hamburger 89

enjoy challenging, cognitive tasks. According to their view, people are ‘high on the need for cognition’ when they are more intrinsically motivated and less affected by situational factors. In comparison, people with a ‘low need for cognition’ are more situationally motivated. Smith et al., (2001) demonstrated that individuals with a high need for cognition put the same degree of effort into collective tasks as they do into coactive tasks, while individuals with a low need for cognition put less effort in collective tasks as compared with coactive tasks. According to the CEM, when a group task is challenging, individuals with a high need for cognition who have experienced success in such tasks will expect their efforts to be positively related to performance and instrumental in achieving valued outcomes. Similarly, according to the search for meaning explanation, people with a high need for cognition enjoy the challenge of dealing with cognitively demanding tasks and feel that they fulfil their need by doing so. Therefore, those people do not loaf when performing a challenging task. However, people with a low need for cognition do not enjoy challenging tasks and therefore perform SL.

Approval orientation Sorrentino and Sheppard (1978) distinguished between those people who are approval oriented (who seek approval and do not fear rejection) and those who are rejection threatened. They found that swimmers who were approval oriented swam faster in a team than as individual competitors. In comparison, rejection-threatened swimmers swam more slowly in a team than as individual competitors. According to the CEM, personality differences create a different set of expectations that come into action in the group swimming task. While the group swimming task affects positively the motivation of the approval-oriented swimmer, the same psychological pressure is too much to handle for the rejection-threatened swimmer.

Situational characteristics Challenging vs. non-challenging tasks Jackson and Williams (1985) suggested that for simple tasks, collective performance is lower than it would have been had the same task been carried out by the same number of people performing the task individually. When the task was more complicated, however, collective work enhanced productivity. Wageman (1999) found that individuals working in a group tend to put more effort into challenging tasks and have a tendency to loaf when the task is less challenging.

90 Understanding Social Loafing

According to the search for meaning explanation, the complicated task creates a challenge, which in turn arouses high motivation, leading to the investment of a high degree of effort into the task. Thus, despite its collective nature, a complex task is meaningful and SL will be minimal.

Intellective vs. judgmental tasks Henningsen, Cruz and Miller (2000) suggested that SL is more likely to occur in groups that perform a task which is perceived by group members as judgmental (a problem that does not have an objective solution) in comparison with a task which is perceived as intellective (a problem for which there is an objective solution). It would appear that in a judgmental task, there is no objective criterion as to when the task has been successfully completed and this leads to a greater tendency by participants, early on in the proceedings, to decide that it is over. As a result, people put less effort into this type of task. The CEM suggests that when group members perceive a problem as having an objective answer, the link between expectancy, instrumentality, and valence is clearer and stronger and therefore they are less likely to loaf. Conversely, when the challenge does not have an objective answer, the link between expectancy, instrumentality, and valence is vague and therefore people will tend to loaf.

Task reinforcement and punishment Zaccaro (1984) found that when a group task was achieved by students who were promised extra course credits for good performance, SL was nonexistent. Similarly, Miles and Greenberg (1993) found that when managers gave individuals rewards for their specific contributions to the group performance, SL was reduced. The rewards enhance the interest of workers in the mutual task by giving them the feeling that their individual contribution is significant, appreciated, and rewarded. This has been acknowledged in several areas of modern life, for example in the awarding of prizes, known as the ‘person of the match’ award for an individual who contributed significantly to a sporting event that was ostensibly between two teams. An additional tool that impacts on the degree of SL is the supervisor’s ability to punish the loafer. SL is reduced by the threat of punishment for workers who are not giving their best performance. Using a group of swimmers, Miles and Greenberg demonstrated that when participants were faced with the threat of punishment, the degree of SL was negligible. In a more careful examination of the impact of reinforcement and punishment, George (1995) found that a contingent reward (where positive reinforcers are awarded to

Yair Amichai-Hamburger 91

subordinates dependent on their performance level) was negatively related to SL, whereas non-contingent rewards (where positive reinforcers are given independent of performance levels) were shown to have no appreciable effect on SL. Conversely, supervisor contingent punishment (where punishments are given to subordinates because of poor performance) was not significantly associated with SL, whereas supervisor noncontingent punishment (where punishments are given to subordinates independently of poor performance) was positively associated with SL (George, 1995). The impact of punishment and reward plays an essential role in determining the factors that affect the valence of outcomes in the CEM. When rewards are contingent even if a group performs the work, the group member will perceive him/herself as identifiable and so the chances of SL are lessened considrably. When punishment is contingent, a conflict is created. The individual is encouraged to work harder so as to avoid punishment, but at the same time, having received punishment, he/she may be less motivated to work for the authority that recently chastised him. When the receipt of a reward is not conditioned by the individual contribution, non-contingent reward has no impact on SL. Conversely, non-contingent punishment creates motivation to take revenge against the punitive authority and so encourages SL as an effective weapon.

Fatigue SL was found to be related to lack of sleep and fatigue. As the individual is more tired and in need of sleep, his/her tendency to practise SL will increase (Hoeksema-van Orden, Gaillard and Buunk, 1998). According to the CEM, fatigue creates a situation in which the individual feels that the link between effort, performance, and desired outcomes is weaker as compared to when he/she is not tired. The search for meaning explanation points out that when people are very tired, they are more concerned with their basic needs than with their work, which in turn leads to loafing. The social comparison explanation argues that when someone is very tired, he/she is more likely to diffuse responsibility.

Group characteristics Group size Group size is one of the main variables affecting SL. As the number of group members increases, so does SL: namely, the larger the group, the lower the level of group productivity (Latane et al., 1979).

92 Understanding Social Loafing

The social evaluation explanation suggests that as group size increases, so does diffusion of responsibility. According to the CEM, individuals work harder when they believe that their contribution is instrumental in bringing about a desired outcome. However, as the size of the group increases, individuals may well perceive a weakening of this link.

Gender composition The gender composition of groups may also be an important moderating factor, although this has yet to be proved conclusively. Anderson and Blanchard (1982) did not find any difference as to the degree of SL performed by male and female participants in same-sex or mixed-sex groups. Kerr and MacCoun (1984) found higher task motivation by both male and female participants in mixed-sex groups than when they participated in same-sex groups or as individuals. They attributed this effect to special evaluation concerns arising in mixed-sex interactions (Kerr and MacCoun, 1984; Kerr and Sullaway, 1983). The CEM suggests that in a mixed group, both men and women are likely to forge a stronger link between expectancy, instrumentality, and valence because of their wish to impress members of the opposite sex. Hence, SL is likely to be lower.

Group cohesiveness Group cohesiveness may be perceived as the degree of attraction felt by individuals toward their group (or team). Karau and Hart (1998) found that members of low cohesive groups engaged in SL, whereas members of high cohesiveness groups worked just as hard collectively as coactively. The authors explained that individuals in a highly cohesive group alter their behaviour in order to gain acceptance and meet the expectations of others. These persons are more likely, therefore, to conform to group efforts and goals rather than loaf (Karau and Hart, 1998). According to the social evaluation explanation, members of highly cohesive groups feel strongly that they should perform at their best so as to achieve a high standard for the whole group. The CEM suggests that people in cohesive groups value their groups and work harder because the level of group cohesiveness directly affects the valence of the outcomes. Conversely, individuals in low cohesive groups do not feel the need to perform to their optimum capabilities and therefore perform SL.

Level of communication within the group Interaction and communication between fellow workers increase the awareness of the presence of others and the level of appreciation of the

Yair Amichai-Hamburger 93

individual towards his or her fellow worker. It was found that when a social dilemma arises, communication between group members reduces SL completely (Matsui, Kakuyama and Onglatco, 1987). The CEM suggests that communication among group members is likely to increase collective effort when it enhances the perception of task importance or social responsibility. However, should the same communication relay negative task attitudes or contribute to feelings of dispensability, collective effort will be hindered (Karau and Williams, 1993).

Intragroup competition Research findings suggest that implicit or explicit competition among the members of an ostensibly cooperative task group can enhance member motivation and performance (Erev, Bornstein and Galili, 1993; Stroebe, Diehl and Abakoumkin, 1996). According to the social evaluation explanation, competition between groups is likely to increase group cohesiveness and improve performance standards. As a result, performance of SL is minimal.

Perceived degree of effort invested by fellow members Harkins and Jackson (1985) manipulated participants’ expectations as to the degree of effort their co-workers will invest in a task. They found that participants matched their efforts to those they expected from their co-workers. When participants thought that co-workers would work hard, they, too, worked hard and did not practise SL. However, this process can be compensated for when the group members are highly motivated. Conversely, Williams and Karau (1991) found that people perform particularly well in order to compensate for the low performance of other group members. However, when they expect others to work hard, they will actually practise SL. The authors suggested that when people perceive the task results as desirable and also believe that they can compensate for the poor performance by other group members, they are likely to become highly motivated and perform at their best. There are times when despite being part of a group which is performing poorly, an individual member will not loaf, but rather will actually perform to his/her maximum capability. According to the CEM, this may be because of the high valence achieved by completing the task. Another suggestion is offered by the search for meaning theory. This argues that when an individual group member achieves his/her own important goal, by means of the collective task he/she is likely to ignore

94 Understanding Social Loafing

the poor performance of the other group members. The social comparison explanation offers no explanation for this phenomenon.

Individualism/collectivism Earley (1993) and Erez and Somech (1996) found that when collectivists work together in a group, SL is minimized because the group members respect one another and are likely to identify their own success with that of their group. Conversely, when the group is composed entirely of members with an individualistic outlook, the level of SL will be high because group members feel that they cannot influence the group outcome and that their separate contribution will not be recognized. In addition, it was found that when people with a collectivist outlook work within a group of people whose values are individualistic, they are likely to practise SL. This is because their fellow group members have individualistic values and so they cannot trust them to put their best efforts into the group task (Erez and Earley, 1993). Erez and Somech (1996) compared patterns of work behaviour among Israeli urban group members with those of members of an Israeli kibbutz, i.e. a self-managing collective village. They found that while urban group members practise more SL, among kibbutz members there was no difference in performance between the individual and the group tasks. Earley (1989) found that American managers practised SL, whereas Chinese managers did not. Meta-analysis of SL studies revealed that work groups composed of people from Western countries show greater levels of SL as compared with groups from Eastern countries (Karau and Williams, 1993). Values in individualistic culture, such as that of the United States, maintain that working hard for a collective goal runs contrary to self-interest. In contrast, collectivist cultures, such as that of China, place group goals ahead of personal goals. This results in individuals giving of their best in order to benefit the whole group, even if they themselves do not reap any individual benefit. They gain their satisfaction and feelings of accomplishment from group outcomes (Earley, 1989). This does not mean that no SL occurs in collectivistic cultures; SL was found among workers in Japan, which is considered to be a highly collectivist society (Kugihara, 1999). However, it is reasonable to expect that collectivists will perform much less SL than their individualist counterparts. The CEM suggests that cultural difference creates different sets of expectations among group members. These expectations have a direct bearing on the link created between expectancy, instrumentality, and valence, and as a result, on the performance in a group task.

Yair Amichai-Hamburger 95

In summary, the current analysis of SL correlates shows that the search for meaning explanation is particularly useful for explaining the correlates in which intrinsic motivation plays a significant role. The social comparison explanation provides a good understanding of the correlates where personal contrasting is the dominant factor for the individual, causing him/her to perform to the best of his/her ability or to loaf. Conversely, the CEM may be described as an umbrella theory. Unlike other theories that concentrate on a single aspect, the CEM provides an explanation for most of the correlates.

Ideas and tactics to reduce SL The negative picture regarding the prevalence of SL in-group tasks is not, however, conclusive. As described above, certain group members may work even harder in some cases to compensate for others’ lack of ability (Karau and Williams, 1997). It is also true that an understanding of both the causes that lead to SL and the variables associated with it may make it possible to minimize this phenomenon and eventually work towards its elimination. The following ideas and practical suggestions to diminish SL are divided into three sections: organizational and group culture, group characteristics, and individual characteristics.

Organizational and group culture.

All explanations of SL suggest tackling SL on a cultural level. The search for meaning explanation suggests creating a culture where people can feel that they can fulfil themselves through achieving group goals. The social evaluation explanation suggests establishing a culture of fairness, where people feel that when they do their best to achieve group goals so do the others. The CEM suggests building a challenging culture where workers become highly motivated to achieve group goals. Research into the cultural impact on SL in collectivistic societies found that the SL rate is lower than the rates found in individualistic societies. The challenge, therefore, is to build a cohesive working group with a relatively collectivistic nature. For those entrusted with the job of reducing SL, it is vital to remember that this is impossible to do without taking larger organizational factors into account. A relevant consideration here is the assessment of how much power the manager has to recreate the organizational culture. Consequently, the entire organization is recommended to adopt a management model that encourages workers and managers to work together in a win-win collaborative paradigm in

96 Understanding Social Loafing

order to integrate interests of workers and management (Thomas, 1976; Tjosvold, 1986). Workers who are involved and committed will create a highly motivated organization (Lee et al., 1992). One relevant management model is ‘Total Quality Management’ (TQM) – an organizational strategy aimed at increasing customer satisfaction by focusing organizational efforts on improving quality (Walton, 1990). This model stresses the need for high levels of communication between all levels of the organization in order to create a high degree of commitment towards the introduction of the quality management. One central tenet of TQM is an emphasis on continual improvement rather than on rigid standards of performance. This leads to processes whereby each achievement is challenged again and again. In many organizations, workers who initially worked hard to achieve the goals of the organization no longer feel challenged and are actually loafing today. By reinforcing continual improvement, TQM eliminates this phenomenon. Currently we have no empirical evidence showing a direct correlation between the implementation of TQM and the reduction of SL. However, given the aforementioned considerations, such linkage seems likely.

Group characteristics.

The search for meaning explanation advocates the importance of building a process whereby the group gains an increasing degree of control on group processes. The social evaluation explanation and the CEM advocate the importance of controlling group characteristics so that individuals cooperating together on group tasks will be able to fulfil their own interests through those interests of the group. The group should consist of a variety of members whose different skills cover a wide spectrum that will enable the group to complete the task at a very high level (Harari, 1995). It is recommended that the group size be kept small (Hogg and Abrams, 1993), with each group member responsible for his/her own area of expertise (Harkins and Petty, 1982). At the early stages of the group development, more emphasis should be placed on providing monetary incentives (Shepperd and Wright, 1989), threats of punishment (Zaccaro, 1984), and a contingent reward system (Miles and Greenberg, 1993). The group should be directed towards specific goals until it becomes independent and, more importantly, is able to use a team approach to set its own agenda (White, Kjelgaard and Harkins, 1995). As group members become more involved, their motivation will strengthen (George, 1992) and the group may set their own group performance goals (Ray and Bronstein, 1995). When members identify with group goals, a cohesive group will be created. The feelings of mutual responsibility may cause group

Yair Amichai-Hamburger 97

members to compensate for the poor performance of weaker colleagues (Karau and Williams, 1997).

Task characteristics.

The search for meaning explanation advocates designing group tasks in a way that will enhance intrinsic motivation. The social evaluation explanation suggests the need to define tasks through which people feel highly accountable so they will not diffuse their responsibility. The CEM suggests designing attractive and challenging group tasks that will enhance intrinsic and extrinsic motivation (Harkins and Petty, 1982; Zaccaro, 1984). It is therefore important that goals should be ambitious but practical, in order to create a feeling of achievement, self-esteem (Karau and Williams, 1993), and self-efficacy (Maurer and Pierce, 1998). Tasks should be built in such a way that individual contributions may be easily identified and evaluated (Kerr and Bruun, 1981; Williams, Harkins and Latane, 1981). Leaders should give continuous feedback to the group on their performance (Erez and Somech, 1996). In a modern organizational setting, this may be achieved using information systems. For example, in an electronic brainstorming group session, each group member may be given access to computer screens showing everyone’s ideas (George, 1995).

Individual characteristics. The social comparison and the CEM explanations help understand the impact of the group composition on the decision made by the individual as to how much effort he/she will put into the task. When someone feels that he/she belongs to a powerful group of unique individuals, he/she is very likely to perform to his/her maximum capabilities. When new staff is hired, it is advised to recruit people who, in addition to their task-related abilities, have a high need for cognition (Smith et al., 2001), are approval oriented (Sorrentino and Sheppard, 1978), and are not arrogant (Huguet et al., 1999). People with these characteristics have a strong internal tendency to perform to their maximum ability and are less likely to be affected by aversive situational factors.

Suggestions for future research A comprehensive SL questionnaire has yet to be designed. Researchers are encouraged to construct a questionnaire that will examine the relative contribution of all the correlates of SL. This examination will enhance our understanding of SL and assist in the building of an organizational strategy to reduce it.

98 Understanding Social Loafing

In an individualistic rather than in a collectivist society, the elimination of SL is clearly an uphill struggle. Transformational leaders may have a chance of success. As research showed that transformational leaders are capable of transforming a group of individuals into a cohesive task group, it is reasonable to expect that they be able to reduce SL (Bass, 1985; Roberts, 1985). Under the tutelage of a transformational leader, it is likely that group members will be less interested in loafing. It is therefore important that future studies examine carefully the impact of transformational leadership on groups. Another important topic for future research is personality as a determinant of SL. People who are Type A or low in conscientiousness or have an external locus of control seem to be more prone to practise SL than others. These should be questions for future research. Lastly, it seems that in the modern organizational world, virtual teams would be more common. In this type of setup, team members work together without being in physical proximity (Johnson, Heimann and O’Neil, 2001). In some of the cases, the virtual group members never actually meet one another. This phenomenon should be studied in the context of SL so that the effectivity of these teams may be evaluated.

References Adams, J. S. (1965) ‘Inequity in social exchange’, in L. Berkowitz (ed.), Advances in Experimental Social Psychology (vol. 2, pp. 267–99). New York: Academic Press. Albanese, R. and Van Fleet, D. D. (1985) ‘Rational behavior in groups: the free riding tendency’, Academy of Management Review, 10, 244–55. Alderfer, C. P. (1972) Existence, Relatedness and Growth: Human Needs in Organizational Settings. New York: Free Press. Anderson, L. R. and Blanchard, P. N. (1982) ‘Sex differences in task and social– emotional behavior’, Basic and Applied Social Psychology, 3, 109–39. Bass, B. M. (1985) Leadership Performance Beyond Expectations. New York: Free Press. Brickner, M. A., Harkins, S. G. and Ostrom, T. M. (1986) ‘Effects of personal involvement: thought-provoking implications for social loafing’, Journal of Personality and Social Psychology, 51, 763–70. Cacioppo, J. T. and Petty, R. E. (1982) ‘The need for cognition’, Journal of Personality and Social Psychology, 42, 116–31. Campion, M. A. and McClelland, C. L. (1993) ‘Follow-up and extension of the interdisciplinary costs and benefits of enlarged jobs’, Journal of Applied Psychology, 78, 339–51. Charbonnier, E., Huguet, P., Brauer, M. and Monteil, J. M. (1998) ‘Social loafing and self-beliefs: people’s collective effort depends on the extent to which they distinguish themselves as better than others’, Social Behavior and Personality, 26, 329–40. Cialdini, R. B. (1993) Influence: Science and Practice. New York: HarperCollins.

Yair Amichai-Hamburger 99 Duffy, M. K. and Shaw, J. D. (2000) ‘The Salieri Syndrome: consequences of envy in groups’, Small Group Research, 31, 3–23. Earley, P. C. (1989) ‘Social loafing and collectivism: a comparison of the United States and the People’s Republic of China’, Administrative Science Quarterly, 34, 565–81. Earley, P. C. (1993) ‘East meets West meets Mideast: further explorations of collectivistic and individualistic work groups’, Academy of Management Journal, 36, 319–48. Erev, I., Bornstein, G. and Galili, R. (1993) ‘Constructive intergroup competition as a solution to the free rider problem: a field experiment’, Journal of Experimental Social Psychology, 29, 463–78. Erez, M. and Earley, P. C. (1993) Culture, Self-Identity and Work. New York: Oxford University Press. Erez, M. and Somech, A. (1996) ‘Is group productivity loss the rule or the exception? Effects of culture and group based motivation’, Academy of Management Journal, 39, 1513–37. George, J. M. (1992) ‘Extrinsic and intrinsic origins of perceived social loafing in organizations’, Academy of Management Journal, 35, 191–202. George, J. M. (1995) ‘Asymmetrical effects of rewards and punishments: the case of social loafing’, Journal of Occupational and Organizational Psychology, 68, 327–38. Hackman, J. R. (1992) ‘Group influences on individuals in organizations’, in M. D. Dunnette and L. M. Hough (eds), Handbook of Industrial and Organizational Psychology (2nd edn, vol. 3, pp. 199–268) Palo Alto, CA: Consulting Psychologists Press. Hackman, J. R. and Oldham, G. R. (1976) ‘Motivation through the design of work: test of a theory’, Organizational Behavior and Human Performance, 16, 250–79. Harari, O. (1995) ‘The dream team’, Management Review, October, pp. 29–31. Harkins, S. G. and Jackson, J. M. (1985) ‘The role of evaluation in eliminating social loafing’, Personality and Social Psychology Bulletin, 11, 457–65. Harkins, S. G. and Petty, R. E. (1982) ‘Effects of task difficulty and task uniqueness on social loafing’, Journal of Personality and Social Psychology, 43, 1214–29. Henningsen, D. D., Cruz, M. G. and Miller, M. L. (2000) ‘Role of social loafing in predeliberation decision making’, Group Dynamics, 4, 168–75. Hoeksema-van Orden, C. Y. D., Gaillard, A. W. K. and Buunk, B. P. (1998) ‘Social loafing under fatigue’, Journal of Personality and Social Psychology, 75, 1179–90. Hogg, M. A. and Abrams, D. (1993) Group Motivation: Social Psychological Perspectives. Hemel Hempstead: Harvester Wheatsheaf. Huguet, P., Charbonnier, E. and Monteil, J. M. (1999) ‘Productivity loss in performance groups: people who see themselves as average do not engage in social loafing’, Group Dynamics, 3, 118–31. Huseman, R. C., Hatfield, J. D. and Miles, E. W. (1987) ‘A new perspective on equity theory: the equity sensitivity construct’, Academy of Management Review, 12, 222–34. Jackson, J. M. and Padgett, V. R. (1982) ‘With a little help from my friends: social loafing and the Lennon–McCartney songs’, Personality and Social Psychology Bulletin, 8, 672–7. Jackson, J. M. and Williams, K. D. (1985) ‘Social loafing and difficult tasks: working collectively can improve performance’, Journal of Personality and Social Psychology, 49, 937–42.

100 Understanding Social Loafing Johnson, P., Heimann, V. and O’Neil, K. (2001) ‘The “wonderland” of virtual teams’, Journal of Workplace Learning, 13, 24–30. Karau, S. J. and Hart, J. W. (1998) ‘Group cohesiveness and social loafing: effects of social interaction manipulation on individual motivation within groups’, Group Dynamics, 2, 185–91. Karau, S. J. and Williams, K. (1993) ‘Social loafing: a meta-analytic review and theoretical integration’, Journal of Personality and Social Psychology, 65, 681–706. Karau, S. J. and Williams, K. D. (1997) ‘The effects of group cohesiveness on social loafing and social compensation’, Group Dynamics, 1, 156–68. Katzenbach, J. R. and Smith, D. K. (1993) ‘The discipline of teams’, Harvard Business Review, 71 (March–April), 111–20. Kerr, N. L. (1983) ‘Motivation losses in small groups: a social dilemma analysis’, Journal of Personality and Social Psychology, 45, 818–28. Kerr, N. L. and Bruun, S. E. (1981) ‘Ringelmann revisited: alternative explanations for the social loafing effect’, Personality and Social Psychology Bulletin, 7, 224–31. Kerr, N. L. and MacCoun, R. J. (1984) ‘Sex composition of groups and member motivation: effects of relative task ability’, Basic and Applied Social Psychology, 5, 255–71. Kerr, N. L. and Sullaway, M. E. (1983) ‘Group sex composition and member task motivation’, Sex Roles, 9, 403–17. Kravitz, D. A. and Martin, B. (1986) ‘Ringelmann rediscovered: the original article’, Journal of Personality and Social Psychology, 50, 936–41. Kugihara, N. (1999) ‘Gender and social loafing in Japan’, Journal of Social Psychology, 139, 516–26. Latane, B. (1981) ‘The psychology of social impact’, American Psychologist, 36, 343–56. Latane, B. and L’Herrou, T. (1996) ‘Spatial clustering in the conformity game: dynamic social impact in electronic groups’, Journal of Personality and Social Psychology, 70, 1218–30. Latane, B., Williams, K. and Harkins, S. (1979) ‘Many hands make light the work: the causes and consequences of social loafing’, Journal of Personality and Social Psychology, 37, 822–32. Lee, T. W., Ashford, S. J., Walsh, J. P. and Mowday, R. T. (1992) ‘Commitment propensity, organizational commitment, and voluntary turnover: a longitudinal study of organizational processes’, Journal of Management, 18, 15–32. Locke, E. A. and Latham, G. P. (1990) A Theory of Goal Setting and Task Performance. Englewood Cliffs, NJ: Prentice-Hall. Luthans, F. and Reif, W. E. (1974) ‘Job enrichment: long on theory, short on practice’, Organizational Dynamics, 2, 30–43. Maslow, A. H. (1943) ‘Deprivation, threat and frustration’, Psychological Review, 50, 370–96. Maslow, A. H. (1971) The Further Research of Human Nature. New York: Viking. Maslow, A. H., Stephens, D. C. and Heil, G. (1998) Maslow on Management. New York: Wiley. Matsui, T., Kakuyama, T. and Onglatco, M. U. (1987) ‘Effects of goals and feedback on performance in groups’, Journal of Applied Psychology, 72, 407–15. Maurer, T. J. and Pierce, H. R. (1998) ‘A comparison of Likert scale and traditional measures of self-efficacy’, Journal of Applied Psychology, 83, 324–9.

Yair Amichai-Hamburger 101 Miles, J. A. and Greenberg, J. (1993) ‘Using punishment threats to attenuate social loafing effects among swimmers’, Organizational Behavior and Human Decision Processes, 56, 246–65. Mulvey, P. W., Veiga, J. F. and Elsass, P. M. (1996) ‘When team mates raise a white flag’, Academy of Management Executive, 10, 40–9. Petty, R. E., Harkins, S. J., Williams, K. D. and Latane, B. (1977) ‘The effects of group size on cognitive effort and evaluation’, Personality and Social Psychology Bulletin, 3, 579–82. Ray, D. and Bronstein, H. (1995) Teaming Up. New York: McGraw-Hill. Roberts, N. C. (1985) ‘Transforming leadership: a process of collective action’, Human Relations, 38, 1023–46. Roy, M. C., Gauvin, S. and Limayem, M. (1996) ‘Electronic group brainstorming: the role of feedback on productivity’, Small Group Research, 27, 215–47. Shepperd, J. A. and Wright, R. A. (1989) ‘Individual contributions to the collective effort: an incentive analysis’, Personality and Social Psychology Bulletin, 15, 141–9. Smith, B. N., Kerr, N. A., Markus, M. J. and Stasson, M. F. (2001) ‘Individual differences in social loafing: need for cognition as a motivator in collective performance’, Group Dynamics, 5, 150–8. Sorrentino, R. M. and Sheppard, B. H. (1978) ‘Effects of affiliation-related motives on swimmers in individual versus group competition: a field experiment’, Journal of Personality and Social Psychology, 36, 704–14. Steiner, I. D. (1972) Group Process and Productivity. New York: Academic Press. Stroebe, W., Diehl, M. and Abakoumkin, G. (1996) ‘Social compensation and the Koehler effect: toward a theoretical explanation of motivation gains in group productivity’, in E. H. Witte and J. H. Davis (eds), Understanding Group Behavior: Small Group Processes and Interpersonal Relations (vol 2., pp. 37–65). Hillsdale, NJ: Erlbaum. Tesser, A. and Campbell, J. (1990) ‘Self-definition: the impact of the relative performance and similarity of others’, Social Psychology Quarterly, 43, 341–7. Thomas, K. W. (1976) ‘Conflict and conflict management’, in M. D. Dunnette (ed.), Handbook of Industrial and Organizational Psychology (pp. 889–935) Chicago: Rand McNally. Tjosvold, D. (1986) Working Together to Get Things Done. Lexington, MA: Lexington Books. Vroom, V. H. (1964) Work and Motivation. New York: Wiley. Wageman, R. (1999) ‘Task design, outcome interdependence, and individual differences: their joint effects on effort in task performing teams’, Group Dynamics, 3, 132–5. Walton, M. (1990) The Demming Management Method at Work. New York: Pedigree. Weldon, E. and Mustari, E. L. (1988) ‘Felt dispensability in groups of coactors: the effects of shared responsibility and explicit anonymity on cognitive effort’, Organizational Behavior and Human Decision Processes, 41, 330–51. White, P. H., Kjelgaard, M. M. and Harkins, S. G. (1995) ‘Testing the contribution of self-evaluation of goal-setting effects’, Journal of Personality and Social Psychology, 69, 69–79. Williams, K., Harkins, S. G. and Latane, B. (1981) ‘Identifiability as a deterrent of social loafing: two cheering experiences’, Journal of Personality and Social Psychology, 40, 303–11.

102 Understanding Social Loafing Williams, K. D. and Karau, S. J. (1991) ‘Social loafing and social compensation: the effects of expectations of co-worker performance’, Journal of Personality and Social Psychology, 61, 570–81. Williams, K. D., Nida, S. A., Baca, L. D. and Latane, B. (1989) ‘Social loafing and swimming: effects of identifiability on individual and relay performance of intercollege swimmers’, Basic and Applied Social Psychology, 10, 73–81. Zaccaro, S. J. (1984) ‘Social loafing: the role of task attractiveness’, Personality and Social Psychology Bulletin, 10, 99–106.

6 Job Burnout and Dysfunctional Work Attitude Thomas J. Kalliath*

Introduction Over the past three decades, organizational researchers have shown that job burnout could result in a number of dysfunctional attitudes affecting individual employee well-being and organizational outcomes, including negative orientation towards clients (Cherniss, 1980), reduced job performance (Kahill, 1988), reduced organizational commitment (Kalliath, O’Driscoll and Gillespie, 1998; Leiter and Maslach, 1988), reduced job satisfaction (Maslach and Jackson, 1981), reduced quality of personal life (Kahill, 1988; Maslach and Jackson, 1981), depression (Jayaratne and Chess, 1983; Jayaratne, Chess and Kunkel, 1986), and guilt (Pines and Kafry, 1981). One common theme running through this stream of research is the effect of job burnout contributing to negative work attitudes. Given the mounting costs of job burnout to organizations and to individual employees (Golembiewski, Hilles and Rick, 1987; ILO, 1992; Pines and Maslach, 1978), the investigation of viable antidotes to job burnout out has become urgent. In the present chapter, I shall present evidence to support the converse hypotheses – that positive work attitudes can combat job burnout. I will show that there are direct and indirect effects of three commonly held work attitudes, namely job satisfaction, organizational commitment, and supervisor satisfaction on reduced levels of job burnout. However, before presenting this evidence, I will define and describe the job burnout construct, discuss the

* Award of an Academic Research Visitor grant from the School of Social Sciences at the University of Waikato to the author supported this research.

103

104 Job Burnout and Dysfunctional Work Attitude

relationship between job burnout with three antecedent variables, namely job satisfaction, organizational commitment and supervisor satisfaction, and then present the hypotheses to test the converse hypotheses.

Job burnout The originators of the construct, Maslach and Jackson, define burnout as ‘a syndrome of emotional exhaustion, depersonalization, and reduced personal accomplishment that can occur among individuals who do “people work”’ (1986, p. 1). More fully, burnout is ‘the gradual loss of caring about the people they work with. Over time they find they simply cannot sustain the kind of personal care and commitment called for in the personal encounters that are the essence of their job’ (Maslach, 1978, p. 58). Maslach states that there are three components to the experience of burnout that have implications for individuals who do ‘people work’. These are: emotional exhaustion, depersonalization and diminished personal accomplishment. Each component of burnout needs to be understood in the context of what happens to the caregivers in the process of caring. Emotional exhaustion. The first component, emotional exhaustion, is generally considered to be the core symptom of burnout, and it is strongly related to other burnout dimensions as well. Emotional exhaustion refers to the feelings of being emotionally over-extended and exhausted by one’s work (Maslach and Jackson, 1986). Emotional exhaustion is characterized by a lack of energy and a feeling that one’s emotional resources are used up (Cordes and Dougherty, 1993). Maslach (1982) describes it as a response to the overwhelming emotional demands of other people. Individuals who are emotionally exhausted lack enough energy to face another day, and often report that they are filled with dread at the prospect of returning to work for another day. Depersonalization. The second component of burnout is depersonalization. Researchers have described depersonalization as a (defensive) coping response to deal with the effects of emotional exhaustion when other coping resources are not available (Lee and Ashforth, 1990; Leiter, 1990; Maslach, 1982). Caregivers develop a detached, callous and even dehumanized responses to the people they are supposed to care for. Visible symptoms of depersonalization are the use of derogatory or abusive language, strict compartmentalization of the situation, withdrawal through longer breaks, and an unfeeling and impersonal response towards recipients of one’s service, care, treatment or instruction (Cordes and Doughtery, 1993).

Thomas J. Kalliath 105

Reduced personal accomplishment. The third component of burnout refers to a sense of reduced personal accomplishment. This negative feeling partly stems from depersonalization – caregivers experience negative feelings about themselves. Caregivers feel distress or guilt about the way they have thought about or mistreated others and this leads to feelings of reduced personal accomplishment. Caregivers feel that all efforts repeatedly fail to produce positive results, and so they ‘quit trying’ (Maslach, 1982). Feelings of diminished personal accomplishment can also result from factors that suggest that one is unappreciated, that one’s efforts are ineffective (Jackson, Turner and Brief, 1987), or that one’s competence is low (Burke, Shearer and Deszca, 1984). The perception of self-efficacy which is defined as ‘people’s judgments of their capabilities to organize and execute courses of action required to attain designated types of performances’ (Bandura, 1986, p. 391), is at the core of the personal accomplishment component (Lee and Ashforth, 1990).

Job burnout and dysfunctional attitudes The descriptions of three dimensions of job burnout in the previous sections clearly suggest that burnout is a dysfunctional work attitude taking considerable toll on the well-being of individual employees and on organizational outcomes. In the past thirty years, organizational researchers have demonstrated that job burnout is a good predictor of related dysfunctional work attitudes such as lowered job satisfaction, reduced organizational commitment and poor supervisor satisfaction. I present next a review of this literature.

Job burnout and job satisfaction Of various work attitudes investigated, the negative impact of burnout on job satisfaction has received the most amount of research attention (Brookings et al., 1985; Burke, 1987; Burke and Greenglass, 1989; Golembiewski and Munzenrider, 1988; Justice, Gold and Klein, 1981; Lee and Ashforth, 1996; Maslach and Jackson, 1981; Maslach and Jackson, 1984; Pines, Kafry and Etzion, 1980). In a study by Pines, Kafry and Etzion (1980) involving 32 nurses, burnout was found to be significantly and negatively correlated with self-ratings of satisfaction from work (r = −.71, p < .01), from life (r = −.35, p < .01) and from oneself (r = −.48, p < .01). In another study involving 135 female human service professionals, Brookings et al. (1985) used the MBI and 4-item Hoppock Job Satisfaction Blank to measure overall job satisfaction. They found that

106 Job Burnout and Dysfunctional Work Attitude

emotional exhaustion was negatively correlated with job satisfaction (r = −0.59, p < .01) and so was depersonalization (r = −0.50, p < .01). There was a moderate, positive correlation between personal accomplishment and job satisfaction (r = 0.32, p < .01). Maslach and Jackson (1981) found that job satisfaction had a moderate negative correlation with both emotional exhaustion (r = −0.23, p < .05) and depersonalization (r = −0.22, p < .02) as well as a slightly positive correlation with personal accomplishment (r = 0.17, p < .06). A meta-analytic examination of the correlates of burnout by Lee and Ashforth (1996) calculated the mean corrected correlation among 61 studies, of which 80 per cent sampled human services providers. The correlation corrected for artefacts between emotional exhaustion and job satisfaction was (rc = −.31, p < .001), between depersonalization and job satisfaction was (rc = −.44, p < .001) and finally between personal accomplishment and job satisfaction was (rc = .29, p < .001). These studies consistently show the strong link between job burnout and job satisfaction. In most studies reviewed, job burnout was a predictor of job satisfaction.

Burnout and organizational commitment A closely related yet distinct construct, organizational commitment refers to the extent to which employees incorporate the values of the organization and have intentions to remain in the organization (Mowday, Steers and Porter, 1979). Jackson, Turner and Brief (1987) investigated whether high levels of burnout among public service lawyers was associated with reduced commitment to their agency. Their study revealed that all three burnout constructs (emotional exhaustion, depersonalization and reduced personal accomplishment) were significantly related to organizational commitment. Leiter and Maslach (1988) examined the contribution of the three burnout components to organizational commitment among nurses in a small general hospital. They found that emotional exhaustion and personal accomplishment exhibited a direct influence on commitment, and exhaustion also showed an indirect effect via depersonalization, but depersonalization itself made no direct contribution to nurses’ commitment scores. Other studies by Leiter have supported burnout as a mediating variable between organizational demands and organizational commitment (Leiter, 1988a; 1991). A meta-analytic examination of the correlates of burnout by Lee and Ashforth (1996) showed correlation corrected for artefacts between organizational commitment and emotional exhaustion was (rc = −.43, p < .001), between organizational commitment and depersonalization

Thomas J. Kalliath 107

was (rc = −.42, p < .001) and finally between personal accomplishment and organizational commitment was (rc = −.02).

Burnout and supervisor support There is now considerable evidence that lack of supervisor support is a major determinant of job burnout (Constable and Russell, 1986; Jackson, Turner and Brief, 1987; Leiter, 1988b; Leiter and Maslach, 1988; Ross, Altmaier and Russell, 1989). In a study of 310 military nurses, Constable and Russell (1986) showed that lack of supervisor support along with work pressure and low job enhancement was one of the variables contributing to job burnout. Leiter and Maslach (1988) found that lack of supervisor support contributed to emotional exhaustion among nurses, while Jackson, Turner and Brief (1987) showed in a study of 391 public service lawyers that a feeling of personal accomplishment was associated with supervisory support. A meta-analysis by Lee and Ashforth (1996) showed the correlation corrected for artefacts between emotional exhaustion and supervisor satisfaction was (rc = −.37, p < .001), between depersonalization and supervisor satisfaction was (rc = −.24, p < .001) and between personal accomplishment and supervisor satisfaction was (rc = .14). In sum, the literature review shows that burnout was a consistent predictor of negative and dysfunctional work attitudes such as reduced job satisfaction, reduced organizational commitment, and poor supervisor satisfaction. The thread of evidence for the converse hypothesis – the effect of positive work attitudes on burnout levels – is less consistent. Dolan (1987) for instance showed that reduced job satisfaction was a reliable predictor of burnout among nurses. However, in another study by Wolpin, Burke and Greenglass (1991) involving 245 school-based educators the authors came to the conclusion that ‘psychological burnout appears to cause lower job satisfaction and not vice versa’ (p. 205). The link between burnout and organizational commitment was shown by Kalliath, O’Driscoll and Gillespie (1998) in a sample of 197 nurses and 110 laboratory technicians. They found that commitment had direct effects on emotional exhaustion and depersonalization, and a weaker indirect effect (via exhaustion) on depersonalization. In another study of 500 hospital nurses, Reilly (1994) showed that nurses who reported greater commitment tended to experience less burnout overall. Constable and Russell (1986) found ‘buffering effects’ of supervisor support on job burnout; as supervisory support increased, the negative relationship between job enhancement and emotional exhaustion

108 Job Burnout and Dysfunctional Work Attitude –

Job satisfaction

Emotional exhaustion

+ –

Depersonalization Figure 6.1 Theoretical model tested Note: The theoretical model with Job satisfaction as predictor of burnout is shown. Other work attitudes tested were: organizational commitment and supervisor satisfaction.

disappeared. They concluded that managers could prevent burnout by increasing supervisory support among nurses. The evidence presented supporting the converse hypotheses although sparse suggests that in stressful work environments, higher levels of positive work attitudes may ameliorate the extent to which burnout is experienced by employees. The positive work attitude reflected in higher levels of job satisfaction, organizational commitment and supervisor satisfaction might shield employees from extreme psychological strain induced by the experience of burnout. In the present study, I explicitly test the hypotheses that higher levels of job satisfaction, organizational commitment and supervisor satisfaction would have a direct effect on emotional exhaustion and depersonalization, and (following Leiter, 1993) that exhaustion would also make a direct contribution to depersonalization. I expected to find also an indirect effect of job satisfaction, organizational commitment and supervisor satisfaction on depersonalization transmitted through emotional exhaustion. Figure 6.1 depicts the proposed theoretical models to be tested.

Method Sample Data were collected from 203 nurses, 145 mangers and 110 laboratory technicians employed by a general community hospital located in a moderately large city in the Midwestern United States (response rate was

Thomas J. Kalliath 109

60 per cent). Mean ages of respondents were 38 years for the nursing sample, and 40 years for managers and 35.9 years for technicians. Average organizational tenure for the nurses was 14.2 years, for managers 15.4 years, and for technicians 18 years. Seventy-five per cent of the nurses and managers had been in the present position for more than 3 years. Ninety per cent of the nurses were females, whereas 72 per cent of the managers were females.

Measures The Maslach Burnout Inventory (MBI; Maslach and Jackson, 1981, 1986) consists of 22 items, of which 9 items are hypothesized to measure emotional exhaustion, 5 items to measure depersonalization, and 8 items to measure personal accomplishment. All 22 items responded to a 7-point frequency scale. Confirmatory factor analyses showed that only two of the three burnout components (emotional exhaustion and depersonalization) could be retained for further analysis (Kalliath et al., 2000). In addition, the results from the confirmatory analyses indicated that a good fit could be obtained only when certain items from the emotional exhaustion and depersonalization subscales were eliminated. The resulting burnout measure comprised five items assessing emotional exhaustion (items 1, 3, 13, 14 and 20 in the MBI) and two items (numbers 10 and 11) tapping depersonalization. The job satisfaction scale measured respondents’ overall job satisfaction. Items were taken from Katzell, Thompson and Guzzo’s (1992) study of job satisfaction and job performance. Katzell et al. (1992) in turn took the best items from the Michigan Organizational Assessment Questionnaire (Cammann et al., 1983) and Overall Job Satisfaction Scale (Quinn and Staines, 1979). The job satisfaction scale responded to a 7-point response scale. Organizational commitment was measured using the 15-item Organizational Commitment Questionnaire (OCQ) (Mowday, Porter and Steers, 1982). The response format uses a 7-point Likert scale with anchors from 1 = strongly disagree to 7 = strongly agree. Mowday, Steers and Porter (1979) have reported internal consistency (Cronbach’s alpha) ranging from .82 to .93 with a median of .90 for the OCQ measure. Supervisor satisfaction was measured using 6 items taken from the Index of Organizational Reactions (Dunhan, Smith and Blackburn, 1977). The response format uses a 5-point Likert scale with anchors from 1 = strongly disagree to 5 = strongly agree. Dunham et al. have reported internal consistency in the range of .86 to .92.

110 Job Burnout and Dysfunctional Work Attitude

Analysis SEM allowed us to test fit, first for the measurement models, then for the hypothesized theoretical relationships between the latent constructs (Anderson and Gerbing, 1988). In the first wave of analysis, we developed good-fitting measurement models for emotional exhaustion, depersonalization, job satisfaction, organizational commitment and supervisor satisfaction. In the second wave of the analysis, we tested the theoretical model depicting the relationship between job satisfaction and burnout outlined earlier. The model was developed with the sample of nurses and then replicated in a sample of managers and laboratory technicians with the parameter estimates constrained to the previously estimated values (Anderson and Gerbing, 1988).

Results Table 6.1 shows the item squared multiple correlations (R2) of all items (i.e. observed variables) used in the structural models. Emotional exhaustion was measured by five items and depersonalization by two items. The R2 values for emotional exhaustion items ranged from .46 to .73 across three samples. These figures correspond to coefficient alphas of .68 to .85, illustrating that a substantial portion of the variability in each indicator was accounted for by the corresponding factor. The two items measuring depersonalization showed R 2 values in the range of .68 to .93 (corresponding coefficient alphas are .82 and .96), which confirms that a large part of the variability in each indicator was accounted for by the depersonalization factor. The bottom half of Table 6.1 shows the three predictor variables (job satisfaction, organizational commitment and supervisor satisfaction) and their corresponding R 2 values, which range from .47 to .84 (the corresponding coefficient alpha are .69 and .92).

Fit statistics Table 6.1 presents the fit statistics for the theoretical model tested in the sample of nurses, managers and technicians. In all, nine models were tested, three models with job satisfaction predicting job burnout (i.e. emotional exhaustion and depersonalization), three with organizational commitment as predictor and the last three models with supervisor satisfaction as predictor. I obtained non-significant chi-squares for all nine structural models tested. Note that the chi-square statistic provides a measure of the discrepancy between the sample covariance matrix and

Table 6.1 Item number

Descriptive statistics across three samples Item description

Nurses

Managers

Technicians

R2

Mean

SD

R2

Mean

3.05 2.75

1.47 .50 1.62 .52

3.45 2.75

1.59 1.70

.53 .49

2.87 2.58

1.69 .48 1.81 .68

2.75 2.58 1.14

1.56 .73 1.60 .46 1.27 .59

3.12 3.21 1.61

1.67 1.80 1.59

.65 .53 .57

2.58 2.7 1.16

1.78 .68 1.79 .69 1.46 .47

Depersonalization 1 I’ve become more callous towards people since I took this job. 2 I worry that this job is hardening me emotionally.

1.61 1.48

1.67 .77 1.61 .85

1.43 1.29

1.64 1.76

.82 .84

1.57 1.55

1.73 .68 1.89 .93

Job satisfaction 1 In general, I like working here. 2 In general, I don’t like my job (R). 3 How well do you like your job at . . . .

5.80 5.45 5.36

1.27 .62 1.41 .65 .96 .73

6.03 6.26 5.70

1.14 1.00 .98

.77 .77 .70

5.89 6.14 5.38

1.18 .77 1.29 .59 1.06 .72

Organizational commitment 1 I talk up ‘O’ to my friends as a great organization to work for. 2 I find that my values and ‘O’s values are very similar. 3 I am proud to tell other that I am part of ‘O’. 4 I am extremely glad that I chose ‘O’ to work for over others. 5 I really care about the fate of ‘O’. 6 For me, this is the best of all possible organizations to work for.

4.72 4.11 4.83 4.92 5.79 3.97

1.50 1.52 1.45 1.50 1.24 1.54

5.17 4.65 5.30 5.18 6.35 4.49

1.50 1.54 1.44 1.41 .88 1.55

.73 .51 .83 .59 .39 .67

5.03 4.37 5.21 5.04 5.98 4.31

1.49 1.49 1.39 1.43 1.14 1.64

Mean Emotional exhaustion 1 I feel emotionally drained from my work. 2 I feel fatigued when I get up in the morning and have to face another day on the job. 3 I feel frustrated by my job. 4 I feel I’m working too hard on my job. 5 I feel like I’m at the end of my rope.

SD

.67 .50 .69 .64 .47 .60

SD

R2

.73 .51 .83 .59 .39 .67

111

112

Table 6.1

(Continued)

Item number

Item description

Supervisory satisfaction 1 My supervisor has more bad traits than good ones. 2 The supervision I receive encourages me to give extra effort. 3 The efforts of my supervisor add little to the success of my unit. 4 I am dissatisfied by the supervision I receive. 5 I feel that I would be better off working under a different supervisor.

Nurses

Managers

Technicians

Mean

SD

R2

Mean

SD

R2

Mean

SD

R2

3.93 3.43 3.72 3.77 3.95

1.08 1.13 1.20 1.19 1.19

.74 .64 .64 .84 .83

4.07 3.61 3.65 3.88 3.95

.99 1.17 1.22 1.20 1.18

.64 .58 .58 .85 .79

3.78 3.35 3.33 3.58 3.65

1.09 1.11 1.27 1.25 1.25

.69 .59 .63 .85 .83

Note: The squared multiple correlation (R2) is a direct index of item performance for each factor; an approximation to the alpha coefficient is attained by taking the square root of R2. An R2 of .49 corresponds to an alpha coefficient of .70. ‘O’ refers to the name of the organization where study was conducted.

Thomas J. Kalliath 113 Table 6.2

Fit statistics of the tested structural models

Tested models Job satisfaction Nurses (N = 203) Managers (N = 145) Technicians (N = 110)

Chi-square

p

df

RMSEA CI for RMSEA GFI

CFI

49.02 48.94 43.25

.07 36 .07 36 .16 35

.05 .05 .05

(.0; .07) (.0; .08) (.0; .08)

.96 .94 .94

.99 .98 .99

Organizational commitment Nurses (N = 203) 91.02 Managers (N = 145) 36.00 Technicians (N = 110) 77.56

.07 72 .08 58 .22 69

.04 .05 .04

(0; .06) (0; .07) (0; .07)

.99 .92 .99

.99 .98 .97

Supervisory satisfaction Nurses (N = 203) Managers (N = 145) Technicians (N = 110)

.65 54 .09 56 .09 51

0.00 .04 .05

(0; .04) (0; .07) (0; .08)

.96 1.00 .93 .99 .91 .99

49.49 70.58 65.37

Note: RMSEA = root mean-square error of approximation; CI for RMSEA = confidence interval for RMSEA; GFI = goodness of fit index; CFI = comparitive fit index. Note that smaller chi-square value indicates better model fit.

the fitted covariance matrix. The statistically non-significant chi-square value obtained in all the three samples revealed good overall fit of the proposed theoretical model with data. Other fit statistics assessed were: the Root Mean Square of Approximation (RMSEA) and the 90 per cent confidence interval of the RMSEA. The RMSEA estimates the overall amount of error; it is a function of the fitting function value relative to the degrees of freedom. The RMSEA point estimate should be .05 or less and the upper limit of the confidence interval should not exceed .08. Table 6.2 shows that in the models tested, RMSEA ranged from .00 to .05 and the Confidence Interval for RMSEA did not exceed the upper limit of .08, which reflected good fit for all the models. Two other fit statistics were assessed to confirm the good fit for the theoretical model with data. First, the Goodness of Fit Index (GFI) which indicates the relative amount of the variances/co-variances in the sample predicted by the estimates of the population. It usually varies between 0 and 1 and a result of .90 or above indicates good model fit. In the models tested, GFI ranged from .91 to .99, indicative of good fit. Second, the Comparative Fit Index (CFI) indicates how much better the model fits as compared to a model of independence (the null model). CFI indicates the proportionate decrease in chi-square from the null model to the theoretical model. A result of .90 or above is indicative of a good model fit. In the present tests, CFI ranged from .97 to 1.00, suggestive of good model fit.

114 Job Burnout and Dysfunctional Work Attitude a. Nurses JS

–.97** (–7.59)

–.34 (–1.87)

EE .60** (4.88)

DP

Figure 6.2 burnout

c. Technicians

b. Managers JS

–.69** (–5.37)

–.14 (–.91)

EE .66** (4.69)

DP

JS

–.87** (–6.53)

–.33 (–1.92)

EE .55** (3.81)

DP

Structural coefficients of models with job satisfaction predicting

Note: JS represents the latent construct job satisfaction, EE represents emotional exhaustion and DP represents depersonalization. Numbers in the brackets are t-values. **p < .01. Indirect effects (not shown in the figures) are: Nurses: JS → EE → DP (−.58, p < .01); Managers: JS → EE → DP (−.41, p < .01); Technicians: JS → EE → DP (−.49, p < .01).

Path coefficients Next, I report path coefficients of the models tested. The first set of models had job satisfaction as predictor. Inspection of the path coefficients (beta weights) in Figure 6.2a for the nurses sample indicated that, as predicted, job satisfaction had a significant direct negative effect on emotional exhaustion (−.97, p < .01), while emotional exhaustion had a direct positive effect on depersonalization (.60, p < .01). There was also a significant indirect effect (see note below figures) of job satisfaction on depersonalization via exhaustion (−.58, p < .01) for the nurses. However, I did not obtain the predicted direct effect of job satisfaction on depersonalization (−.34, ns). It is interesting to note that when the model developed in the sample of nurses was replicated in the sample of managers and laboratory technicians, the path coefficients for managers and technicians followed the same pattern as those for nurses, as shown in Figure 6.2b and 6.2c: direct effects of job satisfaction on emotional exhaustion (mangers = −.69, p < .01; technicians = −.87, p < .01), a direct effect of emotional exhaustion on depersonalization (managers = .66, p < .01; technicians = .55, p < .01) and an indirect effect (see below figures) of job satisfaction on depersonalization through exhaustion (managers = −.43, p < .01; technicians = −.47). Inspection of path coefficients of the three models with organizational commitment as predictors (Figure 6.3a–c) and supervisor satisfaction as predictors (Figures 6.4a–c) showed similar pattern of coefficients as found in the three models with job satisfaction as predictors. Table 6.3

Thomas J. Kalliath 115 a. Nurses

b. Managers

–.38** (–4.73) OC

EE

–.55** (–4.25) OC EE

.68** (6.30)

–.32** (–3.68)

c. Technicians

.78** (7.90)

–.20 (–1.54)

DP

OC

–.59** (–4.96) EE .79** (6.31)

–17 (–1.27)

DP

DP

Figure 6.3 Structural coefficients of models with organizational commitment predicting burnout Note: OC represents the latent construct organizational commitment, EE represents emotional exhaustion and DP represents depersonalization. Numbers in the brackets are t-values. **p < .01. Indirect effects (not shown in the figures) are: Nurses: JS → EE → DP (−.26, p < .01); Managers: JS → EE → DP (−.43, p < .01); Technicians: JS → EE → DP (−.47, p < .01).

serves as a summary table. It shows that out of 36 effects predicted, 27 were significant and 9 non-significant. It may be noted that all nine non-significant results pertain to the lack of a direct effect of predictors (i.e. job satisfaction, organizational commitment and supervisor satisfaction) on depersonalization.

a. Nurses

b. Managers –.43** (–3.86)

–.29** (–3.61) SS

EE

–.24 (–2.73)

c. Technicians

.70** (6.57) DP

SS

–.53** (–4.56) EE

–.11 (–1.01)

.59** (5.47) DP

SS

EE

–.22 (–1.77)

.67** (5.31)

DP

Figure 6.4 Structural coefficients of models with supervisor satisfaction predicting burnout Note: SS represents the latent construct supervisor support, EE represents emotional exhaustion and DP represents depersonalization. Numbers in the brackets are t-values. **p < .01. Indirect effects (not shown in the figures) are: Nurses: JS → EE → DP (−.20, p < .01); Managers: JS → EE → DP (−.25, p < .01); Technicians: JS → EE → DP (−.36, p < .01).

116 Job Burnout and Dysfunctional Work Attitude Table 6.3

Summary of results

Tested paths

SEM results Nurses

Managers

Technicians

JS→EE EE→DP JS→EE→DP JS→DP

Significant Significant Significant ns

Significant Significant Significant ns

Significant Significant Significant ns

OC→EE EE→DP OC→EE→DP OC→DP

Significant Significant Significant ns

Significant Significant Significant ns

Significant Significant Significant ns

SS→EE EE→DP SS→EE→DP SS→DP

Significant Significant Significant ns

Significant Significant Significant ns

Significant Significant Significant ns

Discussion The findings of the present study confirm that job satisfaction, organizational commitment and supervisor satisfaction are significant predictors of emotional exhaustion in the three samples (nine models) tested. The path coefficients obtained in the present study show that job satisfaction, organizational commitment and supervisor satisfaction had both direct and indirect effects on emotional exhaustion. Taken as a whole, these findings support the argument that, in highly stressful work environments, employees reporting higher job satisfaction, organizational commitment and supervisor satisfaction experience reduced emotional exhaustion. These results contradict the findings of Reilly (1994), who reported that careful examination of beta weights showed that higher frequency of stressors was associated with relatively greater emotional exhaustion for more committed nurses. Further, she states that it is not that more committed nurses are more burned-out per se but that there is a stronger link between the occurrence of stressful incidents and the burnout that they report (p. 408). The present study shows that in spite of being exposed to stressful events in a stressful work environment, nurses reporting higher commitment experienced less emotional exhaustion. These results are also in conformity with Kalliath, O’Driscoll and

Thomas J. Kalliath 117

Gillespie (1998), which showed direct and indirect effect of organizational commitment on burnout. I did not obtain the predicted direct effect of reduced job satisfaction, organizational commitment and supervisor satisfaction on depersonalization among nurses, managers or laboratory technicians. Golembiewski and Munzenrider (1988) have argued that depersonalization is the first manifestation of burnout, followed by reduced personal accomplishment, and then the two together contribute to emotional exhaustion. Although the present study does not include personal accomplishment in the tested model, I obtained clear evidence for direct effect of job satisfaction on emotional exhaustion and indirect effect of job satisfaction on depersonalization through exhaustion in three samples. The same pattern of results was repeated for organizational commitment and supervisor satisfaction as predictors. The fact that I did not obtain a direct effect of job satisfaction (or organizational commitment or supervisor satisfaction) on depersonalization in any of the three samples suggests that the major component in burnout is emotional exhaustion, a finding that supports the dominant view in the literature – that of the primacy of emotional exhaustion in burnout (Evans and Fischer, 1993; Lee and Ashforth, 1996). Further, Leiter and Maslach (1988) have argued that the relationship between emotional exhaustion and depersonalization is sequential. Human service workers experience emotional exhaustion first in response to a stressful work environment. As they become overwhelmed by the work demands, they tend to develop a depersonalized attitude towards their service recipients, distancing themselves emotionally from clients and treating them in a callous way. Leiter and Maslach (1988) argued that continued emotional exhaustion leads to depersonalization. The present results provide indirect support to Leiter and Maslach’s (1988) arguments. The lack of a direct effect of all three predictors on depersonalization suggests that those workers who are depersonalized may have become too insensitive and hardened for these to make any difference in their work attitudes. The relative efficacy of measures to increase job satisfaction, commitment, and supervisor support may work only in the early stages of the burnout process, but may not be adequate once a human service worker has reached the state of depersonalization. Indeed, depersonalized workers have become so alienated from their jobs, their clients, their co-workers and their organization that they cannot experience or report high measures of these work attitudes. This finding suggests the need for early action to alleviate/prevent job burnout

118 Job Burnout and Dysfunctional Work Attitude

before human service workers are too far gone down the path of becoming depersonalized. It is noteworthy that the effect of job satisfaction on emotional exhaustion was stronger for the three groups (nurses, managers, and technicians) than the effect of organizational commitment or supervisor satisfaction. This result suggests that among the three work attitudes examined in the present study, the link between job satisfaction and emotional exhaustion is stronger in all three groups than the link between organizational commitment and supervisor satisfaction with emotional exhaustion. In a strategy for combating job burnout, clearly efforts to increase job satisfaction need to be given priority. The present study had several limitations. First, all three samples were taken from the same hospital setting. Although this strengthens the comparative aspect of the design by ruling out organizational goals, structure, and other potential confounds, it also restricts the generalizability of our results. Second, data from the three samples were gathered at the same time, which limits the ability to test causal assumptions regarding the burnout syndrome (MacCallum et al., 1993). Third, the present analyses were carried out to investigate the effect of three commonly held work attitudes on job burnout, so the effect of other work attitudes were excluded from the analysis. Finally, Bandura (1986) has argued that environmental influences and individual behaviour operate interactively to determine each other. He uses the term ‘reciprocal determinism’ to describe the production of effects by certain factors. By restricting the investigation of work attitudes on job burnout in one direction only, the present study may have over-simplified the dynamics of human interaction. Bandura’s approach supported by social cognitive theory recommends the examination of reciprocal influences of factors on each other. Future studies may address these concerns.

Conclusion Although it is well established in the literature that job burnout predicts significant negative consequences such as dysfunctional work attitudes among human service employees, what is not established conclusively in the literature is the converse hypothesis – the benign consequences of higher levels of positive work attitudes on experienced job burnout. In the present chapter, I described the job burnout construct, and presented the relationship of job burnout with three antecedent variables, namely job satisfaction, organizational commitment and supervisor satisfaction. Then I presented some limited evidence from the literature

Thomas J. Kalliath 119

for the converse hypotheses – the effect of positive work attitudes on job burnout experiences. I hypothesized that higher levels of positive work attitudes (e.g. job satisfaction, organizational commitment, and supervisor satisfaction) would predict lower levels of job burnout in three samples, including nurses, managers, and technicians. The results supported the hypotheses that job satisfaction, organizational commitment and supervisor satisfaction were significant predictors of lower levels of emotional exhaustion, and personalization (via exhaustion). The findings of the present study have practical implications for organizational efforts directed at preventing dysfunctional work attitudes among nurses, managers and laboratory technicians. Organizations need to take into account the salutary effects of job satisfaction, organizational commitment and supervisor satisfaction on job burnout. Higher levels of these three work attitudes predicted lower levels of job burnout among managers, nurses, and technicians. This finding suggests that promoting higher levels of these three work attitudes may work as an antidote to job burnout. Moreover, the strongest relationship among the three predictors was that between job satisfaction and job burnout. Hence, a strategy for combating job burnout could well begin with organizations paying careful attention to improving job satisfaction of employees.

References Anderson, J. C. and Gerbing, D. W. (1988) ‘Structural equation modeling in practice: a review and recommended two-step approach’, Psychological Bulletin, 103, 411–23. Bandura, A. (1986) Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall. Brookings, J., Bolton, B., Brown, C. and McEvoy, A. (1985) ‘Self-reported job burnout among female human service professionals’, Journal of Occupational Behavior, 6, 143–50. Burke, R. J. (1987) ‘Burnout in police work: an examination of the Cherniss model’, Group and Organization Studies, 12, 174–88. Burke, R. J. and Greenglass, E. R. (1989) ‘Psychological burnout among men and women in teaching: an examination of the Cherniss model’, Human Relations, 42, 261–73. Burke, R. J., Shearer, J. and Deszca, G. (1984) ‘Burnout among men and women in police work: an examination of the Cherniss model’, Journal of Health and Human Resources Administration, 7, 162–88. Cammann, C., Fichman, M., Jenkins, G. D. Jr. and Klesh, J. R. (1983) ‘Assessing the attitude of organizational members’, in S. E. Seashore, E. E. Lawler III, P. H. Mirvis and C. Cammann (eds), Assessing Organizational Change (pp. 71–138). New York: Wiley.

120 Job Burnout and Dysfunctional Work Attitude Cherniss, C. (1980) Professional Burnout in Human Service Organizations. New York: Praeger. Constable, J. and Russell, D. (1986) ‘The effects of social support and the work environment upon burnout among nurses’, Journal of Human Stress, 12, 20–6. Cordes, C. and Dougherty, T. (1993) ‘A review and integration of research on job burnout’, Academy of Management Review, 18, 621–56. Dolan, N. (1987) ‘The relationship between burnout and job satisfaction in nurses’, Journal of Advanced Nursing, 12, 3–12. Dunham, R. B., Smith, F. J. and Blackburn, R. S. (1977) ‘Validation of the Index of Organizational Reactions with the JDI, the MSQ, and Faces Scales’, Academy of Management Journal, 20(3), 420–32. Evans, B. and Fischer, D. (1993) ‘The nature of burnout: a study of the three-factor model of burnout in human service and non-human service samples’, Journal of Occupational and Organizational Psychology, 66, 29–38. Golembiewski, R. T. and Munzenrider, R. (1988) Phases of Burnout: Developments in Concepts and Applications. New York: Praeger. Golembiewski, R. T., Hilles, R. and Rick, D. (1987) ‘Some effects of multiple OD interventions on burnout and work site features’, Journal of Applied Behavioural Science, 23, 295–313. International Labour Organization (1992) Conditions of Work Digest: Preventing Stress at Work (ed. V. Di Martino). Geneva: ILO. Jackson, S., Turner, J. and Brief, A. (1987) ‘Correlates of burnout among public service lawyers’, Journal of Occupational Behaviour, 8, 339–49. Jayaratne, S. and Chess, W. A. (1983) ‘Job satisfaction and burnout in social work’, in B. A. Farber (ed.), Stress and Burnout in the Human Service Professions. Elmsford, NY: Pergamon Press. Jayaratne, S., Chess, W. A. and Kunkel, D. A. (1986) ‘Burnout: its impact on child welfare workers and their spouses’, Social Work, 31, 53–9. Justice, B., Gold, R. S. and Klein, J. P. (1981) ‘Life events and burnout’, Journal of Psychology, 108, 219–26. Kahill, S. (1988) ‘Symptoms of professional burnout: a review of empirical evidence’, Canadian Psychology, 29, 284–97. Kalliath, T., O’Driscoll, M. and Gillespie, D. (1998) ‘The relationship between burnout and organizational commitment: a structural equations analysis in two samples of health professionals’, Work and Stress, 12, 179–85. Kalliath, T., Gillespie, D., O’Driscoll, M. and Bluedorn, A. (2000) ‘A test of the Maslach Burnout Inventory in three samples of healthcare professionals’, Work and Stress, 14(1), 35–50. Katzell, R. A., Thompson D. E. and Guzzo, R. A. (1992) ‘How job satisfaction and job performance are and are not linked’, in C. J. Cranny, C. P. Smith and E. F. Stone (eds), Job Satisfaction: How People Feel About Their Jobs and How It Affects Their Performance. New York: Lexington Books. Lee, R. and Ashforth, B. (1990) ‘On the meaning of Maslach’s three dimensions of burnout’, Journal of Applied Psychology, 75, 743–7. Lee, R. and Ashforth, B. (1996) ‘A meta-analytic examination of the correlates of the three dimensions of job burnout’, Journal of Applied Psychology, 81, 123–33. Leiter, M. P. (1988a) ‘Burnout as a function of communication patterns. A study of multidisciplinary mental health team’, Group and Organization Studies, 13, 111–28.

Thomas J. Kalliath 121 Leiter, M. P. (1988b) ‘Commitment as a function of stress reactions among nurses: A model of psychological evaluations of work settings’, Canadian Journal of Community Mental Health, 7, 115–32. Leiter, M. P. (1990) ‘The impact of family resources, control coping, and skill utilisation on the development of burnout: a longitudinal study’, Human Relations, 43, 1067–83. Leiter, M. P. (1991) ‘Coping patterns as predictors of burnout: the function of control and escapist coping’, Journal of Occupational Behaviour, 12, 123–44. Leiter, M. (1993) ‘Burnout as a developmental process: consideration of models’, in W. Schaufeli, C. Maslach and T. Marek (eds), Professional Burnout: Recent Developments in Theory and Research. Washington, DC: Taylor & Francis. Leiter, M. and Maslach, C. (1988) ‘The impact of interpersonal environment on burnout and organizational commitment’, Journal of Organizational Behavior, 9, 297–308. MacCallum, R., Wegener, D., Uchino, B. and Fabrigar, L. (1993) ‘The problem of equivalent models in applications of covariance structure analysis’, Psychological Bulletin, 114, 185–99. Maslach, C. (1978) ‘Job burnout: how people cope’, Public Welfare, 36, 56–8. Maslach, C. (1982) ‘Burnout: a social psychological analysis’, in J. W. Jones (ed.), The Burnout Syndrome: Current Research, Theory, Interventions (pp. 30–53). Parke Ridge, IL: London House Press. Maslach, C. and Jackson, S. (1981) The Maslach Burnout Inventory (Research Edition). Palo Alto, CA: Consulting Psychologists Press. Maslach, C. and Jackson, S. (1984) ‘Patterns of burnout among a national sample of public contact workers’, Journal of Health and Human Resources Adminstration, 7, 189–212. Maslach, C. and Jackson, S. (1986) Maslach Burnout Inventory Manual (2nd edn). Palo Alto, CA: Consulting Psychologists Press. Mowday, R., Porter, L. and Steers, R. (1982) Employee–Organisation Linkages: The Psychology of Commitment, Absenteeism, and Turnover. New York: Academic Press. Mowday, R., Steers, R. and Porter, L. (1979) ‘The measurement of organizational commitment’, Journal of Vocational Behavior, 14, 224–47. Pines, A. and Maslach, C. (1978) ‘Characteristics of staff burnout in mental health settings’, Hospital and Community Psychiatry, 29, 233–7. Pines, A. Kafry, D. and Etzion, D. (1980) ‘Job stress from a cross cultural perspective’, in K. Reid (ed.), Burnout and the Helping Professions. Kalamazoo, MI: M.U. Press. Pines, A. and Kafry, D. (1981) ‘Tedium in the life and work of professional women as compared with men’, Sex Roles, 7, 117–34. Quinn, R. P. and Staines, G. L. (1979) The 1977 Quality of Employment Survey. Ann Arbor, MI: Institute of Social Research, University of Michigan. Reilly, N. (1994) ‘Exploring a paradox: commitment as a moderator of the stress– burnout relationship’, Journal of Applied Social Psychology, 24, 397–414. Ross, R. R., Altmaier, E. M. and Russell, D. W. (1989) ‘Job stress, social support, and burnout among counseling center staff’, Journal of Counseling Psychology, 36, 464–70. Wolpin, J., Burke, R. J. and Greenglass, E. R. (1991) ‘Is job satisfaction an antecedent or a consequence of psychological burnout?’, Human Relations, 44, 193–209.

7 Flexible Work Schedules and Their Impact on Employees Lonnie Golden

Introduction The timing of employees’ work hours and the degree to which they are able to exercise control over it are increasingly important determinants of workers’ motivation, behaviour and effort at work. More sectors in the global economy are moving toward operating on a round-the-clock service or continuous production. At the same time, dual-income households and annual and weekly work hours (in the US) are climbing. These forces have contributed to a spreading out of the workday, and thus a growing value being attached to the temporal flexibility in work schedules and the timing of work activities (e.g. Fagan, 2001; Hamermesh, 1999; Presser, 1995). Flexible schedules are perceived to help reduce the chronic pressures imposed on workers by time and role conflicts that arise when work and non-work responsibilities overlap in the same block of time. Flexible work schedules are being implemented by organizations in certain sectors as either or both a prized employee benefit to promote employee retention and as a device to curb forms of employee withdrawal, such as absenteeism, tardiness, quits and on-thejob leisure. Flexibility is thus becoming an ever more important tool for improving individual, organizational and national productivity, including indirectly by attempting to curb various employee misbehaviours. Flexible schedules are those that permit employees some discretion to alter the daily starting or ending times of their workday. From an employee’s standpoint, flexibility generally refers to getting an immediate and proportional adjustment of their schedule toward their desired schedule. Employees may be taking up a formally offered ‘flextime’ 122

Lonnie Golden 123

(written also as ‘flexitime’) programme, where they must be present for ‘core’ hours but are able to vary their own arrival and departure times as long as they work a required number of total hours per day or week. Others arrange flexible start and leave times informally with supervisors. The chapter consists of three discussions: (1) the potential and documented effects of temporal flexibility in work schedules on employees’ well-being and satisfaction, productivity, and indicators of negative behaviour at work, such as absenteeism and turnover; (2) the distribution of flexible schedules by workers’ demographic characteristics and job characteristics such as their industry and occupation; (3) the links between the allocation of flexible schedules and the underlying reasons employers would choose to offer them to particular occupations.

Flexible work schedules, employee effort, satisfaction and behaviour at work The effects of flexible work schedules on individual and organizational productivity Why should an employer offer individualized work schedules or formal flexitime programmes? After all, this will entail additional managerial cost for the organization due to the administrative burden of preventing coordination failures (e.g. ensuring no gaps in coverage servicing customers or clients) and improving communication structures (e.g. meetings), particularly in the adoption phase. The ‘flexible firm’ literature suggests employers increasingly seek internal organizational and work process changes that allow them to more quickly adapt labour input, including the length and timing of work and operating hours, to improve either efficiency, capital utilization or sales revenues. The decision of a firm to adopt more flexible schedules for employees boils down to whether the cost of implementing flexible scheduling is exceeded by the cost of not implementing it represented by their potential benefits in the longer run (see Golden, 1996). First, the additional cost is often negligible (Hohl, 1996, p. 83). Second, there is ample evidence that flexible schedules help to promote higher productivity either directly via increased work effort or time on the job, offsetting the administrative costs. More often, the improved productivity occurs cumulatively and indirectly, by fostering morale and longer-term commitment to the employer and by reducing employee stress, role conflict and ‘on-the-job’ leisure. Employers also benefit because providing schedule flexibility permits employees to work longer hours and suffer less negative impact on their attempts to

124 Flexible Work Schedules

balance work and non-work commitments (Hill et al., 2001). While this may lead to greater productivity per worker, the evidence regarding greater productivity per hour, however, is not as clear. Nevertheless, management interviews confirm higher productivity per hour (e.g. Kim and Campagna, 1981; Leonard, 1999). Moreover, flexible work schedules are found empirically to raise one industry’s productivity, pharmaceuticals, by about 10 per cent (Shephard, Clifton and Kruse, 1996). Any positive effect of flexible schedules on productivity tends to be context sensitive. For example, it appears to be greater for professional than for non-professional jobs ( Johnson and Provan, 1995), or when employees are sharing a computer system or inputs (Ralston, Anthony and Gustafson, 1985), Flexible schedules better promote productivity gains when they involve slight rather sweeping changes, and the gains themselves are short-term rather than long-lasting (Baltes, et al. 1999). In addition, flexible schedules deliver productivity improvement for some demographic groups more than others (Bardoel et al., 1999), when workers are interdependent or seek personal growth (Krausz and Hermann, 1991), or only when facilitating greater work from home (Meyer et al., 2001). Nevertheless, employees on flexitime are usually found to be no less productive than their comparable counterparts on traditional or fixed work schedules (Ahmadi et al., 1986; Hammer and Barbera, 1997; Kopelman, 1986; Kossek et al., 1999). Moreover, workers on flexitime schedules self-report that this makes them relatively more productive. In a survey of job-seeking professionals, 65 per cent agree that their productivity would be greater working on a flexible rather than ‘traditional’ schedule, twice the rate that said their productivity would remain the same (Flexible Resources, 2001). Flexible schedules, however, are more apt to affect productivity or costs indirectly. Greater organizational commitment on the part of employees (Scandura and Lankau, 1997) suggests that organizations offering more flexible schedules gain in the longer run through longer average job tenure and increased rates of retention (McShulskis, 1997). The presence of autonomy and flexibility in scheduling is expected to promote longer attachments to employers, if employees view it as an employee benefit that they are reluctant to sacrifice. Indeed, Almer and Kaplan (2002) found significantly lower rates of turnover among accountants relative to a control group without flexible scheduling. Dalton and Mesch (1990), however, found scant evidence that flexible schedules tend to thwart turnover. In sum, organizations may adopt more flexible scheduling options if they view it as a worthwhile investment in human resources that will curb long-run costs via promoting

Lonnie Golden 125

worker effort, particularly in the context of labour shortages and increasingly expensive capital.

Effect of flexible schedules on individuals’ well-being, morale and satisfaction The daily timing of work and time-off, in addition to the volume of non-work time, may have a profound impact on an individual’s quality of work-life, job satisfaction and general well-being. The daily and weekly scheduling of their shop, office, school, class or store hours are often outside the direct control of the individual. To the extent that scheduling of work interferes or conflicts with workers’ ability to execute their non-work responsibilities, the scheduling of a given number of hours of work itself is crucial. Autonomy and the ease with which work scheduling allows individuals to transition between work and non-work activities often may be a highly valued feature of a job. The rescheduling of work may reduce commuting times, facilitate attendance in college courses, permit the coordinated sequencing of production in the household or inhibit social interaction with friends or family. Moreover, the ability to adjust work schedules flexibly when non-work responsibilities change is surely a vital component of workers’ job satisfaction. The flipside, of course, is that schedule inflexibility may lead to dissatisfaction and additional risk of job stress and misbehaviours to alleviate these. Worker satisfaction with the job, work environment or work schedule itself is found to be higher if the schedules are flexible (Baltes et al., 1999; Christensen and Staines, 1990; Deery and Mahony, 1994; Dunham, Pierce and Castenda, 1987; Krausz and Hermann, 1991; McGuire and Liro, 1987). Worker satisfaction with their job, at least their satisfaction with work schedules, is found to improve if flexible schedules reduce the closeness of supervision (Buckley, Kicza and Crane, 1987). Women managers generally, and in particular those with family responsibilities, exhibited greater job satisfaction if their organizations were perceived to offer flexible work hours such as flexitime (Scandura and Lankau, 1997). Workers’ control over scheduling, independent of shift times, contributes to their general health and psychological well-being (Fenwick and Tausig, 2001; Krausz, Sagie and Bidermann, 2000). Federal government employees using flexitime reported higher satisfaction with both their jobs and their work–family balance (Ezra and Deckman, 1996). Almer and Kaplan (2002) found a significant improvement in job satisfaction among accountants when granted or switching to more flexible work arrangements. Sparks, Faragher and Cooper (2001) also find some evidence in support among their respective samples. No matter what

126 Flexible Work Schedules

the level of work hours, employees whose work schedules were different from what their partner preferred, were more disengaged, distracted, and alienated at work than were their counterparts who were working their preferred schedules, leading to relatively higher risks of burnout or work–family conflict (Barnett, Gareis and Brennan, 1999; Clarkberg, 2000; Fagan, 2001). However, the positive effects of flexitime, particularly for job satisfaction, tend to be either not long-lasting (Baltes et al., 1999) or may be offset by resulting dissatisfaction due to inflexibility of nonwork (home) obligations (Krausz and Freibach, 1983).

Effects of flexible schedules on absences and tardiness Rates of absenteeism are one indicator of employee commitment that affects organizational productivity. Absence rates in the US average 3.6 per cent, for 2001. That is, in a given week, about three and a half of every one hundred employees who usually work full time will work less than a full week, due mainly to own illness or injury, other medical problems, child care or personal obligations, or family leave (Bureau of Labor Statistics, 2002, Employment and Earnings, table 46). The rate is remarkably constant across age groups, though slightly higher among workers aged 55 and over. The average is considerably lower among men, at about 2.5 per cent (higher for older-aged men), and about double the rate among women, at 4.8 per cent. By occupational groups, it is over 4 per cent among clerical workers, service occupations, machine operatives/labourers and technicians. The rate is under 3 per cent among managers and administrators. Lost hours of work time is steepest among operatives/labourers and least among managers, professionals and sales occupations. There is little variation among industries, although the absence rate is highest in public sector jobs and lower in construction. In terms of lost work time, the high amount among government workers is matched by workers employed in transportation and nondurables manufacturing. In theory, a greater ability to vary or delay start times should discourage absenteeism, or at least tardiness. Dalton and Mesch (1990) exhaustively reviewed the evidence regarding the effect of flexible scheduling on absence rates and found it ambiguous, but in their own study found a significant reduction. An inverse association between absences due to sick time and employee control over working time has been identified elsewhere, as well. Organizations using flexitime have experienced reduced absenteeism and use of sick time (Ahmadi et al., 1986; Allen, 1987; Baltes et al., 1999; Dalton and Mesch, 1990; Hammer and Barbera, 1997; Harrick, Vanek and Michlitsch, 1986; Kim and Campagna, 1981;

Lonnie Golden 127

Kopelman, 1986; Kraus and Freibach, 1983; Moss and Curtis, 1985; Ralston and Flanagan, 1985). In addition, it appears to decrease the incidence of tardiness (Moss and Curtis, 1985; Ralston, 1989). Baltes et al. (1999) found lower rates of absences in three-quarters of studies examined. However, such a reduction appears to be significant only for women employees (Ala-Mursula et al., 2002).

The level and distribution of flexible schedules among workers US national data show that the proportion of workers reporting an ability to alter their daily starting and ending times of work has increased dramatically since the 1980s, to over 28 per cent of the workforce in 2001, although it is only slightly greater than in 1997 (see US BLS, 1998, 2002, reporting the results of the May Current Population Survey supplement sample that consists of 50,000 households, including 88,728 individuals). About 11 per cent have a formal flexitime arrangement (US BLS, 2002). Despite the recent growth in availability of flexible schedules, there is ample evidence that it still falls well short of worker demand for such schedules. Workers in the US do not yet face a full continuum of work schedule options (Clarkberg, 2000). Far greater proportions of employers claim to offer flexible work schedules than the proportion of employees reporting having them (see Glass, 1995; Greenwald, 1998; Hein, 1999; Kush and Stroh, 1994; McShulskis 1997), with working parents and dual-income householders in particular desiring greater access (HR Focus, 1998; Kate, 1998). Employees often are uncomfortable using or asking for flexitime, fearing they will damage their career prospects (Galinsky and Bond, 1998; US Department of Labor, 1999). Nevertheless, to attain more flexibility in the timing of their daily work hours, as much as 25 per cent of the workforce would be willing to sacrifice career prospects (Galinsky and Bond, 1998). Moreover, there remains much disparity in employee access to work schedule flexibility by demographic and job characteristics (Golden, 2001; Sharpe et al., 2002). Some statistics relevant to this disparity are presented in Tables 7.1 to 7.4. Differences by demographic group are not large, but do exist. Men actually have greater access to flexible schedules than women, except among teens (consistent with the findings of McCrate, 2002). There is slightly greater access to flexible schedules among the married and those with young children, but also greater for men. Non-whites have significantly less access than do whites. Those enrolled in college get somewhat greater access, but generally age is positively associated with being granted schedule flexibility.

128 Flexible Work Schedules Table 7.1 Proportions of full-time wage and salary workers on flexible and shift schedules, in May, 1985–97: selected years and demographic groups (percent) Flexible schedules May 1985

May 1989

May 1991

May 1997

Total

12.4

11.9

15.1

27.6

Gender: Men Women

13.1 11.3

12.9 10.6

15.5 14.5

28.7 26.2

Race: White Black Hispanic

12.8 9.1 8.9

12.5 8.2 9.0

15.5 12.1 10.6

28.7 20.1 18.4

May, 1997: Age 16 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 64 55 years and over 20 years +

Both 20.7 22.7 28.4 29.1 27.3 27.4 27.8 27.8

Men 16.9 22.4 28.7 30.6 28.9 30.1 31.1 29.0

Women 27.4 23.2 27.9 26.9 25.4 23.8 23.4 26.2

Marital status Never married Married, spouse present Other status

Both 25.4 28.8 26.8

Men 24.9 30.8 25.8

Women 26.1 25.6 27.6

Presence and age of children With own children < 18 With own children 6 to 17 With own children < 6 Without children < 18

Both 28.9 27.9 30.2 26.8

Men 30.5 29.7 31.5 27.5

Women 26.6 25.8 27.8 26.0

The data suggest that employers’ use of flexible scheduling is not equivalent across occupations. Among occupations, the percentage of workers with a flexible schedule ranges widely, from over 60 per cent for natural scientists, down to 11 per cent for machine operators and tenders. However, if a worker’s major industry location is accounted for, only managerial and administrative, sales and handler/labourer and, to a slight extent, technician occupations, get greater access to flexible schedules. Professional and service occupations have greater flexibility in schedules, but this is because their concentration in certain industries offers such flexibility. Craft occupations appear to get significantly less

129 Table 7.2

Flexitime distribution by major industries and occupations: total

Major industry (22 in total)

Agriculture Mining Construction Manfg Manfg Transportation Communication Public Wholesale Retail Finance Private Business Personal Entertainment Hospitals Medical Educational Social Other Forestry/fishing Public

durable non-durable

utilities trade trade insurance households automobile services recreation services services services professional

goods goods

real estate services repair

service

administration

Percent of all flextime schedules

Percent of work force on flexitime

5.2 0.5 7.3 6.7 4.5 3.7 1.2 0.8 4.4 17.9 7.5 1.1 7.4 3.3 2.3 2.8 4.2 4.9 2.5 7.3 0.4 4.2

57.0 24.1 34.9 23.9 22.7 26.1 31.3 22.2 36.7 32.5 39.0 41.7 38.0 37.5 35.4 22.8 28.2 19.3 30.5 49.4 63.9 29.8

100.0 Major occupations (13 in total) Managerial Professional Technicians Sales Admin support, clerical Private household Protective service Other service Craft Operators Transportation operatives Labourers Farming

service

19.8 17.0 3.0 15.7 11.9

46.0 37.1 31.1 41.1 26.8

0.9 0.9 9.0 8.7 2.2 3.1 2.2 5.8

43.3 16.2 24.3 25.6 12.1 23.5 16.9 53.4

100.0

130 Flexible Work Schedules Table 7.3 Detailed occupations with greatest percentage of workers with flexible schedules: rank Rank

Occupation

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Farm Operators Natural Scientists Lawyers and Judges Sales Reps, Finance, Bus. Service Mathematical Scientists Teachers, College and University Forestry Occupations Other Professional Sales Reps, Commodities, Exc. Retail Engineers Managers Sales Supervisors and Proprietors Sales Related Occupations Other Technicians Financial Records, Processing Private Household Service Health Diagnosing Occupations Management Related Public Administrator

% with flex 77.9 60.2 58.6 58.1 55.9 54.6 53.8 50.3 49.8 47.9 47.9 45.7 44.4 44.0 43.5 43.3 42.6 41.2 41.0

access to such flexibility. Finally, being a union member tends to improve worker access to flexible scheduling, although not always significantly so, in contrast to Glass (1995) who had found that unions inhibited flexitime perhaps because unionized occupations tend to have fixed schedules. There is far less differential access among broad industry categories in which workers are employed, but some disparity between the specific (‘detailed’) industries. Within manufacturing, tobacco, primary metals, professional equipment and miscellaneous manufacturing exhibit double the share of workers in textile and leather industries. Public administration exceeds the rate found in hospitals, entertainment– recreation, utilities and other professional services. Thus, employers using flexible scheduling appear to be targeting certain types of employees, perhaps perceived as most responsive in terms of promoting either greater commitment or labour productivity. Flexible schedules are clearly not geared toward workers who are working a standard day shift or 40-hour work-week. Indeed, they are made available relatively more to part-time workers and those putting in very long hours – greater than 50 per week. Indeed, having a flexible schedule is

Table 7.4 Likelihood of having flexible start- and end-times, probit estimates: marginal effects of personal, work and hours-status characteristics Controls:

Variable

None

Coeff

Personal characteristics: Age 0.085 −0.001 Age squared −0.212 PhD Master 0.428 College 0.227 −0.187 Associates −0.062 Some College −0.261 HS −0.660 Less than HS −0.601 Nonwhite −0.237 Female Married 0.125 College student 0.198 Job status: Self-employed Union member Hours status: Usually full-time Usually part-time

44.01 −45.70 −3.11 12.73 8.42 −5.39 −2.31 −10.08 −22.96 −37.97 −23.45 10.94 5.54 64.96 1.94

Coeff

z-stat

FT or PT

Coeff

z-stat

Long hours

Coeff

z-stat

Standard hours Non standard hours Coeff

z-stat

Coeff

z-stat

Shifts

Coeff

z-stat

Hours vary

Coeff

z-stat

−0.008 −2.79 0.012 3.98 −0.001 −0.19 0.002 0.60 0.002 0.60 −0.001 −0.25 −0.011 −0.25 0.000 11.67 0.000 3.58 0.000 8.61 0.000 7.59 0.000 7.50 0.000 8.65 0.000 11.93 0.513 5.45 0.504 5.33 0.488 5.16 0.521 5.49 0.421 4.44 0.466 4.94 0.555 5.81 0.624 15.53 0.619 15.30 0.618 15.21 0.683 16.70 0.511 12.30 0.643 16.02 0.726 17.61 0.469 15.04 0.448 14.22 0.433 13.73 0.439 13.96 0.398 12.35 0.411 13.12 0.482 15.07 0.125 3.13 0.144 3.59 0.148 3.69 0.182 4.54 0.098 2.41 0.137 3.42 0.228 5.61 0.165 5.26 0.111 3.51 0.137 4.34 0.156 4.92 0.073 2.24 0.152 4.84 0.200 6.22 0.115 3.76 0.069 2.23 0.093 3.03 0.121 3.91 0.033 1.05 0.049 1.59 0.127 4.04 −0.144 −4.09 −0.269 −7.51 −0.201 −5.63 −0.177 −4.94 −0.282 −7.75 −0.243 −6.82 −0.146 −4.01 −0.462 −25.41 −0.585 −30.91 −0.549 −29.25 −0.551 −29.29 −0.575 −30.54 −0.528 −27.98 −0.596 −31.21 −0.036 −2.62 −0.100 −7.20 −0.071 −5.13 −0.086 −6.15 −0.059 −4.20 −0.109 −7.84 −0.042 −2.97 0.093 6.78 0.108 7.71 0.077 5.56 0.075 5.37 0.093 6.67 0.106 7.61 0.079 5.64 0.504 11.33 0.278 6.11 0.382 8.49 0.045 0.00 0.357 7.91 0.354 7.78 0.393 8.71 1.075 0.056

43.10 1.42

2.67

1.009 0.037

39.47 0.93

1.013 0.071

39.65 1.77

1.012 0.026

39.45 0.65

1.111 0.068

43.80 1.71

−0.479 −28.50 0.904 21.57

1.202

29.73

1.113

27.24

1.160

28.60

1.147

28.46

0.181

10.95

0.107

0.078 4.39 0.083 4.82 −0.221 −15.35 −0.318 −15.42

0.089

2.19

0.660

15.04

0.155

8.58

−0.145

−9.85

131

Hours 50 + Hours 41–49 Hours 40

1.498 0.073

z-stat

Occupation

(Continued)

Controls:

Variable

132

Table 7.4

None

Coeff

z-stat

Occupation

Coeff

z-stat

FT or PT

Coeff

z-stat

Long hours

Coeff

z-stat

Standard hours Non standard hours Coeff

z-stat

Shift: Standard day shift Evening Night Irregular

Constant PseudoR2 = n=

z-stat

Coeff

z-stat

Hours vary

Coeff

z-stat

0.680

27.60

−0.333 −12.68

Hours 35–39 Hours vary

Occupation: Managerial Professional Technicians Sales Admin support, Clerical Protective service Other service Craft Operators Transportation Labourers

Coeff

Shifts

−0.294 −19.70 −0.245 −16.02 −0.275 −18.37 0.155 4.66 −0.363 −6.79 0.830 30.26 0.674 0.467 0.592 0.708 0.122

4.30 2.95 3.59 4.53 0.78

0.741 0.520 0.615 0.702 0.078

4.78 3.32 3.76 4.54 0.50

0.758 0.570 0.648 0.749 0.199

4.87 3.63 3.96 4.83 1.28

0.746 0.562 0.666 0.707 0.195

4.78 3.57 4.05 4.54 1.25

0.780 0.579 0.619 0.761 0.163

5.04 3.71 3.80 4.93 1.05

0.680 0.515 0.565 0.712 0.132

4.35 3.26 3.44 4.57 0.84

0.635 0.428 0.555 0.601 0.050

4.04 2.69 3.35 3.83 0.32

0.219 0.261 −0.278 0.160 0.036 0.834

1.39 0.067 1.67 0.259 −1.75 −0.235 1.01 0.190 0.22 −0.015 5.23 0.799

0.43 0.160 1.67 0.340 −1.49 −0.230 1.21 0.153 −0.10 0.058 5.06 0.861

1.02 0.113 2.18 0.341 −1.45 −0.209 0.97 0.136 0.37 0.049 5.43 0.829

0.72 0.143 2.18 0.309 −1.32 −0.254 0.86 0.149 0.31 0.031 5.21 0.840

0.92 2.00 −1.62 0.95 0.20 5.33

0.172 1.09 0.294 1.88 −0.230 −1.45 0.099 0.62 0.040 0.25 0.839 5.27

0.112 0.213 −0.285 0.037 −0.025 0.654

0.70 1.35 −1.79 0.23 −0.16 4.08

−2.180 −43.79 −1.319 0.186 0.185 88,728 56,982

−7.85 −1.212 0.208 56,982

−7.27 −1.274 0.198 56,982

−7.62 −1.239 0.207 56,982

−7.39 −1.194 0.208 56,982

−7.17 −1.449 −8.62 0.218 56982

−1.183 0.214 56982

−7.00

Dependent variable=1 if able to vary start or end times of work.

Lonnie Golden 133

positively correlated with workers’ usual work-week length (ρ = 0.55). This positive association is likely not a statistical artifact produced by workers on flexible times tending to systematically exaggerate their own reported average work time (see Jacobs, 1998). However, it is possible that flexible daily schedules are needed psychologically to facilitate long workdays, such as ‘night owls’ working late knowing it is acceptable to arrive late the following day. Similarly, employers may be able to induce greater duration of work by permitting flexible starting and/or ending times of the workday. Workers wishing to gain greater access to flexible daily schedules on their own thus must endure either foregone leisure time, or transition to part-time status, which tend to be associated with current and long-term reductions in income (see Ferber and Waldfogel, 1998)

Interpretation of results and conclusions Workers generally express a growing need for more individualized working time, to set or influence their own work schedules according to their own natural work rhythms and constraints. More flexible scheduling of work addresses workers more as whole persons seeking work–life balance and satisfaction through their multiple roles (Bailyn, Drago and Kochan, 2001). While such individualization may burden organizations with additional scheduling coordination challenges, there is mounting evidence that employers are being paid back in some form in the longer run, via either improved job and/or work–life satisfaction that reduces various employee misbehaviours, greater retention rates or, to some extent, higher labour productivity. The distribution of flexible daily schedules in the US workforce suggests that certain workers are able to obtain greater control over the timing of their daily hours. It is clear is that having a flexible schedule has at least as much if not more to do with a worker’s particular job and occupation than with their personal characteristics. It is determined by many characteristics other than those often thought to be associated with a greater demand for schedule flexibility, such as being an employed woman. Some jobs appear to be targeted more than others. Perhaps such jobs are the lowest hanging fruit, where flexible scheduling is less costly to implement and administer. Alternatively, the occupants of such jobs may be more likely to exhibit greater responsiveness, such as increased retention or productivity. It is not directly observable here whether the use of flexible schedules by organizations is primarily as an employee benefit or as an instrument to enhance organizational productivity. The fact that flexible schedules grew across a broad spectrum of sectors in

134 Flexible Work Schedules

the 1990s in the US suggests that organizational concern with or economic payoff for either employee retention or productivity increased. The fact that the distribution of flexibility remains quite uneven across workers, being relatively less available for non-whites, women, unmarried (and relatively less-educated and public sector) employees, all else constant, reflects either exclusivity in the distribution of rewards or an occupational segregation of workers with such characteristics into jobs that inherently receive less flexible scheduling. Certain occupations (managerial, professional, technical, sales, labourer/helper/cleaner and handler jobs and perhaps service (except protective and household) receive more flexibility than other occupations. These are presumably the occupations where retention, worker morale, withdrawal and effort are a key concern for employers, and where flexible schedules are presumed to remedy these problems. Aggregate data on absences suggest that the elevated presence of flexible schedules among managers and administrators may successfully promote attendance. Among technical workers, which exhibit relatively higher rates of absenteeism, flexible schedules appear to be used more as an employee benefit and retention device than as a way to encourage attendance or effort. The greatest potential for improvement may lie among operatives, who currently experience low flexibility of work schedule and higher absenteeism. Finally, there is clearly room to gain reductions in absenteeism by extending flexible schedules to more employed women.

References Ahmadi, M., Raiszedeh, F. and Wells, W. (1986) ‘Traditional vs. non-traditional work schedules: a case study of employee preference’, Industrial Management, 28(2), 20–3. Ala-Mursula, L., Vahtera, J., Kivim, M., Kevin, M. and Pentti, J. (2002) ‘Employee control over working times: associations with subjective health and sickness absences’, Journal of Epidemiology and Community Health, 56(4), 272–9. Allen, S. (1987) ‘An empirical model of work attendance’, Review of Economics and Statistics, February, 77–87. Almer, E. D. and Kaplan, S. (2002) ‘The effects of flexible work arrangements on stressors, burnout and behavioral job outcomes in public accounting’, Behavioral Research in Accounting, 14, 1–34. Bailyn, L., Drago, R. and Kochan, T. (2001) Integrating Work and Family Life: A Holistic Approach. Report for Alfred P. Sloan Foundation Work-Family Policy Network, September. Baltes, B., Briggs, T., Wright, J. and Neuman, G. (1999) ‘Flexible and compressed workweek schedules: a meta-analysis of their effects on work-related criteria’, Journal of Applied Psychology, 84(4), 496–513.

Lonnie Golden 135 Bardoel, E. A., Moss, S., Smyrnios, K. and Tharenou, P. (1999) ‘Employee characteristics associated with the provision of work-family policies and programs’, International Journal of Manpower, 20(8), 563–77. Barnett, R. C., Gareis, K. and Brennan, R. T. (1999) ‘Fit as a mediator of the relationship between work hours and burnout’, Journal of Occupational Health of Psychology, 4, 307–17. Buckley, M. R., Kicza, D. C. and Crane, N. (1987) ‘A note on the effectiveness of flextime as an organizational intervention’, Public Personnel Management, 16(3), 259–67. Christensen, K. and Staines, L. G. (1990) ‘Flextime: a viable solution to work/family conflict?’, Journal of Family Issues, 11, 455–76. Clarkberg, M. (2000) Balancing Acts: Easing the Burdens and Improving the Options for Working Families. (Washington, DC: Economic Policy Institute). Dalton, D. and Mesch, D. (1990) ‘The impact of flexible scheduling on employee attendance and turnover’, Administrative Science Quarterly, 35, 370–87. Deery, S. and Mahony, A. (1994) ‘Temporal flexibility: management strategies and employee preferences in the retail industry’, Journal of Industrial Relations, September, 332–52. Dunham, R., Pierce, J. and Castenda, M. (1987) ‘Alternative work schedules: two field quasi-experiments’, Personnel Psychology, 40, 215–42. Ezra, M. and Deckman, M. (1996) ‘Balancing work and family responsibilities: flextime and child care in the federal government’, Public Administration Review, 56(2), 174–9. Fagan, C. (2001) ‘The temporal re-organization of employment and the household rhythm of work schedules’, American Behavioral Scientist, 44(7), 1199–212. Families and Work Institute (1993) The 1992 Study of the Changing Work Force. New York, NY. Fenwick, R. and Tausig, M. (2001) ‘Scheduling stress: family and health outcomes of shift work and schedule control’, The American Behavioral Scientist, 44(7), 1179–98. Ferber, M. and Waldfogel, J. (1998) ‘The long-term consequences of non-standard work’, Monthly Labor Review, 121(5), 3–12. Flexible Resources (2001) Profile of candidates seeking flexible work arrangements. May. Galinsky, E. and Bond, J. T. (eds) (1998) The 1998 Business Work-Life Study: A Sourcebook. New York: Families and Work Institute. Glass, J. (1995) ‘Employer characteristics and the provision of family responsive benefits’, Work and Occupations, 22(4), 380–411. Golden, L. (1996) ‘The economics of worktime length, adjusting and flexibility: contributions of three competing paradigms’, Review of Social Economy, Spring 1996, 1–44. Golden, L. (2001) ‘Flexible work schedules: what are workers trading off to get them?’, Monthly Labor Review, 124(3), 50–67. Greenwald, J. (1998) ‘Employers warming up to flexible schedules’, Business Insurance, 32(24), 3–6. Hamermesh, D. (1999) ‘Timing of work over time’, The Economic Journal, 109(452), 37–66. Hammer, L. and Barbera, K. (1997) ‘Toward an integration of alternative work’, Human Resource Planning, 20(2), 28–36.

136 Flexible Work Schedules Harrick, E., Vanek, G. and Michlitsch, J. (1986) ‘Alternative work schedules, productivity, leave usage and employee attitudes: a field study’, Public Personnel Management, 15(2), 159–69. Hein, K. (1999) ‘Finding balance’, Incentive, 173, 7. Hill, E. J., Hawkins, A., Ferris, M. and Weitzman, M. (2001) ‘Finding an extra day a week: positive influence of perceived job flexibility on work and family balance’, Family Relations, 50(1), 49–58. Hohl, L. (1996) ‘The effects of flexible work arrangements’, Nonprofit Management and Leadership, 7(1), 69–86. HR Focus (1998) ‘The work/life haves and have-nots’, 75(8), s6. Jacobs, J. (1998) ‘Measuring time at work: are self-reports accurate?’, Monthly Labor Review, December, 42–53. Johnson, N. and Provan, K. (1995) ‘The relationship between work/family benefits and earnings: a test of competing hypotheses’, Journal of Socio-Economics, 24(4), 571–84. Kate, Nancy Ten (1998) ‘Two careers, one marriage’, American Demographics, 20(4), 28. Kim, J. and Campagna, A. (1981) ‘Effects of flexitime on employee attendance and performance: a field experiment’, Academy of Management Journal, 24(4), 729–41. Kopelman, R. (1986) ‘Alternative work schedules and productivity: a review of the evidence’, National Productivity Review, 5(2), 150–65. Kossek, E., Barber, A. and Winters, D. (1999) ‘Using flexible schedules in the managerial world’, Human Resource Management, 38(1), 33–46. Krausz, M. and Freibach, N. (1983) ‘Effects of flexible working time for employed women upon satisfaction, strains and absenteeism’, Journal of Occupational Psychology, 2, 155–9. Krausz, M. and Hermann, E. (1991) ‘Who is afraid of flexitime? Correlates of personal choice of flexitime schedules’, Applied Psychology, 40(3), 315–26. Krausz, M., Sagie, A. and Bidermann, Y. (2000) ‘Actual and preferred work schedules and scheduling control as determinants of job-related attitudes’, Journal of Vocational Behavior, 56, 1–11. Kush, K. and Stroh, L. (1994) ‘Flextime: the imaginary innovation’, Business Horizons, Fall. Leonard, B. (1999) ‘Flexibility may enhance job performance’, HRMagazine, 44(9), 28–30. McCrate, E. (2002) ‘Working mothers in a double bind wednesday’, Economic Policy Institute, Washington DC, May. McGuire, J. B. and Liro, J. R. (1987) ‘Absenteeism and flexible work schedules’, Public Personnel Management, 16, 47–59. McShulskis, E. (1997) ‘Work and family benefits are increasingly popular’, HRMagazine, 42(7), 26–9. Meyer, C. S., Mukerjee, S. and Sestero, A. (2001) ‘Work-family benefits: which ones maximize profits?’, Journal of Managerial Issues, 13(1), 28–44. Moss, R. L. and Curtis, T. D. (1985) ‘The economics of flexitime’, Journal of Behavioral Economics, Summer, 95–114. Presser, H. (1995) ‘Job, family and gender: determinants of nonstandard work schedules among employed americans in 1991’, Demography, 32(4), 577–98. Presser, H. (1999) ‘Toward a 24-hour economy’, Science, 284, 1778–9.

Lonnie Golden 137 Ralston, D. (1989) ‘The benefits of flextime: real or imagined?’, Journal of Organizational Behavior, 10(4), 369–73. Ralston, D. and Flanagan, M. (1985) ‘The effect of flextime on absenteeism and turnover for male and female employees’, Journal of Vocational Behavior, 26(2), 206–17. Ralston, D., Anthony, W. and Gustafson, D. (1985) ‘Employees may love flexibility, but what does it do to the organization’s productivity?’, Journal of Applied Psychology, 70(2), 272–80. Romano, C. (1994) ‘What is your flexibility factor?’, Management Review, January, 9. Scandura, T. and Lankau, M. (1997) ‘Relationships of gender, family responsibility and flexible work hours to organizational commitment and job satisfaction’, Journal of Organizational Behavior, 18(4), 377–91. Schor, J. (1994) ‘Worktime in a contemporary context: amending the FLSA’, ChicagoKent Law, 70(1), 157–72. Sharpe, D., Hermsen, J. and Billings, J. (2002) ‘Factors associated with having flextime: a focus on married workers’, Journal of Family and Economic Issues, 23(1), 51–72. Shephard, E., Clifton, T. and Kruse, D. (1996) ‘Flexible work hours and productivity: some evidence from the pharmaceutical industry, Industrial Relations, 35(1), 123–39. Sparks, K., Faragher, B. and Cooper, C. (2001) ‘Well-being and occupational health in the 21st century workplace’, Journal of Occupational and Organizational Psychology, 74(4), 489–509. US Department of Labor (1999) Futurework – Trends and Challenges for Work in the 21st Century. Part 3: Work And Family: Flexibility On The Job. Washington, DC. US Department of Labor, Bureau of Labor Statistics (BLS) (1998) ‘Workers on flexible and shift schedules in 1997’, USDL 98–119, BLS News, March 26. US Department of Labor, Bureau of Labor Statistics (BLS) (2002) ‘Workers on flexible and shift schedules in 2001’, USDL 02–225, BLS News, April 18.

8 When Bad Email Happens to Good People: A Case of Information Technology Mismanagement David G. Schwartz

Introduction Email is a powerful communications tool, playing an increasingly important role in our organizations (Fulk and Desanctis, 1995). Yet with power should come responsibility. By far the most dangerous form of power is power wielded by the ignorant. Email use in general is the most widely used form of computer mediated communications in organizations (Kalakota and Whinston, 1996). Broadcast usage of computer mediated communications, in particular through the use of mailing lists, is one of the most frequent uses of interorganizational email (Kettinger and Grover, 1997). This chapter presents a case study of electronic mailing list abuse and mismanagement. The abuse I will discuss takes many forms, as not only is the email system abused, but so too are mailing list participants, and mailing list moderators. My discussion centres on mailing list membership and what happens when mailing list members are not properly trained in the operation of such lists. In addition, a proper definition of mailing list characteristics is suggested. We illustrate these issues through a case study of two different types of mailing lists, the first being a Read-Only Subscription-Based (ROSB) and the second being an UnModerated Mandatory Participation (UMMP) list. The lists have been so-named to protect the anonymity of the list owners and members. We present the ROSB/UMMP mailing list mix-up, analyze the reactions and responses to the mailing list abuse, and present points of action to avoid such problems. The lessons learned are instructive for both IT professionals and users by shaping the requirements and demands for training in the adoption of new information technologies. 138

David G. Schwartz 139

Background This case study focuses on a research and teaching university located in Israel. The university has over 26,000 students and 1,300 faculty members. The level of computerization of the campus is average by Western standards, with network and email access provided in all faculty offices, but with no wireless campus facilities and limited networking of lecture halls and classrooms. The overwhelming majority of departmental and interdepartmental communication is still handled through internal paper mail.

Mailing lists A mailing list is a list of e-mail addresses of people interested in the same subject. When a list subscriber sends a message, the message goes to everyone on the list. A subscriber can reply to the messages, send new messages, or simply read messages without participating. Mailing list email traffic is automatically managed by programs called Mailing List Managers (MLMs). The two most frequently used Unix-based MLMs are Listserve and Majordomo. Two popular web-based MLMs are ListBot (www.listbot.com) and OneList (www.onelist.com). Once a user has subscribed to a mailing list, he or she can send a message to all members of the list by sending a single email message to the email address of the mailing list.

Voluntary subscription-based mailing lists A subscription-based mailing list is a mailing list whose members choose to join by submitting their email addresses to the listserver. One example of such a voluntary list is the ROSB list maintained by a university research centre to disseminate a digest of information related to their area of research. ROSB follows standard subscribe and unsubscribe procedures as presented in Table 8.1. Subscription to a voluntary mailing list requires confirmation in the form of an email message originating from the email account that is to be added to the list. In general, one cannot subscribe to a voluntary mailing list against one’s will unless access to one’s email account has been usurped or forged – and even in such a case he or she can immediately unsubscribe.

Mandatory mailing lists A mandatory mailing list is an email distribution list where one becomes a member by virtue of association, affiliation, or participation in a certain group or activity. Subscription to such a list is contingent on membership in a group and follows automatically from that membership without

140 When Bad Email Happens to Good People Table 8.1

Common mailing list commands

Command syntax

Description

INFO SUBscribe listname SIGNOFF listname Query listname Search listname keyword . . . SET listname options INDex <listname> GET filename filetype

Order documentation (plain text files) Subscribe to a list Sign off from a list Query your subscription options Search list archives Update your subscription options Order a list of LISTSERV files Order a file from LISTSERV

additional verification or subscription process. One example of such a list is the UMMP list maintained by the University for the Space Science Library members. Every faculty member of the Faculty of Space Science is added to the UMMP list without prior consultation. The stated purpose of the UMMP list is to inform faculty members as to new acquisitions and other items related to the functioning of the Space Science library. While in theory a group member could be allowed to unsubscribe from a UMMP list, this would negate the purpose of such a list.

Moderated vs. unmoderated mailing lists A moderated mailing list is a mailing list where an authority (usually a single person, but possibly a group) decides what messages can be posted to the list and be seen by the full list membership. This moderator effectively controls what is seen by members of the list. In an unmoderated mailing list, any member may post any message at any time. Unmoderated lists are essentially self-governing.

Read-only lists An extreme case of the Moderated Mailing List is the Read-only Mailing List. In this type of list, the members only receive messages (posted by the moderator) and cannot respond. This type of list is commonly used for marketing and customer support messages. Combining the list characteristics described above gives us the taxonomy shown in Figure 8.1.

Theoretical background and related work Most research regarding organizational use of email focuses on the issues of security (Segev et al., 1998) and corporate policy ( Jones, 1998).

David G. Schwartz 141

Control level moderated

read-only

unmoderated

participatory

Activity level

voluntary Participation level

Figure 8.1

mandatory

voluntary

mandatory voluntary

mandatory

Electronic mailing-list taxonomy

There has been some work done in the areas of email and human– computer interfaces toward adding context to the use of email (Schwartz and Te’eni, 2000). In a case study of Bank of America, Segev et al. (1998) found that in 5 per cent of cases, organizational users complained of lack of training in the proper use of email systems. Gilsdorf (1998) examined employee perceptions of problems with communications in general, though not specifically electronic communications. Schwartz (1999) looked at threats to organizational communication specific to email messages. Yates and Orlikowski (1999) showed how different cultural genres within a single organization can have an impact on electronic organizational communications. Kettinger and Grover (1997) found that broadcast usage of computer mediated communications, which includes mailing lists, reflects an information-gathering communication function, and is the most frequent use of interorganizational email. More general studies of IT adoption in organizations, such as Avgerou (2000), reported on the importance of the connection between IT innovation and organizational practice. While the close relationship between IT innovation and organizational practice is identified as a primary factor in successful IT adoption, little has been said regarding organizations where this is not the case and interaction between IT and organizational change is minimal or non-existent. The importance of training for proper email use is identified as an important element by Hacker et al. (1998), who found that email policies alone are not sufficient to guarantee proper functioning of an organizational email system.

142 When Bad Email Happens to Good People

Organization awareness as to the importance and potential of electronic communications had been found to be sorely lacking (Parker and Swatman, 1995). Interestingly, many organizations take a narrow view of email and perceive it as a ‘speed enhancer’ rather than an opportunity to re-engineer certain business practices (see discussion by Swatman, 1994). Mailing lists in general are the subject of much research in marketing communications (Arnold and Arnold, 1997; Cyr, 1997; Hodgdon, 1997). That work focuses primarily on mailing list members that are not part of a single organization and are the target of a marketing message, usually as a voluntary Read-Only Subscription-Based list. In one of the only research efforts to highlight the use of mailing lists, Compton et al. (1991) studied two high-tech organizations and found favourable organizational responses to the efficiencies brought about by the use of such lists. The above related work illustrates the importance placed on proper use of email in the organization and the potential positive and negative effect email has on organizational structure, culture, and performance. This work builds on existing research on organizational email and mailing lists, and provides the first documented case of an internal organizational mailing list that examines the effects of mailing list mismanagement from the users’ perspective.

Methodology and case description The Research Center’s ROSB voluntary moderated mailing list is a subscription-based email list that has 6,100 subscribers of various affiliations in 67 countries. It provides weekly updates of news related to the activities of the Research Center. A portion of the ROSB mailing that initiated the activities described in this case appears in Figure 8.2. The UMMP mandatory unmoderated mailing list has 352 subscribers, all of whom are faculty members of the university. It provides occasional updates regarding changes and additions to the Space Science Library at the university. For the remainder of this chapter, the Research Center’s read-only mailing list will be referred to as ROSB and the University Library mailing list as UMMP. It is important to note that ROSB is an inter-organizational list and only UMMP is intra-organizational. During the period between 23/09/99 and 15/10/99 a total of 93 email messages were posted to the UMMP mailing list. All email messages in this period were gathered by a member of the UMMP list. Categorization

David G. Schwartz 143

REVIEW OF RESEARCH ISSUES (ROSB) ROSB NEWS A magazine on [ ] /1999:Issue/September 1999 Direct circulation 6, 100; Total Circulation 8,000 *Serving Readers in 67 Countries*

CONTENTS IMPORTANT NEWS: Redesigned homepage; ROSB journal, Vol. 3, No. 4. SPECIAL REPORT: Research Guide 1. PUBLICATIONS OF INTEREST 2. WEBSITES/GROUPS ONLINE 3. FREE SUBSCRIPTION TO E-MAIL PUBLICATIONS 4. FUNDING/FELLOWSHIPS/WRITING OPPORTUNITIES 5. RESEARCH QUERIES-PLEASE HELP 6. SCHOLAR’S AND AUTHOR’S ALERTS (people report on their books, articles, and activities) 7. ANNOUNCEMENTS OF MEETINGS, CONFERENCES, LECTURES

IMPORTANT ANNOUNCEMENTS: ROSB has completely redone its homepage, making it easier to use, adding over 30 items, updating our link list, and including a search engine that shows all listings for any word. Suggestions welcome. We are now seeking article for the next issue of ROSB journal, Vol. X, No. Y (1999) and welcome submitted items or queries. (6 pages of ROSB Digest Material Removed)

Figure 8.2

Original ROSB posting to UMMP list (edited and abridged)

of the messages was based on the message subject lines. We present a chronological study, showing the progression of email messages and the gradual increase in frustration and misunderstanding that ensued.

Chronology of a mailing list mess On September 23 and 26, 1999, two messages were crossposted from the ROSB list to the UMMP list (See Table 8.2). This was due to an error originating in the Computing Center communications system that was corrected immediately upon discovery. The abuse that followed spanned 11 days until it died down. The fact that the UMMP is unmoderated explains why the list owner did not intervene earlier, but it does not justify the lack of monitoring which may have led to less waste and a quicker resolution.

144 When Bad Email Happens to Good People Table 8.2

Categorization of email messages by topic

1. Official ROSB list postings (that appear on the UMMP list) 2. Request to Unsubscribe from the ROSB list 3. Request to Unsubscribe from the UMMP list 4. Request to Subscribe to the ROSB list 5. Threat of legal or disciplinary action to ROSB list owners 6. Private Messages that were posted instead of sent personally 7. Comments on the Private Messages that were posted 8. Attempts to explain how to fix the problem or unsubscribe 9. Apologies and explanations from the UMMP list moderators 10. Apologies and explanations from the ROSB list moderators Total messages:

2 75 3 2 1 2 2 3 2 1 93

The first series of messages posted to UMMP were from people who had incorrectly understood that they were now a regular member of ROSB and wanted to unsubscribe. A sample of such messages appears in Table 8.3, showing the exact time and date a message was posted to the list as well as the core content of the message. The first unsubscribe request came on Sunday, Oct. 3, 1999 at 12:10 pm.

Escalation and resolution As time progressed with no clear resolution to the confusion, and after around message 16, the messages began to take on a more confused and desperate tone (emphasis not added). Some people felt they were being directly addressed by the senders of the messages: Message 18: ‘For some reason I received messages from 19 people to be dropped from the ROSB list. I am not the address!’ Others could not comprehend why this was happening: Message 74: ‘I do not understand why I still get these messages!!’ And others thought they had reacted properly with unsubscribe messages, to no avail: Message 70: ‘For the fourth time, please kindly remove my name from your list.’ Message 79: ‘For the second time I BEG you REMOVE my name from your mailing list.’

145 Table 8.3

Sample email messages posted to the UMMP list

No.

Date/time stamp*

Core message

1

Sun, 3 Oct 1999 12:10:25 +0200 Sun, 3 Oct 1999 13:09:30 +0200

Please drop my name from the mailing list for ROSB. Please drop my name from the mailing list of ROSB. Thank you in advance. Please, drop my name from ROSB. Please remove me from this list. Please drop my name from the mailing list for ROSB. Please remove me from ROSB. Please remove me from the list

2

3 4 5 6 7 8 9

10 11 12 13 14 15 16 17

18

Sun, 3 Oct 1999 15:26:54 +0200 Sun, 3 Oct 1999 16:10:04 +0200 Sun, 3 Oct 1999 16:17:29 +0200 Sun, 3 Oct 1999 16:34:11 +0200 Sun, 3 Oct 1999 16:58:48 +0200 Sun, 3 Oct 1999 14:59:30 +0000 Sun, 3 Oct 1999 17:48:43 +0200 Sun, 3 Oct 1999 17:49:15 +0200 Sun, 3 Oct 1999 18:48:41 +0200 Sun, 3 Oct 1999 19:22:37 +0200 Sun, 3 Oct 1999 21:08:23 +0200 Sun, 3 Oct 1999 21:24:29 +0200 Sun, 3 Oct 1999 12:40:47 −0700 Sun, 3 Oct 1999 22:27:15 +0200 Sun, 3 Oct 1999 22:59:31 +0200

Sun, 3 Oct 1999 22:59:32 +0200

Note

Please remove me from this mailing list. Thanks Please, I would like to be dropped from the ROSB list like everyone else. Please drop my name from the mailing list for ROSB. Please drop my name from the mailing list for ROSB. Please Remove my name from the ROSB list. Please remove my name from the ROSB list. Please drop my name from ROSB list Please drop my name from the mailing list for ROSB. Please drop from list. For some reason, I receive the mail 19 times!!! Please drop XXXX@YYYY from the mailing list. For some reason I received messages from 19 people to be dropped from the ROSB list. I am not the address! Please drop me as well

Same person as 17

146 When Bad Email Happens to Good People Table 8.3

(Continued)

No.

Date/time stamp*

Core message

Note

19

Sun, 3 Oct 1999 23:05:43 +0200

Same person as 16

20

Sun, 3 Oct 1999 23:07:01 +0200

21

Sun, 3 Oct 1999 23:54:49 +0200

22

Sun, 3 Oct 1999 14:57:22 −0700

23

Mon, 4 Oct 1999 02:51:36 +0200 Mon, 4 Oct 1999 05:31:32 +0200

Why not tell the millions of people who seem to want off the list to use the reply to address ONLY and to say no if they are given a choice of replying to all recipients? I have never received so much garbage mail in one day. Please remove me from this list. I am receiving dozens of letters sent to other faculty concerning the ROSB and I beg you to stop this. Please, do not send this information (UMMP) anymore. I never subscribed in the first place!! Please!!!!drop my name from ROSB mailing list. Please remove me from the list of ROSB

24 ... 70

74 79

Mon, 11 Oct 1999 22:27:15 +0200 Wed, 13 Oct 1999 13:30:36 +0200 Wed, 13 Oct 1999 23:43:56 +0200

Subject: ‘You are driving me crazy’

UMMP

For the fourth time, please kindly remove my name from your list. I do not understand why I still get these messages!! For the second time I BEG you REMOVE my name from your mailing list.

* Time stamp automatically provided by the email system is shown as GMT +/− offset based on location of the sender and indicates the time the message was sent. The messages appear in the table based on the order in which they were received.

By the time Message 80 rolled around, legal action was being threatened: Message 80: ‘Dear Sirs, I regret to say that if you do not stop sending me this junk mail, which is also blocking important messages to my home computer, I will have no recourse but to advise all my colleagues in the Economics Department to ask the legal adviser to the University to stop you from using the University system to distribute this unsolicited journal’.

David G. Schwartz 147

Of course the ‘unsolicited journal’ had only been distributed twice, and that was 79 messages ago. The real problem was not the original posting, but the way in which others reacted to it. Given the purpose of the mailing list, the issue brought to the fore by this message was whether the list characteristics been properly defined. If the Library list been properly defined as a Read-Only Mandatory Participation list, rather than an UnModerated Mandatory Participation list, the issue raised in Message 80 would have been moot. Finally, Message 90 presented a reasoned attempt by one of the subscribers to understand what had happened: Message 90: ‘As with many of the other users who have already written, I have asked to be removed from this ROSB list. I sent my request directly to Prof [in charge of the Center] so as to avoid it going out to everyone on the UMMP list. Since I have not received any additional e-mails from ROSB but rather only the daily bombardment of people requesting to be removed from the distribution list, I will give ROSB the benefit of the doubt and assume that my name was removed from their list and that the reason that we continue to receive messages is that those who wish to be removed are replying to the UMMP list and not directly to Prof [in charge of the Center]. What I would like to know then is who prepared the UMMP list, for what purpose, and why was the original ROSB announcement sent to the list (apparently without authorization)? I assume that UMMP is for distributing official university sanctioned mail and announcements which does not seem to be the case here. In the future any initiatives such as ROSB should prepare their own mailing list and be sure to provide an easy way to unsubscribe and/or to reply to the original sender without involving the entire mailing list.’ This was soon followed by an ‘official’ message from the UMMP owner: Message 93: ‘Dear Members of UMMP, UMMP and other similar ListServs from the other faculties were set up to provide the faculty with updates about the various electronic databases and electronic journals that the University subscribes to via the Consortium of University Libraries. In addition other relevant news of the library is included in LIBRARY NEWS. This is the only use for the UMMP list. Unfortunately security was breached and other material was sent out without my knowledge. The Computer Center has dealt with the problem and I hope that none of you are bothered again with unsolicited mail.’

148 When Bad Email Happens to Good People

Note that the ‘explanation’ did not help shed light on the process itself that led to the transmission of 90 email messages to each of 352 mailing list participants – a total of 31,680 unnecessary email messages.

Deviation Somewhere towards the middle of the sequence of messages, the discussion took an interesting turn as members of the list mistakenly began to exchange private email correspondence to the whole list. The contents ‘shared’ was in no way damaging or embarrassing to those involved, but it did provide a much-needed break for the 350 other frustrated list members. After the two members, hereafter X and Y, exchanged ‘open’ correspondence about an administrative issue, the following was posted: Message 77: ‘Would everybody else like to advise X and Y about how they should sign their letters? At least if we are going to get bombarded with mail that we did not ask for (and if nobody knows how to reply to the sender only, in order to cancel, and instead has to reply to everybody), at least it seems we are getting into something interesting. So X and Y, give us more details about who thinks what of who, and who should sign letters and why etc. etc, and make life more interesting so that we stop cursing ROSB every few minutes.’ This tongue-in-cheek approach generated two addition email messages that concurred with its suggestion. Needless to say, the ‘shared’ email ceased immediately.

Attempt to alleviate The ROSB list owners re-entered the picture at message 43, when they were informed of what was being attributed to their mailing list. To the credit of the list owners, since they were not UMMP members, they had no way of knowing what had transpired at their expense. Message 43: ‘We have been informed that due to some computer error or sabotage, large numbers of messages have been sent in the name of various people – including myself – to many faculty members at the university through different lists. The computer department is investigating. Please note that we did not send ROSB to you, you are not on our list, and you will not receive anything in future. Again, we would never send unwanted material to people.’

David G. Schwartz 149

Misconceptions and misunderstanding The lack of user training in the use of email systems in general and mailing lists in particular came across in a number of messages.

• Misconception no. 1: I am subscribed to list X and don’t wish to be. Most of the participants in the mailing list dialog incorrectly assumed that they had been unknowingly subscribed to the UMMP mailing list and were asking to be removed. The exception to this was: Message 22: ‘Please do not send this information (UMMP) anymore. I never subscribed in the first place.’ While the author of message 22 was correct in his determination that he never had subscribed to UMMP (since it was a mandatory list not requesting permission), he/she still incorrectly assuming that UMMP was the cause of the unwanted ‘information’ he was being sent, when in fact it was his/her colleagues generating the messages.

• Misconception no. 2: You can avoid sending messages to everyone in a certain way. Message 19: ‘Why not tell the millions of people who seem to want off the list to use the reply to address ONLY and to say no if they are given a choice of replying to all recipients?’ However well-intended, this advise was incorrect, since the reply to address of the email messages was the UMMP mailing list. Since this was not the result of an improper cc field (an anachronistic term referring to the ‘carbon copy’ list of recipients who are supposed to receive a message in addition to the main addressee), the cause was not the use of reply all as assumed by the author of this message.

• Misconception no. 3: I am the only one experiencing this. Many of the participants, not being at all acquainted with the workings of mailing lists, assumed that they were in this alone or were the specific target of some problem. Message 47: ‘PLEASE CHECK WHY ALL THE RESPONSES CANCELLING ROSB ARE BEING SENT TO ME! I AM FLOODED.’

150 When Bad Email Happens to Good People

• Misconception no. 4: The problem is being caused by a bug. Message 57: ‘Why do I get all the emails sent to you by tens of people telling you not to send them mail. It drives me crazy. Please see what is the bug.’

• Misconception no. 5: I can subscribe to ROSB by posting to UMMP. Message 59: ‘Please put me on the ROSB list.’ was the first attempt by a user to actually get subscribed to the subscription-based list where the original message had come from. This may have been an attempt at black humour given the circumstances, or a legitimate attempt to join the list. If the latter, the user once again was exhibiting a lack of understanding of mailing lists by trying to subscribe to ROSB through a message to UMMP. The same user followed up with another posting to UMMP as follows: Message 75: Prof [ ] I asked to sign on to your important and interesting list but have yet to receive any messages besides some people wishing to sign off. Perhaps they too are tired of waiting to receive the new journal. When will we receive an update? Message 61 followed in the same spirit: Message 61: ‘Please confirm that you have not dropped me from the list and that I will continue to receive your postings.’ though not making it clear if the desire was to stay on UMMP or ROSB.

Lessons learned There are four major lessons that can be learned from this case, the former two related to proper user education, and the latter two related to proper mailing list definition and management. When establishing a mandatory mailing list, subscribers must be informed of their membership. This is the most rudimentary form of user education for a mailing list. Even if you do not get into details

David G. Schwartz 151

about how the user can interact with the list you at least need to inform them of the list’s existence, purpose, and their association as members. Rudimentary training as to the proper operation of a mailing list requires learning only three command sequences: Subscribe, Unsubscribe, and Post Message. Had the users been properly trained, the bulk of unsubscribe messages (albeit to the wrong list) would have been processed and they would have been removed from the UMMP list. These, however, will be of no use if the user is unaware of list membership and purpose in the first place. User training was so completely lacking in this case that in the scope of 90 messages, not a single user was able to provide the membership with the proper commands to unsubscribe. When defining a new mailing list, determine whether your mailing list should be moderated, unmoderated, or read-only. In the case of ROSB, the list was properly defined as moderated and read-only, since it was meant to be an authoritative source of information to its subscribers. The UMMP list, on the other hand, was improperly defined as an unmoderated participatory list, allowing each member to post at will, when really its sole purpose was to send library announcements to members. In such a list there is little need, if any, to let the members react to the announcement of a new periodical subscription. Had UMMP been read-only, not only would the members not have been able to initiate the snowball of unsubscribe emails in reaction to the ROSB post, the ROSB post itself would never have occurred. Another lesson to be learned by list owners is to properly monitor their own mailing lists. It took 11 days and 90 emails for the UMMP owner to provide a detailed explanation and apology to the list membership for the original posting. Had the list owner been monitoring his or her own list, the situation could have been corrected immediately after Message 1 (Table 8.3) was posted.

To summarize: 1. Let your users know what mailing lists they are subscribed to, and the purpose of each list. 2. Train your users in rudimentary mailing list operation so that they know how to subscribe and unsubscribe to mailing lists that you offer. 3. Make sure mailing list owners monitor all traffic on their lists, even if it is not defined as moderated.

152 When Bad Email Happens to Good People

4. Properly define the mailing list characteristics to match the purpose of the list. Do not allow user participation if there is no justifiable reason for it. 5. Use mailing list headers/footers. A mailing list header or footer with contact information for the list owner can be automatically appended by the MLM to any mailing list message given proper definition of the list. Had the UMMP list contained such information, the disgruntled members would at least have known who to turn to for help.

Conclusion A total of 31,680 unnecessary email messages passed through the email system between the 352 UMMP members in the course of 11 days. Assuming that each member spent only 45 seconds to download, read, and delete each message, there were a total of 396 work-hours wasted. Considering that 352 is a relatively small subscription base for a mailing list, the potential for damage is tremendous. The lessons to be learned from this case are clear, and their implementation should be a matter of course in any rollout of mailing list use in an organization. Training takes time and money, but it is an investment in the smooth future operation of your organization. Had the size of the lists been reversed, with the victims being 6,100 subscribers (i.e. the size of the ROSB list), the damage of 90 messages alone would reach 6,862 lost work-hours. Needless to say, with 6,100 untrained users, the potential for needless unsubscribe request messages would be significantly higher. Thus, responses can be expected to grow exponentially rather than linearly. Can you afford to let this happen in your organization?

References Arnold, S. E. and Arnold, E. S. (1997) ‘Push technology: driving traditional online into a corner’, Econtent, 20(4), 36–46. Avgerou, C. (2000) ‘IT and organizational change: an institutionalist perspective’, Information Technology and People, 13(4), 234–62. Compton, D. C., White, K. and DeWine, S. (1991) ‘Techno-sense: making sense of computer-mediated communication systems’, The Journal of Business Communication, 28(1), 23–43. Cyr, D. (1997) ‘Lists in space’, Catalog Age, 19(3). Jones, V. A. (1998) ‘Developing an e-mail policy’, Office Systems, Oct. Fulk, J. and Desanctis, G. (1995) ‘Electronic communications and changing organizational forms’, Organizational Science, 6, 338–49. Gilsdorf, J. W. (1998) ‘Organizational rules on communicating: how employees are – and are not – learning the ropes’, The Journal of Business Communication, April.

David G. Schwartz 153 Hacker, K. L., Goss, B., Townley, C. and Horton, V. J. (1998) ‘Employee attitudes regarding electronic mail policies: a case study’, Management Communication Quarterly, Feb. Hodgdon, P. N. (1997) ‘The role of intelligent agent software in the future of direct response’, Direct Marketing, 59(9), 10–17. Jones, V. A. (1998), ‘Developing an email policy’, Office Systems, October. Kalakota, R. and Whinston, A. (1996) Frontiers of Electronic Commerce. Reading, MA: Addison-Wesley. Kettinger, W. J. and Grover, V. (1997) ‘The use of computer-mediated communication in an interorganizational context’, Decision Sciences, 28(3), 513–55. King, L. J., Gurbaxani, V., Kraemer, K., McFarlan, F. W., Raman, K. S. and Yap, C. S. (1994) ‘Institutional factors in information technology innovation’, Information Systems Research, 5(2), 136–69. Parker, C. M. and Swatman, P. M. C. (1995) ‘Educating tomorrow’s managers for telecommunications and EDI: a cross-cultural experience’, Information Technology and People, 8(2), 58–79. Schwartz, D. G. (1999) ‘When email meets organizational memories: addressing threats to communication in a learning organization’, International Journal of Human-Computer Studies, 51, 599–614. Schwartz, D. G. and Te’eni, D. (2000) ‘Applying knowledge to action with Kmail’, IEEE Intelligent Systems, April/May. Segev, A., Porra, J. and Roldan, M. (1998) ‘Internet security and the case of Bank of America’, Communications of the ACM, Oct. Swatman, P. M. C. (1994) ‘Business process redesign using EDI: the BHP steel experience’, Australian Journal of Information Systems, 1(2), 55–73. Yates, J. and Orlikowski, W. J. (1999) ‘Explicit and implicit structuring of genres in electronic communication: reinforcement and change of social interaction’, Organization Science, 10(1), 83–103.

9 Can Workplace Deviance Be Constructive? Bella L. Galperin*

Introduction Workplace deviance and misbehaviour are becoming an increasingly important issue for organizations. The prevalence of destructive and harmful behaviours is surprisingly common in the workplace. Approximately 70 per cent of employees have engaged in some form of deviant behaviour, such as losing their temper at work (Bennett and Robinson, 2000). Furthermore, it is estimated that annually more than two million people become victims of work-related crimes (Kondrasuk, Moore and Wang, 2001). Destructive deviant behaviours can result in substantial economic and social costs for organizations. According to the National Safe Workplace Institute, workplace misbehaviour can cost organizations approximately 4.2 billion US dollars in lost productivity and legal expenses (Bensimon, 1994). Other costs include tarnished reputations, weakened employee morale, increased absenteeism and turnover (Filipczak, 1993). To date, employee deviance has generally been conceptualized as destructive. While some authors (e.g. Tripp and Bies, 1997; Vardi and Weiner, 1996) have recognized that deviant behaviours can also be beneficial to organizations, the majority of the theoretical development has focused on destructive deviance. Workplace deviance has generally been used to describe the following behaviours: antisocial behaviour (Giacalone and Greenberg, 1997), misbehaviour (Vardi and Wiener, 1996), workplace * This chapter is based on the author’s doctoral dissertation. I am grateful for the support of my dissertation committee (Muhammad Jamal, Gary Johns, Vishwanath V. Baba and Alfred Jaeger).

154

Bella L. Galperin 155

aggression (Baron and Neuman, 1996; O’Leary-Kelly, Griffin and Glew, 1996), and organizational retaliation (Skarlicki and Folger, 1997). Although these authors have defined deviance as behaviours that cause harm to the organization, deviance can also be defined as behaviours that violate the norm (Bord, 1976). Consequently, deviance may include both positive and negative aspects. While deviant behaviour may be harmful, employee deviance may be functional and constructive as well. Employees who voluntarily violate the organizational norms may be important sources of innovation and entrepreneurship. Employees who engage in nonconforming behaviours, such as champions of innovations or corporate entrepreneurs, can contribute to the innovation process and competitive advantage of the organization (Howell and Higgins, 1990; Howell, Shea and Higgins, 1998). Organizations which are unable to tolerate deviation are often unable to adapt, and consequently are more likely to fail (Dehler and Welsh, 1998). This chapter conceptualizes deviance as either destructive or constructive. It argues that deviance may not always be harmful to the organization. Employees who engage in discrepant behaviours, such as innovative behaviours and internal whistle-blowing, may contribute to the overall well-being of the organization (Dehler and Welsh, 1998). First, this chapter presents a definition of deviance. Using Merton’s (1957) typology, an attempt is then made to integrate the concepts of destructive and constructive deviance into a single theoretical framework in order to gain a more complete understanding of workplace deviance. The concept of constructive deviance is then examined in greater detail. Finally, practical implications and future research directions are discussed.

A definition of workplace deviance Given that workplace deviance is relatively a new area, it is important to establish a clear definition of the phenomenon. According to Robinson and Bennett (1997), a concise definition of workplace deviance is needed because: (1) it clarifies which behaviours are included in the domain of workplace deviance behaviours; (2) it increases the likelihood that researchers will have a shared understanding of the concept despite their independent research efforts; and (3) it helps in the development of future theories, conceptualizations and empirical studies on workplace deviance. In this chapter, employee deviance is defined as: voluntary behaviour that violates organizational norms. First, this definition focuses on the behaviour of employees who are existing members of the organization.

156 Can Workplace Deviance Be Constructive?

Similar to Robinson and Bennett (1997), the focus is placed on the actions of people who are inside the organization, such as organizational members, as opposed to the behaviour of outsiders, such as customers or suppliers. Given that organizations are most concerned with managing behaviours that are controllable, this perspective can be useful for organizations (Robinson and Greenberg, 1998). Second, using this perspective, deviance is conceptualized as purposeful or intentional behaviour. Similar to Robinson and Bennett’s (1995) construct of workplace deviance and Vardi and Wiener’s (1996) construct of misbehaviour, voluntary acts are included and accidental acts are excluded in the definition. However, unlike most conceptualizations of deviance, such as the constructs of workplace aggression (Baron and Neuman, 1996), retaliatory behaviour (Skarlicki and Folger, 1997), and organization-motivated aggression (O’Leary-Kelly et al., 1996), which stress behaviours that are intentionally harmful, the definition of deviance presented in this chapter addresses acts that are both intentionally harmful and intentionally beneficial to the organization. Finally, according to the definition, organizational norms assume a managerial orientation about expectations of employee behaviour and focus on violations of norms that apply across a broad spectrum of employees, rather than violations of norms around specific duties. Similar to Robinson and Bennett (1997), the definition of deviance focuses on the violation of norms adopted by the dominant coalition. Organizations are most likely to be interested in managing the violation of norms held by the dominant group because non-compliance with these norms could harm or benefit the organization. Furthermore, since most organizational members will usually share the norms of the dominant coalition, it is important to focus on these norms as opposed to the norms of a specific department or subculture (Robinson and Bennett, 1997). This definition is also in line with the psychological definition of deviance that characterizes deviants as not obeying or conforming to social norms (Cohen, 1966) or individuals who are less likely to belong to a social network (Freedman and Doob, 1968). As noted by Bord (1976), by defining deviant behaviour as behaviours that violate the norm, deviance can include both positive and negative aspects. While constructive deviance (e.g. innovative behaviours and internal whistleblowing) may contribute to the overall well-being of the organization, destructive deviance (e.g. theft and sabotage) may threaten it. Given that both forms of deviance encompass behaviours that violate the organizational norms, the two forms of deviance are related to each other. Furthermore, employees who engage in constructive deviance can also engage in destructive deviant behaviours because the two forms

Bella L. Galperin 157

of deviance do not preclude each other. Specifically, Merton’s (1957) typology provides us with a better understanding of the association between destructive and constructive deviance.

A framework of destructive and constructive deviance Merton’s (1957) typology of individuals’ responses to societal and organizational goals provides a framework for a greater understanding of the relationship between destructive and constructive deviance. According to Merton (1957), individuals’ responses will differ based on how strongly they identify with societal or organizational goals and the means to attain the goals. Specifically, people will display the following responses: (1) Innovation; (2) Conformity; (3) Ritualism; (4) Retreatism; and (5) Rebellion. First, individuals who engage in innovative behaviours accept the goals of the organization, but disagree with the means by which one must attain the goals. Second, those individuals who conform identify with both the means and goals of the organization. Third, ritualism occurs when the person agrees with the means but not the goals. Ritualism occurs when conformity becomes so highly valued that it becomes an end in itself instead of the accomplishment of goals. Ritualists dutifully follow the organizational rules and regulations. Fourth, retreatism occurs when the individual rejects both the goals and the means, and consequently becomes alien to the organization. Finally, rebellion occurs when people see organizations as barriers to legitimate goals. A person who rebels also rejects the established goals and means, but substitutes the current goals for new ones. Specifically, the responses of innovation and rebellion are most relevant to the constructs of destructive and constructive deviance. As outlined in Merton’s (1957) typology, individuals who do not accept the means and/or goals of the organization will deviate from the majority. The responses of innovators and rebels are similar in that both groups do not conform to the norms of the majority (Merton, 1957). Unlike rebels, innovators strongly identify with the organizational goals. As innovators use deviant or unorthodox means to attain the goals, their behaviour can be viewed as constructive deviance since the innovation is beneficial for the organization. Below, the constructs of destructive and constructive deviance are discussed.

Destructive deviance According to Robinson and Bennett (1995, p. 556), employee deviance is defined as: ‘voluntary behavior that violates significant organizational

158 Can Workplace Deviance Be Constructive?

norms and in so doing threatens the well-being of an organization, its members, or both’. Given the broader conceptualization of deviance adopted in this chapter, Robinson and Bennett’s (1995) definition of deviance is used to define destructive deviance. Robinson and Bennett (1995) argue that destructive deviance includes minor forms of deviance such as absenteeism, lateness, gossiping, and leaving early, as well as more serious forms such as theft, sabotage, aggression, and verbal abuse. These two forms of behaviours can also be further broken down into two categories: (1) interpersonal deviance, behaviours that are directed toward other individuals, such as gossiping and verbal abuse; and (2) organizational deviance, behaviours directed toward the organization, such as stealing from the company and sabotaging equipment (Robinson and Bennett, 1995). This distinction is similar to Greenberg and Scott (1996) who differentiate between employee thefts, which is targeted toward other employees (e.g. stealing a co-worker’s possession), and theft, which is directed toward the organization (e.g. stealing company resources). Other researchers such as Green (1997) and Turner and Stephenson (1993) have also differentiated between targets in their conceptualizations of organizational crimes. While the above conceptualizations of deviance increase our understanding of employee behaviour, these perspectives remain limited in scope. Since employee deviance can also be constructive, it is important to examine both forms of deviance: destructive and constructive deviant behaviours. Below, the concept of constructive deviance is discussed in greater detail.

Constructive deviance Based on the concepts of ‘creative individualism’ (Schein, 1977), ‘productive nonconformity’ (Pepinsky, 1961) and ‘opinion deviance’ (Levine, 1980), Hanke and Saxberg (1985) introduced the concept of constructive deviance. According to Hanke and Saxberg (1985, p. 224), constructive deviance ‘has the positive effect of increasing the diversity of opinion in the group rather than polarizing majority or minority opinions’. More recently, Dehler and Welsh (1998, p. 263) argue that constructive deviance can be conceptualized as ‘breakdowns in the organization’s control systems where employees use discrepant behaviors to advance organizations’ interests’. In line with the definition of workplace deviance presented in this chapter, constructive deviance is defined as: voluntary behaviour that violates significant organizational norms and in so doing contributes to the well-being of an organization, its members, or both. For example, constructive

Bella L. Galperin 159

deviant behaviours may include behaviours that are unauthorized yet facilitate the organizational goals, such as innovative role behaviours, non-compliance with dysfunctional directives, and criticizing incompetent superiors (Ashforth and Mael, 1998). Similarly, employees who engage in discrepant behaviours, such as internal whistle-blowing, may violate the present organizational norms but contribute to the overall well-being of the organization. In this view, internal whistle-blowers may be seen as reformers, whose change efforts are for the organization’s benefit (Graham, 1986; Near and Miceli, 1987). Unlike destructive deviance, constructive deviance has a positive influence on increasing the diversity of a group (Hanke and Saxberg, 1985). Constructive deviants accept the most critical values and goals of the organization but conform selectively to the means for achieving these ends (Hanke and Saxberg, 1985) Dehler and Welsh (1998) argue that selective conformity is central to being a constructive deviant. According to Pepinsky (1961), individuals who do not always conform to the organization’s norms are able to use discrepant cognitions or behaviours to the organization’s advantage. Employees who engage in productive or constructive forms of deviant behaviour may integrate divergent viewpoints more effectively and provide organizations with the necessary creativity. For example, devil’s advocacy, a form of opinion deviance, can play a role in effectively managing conflict. The conflict management literature has recognized that low levels of conflict may be as dysfunctional as high levels of conflict (Brown, 1983). Similarly, the innovation literature suggests that newcomers, entrepreneurs, and other individuals who do not accept the norms of the majority can also be crucial sources of innovation (Dehler and Welsh, 1998; Kanter, 1988; Weick, 1979). Since innovation is the creation and exploitation of new ideas, the very nature of the innovation process departs from the views and opinions held by the majority. For example, Dougherty and Heller (1994) found that product innovation activities in large and mature firms were even considered ‘illegitimate’ because they violated the prevailing organizational norms. Employees who display innovative behaviours can be considered constructive deviants because they engage in acts that initiate a process that departs from the organization’s established routines or systems (Kanter, 1988). Product champions often encounter sharp resistance. Consequently, they must sometimes engage in unorthodox or deviant ways to break away from the current structure and create a new structure (Maidique, 1980).

160 Can Workplace Deviance Be Constructive?

For example, Dr Spencer Silver, a 3M scientist who invented Post-it notes, provides a good illustration of the organizational resistance that constructive deviants must overcome when initiating their novel ideas. In 1968, Dr Silver was looking for a way to improve acrylate adhesives. Instead of discovering a stronger bonding agent, Dr Silver found an adhesive with weaker sticking properties. During the next five years, while holding internal seminars and trying to convince others in his company that his adhesive had great potential, Dr Silver was considered a ‘deviant’ by most of his colleagues. Although Dr Silver was able to get the attention of Geoff Nicholson, 3M’s Commercial Tape Division’s new products development manager in 1973, the two were unsuccessful in swaying the majority opinion. It was only in the late 1970s, when Art Fry, a new product development researcher, was able to overcome 3M’s resistance and the product began commercial trials. After using Dr Silver’s sticking notes on everything, Fry finally managed to convince the engineering and production departments that the product would not be a waste. In 1981, only one year after Post-it notes were introduced, Dr Silver’s invention was named 3M’s Outstanding New Product. Currently, Post-it products are sold in more than a hundred countries across the globe and they come in twenty-seven sizes, fifty-six shapes and fifty colours (Mathews and Wacker, 2002). In sum, the above case illustrates quite dramatically how a ‘deviant’, such as Dr Silver, was able to contribute to the well-being of the organization. Although Dr Silver violated the norms at 3M, he brought much success to the company in the long term. In the sections below, the different forms of constructive deviance and its differentiation from other constructs are highlighted.

What are the different forms of constructive deviance? Constructive deviance can take a variety of different forms. Based on a review of the literature and semi-structured interviews with employees from the pharmaceutical, food and beverage, telecommunication, communications/marketing, computer, and healthcare industries, I identified a range of behaviours needed for developing a measure of constructive deviance. In order to ensure that the content domain of constructive deviance was adequately captured, a group of eight experts were then asked to review the list and provide feedback on the items. After conducting exploratory and confirmatory factor analyses, evidence for a three-factor model of constructive deviance was found (Galperin, 2001).

Bella L. Galperin 161

The first form of constructive deviance, called innovative organizational constructive deviance, reflects innovative behaviours and unconventional ways to help the organization. These behaviours include: searching for innovative ways to perform day-to-day procedures and developing creative solutions to problems. The second form, labelled challenging organizational constructive deviance, describes behaviours that outwardly challenge the existing norms of the organization and break the rules in order to help the organization. Examples include breaking and bending the rules to perform your job and violating company procedures to solve a customer’s problem. Unlike the first two forms which represent acts of constructive deviance that are directed toward the organization, the final form represents acts that are directed toward individuals rather than organizations. Called interpersonal constructive deviance, it includes behaviours performed by people who do not follow the orders of their supervisor or workers who report a wrongdoing to co-workers to bring about a positive organizational change. In sum, these different forms of constructive deviance, while differing slightly, can improve the well-being of the organization even when current norms are violated.

How does constructive deviance differ from other constructs? Constructive deviance is related to concepts such as organizational citizenship behaviour (OCB), whistle-blowing, voice, and role innovation because all the concepts relate to aspects of employee performance that extend beyond the required roles of the organization. Nevertheless, constructive deviance differs from these concepts. OCB has been defined as ‘behaviour that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization’ (Organ, 1990, p. 45). Unlike OCB, which is more of an affiliative behaviour (e.g. helping), and is meant to maintain or improve relationships (LePine and Van Dyne, 1998), constructive deviance violates the organizational norms and may disrupt personal relationships. Moreover, unlike constructive deviance, which requires employees to be proactive, OCB includes dimensions of a more passive nature, such as compliance with procedures (George and Brief, 1992). The concept of constructive deviance is also different from whistleblowing, using out-of-the-company channels or the disclosure of organizational wrongdoing to external parties who can take action (Near and Miceli, 1986). While both constructive deviants and whistle-blowers

162 Can Workplace Deviance Be Constructive?

challenge authority (Near and Miceli, 1987), whistle-blowers may either be viewed as reformers or as dissenters. Unlike constructive deviants who share the goal of improving the organization, the efforts of whistleblowers often threaten the organization’s authority structure (Weinstein, 1979). For example, the whistle-blower who informs authorities outside the organization, such as news media or some agency, can threaten the reputation of the organization. On the other hand, Malin (1983) states that the whistle-blower who uses internal channels may save the company, in the long run, from damaging its reputation by enabling the company to correct its actions before it becomes public knowledge. Miceli and Near’s (1985, p. 541) findings suggest that managers who wish to ‘clean their organization’s dirty laundry in private rather than in public’ may benefit by finding or generating alternatives to current activities that may be viewed as wrong. In line with Miceli and Near (1985), internal whistle-blowing but not external whistle-blowing was found to fit the definition of constructive deviance (Galperin, 2001). In addition, constructive deviance differs from voice or non-required behaviour, which emphasizes expressions for improving rather than merely criticizing the organization (Le Pine and Van Dyne, 1998). Unlike voice, which focuses on making constructive suggestions that may challenge the system, constructive deviance encompasses behaviours which more strongly violate the organizational norms. Finally, constructive deviance differs from role innovation or the introduction of new behaviours into a role (West, 1987). Unlike innovative role behaviour, constructive deviance is a broader concept that includes a larger set of behaviours, such as internal whistle-blowing and behaviours that outwardly challenge the existing norms of the organization. Given the proliferation of constructs in this area, the clarification of construct definition becomes a significant issue to the advancement of the field. In Bennett and Robinson’s (in press, p. 8) upcoming review of workplace deviance research, the authors stress the importance of providing a ‘solid definition of one’s construct before embarking on research’ and note the central role of matching the definition to the operationalizations of the specific constructs. The definitions of destructive and constructive deviance were presented above. The concept of constructive deviance, a multidimensional construct that consists of three distinct but correlated forms, was discussed. While constructive deviance is related to concepts in the literature that relate to aspects of employee performance that extend beyond the required roles of the organization, it differs from other constructs, such as OCB, whistle-

Bella L. Galperin 163

blowing, voice, and role innovation. In the section below, the implications and recommendations for future research are discussed.

Discussion Employees’ adherence to corporate norms, policies, and procedures is essential for an organization’s survival. Employees who fail to follow the accepted rules of behaviour may jeopardize the organization’s overall effectiveness. For example, employee misconduct, such as neglecting to follow one’s boss’s instructions, intentional worker slowdown, lateness, petty theft, and acting rudely toward fellow co-workers, can be very detrimental to the organizational well-being. However, the strict adherence to corporate norms, policies, and procedures may in some cases be undesirable for organizations. When employees strictly follow organizational procedures, innovative and new approaches to problemsolving may be limited. Employees who display innovative behaviours or initiate changes in task objective and processes that are contrary to the work norms, can provide organizations with the necessary innovations and creativity. Given the increasing importance of innovation in today’s environment (Kitchell, 1997), productive deviants can contribute to the overall wellbeing of organizations by developing innovative processes, products, and/or services. Dehler and Welsh (1998) note that organizations have generally not been very successful in developing innovations because of organizational conditions that discourage deviations. Bureaucratic and social strategies are designed to control work processes by increasing desired work behaviour and decreasing the potential for deviance. It is becoming more apparent to organizations that accepting constructive deviance can play a critical role in developing appropriate conditions to increase flexibility. Companies are realizing that when employees display wisdom and imagination, they are able to achieve breakthroughs that can be very profitable for the business (Mariotti, 1999). For example, Motorola invented and dominated the car radio, walkie-talkie, and cell-phone markets. However, Motorola lost its competitive position to Nokia and Ericsson because of the institutionalization of bureaucratic procedures, which failed to sustain wisdom and imagination. The above example illustrates that organizations which are unable to tolerate deviation are often unable to adapt to the changing market and consequently are more likely to fail (Dehler and Welsh, 1998). While it is important to recognize the functional aspects of workplace deviance, the concept of constructive deviance should not be examined

164 Can Workplace Deviance Be Constructive?

in isolation. Destructive and constructive deviance must be integrated into a single theoretical framework. The systematic examination of these two perspectives will provide us with a more coherent understanding of workplace deviance as a broader phenomenon. How is constructive deviance related to destructive deviance? Based on the literature, deviance was conceptualized as both destructive or dysfunctional, and constructive or functional. Given that destructive and constructive deviance both encompasses behaviours that violate the organizational norms, one would expect the two forms of deviance to be positively related to each other. Specifically, it seems that there would be a positive relationship between destructive and constructive deviance. However, since the two forms of deviance differ, the strength of the relationship between constructive and destructive deviance would be low to moderate. Unlike constructive deviance, which comprises discrepant behaviours used to advance the organization’s interests, destructive deviance encompasses behaviours that threaten the wellbeing of the organization. Future research should examine the relationship between destructive and constructive deviance. Specifically, a positive relationship between the two forms of deviance would suggest that employees who engage in harmful deviant acts toward the organization also engage in deviant acts that benefit the organization. Currently, organizations have focused their efforts on identifying individual determinants of deviance. Much of the emphasis has been placed on selecting and recruiting employees who are less likely to be aggressive or dishonest (Collins and Schmidt, 1993). By identifying individuals who are less likely to engage in destructive deviant behaviours, organizations have been able to reduce the occurrence of destructive acts in their organizations. Despite the legal and ethical issues involved in using data based on individual differences, such as criminal records and personality testing, as a basis for hiring decisions, researchers have argued that management should be acquainted with their employees’ histories, traits and behaviours that may be related to destructive deviant behaviours (Martinko and Zellars, 1998). Interestingly, people have not considered the possibility that these individual differences may also be associated with constructive behaviours. By identifying individuals who are less likely to engage in these destructive behaviours, it is possible that organizations may also be selecting individuals who are less likely to challenge the current system in order to bring about a positive change. Future research is needed to uncover the relationship between deviance and individual differences. While studies have not shown a strong

Bella L. Galperin 165

relationship between individual factors and destructive deviance, recent research suggests that the Big Five personality factors, a larger grouping of personality traits, are significantly related to destructive deviance (Cullen and Ones, 2001; Lee, Ashton and Shin, 2001, as cited in Bennett and Robinson, 2000). Researchers should investigate whether the Big Five personality factors are also strongly related to constructive deviance. Moreover, factors in the work environment may also contribute to organizational deviance. Research has shown that the perception of inequitable pay systems is related to increased destructive deviance (Aquino, Lewis and Bradfield, 1999), theft (Greenberg, 1990), and aggression (Greenberg and Alge, 1998). When employees feel that they have been paid unfairly, they are more likely to react negatively. Another factor that appears to be related to employee destructive deviance is powerlessness or the lack of autonomy and participation (Ashforth, 1989). Individuals will engage in greater destructive deviance in order to restore control over their environment. Studies have shown that perceived lack of autonomy and participation are related to increased destructive deviance (Ball, Trevino and Sims, 1994; Bennett, 1998; but see also Vardi and Weiner in this volume for a different view on this issue). One would also expect perceptions of inequitable pay systems and lack of autonomy and participation to be negatively related to constructive deviance. Employees who feel that the pay systems in their organization are inequitable will be less motivated to violate the organizational norms in order to contribute to the well-being of their company. Similarly, environmental factors, such as perceived lack of autonomy and participation, are also expected to be related to constructive deviance. The innovation literature suggests that autonomy may facilitate constructive deviant behaviour (Galbraith, 1982; Kanter, 1988). Employees who have a high degree of autonomy are given the opportunity to deviate from the established routines and procedures. Van de Ven (1986) states that individuals who use this opportunity are engaging in idea generation, the commencement of the innovative process. Future research is needed to examine the relationship between constructive deviance and environmental factors, such as perceptions of justice and job autonomy. The literature suggests that managers should realize that while it is important to have fair compensation systems in place, it is essential that employees perceive the systems to be fair. When employees believe that their organization has a fair pay system, they will be less likely to engage in destructive deviance and may be more likely to engage in

166 Can Workplace Deviance Be Constructive?

constructive deviance. Companies should also recognize that employees must feel that they have control over their own environment. Employees who have perceptions of low job autonomy and participation are more likely to engage in destructive deviant behaviours. They may also be less likely to engage in constructive deviant behaviours. By having a greater understanding of the factors that lead to deviance, managers will be able to create conditions for innovative behaviours that contribute to the organizational effectiveness, as well as reduce the direct and indirect costs associated with employee misconduct. The literature also suggests that employee status may be an important variable in understanding both destructive and constructive deviance. For example, the literature suggests that people who have high status are likely to engage in constructive deviant behaviours (Hanke and Saxberg, 1985). Employees who have high status are more likely to obtain support for deviant perspectives (Hanke and Saxberg, 1985). Similarly, researchers have also examined the relationship between status and destructive deviance. More recently, the findings of Aquino and colleagues (e.g. Aquino, Galperin and Bennett, 2001; Aquino, Tripp and Bies, 2001) suggest that employee status plays a central role in understanding people’s responses to experiences in the workplace. It is only with continued research in the areas of both destructive and constructive deviance that we will be able to gain a more complete understanding of workplace deviance.

Conclusion Unlike the majority of the literature in workplace deviance, this chapter argues that deviance may play a constructive role in enhancing the organization’s well-being. For example, employees who engage in constructive deviant behaviour, such as innovative behaviour, may increase the organization’s competitive advantage. By conceptualizing deviance as constructive or destructive, one is able to obtain a broader perspective of workplace deviance. Merton’s (1957) typology was used to integrate the concepts of destructive and constructive deviance into a single theoretical framework. Furthermore, the definitions of destructive and constructive deviance were outlined. The forms of constructive deviance and its differentiation from other constructs were also presented in order to provide a basis for future research in the area of workplace deviance. I hope that this chapter will shed light on the functional aspects of workplace deviance and inspire scholars to better understand the complex and multifaceted nature of workplace deviance.

Bella L. Galperin 167

References Aquino, K., Lewis, M. U. and Bradfield, M. (1999) ‘Justice constructs, negative affectivity, and employee deviance: a proposed model and empirical test’, Journal of Organizational Behavior, 20, 1073–91. Aquino, K., Galperin, B. L. and Bennett, R. J. (2001) ‘Social status and aggressiveness as moderators of the relationship between interactional justice and workplace deviance’, paper presented at the meeting of the Academy of Management, August, Washington, DC. Aquino, K., Tripp, T. M. and Bies, R. J. (2001) ‘How employees respond to personal offense: the effects of blame attribution, victim status, and offender status on revenge and reconciliation in the workplace’, Journal of Applied Psychology, 86, 52–9. Ashforth, B. E. (1989) ‘The experience of powerlessness in organizations’, Organizational Behavior and Human Decision Processes, 43, 207–42. Ashforth, B. E. and Mael, F. A. (1998) ‘The power of resistance: sustaining valued identities’, in R. M. Kramer and M. A. Neale (eds), Power and Influence in Organizations (pp. 89–119). Thousand Oaks, CA: Sage. Ball, G. A., Trevino, L. K. and Sims, H. P. (1994) ‘Just and unjust punishment: influences on subordinate performance and citizenship’, Academy of Management Journal, 37, 299–322. Baron, R. A. and Neuman, J. H. (1996) ‘Workplace violence and workplace aggression: evidence on their relative frequency and potential causes’, Aggressive Behavior, 22, 161–73. Bennett, R. J. (1998) ‘Perceived powerlessness as a cause of employee deviance’, in S. B. Bacharach (ed.), Dysfunctional Behavior in Organizations: Violent and Deviant Behavior (pp. 221–39). Stamford, Connecticut: JAI Press. Bennett, R. J. and Robinson, S. L. (2000) ‘The development of a measure of workplace deviance’, Journal of Applied Psychology, 85, 349–60. Bennett, R. J. and Robinson, S. L. (in press). ‘The past, present, and future of workplace deviance research’, in J. Greenberg (ed.), Organizational Behavior: The State of the Science, 2nd edn. Hillsdale, NJ: Lawrence Erlbaum. Bensimon, H. F. (1994) ‘Crisis and disaster management: violations in the workplace’, Training and Development, 28, 27–32. Bord, R. J. (1976) ‘The impact of imputed deviant identities in structuring evaluations and reactions’, Sociometry, 39, 108–15. Brown, L. D. (1983) Managing Conflict at Organizational Interfaces. Reading, MA: Addison-Wesley. Cohen, A. K. (1966). Deviance and Control. Englewood Cliffs, NJ: Prentice Hall. Collins, J. M. and Schmidt, F. L. (1993) ‘Personality, integrity, and white collar crime’, Personnel Psychology, 46, 295–311. Cullen, M. J. and Ones, D. S. (2001) ‘The role of conscientiousness and neuroticism in predicting police corruption’, paper presented at the meeting of the Society for Industrial & Organizational Psychology, April. San Diego, CA. Dehler, G. E. and Welsh, M. A. (1998) ‘Problematizing deviance in contemporary organizations: a critical perspective’, in R. W. Griffin, A. O’Leary-Kelly and J. M. Collins (eds), Dysfunctional Behavior in Organizations: Violent and Deviant Behavior (pp. 241–69). Stamford, Conn.: JAI Press. Dougherty, D. and Heller, T. (1994) ‘The illegitimacy of successful new products in large firms’, Organization Science, 5, 200–18.

168 Can Workplace Deviance Be Constructive? Filipczak, B. (1993) ‘Armed and dangerous at work’, Training, July, 39–43. Freedman, J. and Doob, A. (1968) Deviancy: The Psychology of Being Different. New York: Academic Press. Galbraith, J. R. (1982) ‘Designing the innovating organization’, Organizational Dynamics, Winter, 5–25. Galperin, B. L. (2001) ‘The development of a measure of workplace constructive deviance’, unpublished manuscript, Concordia University, Montreal, Canada. George, J. M. and Brief, A. P. (1992) ‘Feeling good–doing good: a conceptual analysis of the mood at work-organizational spontaneity relationship’, Psychological Bulletin, 112, 310–29. Giacalone, R. A. and Greenberg, J. (1997) Antisocial Behavior in Organizations. Thousand Oaks, CA: Sage. Graham, J. W. (1986) ‘Principled organizational dissent: a theoretical essay’, in L. L. Cummings and B. M. Staw (eds), Research in Organizational Behavior, Vol. 8 (pp. 1–52). Greenwich, CT: JAI Press. Green, G. S. (1997) Occupational Crime, 2nd edn. Chicago: Nelson-Hall. Greenberg, J. (1990) ‘Employee theft as a reaction to underpayment inequity: the hidden cost of pay cuts’, Journal of Applied Psychology, 75, 561–8. Greenberg, J. and Alge, B. J. (1998) ‘Aggressive reactions to workplace injustice’, in R. W. Griffin, A. O’Leary-Kelly and J. Collins (eds), Dysfunctional Behavior in Organizations: Violent Behaviors in Organizations, Vol. 23 (pp. 83–118). Greenwich, CT: JAI. Greenberg, J. and Scott, K. S. (1996) ‘Why do workers bite the hands that feed them? Employee theft as a social exchange process’, in B. M. Staw and L. L. Cummings (eds), Research in Organizational Behavior, Vol. 18 (pp. 111–56). Greenwich, CT: JAI. Hanke, J. J. and Saxberg, B. O. (1985) ‘Isolates and deviants in the United States and Japan: productive nonconformists or costly troublemakers?’, in R. F. Tomasson (ed.), Comparative Social Research, Vol. 8 (pp. 219–45). Greenwich, CT.: JAI. Howell, J. M. and Higgins, C. A. (1990) ‘Champions of technological innovation’, Administrative Science Quarterly, 35, 317–41. Howell, J. M., Shea, C. M. and Higgins, C. A. (1998) ‘Champions of product innovations: defining, developing and validating a measure of champion strength’, unpublished manuscript, The University of Western Ontario. Kanter, R. M. (1988) ‘When a thousand flowers bloom: structural, collective, and social conditions for innovation in organization’, Research in Organizational Behavior, 10, 169–211. Kitchell, S. (1997) ‘CEO characteristics and technological innovativeness: a Canadian perspective’, Canadian Journal of Administrative Sciences, 14, 111–25. Kondrasuk, J. N., Moore, H. L. and Wang, H. (2001) ‘Negligent hiring: the emerging contributor to workplace violence in the public sector’, Public Personnel Management, 30, 185–95. Lee, K., Ashton, M. C. and Shin, K. (2001) ‘Personality correlates of workplace antisocial behavior’, paper presented at the meeting of the Academy of Management, August, Washington, D.C. Le Pine, J. A. and Van Dyne, L. (1998) ‘Predicting voice behavior in work groups’, Journal of Applied Psychology, 83, 853–68. Levine, J. M. (1980) ‘Reaction to opinion deviance in small groups’, in P. B. Paulus (ed.), Psychology of Group Influence. Hillsdale, NJ: Lawrence Erlbaum.

Bella L. Galperin 169 Maidique, M. A. (1980) ‘Entrepreneurs, champions, and technological innovation’, Sloan Management Review, Winter, 59–76. Malin, M. H. (1983) ‘Protecting the whistle-blower from retaliatory discharge’, University of Michigan Journal of Law Reform, 16, 277–318. Mariotti, J. (1999) ‘Four formulas for success’, Industryweek.Com, October, 4. Martinko, M. J. and Zellars, K. L. (1998) ‘Toward a theory of workplace violence and aggression: a cognitive appraisal perspective’, in R. W. Griffin, A. O’Leary-Kelly and J. M. Collins (eds), Dysfunctional Behavior in Organizations: Violent and Deviant Behavior (pp. 1–42). Stamford, Conn: JAI Press Inc. Mathews, R. and Wacker, W. (2002) The Deviant’s Advantage: How Fringe Ideas Create Mass Markets. NY: Crown Business. Merton, R. K. (1957) Social Theory and Social Structure. New York: The Free Press. Miceli, M. P. and Near, J. P. (1985) ‘Characteristics of organizational climate and perceived wrongdoing associated with whistle-blowing decisions’, Personnel Psychology, 38, 525–44. Near, J. P. and Miceli, M. P. (1987) ‘Whistle-blowers in organizations: dissidents or reformers?’, in L. L. Cummings and B. M. Staw (eds), Research in Organizational Behavior, Vol. 9 (pp. 321–68). Greenwich, CT: JAI Press. O’Leary-Kelly, A. M., Griffin, R. W. and Glew, D. J. (1996) ‘Organizationmotivated aggression: a research framework’, Academy of Management Review, 21, 225–53. Organ, D. W. (1990) ‘The motivational basis of citizenship behavior’, in B. M. Staw and L. L. Cummings (eds), Research in Organizational Behavior, Vol. 12 (pp. 43–72). Greenwich, CT: JAI Press. Pepinsky, P. (1961) ‘Social exceptions that prove the rule’, in I. A. Berg and B. M. Bass (eds), Conformity and Deviation. New York: Harper & Bros. Robinson, S. L. and Bennett, R. B. (1995) ‘A typology of deviant workplace behaviors: a multi-dimensional scaling study’, Academy of Management Journal, 38, 555–72. Robinson, S. L. and Bennett, R. J. (1997) ‘Workplace deviance: its definition, its nature and its causes’, in R. J. Lewicki, B. H. Sheppard and R. J. Bies (eds), Research on Negotiation in Organization, Vol. 6 (pp. 3–28). Greenwich, CT: JAI. Robinson, S. L. and Greenberg, J. (1998) ‘Employees behaving badly: dimensions, determinants, and dilemmas in the study of workplace deviance’, in C. L. Cooper and D. M. Rousseau (eds), Trends in Organizational Behavior, Vol. 5 (pp. 1–30). New York: Wiley. Schein, E. H. (1977) ‘Organization socialization and the profession of management’, in B. B. Staw (ed.), Psychological Foundations of Organizational Behavior (pp. 9–23). Santa Monica, CA: Goodyear. Skarlicki, D. P. and Folger, R. (1997) ‘Retaliation in the workplace: the roles of distributive, procedural and interactional justice’, Journal of Applied Psychology, 82, 416–25. Tripp, T. M. and Bies, R. J. (1997) ‘Seeking revenge in organizations: an exploration into the hearts and minds of avengers’, in R. J. Lewicki, R. J. Bies and B. H. Sheppard (eds), Research on Negotiations in Organizations, Vol. 6 (pp. 145–62). Greenwich, CT: JAI. Turner, D. L. and Stephenson, R. G. (1993) ‘The lure of white-collar crime’, Security Management, February, pp. 57–8. Van de Ven, A. H. (1986) ‘Central problems in the management of innovation’, Management Science, 32, 590–607.

170 Can Workplace Deviance Be Constructive? Vardi, Y. and Wiener, Y. (1996) ‘Misbehavior in organizations: a motivational framework’, Organizational Science, 7, 151–65. Weick, K. E. (1979) The Social Psychology of organizations. Reading, Mass: AddisonWesley. Weinstein, D. (1979) Bureaucratic Opposition. New York: Pergamon Press. West, M. A. (1987) ‘A measure of role innovation at work’, British Journal of Social Psychology, 26, 83–5.

Part II The Dynamics of Organizational Misbehaviour

This page intentionally left blank

10 Personal and Positional Antecedents of Organizational Misbehaviour Yoav Vardi and Ely Weitz*

Introduction There seems to be a new realization among scholars that intentional acts of misbehaviour are a prevalent aspect of organizational behaviour and therefore merit scientific scrutiny (Bamberger and Sonnenstuhl, 1998; Griffin, O’Leary-Kelly and Collins, 1998; Sackett and DeVore, 2001). Not only are these acts common among members of organizations but their costs for employers and society at large are enormous (Bennett and Robinson, 2000; Murphy, 1993). Our chapter supports this view by presenting a general framework and empirical evidence that some personal and positional variables may be regarded as antecedents of various forms of organizational misbehaviour. We argue that one of the important factors conducive to misbehaviour in the workplace is the actual opportunity to engage in it. Such an opportunity may be related to specific characteristics of the job and of its surroundings. That is to say, the level of control, or alternatively the level of autonomy inherent in the job, creates a structure of opportunity that facilitates not only desirable and normative behaviours, but also various forms of misbehaviour. Most of the research on the effect of job autonomy on employees has, not surprisingly, focused on positive work outcomes such as performance and satisfaction (Breaugh, 1985; Hackman and Oldham, 1980). We have identified only a few studies whose focus was on negative outcomes (Allen and Greenberg, 1980; Dwyer and Fox, 2000; Wortman and Breham, 1975). The field study we report here was designed specifically to examine whether autonomy on the job as perceived by the incumbent *Preliminary versions of this chapter were presented at the 7th ISSWOV conference, Jerusalem, 2000, and the Academy of Management meeting, Washington, DC, 2001.

173

174 Personal and Positional Antecedents of Misbehaviour

is indeed related to various manifestations of misbehaviour, and whether other personal and organizational variables affect these relationships.

Manifestations and definitions of organizational misbehaviour Misbehaviour by organization members has received numerous designations over the years – workplace deviance, counterproductive behaviour, dysfunctional behaviour, antisocial behaviour, workplace aggression, organizational misbehaviour, and so forth (Robinson and Greenberg, 1998). The numerous labels and conceptualizations reflect the complexity and the multi-dimensionality of this phenomenon on the one hand, and the different disciplinary affiliations of the researchers, on the other. For example, Griffin et al. (1998) focuses on dysfunctional behaviour and defines it as motivated behaviour by an employee or a group that has negative consequences for individuals, groups, or the organization as a whole. Giacalone and Greenberg (1997) treated antisocial behaviour in organizations as any behaviour that brings harm, to an organization, its employees, or its stakeholders. Hollinger and Clark (1982) view deviant behaviour in the workplace as behaviour that violates the organization’s formal rules. They distinguish between ‘property deviance’, which refers to damaging the organization’s resources, and ‘production deviance’, which includes failing to comply with the set production standards of quality and quantity. Robinson and Bennett (1995), and Bennett and Robinson (2000) too, view misbehaviour on the job as deliberate action. They define workplace deviance as voluntary behaviour directed against the organizational norms, threatening the survival of the organization and the welfare of its members. Using a scaling statistical method they empirically derived a classification of workplace deviant behaviour based on whether the act is organizational deviance (acts intended to harm the organization) or interpersonal deviance (behaviours directly harmful to other members of the organization), and whether it is severe or minor. The four types are: property deviance (sabotage, theft), characterized as very severe behaviour in the organizational sphere; production deviance (taking many breaks, slow work pace), characterized as non-severe behaviour in the organizational sphere; political deviance (gossip, nepotism) – nonsevere behaviour in the interpersonal sphere; personal aggression (violence, sexual harassment) – severe behaviour in the interpersonal sphere. A recent review by Sackett and DeVore (2001) defines counterproductive behaviour as any intentional behaviour on the part of an organizational

Yoav Vardi and Ely Weitz 175

member viewed by the organization as contrary to its legitimate interests. It encompasses a broad number of domains. Gruys (1999) identified 87 separate counterproductive behaviours appearing in the literature and used a sorting procedure and factor analytic techniques to produce 11 categories: Theft and related behaviour (theft of cash or property, giving away of goods or services, misuse of employee discount); Destruction of property (deface, damage, or destroy property; sabotage production); Misuse of information (reveal confidential information, falsify records); Misuse of time and resources (waste time, alter time card, conduct personal business during work time); Unsafe behaviour (fail to follow safety procedures, failure to learn safety procedures); Poor attendance (unexcused absence or tardiness, misuse sick leave); Poor quality work (intentionally slow or sloppy work); Alcohol use (alcohol use on the job, coming to work under the influence of alcohol); Drug use (possess, use, or sell drugs at work); Inappropriate verbal actions (argue with customers, verbally harass co-workers); Inappropriate physical actions (physically attack co-workers, physical sexual advances toward co-worker). Sackett and DeVore (2001) attribute such behaviours to both distal antecedents (situational features such as company policy and work characteristics) and proximal antecedents (personality and perception variables). They also suggest that proximal variables can serve as mediators between situational factors and counterproductive manifestations. Unlike Robinson and Bennett (1995), Vardi and Wiener (1996) subsume that misbehaviour should not be treated as deviant behaviour (rare or marginal) but rather as an integral and prevalent aspect of everyday work-related activities in organizations. They also argue that such acts are not necessarily intended by perpetrators to be harmful or counterproductive. Vardi and Wiener proposed that organizational misbehaviour (OMB) pertains to any intentional act by an organizational member which violates the organization’s shared norms and/or the accepted social values and standards of proper behaviour. Their general typology of organizational misbehaviours is based on the actor’s main motives underlying these acts: benefiting the employee (OMB Type S), such as cover-up or theft; benefiting the organization (OMB Type O) such as misleading customers; and inflicting damage (OMB Type D) such as harassment and vandalism.

Antecedents of organizational misbehaviour The need to better understand job-related misbehaviours, and the organizational need to control them, led some researchers to propose

176 Personal and Positional Antecedents of Misbehaviour

models defining their possible causes. Griffin et al. (1998) relate work behaviours to choices individuals or groups make, which in turn form the intentions to behave in a functional or dysfunctional way. Robinson and Greenberg (1998) view workplace misconduct as a multiple-step, dynamic model, suggesting a variety of actors and events: perpetrators – intentions – targets – actions – consequences. In a similar vein, the conceptual framework proposed by Vardi and Wiener (1996) posits that certain organizational and personal factors might interact to serve as catalysts of misbehaviour. Among the personal factors are the worker’s level of moral development, the value system internalized by the worker, the level of congruence between organizational and personal values, and the company’s failure to satisfy personal needs. The organizational factors conducive to OMB include job pressures, supervisory style, management and control systems, organizational goals, culture, climate and social cohesiveness. They also proposed that built-in opportunities at both the position and organization levels might interact with certain individual factors in ways that encourage actual misbehaviour. In the present study, therefore, we suggest that autonomy on the job is a plausible, yet under-researched, antecedent of organizational misbehaviour perpetrated and reported by individual members.

Job characteristics The literature discussing job autonomy as an organizational variable can be divided into two domains: control theories and job design theories. Control theories treat organizational control as a variable influencing workers’ attitudes and behaviours. Since autonomy is viewed as a type of control, such theories attempt to explicate its effect on employee behaviour by examining the effects of different types of control (Ganster, 1989; Karasek and Theorell, 1990; Spector, 1986; Thompson, 1981). Job design theories regard autonomy as a form of work enrichment and assume that job autonomy will positively influence the individual’s attitudes and behaviour by increased empowerment (Abdel Halim, 1978; Lee, Ashford and Bobko, 1990; Pines, 1981). The best-known framework conceptually dealing with job design is the Job Characteristics Model developed and advocated by Hackman and Oldham (1976; 1980). They defined the core dimension of autonomy as the way in which the position or job provides individuals with a meaningful degree of freedom, independence and personal initiative in organizing their work and controlling performance. Autonomy is assumed to engender a certain critical psychological state (responsibility for results), which, in turn, may contribute to desired outcomes

Yoav Vardi and Ely Weitz 177

such as attachment, satisfaction, internal motivation and effective performance. Combining these approaches, Breaugh (1985) defines job autonomy as workers’ level of freedom to decide on work methods and schedules and on the criteria for evaluating their work. To assess these characteristics, he developed a research instrument which measures the degree of discretion employees have with regard to various dimensions of their work: work methods – the degree to which individuals are free to conduct and choose processes and methods for performing the work; scheduling – the degree to which workers feel that they have control over their schedule; and performance criteria – the degree to which workers are able to set or choose the standards for evaluating their own performance. Very few studies have suggested job autonomy as a possible cause of misbehaviour. Researchers who did so (e.g. Allen and Greenberg, 1980; Vaught and Smith, 1980; Wortman and Breham, 1975) showed that job autonomy is likely to increase the tendency to take advantage of the lack of external control. Analoui and Kakabadse (1992), on the other hand, noted that possibly one of the motives for engaging in unconventional practices, which they observed both among managers and subordinates, was the desire for more autonomy on the job. Molstad (1988) also found that in the face of routine work and extensive supervision, workers developed various action strategies to give themselves a sense of control and autonomy in their work. For example, they created an impression among the managers that they were working hard and were not free to take on more work, or they developed a private language that the managers could not understand. This autonomous behaviour gave the workers a sense of control, despite the extensive supervision on the part of the managers. Just recently, Dwyer and Fox (2000) provided further, albeit indirect, support for possible negative effects of autonomy. In a longitudinal study of nurses they found that perceptions of job control were positively related to employee hostility. Similarly, Rentsch and Steel (1998) presented longitudinal evidence that autonomy, skill variety, and task identity were correlated with measures of absence from work. In a similar vein, some researchers argue that a great deal of job freedom is precisely what might lead to misbehaviour by providing the actual opportunity to engage in actions that may be viewed as counterproductive or even dysfunctional. Ackroyd and Thompson (1999) referred to this as ‘irresponsible’ autonomy. Mars (1982), for example, found that differences in the level of supervision over the employee’s attendance

178 Personal and Positional Antecedents of Misbehaviour

at work were related to the opportunity structure for deviance. Vaught and Smith (1980) studied the behaviour of miners and showed how lack of control enhances interpersonal misconduct. Both Analoui and Kakabadse (1992) and Klein, Leong and Silva (1996) posit that sabotage on the job could result from independent attempts by employees to alleviate boredom. Vardi and Wiener (1996) also suggested that organizations with sensible control systems are more effective in controlling their employees’ behaviour than organizations with either very restrictive or very flexible systems. The latter are liable to increase the motivation for misbehaviour. Thus, both high and low autonomy might be conducive to improper on-the-job conduct. Overall, the literature on job characteristics and employee behaviour appears to imply that the freedom to make choices, and the variety of opportunities for action that stem from them, may indeed lead to intentional misconduct. This claim has yet to be tested empirically. Therefore, in this study we suggest that such features, alongside their well-established positive effects (Hackman and Oldham, 1980; Herzberg, 1968), are also likely to facilitate manifestations of OMB. As suggested by Breaugh (1985), autonomy, by definition, accords employees a great deal of choice. Consequently, when employees perceive their position as autonomous, these perceptions may facilitate (tempt) such misbehaviours as purposely working at a slow pace, stealing time, engaging in private activities during work hours, and otherwise abusing organizational resources. Based on this argument, we hypothesize that job autonomy and production-related misbehaviour are positively correlated: the more the perceived autonomy on the job, the higher the reported misbehaviour (hypothesis 1).

Additional individual and positional differences The simple hypothesized effect of job autonomy on OMB is sometimes difficult to observe because the job–behaviour relationship may be a complex one. For instance, Hackman and Oldham (1980) argue that the effects of the job’s core dimensions on a person’s job attitudes and behaviours are mediated by certain psychological states and moderated by the person’s levels of growth needs. These relationships may be affected by certain individual variables such as personality traits and dispositions toward work and organizations. Furthermore, models dealing with unethical and improper behaviour (Robinson and Greenberg, 1998; Trevino, 1986; Vardi and Wiener, 1996) posit that not only job characteristics affect the propensity for OMB, but also individual antecedents (e.g. personality and attitudinal variables), and situational

Yoav Vardi and Ely Weitz 179

constraints such as positional and organizational features. In the following sections we review some of the relevant models and studies that relate to our hypotheses regarding the role of some selected and presumably relevant, individual and positional traits.

Type-A personality . The ‘Type-A’ behaviour syndrome is a catch-all phrase for a combination of characteristics such as ambition, competitiveness, impatience, hostility potential, over-involvement in work, aggression, hastiness, high achievement need, over-alertness and constant sense of lack of time. Studies examining the link between Type-A and productivity show that people with Type-A behaviour exhibit faster or higher quality performance (Glass, 1977; Jamal, 1985; Lee and Gillon, 1989). Chesney and Rosenman (1980) claim that Type-As undertake considerable responsibility at work and also function well under conditions of pressure because of their deep devotion and commitment to the accomplishment of the task. Matthews (1982) showed that Type-As set high-level goals for themselves in the belief that they will overcome the difficulties they encounter. For the same reason they also tend to work on several projects simultaneously. Ganster (1989) claims that Type-As respond in a more extreme manner to stressful situations at work, and Goldshmid-Aron (1997) found that among managers measures of Type-A behaviour and workaholism were positively related. These and other studies seem to indicate that Type-A behaviour characteristics may also motivate employees to engage in improper action in the workplace. That is to say, workaholism, competitiveness, aggression, urgency, constant sense of lack of sufficient time, and internal pressure to succeed, are liable to drive Type-As to organizational misbehaviour as an adaptive reaction to such perceived work-related demands. Therefore, we hypothesize that, as compared with non-Type-As, employees with Type-A personality will reveal a higher tendency for engaging in organizational misbehaviour (hypothesis 2). Professional identity and commitment . Meyer, Allen and Smith (1993) identified three dimensions of commitment to one’s profession: commitment as an emotion (affective), commitment as a perception of the cost of leaving the occupation or instrumental continuance, and commitment as a sense of responsibility with regard to remaining in the occupation or the workplace (normative). The authors claim that affective commitment or commitment out of a sense of duty to remain in the occupation, is positively related to task performance and to organizational citizenship behaviour (positive extra-role acts). Focusing on the

180 Personal and Positional Antecedents of Misbehaviour

commitment–behaviour relationship, Cohen (1999) found that, in fact, each of the dimensions of commitment has a negative relation to some aspects of OMB. That is, committed employees are less likely to misbehave on the job. In addition, it has been shown that a strong identification with one’s profession helps to strengthen professional orientation and work ethics (Raelin, 1986). For example, in a study conducted among insurance agents, those with a strong professional identity were found to maintain higher ethical standards (Hutnik, 1989). Overall, such studies suggest that the higher the professional identity or commitment, the lower the tendency to engage in self-benefiting or destructive acts on the job. We hypothesize that there will be a negative correlation between professional identity and misbehaviour: the higher the professional identity the lower the reported misbehaviour (hypothesis 3).

Job dissatisfaction. It is widely assumed that employees who do not like their work will withdraw from it either permanently (turnover) or temporarily (excessive tardiness and absenteeism) (see review of job satisfaction research in Spector, 1997a). Similarly, job dissatisfaction may positively affect OMB. Hollinger and Clark (1982) found a negative relationship between employee deviance and both general satisfaction and facet satisfaction (e.g. with the direct superior, co-workers, work load, promotion and salary). Hirschman (1970) argued that dissatisfied employees might react to unfavourable conditions in the workplace by leaving the organization, actively voicing their complaint, or by neglecting their duties or by complacency. This ‘voice, loyalty and neglect’ paradigm was supported by empirical studies (Farrell, 1983; Meyer et al., 1993). Similarly, Vardi and Wiener (1996) proposed that the intention to engage in OMB is related to the level of dissatisfaction of individual needs, and Spector (1997b) attributed antisocial behaviour at work to job-related experienced frustration. Yosifon (2001) recently reported significant correlations between various facets of work satisfaction and both self-reported and other-reported measures of misconduct. Therefore, we hypothesiz that affective job satisfaction and the intention to misbehave are negatively related (hypothesis 4). As suggested above, we surmised that misbehaviour on the job is not only a function of the level of autonomy inherent in the job, but also of personality and attitudinal variables. Thus, we expected both direct and indirect (interaction) effects. For example, because we assumed that freedom of action permits dissatisfied organization members to express their negative affect in non-compliant behaviour, we can expect that among dissatisfied employees, the higher the level of autonomy the

Yoav Vardi and Ely Weitz 181

stronger the tendency for OMB. Similarly, Type-A personality, when coupled with high autonomy, may result in higher personal OMB, especially when these two conditions interact with some built-in opportunities inherent in certain managerial positions. It is thus hypothesized that the variance in self OMB will be accounted for by both main and interaction effects of autonomy, Type-A, professional identity, dissatisfaction, and position level (hypothesis 5).

The study Data to test these hypotheses were collected in a large privately run personnel placement centre in Israel. All the participants, at the time of the study, were in the process of being evaluated by the centre for some career transition, and thus represented a random variety of individuals, jobs and organizations. This type of heterogeneous research population was necessary for maximizing the variance of the study variables. At the same time the fact that all were responding about a job that they were about to leave, held their perspective about which job to think about, constant. In order to protect the subjects’ anonymity the procedure devised together with the director of the centre was as follows: The house psychologist presented to the participants the research questionnaire and explained its purpose, emphasizing the fact that this survey was not part of their job-transition assessment. Then they were asked to fill out the questionnaire sometime during the day. The psychologist answered their questions as instructed by the research team, and collected the questionnaires when completed. She later delivered them to the research team maintaining full anonymity of the subjects. Our sample comprised 250 employees from various organizations and diverse jobs. It included 134 men and 116 women, ranging in age from 23 to 35. Of these, 32.4 per cent had secondary level education, 47.6 per cent had 15 years of education, and 20 per cent 16–17 years. Their last jobs were in sales (13 per cent), computers (10.8 per cent), engineering (16 per cent), economics (8.4 per cent) or occupations such as clerks, manpower and advertising. Thirty-six per cent of the employees worked in field positions and 50 per cent held managerial roles. Most of the subjects had worked up to 5 years in their last job. The questionnaire consisted of two major scales and four additional measures (for details, see Biran, 1999): 1. Organizational misbehaviour. Thirteen behavioral items pertaining to production misbehaviour were adapted from Robinson and Bennett

182 Personal and Positional Antecedents of Misbehaviour

2.

3.

4.

5.

(1995) and translated into Hebrew. Special care was given to the translation process (see Brislin, 1970) in order to maintain authenticity of the items while adapting them to colloquial Hebrew. Two types of lead questions, one pertaining to misbehaviour by others and one to misbehaviour of the subject, were used in order to generate two OMB subscales: (a) Others’ misbehaviour in the organization. In this part the items were presented as indirect questions, relating to other people in the organization (α = .63). For example, ‘To what extent do workers at your workplace lie about their work hours, or make deliberate mistakes?’ The possible replies ranged from ‘not at all’ (1) to ‘very much’ (7). (b) Self-reported misbehaviour in the organization. The same items were presented as directly relating to the employee’s personal behaviour at work (α = .88). Job autonomy was examined by the subjects’ self-report about their last job. Nine items were adopted from Breaugh (1985). The translation into Hebrew followed the same procedure as described above. The first three items relate to the work methods – for example, ‘I could decide how to do my work’; the next three items examine the degree of autonomy in setting work schedules – for example, ‘My work allows me to decide when to do each action’; and the last three items refer to autonomy in evaluating the outcomes – for example, ‘I could choose what goals and tasks to accomplish and complete’. The subjects were asked to mark the answer on a Likert type scale ranging from ‘not at all’ (1) to ‘to a very large extent’ (7). The mean served as the score for overall job autonomy (α = .93). Professional identity. To measure professional identity and commitment we used 6 items from a commitment questionnaire by Meyer et al. (1993) tapping the level of identification with the profession and the affective commitment to the occupation (α = .89). For example, ‘I am proud to work in my profession’. Answers ranged from ‘not at all’ (1) to ‘very much’ (7). The Hebrew version was adopted from Goldshmid-Aron (1997). Type-A personality was measured with 9 items with a four-point scale ranging from 1 (‘not true at all’) to 4 (‘very true’). For example, ‘Sometimes I feel that I shouldn’t work so hard, but something drives me on’. A subject with a mean score above 2 was defined as Type-A. The internal reliability of the measure was α = .80. The Hebrew version, originally developed by Caplan, Cobb and French (1975) was also adopted from Goldshmid-Aron (1997). Overall job satisfaction was simply measured on a scale of 1 to 5 by five generic faces arranged in a row, with universal expressions showing

Yoav Vardi and Ely Weitz 183 Table 10.1

Means, standard deviations, and correlations among study variables

Variables

Mean

SD

1

1. OMB self 2. OMB other 3. Autonomy 4. Professional identity 5. aType-A 6. Satisfaction 7. Organizational levelb

1.86 2.64 4.69 5.39

0.40 0.70 0.38*** 1.39 0.26*** 1.37 −0.17**

2.95 3.85 –

0.56 −0.02 0.96 −0.20** −0.04 –

2

3

0.14* 0.02

0.31***

4

5

6

0.18** 0.21*** .27*** −0.18* 0.14* 0.33*** 0.16** 0.14* 0.29*** 0.12* 0.22** 0.10

N = 250 a N = 202 b Non managerial level = 0 (N = 123), Managerial level = 1 (N = 127). * p < .05, ** p < .01, *** p < .0 01

varying levels of content or discontent. Subjects were asked to mark the face that best matched their overall sense of satisfaction from their last job. This measure is based on the famous Kunin (1955) General Motors Faces Scale. Finally, participants were asked whether on their last job they had been in charge of other people or whether their position was considered managerial at their organization. The 127 subjects (50.8 per cent) who answered positively were coded as managers and all others as nonmanagers. Before examining the specific hypotheses we looked at the relationship between the self-report of misbehaviour (OMB self) and the report on others (OMB others). Not surprisingly, the mean of reporting on OMB of others was significantly higher (Mean = 2.64) than the mean for reporting self OMB (Mean = 1.64), (t = 12.46; p < .001). Also, there was a significant positive correlation between the two measures (r = .38; p < .001). We took this to mean that while the two measures tap different individual perspectives they are also interrelated; reporting on misbehaviour exhibited by others may, to some extent, project on the self. Table 10.1 presents the means, standard deviations and correlations among the study variables. There is a significant positive correlation between job autonomy and misbehaviour, both in reporting on others and on the self, with the correlation between autonomy and self-reported OMB higher than the correlation between autonomy and OMB reported on others.

184 Personal and Positional Antecedents of Misbehaviour Table 10.2 Relationships between self-OMB and overall job autonomy and its components

Overall job autonomy Autonomy in work methods Autonomy in scheduling Autonomy in setting performance standards

OMB self

OMB others

0.26*** 0.16** 0.27*** 0.24***

0.14* 0.14* 0.09 0.15*

* p < .05, ** p < .01, *** p < .001

Thus, the main research proposition that the perceived autonomy and discretion inherent in the job may be an important antecedent of OMB is supported. In addition to the correlation between general autonomy and OMB, a significant positive correlation was found between each individual dimension of autonomy and OMB of the self and of others, with the exception of the correlation between autonomy in scheduling and OMB of others, which was positive but non-significant (see Table 10.2). Hypothesis 2 concerning a positive correlation between Type-A and OMB was partially supported. A positive correlation was found between Type-A and OMB only when the reporting was on others’ behaviour. Thus, Type-A subjects seem to report more OMB of others than of themselves. Hypothesis 3 concerning the negative correlation between professional identity and OMB was supported in part. A negative correlation was found between professional identity and self-OMB but not with others-OMB. That is, the higher the professional identity, the lower the self-reported OMB. A statistically significant negative correlation was found between satisfaction and misbehaviour in both reporting on self and in reporting on others (hypothesis 4). This may indicate that with the decline in satisfaction there is an increase in reporting on misbehaviour of the self and of others (r = − .20; r = −.18, p < .01, respectively). Finally, there is a positive correlation between rank in the organization and reporting OMB of others (r = .14; p < .05). The correlation with self-OMB, however, was not significant. It seems reasonable that managers, responsible for other employees or to certain functional areas, will report on others’ misconduct more than their own. The overall argument that personal and organizational variables influence the relationship between job autonomy and OMB was examined by analysis of variance. As may be seen in Table 10.3, autonomy, professional identity, and job satisfaction each has a significant main

Yoav Vardi and Ely Weitz 185 Table 10.3 Analysis of variance with self-OMB as the dependent variable: main and interaction effects

Autonomy Type-A Autonomy × Type-A Professional identity Autonomy × Professional identity Satisfaction Autonomy × Satisfaction Managerial job Managerial job × Autonomy Type-A × Managerial job × Autonomy Error

DF

SS

MS

F

2 1 2 2 4 4 8 1 2 2

4.360 0.005 0.620 2.600 9.300 7.620 5.070 0.001 0.800 2.910 198

2.180 0.005 0.310 1.300 2.320 1.900 0.630 0.001 0.440 1.450 77.15

5.600** 0.010 0.800 3.340* 5.970*** 5.790*** 1.930 0.000 1.130 3.740* 0.389

* p < .05, ** p < .01, *** p < .0 01

effect on self-reported OMB. In addition, several interaction terms appear significant: autonomy times identity, autonomy times job satisfaction, and autonomy times position level times Type-A personality. Remember that we wanted to look at these interactions because of the underlying assumption that misbehaviour on the job stems not only from mere seizing of opportunity, but from some combination of opportunity and motivation. While our results fall short of conclusive evidence of moderating effects, they do suggest that the level of autonomy in the job alone does not suffice to predict employees’ intentions to misbehave. Additional factors are in place. Thus, Hypothesis 5 is partially supported.

Discussion During the recent decade, students of organizations began to map and define the misbehaviour phenomena and to identify the causes of such behavioural patterns (Ackroyd and Thompson, 1999; Robinson and Bennett, 1995; Robinson and Greenberg, 1998; Sackett and DeVore, 2001; Vardi and Wiener, 1996). This effort has significantly balanced the positive bias that had dominated the early research on organizational behaviour which has led to the development of models that tend to emphasize normative and desirable attitudes and behaviours, while ignoring negative and dysfunctional ones (Griffin et al., 1998). Here we took the position that the field is now ready for unbiased systematic explorations of the domain that Vaughan (1999) aptly describes as the

186 Personal and Positional Antecedents of Misbehaviour

dark side of the organization. We also surmised that the very factors and variables affecting good behaviour might also explain misbehaviour. Specifically, the purpose of this study was to explore the relationships between job autonomy and organizational misbehaviour, broadly defined as those intentional acts on the job that run counter to standards of proper work conduct. We focused on autonomy precisely because most models and studies of the effects of the job structure on employee behaviour emphasized positive outcomes (Hackman and Oldham, 1980; Sackett and DeVore, 2001). We found that job autonomy could, under certain conditions (for example when it interacts with the extreme personality traits of Type-A), indeed, facilitate acts, which intentionally defy organizational expectations for acceptable conduct. When we broke autonomy down to its components the same effect existed. Thus, actually all four facets of autonomy (planning, scheduling, pacing and appraising) appear to allow for some level of misconduct. However, one should bear in mind that in this study we were able to tap only one type of misconduct commonly referred to as ‘production’-related (counterproductive acts that interfere with proper and expected work performance). We must therefore limit inferences about the generality of our findings. Similar studies in the future should therefore also include measures of interpersonal and property-related OMB. We suggest that the positive effect that job autonomy has on the subjects’ reports about engagement in OMB could be attributed to the opportunity to misbehave and to the need for autonomy. The opportunity to misbehave is related to those characteristics of the job or the organization that actually facilitate some forms of organizational misbehaviour (see Bryant’s 1977 discussion on occupational deviance). This ‘built-in opportunity’ exits when an employee perceives an availability of varied avenues for action, multiple degrees of freedom and high personal discretion coupled with relatively easy accessibility to resources. According to Vardi and Wiener (1996) such a perception positively influences the intention to misbehave. Moreover, if autonomy is perceived as the absence of supervision, or alternatively as managerial weakness, it may actually enhance this intention. This is important because such considerations must be taken into account when jobs and control systems are redesigned, providing more autonomy to job incumbents. Nonetheless, this proposition too must be tested in further studies that control for different levels of opportunities to misbehave. Researchers who have dealt with the negative effects of job autonomy attribute their results to differences in the need for autonomy (Harrell and Alpert, 1979; Langer, 1983). For people with a low need for autonomy,

Yoav Vardi and Ely Weitz 187

a job that is characterized by a great deal of autonomy will create a sense of tension and overload, and even decline in their satisfaction with the job. Moreover, the autonomy may decrease their sense of control and arouse a feeling of confusion. In other words, autonomy is a positive job characteristic if and when the individual is interested in it and is capable of handling it. Indeed, when exercised by professionals and competent employees who strive for such freedom at work, it may become a ‘responsible autonomy’ (Ackroyd and Thompson, 1999). Hackman and Oldham (1980), in their theory of job design, do consider personal growth needs as having a moderating effect. Employees with low growth needs, when granted autonomy, may lower their motivation, and experience lower satisfaction. For such individuals, the added level of discretion actually means an unwanted overload, burden and anxiety. Hackman and Oldham, however, do not raise the possibility that individuals with low growth needs may react to autonomy not only by a decline in such outcomes but also by venting their frustration and misbehaving. Such frustrations may actually become antecedents of misconduct. Spector (1997b), for example, has reported a meta-analysis of studies showing that autonomy is one of the major job-related (or positional) frustrations that can lead to antisocial work behaviour. Overall, the research findings imply that the personal variables examined have a stronger effect than the organizational variables on the link between autonomy and OMB. This proposition, however, should be confirmed in studies that better control for organization level variables such as goals and reward structures, which would be less directly related to autonomy at the position level. As compared with non-Type-As, the Type-A individuals in our sample seemed to report on more production and work process misbehaviour (of others but not of themselves). To date, studies that examined these personality characteristics found contradictory results with regard to the attitudes and behaviours that Type-As exhibit at work. Generally speaking, Type-A persons have higher levels of devotion to the organization, are responsible and committed to their work, persist and insist on doing their work even in conditions of tiredness and when sensing failure (Caplan et al., 1975; Chesney and Rosenman, 1980; Lee et al., 1990; Matthews, 1982). However, they also exhibit a tendency for rigid behaviour and aggression in the workplace. It is not just the sense of time pressure and the multiplicity of simultaneous goals that is liable to lead them to OMB, but also their natural tendency for impulsive, intolerant and impatient behaviour that may be targeted at fellow employees. Also, their need for control may actually lead them to employ aggressive action strategies in order to achieve it

188 Personal and Positional Antecedents of Misbehaviour

(Glass, 1977). Certainly, further studies should look at the effects of other personality traits such as the ‘big five’ (Digman, 1990; Krigel, 2000; Mount and Barrick, 1995) on the autonomy–OMB relationship. For instance, what would be the role of a high level of conscientiousness (Hogan and Ones, 1997) in mitigating the effects of job autonomy on the intention to mishandle human or physical organizational resources? As hypothesized, a negative correlation was found between professional identity and self-reported OMB: the higher the professional identity, the lower the reporting on OMB. This finding corroborates the results of previous studies (Hutnick, 1989; Sorensen and Sorensen, 1974) which show that people with high professional identity tend to be involved, to accept the profession’s values and aims, and to act for the good of the profession. Strong professional identity appears to raise the importance that the individual assigns to his/her job, to work goals, and to such normative work-related values as duty and loyalty (Wiener, 1982). Many studies assume that workers’ behaviours are the result of satisfaction or lack of it (Kreitner and Kinicki, 1995). Our study supports this linkage by including misconduct as an important aspect of work behaviour. The hypothesis concerning a negative correlation between satisfaction at work and OMB was supported: the higher the satisfaction, the lower the reporting on OMB. This negative relationship found between satisfaction and misbehaviour is compatible with the results of other studies that reported positive correlations between dissatisfaction and deviant behaviour such as theft, sabotage of property, or damaging production processes (Hollinger and Clark, 1982; Mangioni and Quinn, 1975). It thus appears reasonable to suggest that maintaining employee satisfaction is not only normatively correct, but also economically justified (Greenberg, 1990) because it may help to lower the staggering costs that organizations accrue due to misconduct. While it was important to us to have a research population as heterogeneous as possible, our design may also present a problem because it included subjects who were in the process of seeking work and in transition between jobs. It is possible that the findings are strongly influenced by both social desirability and attribution biases typical of any research on personally sensitive attitudes and behaviours. However, because of the constraints set by the testing centre, we could not include such control measures in the questionnaire. Thus the findings should be evaluated with due caution and replicated in future research that utilizes such controls. Replicating this research on other groups of employees in different organizational settings, therefore, is warranted. Most importantly,

Yoav Vardi and Ely Weitz 189

Possible antecedents

Intentions

Misbehaviour expressions (self, other)

Individual level Personality Attitudes, values Attribution Affect Stress

Intrapersonal misbehaviour Positional level

Autonomy Discretion Build-in opportunity to misbehave Level of responsibility Job pressures Employment status Group level

Interpersonal misbehaviour

Production misbehaviour Intention to misbehave Property misbehaviour

Norms Cohesiveness Leadership Organization/ Profession level

Political misbehaviour

Goals Climate Culture Ethics Socialization Control systems

Figure 10.1

Antecedents and expressions of OMB intentions: an emergent model

further research should investigate the impact of job autonomy on both proper (OCB) and improper (OMB) conduct simultaneously in order to test for the differential impact autonomy has on behaviour. Furthermore, our findings are based on self-report measures and should be replicated using measures of actual autonomy and actual behaviour.

Conclusion Our review of the literature and this exploratory investigation led us to derive a general framework for the future research of intentional organizational misbehaviour (see Figure 10.1). The framework offers a distinction between antecedents and manifestations of OMB. Antecedent factors are conceived at different levels of definition and observation: individual, positional, group and organization. Manifestations of OMB

190 Personal and Positional Antecedents of Misbehaviour

(both of self and of others) are classified into five types: Intrapersonal misbehaviour pertains to damages inflicted on the self such as drug abuse. Interpersonal misbehaviour pertains to such acts as harassment. Production misbehaviour (studied here) pertains to acts that may harm work processes such as slowdowns or faking information. Property misbehaviour pertains to acts that amount to improper use or handling of resources such as fraud or theft. Political misbehaviour pertains to behaviour that employs improper influence strategies such as deceit and impression management. As suggested above, the present study obviously dealt with just a few selected antecedents and manifestations providing direction for varied avenues of future research. From a practitioner perspective, we believe management of work organizations should realize that certain job characteristics do in fact increase the tendency for misbehaviour. It is important to remember that positional features such as variety and autonomy, while allowing independent decision-making, personal development and prospects, may also unwittingly create a structure of opportunity for engaging in organizational misbehaviour. Such temptations appear to be increasing as more work is carried out independently, in virtual organizations, and using global telecommunication and computer networks. Organizations must consider not only the effects of work environment designs on misbehaviour but also other enhancing factors such as employee characteristics (selection and placement) and control systems (performance appraisal, reward and punishment). No less important is the understanding that culture and atmosphere surrounding different occupational groups and hierarchical levels do play important enhancing or mitigating roles influencing individual misbehaviour. While, admittedly, practical solutions are hard to come by, increased management awareness of these effects is an important first step.

References Abdel Halim, A. A. (1978) ‘Employee affective responses to organizational stress: moderating effects of job characteristics’, Personnel Psychology, 31, 561–78. Ackroyd, S. and Thompson, P. (1999) Organizational Misbehavior. London: Sage. Allen, V. L. and Greenberg, D. B. (1980) ‘Destruction and perceived control’, in A. Baum and J. Singer (eds), Advances in Environmental Psychology (pp. 85–109). Hillside, NJ: Erlbaum. Analoui, F. and Kakabadse, A. (1992) ‘Unconventional practices at work: insight and analysis through participant observation’, Journal of Managerial Psychology, 7, 3–31. Bamberger, P. A. and Sonnenstuhl, W. J. (1998) ‘Introduction: research on organizations and deviance – some basic concerns’, in S. Bacharach (ed.), Research in the Sociology of Organizations (pp. vii–xviii). Greenwich, CT: JAI Press.

Yoav Vardi and Ely Weitz 191 Bennett, R. J. and Robinson, S. L. (2000) ‘Development of a measure of workplace deviance’, Journal of Applied Psychology, 85, 349–60. Biran, A. (1999) ‘Freedom of acting as chance or temptation: the relationship between autonomy, personal and organizational variables and misbehavior at work’, unpublished master’s thesis, Tel Aviv University. Breaugh, J. A. (1985) ‘The measurement of work autonomy’, Human Relations, 38, 551–570. Brislin, R. W. (1970) ‘Back-translation for cross-cultural research’, Journal of CrossCultural Research, 1, 185–216. Bryant, C. D. (1977) Deviant Behavior: Occupational and Organizational Bases. Chicago: Rand McNally. Caplan, R. D., Cobb, S. and French, J. R. P. (1975) ‘Relationships of cessation of smoking with job stress, personality, and social support’, Journal of Applied Psychology, 60, 210–19. Chesney, M. A. and Rosenman, R. H. (1980) ‘Type A behavior in the work setting’, in C. L. Cooper and R. Payne (eds), Current Concerns in Occupational Stress (pp. 187–212). New York: Wiley. Cohen, G. (1999) ‘Relationship between different types of commitment and organizational misbehavior’, unpublished master’s thesis, University of Tel-Aviv. Digman, J. M. (1990) ‘Personality structure: emergence of the five-factor model’, Annual Review of Psychology, 41, 417–40. Dwyer, D. J. and Fox, M. L. (2000) ‘The moderating role of hostility in the relationship between enriched jobs and health’, Academy of Management Journal, 43, 1086–96. Farrell, D. (1983) ‘Exit, voice, loyalty, and neglect as responses to job dissatisfaction: a multidimensional scaling study’, Academy of Management Journal, 26, 596–607. Ganster, D. C. (1989) Stress, Personal Control and Health. London: Wiley. Giacalone, R. A. and Greenberg, J. (1997) Antisocial Behavior in Organizations. Thousand Oaks, CA: Sage. Glass, D. C. (1977) Behavior Pattern: Stress and Coronary Disease. Hillsdale, NJ: Lawrence Erlbaum. Goldshmid-Aron, L. (1997) ‘Individual and organizational antecedents of workaholism’, unpublished master’s thesis, University of Tel-Aviv. Greenberg, J. (1990) ‘Employee theft as a reaction to underpayment inequity: the hidden cost of pay cuts’, Journal of Applied Psychology, 75, 561–8. Griffin, R. W., O’Leary-Kelly, A. M. and Collins, J. (1998) ‘Dysfunctional work behaviors in organizations’, in C. L. Cooper and D. M. Rousseau (eds), Trends in Organizational Behavior (pp. 65–82). New York: Wiley. Gruys, M. L. (1999) ‘The dimensionality of deviant employee performance in the workplace’, unpublished doctoral dissertation, University of Minnesota, Minneapolis, MN. Hackman, J. R. and Oldham, G. R. (1976) ‘Motivation through the design of work: test of a theory’, Organization Behavior and Human Performance, 16, 250–79. Hackman, J. R. and Oldham, G. R. (1980) Work Redesign. Reading, MA: AddisonWesley. Harrell, T. and Alpert, B. (1979) ‘The need for autonomy among managers’, Academy of Management Review, 4, 259–67. Herzberg, F. (1968) ‘One more time: how do you motivate employees?’, Harvard Business Review, January–February, 53–62.

192 Personal and Positional Antecedents of Misbehaviour Hirschman, A. O. (1970) Exit, Voice, and Loyalty: Response to Decline in Firms, Organizations and States. Cambridge, MA: Harvard University Press. Hogan, J. and Ones, D. S. (1997) ‘Conscientiousness and integrity at work’, in R. Hogan, S. Johnson and S. Briggs (eds), Handbook of Personality Psychology (pp. 848–70). San Diego: Academic Press. Hollinger, R. C. and Clark, J. (1982) ‘Employee deviance: a response to the perceived quality of the work experience’, Work and Occupations, 9, 97–114. Hutnik, H. (1989) ‘Factors influencing ethical behavior among insurance agents in Israel’, unpublished master’s thesis, University of Tel-Aviv. Jamal, M. (1985) ‘Type A behavior and job performance: some suggestive findings’, Journal of Human Stress, 11, 60–8. Karasek, R. A. and Theorell, T. (1990) Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York: Basic Books. Klein, R. L., Leong, G. B. and Silva, J. A. (1996) ‘Employee sabotage in the workplace: a biopsychosocial model’, Journal of Forensic Sciences, 41, 52–5. Kram, K. E. (1985) Mentoring at Work: Developmental Relationships in Organizational Life. Glenview, IL: Scott, Foresman & Co. Kreitner, R. and Kinicki, A. (1995) Organizational Behavior. New York: Irwin. Krigel, K. (2000) ‘Relationships between personality traits (big 5), organizational justice, and organizational misbehavior: a conceptual model’, unpublished master’s thesis, University of Tel-Aviv. Kunin, T. (1955) ‘The construction of a new type of attitude measure’, Personnel Psychology, 8, 65–77. Langer, E. J. (1983) The Psychology of Control. Beverly Hills, CA: Sage. Lee, C. and Gillon, D. J. (1989) ‘Relationship of Type A behavior pattern, self efficacy perceptions on sales performance’, Journal of Organization Behavior, 10, 75–81. Lee, C., Ashford, S. J. and Bobko, P. (1990) ‘Interactive effects of “Type A” behavior and perceived control on worker performance, job satisfaction, and somatic complaints’, Academy of Management Journal, 33, 870–81. Mangioni, T. W. and Quinn, R. P. (1975) ‘Job satisfaction, counterproductive behavior, and drug use at work’, Journal of Applied Psychology, 11, 114–16. Mars, G. (1982) Cheats at Work: An Anthropology of Workplace Crime. London: Allen & Unwin. Matthews, K. A. (1982) ‘Psychological perspectives on the Type A behavior pattern’, Psychological Bulletin, 91, 293–323. Meyer, J. P., Allen, N. J. and Smith, C. A. (1993) ‘Commitment to organization and occupation: extension and test of a three component conceptualization’, Journal of Applied Psychology, 78, 538–51. Molstad, C. (1988) ‘Control strategies used by industrial brewery workers: Work avoidance, impression management and solidarity’, Human Organization, 47, 354–59. Mount, M. K. and Barrick, M. R. (1995) ‘The big five personality dimensions: implications for research and practice in human resources management’, Research in Personality and Human Resource Management, 13, 153–200. Murphy, K. R. (1993) Honesty in the Workplace. Grove, CA: Brooks/Cole Publishing Co. Pines, A. (1981) Job Stress and Burnout: Research, Theory and Intervention Perspectives. Beverly Hills, CA: Sage.

Yoav Vardi and Ely Weitz 193 Raelin, J. A. (1986) ‘An analysis of professional deviance within organizations’, Human Relations, 39, 1103–30. Rentsch, J. R. and Steel, R. P. (1998) ‘Testing the durability of job characteristics as predictors of absenteeism over a six-year period’, Personnel Psychology, 51, 165–89. Robinson, S. L. and Bennett, R. J. (1995) ‘A typology of deviant workplace behavior: a multidimensional scaling study’, Academy of Management Journal, 38, 555–72. Robinson, S. L. and Greenberg, J. (1998) ‘Employees behaving badly: dimension, determinants and dilemmas in the study of workplace deviance’, in C. L. Cooper and D. M. Rousseau (eds), Trends in Organizational Behavior (pp. 1–30). New York: Wiley. Sackett, P. R. and DeVore, C. J. (2001) ‘Counterproductive behaviors at work’, in N. Anderson, D. S. Ones, H. K. Sinangil and C. Viswesvaran (eds), Handbook of Industrial, Work and Organizational Psychology (pp. 145–64). London: Sage. Sorensen, J. E. and Sorensen, T. L. (1974) ‘The conflict of professionals in bureaucratic organizations’, in S. L. Sauter, J. J. Hurell and C. L. Cooper (eds), Job Control and Worker Health (pp. 3–25). Chichester: Wiley. Spector, P. E. (1986) ‘Perceived control by employees: a meta analysis of studies concerning autonomy and participation at work’, Human Relations, 39, 1005–16. Spector, P. E. (1997a) Job Satisfaction: Application, Assessment, Causes, and Consequences. Thousand Oaks, CA: Sage. Spector, P. E. (1997b) ‘The role of frustration in antisocial behavior at work’, in R. A. Giacalone and J. Greenberg (eds), Antisocial Behavior in Organizations (pp. 1–17). Thousand Oaks, CA: Sage. Thompson, S. C. (1981) ‘Will it hurt less if I can control it? A complex answer to a simple question’, Psychological Bulletin, 90, 89–110. Trevino, L. K. (1986) ‘Ethical decision making in organizations: a person– situation interactionist model’, Academy of Management Review, 11, 601–17. Vardi, Y. and Wiener, Y. (1996) ‘Misbehavior in organizations: a motivational framework’, Organization Science, 7, 151–65. Vaughan, D. (1999) ‘The dark side of organizations: mistake, misconduct, and disaster’, Annual Review of Sociology, 25, 271–305. Vaught, C. and Smith, D. (1980) ‘Incorporation and mechanical solidarity in an underground coal mine’, Work and Occupations, 7, 159–87. Westman, M. (1992) ‘Moderating effect of decision latitude on stress-strain relationship: does organizational level matter?’, Journal of Organizational Behavior, 13, 713–22. Wiener, Y. (1982) ‘Commitment in organizations: a normative view’, Academy of Management Review, 7, 418–28. Wortman, C. B. and Breham, J. W. (1975) ‘Responses to uncontrollable outcomes: an integration of reactance theory and the learned helplessness model’, in L. Berkowitz (ed.), Advances in Experimental Social Psychology (pp. 278–336). New York: Academic Press. Yosifon, H. (2001) ‘The relationship between job satisfaction and organizational misbehavior among employees’, unpublished master’s thesis, University of Tel-Aviv.

11 Stress, Personality and Counterproductive Work Behaviour Lisa M. Penney, Paul E. Spector and Suzy Fox

Introduction Counterproductive work behaviour (CWB) by employees is an all too common occurrence in organizations. Studies have shown that 95 per cent of employees have engaged in some form of CWB at least once (e.g. Penney, 2002). The costs to American businesses associated with just one type of CWB, employee theft, have been estimated to be more than $200 billion annually (Govoni, 1992). Given the prevalence and economic impact of CWB, the attention given to CWB by organizational researchers is not surprising. Two major threads in organizational research on CWB have developed over the years. One identifies environmental conditions that may serve as antecedents to CWB, such as the presence of job stressors, while the other focuses on the role that individual personality plays in the likelihood that an individual will engage in CWB. The purpose of this chapter is to discuss recent organizational research linking both of these streams, job stressors and personality, to CWB. First, the CWB construct and research is discussed briefly. Next, a conceptual model illustrating the relationships among job stress, personality, and CWB is presented as an organizing framework. Finally, specific job stressors and personality variables and their relationships with CWB are discussed.

Counterproductive work behaviour (CWB) CWB is behaviour by employees that intends to harm an organization or its members (Fox and Spector, 1999) and it encompasses a variety of acts ranging from low intensity behaviour such as taking excessive breaks, spreading false rumours and withholding effort, to more serious 194

Lisa M. Penney, Paul E. Spector and Suzy Fox 195

behaviours such as sabotage, verbal abuse, and physical assault. Researchers have studied similar constructs describing an overlapping set of behaviours, although the constructs are called by different names, including organizational retaliatory behaviours (Skarlicki, Folger and Tesluk, 1999), organization-motivated aggression (O’Leary-Kelly, Griffin and Glew, 1996), workplace aggression (Neuman and Baron, 1998), workplace deviance (Bennett and Robinson, 2000), revenge (Bies and Tripp, 1998) and antisocial behavior (Giacalone and Greenberg, 1997). The different terms used to describe these behaviours are often a reflection of the theoretical perspective of the researchers. For example, Skarlicki et al. (1999) approached this class of behaviour from an organizational justice perspective wherein the term ‘retaliation’ is generally used in order to convey the belief that these behaviours are used as a means to balance the justice/equity equation. Likewise, Neuman and Baron’s use of the term ‘workplace aggression’ reflects the fact that the basis of their work lies in the study of human aggression in general. Despite the differences in terminology in these lines of research, however, the common theme that runs throughout is that these behaviours are intentional and inflict harm on an organization, its employees, or other stakeholders (e.g. customers). Additionally, researchers have described the underlying behavioural dimensions that make up CWB in a number of ways. For example, Baron and Neuman (in press) factor analyzed individual ratings of how often they were the target of 40 different forms of CWB. Their analysis indicated three factors: (a) expressions of hostility – e.g. ‘belittling someone’s opinion to others’, ‘spreading rumours’, ‘acting in a condescending manner’; (b) obstructionism – e.g. ‘failure to return phone calls or respond to memos’, ‘interfering with target’s work’, ‘intentional work slowdowns’; and (c) overt aggression – e.g. ‘physical attacks’, ‘threats of physical violence’, ‘sabotage, destruction, and theft of property’ (p. 396). Fox, Spector and Miles (1999) conducted a factor analysis of participant reports of how frequently they engaged in CWB that yielded five dimensions: (a) overt acts – e.g. theft, physical assault; (b) threats – violent or otherwise; (c) abuse – e.g. ignoring and insulting others; (d) work sabotage – e.g. performing work incorrectly, purposely working slowly; and (e) work avoidance – e.g. taking longer breaks than allowed, leaving early. However, the interpersonal/organizational distinction made by Robinson and Bennett (1995) has received the most attention. Specifically, this two-factor model categorizes CWB according to whether the behaviour targets an individual (e.g. arguing with others, stealing

196 Stress, Personality and Counterproductive Work Behaviour

someone’s personal property) or an organization (e.g. stealing company property, wasting time or materials). Support for the two-factor structure of CWB has been demonstrated by other studies (Aquino, Lewis and Bradfield, 1999; Bennett and Robinson, 2000). Moreover, additional research (Fox and Spector, 1999; Fox, Spector and Miles, 2001; Penney, 2002) suggests that interpersonal and organizational CWB may relate differently to other variables of interest, such as job satisfaction and organizational constraints. In sum, the interpersonal/organizational distinction appears to be an important one. As a whole, the research on CWB and related constructs has focused on identifying the antecedent conditions and individual difference variables associated with CWB. In other words, they have examined the personality traits and environmental events or conditions that may increase the probability that individuals will choose to engage in CWB. Similarly, the job stress literature has focused predominantly on the impact of environmental conditions on individual health and emotional well-being, and to a lesser extent, employee behaviour. Given the overlap between the two areas, it is not surprising that they would eventually converge.

A model of job stress and CWB A conceptual framework is helpful for understanding how environmental conditions and personality can contribute to CWB. Research on the job stress process by Spector (1998) described the relationship between job stressors, or conditions that elicit a negative emotional response, and job strains, the reactions to the stressors. According to Jex and Beehr (1991), job strains can be categorized as either psychological, physical, or behavioural. The majority of the job stress literature has focused primarily on assessing the effects of stress on psychological and physical strains, such as anxiety, anger, job satisfaction, high blood pressure, and illness. For the purposes of the current discussion, however, the focus will be on behavioural strains. Behavioural strains are the behaviours individuals perform in order to cope with stressors either by focusing on reducing the emotions elicited by the stressor (e.g. drinking alcohol, avoiding work) or by focusing on the eliminating stressor itself (e.g. talking with a supervisor, quitting). Behavioural strains can either benefit or harm an organization. For instance, an organization would benefit from an employee who responds to a tremendous workload by developing ways to improve the efficiency of the work process. However, an employee who responds to the same situation by reducing work quality to meet time constraints or by staying home to avoid work, harms the organization. The harmful

Lisa M. Penney, Paul E. Spector and Suzy Fox 197

Environmental stressor

Perceived stressor

Experienced negative emotions

CWB

Personality Figure 11.1

A model of stress, personality and CWB

behaviours performed in response to stress can be considered CWB. Therefore, in keeping with the focus on CWB as a behavioural response to stress, a model of job stress and CWB is presented as an organizing framework for the current discussion (see Figure 11.1). The model begins with an environmental stressor. This can be anything that elicits a negative emotional reaction, although several specific classes of stressors have been identified in research, including role conflict, role ambiguity, workload, perceived control/task autonomy, injustice, interpersonal conflict, and situational constraints. In order for an environmental stressor to have an impact, it must first be perceived as a stressor. Once an individual perceives a stressor, and appraises the situation as a challenge to his or her well-being, the next step may be the experience of negative affect, such as frustration, anger, or anxiety. The experience of negative emotions and the physiological arousal that accompany it may then serve as a catalyst for the final step in the process, a behavioural response, such as CWB. Personality may influence all parts of the model. For example, certain personality traits may increase the propensity to perceive stressors in the environment. Likewise, other personality traits are associated with an increased experience of negative affect. Finally, some personality traits are associated with a greater tendency to perform CWB. The following sections discuss specific stressors and personality as they related to CWB.

Job stress Role stressors At work and in everyday life people have roles to play. Roles are important because they dictate behaviour, or how to respond situations. They are often a function of individual expectations for behaviour and,

198 Stress, Personality and Counterproductive Work Behaviour

especially in organizations, the expectations of others, such as supervisors. When roles become unclear, individuals experience confusion and an uncertainty as to what to do or how to behave. Researchers refer to this experience as role ambiguity. Role ambiguity can lead to frustration and anxiety, especially if the consequence of performing the wrong behaviour are perceived to be high. (For example, a new department manager in a retail store could experience role ambiguity if she is uncertain about her responsibilities regarding inventory maintenance. If she orders new stock when someone else handles ordering, she risks having too much inventory and no where to store it. However, if she does not order new stock, she risks having empty shelves and losing sales revenue.) In addition, individuals often have multiple roles in their jobs, such as manager, mentor, and friend. Unfortunately, different roles can have competing demands and fulfilling one role may mean neglecting another. When this occurs, individuals experience role conflict. Both role ambiguity and role conflict have been linked to a number of affective responses, such as anxiety, frustration, and job dissatisfaction ( Jackson and Schuler, 1985; Jex and Beehr, 1991). A meta-analysis by Jackson and Schuler (1985) investigating the relationships between role conflict and role ambiguity and a variety of outcomes found evidence linking role stressors to narrow forms of CWB. Specifically, both role stressors were positively related to absenteeism, low job involvement, and turnover intention. Moreover, both role conflict and role ambiguity have also been linked to CWB more broadly. An investigation into the relationships between role stressors and CWB was conducted by Chen and Spector (1992). In a survey of four hundred employed persons from a variety of occupations, they found that role ambiguity and role conflict were correlated with a wide range of CWB including sabotage, interpersonal aggression, hostility and complaints, theft, and intention to quit.

Workload Workload refers to the demands that a job places on an individual. Two kinds of workload have been identified in the job stress literature: quantitative and qualitative. Quantitative workload refers to the volume of work an individual has to do in any given amount of time, whereas qualitative workload refers to the relative difficulty of the work an individual has to perform. Thus, individuals with a high quantitative workload have too many things to do, while individuals with a high qualitative workload have tasks that are too difficult for them to perform given their level of ability. Individuals may experience one type of workload without experiencing the other or they may experience them

Lisa M. Penney, Paul E. Spector and Suzy Fox 199

both simultaneously. As is the case with most work stressors, individuals who experience high levels of workload also report high levels of anxiety, frustration, and job dissatisfaction (Chen and Spector, 1992). However, few studies have examined the effects of workload on CWB. The research that is available found that workload was positively related to CWB in general, as well as intentions to quit the job and hostility and complaining about the job (Chen and Spector, 1992; Miles et al., 2002).

Situational constraints Situational constraints refer to conditions at work that prevent individuals who are otherwise willing and capable of performing a task from successfully doing so. Often these situations are beyond the control of employees, such as not receiving needed materials on time or being given incorrect information. Peters and O’Connor (1980) identified eight types of situational constraints: (a) job-related information, (b) tools and equipment, (c) materials and supplies, (d) budgetary support, (e) required services and help from others, (f) task preparation, (g) time availability, and (h) work environment. Other sources of situational constraints include interruptions by other people and organizational rules and procedures. In general, research shows that situational constraints are associated with increased frustration and decreased job satisfaction (Chen and Spector, 1992; Fox and Spector, 1999; Fox et al., 1999). Additionally, a number of studies have found a positive relationship between situational constraints and CWB (Chen and Spector, 1992; Miles et al., 2002; Penney, 2002; Penney and Spector, 2002). In other words, individuals who experienced high levels of situational constraints reported engaging in CWB more frequently than individuals who experienced low levels of constraints. Moreover, because situational constraints are associated with organizational conditions, Fox et al. (1999) and Penney (2002) also examined whether constraints were more strongly related to organizational CWB than personal CWB. In each case, situational constraints were positively related to both personal (r’s = .25 & .26, p < .01) and organizational CWB (r’s = .32 & .33, p < .01). Although, the relationship with organizational CWB was stronger in both studies, the difference was not significant. Taken as a whole, these studies indicate that situational constraints are an important predictor of CWB.

Interpersonal conflict Interpersonal conflict refers to the experience of poor interpersonal treatment by co-workers, supervisors, or others at work. Examples of

200 Stress, Personality and Counterproductive Work Behaviour

interpersonal conflict include being ignored or given the silent treatment, being put down in front of others, being interrupted while speaking, and getting into arguments with others at work. Researchers have investigated similar constructs in the stress and occupational health literature such as employee abuse (Keashly, Hunter and Harvey, 1997; Keashly, Trott and MacLean, 1994), mobbing/bullying (Leymann, 1990; Zapf, Knorz and Kulla, 1996), and workplace incivility (Andersson and Pearson, 1999; Cortina et al., 2001). A number of studies have assessed the impact of interpersonal conflict on individuals. For example research indicates that being the target of interpersonal conflict is associated with higher job dissatisfaction, turnover, absenteeism, depression, and psychosomatic symptoms (e.g. headaches, insomnia). More recently, researchers have begun to examine how conflict relates to CWB. A study by Pearson, Andersson and Porath (2000) surveyed 775 individuals from a variety of organizations around the United States to examine the effects of workplace incivility on subsequent CWB by the original target. Workplace incivility is generally defined as low intensity interpersonal conflict that stops short of being overtly aggressive (e.g. verbal abuse). Of the targets in their sample, more than one-third reported reducing their commitment to the organization, including withholding organizational-citizenship behaviours in response to workplace incivility. Additionally, roughly 25 per cent of participants reported they intentionally withheld effort (e.g. reduced the quality of their work, reduced time spent at work) as a result of being a target of workplace incivility, and 12 per cent of participants left the organization. A smaller percentage of respondents (approximately 5 per cent) admitted to stealing property from the instigator of the incivility and 5 per cent admitted to stealing from the organization as a result of experiencing workplace incivility. Additionally, a recent study by Penney (2002) found strong positive correlations between interpersonal conflict and both organization and person CWB. Moreover, the same study also found that co-worker ratings of interpersonal conflict were positively correlated with both self- and peer-ratings of targets’ CWB. Taken as a whole, these studies suggest that the quality of interpersonal treatment received at work is an important determinant of employee CWB.

Organizational justice Although organizational justice has been tied to CWB in several studies, justice has not traditionally been considered a job stressor. However, given the definition of a stressor as an event that triggers an emotional response, one could reasonably argue that the experience of injustice

Lisa M. Penney, Paul E. Spector and Suzy Fox 201

would arouse negative emotions. In fact, a study by Fox et al. (2001) found that the perceptions of injustice were associated with the experience of negative emotions, including anger and anxiety, as well as CWB. Therefore, classifying organizational justice as a stressor appears to be reasonable. Organizational justice refers to the perceived fairness of interactions between individuals and organizations, and researchers have identified three forms: distributive, procedural, and interactional justice. Distributive justice refers to the perceived fairness of the outcomes received from an employer. According to Equity Theory (Adams, 1965), individuals compare their ratio of organizational outcomes (e.g. pay, promotions) to inputs (e.g. effort, abilities) to the ratios of similar others in order to determine fairness. In contrast, procedural justice refers to the perceived fairness of the processes and decisions that determine organizational outcomes, independent of the fairness of the actual outcomes received (Thibaut and Walker, 1975). Characteristics of fair procedures include consistency, lack of bias, and accuracy of the information used to make the decision (Leventhal, Karuza and Fry, 1980). The third category of organizational justice, termed interactional justice, addresses the quality of interpersonal treatment received during the enactment of organizational procedures (Bies and Moag, 1986). The criteria for fairness in interpersonal treatment identified by Bies (1986) were: (a) truthfulness (e.g. open and honest communication); (b) respect (e.g. courtesy, avoiding intentional rude or attacking behaviours); (c) propriety of questions (e.g. avoiding improper questions); and (d) justification (e.g. providing an explanation of the decision). A number of studies have linked the three forms of organizational justice to different forms of CWB including withdrawal behaviours, such as withholding of effort, and absences (Barling and Phillips, 1993; Fox et al., 1999), and other retaliatory behaviours, such as spreading rumours, sabotage, and theft (Fox etal., 1999; Greenberg, 1990; Greenberg, 1993; Skarlicki and Folger, 1997; Skarlicki, Folger and Tesluk, 1999). Research also indicates that, in addition to independently affecting CWB, the three forms of organizational justice can also combine to affect CWB (Greenberg, 1993, 1994). For example, Greenberg conducted a field study (1990) and several laboratory studies (1993, 1994) examining the effects of low distributive, procedural, and interactional justice on theft. In each case, he found that theft was highest when all three forms of justice were low, and theft was lowest when all three forms of justice were high. Greenberg’s findings also indicated that low interactional

202 Stress, Personality and Counterproductive Work Behaviour

justice and low procedural justice independently increased theft rates in the underpayment condition. Similarly, Skarlicki and Folger (1997) conducted hierarchical regression analyses of the relationships among perceived distributive, procedural, and interactional justice and peerreported CWB. They reported that CWB was highest under conditions of low distributive justice when both procedural and interactional justice were also low. Thus, organizational justice appears to be an important determinant of individual CWB.

Control Control refers to the extent to which employees are able to make decisions about their work including what tasks to perform and when, where, and how to perform them. Although control has also been studied objectively as a job characteristic (i.e. autonomy), perceived level of control is more important than the actual level of environmental control that is available. Some researchers have suggested that control is not itself a job stressor, but instead moderates the relationship between environmental and perceived stressors. In other words, control affects the perception of other stressors in the environment (Spector, 1998). Specifically, high levels of perceived control are thought to reduce the level of perceived stress in the environment and to be associated with more problem-focused coping and less CWB. Unfortunately, few studies have examined the relationship between control and other stressors and CWB. A study by Fox et al. (2001) found a significant correlation between autonomy and organization CWB (r = − .25), but not person CWB. Although they conducted analyses to test for the moderation of autonomy between several job stressors and CWB, only two of the eight moderator tests were significant and the relationships were opposite to expectations. High autonomy was related to greater levels of personal CWB when either situational constraints or interpersonal conflict was high. Put another way, when under stress, individuals with more autonomy engaged in more personal CWB. One possible explanation given by the authors is that individuals with more autonomy are often higher up in the organizational hierarchy, and therefore may have less fear of retribution for personal CWB than individuals in lower positions. This corresponds to an earlier study (Fox and Spector, 1999), in which the strongest predictor of CWB after trait anger was the perceived ability to harm the organization without being caught or punished (see also Chapter 10 in this volume).

Lisa M. Penney, Paul E. Spector and Suzy Fox 203

Personality The Big Five The Big Five model of personality is the most empirically studied model in personality research. As a result, a number of studies have examined the relationship between the Big Five personality dimensions and CWB, although the majority of this research has been related to the validation of personality-based integrity tests for selection purposes, that, according to Ones, Viswesvaran and Schmidt (1993), probably capture a general conscientiousness factor. Recently, Salgado (2002) conducted a meta-analysis to assess the relationship between the ‘big five’ factors of agreeableness, conscientiousness, and emotional stability and CWB. The CWB criterion was separated into four categories, namely absenteeism, accident rates, deviant behaviour – e.g. a composite based on ‘measures of theft, admissions, actual theft, disciplinary problems, substance abuse, property damage, organizational rule breaking, and other irresponsible behaviors’ (p. 118) – and turnover. The results from his meta-analysis indicated that both conscientiousness and emotional stability predicted turnover and deviant behaviour, while agreeableness only predicted turnover. None of the personality variables was related to absenteeism or accident rates. Although the lack of significant findings against the absenteeism and accident rate criteria in Salgado’s study suggest that the value of the ‘big five’ dimensions in predicting CWB is limited, these two criteria reflect very narrow forms of CWB that may occur as a result of other work or nonwork issues. Thus, the observed relations may have been attenuated due to contamination by other factors, such as personal illness or illness of a family member, hazardous working conditions, or inadequate workplace safety training. Moreover, the relationship between some personality dimensions and CWB may be more complex, as Skarlicki et al. (1999) demonstrated in their study on the effects of justice perceptions and agreeableness on CWB. Agreeableness was not directly related to peer-reported CWB (the correlation between the two variables was non-significant). However, agreeableness moderated the relationship between justice and CWB such that justice perceptions predicted CWB only for individuals low on agreeableness. Thus, some personality traits may affect how individuals respond to certain types of situations (see also Chapter 12 in this volume).

Affective traits The model of CWB presented earlier proposes that the experience of negative affect (e.g. anxiety, anger, arousal, distress) mediates the relationship between stressors and CWB. Indeed, several studies have

204 Stress, Personality and Counterproductive Work Behaviour

provided evidence supporting the mediational role of affect (Chen and Spector, 1992; Fox and Spector, 1999; Storms and Spector, 1987). Therefore, it seems reasonable that individuals who experience negative emotions more frequently and intensely should be more likely to engage in CWB. Two affective traits that have been examined in relation to CWB are trait anger and trait anxiety. According to Spielberger, Krasner and Solomon (1988), anger is an emotional state or condition that consists of subjective feelings of irritation, annoyance, fury and rage. Individuals high in trait anger, therefore, experience angry feelings more frequently than others. Trait anxiety, on the other hand, refers to the tendency to experience a broad range of situations as stressful or threatening (Spielberger and Sydeman, 1994). A number of studies have reported significant correlations between trait anger and a variety of CWB including sabotage, interpersonal aggression, absenteeism, theft, and intention to quit (Chen and Spector, 1992; Douglas and Martinko, 2001; Fox and Spector, 1999; Fox etal., 1999; Miles etal., 2002; Penney, 1999). Additionally, Fox and Spector (1999) and Fox et al. (1999) also demonstrated significant relations between trait anxiety and CWB. Another personality trait that has received attention in the CWB literature is negative affectivity (NA). NA is dispositional tendency to experience a variety of negative mood states, such as anxiety, distress, guilt, anger, rejection, and sadness (Watson and Clark, 1984). In general, individuals high in NA are characterized as non-conformist, distrustful, hostile, demanding, and distant (Watson and Clark, 1984, p. 483). In contrast, persons low in NA are fairly calm, secure, conformist, and content. Several studies have found significant relationships between NA and self- and peer-reported CWB (Aquino et al., 1999; Douglas and Martinko, 2001; Penney, 2002; Skarlicki et al., 1999). That is, individuals high in NA are more likely to engage in CWB than individuals low in NA. Moreover, in addition to its direct relationship with CWB, NA has also been found to moderate the relationship between external stressors and CWB. In their study of perceived justice, Skarlicki et al. (1999) found that high-NA individuals were more likely to engage in CWB under conditions of low perceived justice than were low-NA individuals. Taken as a whole, these studies suggest that NA is an important predictor of CWB.

Locus of control As stated previously, high levels of perceived control are expected to be related to less CWB. One individual difference variable related to perceptions of control is locus of control. Locus of control refers to individual beliefs as to the amount of control they have over the

Lisa M. Penney, Paul E. Spector and Suzy Fox 205

reinforcements they receive. Internals believe that control over reinforcements lies largely within themselves. On the other hand, externals believe that control over what happens to them is in the hands of others and outside their control (Rotter, 1966). A study of direct-care workers by Perlow and Latham (1993) indicated that externals were more likely to engage in client abuse than internals. Moreover, Storms and Spector (1987) used a measure of work locus of control in their study of frustration and CWB. In comparison to the general locus of control measure used in the Perlow and Latham study, work locus of control refers to control beliefs that are specific to the work environment (Spector, 1988). They found that work locus of control moderated the relationship between frustration and CWB. Specifically, externals were more likely to engage in CWB under frustrating conditions than internals.

Narcissism Recent theory and research suggests that narcissism, a personality trait not often examined in organizational research, may be useful for predicting CWB. Based on the theory of threatened egotism and aggression (Baumeister, Smart and Boden, 1996), Penney and Spector (2002) proposed that individuals high in narcissism would be more likely to engage in CWB, especially in response to perceived threats to their ego. Individuals high in narcissism have an exaggerated positive self-image that is not based on a realistic assessment of their abilities. Hence, they are more likely than others to encounter information that refutes their positive, but inaccurate self-appraisal and threatens their self-esteem. Penney and Spector found a positive relationship between narcissism and CWB (r = .27). Furthermore, they reported that narcissism moderated the relationship between job constraints and CWB. Specifically, the relationship between job constraints and CWB was stronger for individuals high in narcissism than for individuals low in narcissism, suggesting that there are differences in the way individuals respond to situational cues. The positive relationship between narcissism and CWB was replicated by Penney (2002); however, the moderator effect with constraints and CWB was not. Although the research evidence linking narcissism and CWB is sparse, examining the effects of narcissism on CWB does appear to be a promising area for future research.

Conclusion The research summarized in this chapter illustrates the important roles that job stress and individual personality play in the prediction of

206 Stress, Personality and Counterproductive Work Behaviour

CWB. The model presented earlier proposes that the perception of job stressors in the environment can lead to a negative emotional experience, followed by CWB. Moreover, the model also suggests how personality is expected to affect the job stress–CWB process. Specifically, different personality traits have the potential to influence perceptions of the environment, emotional reactions, behaviour, and hence, even the environment itself. For example, trait anger makes an individual susceptible to perceiving environmental conditions that are annoying, frustrating, or unjust, and thus more inclined to perceive stressors that induce anger (e.g. organizational injustice, situational constraints, interpersonal conflict). Frequent expressions of anger, including CWB such as complaining, yelling at people at work, refusing to take on assignments, and purposely damaging property, may lead to interpersonal conflict and other stressors. In contrast, trait anxiety makes an individual susceptible to perceiving environmental conditions as threatening, dangerous, or otherwise stressful, and thus more inclined to perceive stressors that induce anxiety (e.g. role ambiguity, role conflict, workload). Means of coping with anxiety, including CWB such as staying home, taking longer breaks than allowed, or procrastinating on important projects, may then lead to more stressors, such as increased workload. Individuals with an external locus of control have a greater tendency to feel low levels of control over their environment, and hence perceive more stressors. Externals also have a tendency to respond to stressors counterproductively. Finally, conscientiousness is associated with higher levels of job performance that may result in better supervisor treatment, more autonomy, and less constraints. Individuals low in conscientiousness generally have lower levels of performance, leading to closer supervision, less autonomy, less consideration, and more structure. Clearly, the relationships between job stress, personality, and CWB may vary depending on the particular stressors and traits being studied and the research presented here is by no means exhaustive. Future research into the relationships among these variables should better inform our understanding of the conditions that lead to CWB.

References Adams, J. S. (1965) ‘Inequity in social exchange’, in L. Berkowitz (ed.), Advances in Experimental Social Psychology, Vol. 2 (pp. 267–99). New York: Academic Press. Andersson, L. M. and Pearson, C. M. (1999) ‘Tit for tat? The spiraling effect of incivility in the workplace’, Academy of Management Review, 24(3), 452–71.

Lisa M. Penney, Paul E. Spector and Suzy Fox 207 Aquino, K., Lewis, M. U. and Bradfield, M. (1999) ‘Justice constructs, negative affectivity, and employee deviance: a proposed model and empirical test’, Journal of Organizational Behavior, 20, 1073–91. Baron, R. A. and Neuman, J. H. (in press) ‘Workplace aggression – the iceberg beneath the tip of workplace violence: evidence on its forms, frequency, and targets’, Public Administration Quarterly. Barling, J. and Phillips, M. (1993) ‘Interactional, formal, and distributive justice in the workplace: an exploratory study’, Journal of Psychology, 127(6), 649–56. Baumeister, R. F. (1995) ‘Self and identity: an introduction’, in A. Tesser (ed.), Advanced Social Psychology. New York: McGraw-Hill. Baumeister, R. F., Bushman, B. J. and Campbell, W. K. (2000) ‘Self-esteem, narcissism, and aggression: does violence result from low self-esteem or from threatened egotism?’, Current Directions in Psychological Science, 9(1), 26–9. Baumeister, R. F., Smart, L. and Boden, J. M. (1996) ‘Relation of threatened egotism to violence and aggression: the dark side of high self esteem’, Psychological Review, 3(1), 5–33. Bennett, R. J. and Robinson, S. L. (2000) ‘Development of a measure of workplace deviance’, Journal of Applied Psychology, 85, 349–60. Bies, R. J. (1986) ‘Identifying principles of interactional justice: the case of corporate recruiting’, in R. J. Bies (Chair), Moving Beyond Equity Theory: New Directions in Research on Justice in Organizations. Symposium conducted at the meeting of the Academy of Management, Chicago, IL. Bies, R. J. and Moag, J. S. (1986) ‘Interactional justice: communication criteria of fairness’, Research on Negotiation in Organizations, 1, 43–55. Bies, R. J. and Tripp, T. M. (1998) ‘Revenge in organizations: the good, the bad, and the ugly’, in R. W. Griffin, A. O’Leary-Kelly and J. M. Collins (eds), Dysfunctional Behavior in Organizations: Violent and Deviant Behavior. Monographs in Organizational Behavior and Industrial Relations, Vol. 23, parts A & B (pp. 49–67). Stamford, CT: JAI Press, Inc. Bushman, B. J. and Baumeister, R. F. (1998) ‘Threatened egotism, narcissism, selfesteem, and direct and displaced aggression: does self-love or self-hate lead to violence?’, Journal of Personality and Social Psychology, 75(1), 219–29. Chen, P. Y. and Spector, P. E. (1992) ‘Relationships of work stressors with aggression, withdrawal, theft and substance use: an exploratory study’, Journal of Occupational and Organizational Psychology, 65, 177–84. Cortina, L. M., Magley, V. J., Williams, J. H. and Langhout, R. D. (2001). Incivility in the workplace: Incidence and impact. Journal of Occupational Health Psychology, 6(1), 64–80. Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H. and Sears, R. R. (1939) Frustration and Aggression. New Haven: Yale University Press. Douglas, S. C. and Martinko, M. J. (2001) ‘Exploring the role of individual differences in the prediction of workplace aggression’, Journal of Applied Psychology, 86(4), 547–59. Fox, S. and Spector, P. E. (1999) ‘A model of work frustration–aggression’, Journal of Organizational Behavior, 20, 915–31. Fox, S., Spector, P. E. and Miles, D. (1999) ‘Counterproductive behavior (CPB) in response to job stressors and organizational justice: the moderator effect of autonomy and emotion traits’, paper presented at the annual meeting of the Society for Industrial and Organizational Psychology, Atlanta, April–May 1999.

208 Stress, Personality and Counterproductive Work Behaviour Fox, S., Spector, P. E. and Miles, D. (2001) ‘Counterproductive work behavior (CWB) in response to job stressors and organizational justice: some mediator ad moderator tests for autonomy and emotions’, Journal of Vocational Behavior, 59, 291–309. Giacalone, R. A. and Greenberg, J. (1997) Antisocial Behavior in Organizations. Thousand Oaks, CA: Sage Publications, Inc. Giacalone, R. A. and Knouse, S. B. (1990) ‘Justifying wrongful employee behavior: the role of personality in organizational sabotage’, Journal of Business Ethics, 9, 55–61. Govoni, S. J. (1992) ‘To catch a thief’, CFO, February, 24–32. Greenberg, J. (1990) ‘Employee theft as a reaction to underpayment inequity: the hidden cost of pay cuts’, Journal of Applied Psychology, 75, 561–8. Greenberg, J. (1993) ‘Stealing in the name of justice: informational and interpersonal moderators of theft reactions to underpayment inequity’, Organizational Behavior and Human Decision Processes, 54, 81–103. Greenberg, J. (1994) ‘Restitution and retaliation as motives for inequity-induced pilferage’, unpublished manuscript, Ohio State University. Hogan, J. and Hogan, R. (1989) ‘How to measure employee reliability’, Journal of Applied Psychology, 74(2), 273–9. Jackson, S. E. and Schuler, R. S. (1985) ‘A meta-analysis and conceptual critique of research on role ambiguity and role conflict in work settings’, Organizational Behavior and Human Decision Processes, 36(1), 16–78. Jex, S. M. and Beehr, T. A. (1991) ‘Emerging theoretical and methodological issues in the study of work-related stress’, Research in Personnel and Human Resources Management, 9, 311–65. Keashly, L., Hunter, S. and Harvey, S. (1997) ‘Abusive interaction and role state stressors: relative impact on student residence assistant stress and work attitudes’, Work Stress, 11(2), 175–85. Keashly, L., Trott, V. and MacLean, L. M. (1994) ‘Abusive behavior in the workplace: a preliminary investigation’, Violence and Victims, 9(4), 341–57. Keenan, A. and Newton, T. J. (1984) ‘Frustration in organizations: relationships to role stress, climate, and psychological strain’, Journal of Occupational Psychology, 57, 57–65. Leventhal, G. S., Karuza, J. and Fry, W. R. (1980) ‘Beyond fairness: a theory of allocation preferences’, in G. Mikula (ed.), Justice and Social Interaction. New York: Plenum. Leymann, H. (1990) ‘Mobbing and psychological terror at workplaces’, Violence and Victims, 5, (2), 119–26. Miles, D. E., Borman, W. E., Spector, P. E. and Fox, S. (2002) ‘Building an integrative model of extra role work behaviors: a comparison of counterproductive work behavior and organizational citizenship behavior’, International Journal of Selection and Assessment, 10(1/2), 51–7. Neuman, J. H. and Baron, R. A. (1998) ‘Workplace violence and workplace aggression: evidence concerning specific forms, potential causes, and preferred targets’, Journal of Management, 24, 391–419. O’Leary-Kelly, A. M., Griffin, R. W. and Glew, D. J. (1996) ‘Organization-motivated aggression: a research framework’, Academy of Management Review, 21, 225–53. Ones, D. S., Viswesvaran, C. and Schmidt, F. L. (1993) ‘Comprehensive metaanalysis of integrity test validities: findings and implications for personnel

Lisa M. Penney, Paul E. Spector and Suzy Fox 209 selection and theories of job performance’ (monograph), Journal of Applied Psychology, 78, 679–703. Pearson, C. M., Andersson, L. M. and Porath, C. L. (2000) ‘Assessing and attacking workplace incivility’, Organizational Dynamics, 29(3), 123–37. Penney, L. M. (1999) ‘Antisocial workplace behavior and self-esteem: an examination of the relationship between egotism and aggressive and antisocial behavior’, unpublished master’s thesis, University of South Florida. Penney, L. M. (2002) ‘Workplace incivility and counterproductive workplace behavior (CWB): what is the relationship and does personality play a role?’, unpublished doctoral dissertation, University of South Florida. Penney, L. M. and Spector, P. E. (2002) ‘Narcissism and counterproductive work behavior: do bigger egos mean bigger problems?’, International Journal of Selection and Assessment, 10(1), 59–67. Perlow, R. and Latham, L. L. (1993) ‘Relationship of client abuse with locus of control and gender: a longitudinal study’, Journal of Applied Psychology, 78, 831–4. Peters, L. H. and O’Connor, E. J. (1980) ‘Situational constraints and work outcomes: the influences of a frequently overlooked construct’, Academy of Management Review, 5, 391–7. Peters, L. H. and O’Connor, E. J. (1988) ‘Measuring work obstacles: procedures, issues, and implications’, in D. Schoorman and B. Schneider (eds), Facilitating Work Effectiveness. Lexington, MA: Lexington. Robinson, S. L. and Bennett, R. J. (1995) ‘A typology of deviant workplace behaviors: a multidimensional scaling study’, Academy of Management Journal, 38, 555–72. Rotter, J. B. (1966) ‘Generalized expectancies for internal versus external control of reinforcement’, Psychological Monographs, 80, Whole no. 609. Salgado, J. F. (2002) ‘The Big Five personality dimensions and counterproductive behaviors’, International Journal of Selection and Assessment, 10(1/2), 117–25. Skarlicki, D. P. and Folger, R. (1997) ‘Retaliation in the workplace: the roles of distributive, procedural, and interactional justice’, Journal of Applied Psychology, 82, 434–43. Skarlicki, D. P., Folger, R. and Tesluk, P. (1999) ‘Personality as a moderator in the relationship between fairness and retaliation’, Academy of Management Journal, 42, 100–8. Smalley, R. L. and Stake, J. E. (1996) ‘Evaluating sources of ego-threatening feedback: Self-esteem and narcissism effects’, Journal of Research in Personality, 30, 483–95. Spector, P. E. (1975) ‘Relationships of organizational frustration with reported behavioral reactions of employees’, Journal of Applied Psychology, 60, 635–7. Spector, P. E. (1978) ‘Organizational frustration: a model and review of the literature’, Personnel Psychology, 31, 815–29. Spector, P. E. (1986) ‘Perceived control by employees: a meta-analysis of studies concerning autonomy and participation at work’, Human Relations, 39, 1005–16. Spector, P. E. (1988) ‘Development of the work locus of control scale’, Journal of Occupational Psychology, 61, 335–40. Spector, P. E. (1997) ‘The role of frustration in antisocial behavior at work’, in R. A. Giacaolone and J. Greenberg (eds), Antisocial Behavior in Organizations (pp. 1–17). Thousand Oaks, CA: Sage Publications, Inc.

210 Stress, Personality and Counterproductive Work Behaviour Spector, P. E. (1998) ‘A control theory of the job stress process’, in C. L. Cooper (ed.), Theories of Organizational Stress (pp. 153–69). London: Oxford University Press. Spector, P. E. and Fox, S. (2002) ‘An emotion-centered model of voluntary work behavior: some parallels between counterproductive work behavior (CWB) and organizational citizenship behavior (OCB)’, Human Resources Management Review, 12, 269–92. Spector, P. E. and Jex, S. M. (1998) ‘Development of four self-report measures of job stressors and strain: interpersonal conflict at work scale, organizational constraints scale, quantitative workload inventory, and physical symptoms inventory’, Journal of Occupational Health Psychology, 3, 356–67. Spector, P. E., Dwyer, D. J. and Jex, S. M. (1988) ‘Relations of job stressors to affective, health, and performance outcomes: a comparison of multiple data sources’, Journal of Applied Psychology, 71, 11–19. Spielberger, C. D. and Sydeman, S. J. (1994) ‘State-trait anxiety inventory and state-trait anger expression inventory’, in M. E. Maruish (ed.), The Use of Psychological Tests for Treatment Planning and Outcome Assessment. Hillsdale, NJ: Lawrence Erlbaum. Spielberger, C. D., Krasner, S. S. and Solomon, E. P. (1988) ‘The experience, expression, and control of anger’, in M. P. Janisse (ed.), Healthy Psychology: Individual Differences and Stress. New York: Springer Verlag Publishers. Storms, P. L. and Spector, P. E. (1987) ‘Relationships of organizational frustration with reported behavioral reactions: the moderating effect of perceived control’, Journal of Occupational Psychology, 60, 227–34. Thibaut, J. and Walker, L. (1975) Procedural Justice: A Psychological Analysis. Hillsdale, NJ: Lawrence Erlbaum. Watson, D. and Clark, L. A. (1984) ‘Negative affectivity: the disposition to experience aversive emotional states’, Psychological Bulletin, 96(3), 465–90. Zapf, D., Knorz, C. and Kulla, M. (1996) ‘Mobbing factors, the social work environment and health outcomes’, European Journal of Work and Organizational Psychology, 5, 215–38.

12 The Big-5 Personality and Counterproductive Behaviours Deniz S. Ones and Chockalingam Viswesvaran*

Introduction Employees can engage in a wide spectrum of misbehaviour in organizations. Such counterproductivity costs employers billions of dollars annually worldwide (Ones, 2002). The extent of actual, psychological and societal costs to organizations can be better understood when one considers the multitude of different ways employees can misbehave. Geddes and Baron (1997) found that 69 per cent of managers reported having experienced verbal aggression. Wimbush and Dalton (1997) used multiple methods and samples to estimate the base rate for employee theft, and found ‘Depending on the level one ascribes to nontrivial employee theft, .. . [the different techniques of estimating employee theft] . . . converge on theft rates over 50 per cent’ (p. 756). It is estimated that substance abuse costs the United States alone more than $135 billion each year (DeCresce et al., 1990). Harwood, Fountain and Livermore (1992) estimated that in the United States economy $82 billion in lost potential productivity could be attributed to alcohol and drug abuse in 1992 ($67.7 billion and $14.2 billion, respectively). Various forms of organizational rule-breaking, disobedience, misconduct, wrongdoing, and transgressions are subsumed under a general umbrella of organizational misbehaviour. Such behaviours can be termed (1) counterproductive, as they detract from the productive behaviours at work, (2) disruptive, as they disrupt work-related activities, (3) antisocial, as they violate social norms, and (4) deviant, as they diverge from organizationally desired behaviours. There is a very large degree of overlap

* Both authors contributed equally; the order of authorship is alphabetical. 211

212 The Big-5 Personality and Counterproductive Behaviours

among these terms and we use ‘counterproductive work behaviours’ (CWB) because over the past decade a large body of research has developed around this label. CWB include, but are not limited to, theft and related behaviour, destruction of property, misuse of information, misuse of time, white-collar crime, absenteeism, tardiness, drug and alcohol abuse, disciplinary problems, poor-quality work, sabotage, inappropriate physical and verbal actions, sexual harassment, violence, accidents and unsafe behaviours. Robinson and Bennett (1995) define deviant behaviour as ‘voluntary behavior that violates significant organizational norms and in so doing threatens the well being of an organization, its members, or both’ (p. 556). The central thesis in this chapter is that personality traits are powerful determinants of CWB. In the following sections, we first review theoretical and empirical work on the linkages among different forms of CWB. Second, we present a synopsis of quantitative reviews that have demonstrated personality–CWB relationships. Third, we discuss how and why personality traits come to influence CWB.

Relationships among different forms of counterproductive work behaviours A number of studies have sought to specify the domain of CWB and have explored the dimensionality of the construct (see Sackett, 2002, for an overview). However, there have been differences in approaches among these studies in terms of data utilized. Some have used similarity and likelihood of co-occurrence judgments in clustering CWB. Others have relied on self-reports (i.e. admissions) or ratings of CWB.

Similarity judgements in clustering CWB Multiple dimensions that constitute CWB may manifest themselves in critical incident analyses, task analyses, and job analyses. However, researchers have tended to rely on judgments about the similarities of specific CWB in reaching conclusions about the domain’s dimensionality. In a multidimensional scaling study, Robinson and Bennett (1995) found that deviant behaviour in organizations varies along two continua: (1) organizational/interpersonal, and (2) serious/minor. The resulting typology that crosses these two dimensions produced the following four categories: (1) Property deviance (serious deviance directed at the organization), (2) Production deviance (minor deviance directed at the organization), (3) Personal aggression (serious deviance directed

Deniz S. Ones and Chockalingam Viswesvaran 213

at other individuals), and (4) Political deviance (minor deviance directed at other individuals). In a similar study, Gruys (1999) analyzed judgments about the likelihood of co-occurrence of different forms of CWB. Study participants made judgments about the likelihood that a person who engages in one form of counterproductivity would also engage in another. Gruys (1999) identified two CWB continua: (1) organizationally targeted behaviours (e.g. theft, absences) vs. interpersonally targeted behaviours (e.g. verbal and physical violence toward others); and (2) negative behaviours at work not directly related to job performance (e.g. theft) vs. acts that negatively impact job performance (e.g. absences, making errors). In both Robinson and Bennett (1995) and Gruys (1999) working adults constituted the sample. Adults with work experience vary in their degree of understanding of causes, consequences and particularly psychological processes involved in various CWB. In essence, in their approach to judging CWB similarity and perceived co-occurence, working adults may function like novices. Their implicit theories may or may not be accurate. In the expertise literature, one of the most robust findings is that novices group items being rated according to surface features (Johnson, 1988). ‘Knowledge of patterns and their application is the key to expertise in many domains (Larkin et al., 1980)’ (Johnson, 1988, p. 225). We suspect that experts in criminology and psychology would have differing perceptions of similarity and perceived co-occurence of CWB, though empirical evidence to support this expectation needs to be gathered. Furthermore, as we have noted elsewhere, although there is much to be said about analyzing jobs and obtaining behavioural similarity judgments, dimensions obtained using such analyses often differ from those obtained using other empirical methods (Viswesvaran and Ones, 2000b). In support of this point, Gruys (1999) found that there was a small, negative correlation (–.17) between judgments of co-occurrence and corresponding actual covariation among CWB dimensions. Campbell (1990) and Viswesvaran (1993) suggest that the factor structure of importance or criticality ratings do not mirror dimensions of actual behaviour on the job. Factor structure of similarity judgments may not uncover underlying dimensionality CWB. A more important part of evaluating the factor structure of CWB is an assessment of covariance between different forms of counterproductivity. In fact, Gruys (1999) found a two-factor structure for co-occurrence ratings but a one-dimensional model for self-ratings of counterproductive behaviours.

214 The Big-5 Personality and Counterproductive Behaviours

Covariance among counterproductive work behaviours Relationships among single acts of counterproductivity How high is the relationship among single item measures of counterproductive behaviours? To examine this question, we estimated relationships among single item measures of counterproductive behaviours using the internal consistency reliabilities reported in the literature for (1) overall measures of CWB and (2) aspects of CWB. Using the number of items included in each CWB scale and the reported reliability information (coefficient alphas), we were able to compute mean relationships among single, individual item measures of counterproductivity. Table 12.1 presents the CWB measures examined (column 1), the source of the information (column 2), the number of items included in each scale (column 3), and the reported reliabilities (column 4) used in our computations as well as the mean correlations among single, individual behaviours of counterproductivity (column 5). Indices of workplace counterproductivity typically used in omnibus measures of overall CWB correlate, on average, .28 (SD = .17). This means that engaging in one counterproductive behaviour is correlated positively and non-trivially with engaging in another counterproductive behaviour. Considering that the scales used to assess counterproductivity include a broad spectrum of counterproductive behaviours such as absenteeism, substance abuse, aggression, theft etc., it is remarkable that such disparate behaviours appear to contribute to an overall syndrome of counterproductivity. Of course, there are CWB measures that focus on narrower and more specific clusters of behaviours. To the extent that the measures of CWB focused on the same type of counterproductive behaviours, the inter-item correlation estimated here may overestimate the inter-item correlation across all types of counterproductive behaviours. The bottom part of Table 12.1 displays the mean inter-item correlations for specific dimensions and facets of CWB. As can be observed in Table 12.1, dimensions and more specific facets of CWB tend to have markers that are somewhat more highly correlated. Individual acts that comprise organizationally targeted counterproductivity correlate, on average, .34 (SD = .22). Specific aggressive, hostile and antagonistic behaviours correlate at .39 (SD = .14). Individual behavioural indices of absenteeism/withdrawal and substance abuse both have a mean correlation of .34 (SD = .09, for each). Items that measure poor-quality work and having low personal standards at work correlate .48 (SD = .27). Somewhat lower relationships are encountered among acts of interpersonal deviance (mean r = .23; SD = .15) and theft

215 Table 12.1 Mean correlations among single, individual behaviours of counterproductivity Variable

Internal consistency reliability

Mean correlation among CWB

.71 .72 .82 .90 .96

0.29 0.20 0.17 0.17 0.56

Mean (SD)

0.28 (.17)

8

.77

0.30

Marcus et al. (2002)

32

.85

0.15

Bennett and Robinson (2000) Skarlicki and Folger (1997)

12

.81

0.26

17

.97

0.66

Interpersonal deviance Interpersonal deviance

Marcus et al. (2002)

16

Mean (SD) .70

0.34 (.22) 0.13

Bennett and Robinson (2000)

7

.78

0.34

Interpersonal aggression Aggression Workplace aggression and conflict Antagonistic work behaviours Inappropriate verbal actions Inappropriate physical actions Hostility and complaints

Chen and Spector (1992) Marcus et al. (2002) Jockin et al. (2001)

3

Mean (SD) .80

0.23 (.15) 0.57

9 4

.73 .59

0.23 0.26

Overall CWB measures CWB CWB CWB CWB CWB

Dimensions and facets of CWB Organizational deviance Organizational deviance Organizational deviance Organizational retaliatory behaviour

Study

Number of items

. Duffy et al. (1998) Kelloway et al. (2002) Lee and Allen (2002) Miles et al. (2002) Penney and Spector (2002)

Ashton (1998)

6 10 23 45 19

Lehman and Simpson (1992) Gruys (1999)

5

.60

0.23

8

.82

0.36

Gruys (1999)

7

.82

0.39

Chen and Spector (1992)

6

.85

0.69

216 Table 12.1

(Continued)

Variable

Study

Number of items

Internal consistency reliability

Mean correlation among CWB

Sabotage

Chen and Spector (1992)

9

.93

0.31

Poor attendance Work avoidance

Gruys (1999) Hakstian et al. (2002) Marcus et al. (2002)

5 3

Mean (SD) .77 .66

0.39 (.14) 0.40 0.39

13

.76

0.20

Lehman and Simpson (1992) Lehman and Simpson (1992) Gruys (1999)

4

.58

0.26

8

.84

0.40

13

.90

0.41

Hakstian et al. (2002) Marcus et al. (2002) Gruys (1999) Gruys (1999)

8

Mean (SD) .86

0.34 (.09) 0.43

6

.64

0.23

3 4

Hakstian et al. (2002) Gruys (1999)

3

.59 .71 Mean (SD) .60

0.32 0.38 0.34 (.09) 0.33

10

.77

0.25

Marcus et al. (2002)

6

.46

0.12

Gruys (1999)

4

.66

0.33

Low personal standards Poor-quality work

Hakstian et al. (2002) Gruys (1999)

8

Mean (SD) .76

0.26(.10) 0.28

Misuse of information Unsafe behaviours

Gruys (1999)

5

.86 Mean (SD) .71

0.67 0.48 (.27) 0.33

Gruys (1999)

4

.71

0.38

Mean (SD) Grand Mean (SD)

0.35 (.04) 0.34 (.14)

Absenteeism/ withdrawal Physical withdrawal Psychological withdrawal Misuse of time and resources Substance abuse Substance use Alcohol use Drug use Property theft Theft & related behaviour Theft/property violations Destruction of property

3

Deniz S. Ones and Chockalingam Viswesvaran 217

and related behaviour (mean r = .26; SD = .10). This suggests that measures of these latter CWB facets tend to include more heterogeneous sets of items. Nevertheless, what we find remarkable is that individual acts of counterproductivity correlate positively and as high as they do. The correlations within different domains of counterproductivity average to .34 (SD = .14). It is important to put the magnitude of this correlation in context with other individual differences variables in psychology. For example, in the personality domain, items that comprise scales such as conscientiousness, agreeableness, and extraversion correlate, on average, .12 to .18 (Viswesvaran and Ones, 2000a). Therefore, the behaviours that make up existing CWB measures are about twice as highly correlated as items assessing personality traits. Alternately, the behaviours that make up existing CWB measures may be narrower in their specification of the CWB domain. As instructive as relationships among specific individual acts of CWB are, they do not fully assess the construct-level relationships among dimensions and aspects of CWB. Single measures are inherently unreliable and construct-deficient. Using multiple items to assess constructs increases reliability and construct validity. Even with multiple item scales of CWB, unreliability is a concern. Unreliability in measures systematically depresses observed correlations. Correcting for this attenuation allows for an assessment of the true relationship (the true score correlation, rho) between variables of interest (Hunter and Schmidt, 1990). In our effort to provide estimates for the relationships between different forms of CWB, we corrected observed correlations from the literature for unreliability in both measures. In corrections for attenuation, coefficient alphas were used.

Relationships between CWB facets As already noted, CWB measures are diverse. Different studies focus on different aspects of CWB. Is there evidence that the seemingly different constructs assessed by the different facet measures of CWB are related to each other? The relationships between the absenteeism/withdrawal construct and other CWB are reviewed in Table 12.2. In the following discussion, consistent with contemporary conceptualizations of criterion variables in industrial, work and organizational (IWO) psychology, we use the terms CWB components, facets, factors and dimensions synonymously (Campbell, 1990). Different aspects of CWB are behavioural constructs best viewed as hierarchical representations of the counterproductivity domain (Sackett and DeVore, 2001).

Relations of absenteeism/withdrawal and related constructs with other CWB

Correlate

Absenteeism/withdrawal and related constructs

Study

Aggression Alcohol use Alcohol use Antagonistic work behaviours Antagonistic work behaviours Destruction of property Destruction of property Drug misuse1 Drug misuse1 Drug use Drug use Engaging in off-task behaviours1 Inappropriate physical actions Inappropriate physical actions Inappropriate verbal actions Inappropriate verbal actions Lateness1 Lateness1 Low personal standards Misuse of information Misuse of information Misuse of time & resources Poor attendance Poor-quality work Poor-quality work

Absenteeism/withdrawal Misuse of time & resources Poor attendance Physical withdrawal Psychological withdrawal Misuse of time & resources Poor attendance Engaging in off-task behaviours Tardiness/absenteeism Misuse of time & resources Poor attendance Tardiness/absenteeism Misuse of time & resources Poor attendance Misuse of time & resources Poor attendance Total absences Voluntary absences Work avoidance Misuse of time & resources Poor attendance Poor attendance Misuse of time & resources Misuse of time & resources Poor attendance

Marcus et al. (2002) Gruys (1999) Gruys (1999) Lehman & Simpson (1992) Lehman & Simpson (1992) Gruys (1999) Gruys (1999) Hunt (1996)1 Hunt (1996)1 Gruys (1999) Gruys (1999) Hunt (1996)1 Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Koslowsky et al. (1997)1 Koslowsky et al. (1997)1 Hakstian et al. (2002) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999)

218

Table 12.2

N

r

rho

174 363 363 1,070 1,070 363 363 1,818 1,827 363 363 5,931 363 363 363 363 8,013 8,013 1,019 363 363 363 363 363 363

0.49 0.40 0.29 0.38 0.26 0.43 0.38 – – 0.21 0.27 – 0.36 0.31 0.54 0.47 0.29 0.29 0.31 0.71 0.64 0.76 0.76 0.5 0.53

0.66 0.55 0.43 0.64 0.37 0.56 0.53 0.42 0.26 0.26 0.37 0.76 0.42 0.39 0.63 0.59 0.40 0.41 0.44 0.89 0.87 0.91 0.91 0.57 0.65

Psychological withdrawal Substance abuse Substance use Theft Theft Theft of property Theft/property violations Theft1 Theft1 Unruliness1 Unruliness1 Unsafe behaviour Unsafe behaviour

Physical withdrawal Work avoidance Absenteeism/withdrawal Misuse of time & resources Poor attendance Work avoidance Absenteeism/withdrawal Engaging in off-task behaviours Tardiness/absenteeism Engaging in off-task behaviours Tardiness/absenteeism Misuse of time & resources Poor attendance

Lehman & Simpson (1992) Hakstian et al. (2002) Marcus et al. (2002) Gruys (1999) Gruys (1999) Hakstian et al. (2002) Marcus et al. (2002) Hunt (1996)1 Hunt (1996)1 Hunt (1996)1 Hunt (1996)1 Gruys (1999) Gruys (1999) Unit weighted

1,070 1,019 174 363 363 1,019 174 2,280 2,289 2,839 4,102 363 363 mean

0.43 0.13 0.58 0.74 0.63 0.17 0.40 – – – – 0.48 0.37 0.44

0.62 0.17 0.83 0.89 0.82 0.27 0.68 0.70 0.52 0.71 0.66 0.60 0.50 0.58

Note: rho = true score correlation (i.e. observed correlation corrected for unreliability in both measures). 1 Data from multiple data sets or from metaanalyses.

219

220 The Big-5 Personality and Counterproductive Behaviours

In Table 12.2, the first column notes the specific CWB facet that absenteeism/withdrawal measures have been correlated with. The second column provides a characterization of absenteeism/withdrawal measures used in studies. The third column indicates the source of the information. The fourth and fifth columns report the sample size and observed correlation between the measures listed in columns 1 and 2. The last column of the table reports the correlation corrected for unreliability in both measures (i.e. rho = true score correlation). These are the relationships between the construct of absenteeism/withdrawal and other CWB constructs: the true score relationships of interest for this chapter. Reviewing the last column of Table 12.2 makes it clear that absenteeism/withdrawal from work is highly related to aggression, substance abuse, lateness, misuse of information, unruliness and unsafe behaviour. Absenteeism/withdrawal behaviours appear to be related even to theft (rhos range between .27 and .82). Across all the different constructs of CWB, the mean true correlation with absenteeism/withdrawal is .58. In other words, individuals who exhibit tardiness, absenteeism, and misuse of time at work are those who are also more likely to steal from their employers, use drug and alcohol, do poor-quality work, and engage in unsafe behaviours. Aggression, antagonistic and violent behaviours at work constitute another aspect of CWB. The relationships between these constructs and other aspects of CWB are reviewed in Table 12.3. The specific CWB facets correlate of aggression and measures of aggression, antagonistic and violent behaviours are listed in columns 1 and 2, respectively. The remaining columns are as in Table 12.2. The true score correlations of aggressive and violent work behaviours with other indices of counterproductivity are striking. Substance abuse, theft, unsafe behaviours, absenteeism, doing poor-quality work, among others, were all related positively and substantially to aggressive and violent work behaviours. The mean rho in Table 12.3 is .55. Employees who are antagonistic, aggressive and violent on the job tend to also engage in alcohol and drug use, absenteeism, theft and other distinct forms of counterproductivity. The correlations establishing a link between substance abuse and other types of CWB are presented in Table 12.4. Substance abuse includes both drug and alcohol abuse (Schmidt, Viswesvaran and Ones, 1997). The first and second columns list the CWB correlates and substance abuse measures, respectively. The remaining columns are as in Table 12.2. Substance abuse is positively correlated with other kinds of CWB. These

Table 12.3

Relations of antagonistic behaviours/aggression/violence and related behaviours with other CWB

Correlate

Aggression/violence and related constructs

Study

N

Absenteeism/withdrawal Alcohol use Alcohol use Destruction of property Destruction of property Drug use Drug use Hostility and complaints Misuse of information Misuse of information Misuse of time & resources Misuse of time & resources Tardiness/absenteeism1 Theft Theft Theft/property violations Unsafe behaviour Unsafe behaviour

Aggression Inappropriate physical actions Inappropriate verbal actions Inappropriate physical actions Inappropriate verbal actions Inappropriate physical actions Inappropriate verbal actions Interpersonal aggression Inappropriate physical actions Inappropriate verbal actions Inappropriate physical actions Inappropriate verbal actions Unruliness Inappropriate physical actions Inappropriate verbal actions Aggression Inappropriate physical actions Inappropriate verbal actions

Marcus et al. (2002) 174 Gruys (1999) 363 Gruys (1999) 363 Gruys (1999) 363 Gruys (1999) 363 Gruys (1999) 363 Gruys (1999) 363 Chen & Spector (1992) 387 Gruys (1999) 363 Gruys (1999) 363 Gruys (1999) 363 Gruys (1999) 363 Hunt (1996)1 4,102 Gruys (1999) 363 Gruys (1999) 363 Marcus et al. (2002) 174 Gruys (1999) 363 Gruys (1999) 363 Unit weighted mean

r

rho

0.49 0.29 0.39 0.59 0.52 0.38 0.27 0.56 0.42 0.57 0.36 0.54 – 0.42 0.55 0.39 0.24 0.47 0.42

0.66 0.42 0.56 0.80 0.71 0.50 0.35 0.68 0.55 0.75 0.42 0.63 0.66 0.53 0.69 0.67 0.31 0.62 0.55

Note: rho = true score correlation (i.e. observed correlation corrected for unreliability in both measures). 1 Data from multiple datasets or from metaanalyses.

221

222

Table 12.4

Relations of substance abuse and related behaviours with other CWB

Correlate

Substance abuse and related constructs

Study

Absenteeism/withdrawal Aggression Alcohol use Destruction of property Destruction of property Drug use Engaging in off-task behaviours1 Inappropriate physical actions Inappropriate physical actions Inappropriate verbal actions Inappropriate verbal actions Low personal standards Misuse of information Misuse of information Misuse of time & resources Misuse of time & resources Poor attendance Poor attendance

Substance abuse Substance abuse Drug use Alcohol use Drug use Alcohol use Drug misuse Alcohol use Drug use Alcohol use Drug use Substance abuse Alcohol use Drug use Alcohol use Drug use Alcohol use Drug use

Marcus et al. (2002) Marcus et al. (2002) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Hunt (1996)1 Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Hakstian et al. (2002) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999)

N

r

rho

174 174 363 363 363 363 1,818 363 363 363 363 1,019 363 363 363 363 363 363

0.58 0.32 0.28 0.30 0.44 0.28 – 0.29 0.38 0.39 0.27 0.19 0.35 0.29 0.40 0.21 0.29 0.27

0.83 0.47 0.43 0.48 0.64 0.43 0.42 0.42 0.50 0.56 0.35 0.23 0.54 0.41 0.55 0.26 0.43 0.37

Poor-quality work Poor-quality work Property theft Tardiness/absenteeism1 Theft Theft Theft Theft/property violations Unruliness1 Unsafe behaviour Unsafe behaviour Work avoidance

Alcohol use Drug use Substance abuse Drug misuse Alcohol use Drug misuse Drug use Substance abuse Drug misuse Alcohol use Drug use Substance abuse

Gruys (1999) 363 Gruys (1999) 363 Hakstian et al. (2002) 1,019 Hunt (1996)1 1,827 Gruys (1999) 363 1,701 Hunt (1996)1 Gruys (1999) 363 Marcus et al. (2002) 174 Hunt (1996)1 2,051 Gruys (1999) 363 Gruys (1999) 363 Hakstian et al. (2002) 1,019 Unit weighted mean

0.20 0.38 0.21 0.26 0.38 – 0.27 0.38 – 0.33 0.17 0.13 0.30

0.28 0.49 0.29 0.33 0.56 0.52 0.37 0.70 0.44 0.51 0.24 0.17 0.44

Note: rho = true score correlation (i.e. observed correlation corrected for unreliability in both measures). 1 Data from multiple datasets or from metaanalyses.

223

224 The Big-5 Personality and Counterproductive Behaviours

correlates include aggression, absenteeism, low personal standards, and misuse of information, poor-quality work, theft, unruliness and unsafe behaviour. On average, the construct of substance abuse correlates .44 with other facets of CWB. Employee theft and property violations have been discussed as potentially good marker behaviours for counterproductivity (Ones, Viswesvaran and Schmidt, 1993). The relationships between theft and property violations are reviewed in Table 12.5. The first column lists various CWB facets that have been correlated with theft and property violations. The second column indicates specific theft-related measures used in studies. Columns 3 through 6 provide study-level information, including authors, sample size, observed correlation and unreliability corrected correlation (i.e. rho). As can be seen in Table 12.5, employee theft and property violations are highly related to constructs that include absenteeism, alcohol and drug use, engaging in off-task behaviours, inappropriate physical and verbal actions, low personal standards and a wide gamut of CWB. Across all the different CWB facets, the mean true correlation with theft/property violations is .62. Theft-related behaviours do appear to be prime markers for CWB. Workers who steal from their employers exhibit tardiness, absenteeism, and misuse of time at work, use drugs and alcohol, do poor-quality work, and engage in unsafe behaviours, as well. Earlier in this chapter, we noted that Robinson and Bennett (1995) identified two types of counterproductive behaviours: organizational deviance and interpersonal deviance. These two clusters of behaviours distinguish between the organization and other individuals as the target of the counterproductive behaviour. To date, four studies have reported the relationships between the two. Table 12.6 reports the findings from these four studies. The last column of the table presents the rhos between organizational and interpersonal deviance. The mean rho between these two dimensions of CWB is .82. Lee and Allen (2002) have reported the construct level correlation between the two to be as high as .96. Unquestionably, organizational and interpersonal deviances are two closely related aspects of CWB. Employees who engage in organizational deviance also engage in interpersonal deviance and vice versa. In general, this section of our chapter has established that counterproductivity can best be viewed as a broad construct with many different behavioural manifestations. In reviewing this literature, what was surprising for us is the positive, substantial relationships we observed across so many different CWBs. An extraordinary finding in empirical studies

Table 12.5

Relations of theft/property violations and related behaviours with other CWB

Correlate

Theft/property violations and related behaviours

Study

Absenteeism/withdrawal Aggression Alcohol use Alcohol use Drug misuse1 Drug use Drug use Engaging in off-task behaviours1 Hostility & complaints Inappropriate physical actions Inappropriate physical actions Inappropriate verbal actions Inappropriate verbal actions Interpersonal aggression Low personal standards Misuse of information Misuse of information Misuse of time & resources Misuse of time & resources Poor attendance Poor attendance Poor-quality work Poor-quality work

Theft/property violations Theft Destruction of property Theft Theft Destruction of property Theft Theft Sabotage Destruction of property Theft Destruction of property Theft Sabotage Property theft Destruction of property Theft Destruction of property Theft Destruction of property Theft Destruction of property Theft

Marcus et al. (2002) Marcus et al. (2002) Gruys (1999) Gruys (1999) Hunt (1996)1 Gruys (1999) Gruys (1999) Hunt (1996)1 Chen & Spector (1992) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Chen & Spector (1992) Hakstian et al. (2002) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999) Gruys (1999)

N

r

rho

174 174 363 363 1,701 363 363 2,280 387 363 363 363 363 387 1,019 363 363 363 363 363 363 363 363

0.40 0.39 0.30 0.38 – 0.44 0.27 – 0.45 0.59 0.42 0.52 0.55 0.57 0.11 0.55 0.71 0.43 0.74 0.38 0.63 0.51 0.55

0.68 0.67 0.48 0.56 0.52 0.64 0.37 0.70 0.70 0.80 0.53 0.71 0.69 0.66 0.16 0.80 0.96 0.56 0.89 0.53 0.82 0.68 0.68

225

226

Table 12.5

(Continued)

Correlate

Theft/property violations and related behaviours

Study

N

Substance abuse Substance use Tardiness/absenteeism1 Theft Unruliness1 Unsafe behaviour Unsafe behaviour Work avoidance

Property theft Theft/property violations Theft Destruction of property Theft Destruction of property Theft Property theft

Hakstian et al. (2002) 1,019 Marcus et al. (2002) 174 Hunt (1996)1 2,289 Gruys (1999) 363 Hunt (1996)1 2,513 Gruys (1999) 363 Gruys (1999) 363 Hakstian et al. (2002) 1,019 Unit weighted mean

r

rho

0.21 0.38 – 0.47 – 0.35 0.46 0.17 0.44

0.29 0.70 0.52 0.66 0.74 0.51 0.62 0.27 0.62

Note: rho = true score correlation (i.e. observed correlation corrected for unreliability in both measures). 1 Data from multiple datasets or from metaanalyses.

Deniz S. Ones and Chockalingam Viswesvaran 227 Table 12.6 deviance

The relationship between organizational deviance and interpersonal

Study Marcus et al. (2002) Lee and Allen (2002) Fox et al. (2001) Bennett and Robinson (2000) Unit Weighted Mean

N

r

rho

174 218 292 352

0.59 – 0.63 0.65 0.62

0.76 0.96 0.69 0.86 0.82

Note: rho = true score correlation (i.e. observed correlation corrected for unreliability in both measures).

that have explored these specific dimensions developed independently in so many contexts, is the positive correlation observed across them. It is noteworthy that although many of studies reviewed used self-report instruments, this was not necessarily so. Some authors did not use selfreports. For example, Hunt (1996), Koslowsky et al. (1997), and Skarlicki and Folger (1997) did not rely on self-report measures of CWB. Future research will need to disentangle the impact of measurement method on CWB relationships reported (see work in this vein, but for the construct of overall job performance, Viswesvaran, Schmidt and Ones, 1996). Regardless of the measurement method, we suspect that the relationships among different forms of CWB will remain positive and substantial. Our purpose in this section of our chapter was to describe the linkages among different forms of CWB. Following Hogan and Hogan’s (1989) theory, that organizational delinquency is a broad phenomenon that includes many types of disruptive behaviours on the job, we posited that seemingly different forms of CWB would be aspects of an overall construct. Data appear to support this conclusion. These data come from different situations, settings, types of study participants and even countries. Hence, it is hard for us to attribute unmistakable similarities across study results to situational variables. Together these findings suggest that a potential reason for the positive associations among different forms of counterproductivity is the presence of common individual differences antecedents. Dispositional basis of counterproductivity at work has been examined in studies examining the links between personality and counterproductivity.

228 The Big-5 Personality and Counterproductive Behaviours

Relationships between personality traits and CWB The studies reviewed above implicate common individual differences antecedents of CWB. In this section of our chapter, we review and discuss results of quantitative reviews that have directly demonstrated personality–CWB relationships. Individual differences in personality traits have been studied in connection with different forms of counterproductive behaviours. A typical study collects personality data as well as counterproductive behaviours engaged in by a sample of individuals. The two scores are correlated, and the correlation commonly referred to as the criterion-related validity is taken as an index of the personality–CWB relationship. Rather than discussing individual studies, meta-analytic summaries are used here to draw inferences from the available literature. Meta-analysis is a statistical technique that seeks, among other goals, to determine (a) the mean correlation between two constructs unaffected by statistical artifacts and (b) the extent to which the observed variance of findings across studies results from statistical artifacts. If the validity is substantially dependent on the situation, statistical artifacts will not account for all or nearly all of the observed variation in the validities. In addition to estimating the portion of the observed variance that is due to statistical artifacts, meta-analysis also provides the most accurate obtainable estimate of the mean operational (true) validity. In this case, operational validity refers to the relationship between personality scales and CWB, corrected for unreliability in the criterion and range restriction (where appropriate), but not unreliability in the predictor (personality measures). As such, operational validity is an index of the potential predictive usefulness of personality scale scores. We organize this section of our chapter around two types of personality measures utilized in personnel staffing decisions. First, there are measures of normal adult personality. The aim in the construction of these scales is the accurate description of individual differences in personality. The use of these inventories for personnel screening and selection is only one of their many applications. Second, during the past decade, a number of measures of personality at work have come to the attention of the scientific community. Such measures are generically referred to as occupational personality scales (Ones and Viswesvaran, 2001a, 2001b). The main aim of these measures is the accurate prediction of individual differences in work behaviours of interest. There are four defining characteristics of occupational personality scales: they are (1) paper-and-pencil instruments containing items similar to those found on traditional

Deniz S. Ones and Chockalingam Viswesvaran 229

personality scales, (2) specifically developed to assess personality constructs of relevance for work environments, (3) designed for use with job applicants (this is reflected in their normative data), and, perhaps most important, (4) designed to predict work behaviours.

The relationships between normal adult personality scales and CWB The use of personality variables in personnel selection has seen a resurgence during the past decade (Barrick and Mount, 1991; Hogan and Hogan, 1989; Hough et al., 1990; Ones, Viswesvaran and Schmidt, 1993; Sackett, Burris and Callahan, 1989). This renaissance is partly fuelled by meta-analytic studies demonstrating useful validities for personality variables across situations (Barrick and Mount, 1991; Hough and Oswald, 2000; Ones et al., 1993; Ones and Viswesvaran, 1998; Salgado, 1997), and partly by an emerging consensus among personality researchers that numerous personality measures available in the literature can be expressed within the Big Five framework (Costa and McCrae, 1992; Digman, 1990; Goldberg, 1993; Ones, Viswesvaran and Reiss, 1996). Even though disagreements still exist about the Big Five (e.g. Block, 1995; Eysenck, 1991), the considerable consensus that the personality of an individual can be described in terms of the Big Five factors and their facets has provided a workable taxonomy, thereby facilitating meta-analytic cumulation. (The interested reader is referred to Block, 1995; Digman, 1990; John, 1990a,b; John, Angleitner and Ostendorf, 1988, for detailed descriptions of the history of the Big Five.) The Big Five are robust and generalizable across different types of assessments, rating sources, cultures, language, and gender as well as a variety of factor extraction and rotation methods (e.g. Digman and TakemotoChock, 1981; Goldberg, 1990; McCrae and Costa, 1987, 1997; Norman, 1963; Ostendorf and Angleitner, 1994; Saucier and Goldberg, 1998). The Big Five dimensions of personality are emotional stability (or neuroticism, if measured from its negative pole), extraversion, openness to experience (also termed intellect and culture), agreeableness, and conscientiousness (Costa and McCrae, 1992) Emotional stability refers to the anxiety, anger–hostility, self-consciousness, impulsiveness, vulnerability and depression experienced by an individual. Extraversion refers to the extent to which a person is gregarious, assertive, excitementseeking, warm and active. Positive emotions (enthusiastic, humorous, optimistic, jolly) are also part of extraversion (Costa and McCrae, 1992). Openness to experience has been interpreted as both intellect and culture. Traits commonly associated with this dimension include imagination, curiosity, originality, broad-mindedness, and artistic sensitivity. Traits

230 The Big-5 Personality and Counterproductive Behaviours

associated with agreeableness (likeability) include courteousness, flexibility, trust, good-naturedness, cooperativeness, forgiveness, empathy, softheartedness, and tolerance. The conscientiousness dimension has been called dependability, prudence or conformity. Because of its relationship to a variety of educational achievement measures and its association with volition, the dimension has also been called ‘will to achieve’. Traits associated with this dimension reflect both dependability (i.e. carefulness, thoroughness, responsibility, organization, efficiency, planfulness) and volition (i.e. hard work, achievement orientation, perseverance). Each of the Big Five dimensions is regarded as a continuum. Salgado (2002) used meta-analysis to summarize the scant evidence for the relationships between the Big Five personality dimensions and counterproductive work behaviours. Salgado’s (2002) key findings are summarized in Table 12.7. There were three CWB-related criteria that Salgado (2002) studied: deviant behaviours, accidents and absenteeism. The operational validities (i.e. criterion-related validities corrected for unreliability in the criterion and range restriction, where possible) were disappointing. Neither Table 12.7 CWB

Relationships between the Big Five dimensions of personality and

Criterion

Facet

Deviant behaviour (lack of)

Emotional stability Extraversion Openness to experience Agreeableness Conscientiousness Emotional stability Extraversion Openness to experience Agreeableness Conscientiousness Emotional stability Extraversion Openness to experience Agreeableness Conscientiousness

Accidents (lack of)

Absenteeism (lack of)

N

K

Operational validity

3,107 2,383 1,421 1,299 6,276 2,121 2,341 1,660 1,540 2,094 2,491 1,799 1,339 1,339 2,155

15 12 8 9 13 5 7 5 4 6 12 10 8 8 10

.06 −.01 −.14 .20 .26 .08 .04 −.09 .01 .06 −.04 −.08 −.00 −.04 .06

Note: Findings are summarized from Salgado (2002). Positive correlations mean that high scores on personality measures are positively associated with NOT engaging in deviant behaviour, NOT getting involved in accidents, and NOT being absent. Operational Validity = validity corrected for unreliability in the criterion and range restriction, but NOT for unreliability in the predictor.

Deniz S. Ones and Chockalingam Viswesvaran 231

absenteeism nor accidents were predicted well by any of the Big Five dimensions. However, in the prediction of general deviant behaviours, operational validities for agreeableness and conscientiousness were substantial − .20 and .26, respectively. These findings suggest that individuals who score high on conscientiousness and agreeableness scales tend to refrain from engaging in deviant behaviours at work. Personality traits of conscientiousness and agreeableness appear to be useful correlates of workplace deviance. Of all the Big Five dimensions of personality, conscientiousness has perhaps received the most attention in IWO psychology. This is partly because all meta-analyses examining the personality predictors of job performance have found generalizable validities for conscientiousness (Barrick, Mount and Judge, 2001; Ones and Viswesvaran, 1996). A widespread consensus (e.g. Digman, 1990; Eysenck, 1991; Goldberg, 1993) in the field of personality is that the description of personality can be undertaken in a hierarchy of levels with specific descriptions at the lower levels of the hierarchy. Conscientiousness is no exception. Most contemporary personality and IWO psychologists agree that two of the main sub-dimensions of conscientiousness are achievement orientation (including industriousness) and dependability (including reliability). (Other sub-dimensions posited vary by researcher. For example, Saucier and Ostendorf (1999) suggest additional dimensions of orderliness and decisiveness. Hough and Ones (2001) offer impulse control, order, persistence, and moralistic as additional sub-dimensions.) In a series of papers, Hough and colleagues (Hough et al., 1990; Hough, 1992 and 1998) have reported validities of conscientiousness facets of achievement and dependability with various CWB. Table 12.8 summarizes Hough and colleagues’ findings for law-abiding behaviour, avoiding delinquency, lack of substance abuse, avoiding irresponsible behaviour and avoiding CWB. The general trends in operational validities indicate that both conscientiousness sub-dimensions of achievement and dependability have sizable predictive value in relation to all CWB included in the analyses. In predicting the avoidance of CWB and generally irresponsible behaviours, the two facets of conscientiousness appear to have similar levels of validities. Employees who are achievement striving and dependable tend to refrain from CWB. The strength of these effects is fairly large. For example, dependability correlated .47 with avoiding CWB, across 66 studies and 113,427 individuals. The differences in magnitudes of agreeableness and conscientiousness vis-à-vis lower facets of conscientiousness may be due to second order sampling error (as we already

232 The Big-5 Personality and Counterproductive Behaviours Table 12.8 for CWB

Summary of criterion-related validities for conscientiousness facets

Criterion

Conscientiousness Study facet

Law-abiding behaviour

Achievement Dependability

Delinquency (lack of)

Achievement Dependability

Substance abuse (lack of)

Achievement Dependability

Irresponsible Achievement behaviour (lack of) Dependability

Counterproductive Achievement behaviour (lack of) Dependability

Hough (1992) Hough (1992) Hough et al. (1990) Hough et al. (1990) Hough et al. (1990) Hough et al. (1990) Hough (1992, 1998) Hough (1992, 1998) Hough (1998) Hough (1998)

N

K

Operational validity

2

.421

25,867 22

.581

–

4

.351

–

10

.271

–

–

–

–

25

.281

4

.26

98,676 69

.32

10,062

4

.51

113,427 66

.47

5,918

19,476

Note: Operational Validity = validity corrected for unreliability in the criterion only and not for restriction or for unreliability in the predictor. 1 Observed validities, not corrected for unreliability in the criteria.

noted, some analyses are based on scant data), and differences in the specific criteria examined. Additional future primary studies should be devoted to studying the relationships of the Big Five personality dimensions and lower-level facets with CWB.

The relationships between criterion-focused occupational personality scales (COPS) and CWB In general, occupational personality scales have been developed for particular job families (e.g. managers, sales people, clerical workers) or for predicting particular criteria of interest (violence at work, employee theft). The former can be referred to as job-focused occupational personality scales (JOPS) in that they aim to predict job performance constructs for particular occupational categories. Examples of such

Deniz S. Ones and Chockalingam Viswesvaran 233

scales include Managerial Potential Scales and Sales Potential Scales. The latter can be referred to as criterion-focused occupational personality scales (COPS) in that they aim to predict particular criteria of interest in work environments. Examples include integrity tests (which aim to predict dishonest behaviours at work), violence scales (which aim to predict violent behaviours at work), drug and alcohol avoidance scales (which aim to predict substance abuse at work), stress tolerance scales (which aim to predict handling work pressures well) and customer service scales (which aim to predict serving customers well). Each of these COPS has been used in predicting various forms of CWB. Before we describe the relations between COPS and CWB, we provide a brief overview of different types of COPS. Integrity tests, which were specifically developed to assess dependability, integrity, and honesty of applicants to help predict theft and future on-thejob dishonest behaviours, are considered to be the prototypical criterionfocused occupational personality scales (Ones and Viswesvaran, 2001a; 2001b). Sackett and colleagues (Sackett et al., 1989; Sackett and Wanek, 1996) classified integrity tests into two categories: ‘Overt integrity tests’ and ‘Personality-based tests’. Overt integrity tests are designed to directly assess attitudes regarding dishonest behaviours. Overt integrity tests include the London House Personnel Selection Inventory (PSI), the Employee Attitude Inventory (EAI), the Stanton Survey, the Reid Report, the Phase II Profile, the Milby Profile, the Trustworthiness Attitude Survey, and the Pre-employment Analysis Questionnaire. Ones (1993) found that there are sizable correlations among overt integrity tests. Personalitybased measures, on the other hand, are intended to predict a broad range of counterproductive behaviours at work (e.g. violence on the job, absenteeism, tardiness, drug abuse, in addition to theft) using personality scale items. Unlike overt tests, the initial intent in the construction of these measures has not been the prediction of theft or theft-related behaviours. Examples of personality-based measures that have been used in integrity testing include the Personal Outlook Inventory, the Personnel Reaction Blank, the Employment Inventory (Personnel Decisions, Inc.), and the Hogan Personality Inventory Reliability Scale. Paper-and-pencil drug and alcohol scales (not urinalysis) are measures specifically developed to assess ‘drug and alcohol abuse potential’ of job applicants for the purpose of predicting drug and alcohol abuse. Drug and alcohol scales specifically attempt to tap into attitudes, values and perceptions of test-takers toward drug and alcohol abuse. Drug and alcohol scales contain items inquiring after drug and alcohol related attitudes of job applicants. None of the questions focus on

234 The Big-5 Personality and Counterproductive Behaviours

other counterproductive behaviours, such as theft. It should be noted that users of drug and alcohol scales do not disguise the goals of the scale. They are clear purpose tests in that test-takers can readily discern what is being assessed. Currently available drug and alcohol scales include London House Personnel Selection Inventory Drug Scale, Employee Reliability Inventory Alcohol Substance Use Scale, Accutrac Substance Abuse Scale, Orion Survey Drug and Alcohol Attitudes Scale, and PEAK Procedure Substance Abuse Scale. Stress tolerance scales are paper-and-pencil tests that have been developed for the purpose of identifying job applicants who can ‘handle pressure well’ and who are ‘not tense and anxious’ (Hogan and Hogan, 1992). These scales are used primarily for personnel selection. Examples of stress tolerance scales include Hogan Personality Inventory Stress Tolerance Scale, Employee Attitude Inventory Job Burnout Scale, PEOPLE Survey Wellness Scale, and the PEAK Procedure Stress Scale. Customer service inventories have been developed to assess the potential for customer service among applicants (e.g. Hogan et al., 1984; McLellan and Paajanen, 1994; Sanchez and Fraser, 1993; Saxe and Weitz, 1982). These are paper-and-pencil tests, specifically developed to predict customer service performance. There appear to be at least six such pre-employment selection measures available to US organizations (McDaniel and Frei, 1994). Example inventories include the PDI Customer Service Inventory, Hogan Personality Inventory Service Orientation Scale, and the London House Customer Relations Scale. Violence scales aim to predict violent behaviours at work. Examples of violence scales can be found along with integrity tests on batteries marketed by integrity test publishers (e.g. Inwald Personality Inventory, London House-Personnel Selection Inventory). Recently, Ones and Viswesvaran (2001a and 2001b) showed that these COPS are positively and substantially related to one another. In other words, integrity tests, customer service scales, violence scales, drug and alcohol scales are all related. Furthermore, they all assess a combination of Big Five constructs of conscientiousness, agreeableness and emotional stability. The weights given to these three Big Five constructs vary across the COPS under investigation. The theoretical meaning of such a clustering of personality traits is informed by a study conducted by Digman (1997). He carried out a series of factor analyses on fourteen different databases, examining whether there was evidence of higher order factors than the Big Five. The databases used both self and other ratings of personality. The data were collected over 4,269 individuals. He found evidence for two higher order

Deniz S. Ones and Chockalingam Viswesvaran 235

factors to the Big Five. The first higher order construct was defined by the overlap between conscientiousness, agreeableness, and emotional stability. Digman termed this super factor ‘alpha’. Extraversion and openness to experience defined the second higher order construct (beta). COPS appear to tap into factor alpha. Ones and Viswesvaran (2001b) point out that the conglomeration of traits assessed by COPS correspond to Digman’s (1997) factor alpha (i.e. the socialization higher order factor of personality – a higher order factor than the Big Five) and are particularly relevant in the prediction of behaviours at work. We should note that designers of COPS reviewed did not intend to capture Digman’s (1997) alpha, nor did they intend to assess conscientiousness, agreeableness, and emotional stability in combination. However, they did intend to create personality scales that predicted certain occupational criteria. Items that proved valuable and useful for prediction purposes were later found to assess Digman’s (1997) alpha (Ones and Viswesvaran, 2001a, 2001b). It is reasonable to expect that factor alpha, as measured by COPS, would be particularly well suited to predict CWB. COPS incorporate conscientiousness elements of work ethic, dutifulness, rule-following, impulse control, dependability, and reliability; agreeableness components of trust and non-hostility; and emotional stability element of stress tolerance. Further, COPS were created explicitly to be used in the prediction of work behaviours. Table 12.9 summarizes the operational validities of COPS for both specific facets of CWB and overall counterproductivity. We first review the criterion-related validity evidence for specific facets of CWB. Integrity tests show sizable operational validities for admitted theft (operational r = .42), accidents on the job (operational r = .52), and property damage (operational r = .69). Personality-based integrity tests appear to be quite good predictors of voluntary absenteeism (operational r = .33). However, overt tests have been found to be quite poor in predicting voluntary absenteeism (operational r = .09) and detected theft (operational r = .13). Violence scales appear to have excellent validity in predicting violent behaviours (operational r = .48); while drug and alcohol scales appear to have excellent validity in predicting both drug use (operational r = .33) and alcohol use on the job (operational r = .41). The pattern that emerges from these meta-analyses is that COPS, which all assess Digman’s factor alpha (the higher order factor of personality composed of conscientiousness, agreeableness, and emotional stability), are substantially correlated with diverse facets of CWBs. The only exception to this, overt integrity tests in predicting detected theft and

236

Table 12.9

Operational criterion-related validities of criterion-focused personality scales for predicting CWB

Criterion

Facets of CWB Detected theft1 Admitted theft1 Absenteeism1 Absenteeism1 Accidents on the job Property damage Violent behaviours Drug use Alcohol use on the job Overall broad CWB Counterproductive behaviours1 Counterproductive behaviours1 Counterproductive behaviours Counterproductive behaviours Counterproductive behaviours Counterproductive behaviours

Personality measure

Study

N

K

Operational validity

Overt integrity tests Overt integrity tests Overt integrity tests Personality-based integrity tests Integrity tests Integrity tests Violence scales Drug and alcohol scales Drug and alcohol scales

Ones et al. (1993) Ones et al. (1993) Ones et al. (in press) Ones et al. (in press) Ones & Viswesvaran (1996) Ones & Viswesvaran (1996) Ones & Viswesvaran (2001a) Ones & Viswesvaran (2001b) Ones & Viswesvaran (2001b)

2,434 68,618 8,508 4,922 759 1,970 1,265 931 128

7 63 9 13 5 14 11 3 3

.13 .42 .09 .33 .52 .69 .48 .33 .41

Personality-based integrity tests Overt integrity tests Drug and alcohol scales Violence scales Stress tolerance scales Customer service scales

Ones et al. (1993) Ones et al. (1993) Ones & Viswesvaran (2001b) Ones & Viswesvaran (2001a) Ones & Viswesvaran (2001a) Ones & Viswesvaran (2001a)

93,092 5,598 665 533 594 740

62 10 3 4 5 5

.29 .39 .29 .46 .42 .42

Note: Operational Validity = validity corrected for unreliability in the criterion (where possible) and range restriction, but NOT for unreliability in the predictor. 1 Focusing only on predictive studies conducted on job applicants.

Deniz S. Ones and Chockalingam Viswesvaran 237

voluntary absenteeism, might be due to the nature of the criterion or the specific developmental features of tests under consideration. Ones et al. (1993) discusses the relative merits and deficiencies of the detected theft criterion. Ones et al. (in press) offers potential reasons for the low validity observed between overt integrity tests and voluntary absenteeism. We refer the interested readers to these original sources. Even though COPS such as integrity tests, drug and alcohol scales, customer service scales, and violence scales are each constructed to predict specific behaviours, including specific facets of CWB, they have all been used to predict a broad, overall counterproductivity criterion as well. The bottom half of Table 12.9 presents the meta-analytically obtained operational validities for integrity tests, drug and alcohol scales, violence scales, stress tolerance scales and customer service scales for broad CWB. Recall that the purpose in the constructions of these COPS was the prediction of specific CWB, but each shows substantial validities for a broader CWB criterion. Even the lowest operational validities tabled are notable. Both overt integrity tests and drug and alcohol scales predict CWB with an operational validity of .29. Validities of personality-based integrity tests, violence scales, stress tolerance scales, and customer service scales are extraordinarily similar for overall CWB, .39, .46, .42, and .42, respectively. There is convincing evidence that scales that assess factor alpha appear to be behind the COPS that predict broad CWB. Theoretically, ‘factor alpha represents the socialization process itself. From Freud (1930) to Kohut (1977), from Watson (1929) to Skinner (1971), personality theorists of various persuasion have been concerned with the development of impulse restraint and conscience, and the reduction of hostility, aggression, and neurotic defense. From this point of view, factor alpha is what personality development is all about. Thus, if all proceeds according to society’s blueprint, the child develops superego and learns to restrain or redirect id impulses and to discharge aggression in socially approved ways. Failure of socialization is indicated by neurosis, by deficient superego, or by excessive aggressiveness’ (Digman, 1997, pp. 1249–50). COPS are personality scales that assess individuals who function well according to social rules. Even though these scales have come into existence when industrial psychologists have been faced with organizational personnel selection problems (i.e. in developing personality scales to predict specific criteria such as drug use, customer service behaviours), they are also useful in predicting CWB. Functioning according to society’s rules in organizational settings is an important individual differences characteristic. Hence, factor alpha is perhaps the

238 The Big-5 Personality and Counterproductive Behaviours

most important personality trait that needs to be systematically measured among job applicants. Scoring high on this higher order personality trait predicts a whole spectrum of CWB, from avoiding drug and alcohol use and violent behaviours to refraining from absenteeism to avoiding getting into accidents. In fact, validities as high, if not higher, have been obtained for overall CWB. All in all, the findings reviewed support a dispositional basis of counterproductivity. We suggest that the positive associations among different forms of counterproductivity are likely to be due to individual differences in factor alpha. Thus far in this chapter, we have offered empirical evidence that different forms of counterproductivity are related to each other, implicating common individual differences antecedents. We also offered metaanalytic evidence that both conscientiousness sub-dimensions of achievement and dependability as well as higher order constructs assessed by COPS are likely to fulfil that role. In the next section of this chapter we turn our attention to the question of why.

How and why personality traits come to influence CWB: some potential pathways to counterproductivity In this section of our chapter, based on supporting empirical and theoretical evidence from various authors and based on evidence presented earlier, we propose different causal mechanisms through which personality traits may correlate with CWB. Given that empirical research already reviewed in this chapter has ascertained that conscientiousness facets of achievement and dependability as well as factor alpha (i.e. the higher order construct assessed by COPS – overarching conscientiousness, agreeableness and emotional stability) are linked to CWB, we focus on the roles these particular traits can play in CWB. In explicating their roles, a distinction to keep in mind is that personality determinants of CWB can be viewed as enabling or inhibitory. The mediational causal chains outlined below offer insights into both. Individual differences in overall counterproductivity and CWB facets may be viewed as a function of individual differences in behavioural tendencies and habits; knowledge and skill; affect, emotions, and mood; volition and motivation; and cognition, attitudes, beliefs, values and attributions. These are the proximal determinants of CWB. Personality refers to ‘an individual characteristic patterns of thought, emotion and behavior, together with the psychological mechanisms behind those patterns’ (Funder, 1997, p. 447). Personality traits as measured on personality scales either (1) provide direct assessments of the proximal

Deniz S. Ones and Chockalingam Viswesvaran 239

determinants (e.g. characteristic patterns of thought, behavioural tendencies) or (2) act as determinants of the more proximal determinants of CWB via four main pathways: (a) knowledge and skill, (b) affect and mood, (c) motivation and volition, and (d) cognition, including attitudes, beliefs, values and attributions.

Knowledge and skill CWB cannot occur unless the individual engaging in the behaviour knows how to engage in it. In other words, CWB must be in the person’s behavioural repertoire. For example, an individual first needs to know how to use computers and where data protection system weaknesses are before he or she can hack into the company’s secure database. Even for cognitively simpler forms of CWB, such as unwarranted absenteeism, knowledge and skill about how not to get caught or how to get away with the behaviour is necessary. Individuals’ interest patterns are closely linked to declarative and procedural knowledge acquired (Ackerman and Heggestad, 1997; Campbell, 1990). It may be that individuals who score high on dependability, integrity scales and the like simply do not have the interest in figuring out how to break rules and get away with it or in learning how to cheat their organizations. As such, individuals scoring high on the personality trait of factor alpha have at least one fewer path (i.e. via knowledge and skill) to CWB.

Affect, emotion and mood It is possible that personality traits might come to predict CWB via affective mechanisms that involve emotions and mood. Individual differences in personality traits are related to affect and mood (Diener et al., 1984; Emmons and Diener, 1986; Watson, 1988). Individuals differ in their tendency to respond to classes of environmental stimuli in a predetermined, affect-based manner. Negative affect refers to distress and dissatisfaction (i.e. negative emotions) experienced in interacting with the environment. On the other hand, positive affect refers to goodwill and vibrancy (i.e. positive emotions) experienced in interacting with the environment (Connolly and Viswesvaran, 2000). Affective disposition is a casual antecedent of subjective well-being (Steel and Ones, 2002). Moods are transient (Lane and Terry, 2000; Watson, 2000) subjectively experienced feeling states, experienced as less intense (Davidson, 1994; Ekman, 1994; Isen, 1984; Jacobson, 1957) and lasting longer (Davidson, 1994; Ekman, 1994; Goldsmith, 1994; Kagan, 1994; Lazarus, 1994; Watson, 2000) than pure emotions. Affect, emotions and mood influence thinking, perception, and behaviour (Morris, 1989;

240 The Big-5 Personality and Counterproductive Behaviours

Thayer, 1996). For example, in workplace settings, positive mood is associated with increased prosocial behaviour (George, 1991), decreased absenteeism (George, 1989), and improved sales performance (George and Bettenhausen, 1990). Employee negative emotions of jealousy and envy have been linked to undesirable worker reactions (Vecchio, 2000). Affective experiences associated with past behaviours may be instrumental in shaping future behaviour. Emmons et al. (1985) suggest that ‘affect experienced in the situation will partially determine future decisions to enter or avoid the situation . . . thus, to the extent that individuals experience affect which is compatible with their psychological predispositions, the probability of choosing that situation again in the future will be increased’ (p. 695). Individuals who score high on integrity tests and other COPS, tend to be honest, well socialized, cooperative and stable. Engaging in CWB requires that the individual engaging in such behaviour be comfortable with (at least experience little or no negative affect) and even experience positive affect when breaking rules. On the other hand, engaging in CWB would be expected to create aversive affective reactions in individuals who score high on factor alpha. For these individuals, feelings of guilt and discomfort associated with rule-breaking or going against societal norms would make it more likely that future decisions would be made to avoid CWB. Although such affective mechanisms would work to increase or decrease the likelihood of subsequent counterproductive behaviours, they are of limited use when explaining why individuals choose to behave in counterproductive ways in the first place.

Motivation and volition Motivation refers to the ‘choice to expend effort, choice of level of effort to expend, and the choice to persist in the expenditure of that level of effort’ (Campbell, 1990, p. 706). The main components of all wilful behaviour are direction (i.e. what the person chooses to do), intensity (i.e. how much effort the person chooses to devote to the activity), and persistence (how long the person devotes to the activity). We maintain that CWB will not occur unless there is a choice to engage in the behaviour. Therefore, volition and motivation (consciously or unconsciously) is always a direct determinant of CWB. Resources that individuals can devote to various activities are not limitless (Baumeister, 1997; Kanfer, 1990). When employees engage in counterproductive behaviours at work, it is a choice whether to allocate resources (cognitive resources, effort, etc.) toward that CWB rather than other socially and organizationally sanctioned behaviours. Dependable, reliable

Deniz S. Ones and Chockalingam Viswesvaran 241

individuals behaviourally gravitate toward socially approved behaviours. They are more likely to choose to expend their energies on organizationally sanctioned behaviours such as task performance and organizational citizenship behaviours (Borman et al., 2001; Organ and Ryan, 1995). Motivational and volitional paths to CWB pose a problem for explaining the role that achievement orientation plays in CWB. What explains the negative relationships that we noted between achievement orientation and CWB? After all, an achievement-oriented criminal may choose to expend high levels of effort in criminal activities! Fortunately, the traits of achievement orientation and dependability tend to go hand in hand (Costa and McCrae, 1992). That is, individuals who are high on achievement orientation also tend to play by society’s rules, and therefore choose to expend their energy in activities that are not frowned upon.

Cognition Unlike other types of behaviours in organizations (e.g. task performance, helping co-workers), CWB requires that employees overcome internal controls and inhibitions. The choice to refrain from CWB may be cognitively mediated. That is, personality traits such as integrity and dependability might influence CWB via their impact on cognitions. For example, individual beliefs and values influence the perceptions of legitimacy of a given CWB and its relative personal benefits. Personality traits influence such values, beliefs and attributions. ‘People refrain from behaviors that are not likely to be successful or if bad things are likely to happen to them as a result’ (Megargee, 1997). Some people are more likely to discount the potential consequences of their CWB or overcome personal inhibitions against CWB. Bad attitudes have been linked to bad behaviour (Bolin and Heatherly, 2001). Behaviour is ‘guided by perceptions of the current situation, memories of similar situations encountered in the past, impressions of ourselves and other people in the current environment, attributions concerning their and our own experiences and actions’ (Kihlstrom and Hastie, 1997, p. 712). In processing information about the environment, beliefs, attitudes, attributions, and self-concept all come to bear on perceptions, judgments and choices. For example, individuals process information about the likelihood of being caught, and relative benefits and costs of engaging in the CWB. Their attitudes, beliefs and values allow them to make various attributions in their causal reasoning. The ultimate decision to engage in CWB (e.g. be absent when not sick; steal from the company) is influenced by individuals’ causal reasoning.

242 The Big-5 Personality and Counterproductive Behaviours

Recently, Martinko, Gundlach and Douglas (2002) outlined a causal reasoning model of CWB. According to their model, personality traits exert their influence on CWB wholly through cognitive processing. Individuals who value being dependable (sometimes perhaps more so than being rich) and honest may choose to avoid counterproductivity because such behaviour would not be consistent with their values and self-image (Roccas et al., 2002). CWB in organizations may arise because low-integrity individuals may not share the values of the society at large, or they may through rationalizations and justifications choose to ignore the value systems in place. As Megargee (1997) notes, ‘Rationalization can be used to convince us that an act that appears to be wrong actually is not . . . As the old country preacher stated “A lot of sinners slide into hell on their buts.”’ (p. 604). In the preceding paragraphs, we outlined some potential pathways to CWB in organizations and we attempted to link four proximal determinants of CWB to the personality traits, including conscientiousness sub-dimensions and the higher order personality trait of factor alpha assessed by integrity tests and other COPS. Admittedly, these hypothesized causal chains are probably not the only routes to CWB. Future theoretical work will need to specify and test other potential process mechanisms. For example, some support exists for a mediational role of negative affect in predicting CWB. Fox, Spector and Miles (2001) found that negative affect mediated job stressor–CWB relationships. In addition to the mediational effects hypothesized above, it may also be important to explore personality-based moderator effects. Particularly, the effects of situational determinants of CWB may be moderated by personality traits. We did not review potential moderational effects in this chapter, as Cullen and Sackett (in press) offer an excellent discussion of this topic.

Conclusion The main thesis in this chapter was that personality traits are powerful determinants of CWB. We first showed that multiple, seemingly distinct counterproductive behaviours are positively and substantially correlated, suggesting a dispositional basis for CWB. Our review of meta-analyses examining the relationships between personality traits and CWB revealed that conscientiousness facets of dependability and achievement orientation as well a higher order personality trait (factor alpha, composed of conscientiousness, agreeableness and emotional stability) are correlated with CWB, at the global construct level and at

Deniz S. Ones and Chockalingam Viswesvaran 243

the dimensional level. We also offered theoretical and causal discussions of how and why personality traits come to influence CWB. Theories of counterproductive behaviours at work that ignore personality constructs, especially of conscientiousness facets and factor alpha, are at odds with both (1) voluminous empirical data and (2) contemporary causal theories of work behaviour.

References Ackerman, P. L. and Heggestad, E. D. (1997) ‘Intelligence, personality, and interest: evidence for overlapping traits’, Psychological Bulletin, 121(2), 219–46. Ashton, M. C. (1998) ‘Personality and job performance: the importance of narrow traits’, Journal of Organizational Behavior, 19, 289–303. Barrick, M. R. and Mount, M. K. (1991) ‘The Big Five personality dimensions and job performance: a meta-analysis’, Personnel Psychology, 44, 1–26. Barrick, M. R., Mount, M. K. and Judge, T. A. (2001) ‘Personality and performance at the beginning of the new millennium: what do we know and where do we go next?’, International Journal of Selection and Assessment, 9(1–2), 9–30. Baumeister, R. F. (1997) ‘Identity, self-concept, and self-esteem: the self lost and found’, in R. Hogan, J. Johnson and S. Briggs (eds), Handbook of Personality Psychology (pp. 681–711). San Diego, CA: Academic Press, Inc. Bennett, R. J. and Robinson, S. L. (2000) ‘Development of a measure of workplace deviance’, Journal of Applied Psychology, 85(3), 349–60. Block, J. (1995) ‘A contrarian view of the five-factor approach to personality description’, Psychological Bulletin, 117, 187–215. Bolin, A. and Heatherly, L. (2001) ‘Predictors of employee deviance: the relationship between bad attitudes and bad behavior’, Journal of Business and Psychology, 15(3), 405–18. Borman, W. C., Penner, L. A., Allen, T. D. and Motowidlo, S. J. (2001) ‘Personality predictors of citizenship performance’, International Journal of Selection and Assessment, 9, 52–69. Campbell, J. P. (1990) ‘The role of theory in Industrial and organizational psychology’, in M. D. Dunnette and L. M. Hough (eds), Handbook of Industrial and Organizational Psychology, Vol. 1 (pp. 39–73). Palo Alto: Consulting Psychologists Press, Inc. Chen, P. Y. and Spector, P. E. (1992) ‘Relationships of work stressors with aggression, withdrawal, theft and substance use: an exploratory study’, Journal of Occupational and Organizational Psychology, 65, 177–84. Connolly, J. J. and Viswesvaran, C. (2000) ‘The role of affectivity in job satisfaction: a meta-analysis’, Personality and Individual Differences, 29(2), 265–81. Costa, P. T. Jr. and McCrae, R. R. (1992) ‘Four ways five factors are basic’, Personality and Individual Differences, 13, 653–65. Cullen, M. J. and Sackett, P. R. (in press) ‘Personality and counterproductive workplace behavior’, in M. Barrick and A. M. Ryan (eds), Personality and Work. San Francisco: Jossey-Bass. Davidson, R. J. (1994) ‘On emotion, mood, and related affective constructs’, in P. Ekman and R. J. Davidson (eds), The Nature of Emotion: Fundamental Questions (pp. 51–5). New York: Oxford University Press.

244 The Big-5 Personality and Counterproductive Behaviours DeCresce, R., Mazura, A., Lifshitz, M. and Tilson, J. (1990) Drug Testing in the Workplace. Chicago: American Society of Clinical Psychologists Press. Diener, E., Larsen, R. J. and Emmons, R. A. (1984) ‘Person * Situation interactions: choice of situations and congruence response models’, Journal of Personality and Social Psychology, 47(3), 580–92. Diener, E., Larsen, R. J., Levine, S. and Emmons, R. A. (1985) ‘Intensity and frequency: dimensions underlying positive and negative affect’, Journal of Personality and Social Psychology, 48(5), 1253–65. Digman, J. M. (1990) ‘Personality structure: emergence of the five-factor model’, Annual Review of Psychology, 41, 417–40. Digman, J. M. (1997) ‘Higher order factors of the Big Five’, Journal of Personality and Social Psychology, 73, 1246–56. Digman, J. M. and Takemoto-Chock, M. K. (1981) ‘Factors in the natural language of personality: re-analysis and comparison of six major studies’, Multivariate Behavioral Research, 16, 149–70. Duffy, M. K., Ganster, D. C. and Shaw, J. D. (1998) ‘Positive affectivity and negative outcomes: the role of tenure and job satisfaction’, Journal of Applied Psychology, 83(6), 950–9. Ekman, P. (1994) ‘Mood, emotions, and traits’, in P. Ekman and R. J. Davidson (eds), The Nature of Emotion: Fundamental Questions (pp. 56–8). New York: Oxford University Press. Emmons, R. A. and Diener, E. (1986) ‘An interactional approach to the study of personality and emotion’, Journal of Personality, 54(2), 371–84. Emmons, Robert, A., Diener, Ed and Larsen, Randy, J. (1985) ‘Choice of situations and congruence models of interactionism’, Personality and Individual Differences, 6(6), 693–702. Eysenck, H. J. (1991) ‘Dimensions of personality: 16, 5, or 3-criteria for a taxonomic paradigm’, Personality and Individual Differences, 12, 773–90. Fox, S., Spector, F. S. and Miles, D. (2001) ‘Counterproductive work behavior (CWB) in response to job stressors and organizational justice: some mediator and moderator tests for autonomy and emotions’, Journal of Vocational Behavior, 59(3), 291–309. Freud, S. (1930) Civilization and its Discontents. New York: Norton. Funder, D. C. (1997) The Personality Puzzle. New York: W. W. Norton & Co. Geddes, D. H. and Baron, R. A. (1997) ‘Workplace aggression as a consequence of negative performance feedback’, unpublished manuscript, Temple University. George, J. M. (1989) ‘Mood and absence’, Journal of Applied Psychology, 74(2), 317–24. George, J. M. (1991) ‘State or trait: effects of positive mood on prosocial behaviors at work’, Journal of Applied Psychology, 76, 299–307. George, J. M. and Bettenhausen, K. (1990) ‘Understanding prosocial behavior, sales performance, and turnover: a group-level analysis in a service context’, Journal of Applied Psychology, 75(6), 698–709. Goldberg, L. R. (1990) ‘An alternative “description of personality”: the big five factor structure’, Journal of Personality and Social Psychology, 59, 1216–29. Goldberg, L. R. (1993) ‘The structure of phenotypic personality traits’, American Psychologist, 48, 26–34. Goldsmith, H. H. (1994) ‘Parsing the emotional domain from a developmental perspective’, in P. Ekman and R. J. Davidson (eds), The Nature of Emotion: Fundamental Questions (pp. 68–73). New York: Oxford University Press.

Deniz S. Ones and Chockalingam Viswesvaran 245 Gruys, M. L. (1999) ‘The dimensionality of deviant employee performance in the workplace’, unpublished doctoral dissertation, University of Minnesota, Minneapolis, MN. Hakstian, A. R., Farrell, S. and Tweed, R. G. (2002) ‘The assessment of counterproductive tendencies by means of the California Psychological Inventory’, International Journal of Selection and Assessment, 10, 58–86. Harwood, H., Fountain, D. and Livermore, G. (1992) The Economic Costs of Alcohol and Drug Abuse in the United States, 1992. National Institute on Drug Abuse. Hogan, J. and Hogan, R. (1989) ‘How to measure employee reliability’, Journal of Applied Psychology, 74, 273–9. Hogan, R. and Hogan, J. (1992) Hogan Personality Inventory Manual. Tulsa, OK: Hogan Assessment Systems. Hogan, J., Hogan, R. and Busch, C. (1984) ‘How to measure service orientation’, Journal of Applied Psychology, 69, 167–73. Hollinger, R. C. and Clark, J. P. (1983) Theft By Employees. Lexington, MA: DC Heath & Company/Lexington Books. Hough, L. M. (1992) ‘The “Big Five” personality variables – construct confusion: description versus prediction’, Human Performance, 5, 139–56. Hough, L. M. (1998) ‘Personality at work: issues and evidence’, in M. Hakel (ed.), Beyond Multiple Choice: Evaluating Alternatives to Traditional Testing for Selection (pp. 131–59). Hillsdale, NJ: Erlbaum Associates, Inc. Hough, L. M. and Ones, D. S. (2001) ‘The structure, measurement, validity, and use of personality variables in industrial, work, and organizational psychology’, in N. Anderson, D. S. Ones, H. Sinangil and C. Viswesvaran (eds) Handbook of Industrial, Work, and Organizational Psychology: Vol. 1 (pp. 233–77). London: Sage. Hough, L. M. and Oswald, F. (2000) ‘Personnel selection: looking toward the future – remembering the past’, Annual Review of Psychology, 51, 631–64. Hough, L. M., Eaton, N. K., Dunnette, M. D., Kamp, J. D. and McCloy, R. A. (1990) ‘Criterion-related validities of personality constructs and the effect of response distortion on those validities’ (monograph), Journal of Applied Psychology, 75, 581–95. Hunt, S. T. (1996) ‘Generic work behavior: an investigation into the dimensions of entry-level, hourly job performance’, Personnel Psychology, 49(1), 51–84. Hunter, J. E. and Schmidt, F. L. (1990) Methods of Meta-Analysis: Correcting Error and Bias in Research Findings. Newbury Park, CA: Sage. Isen, A. M. (1984) ‘Toward understanding the role of affect in cognition’, in R. S. Wyer and T. K. Srull (eds), Handbook of Social Cognition (pp. 179–236). Hillsdale, NJ: Lawrence Erlbaum. Jacobson, E. (1957) ‘Normal and pathological moods: their nature and functions’, in R. S. Eisler, A. F. Freud, H. Hartman and E. Kris (eds), The Psychoanalytic Study of the Child, Vol. 12 (pp. 73–113). New York: International University Press. Jockin, V., Arvey, R. D. and McGue, M. (2001) ‘Perceived victimization moderates self-reports of workplace aggression and conflict’, Journal of Applied Psychology, 86(6), 1262–9. John, O. P. (1990a) ‘The “Big Five” factor taxonomy: dimensions of personality in the natural language and in questionnaires’, in L. A. Pervin (ed.), Handbook of Personality: Theory and Research. New York: Guilford Press.

246 The Big-5 Personality and Counterproductive Behaviours John, O. P. (1990b) ‘The search for basic dimensions of personality’, in P. McReynods, J. C. Rosen and G. J. Chelune (eds), Advances in Psychological Assessment (pp. 1–37). New York: Plenum Press. John, O. P., Angleitner, A. and Ostendorf, F. (1988) ‘The lexical approach to personality: a historical review of trait taxonomic research’, European Journal of Personality, 2, 171–203. Johnson, E. J. (1988) ‘Expertise and decision under uncertainty: performance and process’, in M. T. Chi, R. Glaser and M. J. Farr (eds), The Nature of Expertise (pp. 209–28). Hillsdale, NJ: Lawrence Erlbaum. Kagan, J. (1994) ‘Distinctions among emotions, moods, and temperamental qualities’, in P. Ekman and R. J. Davidson (eds), The Nature of Emotion: Fundamental Questions (pp. 74–8). New York: Oxford University Press. Kanfer, R. (1990) ‘Motivation theory and industrial and organizational psychology’, in M. D. Dunnette and L. M. Hough (eds), Handbook of Industrial and Organizational Psychology, Vol. 1 (pp. 75–170). Palo Alto: Consulting Psychologists Press, Inc. Kelloway, E. K., Loughlin, C., Barling, J. and Nault, A. (2002) ‘Self-reported counterproductive behaviors and organizational citizenship behaviors: separate but related constructs’, International Journal of Selection and Assessment, 10, 143–51. Kihlstrom, J. F. and Hastie, R. (1997) ‘Mental representations of persons and personality’, in R. Hogan, J. Johnson and S. Briggs (eds), Handbook of Personality Psychology (pp. 712–36). San Diego, CA: Academic Press, Inc. Kohut, H. (1977) The Restoration of the Self. New York: International Universities Press. Koslowsky, M., Sagie, A., Krausz, M. and Singer, A. D. (1997) ‘Correlates of employee lateness: some theoretical considerations’, Journal of Applied Psychology, 82(1), 79–88. Lane, A. M. and Terry, P. C. (2000) ‘The nature of mood: development of a conceptual model with a focus on depression’, Journal of Applied Sport Psychology, 12, 16–33. Lazarus, R. (1994) ‘The stable and unstable in emotion’, in P. Ekman and R. J. Davidson (eds), The nature of Emotion: Fundamental Questions (pp. 79–85). New York: Oxford University Press. Lee, K. and Allen, N. J. (2002) ‘Organizational citizenship behavior deviance: the role of affect and cognitions’, Journal of Applied Psychology, 87(1), 131–42. Lehman, W. E. K. and Simpson, D. D. (1992) ‘Employee substance use and on the job behaviors’, Journal of Applied Psychology, 77, 309–21. Marcus, B., Schuler, H., Quell, P. and Hümpfner, G. (2002) ‘Measuring counterproductivity: development and initial validation of a German self-report’, International Journal of Selection and Assessment, 10, 18–35. Martinko, M. J., Gundlach, M. J. and Douglas, S. C. (2002) ‘Toward an integrative theory of counterproductive workplace behavior: a causal reasoning perspective’, International Journal of Selection and Assessment, 10, 36–50. McCrae, R. R. and Costa, P. T. Jr (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. and Costa, P. T. Jr (1997) ‘Personality trait structure as a human universal’, American Psychologist, 52, 509–16.

Deniz S. Ones and Chockalingam Viswesvaran 247 McDaniel, M. A. and Frei, R. L. (1994) ‘Validity of customer service measures in personnel selection: a meta-analysis’, paper presented at the ninth annual meeting of the Society of Industrial and Organizational Psychology, Nashville, TN. McLellan, R. A. and Paajanen, G. (1994) PDI Customer Service Inventory Manual. Minneapolis, MN: PDI. Megargee, E. I. (1997) ‘Internal inhibitions and controls’, in R. Hogan, J. Johnson and S. Briggs (eds), Handbook of Personality Psychology (pp. 581–614). San Diego, CA: Academic Press, Inc. Miles, D. E., Borman, W. E., Spector, P. E. and Fox, S. (2002) ‘Building an integrative model of extra role work behaviors: a comparison of counterproductive work behavior with organizational citizenship behavior’, International Journal of Selection and Assessment, 10, 51–7. Morris, W. N. (1989) Mood: The Frame of Mind. New York: Springer-Verlag. Norman, W. T. (1963) ‘Personality measurement, faking, and detection: an assessment method for use in personnel selection’, Journal of Applied Psychology, 47, 225–41. Ones, D. S. (1993) ‘The construct validity of integrity tests’, unpublished Doctoral Dissertation, University of Iowa City, Iowa. Ones, D. S. (2002) Introduction to the Special Issue on Counterproductive Behaviors at Work, International Journal of Selection and Assessment, 10, 1–4. Ones, D. S. and Viswesvaran, C. (1996) ‘Bandwidth-fidelity dilemma in personality measurement for personnel selection’, Journal of Organizational Behavior, 17, 209–26. Ones, D. S. and Viswesvaran, C. (1998) ‘Integrity testing in organizations’, in R. W. Griffin, A. O’Leary-Kelly and J. M. Collins (eds), Dysfunctional Behavior in Organizations: Vol. 2. Nonviolent Behaviors in Organizations. Greenwich, CT: JAI press. Ones, D. S. and Viswesvaran, C. (2001a) ‘Integrity tests and other criterionfocused occupational personality scales (COPS) used in personnel selection’, International Journal of Selection and Assessment, 9, 31–9. Ones, D. S. and Viswesvaran, C. (2001b) ‘Personality at work: criterion-focused occupational personality scales (COPS) used in personnel selection’, in B. Roberts and R. T. Hogan (eds), Applied Personality Psychology, Washington, DC: American Psychological Association. Ones, D. S., Viswesvaran, C. and Reiss, A. D. (1996) ‘Role of social desirability in personality testing for personnel selection: the red herring’, Journal of Applied Psychology, 81, 660–79. Ones, D. S., Viswesvaran, C. and Schmidt, F. L. (1993) ‘Comprehensive meta-analysis of integrity test validities: findings and implications for personnel selection and theories of job performance’, Journal of Applied Psychology (Monograph), 78, 679–703. Ones, D. S., Viswesvaran, C. and Schmidt, F. L. (in press) ‘Personality and absenteeism: a meta-analysis of integrity tests’, European Journal of Personality. Organ, D. W. and Ryan, K. (1995) ‘A meta-analytic review of attitudinal and dispositional predictors of organizational citizenship behavior’, Personnel Psychology, 48, 775–802. Ostendorf, F. and Angleitner, A. (1994) ‘The five-factor taxonomy: robust dimensions of personality description’, Psychologica Belgica, 34(4), 175–94.

248 The Big-5 Personality and Counterproductive Behaviours Penney, L. M. and Spector, P. (2002) ‘Narcissism and counterproductive work behavior: do bigger egos mean bigger problems?’, International Journal of Selection and Assessment, 10, 126–34. Robinson, S. L. and Bennett, R. J. (1995) ‘A typology of deviant workplace behaviors: a multidimensional scaling study’, Academy of Management Journal, 38(2), 555–72. Roccas, S., Sagiv, L., Schwartz, S. H. and Knafo, A. (2002) ‘The Big Five personality factors and personal values’, Personality and Social Psychology Bulletin, 28(6), 789–801. Sackett, P. R. (2002) ‘The structure of counterproductive work behaviors: dimensionality and relationships with facets of job performance’, International Journal of Selection and Assessment, 10, 5–11. Sackett, P. R. and DeVore, C. (2001) ‘Counterproductive behaviors at work’, in N. Anderson, D. S. Ones, H. Sinangil and C. Viswesvaran (eds) Handbook of Industrial, Work, and Organizational Psychology, Vol. 1 (pp. 145–64). London: Sage. Sackett, P. R. and Wanek, J. E. (1996) ‘New developments in the use of measures of honesty, integrity, conscientiousness, dependability, trustworthiness, and reliability for personnel selection’, Personnel Psychology, 49, 787–830. Sackett, P. R., Burris, L. R. and Callahan, C. (1989) ‘Integrity testing for personnel selection: an update’, Personnel Psychology, 42, 491–529. Salgado, J. F. (1997) ‘The Five Factor Model of personality and job performance in the European Community’, Journal of Applied Psychology, 82, 30–43. Salgado, J. F. (2002) ‘The Big Five personality dimensions and counterproductive behaviors’, Journal of Selection and Assessment, 10, 117–25. Sanchez, J. I. and Fraser, S. L. (1993) Development and Validation of the Corporate Social Style Inventory: A Measure of Customer Service Skills (Technical Report No. 93–108). Cambridge, Massachusetts: Marketing Science Institute. Saucier, G. and Goldberg, L. R. (1998) ‘What is beyond the Big Five?’, Journal of Personality, 66, 495–524. Saucier, G. and Ostendorf, F. (1999) ‘Hierarchical subcomponents of the Big Five personality factors: a cross-language replication’, Journal of Personality and Social Psychology, 76(4), 613–27. Saxe, R. and Weitz, B. (1982) ‘The SOCO scale measure of the customer orientation of salespeople’, Journal of Marketing Research, 19, 343–51. Schmidt, F. L., Viswesvaran, C. and Ones, D. S. (1997) ‘Validity of integrity tests for predicting drug and alcohol abuse: a meta-analysis’, in W. J. Bukoski (ed.), Meta-Analysis of Drug Abuse Prevention Programs (pp. 69–95). Rockville, MD: NIDA Press. Skarlicki, D. P. and Folger, R. (1997) ‘Retaliation in the workplace: the roles of distributive, procedural, and interactional justice’, Journal of Applied Psychology, 82(3), 434–43. Skinner, B. F. (1971) Beyond Freedom and Dignity. New York: Knopf. Steel, P. and Ones, D. S. (2002) ‘Personality and happiness: a national level analysis’, Journal of Personality and Social Psychology, 83(3), 767–81. Thayer, R. E. (1996) The Origin of Everyday Moods: Managing Energy, Tension, and Stress. New York: Oxford University Press. Vecchio, R. P. (2000) ‘Negative emotion in the workplace: employee jealousy and envy’, International Journal of Stress Management, 7(3), 161–79. Viswesvaran, C. (1993) ‘Modeling job performance: is there a general factor?’, unpublished doctoral dissertation, University of Iowa, Iowa City, IA.

Deniz S. Ones and Chockalingam Viswesvaran 249 Viswesvaran, C. and Ones, D. S. (2000a) ‘Measurement error in “Big Five Factors” of personality assessment: reliability generalization across studies and measures’, Educational and Psychological Measurement, 60, 224–35. Viswesvaran, C. and Ones, D. S. (2000b) ‘Perspectives on models of job performance’, International Journal of Selection and Assessment, 8, 216–27. Viswesvaran, C., Schmidt, F. L. and Ones, D. S. (1996) ‘Modeling job performance: is there a general factor?’, poster presented at the eleventh annual meeting of the Society for Industrial and Organizational Psychology, San Diego, CA. Watson, D. (1988) ‘Intraindividual and interindividual analyses of positive and negative affect: their relation to health complaints, perceived stress, and daily activities’, Journal of Personality and Social Psychology, 54, 1020–30. Watson, D. (2000) Mood and Temperament. New York: Guilford Press. Watson, J. B. (1929) Psychology from the Standpoint of a Behaviorist, 3rd edn. Philadelphia: Lippincott. Wimbush, J. C. and Dalton, D. R. (1997) ‘Base rate for employee theft: convergence of multiple methods’, Journal of Applied Psychology, 82(5), 756–63.

13 Type-A Behaviour Pattern: A Canadian Study Muhammed Jamal and Vishwanath V. Baba*

Introduction Modern society and its constituent organizations, regardless of sector or type, have been plagued with a variety of dysfunctional attitudes and resultant misbehaviour among its members. No organization, economic, political, social, cultural, religious, and academic, is immune to negative attitudes and attendant behavioural malfunctions. The cost of such misbehaviour runs in the billions of dollars annually due to poor performance, loss of productivity, poor physical and mental health, accidents, absenteeism, turnover, theft, etc. Therefore it is both important and timely to address the issue of organizational misbehaviour, its antecedents and its consequences via systematic study and research. This chapter is one such attempt where we look at the impact of personality on dysfunctional attitudes that lead to organizational misbehaviour. More specifically, we are interested in documenting the influence of Type-A behaviour pattern on the onset of poor job satisfaction, low levels of organizational commitment, stress, absenteeism and turnover intention. A recent theoretical work (Baba, 2000) on occupational mental health suggests that personal, role, job, and organizational inadequacies could result in stress and its variants such as poor health, anxiety, depression, and burnout. It further states that when occupational demands are to be met in the absence of adequate resources and in the face of serious

* This study was partly supported by grants from the Social Science and Humanities Research Council of Canada (1999–2002: S00318) and Fonds pour la Formation de Chercheurs de l’Aide à la Recherché (1999–2002: F00269) of the Province of Quebec.

250

Muhammed Jamal and Vishwanath V. Baba 251

negative consequences, distress results. Repeated episodes without any appropriate alleviating mechanism lead to sadness, negative self-assessment, emotional exhaustion, cynicism, and a diminished sense of accomplishment. This results in negative work attitudes, withdrawal, and indifferent individual performance. In turn, it contributes to inefficient, ineffective organizational performance and sub-optimal productivity. This chapter is an empirical investigation of an important personal factor, Type-A behaviour pattern and its influence on stress, negative work attitudes and withdrawal behaviour, as suggested by the theory. In the last three decades, a large number of empirical studies have been conducted by social scientists, health professionals and management researchers in understanding the dynamics of coronary-prone Type-A behaviour. This emphasis has been rightfully attributed to the work of two cardiologists (Friedman and Rosenman, 1974), even though others have pointed out similar behavioural factors in coronary heart disease (CHD). As early as 1868, a German physician, Von Dutch, noted that excessive work involvement and other behavioural characteristics seemed typical of people who develop CHD. Similarly, in 1882, Osler described the coronary-prone person as a ‘keen and ambitious man, the indicator of whose engines are set at full speed’. In the 1940s and 1950s, a number of researchers in health sciences noted some commonalities in CHD patients such as hard-driving, goal-directed, long working hours, and taking fewer vacation periods than other employees (Chesney and Rosenman, 1980; Matthews, 1988). Type-A behaviour has been described as an action emotion complex that can be observed in any person who is aggressively involved in a chronic incessant struggle to achieve more and more in less and less time, and if required to do so, against the opposing efforts of other things and other persons (Friedman and Rosenman, 1974). Some of the overt symptoms of Type-A behaviour are induced explosiveness, accelerated speech, high achievement ambitions, heightened pace of living, polyphasic activities, a tendency to challenge and to compete with others, impatience with slowness, free-floating hostility and the general appearance of tension (Booth-Kewley and Friedman, 1987; Ganster, 1987; Rosenman, 1991; Taylor and Cooper, 1989). Type-B behaviour is generally characterized by opposite characteristics (Matthews, 1988). Many of the above-mentioned characteristics of Type-A and Type-B behaviours have been empirically validated among different samples of people in various occupational and cultural settings (Howton, Lindoerfer and Marriott, 1998; Jamal and Baba, 1991; Jamal, 1999a; Lavanco, 1997; Lee, Jamieson and Earley, 1996; Mudrack, 1999).

252 Type-A Behaviour Pattern: A Canadian Study

There exists a large number of empirical studies, from both laboratory and field settings, relating Type-A and Type-B differences in physiological and health variables, and work-related variables such as job satisfaction, performance, burnout and absenteeism ( Jamal, 1985; Jamal and Baba, 2000). Most of these studies relied almost exclusively on the global measure of Type-A behaviour. Although the Global Type-A behaviour has been fairly consistent in showing a modest relationship with a number of physiological, health and work-related variables, it has recently been suggested that because of the multidimensional nature of the Type-A behaviour pattern, the usage of the component measures of Type-A behaviour might improve the predictive validity of the Type-A behaviour construct (Edwards and Baglioni, 1991). There are only a few studies in existence in which both the global Type-A measure and the component measures of Type-A were employed (Anderson and Meininger, 1993; Barling and Charbonneau, 1992; Edwards and Baglioni, 1991; Jamal and Baba, 2000; Lee, Jamieson and Earley, 1996; Sibilia, Picozzi and Nardi, 1995; Spence, Helmreich and Pred, 1987). Most of these studies used physiological and health variables as dependent variables. A few of these studies used performance measure along with mental and physiological health variables among university student samples (Barling and Charbonneau, 1992). These studies in general suggest that the components of the Type-A behaviour pattern, such as time pressure and hard-driving/competitiveness, may be differentially related to both health and performance despite wide variations in effect size. However, none of these studies used both individual and organizational outcome variables among gainfully full-time employed people in real work settings to test that notion. Our purpose here is to explore and document any differential effect that the components of the Type-A behaviour pattern may have on poor physical and mental health, negative work attitudes, dysfunctional behaviours and behavioural intentions. In addition, the possible effect of gender on the impact of the Type-A behaviour pattern on the outcome variables mentioned above has neither been well understood nor has been well documented. Again, our aim here is to investigate such gender effects as there may be on the relationship between the Type-A behaviour pattern and relevant health, work attitudes, behavioural intentions and behaviour in this context. Given that there is little guidance by way of theory, we are looking at it as a preliminary examination of the phenomenon and are not providing any directional hypotheses. The present study was conducted among employees of two different organizations. Both global Type-A behaviour and the component

Muhammed Jamal and Vishwanath V. Baba 253

measures of Type-A were used. In line with the suggestion of Edwards and Baglioni (1991), the Framingham scale (Haynes, Feinleib and Kannel, 1980) was divided into two components – time pressure and hard-driving/ competitiveness. Both individual and organization outcomes were used as dependent variables. Individual outcomes were operationalized in terms of job stress, psychosomatic health problems and overall job satisfaction. Organizational outcomes were operationalized in terms of organizational commitment, turnover intention and absenteeism. A number of hypotheses were developed in line with the extensive existing literature on Type-A behaviour ( Jamal and Baba, 1991; Jamal, 1999a). Specifically the following hypotheses were tested among hospital employees (N = 175) and telecommunication employees (N = 110) in Canada: Hypothesis 1: Global Type-A behaviour and its components (time pressure and hard-driving/competitiveness) will be positively correlated with job stress and health problems. Hypothesis 2: Global Type-A behaviour and its two components (time pressure and hard-driving/competitiveness) will be negatively correlated with organizational commitment and job satisfaction. Hypothesis 3: Global Type-A behaviour and its two components (time pressure and hard-driving/competitiveness) will be positively correlated with turnover intention and absenteeism.

Method Research setting The present study was conducted in two different settings in a large Canadian city on the east coast. Employees working in a large hospital were invited to participate. They will be referred to as ‘hospital’ employees, henceforth. The other research site was a large telecommunication organization employing over 5,000 employees. A randomly selected number of employees were invited to participate in the study. They will be referred to as ‘telecommunication’ employees, henceforth.

Procedure In both locations, data were collected by means of a structured questionnaire. Respondents were asked to mail back the completed questionnaire in the postage-prepaid envelope directly to the researcher at the university address. In the hospital sample, approximately 340 copies of the questionnaire were distributed among hospital employees.

254 Type-A Behaviour Pattern: A Canadian Study

With one follow-up, 175 (51 per cent) completed questionnaires were returned. In the telecommunication sample, approximately 245 copies of the questionnaire were distributed among a randomly drawn sample of employees. With one follow-up, 110 (45 per cent) completed questionnaires were returned.

Sample characteristics In the hospital sample, the majority of the respondents were female (67 per cent) and had English as their mother-tongue (56 per cent). The average age was 39.3 years, average seniority was 12.1 years and the average education was 16 years. The respondents were quite similar to the non-respondents with regard to a number of background and sociodemographic variables. In the telecommunication sample, the majority of the respondents were male (80 per cent) and were married (56 per cent). The average age was 40 years, average seniority was 17.02 years and the average education was 14 years. No significant differences were found between respondents and non-respondents on a number of background variables such as age, education, experience, marital status and mother-tongue.

Measures In both research settings, the same standardized scales were employed to assess variables in order to make results meaningful. However, data on absenteeism in the telecommunication sample were obtained from company records, whereas, in the hospital sample, absenteeism data were provided by respondents. Type-A behaviour. Type-A behaviour pattern was assessed with the Framingham Scale (Haynes, Feinleib and Kannel, 1980). This scale has 10 items with varied response options. The scale has been used often in social sciences and health sciences to tap Type-A and B behaviours and is validated against coronary heart diseases (Edwards and Baglioni, 1991). A higher score on this scale indicated a Type-A behaviour and a low score indicated a Type-B behaviour. Job stress. Job stress was assessed with the 13-item scale developed by Parker and DeCotiis (1983). It is a Likert-type scale with 1–5 response options, 1 indicating a strong agreement and 5 indicating a strong disagreement with the item. A higher score on the scale indicated a higher degree of job stress. This scale is frequently used to tap overall job stress and has good psychometric properties across different occupational groups and cultures (Jamal, 1999b; Jamal and Badawi, 1993).

Muhammed Jamal and Vishwanath V. Baba 255

Health problems. Health problems were assessed by adopting measures from the Michigan Studies of Workers’ Health (Caplan et al., 1982). Health problems assessed in the present study include headaches, upset stomach, insomnia, gas and bloated feelings, changes in bowel movement, loss of appetite, early morning sickness, dizziness during the day, nervousness or shakiness inside and inability to relax. Each health problem has 1–5 response options, 1 indicating having to face the problem less than once a month and 5 indicating having to face the problem several times a week. Individuals’ responses on various items were combined to create the index of health problems. A higher score on the index indicated a higher degree of health problems. Job satisfaction. Job satisfaction was assessed by using the Hoppock Scale (McNichols, Stahl and Manley, 1978). The scale has four items and each item has well-explained 1–7 response categories. A high score on this scale indicated a higher degree of job satisfaction. Organizational commitment. Organizational commitment was assessed with the 16-item scale developed by Meyer and Allen (1984). It is a Likert-type scale with response options from 1 to 5, indicating a strong agreement to a strong disagreement with each item. A high score on this scale indicated a higher degree of organizational commitment. This scale is regularly used to tap organizational commitment and has good psychometric properties. Turnover intention. Turnover intention was assessed by asking each respondent to give his or her probability of leaving the present employer two years from the day the questionnaire was completed. This measure has been reported as a reasonably valid measure of actual turnover. A higher score on this measure indicated a higher degree of turnover intention. Absenteeism. As mentioned previously, absenteeism was assessed differently in the two research settings. In the hospital sample, respondents were asked to report on how many days they had missed from the job in the last two months. In the telecommunication sample, employees were asked to put their ID number on the questionnaire. Absenteeism data for the previous two months were obtained from the company records for these employees. In both samples, the frequency of absence was measured instead of duration of absence. This is a measure of voluntary absence. A higher score on this measure indicates a higher degree of absenteeism.

Results The means, standard deviations and reliability coefficients of the study variable are presented in Table 13.1. Reliabilities varied from .66 (Type-A

256 Type-A Behaviour Pattern: A Canadian Study Table 13.1 Means, standard deviations and reliability coefficients for hospital and telecommunication employees Variable

1. Type-A behaviour 2. Job stress 3. Health problems 4. Job satisfaction 5. Organizational commitment 6. Turnover motivation 7. Absenteeism

Number of items

Means

SD

10 a (10) 13 (13) 10 (10) 4 (4) 16 (16) 1 (1) 1 (1)

22.6 (24.50) 29.81 (34.03) 17.10 (26.75) 19.12 (17.93) 48.30 (50.72) 2.55 (2.69) 1.22 (1.43)

6.5 (7.30) 9.61 (13.06) 6.31 (18.19) 3.66 ( 4.96) 7.91 (15.16) 1.44 (1.57) 1.41 (1.99)

Alpha reliability .66 (.79) .84 (.86) .93 (.90) .75 (.83) .72 (.79) − (−) − (−)

a Numbers without parentheses are from the hospital sample and numbers within parentheses are from the telecommunication sample.

behaviour) to .93 (health problems) in the hospital sample. In the telecommunication sample, reliabilities varied from .79 (Type-A behaviour) to .90 (health problems). In both samples, reliabilities were judged to be satisfactory for the survey research. Intercorrelations among dependent variables were computed and are presented in Table 13.2. Job stress was moderately correlated with health problems, job satisfaction and turnover intention in both samples. Job satisfaction, organizational commitment and turnover intention were also intercorrelated in both samples. Absenteeism was only marginally correlated with other variables in both samples. Since none of the intercorrelations appeared to be excessively high, they were viewed as fairly independent of each other and were retained for further analysis. Pearson correlations were computed to examine the relationship of Global Type-A behaviour and its two components with individual and organizational outcome variables. These correlations are presented in Table 13.3 for both samples. Global Type-A and the components of time pressure and hard-driving/competitiveness were significantly and positively correlated with job stress and health problems in both samples, thus supporting hypothesis 1. However, the differential effects

Muhammed Jamal and Vishwanath V. Baba 257 Table 13.2

Intercorrelation among dependent variables

Variable 1. Job stress 2. Health problems 3. Job satisfaction 4. Organizational commitment 5. Turnover intention 6. Absenteeism

1

2

3

4

5

6

− (−) .55 a (.48) b −.34 (−.44) −.20 (−.23) .31 (.38) .06 (.19)

− (−) −.32 (−.29) −.09 (−.13) .22 (.25) .07 (.29)

− (−) .53 (.61) −.62 (−.39) −.25 (−.32)

− (−) −.57 (.44) −.28 (−.23)

− (−) .10 (.19)

− (−)

Correlations without parentheses are from the hospital sample: N = 175, r = .16, p < .05, r = .20, P < .01 b Correlations in a parenthesis are from the telecommunication sample: N = 110, r = .18, p < .05, r = .22, p < .01 a

of Type-A components on health problems was noted for the hospital sample only. Time pressure was more strongly correlated with health problems than was the component of hard-driving/competitiveness (t = 3.85, p < .05). Hypothesis 2 postulated inverse relationships between Global Type-A, its two components and job satisfaction and organizational commitment. In both samples, Global Type-A, time pressure and hard-driving/ competitiveness were significantly and negatively correlated with job satisfaction and organizational commitment, thus supporting hypothesis 2. No significant differential effect was found of the Type-A two components on job satisfaction and organizational commitment in either sample. Global Type-A, time pressure and hard-driving/competitiveness were significantly related with turnover intention in both samples. However, Global Type-A and its two components were not significantly correlated with absenteeism in both samples. Thus, hypothesis 3 was only partially supported. No significant differential effect was found of time pressure and hard-driving/competitiveness on turnover intention and absenteeism in either sample. Moderated multiple regression was used to examine the role of gender as a moderator of Type-A behaviour and outcome relationship.

258 Type-A Behaviour Pattern: A Canadian Study Table 13.3 Pearson correlation of Global Type-A behaviour and its components (time pressure and hard driving/competitiveness) with dependent variables for both samples Variable 1. Job stressa 2. Health problems 3. Job satisfaction 4. Organizational commitment 5. Turnover motivation 6. Absenteeism

Global Type A

Time pressure

Hard-driving/ competitiveness

.55** (.49)** .50** (.53)** −.35** (−.29)** −.19*

.57** (.51)** .68** (.61)** −.32** (−.28)** −.25**

.53** (.47)** .33** (.46)** −.37** (−.29)** .15**

(−.29)** .25**

(−.32)** .26**

(−.27)* .25**

(.32)** .01 (.14)

(.35)** .05 (.12)

(.29)** .06 (.15)

a Correlations without a parenthesis are from the hospital sample and correlations in a parenthesis are from the telecommunications sample. * p < .05 ** p < .01

To determine the joint contribution of Type-A and gender on dependent variables, we performed a hierarchical regression analysis in which Type-A behaviour pattern was entered first, followed by gender and then Type-A behaviour pattern × gender. Out of a possible twelve interaction effects in two samples (6 dependent variables × 2 settings), no significant interaction effects were found involving Global Type-A behaviour, gender and outcome variables. Similar results were obtained with the two components of the Type-A behaviour pattern. Thus, in the present study, gender did not appear to moderate the relationship between Type-A and individual and organizational outcomes.

Discussion The results lend credence to the notion that personality has a role to play in the development of negative work attitudes and the expression of dysfunctional organizational behaviour. The findings derived from two occupational groups indicated that the Global Type-A behaviour and the components of time pressure and hard-driving/competitiveness

Muhammed Jamal and Vishwanath V. Baba 259

were significantly related to job stress, health problems, job satisfaction, organizational commitment and turnover intention. However, in both samples, Global Type-A and its components were not related to absenteeism. Limited support for the differential effects of Type-A components on outcomes was found. Before the findings are discussed any further, a note of caution is needed about the perceptual nature of various measures used in the present study. The findings of the adverse effects of Type-A behaviour on employee and organizational well-being are consistent with the bulk of the literature on the topic (Jamal, 1999a; Jamal and Baba, 1991 and 2000; Lavanco, 1997; Rosenman, 1991; Schaubroeck, Ganster and Kemmerer, 1994). For example, in a study of teachers (N = 335) in a developing country, it was found that Type-A behaviour was positively related to burnout and turnover intention and negatively related to work satisfaction (Jamal, 1999a). Similarly, in another study conducted among Canadian teachers (N = 420), it was found that Type-A behaviour was negatively related to work satisfaction (JDI) and positively related to turnover intention (Jamal and Baba, 2000). In another study of Canadian nurses (N = 1148), Type-A behaviour was found to be significantly and positively related to job stress and turnover intention ( Jamal and Baba, 1991). In a recent study of nurses and teachers in Sicily, Global Type-A behaviour was found to be significantly related to burnout and job satisfaction (Lavanco, 1997). Finally, in a longitudinal study conducted in the US among police and fire department employees, it was found that Type-A behaviour predicted cardiovascular disorder over time (Schaubroeck, Ganster and Kemmerer, 1994). As mentioned previously, limited support was found for the differential effects of the component measures of Type-A behaviour on outcome variables used in the present study. There are only a few empirical studies in existence in which both global and component measures of Type-A behaviours have been employed to examine their impact on some individual and organizational outcomes (Barling and Charbonneau, 1992; Edwards and Baglioni, 1991; Jamal and Baba, 2000; Lee, Ashford and Jamieson 1993; Lee, Jamieson and Earley, 1996). Barling and Charbonneau (1992) in a sample of undergraduate students (N = 113) found that achievement striving (AS) predicted GPA and proof-reading performance, and impatience–irritability (II) predicted headaches and sleep habits, but neither respiratory infections nor digestive problems. In another study of undergraduate students, time pressure was found to be related to somatic complaints, anxiety and severe depression, but competitiveness was generally unrelated to such symptoms (Lee, Jamieson and

260 Type-A Behaviour Pattern: A Canadian Study

Earley, 1996). In both of these studies correlations with Global Type-A behaviour with outcome variables were not reported. In a comprehensive study of executives, Edwards and Baglioni (1991) found some support for the differential effects of Type-A behaviour components on mental and physical health. They used ten different measures of mental and physical health and also provided correlations of Global Type-A and its components with the measures of health. Framingham (1980) Global Type-A scale showed significant relationships with five of the ten measures of health. The component of time pressure was also significantly related to five of the ten measures, while the component of hard-driving/ competitiveness was significantly related to only one of the ten measures of health. Finally, in a recent study conducted among Canadian teachers, no support was found for the differential effects of Global Type-A behaviour and the measures of component on burnout, social support, work satisfaction and job performance ( Jamal and Baba, 2000). While the findings did not quite support the hypothesized impact of the Type-A behaviour pattern on absence despite the theoretical suggestion, it can be readily seen that both hospitals and TV stations where the data came from have stringent absence control policies. They cannot afford a high degree of uncertainty in the attendance behaviour of their employees, be they nurses or TV technicians. What these policies do in effect is to constrain variance on absence. It makes managerial sense to institute a policy that allows a certain degree of predictability and therefore control of absence. From a research point of view, this variable manifests only limited variance and therefore empirically less useful in this context. In the present study, differential effects of component measures were found for health problems in the hospital sample only. Generally speaking, the effect of time pressure was stronger than either the global measure of Type-A behaviour pattern or hard-driving/competitiveness for both samples. This is where we have to take the context into account. Both the hospital nursing climate and the TV programme production environment are more sensitive to time urgency rather than competition. In fact, both environments depend more on cooperation among members than on competition. The broad point here is that it is fair to expect differential component effects when the context warrants them. However, the effects were more pronounced in the hospital context than in the communication industry context. They were also stronger for the healthrelated variables than for other attitudinal or behavioural criteria. Thus, at this time, evidence of differential effects of global vs. component measures of Type-A behaviour tends to be somewhat supportive only

Muhammed Jamal and Vishwanath V. Baba 261

with regard to health-related outcomes. More rigorous research with a variety of outcome variables from individual and organizational perspectives has to be conducted in real work organizations before more definitive conclusions could be drawn. However, it is strongly recommended that this type of research is needed to fully comprehend the dynamics of coronary-prone Type-A behaviour pattern in work settings (Rosenman, 1991). In summary, this study examined the relationship of Global Type-A behaviour and its components (time pressure and hard-driving/competitiveness) with individual and organizational outcomes. Data were collected by means of a structured questionnaire from hospital employees (N = 175) and telecommunication employees (N = 110) in a large Canadian city. Global Type-A behaviour and its two components were significantly related to job stress, health problems, job satisfaction, organizational commitment and turnover intention in both samples. Limited support for the differential effects of Type-A component measures on outcomes was found.

References Anderson, J. R. and Meininger, J. C. (1993) ‘Component analysis of the structured interview for assessment of Type-A behavior in employed women’, Journal of Behavioral Medicine, 16, 371–85. Baba, V. V. (2000) ‘Mental health in the work place: implications for management’, The HRM Research Quarterly, 4(2), 1–5 Barling, J. and Charbonneau, D. (1992) ‘Disentangling the relationship between the achievement striving and impatience–irritability dimensions of Type-A behavior, performance and health’, Journal of Organization Behavior, 13, 369–77. Booth-Kewley, S. and Friedman, H. S. (1987) ‘Psychological prediction of heart disease: a quantitative review’, Psychological Bulletin, 101, 343–62. Caplan, R. D., Cobb, S., French, J. R. P., Harrison, P. and Pinneau, S. R. (1982) Job Demands and Worker Health. Washington, DC: Government Printing Press. Chesney, M. A. and Rosenman, R. H. (1980) ‘Type-A behavior in work setting’, in C. L. Cooper and R. Payne (eds), Current Concern in Occupational Stress (pp. 187–212). London: John Wiley. Edwards, J. R. and Baglioni, A. J. (1991) ‘Relationship between Type-A behavior and mental and physical health symptoms: a comparison of global and com-ponents measures’, Journal of Applied Psychology, 75, 440–54. Friedman, M. and Rosenman, R. H. (1974) Type-A Behavior and Your Heart. New York: Knopf Publications. Ganster, R. (1987) ‘Type-A behavior and occupational stress’, in J. M. Ivancevich and D. C. Ganster (eds) Job Stress: From Theory to Suggestion (pp. 61–84). New York: Haworth Press. Haynes, S. G., Feinleib, M. and Kannel, W. B. (1980) ‘The relationship of psychosocial factors to coronary heart disease in the Framingham Study’, American Journal of Epidemiology, 111, 37–58.

262 Type-A Behaviour Pattern: A Canadian Study Howton, M. F., Lindoerfer, J. S. and Marriott, R. C. (1998) ‘Time urgency and imagery in the Type-A behavior pattern’, Perceptual and Motor Skills, 86, 1323–34. Jamal, M. (1985) ‘Type-A behavior and job performance: some suggestive findings’, Behavioral Medicine, 11(2), 60–8. Jamal, M. (1999a) ‘Job stress, Type-A behavior and well-being: a cross-cultural examination’, International Journal of Stress Management, 6, 57–68. Jamal, M. (1999b) ‘Job stress and employee well-being: a cross-cultural empirical study’, Stress Medicine, 15, 153–8. Jamal, M. and Baba, V. V. (1991) ‘Type-A behavior, its prevalence and consequences among woman nurses’, Human Relations, 44, 1213–28. Jamal, M. and Baba, V. V. (2000) ‘Type-A behavior, job performance and well being: an empirical assessment’, Working paper, Faculty of Commerce and Administration, Concordia University, Montreal, Quebec, Canada Jamal, M. and Badawi, J. A. (1993) ‘Job stress among Muslim immigrants in North America: moderating effects of religiosity’, Stress Medicine, 9, 145–51. Lavanco, G. (1997) ‘Burnout syndrome and Type-A behavior in nurses and teachers in Sicily’, Psychological Reports, 81, 523–8. Lee, C., Ashford, S. J. and Jamieson, L. F. (1993) ‘The effects of Type-A behavior dimensions and optimism on coping strategy, health and performance’, Journal of Organizational Behavior, 14, 143–57. Lee, C., Jamieson, L. F. and Earley, P. C. (1996) ‘Beliefs and fear of Type-A behavior’, Journal of Organizational Behavior, 17, 151–77. Matthews, K. A. (1988) ‘Coronary heart disease and Type-A behavior: update on and alternative to the Booth-Kewley and Friedman (1987) quantitative review’, Psychological Bulletin, 104, 373–80. McNichols, C. W., Stahl, J. J. and Manley, T. R. (1978) ‘A validation of Hoppock’s job satisfaction measure’, Academy of Management Journal, 21, 737–42. Meyer, J. P. and Allen, N. J. (1984) ‘Testing the “side bet theory” of organizational commitment: some methodological consideration’, Journal of Applied Psychology, 69, 372–8. Mudrack, P. E. (1999) ‘Time structure and purpose, Type-A behavior and protestant work ethic’, Journal of Organizational Behavior, 20, 145–58. Parker, D. F. and DeCotiis, T. A. (1983) ‘Organizational determinants of job stress’, Organizational Behavior and Human Performance, 32, 160–7. Rosenman, R. H. (1991) ‘Type-A behavior pattern: a personal overview’, in M. J. Strube (ed.), Type-A Behavior (pp. 1–24). Newbury Park, CA: Sage Publications. Schaubroeck, J., Ganster, D. C. and Kemmerer, B. C. (1994) ‘Job complexity, Type-A behavior and cardiovascular disorder: a prospective study’, Academy of Management Journal, 37, 426–39. Sibilia, L., Picozzi, R. P. and Nardi, A. M. (1995) ‘Identifying psychological profile of Type-A behavior pattern’, Stress Medicine, 11, 263–70. Spence, J. T., Helmreich, R. L. and Pred, R. S. (1987) ‘Impatience versus achievement striving in the Type-A pattern: differential effects on student health and academic achievement’, Journal of Applied Psychology, 72, 522–8. Taylor, H. and Cooper, C. L. (1989) ‘The stress-prone personality: a review of the research in the context of occupational stress’, Stress Medicine, 5, 17–27.

Author Index Abakoumkin, G. 93 Abdel Halim, A. A. 176 Abrams, D. 96 Ackerman, P. L. 239 Ackroyd, S. 3, 177, 185, 187 Adams, J. S. 84, 201 Ahmadi, M. 124, 126 Ala-Mursula, L. 127 Albanese, R. 79 Albrecht, S. 17 Alderfer, C. P. 83 Alge, B. J. 165 Allen, N. J. 179, 224 Allen, R. E. 33, 37, 38, 43, 45 Allen, S. 126 Allen, V. L. 173, 177 Almer, E. D. 124, 125 Alsagoff, S. K. 63 Altmaier, E. M. 107 Analoui, F. 177, 178 Anderson, J. C. 110 Anderson, J. R. 252 Anderson, L. R. 92 Andersson, L. M. 14, 15, 22, 26, 200 Anderton, D. L. 47 Angleitner, A. 229 Anthony, W. 124 Aquino, K. 38, 42, 43, 44, 46, 47, 49, 165, 166, 196, 204 Arbetter, L. 16 Arnold, E. S. 142 Arnold, S. E. 142 Arvey, R. D. 22, 23 Ashford, S. J. 176, 259 Ashforth, B. 14, 104, 105, 107, 117, 159, 165 Ashton, M. C. 165 Avgerou, C. 141 Baba, V. V. 250, 251, 252, 253, 259, 260 Bachman, R. 24

Badawi, J. A. 254 Baglioni, A. J. 252, 253, 254, 259 Bailyn, L. 133 Ball, G. A. 165 Baltes, B. 124, 125, 126, 127 Bamberger, P. A. 173 Bandura, A. 105, 118 Barbera, K. 124, 126 Bardoel, E. A. 124 Barling, J. 14, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 201, 252, 259 Barnes-Farrell, J. L. 20 Barnett, R. C. 126 Baron, R. A. 13, 14, 18, 19, 26, 33, 34, 35, 36, 37, 38, 41, 42, 43, 47, 48, 50, 52, 155, 156, 195, 211 Baron, R. M. 19 Barrick, M. R. 188, 229, 231 Bass, B. M. 98 Baumeister, R. F. 205, 240 Beale, D. 21 Beehr, T. A. 196, 198 Bennett, R. J. 3, 4, 14, 20, 34, 35, 154, 155, 156, 157, 158, 162, 165, 173, 174, 175, 181, 185, 195, 196, 212, 213, 224 Bensimon, H. F. 154 Berkowitz, L. 39, 43, 45 Bettencourt, B. A. 38 Bettenhausen, K. 240 Beu, D. 57, 58, 61 Bidermann, Y. 125 Bies, R. J. 154, 166, 195, 201 Biran, A. 181 Bjorkqvist, K. 33, 36, 38, 41, 42, 43, 46, 48 Bjorn, L. 44 Blackburn, R. S. 109 Blanchard, P. N. 92 Block, J. 229 Bobko, P. 176 Boden, J. M. 205 Bolin, A. 241

263

264 Author Index Bond, J. T. 127 Booth-Kewley, S. 251 Bord, R. J. 155, 156 Borgida, E. 41, 44 Borman, W. C. 241 Bornstein, G. 93 Bosma, H. 20 Bounds, G. 64 Bowes-Sperry, L. 34 Bowman, J. S. 16, 25 Bradfield, M. 165, 196 Braverman, M. 14, 21, 27 Breaugh, J. A. 173, 177, 178, 182 Breham, J. W. 173, 177 Brennan, R. T. 126 Brickner, M. A. 81 Brief, A. 105, 106, 107, 161 Brislin, R. W. 182 Bronstein, H. 96 Brookings, J. 105 Brown, L. D. 159 Brumback, G. B. 3 Bruun, S. E. 83, 97 Buckley, M. R. 125 Buckley, R. M. 57, 58, 61 Budd, J. W. 22, 23 bullying 24, 33, 40 Burgess, D. 44 Burke, R. J. 105, 107 Burris, L. R. 229 Buunk, B. P. 91 Cacioppo, J. T. 88 Callahan, C. 229 Cammann, C. 109 Campagna, A. 124, 126 Campbell, J. 84, 213, 217, 239, 240–1 Campion, M. A. 82 Caplan, R. D. 182, 187, 255 Carcione, S. G. 24, 25 Castaneda, M. 125 Cauffman, E. 27 Charbonneau, D. 252, 259, 260 Charbonnier, E. 88 Chen, P. Y. 198, 199, 204 Cherniss, C. 103 Chesney, M. A. 179, 187, 251 Chess, W. A. 103 Christensen, K. 125

Cialdini, R. B. 84 Clark, J. 174, 180, 188 Clark, L. A. 204 Clarkberg, M. 126, 127 Clifton, T. 124 Cobb, S. 182 Cochran, C. C. 48 Cohen, A. 24, 25, 156 Cohen, G. 180 Cohn, E. G. 38 Collins, J. 3, 164, 173 Colquitt, J. A. 19 Compton, D. C. 142 Connolly, J. J. 239 Constable, J. 107 Cook, S. E. 46 Cooper, C. 125, 251 Cooper, C. L. 24 Cordes, C. 104 Cortina, L. M. 200 Costa, P. T. 229, 241 Cox, T. 18, 21 Crane, N. 125 Cropanzano, R. 19 Cruz, M. G. 90 Cullen, M. J. 165 Curtis, T. D. 127 Cyr, D. 142 Dall’Ara, E. 40, 42, 45 Dalton, D. R. 65, 124, 126, 211 Davidson, R. J. 239 Dawson, L. M. 61 Day, R. C. 20 Deckman, M. 125 DeCresce, R. 211 Deery, S. 125 DeFrank, R. S. 59 Dehler, G. E. 155, 158, 159, 163 Dekker, I. 17, 21, 26 DeMore, S. W. 19 Desanctis, G. 138 Deszca, G. 105 DeVore, C. J. 173, 174, 175, 185, 217 Diehl, M. 93 Diener, E. 239 Digman, J. M. 188, 229, 231, 234, 235, 237 Dodge, K. A. 18

Author Index 265 Dolan, N. 107 Donovan, M. A. 19 Doob, A. 156 Dougherty, D. 159 Dougherty, T. 104 Douglas, S. C. 17, 18, 38, 204, 242 Doverspike, D. 39 Drago, R. 133 Drasgow, F. 19, 35 Driscoll, D. M. 41, 42 Dubinsky, A. J. 62 Duffy, M. K. 84 Dunham, R. B. 109, 125 Dupré, K. E. 19, 20, 21, 27 Dwyer, D. J. 173, 177 Eagly, A. H. 38 Earley, P. C. 94, 251, 252 Edwards, J. R. 252, 253, 254, 259, 260 Ehrensaft, M. K. 19 Einarsen, S. 14 Eisenhardt, K. 59 Ekman, P. 239 Elangovan, A. R. 58 Elias, R. 49 Elkins, T. J. 19 Emmons, R. A. 239, 240 Erev, I. 93 Erez, M. 94 Etzion, D. 105 Evans, B. 117 Emmons, R. A. 239, 240 Eysenck, H. J. 229, 231 Ezra, M. 125 Fagan, C. 122 Fain, T. C. 47 Faragher, B. 125 Farrell, D. 180 Feinleib, M. 253, 254 Felson, R. B. 22 Fenwick, R. 125 Ferber, M. 133 Fiesta, J. 25 Filipczak, B. 154 Fischer, D. 117 Fisher, J. D. 19 Fitzgerald, L. F. 35, 36, 44

Flanagan, M. 127 Folger, R. 14, 17, 19, 24, 33, 37, 38, 155, 156, 195, 201, 202, 227 Fountain, D. 211 Fox, J. A. 17, 19 Fox, M. L. 173, 177 Fox, S. 3, 194, 195, 196, 199, 201, 202, 204, 242 Fraser, S. L. 234 Frazier, P. A. 48 Freedman, J. 156 Frei, R. L. 234 Freibach, N. 126, 127 French, J. R. P. 182 Freud, S. 237 Friedman, M. 251 Frink, D. D. 49 Fry, Art 160 Fry, W. R. 201 Fulk, J. 138 Funder, D. C. 238 Gaillard, A. W. K. 91 Galbraith, J. R. 165 Galili, R. 93 Galinsky, E. 127 Galperin, B. L. 160, 162 Ganster, D. C. 176, 179, 259 Ganster, R. 251 Gareis, K. 126 Gauvin, S. 81 Geddes, D. 33, 43, 211 Gelfand, M. J. 35, 36 George, J. M. 90–1, 96, 97, 161, 240 Gerbing, D. W. 110 Gerhart, B. 24 Giacalone, R. A. 3, 154, 174, 195 Gibson, A. M. 58 Gillespie, D. 103, 107, 117 Gillon, D. J. 179 Gilsdorf, J. W. 141 Glass, D. C. 179, 187 Glass, J. 127, 130 Glew, D. J. 20, 33, 155, 195 Gold, R. S. 105 Goldberg, L. R. 229, 231 Golden, L. 123, 127 Goldshmid-Aron, L. 179, 182

266 Author Index Goldsmith, H. H. 239 Golembiewski, R. T. 103, 105, 117 Govoni, S. J. 4, 194 Graham, J. P. 18 Graham, J. W. 159 Green, G. S. 158 Greenberg, D. B. 173, 177 Greenberg, J. 3, 81, 82, 90, 96, 154, 156, 158, 165, 174, 178, 185, 188, 195, 201 Greenberg, L. 14, 17, 18, 19, 20, 22, 26 Greenglass, E. R. 105, 107 Greenwald, J. 127 Gregory, P. R. 59 Griffin, R. W. 3, 20, 33, 34, 155, 173, 174, 176, 185, 195 Grover, S. L. 58 Grover, V. 138, 141 Gruber, J. E. 44 Gruys, M. L. 4, 175, 213 Gundlach, M. J. 242 Gustafson, D. 124 Gutek, B. A. 44 Guynn, J. 33 Guzzo, R. A. 109 Hacker, K. L. 141 Hackman, J. R. 82, 83, 173, 176, 178, 187 Hamblin, R. L. 20 Hamermesh, D. 122 Hammer, L. 124, 126 Hanke, J. J. 158, 159, 166 Harari, O. 82, 96 Harkins, S. 79, 81, 86, 93, 96, 97 Harrick, E. 126 Harris, J. R. 64 Hart, J. W. 92 Harvey, S. 200 Harwood, H. 211 Hastie, R. 241 Hatfield, J. D. 84 Haynes, S. G. 253, 254 Heatherly, L. 241 Heggestad, E. D. 239 Heil, G. 82 Heimann, V. 98 Hein, K. 127

Heller, T. 159 Helmreich, R. L. 252 Henderson, W. L. 41, 42 Henningsen, D. D. 90 Henson, K. D. 47 Hermann, E. 124, 125 Hermodson, A. 57, 58, 61, 64 Herzberg, F. 178 Higgins, C. A. 155 Hill, E. J. 124 Hilles, R. 103 Hirschman, A. O. 180 Hjelt-Back, M. 33, 38, 43 Hodgdon, P. N. 142 Hoeksma-van Orden, C. Y. D. 91 Hoel, H. 24 Hogan, J. 188, 227, 229, 234 Hogan, R. 227, 229, 234 Hogg, M. A. 96 Hohl, L. 123 Hollinger, R. C. 174, 180, 188 Hough, L. M. 229, 231 Howell, J. M. 155 Howton, M. F. 251 Huesman, L. R. 18 Huguet, P. 88, 97 Hui, C. 58 Hunt, S. T. 227 Hunter, J. E. 217 Hunter, S. 200 Hurrell, J. J. 20 Huseman, R. C. 84 Hutnik, H. 180, 188 Hyde, J. S. 38 Ingram, T. N. 62 Isen, A. M. 239 Ivancevich, J. M. 59 Jackson, J. M. 81, 89, 93 Jackson, S. 103, 104, 105, 106, 107, 109, 198 Jacobs, J. 133 Jacobson, E. 239 Jamal, M. 179, 251, 252, 253, 254, 259, 260 Jamieson, L. F. 251, 252, 259 Jayaratne, S. 103 Jenkins, E. L. 14

Author Index 267 Jenkins, P. 27 Jex, S. M. 196, 198 John, O. P. 229 Johnson, E. J. 213 Johnson, N. 124 Johnson, P. 98 Jones, V. A. 140 Jossi, F. 21, 24 Joy, D. S. 17 Jude-Zigmond, C. 16, 25 Judge, T. A. 231 Justice, B. 105 Kafry, D. 103, 105 Kagan, J. 239 Kahill, S. 103 Kakabadse, A. 177, 178 Kakuyama, T. 93 Kalakota, R. 138 Kalliath, T. 103, 107, 109, 116 Kan, P. C. 61, 65 Kanfer, R. 240 Kannel, W. B. 253, 254 Kanter, R. M. 159, 165 Kaplan, S. 124, 125 Karasek, R. A. 176 Karau, S. J. 81, 82, 85, 86, 88, 92, 93, 94, 95, 97 Karuza, J. 201 Kate, Nancy Ten 127 Katzell, R. A. 109 Katzenbach, J. R. 82 Keashly, L. 14, 200 Kelloway, E. K. 14, 23 Kelly, J. R. 41, 42 Kemmerer, B. C. 259 Kerr, N. L. 83, 92, 97 Kesner, I. F. 65 Kettinger, W. J. 138, 141 Kicza, D. C. 125 Kihlstrom, J. F. 241 Kim, J. 124, 126 King, G. 57, 58, 61, 64 Kinicki, A. 188 Kitchell, S. 163 Kjelgaard, M. M. 96 Klein, J. P. 105 Klein, R. L. 178 Klimoski, R. J. 49

Knapp, D. E. 48 Knorz, C. 200 Kochan, T. 133 Kohut, H. 237 Kondrasuk, J. N. 154 Konrad, A. M. 44 Kopelman, R. 124, 127 Koslowsky, M. 227 Kossek, E. 124 Krasner, S. S. 204 Krausz, M. 124, 125, 126, 127 Kreitner, R. 188 Krigel, K. 188 Krueger, R. F. 17 Kruse, D. 124 Kugihara, N. 88, 94 Kulik, C. T. 39 Kulla, M. 200 Kunkel, D. A. 103 Kush, K. 127 Lafontaine, E. 44, 48 Lagerspetz, K. M. J. 33, 36, 38, 41, 42, 46 Lane, A. M. 239 Lankau, M. 124, 125 Latane, B. 79, 80, 81, 82, 84, 91, 97 Latham, G. P. 82 Latham, L. L. 17, 205 Lavanco, G. 251, 259 Lavite, C. M. 50 Lawless, P. 22, 23 Lay, Ken 2 Lazarus, R. 239 Leather, P. 18, 21, 23 LeBlanc, M. M. 14, 23 Lee, C. 176, 179, 187, 251, 252, 259 Lee, K. 165, 224 Lee, R. 104, 105, 107, 117 Lee, T. W. 96 Leiter, M. P. 103, 104, 106, 107, 108, 117 Leonard, B. 124 Leong, G. B. 178 LePine, J. A. 161, 162 Leventhal, G. S. 201 Levin, J. 17, 19 Levine, J. M. 158 Levy, A. C. 49

268 Author Index Lewis, M. U. 165, 196 Leymann, H. 23, 200 L’Herrou, T. 84 Limayem, M. 81 Lindoerfer, J. S. 251 Lion, J. R. 15 Liro, J. R. 125 Livermore, G. 211 Locke, E. A. 82 Lorence, J. 27 Lucas, J. R. 58 Lucero, M. A. 33, 34, 37, 38, 39, 43, 45 Luthans, F. 82 Luthar, H. K. 34, 47 Maass, A. 40, 42, 45 MacCallum, R. 118 MacCoun, R. J. 92 MacEwen, K. E. 25 MacLean, L. M. 200 Mael, F. A. 159 Mahony, A. 125 Maidique, M. A. 159 Malin, M. H. 162 Malone, J. 18 Mangioni, T. W. 188 Mantell, M. 17 Mariotti, J. 163 Marmot, M. G. 20 Marriott, R. C. 251 Mars, G. 177 Martindale, M. 47 Martinko, M. J. 17, 18, 38, 164, 204, 242 Maslach, C. 103, 104, 105, 106, 107, 109, 117 Maslow, A. H. 82, 83 Mathews, R. 160 Matsui, T. 93 Matthews, K. A. 179, 187, 251 Maurer, T. J. 97 McCarthy, D. J. 59 McClelland, C. L. 82 McCrae, R. R. 229, 241 McCrate, E. 127 McDaniel, M. A. 234 McDonald, G. M. 61, 65 McGovern, P. 24

McGuire, J. B. 125 McLellan, R. A. 234 McShulskis, E. 57, 124, 127 Megargee, E. I. 241, 242 Meininger, J. C. 252 Melamed, S. 20 Mello, J. P. 24 Merrill, G. L. 15 Merton, R. K. 155, 157, 166 Mesch, D. 124, 126 Meyer, C. S. 124 Meyer, J. P. 179, 180, 182 Miceli, M. P. 159, 161, 162 Michlitsch, J. 126 Miles, D. 195, 196, 199, 204, 242 Miles, E. W. 84 Miles, J. A. 81, 82, 90, 96 Miller, M. L. 90 Miller, N. 38 Mischel, W. 17 Moag, J. S. 201 Molstad, C. 177 Monteil, J. M. 88 Moore, H. L. 154 Moorman, R. H. 19 Morris, W. N. 239 Mortimer, J. 27 Moss, R. L. 127 Mount, M. K. 188, 229, 231 Mowday, R. 106, 109 Mudrack, P. E. 251 Mulvey, E. P. 27 Munson, L. J. 19 Munzenrider, R. 105, 117 Murphy, C. M. 22, 26 Murphy, J. D. 41 Murphy, K. R. 173 Murphy, L. R. 20 Mustari, E. L. 79 Nardi, A. M. 252 Near, J. P. 159, 161, 162 Neuman, J. H. 13, 14, 18, 19, 26, 33, 36, 38, 41, 42, 47, 48, 50, 52, 155, 156, 195 Newstrom, J. W. 63 Nir-Dotan, M. 20 Nolte, M. L. 25 Norman, W. T. 229

Author Index 269 O’Connor, E. J. 199 O’Donohue, W. 34 O’Driscoll, M. 103, 107, 116 Oestreich, D. K. 59, 77 Oldham, G. R. 82, 83, 173, 176, 178, 187 O’Leary, D. K. 18 O’Leary, K. D. 22, 26 O’Leary-Kelly, A. M. 3, 20, 33, 34, 37, 38, 39, 43, 45, 49, 50, 155, 156, 173, 195 Olson, A. M. 48 Olweus, D. 40, 43, 44, 45 O’Neill, K. 98 Ones, D. S. 165, 188, 203, 211, 213, 217, 220, 224, 227, 229, 231, 233, 234, 235, 237, 239 Onglatco, M. U. 93 Organ, D. W. 161, 241 Orlikowski, W. J. 141 Ostendorf, F. 229, 231 Osterman, K. 33, 36, 38, 41, 42, 43, 46, 48 Ostrom, T. M. 81 Oswald, F. 229 Paajanen, G. 234 Padgett, V. R. 81 Paetzold, R. L. 34 Pal, Lou 2 Paludi, M. A. 49 Parker, C. M. 142 Pastille, C. 34, 47 Pearson, C. M. 14, 15, 22, 26, 200 Penney, L. M. 194, 196, 199, 200, 204, 205 Pepinsky, P. 158, 159 Perlow, R. 17, 205 Perry, E. L. 39 Perry, P. M. 25 Peters, L. H. 199 Petty, R. E. 81, 86, 88, 96, 97 Phillips, J. S. 19 Phillips, M. 201 Picozzi, R. P. 252 Pierce, H. R. 97 Pierce, J. 125 Pines, A. 103, 105, 176

Porath, C. L. 200 Porter, L. 106, 109 Pred, R. S. 252 Presser, H. 122 Provan, K. 124 Pryor, J. B. 40, 41, 51 Puffer, S. M. 59 Quinn, R. P.

109, 188

Raelin, J. A. 180 Ralston, D. 124, 127 Randall, D. M. 58 Ray, D. 96 Rayner, C. 24 Reif, W. E. 82 Reilly, N. 107, 116 Reiss, A. D. 229 Rentsch, J. R. 177 Richer, S. F. 20 Richman, J. A. 33 Rick, D. 103 Riggs, D. S. 18 Riordan, C. M. 39 Robb, L. A. 39 Roberts, N. C. 98 Robinson, S. L. 3, 4, 14, 33, 34, 35, 38, 50, 154, 155, 156, 157, 158, 162, 173, 174, 175, 178, 181, 185, 195, 196, 212, 213, 224 Roccas, S. 242 Rogers, A. G. 23 Rogers, J. K. 47 Rogers, K. A. 23 Rosenman, R. H. 179, 187, 251, 259, 261 Rospenda, K. M. 52 Ross, R. R. 107 Rotter, J. B. 205 Rotton, J. 38 Roy, M. C. 81 Ruch, W. A. 63 Rudman, L. A. 41 Russell, D. 107 Ryan, K. D. 59, 77, 241 Sackett, P. R. 173, 174, 175, 185, 212, 217, 229, 233 Sager, J. K. 24

270 Author Index Sagie, A. 125 Salgado, J. F. 203, 229, 230 Sanchez, J. I. 234 Saucier, G. 229, 231 Sauter, S. L. 20 Saxberg, B. O. 158, 159, 166 Saxe, R. 234 Sbraga, T. P. 34 Scandura, T. 124, 125 Schat, A. C. 23 Schaubroeck, J. 259 Schein, E. H. 158 Schmidt, F. L. 164, 203, 217, 220, 224, 227, 229 Schmidtke, J. M. 39 Schuler, R. S. 198 Schwartz, D. G. 141 Scott, K. S. 158 Segev, A. 140, 141 Serwinek, P. J. 62 Shapiro, D. L. 58 Sharpe, D. 127 Shaw, J. D. 84 Shea, C. M. 155 Shearer, J. 105 Shephard, E. 124 Sheppard, B. H. 89, 97 Shepperd, J. A. 96 Shin, K. 165 Shirom, A. 20 Sibilia, L. 252 Silva, J. A. 178 Silver, Spencer 160 Sims, H. P. 165 Skarlicki, D. P. 14, 17, 19, 24, 38, 155, 156, 195, 201, 202, 203, 204, 227 Skilling, Jeff 2 Skinner, B. F. 237 Skogstad, A. 14 Slora, K. 17 Smart, L. 205 Smith, B. N. 89, 97 Smith, C. A. 179 Smith, D. 177, 178 Smith, D. K. 82 Smith, F. J. 109 Snyder, W. 15 Solomon, E. P. 204 Somech, A. 94

Sonnenstuhl, W. J. 173 Sorensen, J. E. 188 Sorensen, T. L. 188 Sorrentino, R. M. 89, 97 Sparks, K. 125 Spector, F. S. 242 Spector, P. E. 3, 19, 20, 176, 180, 187, 194, 195, 196, 198, 199, 202, 204, 205 Spence, J. T. 252 Spielberger, C. D. 18, 204 Staines, G. L. 109 Staines, L. G. 125 Stansfield, S. A. 20 Stanton, J. M. 20 Steadman, H. J. 22 Steel, P. 239 Steel, R. P. 177 Steers, R. 106, 109 Steffen, J. 38 Steiner, I. D. 79, 80 Stephens, D. C. 82 Stephenson, R. G. 158 Storms, P. L. 19, 204, 205 Stroebe, W. 93 Stroh, L. 127 Stuart, P. 18 Sullaway, M. E. 92 Swatman, P. M. C. 142 Sydeman, S. J. 204 Tajfel, H. 44 Takemoto-Chock, M. K. 229 Tausig, M. 125 Taylor, H. 251 Te’eni, D. 141 Tepper, B. J. 20 Terpstra, D. E. 46 Terris, W. 17 Terry, P. C. 239 Tesluk, P. 17, 19, 24, 195, 201 Tesser, A. 84 Thayer, R. E. 240 Theorell, T. 176 Thibaut, J. 201 Thomas, K. M. 39 Thomas, K. W. 96 Thompson, D. E. 109 Thompson, P. 3, 177, 185, 187

Author Index 271 Thompson, S. C. 176 Tjosvold, D. 96 Townsend, J. 19 Tredeau, L. 44 Trevino, L. K. 58, 165, 178 Tripp, T. M. 154, 166, 195 Trott, V. 200 Tsui, A. S. 44 Turner, D. L. 158 Turner, J. 105, 106, 107 Turner, J. C. 44 Tyree, A. 18 Vallerand, R. J. 20 Van de Ven, A. H. 165 Van Dyne, L. 161, 162 Van Fleet, D. D. 79 Vanek, G. 126 Vardi, Y. 3, 154, 156, 175, 176, 178, 180, 185 Vaughan, D. 185 Vaught, C. 177, 178 Vecchio, R. P. 240 Viswesvaran, C. 203, 213, 217, 220, 224, 227, 229, 231, 233, 234, 235, 239 Vroom, V. H. 85 Wacker, W. 160 Waldfogel, J. 133 Walker, L. 201 Walton, M. 96 Wanek, J. E. 233

Wang, H. 154 Watson, D. 204, 237, 239 Wayne, J. H. 39 Weick, K. E. 159 Weinstein, D. 162 Weitz, B. 234 Weldon, E. 79 Welsh, M. A. 155, 158, 159, 163 Welsh, S. 34 West, M. A. 162 Whalen, N. J. 41 Whinston, A. 138 White, P. H. 96 Whittington, R. 23 Wiener, Y. 3, 154, 156, 175, 176, 178, 180, 185, 188 Williams, K. 79, 81, 82, 85, 86, 88, 89, 93, 94, 95, 97 Wimbush, J. C. 211 Winter, S. 44 Wolpin, J. 107 Wortman, C. B. 173, 177 Wright, R. A. 96 Wykes, T. 23 Yates, J. 141 Yosifon, H. 180 Youngblood, S. A.

58

Zabid, A. R. M. 63 Zaccaro, S. J. 90, 96, 97 Zapf, D. 200 Zellars, K. L. 17, 164

Subject Index absenteeism 6, 158, 217, 220, 231, 235, 255 aggression and violence at work and 24 flexible work schedules and 126–7, 134 accidents 231 accomplishment, personal, job burnout and 105 Adelphia 3 affective traits 203–4, 239–40 age absenteeism and 126 covering-up behaviour and 61–2, 71, 74–5 flexible work schedules and 127 aggression and violence at work 4–5, 9, 13–27, 33, 156, 158, 220 costs 26 definition and forms of 14–15 escalation of 21–2, 25–6 future research 25–7 incidence of 15–16, 26 outcomes of 22–5; individual 22–3; organizational 24–5; societal 25 policies against 26 predictors of 16–21, 234, 235; individual factors 18–19; workplace factors 19–21 transference 26 young workers 27 see also sexual harassment agreeableness 231 alcohol 18, 211, 220, 233–4, 235 Andersen 2 anger, aggression and violence at work and 18, 19 approval orientation 89 autonomy 165 control and 176–7 misbehaviour and 177–8, 184–5, 186–7

Big Five model of personality 203, 211–43 relationship among different forms of counterproductive work behaviours 212–38; criterion-focused scales and CWB 232–8; personality traits and CWB 228–9, 238–42; single acts of counterproductivity 214–17 burnout see job burnout Canada aggression and violence at work in 16 Type-A behaviour pattern study 252–3; methodology 253–5; results 255–8 challenge constructive deviance and 161 social loafing and 89–90 China, social loafing in 94 citizenship, organizational 161 cognition 241–2 social loafing and 88–9 collective effort model (CEM) 85–6, 88–95 collectivism, social loafing and 94 commitment 255 job burnout and 106–7, 116–17 professional identity and 179–80, 188 communication social loafing and 92–3 see also email mailing lists conflict conflict management 159 stress and 199–200 conformity to norms 84, 157, 163 selective 159 conscientiousness 231 control aggression and violence at work and 19–20, 27

272

Subject Index 273 autonomy and 176–7 deviance and 165, 166 personality as locus of control 204–5 stress and 202 supervision of employees 19–20, 27 counterproductive work behaviour (CWB) see misbehaviour in organizations covering-up behaviour 5, 57–77 firm/organizational characteristics and 64–6, 67, 72–3, 76 job/occupational characteristics and 62–4, 66, 72, 76 methodology of study 66–8 multi-dimensional model of 59–60 personal characteristics and 61–2, 66, 70–2, 74–6 research hypotheses 61–6 results of study 68–74 culture covering-up behaviour and 59, 76–7 organizational 95–6, 141 social loafing and 94 customers/clients aggression and violence by 15 customer service inventories 234 cutting corners 57 depersonalization, job burnout and 104, 117–18 design, job design theories 176–7, 187 deviance in workplace 7, 154–66, 174 constructive 155, 156, 157, 158–66; difference from other constructs 161–3; forms of 160–1 definition 155–7 destructive 154, 156, 157–8, 164 devil’s advocacy 159 drugs 211, 220, 233–4, 235 Edgewater Technology Inc. 13 education levels, covering-up behaviour and 62, 72 email mailing lists 7, 138–52 problems with 143–50; lessons learned from 150–2; methodology of study 142–3

research on 140–2 types of 139–40 emotion 239–40 emotional exhaustion, job burnout and 104, 116, 117, 118 Enron 1–2 envy 84 expectancy theory 85 export sector covering-up behaviour in 74 export sector, covering-up behaviour in 64–5, 72 fairness 84 fatigue, social loafing and 91 fear of aggression 23 covering-up behaviour and 59, 77 flexible work schedules 6–7, 122–34 absence/lateness and 126–7 prevalence of 127–33 productivity and 123–5 well-being, morale and satisfaction and 125–6 fraud 2 free-riding 79 frustration, aggression and violence at work and 19, 38 gender absenteeism and 126 aggression and violence at work and 38 covering-up behaviour and 61, 70, 74, 75–6 flexible work schedules and 127 social loafing and 88, 92 Type-A behaviour and 252 see also sexual harassment goal-setting theory 82 gossiping 158 groups, productivity problems in see social loafing growth of firms, covering-up behaviour and 66, 73, 74, 76 health problems 255 hierarchical relationships aggression and violence at work and 20

274 Subject Index hierarchical relationships (continued) constructive deviance and 166 covering-up behaviour and 62–3, 72 sexual harassment and 43–4, 45–7 supervisor support and job burnout 107–8 homicide 15 incidence of 15–16 individualism, social loafing and 94 information technology 97 see also email mailing lists innovation, constructive deviance and 155, 157, 159, 161, 162, 163, 165 integrity tests 233, 235 intellective tasks, social loafing and 90 International Labour Organization (ILO) 16 International Society for the Study of Work and Organizational Values 1 Israel covering-up behaviour in 66–77 email mailing list problems in 139, 142–52 social loafing in 94 study of organizational misbehaviour in 181–9 Japan, social loafing in 88, 94 job burnout 6, 103–19 job satisfaction and 105–6 meaning of 104–5 method of study 108–10 organizational commitment and 106–7 results of study 110–15 supervisor support and 107–8 job characteristics theory 83 job design theories 176–7, 187 job satisfaction 255 dissatisfaction 180–1, 184, 188, 198 flexible work schedules and 125–6 job burnout and 105–6 judgmental tasks, social loafing and 90 justice aggression and violence at work and 19, 27, 37–8 stress and 200–2

knowledge

239

lateness 6, 158 flexible work schedules and 126–7 length of service, covering-up behaviour and 64, 72, 76 mailing lists see email mailing lists manufacturing sector, covering-up behaviour in 64, 72 marital status covering-up behaviour and 62, 71 flexible work schedules and 127 sexual harassment and 47–8 marketing 142 misbehaviour in organizations 1–4, 194–7, 211–43 antecedents 175–81; individual and personal differences 178–81; job characteristics 176–8 definitions 3–4, 155–7, 174–5 deviance in workplace 154–66, 174; constructive 155, 156, 157, 158–66; destructive 154, 156, 157–8, 164 manifestations 174–5 personal and positional antecedents 173–90 prevalence 4 relationships among forms of 212–13; covariance 214–38 stress and 196–7 study of 181–5 see also individual topics Mitsubishi 36, 38 mood 239–40 motivation 240–1 social loafing and 80–1, 83 team building and 82 Motorola 163 narcissism 205 needs hierarchy 82, 83 negative affectivity 204 non-traditional employment contracts, sexual harassment and 47 norms conformity to 84, 157, 159, 163 violation of 156

Subject Index 275 Northwestern National Life Insurance Co. 16 occupational personality scales 228 criterion-focused scales (COPS) 232–8 Ottawa Carleton Transpo 13, 22 part-time workers covering-up behaviour and 63, 72 flexible work schedules and 130, 133 payment systems 165 performance aggression and violence at work and 24 monitoring of 20 personal accomplishment, job burnout and 105 personality affective traits 203–4 Big Five model 203, 211–43; relationship among different forms of counterproductive work behaviours 212–38 deviance and 165 locus of control 204–5 narcissism 205 personality traits and counterproductive behaviour 227–42; criterion-focused occupational personality scales 232–8; normal adult personality scales 229–32; potential pathways to counterproductivity 238–42 Type-A 179, 184, 187, 250–61; methodology of study 253–5; results 255–8 Polaroid 21 policies, organizational against aggression and violence at work 26 against sexual harassment 49–50 Post-it notes 160 powerlessness 165 productivity aggression and violence at work and 24

flexible work schedules and 123–5 problems in teams see social loafing professional identity, commitment and 179–80, 188 punishment, social loafing and 90–1 quality control 64, 65 cutting corners on 57 rebellion 157 rejection threatened people 89 reporting of sexual harassment 48–9 responsibility, diffusion of 84 retreatism 157 reward, social loafing and 90–1 ritualism 157 role, stress and 197–8 Russia, covering-up behaviour in

59

sabotage 158 safety of employees 24–5 sanctions against employees 20–1 self-actualization 83 self-interest, covering-up behaviour and 58–9 self-monitoring, sexual harassment and 42 service sector, covering-up behaviour in 64, 72 sexual harassment 2, 5, 33–52 aggression and: harassment as form of aggression 34–7; similarity to other forms of aggression 34–7, 51 contextual factors 49–50 perpetrators 38–43 sanctions against 21 victims: provocative 43–5; submissive 45–9 situational constraints, stress and 199 size of firms, covering-up behaviour and 65, 72, 76 skill 239 social evaluation theory 84–5 social identity theory 44–5 social loafing 5–6, 79–98 background 80–1

276 Subject Index social loafing (continued) correlations with 86–95; group characteristics 91–4; individual level 88–9; individualism/ collectivism 94; situational characteristics 89–91 future research 97–8 ideas and tactics for reduction of 95–7 reasons for 83–6; collective effort model (CEM) 85–6, 88–95; search for meaning 83; social evaluation 84–5 scope of 81–2 teams and 82–3 spouses of employees, aggression and violence by 15 standardization, covering-up behaviour and 65, 72, 76 stress 254 control and 202 interpersonal conflict and 199–200 misbehaviour in organizations 196–7 organizational justice and 200–2 role stressors 197–8 situational constraints and 199 tolerance scales 234 workload and 198–9 supervision of employees 19–20, 27 supervisor support and job burnout 107–8 surveillance 20 tardiness see lateness task reinforcement, social loafing and 90–1 teams 82 productivity problems in see social loafing tenure, covering-up behaviour and 63–4, 72, 74, 76 theft 3, 4, 158, 200, 224, 235 Total Quality Management (TQM) 96 trade unions, flexible work schedules and 130 training aggression and violence at work and 21, 39 in email use 141, 151

turnover of employees 255 aggression and violence at work and 24 flexible work schedules and 124 Type-A personality 179, 184, 187, 250–61 methodology of study 253–5 results of study 255–8 unfair treatment, aggression and violence at work and 19 unions, flexible work schedules and 130 uniqueness 88 United States of America absenteeism in 126 aggression and violence at work in 4, 15–16, 24, 25 covering-up behaviour in 59 deviance in workplace in 154 flexible work schedules in 122, 127, 133 job burnout in 108–18 misbehaviour in organizations in 194 sexual harassment in 35, 36, 42 social loafing in 94 substance abuse in 211 theft in 4 United States Postal Services 21 University of Arkansas 13 violence see aggression and violence at work virtual teams 98 voice 162, 180 volition 240–1 whistle-blowing 159, 161–2 working hours see flexible work schedules workload, stress and 198–9 WorldCom 3 young people aggression and violence at work and 27 covering-up behaviour and 61–2, 71