Gender, Equality and Education from International and Comparative Perspectives (International Perspectives on Education and Society, Vol. 10)

LIST OF CONTRIBUTORS Sonya Anderson First Five Years Fund, Chicago, IL, USA Rima Aranha Department of Educational Lea...

Author: David Baker | Alexander Wiseman

13 downloads 1493 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

LIST OF CONTRIBUTORS Sonya Anderson

First Five Years Fund, Chicago, IL, USA

Rima Aranha

Department of Educational Leadership and Policy, State University of New York, Buffalo, NY, USA

Monisha Bajaj

Department of International and Transcultural Studies, Teachers College, Columbia University, NY, USA

David P. Baker

Educational Theory and Policy Department, Pennsylvania State University, PA, USA

Carolyn Barber

Division of Counseling and Educational Psychology, University of Missouri, Kansas City, MO, USA

Elizabeth A. Constantine College of Education, The University of Iowa, Iowa City, IA, USA Joan DeJaeghere

Department of Educational Policy and Administration, University of Minnesota, MN, USA

Alan J. DeYoung

College of Education, University of Kentucky, KY, USA

Rachel Fix Dominguez

Department of Educational Leadership and Policy, State University of New York, Buffalo, NY, USA

Parfait M. Eloundou-Enyegue

Development Sociology Department, Cornell University, NY, USA ix

x

LIST OF CONTRIBUTORS

Sarah C. Giroux

Development Sociology Department, Cornell University, NY, USA

Lihong Huang

NOVA – Norwegian Social Research, Oslo, Norway

Alice Kagoda

School of Education, Makerere University, Uganda

Julia Kaufman

School of Education, University of Pittsburgh, PA, USA

Daniel Kirk

American University of Sharjah, United Arab Emirates

Fouad Makki

Development Sociology Department, Cornell University, NY, USA

Shirley J. Miske

Miske Witt & Associates, Inc., MN, USA

Yuri Nakajima

Department of Educational Leadership and Policy, State University of New York, Buffalo, NY, USA

Diane Napier

College of Education, University of Georgia, GA, USA

Yoshiko Nozaki

Department of Educational Leadership and Policy, State University of New York, Buffalo, NY, USA

Francisco O. Ramirez

School of Education, Stanford University, CA, USA

Catherine Riegle-Crumb

The University of Texas, Population Research Center, TX, USA

Jill Sperandio

College of Education, Lehigh University, PA, USA

Judith Torney-Purta

Department of Human Development, University of Maryland, MD, USA

xi

List of Contributors

Alexander W. Wiseman

College of Education, Lehigh University, PA, USA

Liqun Yin

School of Education, University of Pittsburgh, PA, USA

INTERNATIONAL PERSPECTIVES ON EDUCATION AND SOCIETY Series Editor: Abraham Yogev Volume 1:

International Perspectives on Education and Society

Volume 2:

Schooling and Status Attainment: Social Origins and Institutional Determinants

Volume 3:

Education and Social Change

Volume 4:

Educational Reform in International Perspective

Series Editor from Volume 5: David P. Baker Volume 5:

New Paradigms and Recurring Paradoxes in Education for Citizenship: An International Comparison

Volume 6:

Global Trends in Educational Policy

Volume 7:

The Impact of Comparative Education Research on Institutional Theory

Volume 8:

Education for All: Global Promises, National Challenges

Volume 9:

The Worldwide Transformation of Higher Education

INTERNATIONAL PERSPECTIVES ON EDUCATION AND SOCIETY VOLUME 10

GENDER, EQUALITY AND EDUCATION FROM INTERNATIONAL AND COMPARATIVE PERSPECTIVES EDITED BY

DAVID P. BAKER Pennsylvania State University

ALEXANDER W. WISEMAN Lehigh University

United Kingdom – North America – Japan India – Malaysia – China

JAI Press is an imprint of Emerald Group Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2009 Copyright r 2009 Emerald Group Publishing Limited Reprints and permission service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. No responsibility is accepted for the accuracy of information contained in the text, illustrations or advertisements. The opinions expressed in these chapters are not necessarily those of the Editor or the publisher. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-84855-094-0 ISSN: 1479-3679 (Series)

Awarded in recognition of Emerald’s production department’s adherence to quality systems and processes when preparing scholarly journals for print

PREFACE Much has been written about the global progress made toward gender parity in enrollment and curriculum in nations around the world. And there is much to tout in these areas. Although gender parity is not yet the global norm, the expectation of gender equality increasingly is. Some have gone so far as to say that the global expansion of modern mass schooling has created a world culture of gender equality in education. Yet, while there have been many positive advances regarding girls’ and women’s education around the world, there are still signiﬁcant differences that are institutionalized in the policies and administrative structures of national education systems. For example, some of the strongest evidence of gendered inequality in schooling is the fact that in many developing countries there are large proportions of school-age children who are not in school – many if not most of whom are girls. So, the question remains whether gender equality in education is really being achieved in schools around the world. This volume of International Perspectives on Education and Society investigates the often controversial relationship among gender, equality and education from international and comparative perspectives. Some of the most interesting recent comparative work on gender inequality in schools focuses on how gender intersects with ‘real world’ inﬂuences on equality like race, ethnicity, and class. Although the differences in schooling between boys and girls exist worldwide, evidence also suggests that girls’ and women’s education is strongly contextualized by the political, social, and economic environments of both local schools and national educational systems. Since the World Conference on Education for All (EFA) in Jomtien, Thailand in 1990, the push for participation in modern mass schooling has become a primary focus of national education policymakers and researchers around the world. This push has made the international community painfully aware of widespread and speciﬁc gendered inequality in schooling around the world as well as the beneﬁts to individuals and communities that result from educating girls and women that many nations and communities were missing because of these gendered inequalities. Awareness of this is evident in the fact that half of the EFA goals speciﬁcally mention the global importance of improving and equalizing education between boys and girls. In fact, in several xiii

xiv

PREFACE

countries, the lingering gender differences in academic course participation and achievement in secondary school is tied more to community attitudes and labor market potential than to actual opportunities to learn in schools. At the same time, there has been signiﬁcant progress in eliminating gender inequality in education. In many nations, formal differentiation in schools by gender has largely shifted to differentiation by academic achievement rather than by gender in modern mass schooling systems. For example, the ofﬁcial criterion for promotion into advanced academic courses is not a student’s gender anymore in most schools around the world. Instead, in most cases, the ofﬁcial criterion for advanced academic course enrollment is now either demonstrated or anticipated academic ability instead of gender and other human or social characteristics. Mass schooling has played a large part in making this happen. Once the entire school-age population was both given the opportunity to enroll in and (in many nations) compelled to attend school, traditional gendered differentiation was no longer a legitimate or effective formal educational policy. There is substantial evidence that the global expansion and legitimization of modern mass schooling has hastened a shift in world cultural norms and values from one of differentiation to one of egalitarianism. As a result, the institutional structure of schools shapes and is shaped by world culture. In the 21st century, the concept of gender equality pervades world culture in every nation around the world – even in nations where gender equality has not been and may never be fully achieved. This is the international environment in which the contributors to this volume of the International Perspectives on Education and Society series address issues of gender and both inequality and efforts to achieve equality in education. The chapters in this volume address a variety of national contexts ranging from the small African country of Benin to Uganda, Zambia, Vietnam, China, Japan, Mongolia, India, United Arab Emirates, and Norway. Two chapters address regional issues related to gender and education in SubSaharan Africa and Central Asia. And, two other chapters speciﬁcally use cross-national data from the International Association for the Evaluation of Educational Achievement (IEA) studies on civics as well as math and science achievement (CivEd and TIMSS, respectively). These two chapters address global trends in gender differences or gender effects in education. Although some of these chapters deal speciﬁcally with issues of gender and educational achievement, others address the signiﬁcance of gender in development aid and policymaking in speciﬁc countries or regions of the world. Still others address the importance of gender as it overlaps with ethnicity, political ideology, labor market participation, and societal trends.

xv

Preface

Some of the chapters also deal with the impact that speciﬁc reforms or adjustments to the structure and opportunity of schooling have on students by gender, speciﬁcally the impact of single-sex schooling, participation in higher education, and universal access to education. The overall ﬁndings of the chapters in this volume suggest that the phenomenon of the spread of gender egalitarianism as a world cultural norm in education and society is supported with empirical evidence, but that this ‘‘advance’’ in favor of gender equality is still strongly tempered by very real and difﬁcult to overcome divisions by gender in schools and educational systems around the world.

NOTE FROM SERIES SENIOR EDITOR With this Volume 10, I end my tenure as senior editor of the series International Perspectives on Education and Society. When I was asked in 2001 to revive the moribund series there had been only four volumes published over the preceding decade. The vision proposed then was to create and maintain a series that would represent an annual review of examples of the best scholarship on selected topics on education from an international perspective. With the intellectual support of the expanding and dynamic community of scholars of comparative education throughout the world this vision has become a reality in the six annual volumes published since 2002. To the publishers and contributing authors, as well as to the scores of reviewers of proposed papers who generously donated their time and expertise in selecting and strengthening the ﬁnal chapters, I want to extend my appreciation – it has been an honor to work with all of you. It is a great pleasure to announce that the series’ current publishing house, the Emerald Group Publishing Limited (U.K.) has selected Professor Alexander Wiseman of Lehigh University (U.S.) to serve as the next senior editor of the series. Since Volume 5, Alexander Wiseman has been the collaborating editor of the series and was instrumental in its revival. His considerable editorial experience and broad scholarly interests in the comparative study of education make him an excellent choice to steward the series into the near future. David P. Baker Alexander W. Wiseman Editors

SEX VERSUS SES: A DECLINING SIGNIFICANCE OF GENDER FOR SCHOOLING IN SUB-SAHARAN AFRICA?$ Parfait M. Eloundou-Enyegue, Fouad Makki and Sarah C. Giroux ABSTRACT Recent worldwide gains in girls’ schooling are raising new questions about the continued relevance of gender for educational inequality. At issue is whether the time has come to shift the policy focus away from gender to socioeconomic status. Answers to this question, we suggest, depend on how gender gaps close, i.e., do they close irreversibly, evenly, and faster than socio-economic (SES)-related inequality? Against this background and building on contrasted sociological perspectives on inequality, our chapter examines the recent convergence trajectories of several sub-Saharan countries, asking if these trajectories warrant a policy shift away from gender. Our ﬁndings are mixed. Although, the magnitude of sex-related inequality in schooling is consistently smaller than SES-related $

The analyses in this chapter are based on data from Demographic and Health Surveys.

Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 1–37 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010004

1

2

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

inequality, the process of gender convergence remains reversible and it unfolds in top-down fashion. Such ﬁndings warrant continued attention to gender in sub-Saharan Africa, but with particular focus on poor girls and on synergies that address both female and poor children. This conclusion supports theoretical advances that transcend the Manichean divide between focus on cultural recognition and socioeconomic redistribution.

1. INTRODUCTION There is much consensus that developing countries are achieving some gender convergence in schooling. In the last two decades, the female-to-male ratio in secondary enrollment rose from 0.86 to 0.92 for the developing world and despite considerable variation across regions, this global convergence is widely acknowledged (UNICEF, 2004; UNFPA, 2005).1 In contrast, the implications of this convergence are in dispute. One issue is over socioeconomic dividends, i.e., whether gains in women’s schooling translate into parallel gains in employment or social status (Malhotra, Pande, & Grown, 2003). A second, more fundamental, issue is the continued relevance of gender itself in global education policy: Given the worldwide gains in women’s schooling, is a gender focus in education policy still warranted or has the time come to declare victory in the battle for gender equity in schooling? To some, the continued promotion of girls’ schooling is increasingly anachronistic and ‘‘missing the mark,’’ at a time when gender gaps are narrowing but socioeconomic gaps are growing. Perhaps, as Knodel and Jones (1996, p. 684) suggested ‘‘a strong policy emphasis on closing the [gender] gap is no longer needed’’ and emphasis should shift to the socioeconomic gap. Yet, gains in women’s schooling remain in some instances modest, reversible or selective (DeRose & Kravdal, 2007; Hewett & Lloyd, 2003; Kim, Alderman, & Orazem, 1999; Subrahmanian, 2002) and it may therefore be premature to shift attention away from gender. In principle, therefore, two distinct positions have emerged with regard to the continued relevance of gender for schooling. The ﬁrst – which we label ‘‘pro-shift’’ – not only recognizes the recent gains in closing gender gaps but is optimistic about future ones, whereas the second, ‘‘anti-shift,’’ position remains wary of the depth of past gains in women’s schooling or the inevitability and pace of future ones. Stated differently, from an ‘‘anti-shift’’ perspective, the glass of past progress is half empty and must beg for more. From a ‘‘pro-shift’’ position, however, the glass is half full and likely to ﬁll

Sex versus SES

3

up steadily. Which of these two positions is correct is clearly a matter of great signiﬁcance for education policy and accordingly, the issue requires empirical evaluation. Fortunately, a tentative evaluation is now possible. Despite clear differences in prescriptions, the ‘‘pro-shift’’ and ‘‘anti-shift’’ positions share common assumptions about the pre-conditions needed before a policy shift away from gender. Three pre-conditions stand out: First, gender gaps must have closed enough to become smaller than socioeconomic gaps. Second, progress in gender equity must be expected to continue irreversibly, with limited policy intervention. Third, the process of gender convergence must pervade all socioeconomic groups. Overall, whether it is time to move away from gender depends on the relative magnitude, reversibility, and socioeconomic ubiquity of gender convergence. These three criteria can thus serve as empirical benchmarks for debating the wisdom of continued focus on gender in global education policy. Evidence on these benchmarks remains scattered so far. Although studies have monitored how much gender gaps in schooling have closed, few have tracked convergence trajectories, i.e., how gender gaps close: Do they close at the expense of (or alongside with) SES-based inequality? Do they close irreversibly? Do they close uniformly across all socioeconomic groups or, instead, is convergence initially conﬁned to top SES groups before tricking down? Answers to these questions are critical for research and policy. In policy, they can help adjust the focus of global education policy, at a time when the development community seeks to extend education to all, but when limited resources often force hard choices between various vulnerable constituencies – including female and poor children – vying for policy attention.2 Beyond policy, these answers would inform long-standing debates in academia about the roots of social inequality, notably the relative importance of cultural and ascriptive categories (e.g., sex, race) versus ostensibly universal class and socioeconomic ones. Although this debate has been recently revisited in theory (Wilson, 1980; Fraser, 1995; Young, 1997; Butler, 1998; Fraser & Honneth, 2003), empirical discussions grounded in sub-Saharan contexts are rare. Against this background, our chapter explores the changing signiﬁcance of gender for educational inequality, focusing on sub-Saharan Africa. This chapter is organized as follows. Section 2 describes the theoretical background. Section 3 reviews the criteria and Section 4 the methods. This approach is then applied (Section 5) to African countries. The results are reviewed in Section 6 and the ﬁnal section summarizes the implications for

4

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

relevant policy and academic debates, including some of the debate examined in this volume. Two early caveats can mitigate potential misunderstandings. First, our three criteria for a policy shift are narrowly statistical, as they overlook economic and political considerations, and sociological differences in the nature of the disadvantages suffered by girls as a group.3 Equally narrow is our empirical test itself, as it covers few (8) countries and a relatively brief time span (only a decade or slightly more). For these reasons, our study is more illustrative and exploratory than probative. Its main contribution is to develop and illustrate an approach for studying gender convergence trajectories.

2. THEORETICAL BACKGROUND The debate on the changing signiﬁcance of gender for education is a subset of broader theoretical debates on equity, which are themselves variants of the paradigmatic debate between ‘‘conﬂict’’/‘‘consensus’’ perspectives. These are of course ideal types and they are used here simply for heuristic purposes. The most relevant variants for our purposes are debates over ‘‘meritocracy versus social reproduction’’ in education, ‘‘modernization versus dependency’’ in economic development, and ‘‘women in development versus women and development’’ in understanding gender and development. Whereas a full review of these variants is beyond the scope of this chapter, we offer a compressed account to extract implications for the likely course of gender convergence trajectories.

2.1. Meritocracy versus Social Reproduction In explaining group differences in education and socioeconomic attainment, these two perspectives place different emphasis on levels of human capital versus various mechanisms of ascription and discrimination. Meritocracy stresses differences in ability in educational attainment and, in turn, differences in educational attainment explain occupational achievement (Blau & Duncan, 1967; Brint, 1998; Grusky, 1994). Social reproduction, however, emphasizes discrimination and bias based on ascribed characteristics of individuals. In this perspective, powerful groups maintain a social structure that systematically discriminates against less powerful groups. Embedded in this perspective is a distinction between open and closed societies, with the ﬁrst permitting mobility across classes and the latter restricting

5

Sex versus SES

movement and reproducing inequality across generations. A common expectation, however, is that in the course of modernization, societies open and gradually converge toward meritocracy. Within this perspective, gender inequality in schooling is rooted in either the preferences/predispositions of children or, conversely, in systematic restrictions placed on girls and women. These restrictions may be in the home (domestic sexual division of labor, parental restrictions, or resource control), the school system (curricula, infrastructure, and instruction), and the labor market (hiring and promotion practices). Clearly, Manichean opposition between meritocracy and social reproduction is simplistic, and analysts can use decomposition methods to apportion between merit and discrimination in occupational attainment. Nonetheless, this opposition has didactic value, as it contrasts two distinct viewpoints that closely mirror the distinctions between the ‘‘pro-shift’’ and ‘‘anti-shift’’ positions enunciated earlier. The ‘‘pro-shift’’ position clearly shares some of the benign assumptions embedded in the meritocratic perspective, whereas the ‘‘anti-shift’’ position shares the politically contentious assumptions of the social reproduction perspective.

2.2. Modernization versus Dependency Modernization, an early formulation of development theory, makes a central distinction between ‘traditional’ and ‘modern’ societies. In its classic articulation, modernization involves a transition from the former to the latter and entails a series of co-varying changes: from subsistence to market economies, from subject to participant political cultures, from ascribed to meritocratic status systems and from religious to secular ideologies. Development was conceived as a linear process whereby the adoption of Western technology, institutions, and values would, through a series of stages, lead to global societal convergence. The challenge was therefore to identify the obstacles to a self-sustaining dynamic of growth, whether institutional, technological, educational, or cultural. Relevant to our analysis here is modernization theorists’ assumption of a linear and politically benign rise in meritocratic values and a concomitant decline in both class and gender inequality, regardless of speciﬁc efforts to address these inequities. In contrast to modernization approaches, dependency theorists view underdevelopment not as some original or residual condition but an outcome of the colonial incorporation of Third World regions into the capitalist world economy. This incorporation distorts the colonized economies, leading to

6

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

what Gunder Frank (1967) described as ‘‘the development of underdevelopment.’’ Dependency theory inverts many of the assumptions of the modernization approach: the transfer of Western institutions, technology, and ﬁnance is viewed as perpetuating dependency and underdevelopment rather than self-sustaining development. The solution to this structural dependence, they argued, was a strategy of import-substitution industrialization based on the domestic market and secured through various monetary policies and tariff protections. The dependency critique proved appealing to many scholars and some policy makers in the global south, and it led to calls for a New International Economic Order (NIEO) and the recognition by the World Bank and other multilateral agencies for a basic needs approach that emphasized the necessity to tackle poverty directly.

2.3. Women in Development versus Women and Development The issues of women in development were originally framed within the modernization approach, with inclusion in existing development projects as a focal point of women’s activism. Both development specialists and political authorities assumed that macroeconomic policies were gender neutral and that the beneﬁts of modernization policies would spread to all classes and gender groups. In other words, they assumed a scenario of horizontal convergence similar to the scenario in Fig. 1. This assumption

Fig. 1.

Illustration of Two Hypothetical Gender Convergence Trajectories.

Sex versus SES

7

was closely related to the modernization paradigm’s presuppositions and simpliﬁed conceptual apparatus that abstracted from the complex historical process (Tipps, 1973). Feminists involved with development issues challenged the assumption that modernization would automatically enhance gender equality, and began to use the term women in development (WID) in their efforts to inﬂuence development policies. To improve women’s access to development, they called for more accurate information and data on women’s work and for women’s better access to education, job training, property, and credit. Women had to be consciously integrated into development projects and given a voice in policy design and implementation. This WID approach was informed by an established liberal feminism that mostly eschewed structural analyses of women’s subordination. Liberal feminists called for the legal removal of existing obstacles to various social and institutional resources such as education and employment. Women’s inequality was viewed as a consequence of discrimination that could be removed by the enactment of new laws and codes to end sex-based prejudices (Wollstonecraft, 1972).4 In her landmark study, Ester Boserup (1970) found gender inequality to be exacerbated by modernization projects that often marginalized or displaced women through the provisioning of technological training to men. Policies promoting the cash economy and integration into the market likewise moved male labor from kinship and family to the wage form, giving men access to economic and other social resources in ways that profoundly changed gender dynamics. Boserup questioned the assumption that beneﬁts from the development projects would automatically trickle-down to women and concluded that if the aggregate effects of these policies were to be redressed, it was essential to integrate women more fully into the planning and implementation of development projects (Reiter, 1975). Given its exclusive focus on inclusion, the WID approach was selflimiting and incapable of questioning the modernization framework itself. Its horizon of demands for reducing inequalities between the sexes was too often restricted to calls for improved access to educational establishments and the paid labor force (Kelly, 1994). It rarely addressed fundamental questions about women’s subordination or the impact of global processes on the lives of women. It also failed to connect inequities in the realm of development to women’s inequality across various domains of social life, and to draw the distinction between gender and sex. Feminists have argued that gender relations are expressed in almost all spheres of social life even if they take different forms in different places and historical periods. From the division of labor to the organizing of the state

8

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

and the structuring of everyday life, feminist theory examined the various ways in which ‘development’ itself was constituted as a gendered process. One upshot of all this was a new focus on the relational history and dynamics between men and women that displaced the earlier focus on the subjected sex, a conception that increasingly came to be viewed as no more legitimate than the exclusive focus on workers or peasants by sociologists of class. Radical feminists involved in the development process argued that women needed autonomous development projects and social institutions that were designed to meet their speciﬁc needs. This contention found resonance among social activists and intellectuals who argued for a development approach to women that was cognizant of the limits imposed by patriarchal institutions and recognized the need for women-only projects. This perspective has sometimes been referred to as the women and development (WAD) approach (Parpart, 1989; Rathgeber, 1990). Although, the WAD (later referred to as GAD or Gender and Development) approach offered an important corrective to the WID assumption that male-dominated states can be used to alter gender inequities, it also exhibited important weaknesses of its own.5 In the ﬁrst instance, the WAD approach often viewed women as a distinct and homogenous collectivity, downplaying differences among women along class, racial, and ethnic lines. In other words, by evading differences within women, radical feminists did not acknowledge the kinds of interactivity examined later in this chapter.

3. TO SHIFT OR NOT TO SHIFT? THE PRE-CONDITIONS Since at least the 1970s, gender has become a focal point of global educational policy. As a historically disadvantaged constituency, girls provide a critical marker in efforts to promote educational equity, a position that has been reafﬁrmed at key international conferences since the 1980s (Diaw, 2002; Whitehead & Lockwood, 1999). Yet there is increasing controversy as to whether this focus obscures other forms of inequalities, notably social disparities among girls themselves as well as more general forms of socioeconomic inequalities (Knodel & Jones, 1996). These questions can be understood against the backdrop of recent debates concerning the importance of social and cultural ascription in shaping life outcomes (Wilson, 1980; Fraser, 1997); and the more practical efforts to expand schooling opportunity in developing countries.

Sex versus SES

9

The United Nations’ commitment to reduce both gender and socioeconomic inequalities in schooling under the aegis of the Millennium Development Project presents in this respect a particularly formidable challenge for many African countries. Given the disparity between subSaharan Africa’s large school-age population and budgetary allocations for education that on average amount to no more than 4–5% of Africa’s gross domestic product (GDP), levels of expenditure per student are very small and make resource allocation a critical challenge for post-colonial states. How to adjudicate between competing constituencies such as girls, children from rural areas, those from low-income families, and children from ethnic minorities? How much emphasis to place on gender in relation to other considerations, and how to modulate this emphasis in response to changing conditions? Shifting away from a a` priori emphasis on sex might seem most justiﬁed where the gender-equity gap has closed. But this clearly does not ﬁt much of Africa’s situation where the female-to-male ratio in secondary enrollment remains below 0.80 in more than half the countries (UNESCO, 2008). Although the gender gap in education has progressively narrowed through the 1990s, the pace of this narrowing has been too slow to expect a convergence by 2015 (World Bank, 2001). Indeed, the convergence process seems to have stalled or slightly reversed, as secondary enrollments declined from 82 in 1999 to 79 in 2004 (UNESCO, 2008). For most African countries therefore, the Millennium Development Goal (MDG) of closing gender gaps within the projected time frame remains elusive. Even in such contexts, however, is a continued focus on gender justiﬁed, especially in light of other forms of inequality? The implications of this reconsideration will be examined both theoretically and empirically in what follows. The review in Section 2 has contrasted two sets of perspectives. The ‘‘consensus’’ perspective and its variants (meritocracy, modernization, women in development) assume politically benign processes of linear development with inclusion, whereas variants of the ‘‘conﬂict’’ perspective (social reproduction, dependency, women and development) assume greater political contestation and conﬂict. Again, this contrast is closely reﬂected in the ‘‘pro’’ and ‘‘anti’’ shift positions in the gender convergence debate. Like consensus, the pro-shift position envisions a politically benign view where convergence is achieved steadily, consensually, and automatically as a byproduct of linear development. Like conﬂict, the anti-shift position sees convergence as a more belabored and politically contingent process requiring continuous advocacy. Despite their clear opposition in prescriptions, pro- and anti-shift positions in fact agree on pre-conditions that have

10

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

to be met before a policy shift away from gender is warranted. Boiled down to essentials, these pre-conditions center on the relative magnitude, reversibility, and socioeconomic ubiquity of progress in closing gender gaps. As we show below, each of these three criteria takes on a slightly different conﬁguration depending on whether it is evaluated in static versus historical perspective. Thus, when evaluated historically, the ﬁrst criterion (‘‘relative magnitude’’) becomes the ‘‘changing signiﬁcance of gender’’; the second criterion (‘‘reversibility’’) becomes ‘‘momentum’’; the third criterion (‘‘ubiquity’’) becomes ‘‘directionality’’ of change. All three criteria are discussed in turn.

3.1. Relative Magnitude/Declining Signiﬁcance A strong policy focus on gender is justiﬁed if much of the national inequality in schooling unfolds along gender lines. In other words, gender-based inequality should account for a large share of overall inequality in schooling. A gender focus becomes less defensible if (1) gender gaps are very small relative to socioeconomic gaps, for instance or (2), in a dynamic perspective, gender gaps are steadily declining relative to socioeconomic gaps. Although comparison between sex versus SES-based inequality in schooling has often been made in static perspective (Bloom, 2006; Stromquist, 1990), it is rarely explored historically. Knodel and Jones (1996) are one exception but their analysis focused on two countries only, Vietnam and Thailand. Furthermore, even when studies compare sex and SES-based inequality, they use statistical differentials rather than fuller measures of inequality (Giroux, Eloundou-Enyegue, & Lichter, 2008). We advance analysis in this area by using fuller measures of inequality such as the education Gini or Theil, which combine information on both statistical difference in education between groups and information on group size.

3.2. Irreversibility/Momentum Even before gender gaps fully close, policy makers can contemplate an early shift away from gender if they expect girls’ schooling to continue to improve irreversibly on its own, under the momentum of past gains or broad social change. Such momentum can stem from intergenerational processes within families, as each generation surpasses the achievements of the parental generation. Girls’ education especially beneﬁts from parental education and

11

Sex versus SES

the presence of older sisters (Filmer, 1999; Marteleto, 1996; Parish & Willis, 1993). Gains in one generation thus spill over to the next generation through emulation, role modeling,6 and women’s increased control of domestic resources. If female-headed households invest more than male-headed ones in children’s human capital (Lloyd, 2006; Lloyd & Blanc, 1996), women’s current gains in the household, labor market, and political arena will support future gender convergence in schooling. In the demographic arena, fertility transitions (and the resulting smaller size of families) boost schooling, if resource-strapped households discriminate against girls (Bloom, 2006; Lloyd & Gage-Brandon, 1994; World Bank, 2001). In the economic arena, women’s entry into the labor force augments the household resources, whereas also improving women’s control of domestic resources (Herz & Sperling, 2004). Finally, in the policy arena, women’s increasing representation improves opportunities for the next generation of girls. Momentum can also stem from macroeconomic change. As formal employment opportunities for women improve, gender discrimination in educational investment is likely to wane (Buchmann, 2000; Csapo, 1981). Momentum can also arise externally from cultural globalization (Ilon, 1998). Sub-Saharan Africa has seen a jump in various indicators of global connectivity,7 and a surge in NGOs since the mid-1980s, including externally funded NGOs that directly or indirectly promote cultural change (Sakabe, Kandiwa, & Eloundou-Enyegue, 2006). These cultural vectors have the potential to transform household production and reproduction, as well as gender dynamics within households in ways that affect the distribution of resources between boys and girls. In summary, several intergenerational, cultural, and economic inﬂuences can sustain the momentum of gender convergence in countries where this process has begun.

3.3. Socioeconomic Ubiquity/Directionality In monitoring gender equity, analysts have often focused on differences between boys and girls, to the exclusion of possible differences among girls, and how these internal differences may evolve as the aggregate gender gap narrows down. The implicit assumption is thus that gender convergence occurs ubiquitously/evenly among girls in all socioeconomic groups. Whether this assumption holds is important. First, it affects the relative importance of gender vis-a`-vis socioeconomic gaps, and how this importance changes historically. Second, it determines whether efforts to narrow gender gaps will also reduce socioeconomic inequality.

12

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

To illustrate this, consider Fig. 1, which sketches two opposite convergence trajectories. The bottom rows on the ﬁgure show how aggregate gender gaps change in each of two countries, with darker colors indicating a narrowing of the gender gap. On the aggregate, the two countries appear to have experienced the same extent of convergence. Yet their detailed trajectories are distinct. Society 1 experienced a ‘‘vertical’’ convergence, with gains beginning among high SES groups and gradually trickling down, whereas gains in society 2 occurred uniformly across all SES groups. Clearly, these two patterns have distinct implications for the socioeconomic ubiquity of convergence, and for internal inequality among girls. As these societies move through their process of convergence, the internal inequality among girls increases (during the intermediary stages) in society 1, but it remains the same in society 2. In terms of relative importance, society 2 shows little intrinsic inequality among girls. Any inequality among girls derives entirely from class differences, suggesting that gender has greater salience in terms of group identiﬁcation than does class in this society: A poor girl living in society 2 can relate to girls in higher SES groups insofar as they share a common experience of disadvantage. She may not relate to boys in the same SES group, despite their common class afﬁliation. In that context, broad gender reforms (e.g., generic expansion of gender-equity norms) can effectively address the gender gap in schooling. This is less true of society 1 where class is more salient. At all but the very ﬁrst and ﬁnal stages of this country’s educational transition, girls in the lower SES group do not share a common fate with girls in higher SES groups. However, children in the same socioeconomic class have identical outcomes, regardless of gender. Class, not gender, is the important dimension of stratiﬁcation. Efforts to address schooling inequality are thus best oriented towards addressing class differences. The two distinct convergence scenarios also have different implications for the likely roots of gender convergence and the continued importance of policy. In terms of roots, the uniform convergence in society 2 is consistent with a generic process of cultural globalization that can sustain convergence even in the absence of favorable economic conditions or policies. However, the vertical convergence in society 1 requires speciﬁc targeting of girls, mostly in low SES groups. Even as convergence occurs, the targeting of girls remains warranted so as long as efforts focus on lower income girls. Looking at the fourth time point in society 1’s transitions, one can see at least three reasons why such targeting is warranted. First, it would sufﬁce to close the national gender gap, whereas it

13

Sex versus SES

would not be similarly effective in society 2. Second, it need only to cover a limited number of girls, whereas while broader – albeit thinner – coverage would be required in society 2. Third, it would reduce both gender and SES gaps. In summary, continued focus on girls with an emphasis on boosting enrollment among poor girls in particular is warranted in society 1, but not in society 2. A society experiencing the horizontal convergence seen in society 2 will likely suffer less from an early policy shift away from gender. Its horizontal convergence implies greater potential for momentum, lower salience of gender, and less potential for interactivity. Also, efforts to reduce gender inequality in such a setting are less likely to beneﬁt the poor. In contrast, the ‘‘top-down’’ convergence in society 1 implies less momentum, and targeted policy efforts to reach poor girls can address both the gender and SES gaps. In this second scenario, it is less defensible to shift away from gender until convergence is nearly complete. Overall, whether or not an early policy shift away from gender is warranted depends on convergence trajectories, making the documentation of these trajectories necessary.

4. METHODS AND DATA 4.1. Measures The three criteria described earlier (relative importance, reversibility, and ubiquity) can help evaluate the continued salience of gender for educational inequality. As indicated previously, each takes on a slightly different conﬁguration, depending on whether it is evaluated in static or historical perspective. The indices used to evaluate these three criteria are described in turn. 4.1.1. Relative Importance/Changing Signiﬁcance (R) The index of relative importance (R) is the ﬁrst criterion. The idea here is to compare the extent of sex- to SES-related inequality. Contrary to previous studies that focus on differentials, we use full-information measures of inequality (here the Theil8) that integrate information about both effect size and group size. Key to our analysis is the distinction between inequality across SES (BSES) or across the sexes (Bsex). These two components are computed as: BSES ¼ ðsmale T male Þ þ ðsfemale T female Þ

(1)

14

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

Bsex ¼ ðspoor T poor Þ þ þ ðsrich T rich Þ

(2)

where si is the share of the sum of the education of each sex (SES group, respectively) relative to the national sum and Ti is the Theil inequality index across SES (sex) for the corresponding sex (SES group, respectively). Ultimately, R is measured as the ratio of Bsex over BSES. It indicates whether inequality among the sexes supersedes inequality among the socioeconomic groups. Again, the comparison based on this fuller information measure is better than one derived from standard regression analysis because it combines information about both groups differences and group size. As indicated earlier, it is useful to monitor how R changes over time, i.e., whether the relative importance of gender declines or increases historically. These historical changes were captured by plotting R values against historical time or against stages of educational development. 4.1.2. Irreversibility/Momentum Irreversibility and momentum are the second criterion. Both concepts capture the same idea, except the second is evaluated in historical context. Irreversibility is the likelihood that gender convergence will continue (and not reverse) once initiated. ‘‘Momentum’’ additionally suggests that irreversibility should become more entrenched (i.e., reversals become less likely) as countries advance in their educational transitions. The basic indicator of irreversibility (I ) is therefore measured as the infrequency of reversals in gender gaps, i.e., how infrequently female-to-male enrollment ratios (FMER) among 10–19-year-olds (F) turn out to be lower than values observed during an earlier time period. In other words, I indicates the percentage of instances when reversals are not observed. It ranges from 0 to 1, with higher values indicating greater irreversibility. When irreversibility is considered in historical perspective, it serves to evaluate the extent of momentum. Momentum is said to exist, if reversals become less and less frequent as countries advance in their educational transition. By recording information on countries’ stage in educational transition and plotting it against the corresponding I values, one thus has an indication of the extent of momentum. 4.1.3. Socioeconomic Ubiquity/Directionality (U) Like the two previous concepts, the notions of ubiquity/evenness and directionality of gender convergence capture the same idea, except one (direction of convergence) is framed in more historical context. In its basic sense here, socioeconomic ubiquity refers to whether the extent of gender inequality is the same across all socioeconomic groups. The minimum test of

15

Sex versus SES

ubiquity involves cross-sectional comparison of FMER values across SES groups, to see if values are lower among low SES groups. To facilitate synthesis, comparison can focus on extreme groups, here the bottom two SES groups versus the top two SES groups.9 In this simple case, ubiquity is simply the ratio of FMER within bottom versus top SES categories (U ¼ Fbot =Ftop ). U is close to 1 if no interaction exists between gender and SES. It is signiﬁcantly lower (or higher) than 1, depending upon whether gender inequality is more severe at the bottom (top, respectively).10 Beyond cross-sectional comparison (that mostly documents ubiquity), a fuller test to look at directionality of convergence requires a historical perspective that shows how U changes as countries advance in their educational transition. If convergence occurs ‘‘horizontally,’’ U values should remain close to 1 throughout the educational transition. However, if convergence occurs ‘‘vertically’’ (from the top-down) then U values should be small in early stages, before gradually converging towards one (1). By plotting U values (y-axis) against countries’ stage in educational transition (x-axis), one can visually infer the direction of convergence. These visual insights can be complemented with more formal analyses of the statistical relationship between countries’ transition stage and U values. This statistical analysis should yield a large R2 value and a signiﬁcant regression coefﬁcient in the case of vertical convergence. A small R2 and a ﬂat line are expected in the case of horizontal convergence. To evaluate ubiquity across various socioeconomic groups, we focused on eight (8), SES groups deﬁned by housing characteristics and ownership of various consumer durables.11 Overall, when used in combination, the values and the changes in R, U, and I indice fully describe national convergence trajectories. In a ‘‘horizontal’’ convergence scenario (case 2 of Fig. 1) I and U values (irreversibility/ ubiquity) should be close to unity (1) whereas they would be substantially different from unity (generally less than one) in the case of a vertical convergence. Similarly, in the case of horizontal convergence, the value of R (relative magnitude) tends to remain stable over time, whereas it is expected to decline historically in the case of vertical convergence. One can thus refer to these values to evaluate the extent which a convergence trajectory approximates a vertical or a horizontal mode.

4.2. Data Analyses use data from the Demographic Health Survey (DHS). Over the last two decades, the DHS have conducted nationally representative and

16

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

large surveys on a wide range of issues, including health, education, and nutritional status in over 75 countries mostly in sub-Saharan Africa, South and South East Asia, and Latin America. Interestingly for our purposes, the surveys have been repeated every few years (5 or so) in some countries, including the eight sub-Saharan countries – Burkina Faso, Cameroon, Ghana, Kenya, Madagascar, Niger, Tanzania, and Zambia – included in our sample. Although this subset of countries does not represent the region in a statistical sense, it does cover a wide range of circumstances. Gender gaps in this subset of countries range from small (Madagascar, Zimbabwe) to very large (Burkina Faso), and the overall schooling levels vary similarly from a low of 23% in Niger in 1998 to a high of 78% in Uganda in 2005. Adding to this cross-sectional variability, the surveys span a period (the 1990s and early 2000s) when many African countries underwent dramatic socioeconomic changes that arguably affected schooling opportunities. Noteworthy among these are economic downturns and structural adjustment programs that raised the costs of secondary and university education, as well as reduced employment opportunities with the public sector while also reducing household incomes in real terms (Courade, 1994). In addition, incipient declines in fertility, delays in marriage, and a surge in AIDS-related mortality changed the family environments under which decisions about schooling were being made. Our focus on this eventful time period and on a diverse set of countries permits both geographic and historical contrast.

5. FINDINGS Table A1 shows the raw data on national enrollment and gender gaps in enrollment. These raw data underscore the variability in national conditions and outcomes. National enrollment levels in study countries range from 0.24 in Burkina Faso (1998) to around 0.75 in Cameroon, Zimbabwe, and Uganda for their last survey years. The national FMER range from near parity in Zimbabwe and Madagascar to 0.64 in Niger (1996). Even greater inequality is found when one compares gender gaps across various SES groups. Sub 0.50 values (including 0.16 in Burkina Faso and 0.42 in Niger) are found among the lowest SES group, whereas values above 1 are found in the higher or even in middle SES groups. The detailed data in this table also reveals whether gender gaps close as a result of negative convergence (male enrollments declining, as was the case of Zambia between 1992 and 1996) or positive convergence (as was the case of Burkina Faso between 1998 and 2003). Based on the information in Table A1, summary indices were

17

Sex versus SES

computed for the relative importance of gender and SES (R), the ubiquity (U ), and irreversibility (I ). These various indices are summarized in Table A2. The table shows data for each country period, where appropriate, as well as the unweighted average value for all the country periods. For ease of presentation, much of the discussion of these ﬁndings is based on graphical summaries of the information contained in Tables A1 and A2.

5.1. Relative Importance/Declining Signiﬁcance Table A2 shows the estimates of sex- and SES-related inequality for all country periods. They show substantial inequality along both gender and class lines, but with larger values for SES-related inequality. Whereas the average value for SES-related is approximately 0.21, it is only 0.08 for inequality between the sexes. On average, sex-related inequality is only half (0.48) the value of SES-related inequality. However, this ratio (R) varies markedly across country periods, from a low of 0.10 in Madagascar to a high of 0.95 in Zimbabwe where both SES- and sex-related inequality are low. It is important to monitor how the relative importance of sex-related inequality evolves historically. Fig. 2 thus plots R values against the country’s advance in educational transition, i.e., how does the relative importance of sex-related inequality change as countries expand their total enrollments? The ﬁndings show that the country’s advance in the educational transition does to matter. For the subset of countries and the time period investigated here, the relative importance of sex does in fact increase as countries advance in their educational transition. A good third of all the variation observed in the relative importance of sex inequality is associated with countries’ stage in educational transition. It is important to note that the relative importance of sex increases, rather than decrease as theoretically expected. Although these ﬁndings run against our theoretical expectation of a historical decline, they are understandable. For countries in the early stages of their transitions, school enrollment, and physical access to schools are low, especially in rural areas. Boys and girls, regardless of parental inclinations, will experience similar (poor) schooling outcomes. Because much of the education is conﬁned to urban settings, class rather than gender will be the prime factor. As school infrastructure and schooling opportunities become available, there is greater room for differentiation between boys and girls. There remains of course the possibility that the relative importance of gender may still decline after some later threshold but

18

Fig. 2.

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

Relative Magnitude of Sex-Related Inequality (R)* across Stages of Educational Transition.

for now, the importance of gender is rising during the range of transition stages observed in this sample. The evidence on the relative importance of gender thus leads us to two opposite conclusions. On the one hand, the magnitude of sex-related inequality is consistently smaller than that of SESrelated inequality, a ﬁnding that warrants reduced attention to gender as a focal point of educational equity policy. However, the relative magnitude of this sex-related inequality is expected to increase during the early transition stages where many African countries currently ﬁnd themselves. 5.2. Irreversibility/Momentum The purpose in this analysis was to estimate the likelihood of historical reversals/stalling in gender convergence, especially after countries reach some threshold in school participation. The raw data in Table A2 show a high incidence of reversals. I values range from lows of 0.25 (a high

Sex versus SES

19

incidence of reversals/stalling) in Madagascar 1997 and Cameroon 1998 to high values of 1 (no incidence of reversals) in Ghana (1998) and Uganda (2000), with an average of 0.64: In nearly 1 out of 3 cases, the FMER values registered within a socioeconomic group turn out to be equal or smaller than values registered during an earlier survey. The reversals noted in some of the study countries have been documented elsewhere and reﬂect in part the inﬂuence of adverse economic conditions (Courade, 1994; Hewett & Lloyd, 2003; DeRose & Kravdal, 2007). Nonetheless, the sheer frequency of such stalling/reversals deserves note. A fuller understanding of momentum requires, however, exploring whether reversals become less likely as countries advance in their educational transition. Fig. 3 thus plots I values against the country’s advance in educational transition, as indicated by rising national enrollment rates. This analysis shows no signiﬁcant historical variation: I values remain similar through the educational transition. Formal statistical analysis

Fig. 3. Relationship between Irreversibility Index (I ) and Stage in Educational Transition. Note: I Indicates as Probability that Later values for Gender Gaps are Smaller that Equal the Gaps Observed during Period. A Value of I Indicates Irreversibility.

20

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

conﬁrms this lack of historical trend (see Fig. 3), as does a more detailed line chart showing the relationship between momentum and national enrollment levels, within individual countries (data not shown). It is well to note, however, that even as the relationship is not statistically signiﬁcant at the 0.05 level, the direction of the change in the irreversibility/momentum index is in the right direction. With every 10 percentage points increase in national enrollment rates, the probability that gains in women’s schooling will stall or reverse declines by roughly 0.02 units. A more extensive test involving a larger number of countries is warranted. So far however, the data show little evidence of momentum, and gains in gender gaps appear to stall or reverse even at relatively advanced transition stages. Such evidence clearly cautions against shifting emphasis away from gender, on grounds that future gains will continue automatically.

5.3. Ubiquity/Direction A third criterion in deciding policy emphasis involves examining whether the process of gender convergence is widely shared among all socioeconomic groups. We examine this in both static and historical perspectives. In the static perspective, we examine the similarity of FMER across all SES groups, i.e., whether gender inequality (or equality) is ubiquitously found among all socioeconomic groups. This similarity is captured by U values, which measure the ratio of FMER values among bottom SES groups to FMER values among top SES groups. Ubiquity or similarity in gender gaps across all groups will result in U values close to 1, whereas large dissimilarity will yield small U values, less than 1 and closer to 0. The summary results, shown in Table A2, show an average U value of 0.91, suggesting substantial similarity across SES groups. However, the results also indicate substantial variation, from a minimum of 0.45 to a maximum of 1.13. In essence, there are many cases when gender gaps are substantially larger at the bottom of the SES distribution but also cases when the class differences in gender gaps are small or, in fact, when gaps appear to be a little larger among top SES groups. Given this variability, it is interesting to explore how U values change as countries advance in their educational transition (Fig. 4). The relationship is strong and curvilinear. As countries advance in their educational transition, the initially large gender gaps among lower SES groups begin to close and match those found in high SES groups as well. Nearly 2/3 of all variation in U values is associated with the country’s stage in educational transition. By the time countries reach a 70% enrollment threshold,12 the magnitude of

21

Sex versus SES

Fig. 4. Relationship between the Socioeconomic Ubility Index (U) and Stage in Educational Transition. Note: U Indicates Socioeconomic Ubiquity. It is Speciﬁcally Measured as the Ratio of Female-to-Male Enrollment among the Bottom-Two SES Groups against the FMER Ratio among the Top-Two SES Groups. A Value of 1 Indicates that Gender Gaps are Identical across the SES Groups. Values below 1 Indicate that Gender Gaps are Larger among the Lower SES Groups.

gender gaps should be roughly the same among bottom SES groups than among top SES groups. In sum, convergence appears to follow a vertical, rather than horizontal pattern. For countries still below a 70% threshold in enrollment, continued investments in girls in low SES groups is warranted and it will likely yield gains for both girls and the poor. The curvilinearity in the predictive equation suggests however that beyond this 70% threshold, the trend toward ubiquity tends to slow down or even reverse. This suggests that, at a minimum, one cannot take for granted the expectation of a steady uniformization of girls’ experiences across all socioeconomic groups.

6. CONCLUSION As gender gaps close worldwide, there are increasing questions about the relative importance of gender as a focal point for addressing schooling

22

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

inequality in developing countries. How policy and research adjust to these new trends is important. One issue in sub-Saharan Africa is whether the time has come to gradually shift emphasis from sex-related inequality to socioeconomic inequalities that may be growing in the wake of economic and demographic changes, as well as a possible weakening of informal solidarity networks that had historically equalized schooling outcomes. Arbitrage on this shift can be made on several grounds, a subset of which is explored in this chapter. This subset focuses on aspects of convergence trajectories that include changes in the relative importance of sex-based inequality, the reversibility of gains in closing gender gaps, and the socioeconomic ubiquity of gender convergence. We postulate that early shifts away from gender may be more warranted if national convergence appears to unfold along a ‘‘horizontal pattern’’ that implies strong momentum and little interactivity with class, and in which gender becomes a comparatively less salient dimension of inequality. Based on this premise, the empirical task is then to document recent convergence trajectories in various sub-Saharan countries and how these transitions depend on a host of contextual variables. Our ﬁndings are mixed. On the one hand and consistent with Knodel and Jones (1996), schooling inequality between the sexes is consistently smaller than SES-related inequality. On the other hand, there is little evidence of a momentum: the process of gender convergence is found to stall or reverse regardless of the fertility transition stage. Furthermore, it unfolds in topdown, rather than horizontal fashion: gains in gender equality typically begin among the top SES groups before trickling down. On the whole however, the ﬁndings for this limited subset of countries are more consistent with a ‘‘vertical’’ rather than ‘‘horizontal’’ convergence pattern, and would lead us to caution against dramatic shift away from gender at this point. More plausible perhaps is a recommendation to increasingly focus on poor and rural girls, and on win–win strategies that address both female and poor children. Again, our small sample and narrow historical coverage preclude strong conclusions, and therefore our analysis is more illustrative than probative. Nonetheless it can advance future research and theoretical debate on these issues. In terms of research, the tentative inference from our ﬁndings is that progress in narrowing gender gaps is neither irreversible and linear, nor horizontal. Rather, the advance in the schooling of African girls is prone to stalling and socioeconomic differentiation. More importantly, our research draws attention to the importance of convergence trajectories and outlines a method for describing these trajectories and discussing their implications.

Sex versus SES

23

Empirical investigations can thus be advanced in two directions. One is to focus on individual experiences. For countries where detailed historical data on schooling by gender and class is available, convergence trajectories can be studied in greater detail than was possible here. Countries can further draw from qualitative evidence and deep understanding of history to reﬁne description of the individual convergence trajectories of a given nation, and implications for its future prospects. A second line of inquiry is to build on multi-country studies that shed light on the contextual inﬂuences and policies that promote stalling, provided that appropriate data is available for multiple settings. In terms of theoretical debates, our conclusions dovetail with calls to treat misrecognition as a question of social status. Fraser, in particular suggests reconceptualizing gender inequalities in education as a form of institutionalized subordination, it is possible to integrate it with questions of socioeconomic inequalities. Misrecognition and maldistribution can therefore be understood as mutually irreducible impediments to parity of opportunity and participation. This does not mean, as some of Fraser’s critics have argued, that misrecognition is in some sense derivative or secondary to those of material and economic subordination. Fraser’s normative distinction between misrecognition and maldistribution does not imply some necessary hierarchy, but is rather intended to emphasize the intersections of the two forms of injustices. There is obviously great diversity in the living conditions of rich and poor, and between those belonging to different ethnic, racial, caste, and other similar distinctions, and this diversity necessarily makes the social construction of gender highly contradictory and complex. Although we can celebrate cultural diversity as a vital expression of human societies, perceived cultural difference has also often been foundational to social inequality and exclusion. As Lewis and Lockheed point out in a valuable study of the neglect of girls’ schooling, almost two-thirds of the 60 million girls not in school belong to culturally or religiously deﬁned minority groups that are subjected to discriminatory exclusion. The politics of difference in fact accounts for the more pronounced ‘‘inexcusable absence’’ of girls from educational institutions in culturally diverse societies: ‘‘In homogeneous countries, higher shares of girls complete primary school, enroll in secondary school, and see higher achievement than those in heterogeneous countries’’ (Lewis & Lockheed, 2006, p. 4). It consequently makes little sense to talk about ‘‘African’’ or ‘‘third world’’ women as if they were a monolithic category. At the same time, while questions of gender equity cannot be relegated to a separate sphere, it is crucial to emphasize that all social processes have gendered

24

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

dimensions which demand continued scholarly and political recognition. Women’s participation in the paid labor force has for instance risen signiﬁcantly over the past decades, but it has been accompanied by a concomitant rise in the feminization of poverty, making the sharp conceptual polarity between gender and poverty untenable (Beneria, 2003). Informed by the implications of this rethinking, the analysis above has sought to outline one approach to the policy conundrum in developing countries over targeting of educationally vulnerable groups, and to broader academic debates on the changing salience of sex and SES in the course of development and educational transitions.

NOTES 1. Much of Latin America has now achieved parity, whereas gender gaps remain substantial in sub-Saharan Africa and South East Asia. Further variation is found within regions, especially in sub-Saharan Africa where female-to-male enrollment ratios at the secondary level range from parity in South Africa, Swaziland, Botswana, Kenya to below 0.60 in lagging countries such as Chad, Benin, Guinea, or Togo. 2. The stated Millennium Development Goals (MDGs) in the schooling arena to ‘‘ensure that all boys and girls [regardless of poverty status] complete a full course of primary schooling’’ and ‘‘eliminate gender disparity in primary and secondary education, preferably by 2005, and at levels by 2015.’’ As these formulations make clear, the gender mandate is more explicit and ambitious than the goal in the socioeconomic arena. Yet a simultaneous commitment to girls and the poor pervades this agenda. 3. Even our list of statistical criteria is not exhaustive. One important criterion, not considered here is ‘‘non-zero sumness,’’ i.e., the extent to which policies and programs to address gender inequality have spillover beneﬁts in reducing SES-based inequality. Spillovers can result from emulation, substitution, or free riding. Financial incentives targeting girls can indirectly motivate boys who are intent on keeping up with girls’ performance (Kremer, Miguel, & Thornton, 2004; Kim et al., 1999). Spillovers can also stem from substitution in family budgets. Monetary assistance to girls eases the ﬁnancial constraints of poor families who can then allocate the freed resources to boys. Finally, spillovers arise when policies address dropout reasons that simultaneously affect girls and the poor. Reducing distance to schools will likely boost girls’ enrollments in communities where parents are concerned for their daughters’ safety (Subrahmanian, 2002) but also help poor families avoid the extra costs associated with distant schools. Spillovers can also result from free riding if advocacy for girls ends up indirectly beneﬁting the poor because both groups drop out of school for similar reasons. Conversely, little spillover will result if major dropout reasons are sex-speciﬁc, as may be the case for teenage pregnancy or early marriage (Eloundou-Enyegue, 2004; Mensch & Lloyd,

Sex versus SES

25

1998). If no interaction exists, zero-sum policy must be envisioned in which investments in one group come at the expense of the other, and in which each constituency requires individualized attention. 4. The tradition of liberal feminism has a long lineage. As early as 1792, Mary Wollstonecraft had argued in A Vindication of the Rights of Woman with Strictures on Political and Moral Subjects, that women’s capacity to reason was equal to that of men and the primary reason women appeared to be intellectually inferior was due to their inferior education, which was a consequence of inequality that was premised on a denial of women’s political rights. 5. Radical feminists and dependency theorists had focused their critique not just at modernization approaches, but at what they considered to be economically reductionist Marxist accounts as well. Marxist feminists were subsequently criticized for failing to acknowledge the sphere of reproduction as a source of value-creation and exploitation, and thus traced the roots of gender inequality to women’s exclusion from capitalist development, reverting once again to cultural explanations to account for gender hierarchies within capitalism. Marxist accounts were also subjected to criticism for presuming that only with the abolition of the social–property relations of capitalism will women’s subordination be overcome. 6. At the aggregate, however, intergenerational transmission can be slowed by differential reproductive behavior: If less educated women marry earlier and bear more children, they will disproportionately inﬂuence the composition of the next generation of children, in ways that slow the potential gain from intergenerational transmission (Maralani & Mare, 2005). 7. The prevalence of ﬁxed and mobile phone users in the region has jumped from 8 per 1,000 persons in 1980 to 140 per 1,000 persons in 2005 (WDI, 2008). A similar rise in television ownership is observed, from 7 per 1,000 persons in 1993 to 14 per 1,000 persons in 2002, and in internet subscribers, from 0 per 1,000 persons in 1990 to 29 per 1,000 P persons in 2005 (WDI, 2008). 8. T ¼ j pj rj log rj , where p is the proportion of children in group j and r the ratio of enrollment in this group to average enrollment. 9. Using two groups (instead of a single group) at each extreme helps reduce bias associated with small sample size. We also ran a variant of the same measure using all SES groups (rather than the two extreme groups). This more comprehensive index measured the average difference between adjacent SES categories U2 ¼ P ðFi Fi1 Þ wi and it should be close to zero when little systematic interaction exists; it will be positive where gender disadvantage is greater among low SES groups and negative in the opposite circumstances. In computing U2, we considered both unweighted and weighted averages, with the weight being deﬁned by the number of children in the two adjacent SES categories being compared. The results presented here are based on the simpler index described in the main text. 10. One could also extend comparison to all SES groups and estimate the average difference in F values between adjacent groups. This more detailed analysis is not included here. 11. Speciﬁcally, we used characteristics of the ﬂoor, source of drinking water, toilet facilities, radio, electricity, TV, and ownership of a private car. Individuals were assigned to various categories as follows: 1 ¼ poor ﬂoor, poor drinking water, and poor toilet; 2 ¼ 2 of the following: (poor ﬂoor, poor drinking water, and poor toilet);

26

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

3 ¼ 1 of the following: (poor ﬂoor, poor drinking water, and poor toilet) and no radio; 4 ¼ 1 of the following: (poor ﬂoor, poor drinking water, and poor toilet) and a radio; 5 ¼ 1 of the following: (poor ﬂoor, poor drinking water, and poor toilet) and electricity; 6 ¼ 1 of the following: (poor ﬂoor, poor drinking water and poor toilet) and a television; 7 ¼ 1 of the following: (poor ﬂoor, poor drinking water and poor toilet) and a fridge; and 8 ¼ 1 of the following: (poor ﬂoor, poor drinking water, and poor toilet) and a car. 12. This threshold value is obtained visually or from the predictive equation in Fig. 4. Taking the ﬁrst order derivative (Yu ¼ 2.443.62X) and solving for 0, the threshold value is approximately 2.44/3.62 or roughly 0.67.

ACKNOWLEDGMENTS The authors thank Macro-DHS for making the data available. Earlier drafts or ideas in this chapter were presented at Cornell, at the 2008 annual meeting of the Comparative International Education Society (CIES) in New York City, and at a joint seminar organized by the department of Sociology and Institute for International Education (IIE) at Korea University. The authors are responsible for any remaining inaccuracies in this chapter.

REFERENCES Beneria, L. (2003). Gender, development, and globalization: Economics as if people mattered. New York: Routledge. Blau, P. M., & Duncan, O. D. (1967). The American occupational structure. New York: Wiley. Bloom, D. (2006). Measuring global educational progress. In: J. Cohen, D. Bloom & M. Malin (Eds), Educating all children: A global agenda. Cambridge, MA: American Academy of Arts and Sciences. Boserup, E. (1970). Women’s role in economic development. New York: St. Martins. Brint, S. (1998). Schools and societies. Thousands Oaks, CA: Pine Forge Press. Buchmann, C. (2000). Family structure, parental perceptions, and child labor in Kenya: What factors determine who is enrolled in school? Social Forces, 78, 1349–1378. Butler, J. (1998). Merely cultural. New Left Review, 227, 33–44. Courade, G. (Ed.) (1994). Le Village Camerounais a` l’Heure de l’Ajustement. Paris: Karthala. Csapo, M. (1981). Religious, social and economic factors hindering the education of girls in Northern Nigeria. Comparative Education, 17, 311–319. DeRose, L. F., & Kravdal, O. (2007). The effects of educational reversals on ﬁrst births in subSaharan Africa: A dynamic multi-level perspective. Demography, 44, 59–77. Diaw, C. (2002). Gender and education in sub-Saharan Africa: The women in development approach (WID) and its discontents. LLC Review, 1, 54–66. Eloundou-Enyegue, P. M. (2004). Pregnancy-related dropouts and gender inequality in education: A life-table approach and application to cameroon. Demography, 41, 509–528.

Sex versus SES

27

Filmer, D. (1999). Educational attainment and enrollment proﬁles: A resource book based on an analysis of demographic and health survey data. Policy Research Working Paper No. 2268. World Bank, Washington, DC. Frank, A. G. (1967). Capitalism and underdevelopment in Latin America. New York: Monthly Review Press. Fraser, N. (1995). From redistribution to recognition? Dilemmas of justice in a ‘Post-Socialist’ age. New Left Review, 212, 68–93. Fraser, N. (1997). A rejoinder to iris young. New Left Review, 223, 126–129. Fraser, N., & Honneth, A. (2003). Redistribution or recognition: A political-philosophical exchange. London: Verso. Giroux, S., Eloundou-Enyegue, P. M., & Lichter, D. (2008). Recent trends in fertility inequality in sub-Saharan Africa: Differentials versus overall inequality. Studies in Family Planning, 39(3), 187–198. Grusky, D. B. (Ed.) (1994). Social stratiﬁcation: Class, race, and gender in sociological perspective. Boulder, CO: Westview. Herz, B., & Sperling, G. B. (2004). What works in girls’ education: Evidence and policies from the developing world. Council on Foreign Relations. Hewett, P. C., and Lloyd, C. B. (2003). Progress toward education for all: Trends and current challenges for sub-Saharan Africa. Population Council Working Paper No. 196. Population Council, New York. Ilon, L. (1998). The effects of international economic trends on gender inequality in schooling. International Review of Education, 44, 335–356. Kelly, M. (1994). Broadening the scope: Gender and the Study of International Relations. In: A. Douglas Kincaid & A. Portes (Eds), Comparative national development: Society and economy in the new global order. Chapel Hill: University of North Carolina Press. Kim, J., Alderman, H., & Orazem, P. (1999). Can private school subsidies increase enrollment for the poor? The Quetta Urban Fellowship Program. World Bank Economic Review, 13, 443–465. Knodel, J., & Jones, G. W. (1996). Post-cairo population policy: Does promoting girls’ schooling miss the mark? Population and Development Review, 22. Kremer, M., Miguel, E., & Thornton, R. (2004). Incentives to learn. NBER Working Paper No. W10971. National Bureau of Economic Research, Cambridge, MA. Lewis, M., & Lockheed, M. (2006). Inexcusable absence: Why 60 million girls aren’t in school and what to do about it. Washington, DC: Center for Global Development. Lloyd, C. B. (2006). Schooling and adolescent reproductive behavior in developing countries. Public choices, private decisions: Sexual and reproductive health and the Millennium Development Goals. New York: UN Millennium Project. Lloyd, C. B., & Blanc, A. K. (1996). Children’s schooling in sub-Saharan Africa: The role of fathers, mothers and, others. Population and Development Review, 22(2), 265–298. Lloyd, C. B., & Gage-Brandon, A. J. (1994). High fertility and children’s schooling in Ghana: Sex differences in parental contributions and educational outcomes. Population Studies, 48, 293–306. Malhotra, A., Pande, R., & Grown, C. (2003). Impact of investments in female education on gender equality. Washington, DC: International Center for Research on Women. Maralani, V., & Mare, R. (2005). Demographic pathways of intergenerational effects: Fertility, mortality, marriage and women’s schooling in Indonesia. California Center for Population

28

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

Research On-Line Working Paper Series. Accessed online from: http://computing. ccpr.ucla.edu/ccprwpseries/ccpr_019_05.pdf Marteleto, L. (1996). Children’s schooling in Brazil: Do number and composition of siblings matter? Population and Development Review, 22, 47–75. Mensch, B., & Lloyd, C. (1998). Gender differences in the schooling experiences of adolescents in low-income countries: The case of Kenya. Studies in Family Planning, 29. Parish, W., & Willis, R. (1993). Daughters, education, and family budgets: Taiwan experiences. Journal of Human Resources, 28, 863–898. Parpart, J. (Ed.) (1989). Women and development in Africa. Lanham, MA: University Press of America. Rathgeber, E. M. (1990). WID, WAD, GAD: Trends in research and practice. Journal of Developing Areas, 24, 502. Reiter, R. (Ed.) (1975). Toward an anthropology of women. New York: Monthly Review Press. Sakabe, Y., Kandiwa, V., & Eloundou-Enyegue, P. (2006). The rise of African NGOs: Functional or opportunistic rise? Current, 10. Stromquist, N. P. (1990). Women and illiteracy: The interplay of gender subordination and poverty. Comparative Education Review, 34, 95–111. Subrahmanian, R. (2002). Gender and education: A review of issues for social policy. Social Policy and Development Programme Paper 9. UNRISD, Gene`ve. Tipps, D. (1973). Modernization theory and the comparative study of societies: A critical perspective. Comparative Studies in Society and History, 15, 22. UNESCO. (2008). UNESCO Institute for Statistics (Accessed online at: http://stats.uis. unesco.org/unesco/TableViewer/document.aspx?ReportId ¼ 143&IF_Language ¼ eng). UNFPA. (2005). The state of the world population 2005. New York: United Nations Population Fund. UNICEF. (2004). The state of the world’s children 2004. New York: United Nations Children’s Fund. Whitehead, A., & Lockwood, M. (1999). Gendering poverty: A review of six World Bank African poverty assessments. Development and Change, 30, 525–555. Wilson, W. J. (1980). The declining signiﬁcance of race: Blacks and changing American institutions (2nd ed.). Chicago: University of Chicago Press. Wollstonecraft, M. (1972). Thoughts on the education of daughters: With reﬂections on female conduct in the more important duties of life. Clifton: Kelley. World Bank. (2001). Engendering development enhancing development through attention to gender. Washington, DC: World Bank and Oxford University Press. World Development Indicators (WDI). (2008). World development indicators online (Accessed online at: http://publications.worldbank.org/WDI/). Washington, DC: World Bank. Young, I. M. (1997). Unruly categories: A critique of Nancy Fraser’s dual systems theory. New Left Review, 222, 147–160.

School Enrollment Rates by Country, Sex, and SES, among 10–19-Year-Olds. Socioeconomic Groups

All SES

2

3

4

5

6

7

8

Cameroon 1992 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.48 112 0.42 90 0.46 202 0.88

0.57 458 0.47 508 0.52 966 0.81

0.62 366 0.50 341 0.56 707 0.81

0.70 411 0.56 404 0.63 815 0.81

0.84 294 0.68 330 0.76 624 0.81

0.83 196 0.76 208 0.79 404 0.91

0.93 292 0.87 321 0.90 613 0.93

0.89 240 0.85 286 0.87 526 0.95

0.73 2370 0.63 2489 0.68 4859 0.87

Cameroon 1998 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.55 71 0.36 102 0.44 173 0.66

0.65 732 0.52 684 0.59 1416 0.79

0.58 380 0.40 389 0.49 769 0.69

0.68 304 0.53 298 0.60 602 0.78

0.71 572 0.68 584 0.69 1156 0.96

0.83 371 0.72 372 0.77 743 0.87

0.86 413 0.81 419 0.83 832 0.95

0.82 240 0.78 252 0.80 492 0.95

0.72 3101 0.62 3115 0.66 6216 0.86

Cameroon 2004 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.62 123 0.53 107 0.58 230 0.86

0.79 1322 0.64 1341 0.71 2663 0.81

0.73 826 0.56 802 0.65 1628 0.77

0.76 1081 0.59 1021 0.68 2102 0.78

0.78 1046 0.76 1013 0.77 2059 0.97

0.85 753 0.79 825 0.82 1578 0.93

0.91 683 0.86 685 0.88 1368 0.95

0.92 467 0.78 478 0.85 945 0.85

0.80 6322 0.69 6290 0.75 12612 0.86

29

1

Sex versus SES

APPENDIX Table A1.

30

Table A1.

(Continued)

Socioeconomic Groups

All SES

2

3

4

5

6

7

8

Burkina Faso 1993 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.11 90 0.05 83 0.08 173 0.43

0.16 1759 0.08 1650 0.12 3409 0.48

0.38 495 0.24 451 0.31 946 0.64

0.49 1065 0.35 979 0.42 2044 0.72

0.64 133 0.42 143 0.53 276 0.66

0.73 199 0.56 237 0.64 436 0.76

0.74 117 0.55 187 0.62 304 0.75

0.74 253 0.58 323 0.65 576 0.78

0.37 4134 0.27 4077 0.32 8211 0.72

Burkina Faso 1998 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.10 123 0.02 125 0.06 248 0.16

0.16 2113 0.08 2124 0.12 4237 0.51

0.26 290 0.22 284 0.24 574 0.86

0.38 802 0.28 812 0.33 1614 0.74

0.67 52 0.42 67 0.53 119 0.62

0.72 147 0.57 197 0.64 344 0.80

0.71 133 0.51 164 0.60 297 0.72

0.70 135 0.48 216 0.56 351 0.68

0.28 3805 0.20 4004 0.24 7809 0.71

Burkina Faso 2003 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.12 825 0.06 674 0.09 1499 0.53

0.19 3240 0.13 2828 0.16 6068 0.67

0.31 625 0.26 595 0.28 1220 0.82

0.33 1951 0.26 1715 0.30 3666 0.78

0.51 170 0.51 165 0.51 335 0.99

0.61 482 0.53 561 0.57 1043 0.87

0.76 205 0.57 313 0.65 518 0.75

0.62 277 0.52 347 0.57 624 0.84

0.29 7779 0.24 7205 0.27 14984 0.83

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

1

0.61 1050 0.48 1055 0.54 2105 0.78

0.64 801 0.54 937 0.58 1738 0.83

0.76 760 0.66 802 0.71 1562 0.86

0.81 502 0.69 516 0.75 1018 0.86

0.86 131 0.70 159 0.78 290 0.82

0.87 277 0.76 297 0.81 574 0.87

0.79 265 0.74 281 0.76 546 0.93

0.70 4341 0.58 4656 0.63 8997 0.83

0.54 359 0.43 313 0.49 672 0.78

0.54 1382 0.46 1402 0.50 2784 0.85

0.59 1126 0.48 1169 0.53 2295 0.81

0.71 884 0.62 819 0.67 1703 0.87

0.80 157 0.68 180 0.73 337 0.85

0.83 508 0.73 562 0.78 1070 0.87

0.90 244 0.81 313 0.85 557 0.90

0.84 148 0.77 180 0.80 328 0.92

0.65 4832 0.56 4959 0.60 9791 0.86

Zambia 2001 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.56 304 0.42 303 0.49 607 0.75

0.63 1382 0.48 1337 0.55 2719 0.76

0.69 1119 0.57 1116 0.63 2235 0.83

0.71 822 0.63 754 0.67 1576 0.88

0.80 94 0.70 123 0.74 217 0.88

0.83 468 0.74 445 0.78 913 0.89

0.84 394 0.76 449 0.80 843 0.91

0.84 157 0.85 161 0.85 318 1.01

0.70 4762 0.59 4706 0.65 9468 0.85

Tanzania 1992 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.38 292 0.36 280 0.37 572 0.93

0.54 1935 0.46 1887 0.50 3822 0.85

0.54 1805 0.49 1780 0.51 3585 0.92

0.61 1151 0.57 1169 0.59 2320 0.95

0.62 210 0.56 223 0.59 433 0.89

0.75 8 0.59 17 0.64 25 0.78

0.66 82 0.57 82 0.62 164 0.87

0.76 96 0.52 110 0.63 206 0.68

0.55 5635 0.50 5609 0.52 11244 0.89

31

0.56 535 0.4 582 0.48 1117 0.72

Sex versus SES

Zambia 1992 Male enrollment N males Female enrollment N females Both sexes Total N FMER Zambia 1996 Male enrollment N males Female enrollment N females Both sexes Total N FMER

32

Table A1.

(Continued)

Socioeconomic Groups

All SES

2

3

4

5

6

7

8

Tanzania 1996 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.47 233 0.33 218 0.40 451 0.71

0.61 1371 0.55 1332 0.58 2703 0.89

0.57 1248 0.56 1171 0.57 2419 0.98

0.62 1169 0.62 1163 0.62 2332 1.00

0.74 231 0.58 284 0.65 515 0.79

1.00 20 0.62 26 0.78 46 0.62

0.69 84 0.64 104 0.66 188 0.92

0.79 76 0.60 88 0.69 164 0.76

0.61 4470 0.57 4427 0.59 8897 0.93

Tanzania 2004 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.56 142 0.44 172 0.49 314 0.77

0.66 1680 0.58 1618 0.62 3298 0.88

0.66 1193 0.63 1130 0.64 2323 0.95

0.73 2209 0.66 2126 0.69 4335 0.91

0.72 195 0.69 220 0.70 415 0.95

0.78 167 0.71 213 0.74 380 0.92

0.74 176 0.65 261 0.68 437 0.88

0.71 96 0.61 147 0.65 243 0.86

0.69 5883 0.63 5920 0.66 11803 0.91

Zimbabwe 1994 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.70 293 0.58 260 0.64 553 0.82

0.74 1224 0.68 1122 0.71 2346 0.91

0.78 1080 0.74 1020 0.76 2100 0.94

0.75 673 0.73 591 0.74 1264 0.96

0.71 241 0.60 284 0.65 525 0.84

0.79 198 0.69 239 0.73 437 0.87

0.86 158 0.68 219 0.76 377 0.79

0.75 179 0.68 190 0.72 369 0.91

0.76 4057 0.69 3932 0.72 7989 0.91

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

1

0.73 940 0.69 874 0.71 1814 0.94

0.76 1100 0.73 976 0.74 2076 0.97

0.78 678 0.75 660 0.76 1338 0.96

0.64 243 0.59 229 0.61 472 0.92

0.73 330 0.71 337 0.72 667 0.97

0.81 287 0.70 333 0.75 620 0.86

0.82 202 0.69 219 0.76 421 0.84

0.75 3965 0.71 3807 0.73 7772 0.95

Zimbabwe 2006 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.69 54 0.60 48 0.65 102 0.88

0.69 1407 0.66 1391 0.68 2798 0.95

0.76 1489 0.76 1370 0.76 2859 1.00

0.72 705 0.73 648 0.72 1353 1.02

0.72 229 0.60 249 0.65 478 0.84

0.75 636 0.73 630 0.74 1266 0.97

0.82 603 0.71 783 0.76 1386 0.86

0.78 365 0.73 413 0.75 778 0.93

0.74 5498 0.71 5544 0.72 11042 0.95

Uganda 1995 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.52 277 0.37 247 0.45 524 0.72

0.68 1601 0.51 1606 0.60 3207 0.75

0.64 728 0.44 762 0.54 1490 0.69

0.75 780 0.62 836 0.69 1616 0.83

0.73 255 0.59 364 0.65 619 0.81

0.83 144 0.60 228 0.69 372 0.73

0.75 40 0.69 55 0.72 95 0.92

0.81 149 0.64 192 0.71 341 0.80

0.69 3985 0.53 4305 0.61 8290 0.77

Uganda 2000 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.79 105 0.57 120 0.67 225 0.72

0.79 1008 0.70 1019 0.75 2027 0.88

0.80 980 0.69 1018 0.74 1998 0.86

0.83 1578 0.77 1699 0.80 3277 0.93

0.80 190 0.65 241 0.71 431 0.81

0.86 285 0.76 358 0.80 643 0.89

0.86 135 0.79 198 0.82 333 0.92

0.89 209 0.77 267 0.82 476 0.87

0.82 4517 0.73 4947 0.77 9464 0.89

33

0.66 166 0.69 156 0.68 322 1.04

Sex versus SES

Zimbabwe 1999 Male enrollment N males Female enrollment N females Both sexes Total N FMER

34

Table A1.

(Continued)

Socioeconomic Groups

All SES

2

3

4

5

6

7

8

Uganda 2005 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.33 168 0.35 180 0.34 348 1.05

0.71 917 0.67 984 0.69 1901 0.94

0.84 1466 0.76 1417 0.80 2883 0.90

0.85 2539 0.81 2522 0.83 5061 0.96

0.84 130 0.75 124 0.80 254 0.89

0.88 195 0.84 228 0.86 423 0.95

0.81 120 0.70 159 0.75 279 0.87

0.96 132 0.79 192 0.86 324 0.82

0.81 5704 0.76 5843 0.78 11547 0.93

Madagascar 1992 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.33 919 0.31 810 0.32 1729 0.95

0.34 1321 0.31 1234 0.32 2555 0.91

0.54 462 0.44 484 0.49 946 0.83

0.63 629 0.65 616 0.64 1245 1.02

0.80 268 0.73 319 0.77 587 0.92

0.86 165 0.78 178 0.82 343 0.91

0.80 46 0.76 66 0.78 112 0.94

0.77 112 0.79 119 0.78 231 1.03

0.48 3922 0.46 3826 0.47 7748 0.97

Madagascar 1997 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.26 94 0.22 87 0.24 181 0.85

0.32 1579 0.26 1510 0.29 3089 0.83

0.37 921 0.34 925 0.35 1846 0.94

0.52 771 0.52 688 0.52 1459 1.00

0.69 220 0.72 214 0.70 434 1.04

0.83 303 0.70 342 0.76 645 0.84

0.85 59 0.74 53 0.79 112 0.87

0.78 69 0.74 69 0.76 138 0.94

0.44 4022 0.41 3893 0.42 7915 0.92

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

1

0.47 1057 0.46 942 0.47 1999 0.97

0.48 560 0.50 534 0.49 1094 1.06

0.65 1036 0.64 941 0.65 1977 0.99

0.79 270 0.69 273 0.74 543 0.86

0.83 1053 0.78 1091 0.81 2144 0.94

0.84 189 0.70 195 0.77 384 0.83

0.83 207 0.74 217 0.78 424 0.89

0.64 4570 0.62 4396 0.63 8966 0.96

Ghana 1993 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.52 56 0.40 38 0.47 94 0.76

0.56 452 0.40 380 0.48 832 0.71

0.69 873 0.57 731 0.64 1604 0.83

0.74 461 0.65 367 0.70 828 0.88

0.76 310 0.63 344 0.69 654 0.83

0.84 141 0.68 167 0.75 308 0.82

0.80 188 0.67 245 0.73 433 0.84

0.81 95 0.59 106 0.69 201 0.72

0.70 2576 0.58 2378 0.64 4954 0.83

Ghana 1998 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.34 94 0.29 58 0.32 152 0.86

0.45 560 0.39 453 0.42 1013 0.87

0.68 632 0.58 552 0.63 1184 0.85

0.68 410 0.69 355 0.69 765 1.01

0.69 370 0.64 365 0.66 735 0.93

0.78 221 0.73 219 0.75 440 0.94

0.83 300 0.71 356 0.76 656 0.85

0.82 78 0.74 107 0.77 185 0.90

0.65 2670 0.60 2466 0.63 5136 0.92

Ghana 2003 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.43 110 0.34 61 0.40 171 0.81

0.50 519 0.46 427 0.48 946 0.92

0.68 382 0.60 375 0.64 757 0.88

0.67 941 0.64 752 0.66 1693 0.95

0.77 515 0.69 458 0.73 973 0.91

0.81 285 0.75 257 0.78 542 0.93

0.82 364 0.72 469 0.76 833 0.87

0.77 168 0.72 229 0.74 397 0.93

0.69 3294 0.64 3030 0.66 6324 0.93

35

0.38 192 0.37 197 0.37 389 0.96

Sex versus SES

Madagascar 2003 Male enrollment N males Female enrollment N females Both sexes Total N FMER

36

Table A1.

(Continued)

Socioeconomic Groups

All SES

2

3

4

5

6

7

8

Niger 1992 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.24 90 0.27 81 0.26 171 1.11

0.16 2136 0.06 2263 0.11 4399 0.40

0.36 317 0.20 329 0.28 646 0.57

0.44 486 0.29 511 0.36 997 0.65

0.55 119 0.34 113 0.44 232 0.62

0.62 183 0.48 185 0.55 368 0.77

0.66 172 0.66 181 0.66 353 1.00

0.60 245 0.55 232 0.58 477 0.91

0.30 3755 0.19 3906 0.25 7661 0.64

Niger 1998 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.16 73 0.07 101 0.11 174 0.42

0.16 2596 0.07 2685 0.11 5281 0.43

0.31 370 0.23 396 0.27 766 0.75

0.44 369 0.28 432 0.35 801 0.64

0.54 121 0.42 156 0.47 277 0.79

0.74 172 0.60 183 0.66 355 0.81

0.78 131 0.70 148 0.74 279 0.90

0.65 214 0.62 203 0.64 417 0.96

0.28 4061 0.19 4320 0.23 8381 0.67

Niger 2006 Male enrollment N males Female enrollment N females Both sexes Total N FMER

0.27 70 0.21 72 0.24 142 0.77

0.30 3293 0.17 3456 0.23 6749 0.58

0.43 226 0.33 234 0.38 460 0.78

0.51 485 0.33 477 0.42 962 0.64

0.54 230 0.46 228 0.50 458 0.85

0.63 335 0.58 347 0.60 682 0.91

0.79 291 0.77 300 0.78 591 0.98

0.79 358 0.70 344 0.74 702 0.88

0.41 5300 0.30 5474 0.35 10774 0.71

Note: FMER: Female-to-male enrollment ratio.

PARFAIT M. ELOUNDOU-ENYEGUE ET AL.

1

37

Sex versus SES

Table A2. Country

Summary Indices for the Momentum, Ubiquity, and Relative Importance of Gender and SES Convergence. Year Ubiquity Irreversibility Index (U) Index (I )

Index of Relative Importance Theil values (T )

R

Between SES(1) Between Sex(2) (2)/(1) Burkina Faso Burkina Faso Burkina Faso Cameroon Cameroon Cameroon Ghana Ghana Ghana Madagascar Madagascar Madagascar Niger Niger Niger Tanzania Tanzania Tanzania Uganda Uganda Uganda Zambia Zambia Zambia Zimbabwe Zimbabwe Zimbabwe

1993 1998 2003 1992 1998 2004 1993 1998 2003 1992 1997 2003 1992 1998 2006 1992 1996 2004 1995 2000 2005 1992 1996 2001 1994 1999 2006

0.62 0.71 0.82 0.88 0.81 0.90 0.89 1.01 1.03 0.93 0.91 1.11 0.45 0.46 0.62 1.13 1.04 0.99 0.91 0.97 1.12 0.84 0.92 0.81 1.05 1.12 1.08 0.91

– 0.5 0.88 – 0.25 0.63 – 1.00 0.50 – 0.25 0.50 – 0.63 0.88 – 0.63 0.50 – 1.00 0.75 – 0.63 0.75 – 0.75 0.50 0.64

0.31 0.29 0.28 0.21 0.11 0.14 0.13 0.10 0.07 0.70 0.31 0.18 0.56 0.68 0.60 0.09 0.04 0.04 0.12 0.05 0.04 0.18 0.28 0.15 0.09 0.07 0.08 0.21

0.05 0.04 0.04 0.07 0.06 0.10 0.10 0.02 0.01 0.20 0.03 0.02 0.20 0.22 0.30 0.05 0.03 0.03 0.10 0.05 0.01 0.11 0.11 0.09 0.08 0.05 0.05 0.08

0.15 0.12 0.14 0.32 0.50 0.68 0.77 0.17 0.19 0.28 0.10 0.13 0.36 0.32 0.51 0.53 0.69 0.68 0.83 0.92 0.32 0.61 0.40 0.59 0.95 0.70 0.61 0.48

Notes: – refers to cells where index could not be calculated due to lack of data from previous period groups to the FMER among the top two SES groups; irreversibility (I ) refers to percent of times when later FMER values are greater than earlier FMER values; and relative importance (R) refers to the ratio of sex-related inequality to SES-related inequality.

THE PEDAGOGY OF DIFFERENCE: UNDERSTANDING TEACHERS’ BELIEFS AND PRACTICE OF GENDER EQUITY IN BENIN Sonya Anderson MOVE TOWARD GENDER EQUALITY IN EDUCATION With women constituting more than two-thirds of the world’s approximately 800 million illiterate adults, the importance of gender and equality in education has never been clearer. The education of girls and women has long been associated with positive outcomes such as lower childbearing rates, improved health for women and their families, increased participation in household decision-making, and increased community participation (Wolf & Odonkor, 1997). Consequently, women’s literacy and numeracy are widely understood as critical to individual and national development (Floro & Wolf, 1990). With more than 60 million school-aged girls currently not enrolled in school, however, the status of gender equality in education remains troubling, particularly in sub-Saharan Africa. Many post-colonial education reforms in Africa have focused on increasing the supply of schools and the number of students and qualiﬁed teachers to ﬁll them (Moulton & Mundy, 2002). In addition, growing Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 39–88 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010005

39

40

SONYA ANDERSON

international attention to women’s development has resulted in concerted efforts by governments to increase educational equality for girls, particularly at the primary school level. Despite these efforts, attempts to improve educational equality for girls have been complicated by the recognition that increased access to education must be accompanied by improvements in the quality of girls’ experiences once they enter the classroom (Hyde, 1993). Accordingly, many African countries have designed activities to increase girls’ enrollment and retention by creating school environments that are ‘‘friendly’’ to girls (Tietjen, 1991). As a result, the push for girls’ education has expanded from a focus on increasing the quantity of girls in school to improving the quality of their educational experience. Girls’ education efforts in many African countries aim to foster greater levels of school retention and completion by supporting curricular, pedagogic, and infrastructural innovations that will lead to more gender-equitable learning environments (O’Gara & Kendall, 1996). Since independence from French colonial rule in 1960, the Republic of Benin has been plagued with an education system that has routinely underserved its school-aged population. In response, the Ministry of Primary and Secondary Education (MEPS) has undertaken ambitious reforms to increase participation in primary and secondary education – especially for rural and girl students. MEPS has supported efforts to raise teachers’ awareness of gender bias in the curricula and in their own interactions with students. Notably, MEPS developed a series of ‘‘equity in the classroom’’ (EIC) training modules to equip teachers with pedagogical strategies to foster gender-equitable, ‘‘girl-friendly’’ classrooms. However, the impact of these efforts remains unclear.

Literature on Girls in Sub-Saharan African Schools The literature on the classroom experiences of boys and girls in sub-Saharan African schools suggests that their realities are far from equal. Girls are commonly perceived as less academically motivated and capable than boys, and this perception often serves as the backdrop for the instructional decisions that teachers make. In Biraimah’s (1980) study of a secondary school in Togo, teachers were given a list of characteristics to apply to male or female students. Although girls were assigned qualities such as ‘‘lacks interest in school,’’ and ‘‘quiet/submissive,’’ boys were described as ‘‘responsible’’ and possessing ‘‘leadership qualities’’ (p. 200). Similarly, Davison and Kanyuka (1992) found that teachers in Malawi characterize

Pedagogy of Difference

41

girls as ‘‘shy’’ (p. 463), and 60% of the teachers interviewed concluded that ‘‘girls lack the ambition to work hard because they are lazy’’ (p. 463). Anderson-Levitt, Bloch, and Soumare´ (1998) documented similar attitudes in their study of primary and junior high schools in Guinea where teachers reported familiar stereotypes: boys learn their lessons well, are ambitious, and give good responses. In addition to the lower expectations that some teachers hold for their girl students, teachers may also devalue the importance of education for girls. In Davison and Kanyuka’s (1992) Malawi study, several teachers expressed the belief that girls beneﬁt less from education than boys since many will likely marry at the end of primary school and abandon their educational careers for a life of domesticity. Consequently, all but one of the 14 Malawian teachers surveyed advocated subjects such as home economics for girls, viewing them as ‘‘critical to girls’ education’’ (p. 456). Biraimah (1980) cites similar ﬁndings in her Togolese study saying, ‘‘When asked to list the most likely careers to be pursued by their female students, the . . . teachers relegated them to low-status jobs . . . ofﬁce worker, hairdresser, seamstress . . . ’’ (p. 201). Finally, when asked about the knowledge and skills necessary to prepare boys and girls for adult life, Mensch and Lloyd (1998) found that teachers believed that mastering mathematics was more important for boys than for girls (p. 178). Similarly, Davison and Kanyuka (1992) found that teachers viewed home economics as important for girls to master. Hyde (1993) also reports teachers’ belief that girls should take home economics and needlecraft, while boys should study woodwork. The gendered dynamics that teachers create through their implementation of the formal curriculum may also become intertwined with subtexts from the ‘‘hidden’’ curricula. For example, in her Nigerian primary school study, Biraimah (1989) found that while boys received more of the teacher’s time and attention, girls ‘‘were often assigned ‘housekeeping’ chores or received messages that underscored perceived intellectual inadequacies . . . ’’ (p. 59). This tendency to delegate domestic tasks to girls – particularly when such tasks take away from instructional time – underscores the perception of girls as tangential to the learning enterprise and potentially limits their training to a narrow set of pre-determined, gender-prescribed goals. The formulation of these gender-prescribed goals is aptly described by Madeline Arnot (2002) and her concept of gender codes. These codes result from the structures found in schools that serve to reproduce gender relations. Gender codes are bolstered by systems of gender classiﬁcation and gender framing that function to regulate the content of educational

42

SONYA ANDERSON

knowledge and its transmission from teacher to student. Speciﬁcally, gender classiﬁcation maintains power relations between the sexes by prescribing different domains of knowledge to boys and girls, whereas gender framing governs the pedagogical relationship between the teacher and the student and instructs students on the different expectations associated with their gendered identities. Through gender classiﬁcation and framing, the hierarchy between the two sexes is deﬁned, and the boundaries between appropriate behavior, activities, etc. for boys and girls are delineated. Consequently, teachers’ ideas about ‘‘good pupils’’ are tailored to incorporate gender-based distinctions that correspond to their expectations of girls’ and boys’ ability, comportment, and potential for educational success. Finally, Arnot describes the process of gender recontextualization whereby knowledge is framed so that ‘‘notions of appropriate behavior for each sex are converted into appropriate academic disciplines’’ (p. 10). As a result, the different depictions of masculinity and femininity found in textbooks and in the curricula play an integral role in the complex process of gender identity development.

EDUCATIONAL LANDSCAPE IN BENIN Public Primary Education MEPS is responsible for primary and secondary education in Benin. The primary cycle is comprised of six levels: Cours d’Initiation (CI), Cours Preparatoire (CP), Cours Elementaire 1 (CE1), Cours Elementaire 2 (CE2), Cours Moyen 1 (CM1), and Cours Moyen 2 (CM2). The primary education sector has experienced considerable growth over the past 10–15 years. However, despite these gains, Table 1 shows that geographic disparities exist among Benin’s 12 departments (Fig. 1), particularly in their ability to keep up with increased student enrollment – either by building additional schools or hiring additional teachers.1 Primary education in Benin faces a number of major challenges. Overcrowding, particularly at the CI and CP levels (ﬁrst and second grade), often results in classes with 50–100 students. To compound the situation, many schools do not offer the full 6 years of primary school, which limits students’ access to a complete primary education. In addition, access to quality education is limited by an inadequate supply of well-trained teachers. These challenges have created a system that is marred by low internal efﬁciency, with low promotion rates and high repetition and

Number of Public Primary Students, Teachers, and School During the 2001, 2002, and 2003 Academic Years, and the Percent Change in these Numbers between 2001 and 2003.

Department

Number of Students 2001

2002

2003

Atacora–Donga 136,114 151,626 156,273 Atlantique–Littoral 187,112 199,693 195,078 Borgou–Alibori 149,654 163,472 161,710 Mono–Couffo 199,174 218,579 225,227 Oue´me´–Plateau 194,474 208,274 192,251 Zou–Collines 83,461 223,357 217,863 Total 1,068,862 1,165,001 1,148,402

Number of Teachers Change (%) 14.8 4.3 8.1 13.1 1.1 161 7.4

2001

2002

2003

2,432 2,035 2,630 3,588 3,015 3,799 2,732 2,897 3,033 3,201 3,269 2,250 3,754 2,976 3,736 3,615 3,614 3,795 19,322 17,806 19,243

Pedagogy of Difference

Table 1.

Number of Schools

Change (%) 2001 2002 2003 Change (%) 8.1 5.9 11 29.7 .48 4.9 .41

667 731 786 651 668 693 731 763 783 669 713 753 733 756 729 805 875 898 4,256 4,506 4,642

17.8 6.5 7.1 12.6 .55 11.6 9.1

Source: Ministe`re des Enseignements Primaire et Secondaire (2002, 2003, 2004). Annuaire Statistique, Anne´e Scolaire -2001–2002, 2002–2003, and 2002–2004.

43

44

SONYA ANDERSON

Fig. 1.

Administrative Map of Benin.

dropout rates. However, as crippling as the barriers to educational quality are for all children, girls are especially disadvantaged since they are more often affected by the limited supply of schools, poorly trained teachers, and cultural or economic obstacles to their participation in schooling.

Challenge of Girls’ Education In Benin, the focus on girls’ education became central to education reform shortly after the country’s transition from Marxism–Leninism to democracy in 1990. During the years that followed, a number of studies investigated the factors that negatively affect girls’ school participation. From an economic perspective, school and material fees pose a signiﬁcant barrier to girls’ enrollment. Although MEPS waived school fees for rural girls in 1993, additional costs such as mandatory uniforms, school supplies, etc. continue

Pedagogy of Difference

45

to constrain many families. Moreover, the loss in revenue that schools experienced from the elimination of these fees had the unexpected effect of lowering enrollment rates in some areas, as principals turned girls away because their enrollment would bring no income (Debourou, 1995). Culturally, among those girls who do enroll in school, their promotion rates often lag behind boys’, with greater household responsibilities and early marriage or pregnancy often impeding their completion of the primary cycle (World Bank, 1994). From a structural perspective, the long distances between schools and many villages, coupled with inadequate school infrastructure, constrain girls’ participation in education. Parents are often reluctant to send their daughters to schools that require them to walk long distances or that do not have appropriate hygiene facilities for girls, particularly once the girls reach puberty. Other factors that discourage girls’ schooling include gender-biased teachers, high failure rates, and school policies that expel pregnant girls but not the schoolboy fathers (Service de Promotion de la Scolarisation des Filles, 2004). Finally at the political level, the insufﬁcient supply of schools, especially in rural parts of Benin; the relatively small number of female teachers and principals; and insufﬁcient data about the factors that inﬂuence parents’ attitudes toward girls’ education all conspire to create a macro-policy environment that undermines efforts to increase girls’ primary school participation (Ministe`re des Enseignements Primaire et Secondaire, Ministe`re de l’Enseignement Technique et de la Formation Professionnelle, & Ministe`re de la Culture, de l’Artisanat et du Tourisme, 2003). Despite the strides that Benin has made, the persistence of these barriers ensures that a quality and equitable education for girls remains a challenge. Table 2 shows that despite a general increase between 1990 and 2002, primary school gross enrollment rates (GER) are consistently lower for girls than for boys. The effects of girls’ under-enrollment are intensiﬁed by geography, with the northern rural departments of Atacora–Donga and Borgou–Alibori reporting the lowest rates in 1996 and 2002. Although there has been signiﬁcant progress over the past 10–15 years – for example, Mono–Couffo’s rise from the department with the lowest girls’ GER in 1990 (21%) to the department with the highest rate in 2002 (97%) – other departments have not fared as well. Atlantique–Littoral saw only a slight increase in girls’ enrollment between 1996 and 2002, and Borgou–Alibori fell from fourth to last place among the departments. These ﬁndings reﬂect the ongoing need to address the micro- and macro-level factors that negatively impact girls’ school participation.

46

SONYA ANDERSON

Table 2.

Evolution of Gross Enrollment Rates (%) by Department between 1990, 1996, and 2002.

Department

1990

Atacora–Donga Atlantique–Littoral Borgou–Alibori Mono–Couffo Oue´me´–Plateau Zou–Collines All departments

1996

2002

Boys

Girls

Total

Boys

Girls

Total

Boys

Girls

Total

47.41 72.48 42.41 50.90 84.62 68.10 62.13

21.30 51.43 24.16 21.07 46.28 37.38 35.72

35.36 62.14 33.76 37.45 66.58 53.66 49.68

61.72 110.68 50.34 95.69 101.01 87.22 84.58

33.3 80 31.93 47.55 59.95 53.44 51.97

48.28 95.51 41.52 72.28 81.38 70.86 68.84

103.77 107 76 147.24 119.58 119.51 110.46

65.55 85.69 58.04 96.95 80.4 82.57 78.10

84.87 96.47 67.15 121.71 100 100.85 94.33

Source: Ministe`re des Enseignements Primaire et Secondaire et al. (2003).

PEDAGOGY OF DIFFERENCE IN BENIN The research presented here seeks to explore some of the micro-level factors that shape girls’ experiences in the classroom in two ways: ﬁrst, by examining teachers’ beliefs about the girls and boys that they teach, and second, by investigating the relationship between these beliefs and teachers’ use of EIC teaching methodologies with their girl students. Although data are available for a similar investigation of teachers’ pedagogy with their boy students, the substantive area of interest for this study focuses on girls’ experiences in the classroom. Sample In June 2005, 324 CM1 (ﬁfth grade) teachers were surveyed from across all 12 of Benin’s departments. These teachers came from 36 (42%) of Benin’s 85 school districts (circonscriptions scolaires), and were selected using a random multi-site cluster sample, with sampling occurring at both the school and teacher levels. Normalized weights were applied to the data to counter the over- and under-sampling of teachers from circonscriptions scolaires of different sizes and to ensure that the characteristics of this sample more closely represented the true population of CM1 public school teachers. Table 3 reveals that the average teacher in this sample is approximately 34 years old, has approximately 8.6 years of teaching experience, and has received about 21 days of training in the current education reform, which includes training in gender equity. The sample also reﬂects a gender and

47

Pedagogy of Difference

Table 3. Variable AGE EXPER

Teacher and School Characteristics of the Sample. Description

Age of teacher Years of primary school teaching experience TRAIN Number of days of training in current education reform DIPLOMA Highest educational diploma or degree CERT Professional certiﬁcation TEACHSIZE Total number of teachers at current school SCHSIZE Total number of students at current school NGIRLS Number of CM1 girls currently taught NBOYS Number of CM1 boys currently taught

Observation

Mean (SE)

Mode

Range

324 323

34.01 (.811) 8.64 (.691)

42 2

38 30

322

21.13 (.757)

20

75

316

2.18 (.040)

2a

5

315 323

.859 (.084) 5.51 (.129)

0b 6

2 13

305

298.25 (10.69)

285

665

323

18.49 (.761)

11

47

323

28.24 (.874)

20

63

a

The modal value of 2 represents the junior secondary school diploma, the Brevet d’Etudes du Premier Cycle (BEPC). b The modal value of 0 indicates that teachers with no professional certiﬁcation are the single largest group of teachers in this sample.

geographic skew that is not uncommon among primary school teachers in Benin. The problem of under-qualiﬁed teachers highlighted earlier is also evident in the sample, with 265 (83.8%) of the 316 teachers who responded to this item holding only the Brevet d’Etudes du Premier Cycle (BEPC), a junior secondary school diploma. Additionally, approximately half of the sample (52.1%) holds no professional teaching certiﬁcation. Finally, the sample illustrates the problem of incomplete primary cycles and overcrowding in primary schools. The average school has slightly fewer than 6 teachers and almost 300 students. This indicates that, in a primary system based on a 6-year cycle, the ‘‘average’’ school does not offer the full 6 years in any given academic year; that teachers may be responsible for teaching more than one grade during any given year; or that children are combined in multi-grade classrooms in order to offer the full cycle where there are insufﬁcient teachers. Furthermore, the average 300 students per school yields 50–60 students per teacher, thus underscoring the problem of overcrowded classrooms.

48

SONYA ANDERSON

Survey Design Data were collected using a 218-item self-designed survey. To develop the instrument, an initial 25-item Likert-scale survey was piloted with 6 primary school teachers in Benin, followed by a focus group to generate feedback on the content, organization, and relevance of survey items. The literature on questionnaire design (Peterson, 2000; Rea & Parker, 1997; Sudman & Bradburn, 1982) was reviewed, as well as items from similar instruments used in other research on related topics. From there, a second survey was developed, with each item mapped to the relevant literature or research. Then, feedback on the content of the revised survey was solicited from individuals with experience working in West African schools, as well as from researchers with experience in survey design. The survey was revised again and translated into French. Finally, in Benin, a group of teachers provided detailed feedback on the content, organization, and length of the survey and offered contextual suggestions about terminology and interpretation, in order to ensure alignment with the Beninese dialect. The ﬁnal survey contained 218 items organized along three different question constructs (academic capability/performance; importance/usefulness of education; and management of non-academic behavior); a single outcome construct (self-reported practice of ‘‘EIC’’); and a number of demographic/ background questions.2 The ‘‘EIC’’ construct speciﬁcally interrogated teacher behavior across ﬁve categories: (1) learning styles and instructional interaction, (2) instructional support, (3) language patterns, (4) roles in the classroom, and (5) management of students’ time and space. These categories included items that examined the way that teachers create cooperative learning and/or leadership opportunities for students; pose questions and give feedback; praise or scold students; provide male and female role models for students; engage students in other non-academic tasks in the classroom, etc. Teachers’ perceptions of their boy and girl students were measured on a 1–7 scale, indicating the extent to which they agreed or disagreed with the survey items, or more speciﬁcally, the degree of negative or positive perception that they held of boys or girls. Very Negative 1

Negative

2

Somewhat Negative 3

Neutral

Somewhat Positive

Positive

Very Positive

4

5

6

7

Similarly, the outcome construct was measured on a scale ranging from 0 to 30, with higher values representing teachers’ more frequent

49

Pedagogy of Difference

implementation of EIC strategies across the ﬁve different categories. Never 0-6

Rarely

Sometimes

Usually

Always

7-12

13-18

19-24

25-30

Although this survey was thoughtfully designed and tested, it relies entirely on teachers’ self-reported data. As such, it faces the same veracity challenges that attitude surveys typically face (Sudman & Bradburn, 1982), since self-reporting is generally less reliable than direct observation. For example, widespread attention to girls’ education may have raised teachers’ awareness of the importance of this issue and possibly their perception of what the ‘‘right’’ answer should be. Teachers may overestimate the frequency with which they practice EIC strategies, or conversely, they may be reluctant to admit that, despite hours of intensive training, they do not regularly incorporate EIC strategies into their practice. Despite these limitations, this study contributes importantly to the discourse about gender, equality, and education in Africa. As such, it is intended to provide a quantitative complement to the existing qualitative research, particularly since much of this research is limited by the very nongeneralizable nature of qualitative research itself. The current study, with its use of a nationally representative sample, allows for ﬁndings that are generalizable to a larger population. Moreover, these ﬁndings underscore the importance of qualitative and quantitative methodologies to untangle the complex relationship between beliefs and behavior and to provide the scope and depth necessary to understand teachers’ behavior.

Measures Because of the large number of survey items, composite variables were formed to parsimoniously represent the question and outcome constructs for this study. Principal Components Analysis was applied to the items within the survey to identify substantive dimensions. Eigenvectors from the ﬁrst and second principal components were plotted to examine how items clustered together by dimension, and the speciﬁc items that comprised each dimension. Substantive relevance was then applied to determine which clusters and items would be included in the analyses. Finally, classical item analyses (Cronbach’s alpha-reliability estimation) were used to identify internally consistent groups of indicators that could be combined to reliably measure the relevant constructs.3

50

SONYA ANDERSON

This process yielded 10 composite variables – 8 predictors and 2 outcomes – that were used in subsequent analyses: Outcome B_EIC G_EIC Questions B_CAP B_PERFORM B_SUBJECTS B_NONAC G_CAP G_PERFORM G_SUBJECTS G_NONAC

Frequency of use of EIC strategies with boy students Frequency of use of EIC strategies with girl students Beliefs about boys’ academic capability Beliefs about boys’ academic performance Beliefs about the importance of particular subjects to boys’ education Beliefs about boys’ non-academic characteristics Beliefs about girls’ academic capability Beliefs about girls’ academic performance Beliefs about the importance of particular subjects to girls’ education Beliefs about girls’ non-academic characteristics

Since the survey solicited teachers’ beliefs about boys and girls on unipolar yet parallel sets of items, the composites of teachers’ perceptions were comparable, irrespective of the gender of the student. Three dummy variables were also used to represent the professional education certiﬁcates that teachers might have.4 NO_CERT CAP_CERT CEAP_CERT

Served as the reference category to indicate teachers with no certiﬁcation Teachers with the Certiﬁcat d’Aptitude Pe´dagogique (CAP) Teachers with the Certiﬁcat Ele´mentaire d’Aptitude Pe´dagogique (CEAP) Analyses

To conduct the descriptive analysis of the data, the mean values of the question variables for boys and girls were examined and then compared to indicate teachers’ ‘‘preference’’ for girls or boys.5 Then, multiple regression was used to examine the main effects of teachers’ beliefs about their girl students on their self-reported use of EIC methodologies with those girls. Interaction effects among the question variables and also between the

51

Pedagogy of Difference

question and control variables (e.g., gender and experience of the teacher) were tested to evaluate whether the question effects differed among subgroups of teachers. Though the substantive interest of this study focuses on teachers’ instructional relationship with their girl students, the relationship with boys was also examined. Despite statistically signiﬁcant differences in teachers’ perceptions of boys and girls across the four predictor domains, reported differences in teachers’ use of EIC methodologies with boys versus with girls are small. Teachers, on average, reported that they ‘‘usually’’ incorporate EIC strategies into their pedagogy, with a mean value of 22.97 with boy students and 22.89 with girls. The difference between these values, however, is small and not statistically signiﬁcant (t ¼ 1.16, p ¼ .253). Moreover, the differences in teachers’ use of the individual methodologies that comprise the EIC composite variable are quite small, with only the items measuring the frequency with which teachers call on boys or girls to answer questions (items 144 and 145) showing a statistically signiﬁcant difference (.058) at the .05 level (t ¼ 2.27, p ¼ .03).

TEACHERS’ BELIEFS AND PRACTICE OF GENDER EQUITY What Do Teachers Believe? Table 4 lists the average value of the composite variables that summarize teachers’ perceptions of girls and boys in each of the question domains Table 4. Variable CAP PERFORM SUBJECTS

NONAC

Weighted Mean Comparison of Teachers’ Perceptions of Boys and Girls (N ¼ 324). Description Teachers’ perceptions of academic capability Teachers’ perceptions of academic performance Teachers’ perceptions of subjects important to education Teachers’ perceptions of non-academic behavior

Bp r .10, *pr .05, **pr .01, ***pr .001.

Boys’ Mean (SE) Girls’ Mean (SE) Difference 4.28 (.071) 5.03 (.055) 5.54 (.059)

3.73 (.071) 4.47 (.058) 5.71 (.050)

4.43 (.060)

4.82 (.049)

.552*** .555*** .170***

.392***

52

SONYA ANDERSON

measured in this survey. As previously noted, these perceptions represent teachers’ self-reported beliefs and were not substantiated by objective measures such as grades, test scores, etc. Academic Capability Teachers reported, on average, a statistically signiﬁcant more favorable perception of boys’ academic capability than girls’. The difference in teachers’ opinions about boys and girls is most marked with respect to the frequency of their classroom participation, and their general academic ability. Teachers believe that boys experience more academic success than girls, with an almost 1-point difference in their perceptions (4.81 versus 3.93). Moreover, teachers also note differences in boys’ and girls’ ability in speciﬁc subjects, with the largest difference occurring in mathematics – both in terms of how easily students learn mathematics (4.09 versus 3.62) and the perception of the best math student (4.05 versus 3.02). The second largest difference occurs in science – again, with boys viewed more favorably with learning science (4.11 versus 3.73) and doing well in it (3.87 versus 3.30). Given the popular perception that math and science are more well-suited to boys, the size of these gaps in perception is not inconsequential. Academic Performance Overall, teachers view boys’ academic performance more positively than girls’ (5.03 versus 4.47). They believe that they have to spend less time explaining classroom lessons to boys (4.77 versus 4.39), that boys want to answer questions more frequently in class (5.56 versus 4.52), that they are better at answering difﬁcult questions than girls (5.05 versus 4.55), and that they answer these questions incorrectly less often than girls (5.07 versus 4.72). Importantly, teachers also believe that boys take their education more seriously than girls – an observation underscored by the full 1-point difference in their ratings (4.85 versus 3.81). This perception is evident in the literature which has found that teachers believe that girls ‘‘lack interest in school’’ (Biraimah, 1980, p. 200). The perception of girls as less serious students is compounded by the belief that girls complete their homework assignments less often than boys (4.36 versus 4.89). This perception does not take into consideration the considerable research examining the inverse relationship between girls’ household responsibilities – which often surpasses boys’ – and the time left for them to study, complete homework, etc. (Tietjen, 1991).

Pedagogy of Difference

53

Most Important Subjects In general, teachers hold more positive perceptions of the importance of SUBJECTS to girls’ education (5.71) than to boys’ (5.54).6 They believe that advanced math (5.14 versus 4.93), accounting (5.77 versus 5.56), and French (6.19 versus 5.99) are more important for girls to study than boys, whereas biology (5.71 versus 5.70) and chemistry (5.21 versus 5.17) are more important for boys; however, the difference between the ratings for boys and girls for these two subjects is not statistically signiﬁcant. Despite teachers’ seeming acceptance that certain subjects are equally if not more important to girls’ education as boys’, this belief shifts in the face of more vocational and/or domestic studies such as typing, cooking, sewing, and mechanics. For example, while cooking ranks seventh for boys on the list of 11 subjects, it ranked third for girls. Similarly, mechanics placed ﬁfth for boys and eleventh for girls. Although the rankings for typing (10 for boys and 8 for girls) and sewing (8 for boys and 7 for girls) are nearly equal, the difference between the actual values for each of these two subjects is nearly twice the size of the next largest one (advanced math). Thus, such large differences in teachers’ beliefs about the importance of boys’ and girls’ learning to cook, sew, type, and practice mechanics seem to follow traditional gender divisions, with domestic and secretarial skills identiﬁed as important to girls’ education and mechanics as important to boys’.

Non-Academic Characteristics In general, teachers have a more positive opinion of girls’ non-academic characteristics than boys’ (4.82 versus 4.43).7 Teachers ﬁnd girls to be more trustworthy (4.88 versus 4.59) and cooperative (5.34 versus 4.96) than boys; however, consonant with earlier ﬁndings, teachers believe that boys are tardy (4.92 versus 4.33) and absent from school less (4.72 versus 4.40) often than girls. The biggest difference in teachers’ opinions about boys and girls are noisiness (5.01 versus 3.36) and ease of discipline (5.15 versus 3.96). In both cases, perceptions of girls are a full 1-point higher than boys. Ironically, though teachers may believe that the description of girls in this way is complimentary, such images may, in fact, negatively stereotype girls. Miske and VanBelle-Prouty (1997) caution that the ‘‘good girl’’ archetype may ultimately reinforce girls’ invisibility in the classroom by reducing the amount of teacher attention that they receive. Boys, in contrast, through their boisterousness and assertiveness, receive increasing amounts of instructional attention as they progress through the schooling cycle (Biraimah, 1989).

54

SONYA ANDERSON

How Do Teachers Teach? At its heart, this study seeks to examine whether teachers’ beliefs about girls impact their use of EIC strategies with their girl students. Table 5 presents a subset of the ﬁnal ﬁtted multiple regression models, illustrating the main effects of the four question variables (academic capability, academic performance, most important subjects, and non-academic characteristics) and the teacher certiﬁcation variables on the outcome, teachers’ use of EIC strategies. The model also shows the effects of a two-way interaction between two of the question variables.8 Note that in the listed ﬁnal model, Table 5. Parameter Estimates (Standard Errors) and Approximate p-Values from a Nested Taxonomy of Fitted Regression Models Testing the Main and Interaction Effects of Teachers’ Professional Certiﬁcation and Teachers’ Perceptions of Girls’ Academic Capability, Academic Performance, Subjects Important to Girls’ Education, and Girls’ Non-Academic Behavior on the Frequency of Teachers’ Use of EIC Strategies in their Practice (N ¼ 324).

Intercept CEAP certiﬁcation CAP certiﬁcation

Model 1

Model 2

Model 3

23.22*** (.294) 1.48* (.629) .424 (.364)

19.75*** (2.14) 1.44* (.638) 4.68 (.346) .212 (.229) .139 (.254) .411* (.198) .199 (.262)

10.95* (4.52) 1.51* (.596) .479 (.349) .225 (.231) 1.77B (.976) 1.97* (.743) .225 (.254) 3.45* (1.70)

Academic capacity Academic performance Subjects important to education Non-academic characteristics Interaction between performance and subjects Summary statistics

GLH tests: CEAP_CERT and CAP_CERT

R2 F(df) p of F F p of F

Bp r .10, *pr .05, **pr .01, ***pr .001.

.019 2.79 (2, 34) .075 2.79 .076

.047 1.71 (6, 30) .151 2.60 .089

.059 2.85 (7, 29) .022 3.19 .054

Pedagogy of Difference

55

Model 3, the predictors representing academic capability and non-academic behavior were retained because of their substantive interest to the research, though they were not statistically signiﬁcant at the .05 or the .10 levels. Although not statistically signiﬁcant, the dummy predictor representing the CAP teaching certiﬁcate was also retained so that the effect of the CEAP certiﬁcate could be examined solely against the effects of no certiﬁcate at all. The ﬁndings from the ﬁnal model, represented in Figs. 2 and 3, show that the effects of teachers’ perceptions of girls’ academic performance (G_PERFORM) on their practice of EIC (G_EIC) vary, depending on how teachers feel about the subjects that they believe are important for girls to learn (G_SUBJECTS). Because of the statistically signiﬁcant interaction between G_SUBJECTS and G_PERFORM in the ﬁnal model, ﬁndings are reported separately for two groups of teachers – those who believe that math, biology, chemistry, etc. (G_SUBJECTS) are not important to girls’ education (Low G_SUBJECTS) versus those who believe that they are (High G_SUBJECTS).

School Subjects Not Important (Low G_SUBJECTS) In Fig. 2, ﬁtted Plot A shows that teachers who do not believe that math, biology, chemistry, etc. are particularly important to girls’ education (‘‘Low G_SUBJECTS’’ teachers), on average, use EIC teaching strategies less often with girls whom they perceive as low performing and more often with girls whom they perceive as high performing. Consequently, girls whom teachers believe can perform well academically receive the beneﬁts of EIC strategies more often than their lesser-regarded counterparts. This occurs despite the fact that teachers may not attach much importance to the subject matter itself. More importantly, Plot A demonstrates that when teachers deem that certain subjects are not important to girls’ education and they believe that girls are not strong academic performers, they incorporate EIC methodologies into their practice the least often of all subsets of teachers. Consequently, girls who are perceived as lower performing are instructed in ways that may not maximize their full classroom participation.

School Subjects Important (High G_SUBJECTS) In contrast, Plot B shows that teachers who do believe that math, biology, chemistry, etc. are important to girls’ education (‘‘High G_SUBJECTS

56

SONYA ANDERSON

Plot A: Low G_SUBJECTS Teachers 30 Low G_SUBJECTS

27 24

G_EIC

21 18 15 12 9

No Certification Certificat d’Aptitude Pédagogique (CAP) Certificat Elémentaire d’Aptitude Pédagogique (CEAP)

6 3 0

1

2

3

4

5

6

7

G_PERFORM Plot B: High G_SUBJECTS Teachers 30 27 24

G_EIC

21

High G_SUBJECTS

18 15 12 9 No Certification Certificat d’Aptitude Pédagogique (CAP) Certificat Elémentaire d’Aptitude Pédagogique (CEAP)

6 3 0 1

2

3

4

5

6

7

G_PERFORM

Fig. 2. The Effect of G_PERFORM on G_EIC for Teachers with Low or High G_SUBJECTS Scores, by Teachers’ Level of Professional Certiﬁcation.

57

Pedagogy of Difference 30 Low G_SUBJECTS

27 24

G_EIC

21

High G_SUBJECTS

18 15 12 9 6 3 0 1

2

3

4

5

6

7

G_PERFORM No Certification Certificat d’Aptitude Pédagogique (CAP) Certificat Elémentaire d’Aptitude Pédagogique (CEAP)

Fig. 3.

The Interaction Effect of G_PERFORM and G_SUBJECTS on G_EIC, by Teachers’ Level of Professional Certiﬁcation.

teachers’’) incorporate EIC strategies into their teaching more often with girls who are perceived as low-performers. As the opinion of performance begins to improve, however, the use of EIC strategies becomes less frequent. This ﬁnding may reﬂect to teachers’ commitment to ensuring that academically weaker girls master particular subject content – hence their more frequent use of strategies to increase girls’ classroom participation and learning. However, teachers may feel less urgency to use EIC strategies with girls perceived as academically stronger and who will likely master the subject matter easily, regardless of the pedagogical approach. This explanation is speculative, but it does offer one possible explanation for the negative relationship between G_EIC and G_PERFORM for High SUBJECTS teachers.

Comparing Low and High G_SUBJECTS Teachers Fig. 3 illustrates the difference between the use of EIC strategies between the two groups of teachers. The plot reveals that when the perception of girls’

58

SONYA ANDERSON

academic performance (G_PERFORM) is low, teachers who view math, biology, chemistry, etc. (G_SUBJECTS) as important to girls’ education tend to incorporate EIC strategies into their practice more frequently than those teachers who do not. Interestingly, this pattern changes when opinions about girls’ performance begin to approach the top of the perceptual range. Once the perception of girls’ performance reaches approximately 5.5 or higher, teachers who score low on G_SUBJECTS suddenly outperform their high G_SUBJECTS counterparts, using EIC strategies, on average, more often than teachers with high G_SUBJECTS scores. Fig. 3 also reveals that the difference in the frequency with which the two groups of teachers use EIC strategies is greater when the perception of girls’ academic performance is low (i.e., low values of G_PERFORM) than when it is high. Consequently, the difference between teachers in their the use of EIC strategies for high performing girls is relatively small, regardless of whether or not those teachers believe in the importance of the SUBJECTS variable.

Effects of Teacher Certiﬁcation Figs. 2 and 3 illustrate the rather surprising ﬁnding that professional certiﬁcation seems to have a negative impact on teachers’ use of EIC strategies with their girl students. Of the three categories used to measure certiﬁcation in this study, teachers who reported having no certiﬁcate whatsoever are predicted to include EIC methodologies in their practice the most often. Teachers with the Certiﬁcat d’Aptitude Pe´dagogique (CAP) – the higher of the two teaching certiﬁcates – fall only slightly behind their non-certiﬁed colleagues in their practice of EIC; however, the difference is not large enough to be statistically signiﬁcant. However, the frequency with which teachers who hold the Certiﬁcat Ele´mentaire d’Aptitude Pe´dagogique (CEAP) – the lower teaching certiﬁcate – incorporate EIC strategies is, on average, 1.51 points lower than the frequency score of non-certiﬁed teachers – a difference that is statistically signiﬁcant. This ﬁnding poses an interesting problem since the milestones that must be accomplished in order to acquire certiﬁcation are generally assumed to expand teachers’ professional knowledge base and generally improve teacher practice. One possible explanation for non-certiﬁed teachers’ more frequent use of EIC strategies may be that these teachers are more receptive to the training provided by MEPS, as they seek to enhance their pedagogy and align more closely to their professionally certiﬁed colleagues. Nevertheless, this negative

Pedagogy of Difference

59

relationship between professional certiﬁcation and the implementation of EIC strategies uncovers a puzzle for additional research.

IMPLICATIONS FOR RESEARCH, POLICY, AND PRACTICE Revisiting the Literature on Girls’ Education The ﬁndings of this research resonate with the research on gender and schools presented earlier in this chapter. Moreover, the previous discussion of Arnot’s (2002) theories of gender classiﬁcation, framing, and recontextualization are replete within this study. Teachers’ classiﬁcation of educational knowledge along gender lines is evident in their different assessments of the academic subjects important for boys to learn versus those perceived as important for girls. While teachers generally believe that the study of a wide range of subjects is more important to girls’ education than to boys’, they also believe that girls should be schooled in those subjects most closely associated with female work. The large and statistically signiﬁcant difference between teachers’ attitudes about girls’ versus boys’ study of cooking, sewing, and typing reﬂects a clear classiﬁcation and delineation of these subjects by gender. Similarly, the priority that teachers place on boys’ study of mechanics represents a comparable perception of this subject as one that is more appropriate for boys. The gender recontextualization represented by this classiﬁcation of educational knowledge according to sex is consonant with other ﬁndings from the literature on gender dynamics in sub-Saharan African schools. In addition, this study also found that teachers practice gender framing to the extent that they perceive boys as better students than girls. The statistically signiﬁcant difference between teachers’ perceptions of boys’ and girls’ academic capability and performance underscores their more favorable opinion of boys in these two areas. Here again, these views reﬂect the ﬁndings of researchers such as Anderson-Levitt et al. (1998), Biraimah (1980), and Mensch and Lloyd (1998) who found that teachers often hold low regard for girls’ academic ability and low expectations for their academic success. Moreover, teachers’ perception of girls as less noisy, less unruly, and more cooperative than boys may put girls at risk of being overlooked during classroom activities. Although teachers reported that they use EIC with boys (22.85) and girls (22.75) almost equally, Davison and

60

SONYA ANDERSON

Kanyuka (1992) caution that in academic settings where ‘‘boys have been socialized from birth to be verbal and assertive and girls to be submissive and quiet, it is more often the case that the boys will dominate’’ (p. 457). Interestingly, the fact that teachers report that they spend more time explaining lessons to boys than they do to girls (items 8 and 9); they ﬁnd it easier to teach boys than girls (items 47 and 48), and they believe that boys are naturally smarter in school than girls (items 1 and 13) may belie their later reports of near-parity between their practice of EIC toward boys and their practice toward girls.

FINAL THOUGHTS In Benin and elsewhere, efforts to spur educational reform focus heavily on changing teacher practice. One of the great puzzles of educational reform, however, is the question of when and why do teachers adopt pedagogical change (McLaughlin, 1998). The answers lie, in part, in understanding the individual factors that encourage teachers’ pedagogical experimentation (Richardson, 1990). Among these individual factors, the role of teachers’ beliefs about instruction, student ability, etc. is well documented (Flores, 2001; Graham, Harris, MacArthur, & Fink, 2001). These beliefs bring an affective component to teaching that can inﬂuence behavior even more than objective knowledge (Ethell & McMeniman, 2002). Moreover, these beliefs may act as a ﬁlter through which teachers make decisions about pedagogy (Ernst, 1989; Pajares, 1992; Tobin, 1987). Theories about the role of beliefs in pedagogical decision-making are important to the present study since efforts to raise awareness about gender equity in school may conﬁrm or challenge teachers’ existing beliefs about boys’ and girls’ ability to succeed in an academic setting. Although teachers receive training in ‘‘EIC’’ and other such gender-equitable strategies, it is uncertain whether they uniformly return to their schools at the end of the training period committed and able to remedy the subtle ways that gender bias functions in classroom life. This study is but a single contribution to what should be a larger research effort to improve girls’ experiences in schools. Additional research should not only focus on the personal factors (such as beliefs) that inﬂuence teacher practice, but should examine the impact of structural and institutional factors as well, in order to understand the full spectrum of complex factors that affect teachers’ pedagogical decision-making, and to advance the efforts

61

Pedagogy of Difference

of educators, researchers, and policy makers endeavoring to ensure equality of access and outcomes for girls.

NOTES 1. In 1999, Benin split each of its existing 6 geographic regions (known as departe´ments) in half to create 12 departments found today. These departments are often grouped in the following pairs: Atacora–Donga, Atlantique–Littoral, Bourgou–Alibori, Mono–Couffo, Oue´me´–Plateau, and Zou–Collines. 2. The full version of this survey is presented in Appendix 1. 3. Appendix 2 lists the individual items contained in each composite along with their alpha-reliability estimates. 4. Between the CEAP and the CAP, the CEAP is the lower of the two certiﬁcates. For both certiﬁcates, teachers must enroll in a teacher training college and complete a practicum. However, teachers entering the college with a terminal baccalaureate degree are eligible to obtain the CAP through a shorter program of study and practicum than teachers who enter with a junior secondary school diploma only (BEPC). For these teachers, the study and practicum period is longer and results in the CEAP. These teachers must also complete an additional (minimum) 3 years of training during the course of their teaching in order to advance from CEAP to CAP certiﬁcation. 5. Please see Appendix 3 for the mean value, standard error, and difference in teachers’ perceptions of boys and girls across each of the four question composite variables, as well as the individual items that comprise these variables. 6. The SUBJECTS composite variable includes basic math, advanced math, typing, accounting, chemistry, French, sewing, cooking, mechanics, biology, and computer skills. 7. The NONAC composite variable includes: time management, tardiness, noisiness, absenteeism, ease of discipline, trustworthiness, and cooperativeness. 8. These models are a small subset of the regression models that were ﬁtted to explore the relationship among the question, control, and outcome variables.

REFERENCES Anderson-Levitt, K. M., Bloch, M., & Soumare´, A. (1998). Inside classrooms in Guinea: Girls’ experiences. In: M. Bloch, J. Beoku-Betts & B. Tabachnick (Eds), Women and education in sub-Saharan Africa: Power, opportunities, and constraints (pp. 99–129). Boulder, CO: Lynne Rienner. Arnot, M. (2002). Reproducing Gender? Essays on educational theory and feminist politics. London: Routledge-Falmer. Biraimah, K. (1989). The process and outcomes of gender bias in elementary schools: A Nigerian case. Journal of Negro Education, 58(1), 50–67. Biraimah, K. C. (1980). The impact of Western schools on girls’ expectations: A Togolese case. Comparative Education Review, 24(2), 196–208.

62

SONYA ANDERSON

Davison, J., & Kanyuka, M. (1992). Girls’ participation in basic education in Southern Malawi. Comparative Education Review, 36, 446–466. Debourou, D. M. (1995). The case of Benin. In: Development of African Education Biennial (Ed.), Formulating education policy: Lessons and experience from sub-Saharan Africa (pp. 39–61). Paris: Association for the Development of African Education. Ernst, P. (1989). The knowledge, beliefs and attitudes of the mathematics teacher: A model. Journal of Education for Teaching, 15(1), 13–33. Cited in A. Wilcox-Herzog (2002). Is there a link between teachers’ beliefs and behaviors? Early Education and Development, 13(1), 81–106. Ethell, R., & McMeniman, M. (2002). A critical ﬁrst step in learning to teach: Confronting the power and tenacity of student teachers’ beliefs and preconceptions. In: C. Sugrue & C. Day (Eds), Developing teachers and teaching practice: International research perspectives (pp. 216–234). London: Routledge-Falmer. Flores, B. B. (2001). Bilingual education teachers’ beliefs and their relation to self-reported practices. Bilingual Research Journal, 25(3), 275–299. Floro, M., & Wolf, J. (1990). The economic and social impacts of girls’ primary education developing countries. Advancing Basic Education and Literacy Project. US Agency for International Development. Graham, S., Harris, K. R., MacArthur, C., & Fink, B. (2001). Primary grade teachers’ theoretical orientations concerning writing instruction: Construct validation and a nationwide survey. Contemporary Educational Psychology, 27, 147–166. Hyde, K. (1993). Barriers to educational opportunity in Malawi. In: E. King & M. A. Hill (Eds), Women’s education in developing countries: Barriers, beneﬁts, and policies (pp. 131–147). Baltimore: Johns Hopkins Press. McLaughlin, M. (1998). Listening and learning from the ﬁeld: Tales of policy implementation and situated practice. In: A. Hargreaves, A. Lieberman, M. Fullan & D. Hopkins (Eds), International Handbook of Educational Change (pp. 70–84). Norwell, MA: Kluwer Academic Publishers. Mensch, B., & Lloyd, C. (1998). Gender differences in the schooling experiences of adolescents in low-income countries: The case of Kenya. Studies in Family Planning, 28, 167–184. Ministe`re des Enseignements Primaire et Secondaire, Ministe`re de l’Enseignement Technique et de la Formation Professionnelle, & Ministe`re de la Culture, de l’Artisanat et du Tourisme. (2003). Plan d’actions national du Benin pour la mise en oeuvre du programme Education Pour Tous. Miske, S., & VanBelle-Prouty, D. (1997). Schools are for girls too: Creating an environment of validation. U.S. Agency for International Development, Ofﬁce of Sustainable Development, Technical Paper #41. Africa Bureau Information Center, Washington, DC. Moulton, J., & Mundy, K. (2002). Introduction: Implementation research and educational reform in sub-Saharan Africa. In: J. Moulton, K. Mundy, M. Welmond & J. Williams (Eds), Education reforms in sub-Saharan Africa: Paradigm lost? (pp. 1–11). Westport, CT: Greenwood Press. O’Gara, C., & Kendall, N. (1996). Beyond enrollment: A handbook for improving girls’ experiences in primary education, 1996, Advancing Basic Education and Literacy Project. US Agency for International Development. Pajares, M. F. (1992). Teachers’ beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62, 307–322. Peterson, R. A. (2000). Constructing effective questionnaires. Thousand Oaks, CA: Sage.

Pedagogy of Difference

63

Rea, L. M., & Parker, R. A. (1997). Designing and conducting survey research: A comprehensive guide. San Francisco, CA: Jossey-Bass Inc. Richardson, V. (1990). Signiﬁcant and worthwhile change in teaching practice. Educational Researcher, 19, 10–18. Service de Promotion de la Scolarisation des Filles. (2004). La service de promotion de la scolarisation des ﬁlles. Ministe`re des Enseignements Primaire et Secondaire, Direction de la Programmation et de la Prospective. Sudman, S., & Bradburn, N. M. (1982). Asking questions: A practical guide to questionnaire design. San Francisco: Jossey-Bass Inc. Tietjen, K. (1991). Educating girls: Strategies to increase access, persistence, and achievement. Advancing Basic Education and Literacy Project. US Agency for International Development. Tobin, K. (1987). Forces which shape the implemented curriculum in high school science and mathematics. Teaching and Teacher Education, 3(4), 287–298. Wolf, J., & Odonkor M. (1997). How educating a girl changes the woman she becomes: An intergenerational study in Northern Ghana. Technical Paper No. 8. Advancing Basic Education and Literacy Project. US Agency for International Development. World Bank. (1994). Staff appraisal report: Republic of Benin, education development project. Population and Human Resources Division. Occidental and Central Africa Department.

APPENDIX 1. TEACHER ATTITUDES SURVEY Thank you for your participation in this study. The following survey asks you to describe your attitudes about children’s education and your pedagogic practices. Please read the directions for each set of questions and answer each question as best you can. There are no right or wrong answers; I am interested in your honest opinion. Please be assured that your answers will be held conﬁdential and that you will never be named personally in any report or publication.

PART 1: STUDENTS’ ACADEMIC CAPABILITY AND PERFORMANCE 1. Below are statements about students’ academic capability. Please put an ‘‘x’’ in the box that represents the extent to which you agree or disagree with each one. Please choose only 1 response for each statement.

1.

In general, boys are naturally smart in school.

2.

Boys can do well in all school subjects.

3.

Boys cannot master some school subjects very well.

4.

Science is easy for boys to learn.

5.

Math is easy for boys to learn.

6.

French is easy for boys to learn.

7.

The girls in my class do not have the intellectual capacity to pass the CEP.

8.

I spend a lot of time explaining classroom lessons to boys.

9.

I spend a lot of time explaining classroom lessons to girls.

10.

French is easy for girls to learn.

11.

Science is easy for girls to learn.

12.

Math is easy for girls to learn.

13.

In general, girls are naturally smart in school.

14.

Girls can do well in all school subjects.

15.

Girls cannot master some school subjects very well.

16.

The boys in my class do not have the intellectual capacity to pass the CEP.

Strongly Disagree

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

2. Below are statements about students’ academic performance. Please put an ‘‘x’’ in the box that represents the extent to which you agree or disagree. Choose only 1 response per statement. In my current CM1 class . . .

17.

Girls are usually the best students in math.

18.

Girls are usually the best students in French.

19.

Girls are usually the best students in science.

20.

Boys most often answer questions correctly in class.

21.

Boys most often answer questions incorrectly in class.

22.

Boys most often want to answer questions in class.

23.

Boys least often want to answer questions in class.

24.

Boys do not take their education very seriously.

25.

Boys’ interest in education decreases as they get older.

26.

Boys usually cannot answer difﬁcult questions well.

Strongly Disagree

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

27.

Girls usually cannot answer difﬁcult questions well.

28.

Girls are usually very academically successful.

29.

Girls do not usually complete their homework assignments.

30.

When girls do not volunteer to answer questions in class, it is usually because they are shy.

31.

When girls do not volunteer to answer questions in class, it is often because they do not know the right answer.

32.

When girls do not volunteer to answer questions in class, it is often because they are not paying attention.

33.

Boys are usually very academically successful.

34.

Boys do not usually complete their homework assignments.

35.

When boys do not volunteer to answer questions in my class, it is often because they are shy.

Strongly Disagree

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

36.

When boys do not volunteer to answer questions in class, it is often because they do not know the right answer.

37.

When boys do not volunteer to answer questions in class, it is often because they are not paying attention.

38.

Girls most often answer questions correctly in class.

39.

Girls most often answer questions incorrectly in class.

40.

Girls most often want to answer questions in class.

41.

Girls least often want to answer questions in class.

42.

Boys are usually the best students in math.

43.

Boys are usually the best students in French.

44.

Boys are usually the best students in science.

45.

Girls do not take their education very seriously.

46.

Girls’ interest in education decreases as they get older.

47.

In general, I ﬁnd it easy to teach boys.

48.

In general, I ﬁnd it easy to teach girls.

68

PART 2: THE IMPORTANCE OF EDUCATION 1. Below are statements about the importance of education. Please ﬁnish the sentence by placing an ‘‘x’’ in the box that represents the extent to which you agree or disagree. Please choose 1 response for each statement. In order to gain the knowledge and skills that they need to prepare for adulthood in contemporary Beninese society . . .

Boys should learn basic math.

50

Boys should learn advanced math.

51.

Boys should learn typing.

52.

Boys should learn accounting.

53.

Boys should learn chemistry.

54.

Boys should learn French.

55.

Boys should learn sewing.

56.

Boys should learn cooking.

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

SONYA ANDERSON

49.

Strongly Disagree

Boys should learn mechanics.

58.

Boys should learn biology.

59.

Boys should learn computer skills.

60.

Girls should learn basic math.

61.

Girls should learn advanced math.

62.

Girls should learn typing.

63.

Girls should learn accounting.

64.

Girls should learn chemistry.

65.

Girls should learn French.

66.

Girls should learn sewing.

67.

Girls should learn cooking.

68.

Girls should learn mechanics.

69.

Girls should learn biology.

70.

Girls should learn computer skills.

Pedagogy of Difference

57.

69

70

2. Please ﬁnish the sentence below by placing an ‘‘x’’ in the one box that best represents your opinion. In my opinion, one of the most important beneﬁts of education is that . . .

Boys develop literacy skills.

72.

Boys develop leadership skills.

73.

Boys learn domestic skills.

74.

Boys are able to ﬁnd a good job.

75.

Boys are able to attract a better marriage partner.

76.

Boys develop childrearing skills.

77.

Boys develop interpersonal skills.

78.

Girls develop literacy skills.

79.

Girls develop interpersonal skills.

80.

Girls develop leadership skills.

81.

Girls learn domestic skills.

82.

Girls are able to ﬁnd a good job.

83.

Girls are able to attract a better marriage partner.

84.

Girls develop childrearing skills.

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

SONYA ANDERSON

71.

Strongly Disagree

85.

The beneﬁts of sending a girl to school do not outweigh the ﬁnancial costs to her family.

86.

The beneﬁts of sending a boy to school do not outweigh the ﬁnancial costs to his family.

87.

Boys should be allowed to take any subject that they wish to take in school, regardless of their gender.

88.

Girls should be allowed to take any subject that they wish to take in school, regardless of their gender.

89.

Primary school is the highest level of education that girls should complete.

90.

Secondary school is the highest level of education that girls should complete.

91.

University is the highest level of education that girls should complete.

Strongly Disagree

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

Pedagogy of Difference

3. Below are more statements about the importance of education. Put an ‘‘x’’ in the box that represents the extent to which you agree or disagree with each, and please choose only 1 response.

71

72

Primary school is the highest level of education that boys should complete.

93.

Secondary school is the highest level of education that boys should complete.

94.

University is the highest level of education that boys should complete.

95.

A girl should get married when she ﬁnds an appropriate partner, even if she is still in school.

96.

A boy should get married when he ﬁnds an appropriate partner, even if he is still in school.

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

SONYA ANDERSON

92.

Strongly Disagree

1. Below are statements about students outside of the teaching and learning that normally occurs in class. Please put an ‘‘x’’ in the box that represents the extent to which you agree or disagree with each sentence. Please choose only 1 response for each statement.

Boys are best at cleaning the classroom.

98.

Some chores are only appropriate for boys to do.

99.

In the classroom, it is better for boys to be seated next to other boys.

100.

It is important for boys to be seated near the front of the classroom.

101.

Boys should be allowed to participate in any school sport that they like.

102.

In general, boys do not manage their time in the classroom well.

103.

In general, boys are very polite.

104.

Boys are often tardy.

105.

Boys are often noisy in class.

106.

Boys are often absent from school.

107.

Boys are often difﬁcult to discipline.

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

73

97.

Strongly Disagree

Pedagogy of Difference

PART 3: THE MANAGEMENT OF NON-ACADEMIC BEHAVIOR

109.

Boys are often uncooperative.

110.

Some chores are only appropriate for girls to do.

111.

In the classroom it is better for girls to be seated next to other girls.

112.

Girls are best at cleaning the classroom.

113.

It is important for girls to be seated near the front of the classroom.

114.

Girls should be allowed to participate in any school sport that they like.

115.

Girls are often absent from school.

116.

In general, girls are very polite.

117.

Girls are often tardy.

118.

Girls are often difﬁcult to discipline.

119.

In general, girls do not manage their time in the classroom well.

120.

Girls are generally untrustworthy.

121.

Girls are often noisy in class.

122.

Girls are often uncooperative.

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

SONYA ANDERSON

Boys are generally untrustworthy.

Disagree

74

108.

Strongly Disagree

A. Learning Styles and Instructional Interaction 1. Below are some questions about your teaching practice. Please put an ‘‘x’’ in the box that represents the frequency with which these actions occur in your classroom. Please choose only 1 response for each statement. In general, how often do you . . .

Organize small cooperative learning groups for students?

124.

Create leadership opportunities for girls in classroom activities?

125.

Create leadership opportunities for boys in classroom activities?

126.

Use role-playing in your classroom activities?

127.

Give individualized instruction to a girl in your class?

128.

Give individualized instruction to a boy in your class?

129.

Give positive feedback to a girl about the content of her work?

130.

Give positive feedback to a girl about the appearance of her work?

131.

Give positive feedback to a boy about the content of his work?

132.

Give positive feedback to a boy about the appearance of his work?

Usually

Sometimes

Rarely

Never

1

2

3

4

5

75

123.

Always

Pedagogy of Difference

PART 4: TEACHING AND LEARNING

76

Give a girl another opportunity to respond when she answers incorrectly to a question the ﬁrst time?

134.

Give a boy another opportunity to respond when he answers incorrectly to a question the ﬁrst time?

135.

Scold or punish girls for tardiness?

136.

Scold or punish girls for incorrect responses in class?

137.

Scold or punish girls for not having their homework?

138.

Scold or punish girls for not paying attention in class?

139.

Scold or punish boys for tardiness?

140

Scold or punish boys for incorrect responses in class?

141.

Scold or punish boys for not having their homework?

142.

Scold or punish boys for not paying attention in class?

143.

Called on a student who had not raised his/her hand to volunteer?

144.

Call on a boy to answer a question in class?

145.

Call on a girl to answer a question in class?

146.

Practice ‘‘wait time’’ (waiting a few seconds) before calling on a student to answer?

Usually

Sometimes

Rarely

Never

1

2

3

4

5

SONYA ANDERSON

133.

Always

Pose ‘‘open’’ questions (questions that require a full description or explanation as a response) to a girl in your class?

148.

Pose ‘‘open’’ questions (questions that require a full description or explanation as a response) to a boy in your class?

Pedagogy of Difference

147.

B. Instructional Support 1. Below are statements about the types of instructional support in your classroom. Please put an ‘‘x’’ in the box that represents the extent to which you agree or disagree with each one. Please choose 1 response for each statement.

Currently, boys in my classroom have good access to textbooks.

150.

Currently, girls in my classroom have good access to textbooks.

151.

Currently, there are many examples of girls’ work on display in my classroom.

152.

Currently, there are many examples of boys’ work on display in my classroom.

153.

I occasionally supplement textbooks and curricula with positive images, stories, etc. of girls/ women?

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

77

149.

Strongly Disagree

I have invited women ‘‘role models’’ from the community, etc. to visit my classroom at least once during this academic year.

155.

I occasionally supplement textbooks and curricula with positive images, stories, etc. of boys/men?

156.

I have invited male ‘‘role models’’ from the community, etc. to visit my classroom at least once during this academic year.

Disagree

Somewhat Disagree

Neutral

Somewhat Agree

Agree

Strongly Agree

1

2

3

4

5

6

7

78

154.

Strongly Disagree

C. Language Patterns 1. Please put an ‘‘x’’ in the box that represents the frequency with which these actions take place in your classroom. Please choose only 1 response for each statement.

157.

Do you praise a girl for her attractiveness?

158.

Do you praise a girl for her compliance with classroom rules?

159.

Do you praise a girl for showing caring?

Always

Usually

Sometimes

Rarely

Never

1

2

3

4

5

SONYA ANDERSON

In general, how often . . .

Do you praise a girl for her competence with domestic tasks?

161.

Do you praise a girl for her academic effort?

162.

Do you praise a girl for her creative or innovative ideas?

163.

Do you praise a boy for his attractiveness?

164.

Do you praise a boy for his academic effort?

165.

Do you praise a boy for his compliance with classroom rules?

166.

Do you praise a boy for his competence with domestic tasks?

167.

Do you praise a boy for showing caring?

168.

Do you praise a boy for his creative or innovative ideas?

169.

Does a boy tease or insult a girl in your class?

170.

Does a girl tease or insult a boy in your class?

171.

Do you scold or punish a boy for teasing or insulting a girl?

172.

Do you scold or punish a girl for teasing or insulting a boy?

Pedagogy of Difference

160.

79

80

D. Roles in the Classroom 1. Below are statements about the roles that students play in your classroom. Please put an ‘‘x’’ in the box that represents the extent to which you agree or disagree with each one. Please choose 1 response for each statement.

When you organize role-playing activities in class, how often do you select girls to play traditionally female roles (nurse, secretary)?

174.

When you organize role-playing activities in class, how often do you select boys to play traditionally male roles (doctor, businessman)?

175.

When you organize role-playing activities in class, how often do you mix the roles so that boys and girls play a variety of different characters, regardless of gender?

176.

When there are opportunities to serve in a leadership position in your classroom, how often do you choose students to ﬁll those positions?

177.

When there are opportunities to serve in a leadership position in your classroom, how often do you allow students to volunteer to ﬁll those positions?

178.

When there are opportunities to serve in a leadership position in your classroom, how often does a girl ﬁll the position?

179.

When there are opportunities to serve in a leadership position in your classroom, how often does a boy ﬁll the position?

Usually

Sometimes

Rarely

Never

1

2

3

4

5

SONYA ANDERSON

173.

Always

1. Please complete the sentences below by putting an ‘‘x’’ in the box that represents the frequency with which these activities occur in your classroom. Please choose 1 response for each statement. In general, I . . .

Ask the girls in my class to sweep the ﬂoor.

181.

Ask the girls in my class to sweep the courtyard.

182.

Ask the girls in my class to clean the chalkboard.

183.

Ask the girls in my class to wipe the tables.

184.

Ask the girls in my class to wipe the chairs.

185.

Ask the girls in my class to cook food.

186.

Ask the girls in my class to fetch the food.

187.

Ask the girls in my class to fetch water.

188.

Ask the girls in my class to clean the latrine.

189.

Ask the boys in my class to clean the latrine.

190.

Ask the boys in my class to sweep the courtyard.

191.

Ask the boys in my class to clean the chalkboard.

192.

Ask the boys in my class to sweep the ﬂoor.

Usually

Sometimes

Rarely

Never

1

2

3

4

5

81

180.

Always

Pedagogy of Difference

E. Management of Students’ Time and Space

82

Ask the boys in my class to wipe the chairs.

194.

Ask the boys in my class to cook food.

195.

Ask the boys in my class to fetch the food.

196.

Ask the boys in my class to wipe the tables.

197.

Ask the boys in my class to fetch water.

198.

Allow boys to choose which chores they will perform in the classroom.

199.

Assign cleaning the classroom as a form of punishment for boys.

200.

Allow girls to choose which chores they will perform in the classroom.

201.

Assign cleaning the classroom as a form of punishment for girls.

202.

Allow the girls in my classroom to sit where they please.

203.

Seat most of the girls in the back of the classroom.

204.

Seat girls throughout the classroom.

205.

Allow the boys in my classroom to sit where they please.

206.

Seat most of the boys in the back of the classroom.

207.

Seat boys throughout the classroom.

Usually

Sometimes

Rarely

Never

1

2

3

4

5

SONYA ANDERSON

193.

Always

208. How many students are enrolled in your school? 209. How many teachers work at your school? 210. How many girls do you teach in your current class? 211. How many boys do you teach in your current class?

Pedagogy of Difference

PART 5: GENERAL INFORMATION

212. Your school is located in a _______ area. (Please circle 1 response.) a) Rural b) Urban 213. During this academic year, approximately how many days of training have you received in the Nouveau Programme? 214. What is your gender, male of female? 215. How old are you? 216. How many years have you been teaching primary school? (Please do not count time spent as a student teacher) 217. What is the highest educational diploma/certiﬁcate that you have received? 218. What type of teaching certiﬁcate do you hold? Thank you very much for your participation in this study!

83

84

SONYA ANDERSON

APPENDIX 2. CONTENT AND CRONBACH’S ALPHA VALUES OF THE SUBSET OF ITEMS USED TO FORM THE QUESTION AND OUTCOME COMPOSITE VARIABLES Item No.

Content

Alpha

Implementation of EIC with Boys (B_EIC) Create leadership opportunities for boys in classroom activities? Give individualized instruction to a boy in your class? Give positive feedback to a boy about the content of his work? Give a boy another opportunity to respond when he answers incorrectly to a question the ﬁrst time? Call on a boy to answer a question in class? Pose ‘‘open’’ questions (questions that require a full description or explanation as a response) to a boy in your class?

.452 .353 .448 .411 .405

Implementation of EIC with Girls (G_EIC) Create leadership opportunities for girls in classroom activities? Give individualized instruction to a girl in your class? Give positive feedback to a girl about the content of her work? Give a girl another opportunity to respond when she answers incorrectly to a question the ﬁrst time? Call on a girl to answer a question in class. Pose ‘‘open’’ questions (questions that require a full description on explanation as a response) to a girl in your class?

.440 .368 .416 .386 .383

1 4 5 6 20 22 33 42 43 44 47

Boys’ Academic Capability (B_CAP) In general, boys are naturally smart in school. Science is easy for boys to learn. Math is easy for boys to learn. French is easy for boys to learn. Boys most often answer questions correctly in class. Boys most often want to answer questions in class. Boys are usually very academically successful. Boys are usually the best students in math. Boys are usually the best students in French. Boys are usually the best students in science. In general, I ﬁnd it easy to teach boys.

.873 .866 .859 .858 .859 .867 .873 .869 .856 .854 .854 .868

10 11 12

Girls’ Academic Capability (G_CAP) French is easy for girls to learn. Science is easy for girls to learn. Math is easy for girls to learn.

.841 .822 .821 .824

125 128 131 134 144 148

124 127 129 133 145 147

.411 .417

.416 .403

85

Pedagogy of Difference

APPENDIX 2. (Continued ) Item No. 13 17 18 19 28 38 40 48

Content

Alpha

In general, girls are naturally smart in school. Girls are usually the best students in math. Girls are usually the best students in French. Girls are usually the best students in science. Girls are usually very academically successful. Girls most often answer questions correctly in class. Girls most often want to answer questions in class. In general, I ﬁnd it easy to teach girls.

.828 .829 .824 .823 .828 .830 .838 .837

Boys’ Academic Performance (B_PERFORM) I spend a lot of time explaining classroom lessons to boys. The boys in my class do not have the intellectual capacity to pass the CEP. Boys most often answer questions incorrectly in class. Boys least often want to answer questions in class. Boys do not take their education very seriously. Boys’ interest in education decreases as they get older. Boys usually cannot answer difﬁcult questions well. Boys do not usually complete their homework assignments. When boys do not volunteer to answer questions in class, it is often because they are not paying attention.

.735 .729 .723

.729 .706

39 41 45 46

Girls’ Academic Performance (G_PERFORM) The girls in my classroom do not have the intellectual capacity to pass the CEP. I spend a lot of time explaining classroom lessons to girls. Girls usually cannot answer difﬁcult questions well. Girls do not usually complete their homework assignments. When girls do not volunteer to answer questions in class, it is often because they are not paying attention. Girls most often answer questions incorrectly in class. Girls least often want to answer questions correctly in class. Girls do not take their education very seriously. Girls’ interest in education decreases as they get older.

49 50 51 52

Subjects Important to Boys’ Education (B_SUBJECTS) Boys should learn basic math. Boys should learn advanced math. Boys should learn typing. Boys should learn accounting.

8 16 21 23 24 25 26 34 37

7 9 27 29 32

.707 .715 .681 .712 .694 .706 .727

.723 .697 .689 .715 .708 .719 .673 .721 .891 .891 .892 .883 .874

86

SONYA ANDERSON

APPENDIX 2. (Continued ) Item No.

Content should should should should should should should

learn learn learn learn learn learn learn

chemistry. French. sewing. cooking. mechanics. biology. computer skills.

Alpha

53 54 55 56 57 58 59

Boys Boys Boys Boys Boys Boys Boys

.875 .880 .878 .885 .877 .876 .880

60 61 62 63 64 65 66 67 68 69 70

Subjects Important to Girls’ Education (G_SUBJECTS) Girls should learn basic math. Girls should learn advanced math. Girls should learn typing. Girls should learn accounting. Girls should learn chemistry. Girls should learn French. Girls should learn sewing. Girls should learn cooking. Girls should learn mechanics. Girls should learn biology. Girls should learn computer skills.

.882 .878 .884 .870 .864 .867 .869 .866 .875 .878 .864 .869

102 104 105 106 107 108 109

Boys’ Non-Academic Behavior (B_NONAC) In general, boys do not manage their time in the classroom well. Boys are often tardy. Boys are often noisy in class. Boys are often absent from school. Boys are often difﬁcult to discipline. Boys are generally untrustworthy. Boys are often uncooperative.

.815 .813 .788 .805 .779 .777 .776 .792

115 117 118 119 120 121 122

Girls’ Non-Academic Behavior (G_NONAC) Girls are often absent from school. Girls are often tardy. Girls are often difﬁcult to discipline. In general, girls do not manage their time in the classroom well. Girls are generally untrustworthy. Girls are often noisy in class. Girls are often uncooperative.

.737 .699 .708 .706 .697 .688 .724 .718

Boys’ Mean Girls’ Mean (S.E.) (S.E.)

Boy-Girl Difference

4.28 3.86 4.11 4.09 3.86 4.75 5.52 4.81 4.05 3.77 3.87 3.86

(.071) (.115) (.102) (.112) (.109) (.104) (.103) (.079) (.114) (.092) (.103) (.123)

3.73 3.53 3.73 3.62 3.75 4.44 4.42 3.93 3.02 3.39 3.30 3.44

(.071) (.131) (.101) (.094) (.110) (.118) (.089) (.113) (.082) (.108) (.107) (.095)

.552*** .331* .386*** .479*** .114 .316* 1.11*** .882*** 1.03*** .382** .564*** .418***

Academic Performance (PERFORM) 12. I do not spend a lot of time explaining classroom lessons to boys/girls. 13. The boys/girls in my class do have the intellectual capacity to pass the CEP. 14. Boys/Girls do not most often answer questions incorrectly in class. 15. Boys/Girls do not least often want to answer questions in class. 16. Boys/Girls take their education very seriously. 17. Boys’/Girls’ interest in education does not decrease as they get older. 18. Boys/Girls usually can answer difﬁcult questions well. 19. Boys/Girls usually complete their homework assignments. 20. When boys/girls do not volunteer to answer questions in class, it is not because they are not paying attention.

5.03 4.77 5.66 5.07 5.56 4.85 4.95 5.05 4.89 3.78

(.055) (.125) (.119) (.078) (.081) (.094) (.116) (.091) (.081) (.119)

4.47 4.39 5.82 4.72 4.52 3.81 3.56 4.55 4.36 4.26

(.058) (.125) (.082) (.084) (.128) (.143) (.099) (.085) (.094) (.117)

.555*** .382*** .158 .344*** 1.04*** 1.04*** 1.39*** .495*** .541*** .473***

87

Academic Capability (CAP) 1. In general, boys/girls are naturally smart in school. 2. Science is easy for boys/girls to learn. 3. Math is easy for boys/girls to learn. 4. French is easy for boys/girls to learn. 5. Boys/Girls most often answer questions correctly in class. 6. Boys/Girls most often want to answer questions in class. 7. Boys/Girls are usually very academically successful. 8. Boys/Girls are usually the best students in math. 9. Boys/Girls are usually the best students in French. 10. Boys/Girls are usually the best students in science. 11. In general, I ﬁnd it easy to teach boys/girls.

Pedagogy of Difference

APPENDIX 3. MEAN, STANDARD ERROR, AND DIFFERENCE IN TEACHERS’ PERCEPTIONS (WITH P-VALUES) OF GIRLS’ AND BOYS’ ACADEMIC CAPABILITY, ACADEMIC PERFORMANCE, SUBJECT MATTER IMPORTANCE, AND NON-ACADEMIC BEHAVIOR AND AND CHARACTERISTICS. (N ¼ 324)

88

APPENDIX 3. (Continued ) Boys’ Mean Girls’ Mean (S.E.) (S.E.)

Boy-Girl Difference

5.54 (.059) 5.88 (.102) 4.93(.117) 5.16 (.109) 5.56 (.092) 5.21 (.093) 5.99 (.098) 5.22 (.101) 5.47 (.086) 5.56 (.088) 5.71 (.079) 5.96 (.067)

5.71 5.90 5.14 5.57 5.77 5.17 6.19 5.69 6.07 5.02 5.70 6.08

(.050) (.091) (.109) (.094) (.088) (.094) (.091) (.100) (.077) (.104) (.078) (.073)

.170*** .021 .213*** .403*** .208*** .035 .197** .467*** .601*** .537*** .003 .112

Non-Academic Behavior (NONAC) 32. Manage Time Well 33. Seldom Tardy 34. Seldom Noisy 35. Seldom Absent 36. Easy to Discipline 37. Trustworthy 38. Cooperative

4.43 4.34 4.92 3.36 4.72 3.96 4.59 4.96

4.82 4.38 4.33 5.01 4.40 5.15 4.88 5.34

(.049) (.091) (.085) (.089) (.093) (.074) (.107) (.095)

.392*** .040 .587*** 1.65*** .322* 1.19*** .293* .378***

Bpr.10, *pr.05, **pr.01, ***pr.001

(.060) (.106) (.079) (.087) (.089) (.103) (.085) (.082)

SONYA ANDERSON

Subject Matter Importance (SUBJECTS) 21. Basic Math 22. Advanced Math 23. Typing 24. Accounting 25. Chemistry 26. French 27. Sewing 28. Cooking 29. Mechanics 30. Biology 31. Computer Skills

LEARNING FROM EXPERIENCE: IMPROVING EQUALITY OF ACCESS AND OUTCOMES FOR GIRLS IN UGANDA’S UNIVERSAL POST-PRIMARY EDUCATION AND TRAINING INITIATIVE Jill Sperandio and Alice Kagoda INTRODUCTION Girls’ access to education has improved in many of the world’s developing countries. These countries are striving to meet the United Nations Millennium Development Goals (MDGs) requiring them to provide gender equality, promote the empowerment of women, and establish universal primary education (UPE) by 2015. The success of UPE in achieving gender equality in enrollment in those countries able to institute it is encouraging. Where previously girls trailed boys in their ability to access education due to parent inability or reluctance to pay the costs, they are now entering primary schools in comparable numbers (UNESCO, 1999, 2006). The focus on post-primary education is particularly critical for girls, as gender differences in participation and achievement are more pronounced at

Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 89–121 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010006

89

90

JILL SPERANDIO AND ALICE KAGODA

this level (Mugo, 2007). Even at the primary level, equality of access does not necessarily equate to equal outcomes, and the girls newly included in an education system may still ﬁnd themselves disadvantaged in comparison to boys in their ability to access limited and expensive post-primary educational opportunities. Universal and free post-primary education is seen as at least part of the solution to this continuing process of establishing gender equality societies. However, the limited numbers of post-primary institutions, and their geographical distribution, make gender equal access and outcomes a harder goal to achieve (Mbabazi, 2007). Uganda, which will be used as a case study for this chapter, launched a UPE initiative in 1997 and a universal post-primary education and training initiative (UPPET) in 2007. As the ﬁrst cohort of children who beneﬁted from UPE completed the 7-year primary course and exit examinations, a new dilemma faced the country. Secondary school positions were both limited in number and expensive in respect to the school fees charged for them. Consequently, the majority of the children successfully completing the primary school exit examinations had no choice but to end their formal education at that point. UPPET was a response to this situation. The initiative, which represents an important change in the education system of the country, dramatically increased access to secondary schools by removing much of the cost for those children reaching a given standard on the primary school exit examination. Clearly the costs to the government of undertaking this initiative were high, the more so given the context of ﬁnancial constraint that marks all aspects of Ugandan society. Thus, the move to provide free post-primary education has raised concerns both inside and outside the country that quality in education will be sacriﬁced to quantity, both at the primary and secondary levels, and that this impact the outcomes of education for girls. The purpose in this chapter is to predict the level of success of Uganda’s ﬂedgling post-primary education initiatives in achieving gender equality in both access and student outcomes. Uganda’s ability to meet the gender equality goals in education to which the country aspires is dependent on two conditions. The ﬁrst is the adoption of a change initiative designed to achieve gender equal access to post-primary education. The second is the provision of a classroom and school environment that facilitates gender equal student outcomes. Uganda provides a good illustration of the importance of context in the study of educational systems, a factor increasing recognized by researchers and practitioners involved in the development of cross-cultural education

91

Learning from Experience

initiatives (Dimmock & Walker, 2005). In common with most developing countries, policy-making for educational reform in Uganda is inﬂuenced by a worldwide concern with the provision of gender equal universal primary and secondary education. However, the historical development of education in Uganda has created unique features in its secondary education system. In addition, the recent provision of UPE has dramatically changed the characteristics of potential secondary school students by including many from rural areas, and a much larger numbers of girls. A successful change process must accommodate both of these context-related features. In order to allow the reader to better assess the likely success of UPPET with regard to gender equality, we begin with a consideration of worldwide and local factors that have shaped the post-primary initiative in Uganda. This is followed by an analysis of UPPET using models of successful educational change. We conclude with an examination of plans for the future development of UPPET using as a framework those courses of action that have proved, in Uganda and elsewhere, to be of particular beneﬁt to girls in gaining gender parity in access and outcome at all levels of education.

UNIVERSAL PRIMARY AND POST-PRIMARY EDUCATION: THE WORLD CONTEXT Primary Education Primary, or elementary education, the commonly accepted terms used to describe a period of formal education extending from the ages of 5 to 12 during which time children acquire basic literacy and numeracy and learn to use them in different ways, is now accepted as a child’s right by national governments worldwide. Until recently, however, few nations have had the resources to make the provision of such education universally available to all children. To do so requires that the education be supplied free of cost, and that parents be legally required to enroll their children in schools offering it. The signiﬁcant improvement in the provision of primary education in developing countries since the mid-1980s has been ascribed to differing causes. These including the spread of nationalism, increased expectations of citizens resulting from access to worldwide media, and the growing power of international and donor organizations to inﬂuence national policy-making

92

JILL SPERANDIO AND ALICE KAGODA

(Meyer, Ramirez, & Soysal, 1992). However, the United Nations Educational, Scientiﬁc and Cultural Organization (UNESCO) sponsored agreements of the Education For All (EFA) initiative promoted at Jomtein, Thailand in 1990 and reworked as the MGDs in Dakar, Senegal in 2000, focused attention on the issue of universal primary education and provided a framework for allocating resources to promote and monitor its development. As a result, enrollment, attendance, and completion at the primary school level have increased rapidly between both boys and girls in many countries, although worldwide coverage remains elusive (Bruns, Mingat, & Rakotomalala, 2003; Ridell, 2003).

Post-Primary Education Post-primary education commonly takes the form of secondary, or high school, education provided for students of ages 11–18 and is frequently divided into lower and upper levels. Vocational training may parallel academic schoolwork as an option for this age group, or form part of tertiary education, paralleling college and university courses. The context and objects of secondary education show great variability from one country context to another. The provision of universal secondary education (USE) is seen as the next challenge to countries that have embarked on UPE. In most developing countries, secondary school provision is extremely limited and high cost. This forces even those children who have completed primary education and have the ﬁnancial resources to continue in school, to compete for places. With increasing numbers of children completing primary education as a result of UPE, and the sensitization of populations regarding universal rights to education that UPE has brought, governments face increasing pressure from their citizenry to provide greater access to secondary education. There is a demand for both the creation of more school places and the removal of ﬁnancial barriers in accessing them. Discussions about what drives the demand for secondary education, how to provide it, and what it should contain are ongoing and varied. Clemens (2004) argued that universal primary enrollment hinges on progress in the provision of secondary and tertiary education given that no country has reached 90% net primary enrollment without attaining 45% secondary enrollment. Donor interest in, and support of, the expansion of secondary education has been encouraged by studies suggesting links between secondary education and social development in developing countries

Learning from Experience

93

(Lewin & Calloids, 2001; Bennell, 1996; Psacharopoulos & Patrinos, 2004; Mingat & Tan, 1996). The World Bank has promoted the need to reassess the importance of support to secondary education reform, not only to allow individuals to acquire skills and competencies that enhance their ability to participate fully in society and continue learning at the postsecondary level, but also to enable countries ‘‘to train the manpower required to beneﬁt from the ICT and knowledge revolution, and to compete successfully in the new globalized, knowledge-based economy’’ (Liang, 2002, p. iii).

Post-Primary Education in Sub-Saharan Africa The increasing interest in the provision of secondary education, the justiﬁcation for it, and the attempts currently being made by international organizations to determine the direction this expansion of formal educational provision should take, is clearly demonstrated in the context of subSaharan Africa (SSA). After years of focusing attention and resources on primary education, interest returned to the secondary school level in SSA in the late 1990s. Between 1996 and 2001, UNESCO delineated strategies and developed programs for the reform of secondary education in the region. These focused on expanding payments systems for secondary education services by diversifying institutional structures, reforming curriculum and pedagogy to include the use of information technology, and eradicating inequalities. These themes were developed at the Conference of the Organization of African Unity (OAU) in 1999 and the meeting of the Consortium on Secondary Education organized by UNESCO (Paris, 10–11 June 1999), driven by a common understanding that the challenges of the 21st century could only be met by a reformed secondary education sector. The UNESCO/BREDA World Bank Regional Workshop in Mauritius (UNESCO, 2001) directly addressed the renewal of African secondary education. Three common problems emerged during the proceedings; inadequate infrastructures, limited educational equipment, and few qualiﬁed teachers. Additionally, four justiﬁcations for renewed interest in secondary education were presented. These were a societal demand for an education that equipped students with the skills to enter the job market (who are functionally ready for work), preparing students for higher education, meeting the expectations of an expanding population of qualiﬁed primary school leavers and educating adolescents about societal problems such as the environment, human rights, drug addiction, AIDS, poverty, and

94

JILL SPERANDIO AND ALICE KAGODA

unemployment. The Workshop acknowledged the priorities set forth by the Conference of African Ministers of Education (COMEDAF) and World Education Forum (WEF) at Dakar in 2000 for reorganizing the secondary education sector, which included ensuring equity and quality, developing complimentary learning methods, reinforcing coordination skills, follow-up and evaluation, and diversifying learning methods. It also called for recognition of the role of the informal sector of the economy in impacting poverty eradication and youth employment. Following the Mauritius workshop, secondary education in Africa (SEIA) became a major preoccupation of the African section of the World Bank. The Africa Human Development Department launched a regional study bearing this title in 2002, involving the SSA countries and public and private African educators and stakeholders in a 5-year study. The aim of the study was to ‘‘produce and disseminate information and knowledge to assist SSA countries in the development and reform of secondary education systems, and to promote coordination and exchange of information between African secondary education stakeholders, private sector and civil society organizations, and the donor community’’ (Bregman & Armstrong, 2003, p. 1). The study developed around three conferences – Kampala, 2003; Dakar, 2005; and Accra, 2007. It produced an overview of reform trends and best practices in secondary education in the countries of the Organization for Economic Cooperation and Development (OECD), a data base for secondary education for operational purposes, seven thematic studies on speciﬁc issues relevant to secondary education (including gender equity) and speciﬁc secondary education development strategies for selected African countries, of which Uganda was one. Initial interest in SEIA focused on the quality, content, and organization of post-primary education and was driven by societal perceptions that what was on offer was not well suited to societal needs. The SEIA studies continued to couch the logic for investing in reform of secondary education in terms of the social and private returns it could yield. ‘‘Secondary education is crucial for economic growth. Globalization, the increasing importance of ICT in the 21st century, and rapid technological change have made knowledge a critical determinant of competitiveness in the world market’’ (Bregman & Armstrong, 2003, p. 4). Secondary education was described as the vehicle for providing countries with critical higher level skills and knowledge needed for economic growth, including further learning and training of professional such as technicians, scientists, and entrepreneurs. However, initial plans for long-term radical reform of secondary education systems had to be put aside when it became clear that the SSA countries that

Learning from Experience

95

had made signiﬁcant progress with providing UPE faced immediate decisions regarding expanding access to secondary education that had considerable political and ﬁnancial implications. Bregman and Armstrong (2003), describing this increased pressure from civil society groups, noted: The African youth age group of 12–19 year olds holds the future of the continent, and we must ensure they can get secondary education (by various ﬂexible means and routes) in the next decades. The current average 20% secondary enrollment . . . will not lead to rapid poverty reduction and better lives for African citizens. (p. 12)

Mamadou Ndoye, Executive Secretary for the Association for the Development of Education in Africa, stated that expanding access to secondary education for young people from ‘‘all strata and in all situations’’ was ‘‘a duty of equity and development’’ (quoted in Bregman & Armstrong, 2003, p. 4).

GENDER ISSUES IN POST-PRIMARY EDUCATION WORLDWIDE Access to Post-Primary Education Interest in girls’ education developing countries from 1980 onwards produced research indicating the extent of girls’ under representation and underperformance at all levels of education (Lee & Lockheed, 1990; Hyde, 1989; Ridell & Nyagura, 1991). Some studies argued the importance of education for girls from the standpoint of economic outcomes in the labor market (Elliot & Kelly, 1980) and social beneﬁts connected to investments in educating females (Floro & Wolf, 1990; Herz, Subbarao, Habib, & Raney, 1991; King & Hill, 1993). Research undertaken by Swainson (1996) and Swainson, Bendera, Gordon, and Kadzamira (1998) suggested that many of the issues affecting girls’ access to education and outcomes from it in developing countries could only be explained through the complex web of relations between household, community, and school. She noted, however, that donor investment to improve girls’ education was on a project-by-project basis, often consisting of single interventions based on a simple supply-anddemand framework. Despite publicity by United Nations Children’s Fund (UNICEF) and the Swedish International Development Cooperation Agency (SIDA) during the late 1980s and early 1990s to sensitize governments and donor agencies to the special needs of the ‘girl child’ and her education, most developing countries failed to meet the initial 2005 target date for MDG gender and

96

JILL SPERANDIO AND ALICE KAGODA

education goals. A Global Campaign for Education Report (GCE, 2008) suggested that the initial 2005 girls’ education target had been neither unrealistic nor unaffordable, but it had not been met because both the international community and national governments had given insufﬁcient political attention and inadequate money to meeting it. It was only through the implementation of universal primary education by a number of countries worldwide that a rapid and signiﬁcant change in girls’ access to schooling occurred. In many countries, the gender gap between the numbers of boys and girls completing primary education has narrowed to the point of insigniﬁcance. Lewis and Lockheed (2006) commented ‘‘progress in getting children into school has been impressive over the past decade, with most of the beneﬁts for girls’’ (p. 19). They noted that rapid expansion of the primary education sector automatically favors girls because of their previous neglect. In some countries, girls have surpassed boys in enrollment and attainment. However, 31 of the 196 countries in the world are still at high risk of not achieving gender parity in enrollment at even the primary level by 2015, the adjusted target date for the MDGs (UNESCO, 2006). The majority of these countries are in SSA. Although great strides have been made in giving girls access to primary education in SSA countries, only 17% of girls in these countries ‘‘achieve the promise of a quality secondary education’’ (Rihani, 2006, p. 8). Studies conducted from the mid-1990s onward clariﬁed the processes at work affecting both access and outcomes for girls (Fuller & Clark, 1994; Mensh, Bruce, & Greene, 1998; Raymond & Sadoulet, 2003; Rose, 2003; Herz & Sperling, 2004; Stash & Hannum, 2001). Although, the reasons for the disparity between male and female enrollment at the secondary school level are country-speciﬁc to a degree, considerable commonality exists in the causes of the low showing of girls in the secondary school population. These include the costs to families of the loss of a daughter’s labor, travel time to school, security on the journey to school, and school environments unfriendly to girls. These issues, and promising methods of addressing them in SSA, have been described most recently in a study by Sutherland-Addy (2008).

Equal Outcomes Some of the ﬁrst gender analyses of educational outcomes were conducted in the countries of the developed world and revealed a pattern of underachievement of girls in school, particularly in mathematics and science,

Learning from Experience

97

despite universal access to state-funded education (Steinkamp & Maehr, 1984; Yates, 1993). In SSA even those girls who manage to enroll in postprimary education still face many challenges. Although enrollment explains who enters, the survival and completion rates indicate that many girls fail to stay the post-primary school course. The completion rates for SSA are the lowest in the world, with 19 countries having a completion rate of less than 50% (UNESCO, 2006). Studies of countries across the globe, both developed and developing, have indicated a variety of factors that contribute to this situation (Stromquist, 1989; Lloyd, Mensh, & Hewett, 2000). These include pressure from families to drop out and marry, differentiated delivery and expectations for girls and boys in the classroom, teaching methods that are not ‘girl friendly’, and additional work-loads in the home environment that affect the performance of girls in the classroom. Rihani (2006) suggests that a clear understanding of the complexities of the secondary school experience for girls is needed before meaningful interventions can be designed. She describes how, for an African or SouthAsian girl: Her ﬁrst challenge is access to a safe school where she can focus on her schooling without threat of violence, harassment, physical labor, gender discrimination or personal hygiene concerns. Her second challenge is to ﬁnd quality education that is purposeful, relevant and meaningful so it will have value to the girl and her family as an end in itself. Finally, education becomes more challenging and the obstacles become greater . . . as they progress through the school system, girls must be given the motivation, through the assurance of gainful employment, for example, to continue despite family or societal pressure that may tell her than an education is of little value. (p. 66)

For both girls and boys in SSA, secondary education is a ‘‘grueling, highly selective obstacle course’’ (Noye, in Bregman & Armstrong, 2003, p. 4), and for girls, the obstacles are often insurmountable.

UGANDAN CONTEXT: UNIQUE FEATURES OF THE EDUCATIONAL SYSTEM Uganda, a country of 25 million people in East Africa, is predominantly rural and agricultural. British colonial history has left its legacy in both the use of English as a national language and the structure of its education system. From relative prosperity and political stability at the time of its independence from Britain in 1962, Uganda plunged into a devastating period of civil strife between 1972 and 1986. A new government led by President Museveni embarked on a recovery program in 1987, which

98

JILL SPERANDIO AND ALICE KAGODA

included the fostering of a women’s movement devoted to improving women’s rights and the protection of the girl child. However, the country was faced with a new crisis as HIV/AIDS spread rapidly through the population. This has led to well over a million orphans in the 7.6 million children of school age (6–18) representing 35% of the population, while life expectancy has plunged to below 50. A low gross national product per capita of US$350 in combination with other life quality indicators, ranks Uganda 154 out of 177 countries on the United Nations Development Program Human Development Index in 2008. The Ugandan education system follows a seven-four-two pattern. The ﬁrst 7 years of primary education are followed by 4 years of lower secondary (Ordinary) level, 2 years of upper secondary (Advanced) and 3–4 years of university education, or between 1 and 3 years of vocational or technical education. Selection and certiﬁcation occur at the end of each level when national examinations are conducted. The ofﬁcial age of entry is six for primary education. However, as in many developing countries, most children have traditionally entered at a later age, and there has been no legally imposed age of entry for post-primary education, which may be as high as 15 or older, given late entry and the repetition of grades in primary school. The characteristics of the present education system reﬂect Uganda’s multi-ethnic population, colonial heritage and traditional societal attitudes to gender, which the following historical overview will attempt to highlight.

Gender Differentiation Gender-differentiated access and outcomes in education at all levels are not new to Uganda. An informal system of education operated in Uganda before the arrival of missionary groups in the early 1890s. Boys were sent to the homes of tribal chiefs and inﬂuential persons to learn tribal law and acceptable behavior. Girls were taught by their mothers, relatives and friends to be homemakers, good wives and mothers, and efﬁcient workers on the family farm. Onto this system of traditional and gender-differentiated education, Catholic and Protestant missionary groups entering the country in the 1890s imposed a formal school system, which reﬂected two elements of the British education system of the time. The ﬁrst was the primary school, which eventually provided up to 6 years of ‘basic’ education, with reading, writing, and arithmetic taught in English. At the secondary level, an adaptation of British boarding schools was introduced (Tiberondwa, 1998).

Learning from Experience

99

These schools took children out of the village environment, and brought them together with their peers from other regions and tribal groups. They developed a school climate promulgating loyalty and service to one’s school and country. High academic achievement was reinforced by examinations. Personal leadership and responsibility were promoted by the prefect system, and sports were used as a means of generating team spirit. These two elements, operating in a Euro-Christian environment, dominated the colonial education system in Uganda until the 1950s. Awareness of the under representation of girls in the education system, and attempts to rectify this also have a long history. British government reports from 1937 to Independence in 1962 called for improved access to education for girls through the construction of more secondary boarding schools, and for reducing the disparity between boys and girls in the classroom by the provision of bursaries for girls. However, village families in Uganda still considered educating girls a waste of money and a threat to the structure of traditional society. Although primary schools were coeducational by the 1950s, debates raged as to whether secondary schools should follow suit (Castle, 1996). The report of the Uganda Education Commission in 1963, a year after Uganda’s independence, noted that nearly 60% of female primary school students failed to complete the 6 years of primary education, in contrast to 35% of boys. The commission laid responsibility for this disparity on parents’ desire for girls to marry early, parental dislike of coeducational schools, fear of unsupervised journeys to school, and a disregard for girl’s education in general. It suggested that special ‘catch-up’ grants be offered to certain regions of Uganda to ﬁnance girls’ education and for inducements such as rewards for attendance and bursaries be given to girls. Other suggestions included free lunches, free tuition, and free uniforms for girls of poor families, together with special allowances for women teachers (Ssekamwa, 1997). Secondary education remained the focus of educational development throughout the 1960s and early 1970s, but the structure continued to favor boys. The number of schools increased so that by 1967 there were 15 day and 58 boarding schools, 12 for girls only, 32 for boys, and 29 coeducational. A clear hierarchy of schools was also emerging. A policy was adopted allowing students to choose schools, and the students with the highest scores on the primary school exit examination among the applicants to each school would be awarded the school places. Students’ awareness of schools, parental religious afﬁliation, and the existing informal communications network worked to make some schools more popular than others.

100

JILL SPERANDIO AND ALICE KAGODA

These schools were then in a position to achieve the best secondary examination results, increasing their popularity and ability to attract the best students. As the single sex boarding schools had both the longest history and clearly established traditions of excellence, they topped the hierarchy by choice for girls, followed by the coeducational boarding schools (most of which had been boys’ schools that made places for girls) and then the day schools. The years between 1972 and 1985 were a time of political and economic upheaval in Uganda after a military coup brought dictator Idi Amin to power. AIDS wreaked havoc in the country, and civil war drove people into the cities. Despite all this, the education system remained surprisingly intact. A shift in priority to primary education took up what little funding was available for education and resulted in very limited growth in secondary school places. Competition for the positions at secondary schools increased as primary enrollment grew (Lukenge, 1972). As demand outstripped supply, a rapid growth of private secondary schools took place. As the economy broke down, and government economic support to the public school system dried up, the schools sought parental support to continue to operate and this further developed the school hierarchy. Parent–Teacher Associations (PTAs) and alumni groups were set up and introduced charges to supplement the low salaries of teachers. Thus, parental education levels also began to deﬁne the hierarchy of schools. With the growth of the AIDS crisis and civil disturbance, boarding school education became even more attractive to parents of teenage girls (Ssekamwa, 1997). When stability began to be reestablished after 1985, secondary education once again received priority with an increase in school-building taking place. In 1987, an Education Policy Review Commission was set up by the Ministry of Education to develop a long-term plan for the development of the education system, and speciﬁcally to address the lower numbers of girls completing primary education and entering secondary schools. The committee’s ﬁndings were discussed in parliament and resulted in a Government White Paper on Education, published in 1992. The White Paper contained a number of guidelines for the development of the education system at the secondary and tertiary levels that were regarded as controversial by groups within Uganda concerned with increasing access of girls to secondary schooling. However, little implementation of these recommendations for the secondary sector took place as world interest and donor funding became focused on primary education (Hansen & Twaddle, 1998).

101

Learning from Experience

UPE Initiative UPE was introduced in Uganda in January 1997, following a political commitment by President Museveni in 1996 that the government would meet the costs of primary education of four children per family. All tuition fees were abolished, including PTA charges, leaving parents with just the costs of writing materials, school uniforms and their labor for classroom construction. This commitment was soon extended to allow all people that wanted to access primary education to do so, but enrollment was not made compulsory. The government’s declared objectives for the UPE initiative were to provide the facilities and resources to enable every child to enter and remain in school to complete the 7-year primary course, to increase equity of access resulting from direct and indirect costs of education and to improve the quality of life by providing every individual with basic numeracy and literacy skills (MoES, 1998). Capitation grants of approximately US $8 a month are supplied to schools for each student enrolled. Gross enrollments in primary education rose dramatically as a result of this initiative from 3.1 million in 1996 to 7.6 million in 2003, a 145% increase in comparison with a 39% increase in 10 years previous to UPE. The doubling of the funding needed to support the primary sector came in large part from debt relief and donor agency support. The success of the initiative has been attributed to a number of reasons. These include the government’s commitment to honor promises made during the presidential political campaigns including committing 65% of the educational budget to UPE (MoES, 2006), external funding forthcoming because of Uganda’s excellent track record of compliance with donor recommendations and because UPE was being touted as the answer to world poverty, and to parent willingness to send their children to school (Murphy, Bertoncino, & Wangi, 2002). The numbers of parents willing to do so indicate that school fees were, in fact, an impediment to accessing education for poor families. Since the introduction of the program, measures have been underway to improve both the primary school environment and student outcomes. These measures include countering the misuse of funds allocated to schools, poor classroom construction, high student/teacher and student/classroom ratios, the recruitment of untrained teachers for rural schools, and the lack of community involvement in school improvement. Despite suggestions that education performance has declined following the introduction of UPE and there is a high dropout rate (reported at 25%), the ‘‘UPE programme in Uganda demonstrates that a poor country with a committed government

102

JILL SPERANDIO AND ALICE KAGODA

and donor support can ﬁght poverty through ensuring universal access to education for its citizens’’ (DFID, 2006, p. 25).

Secondary Education Before UPPET Performance on the primary level exam (PLE) determined selection in post-primary and training. A survey by Liang (2002) of secondary school enrollment noted a number of areas of gender bias. Because only the topscoring children gained entrance to limited numbers of secondary schools, and a number of these were girl’s schools, the ‘gender gap’ in enrollment pre-UPPET – 44% girls to 56% boys in secondary school-looked small at this time, but concealed the low numbers of girls proceeding to the post-primary level. High dropout rates resulted in more than 20% of boys being in secondary school at the age of 20, compared to less that 5% of girls. The report assumed that the girls’ dropout rate was linked to marriage. Geographically, the northern region of Uganda has the lowest female participation due to social norms. Female students in rural, non-governmentaided schools have the highest dropout rate of 16.9%. A survey of secondary school quality included in Liang’s report indicated that secondary schoolgirls did signiﬁcantly worse in mathematics than boys, with 46% of girls failing the math test at the completion of the ﬁrst year of secondary school compared to 36% of boys. Liang (2002) also highlighted the impact of poverty on secondary school selection: Compared to their wealthier counterparts, children from poor households, in addition to having fewer chances of enrolling in secondary school, are more likely to go to a lower quality school if they do enroll. And since education is a powerful instrument for upward mobility – social and economic – the poor, lacking the opportunity to go to a quality school, will always remain behind. (p. 30)

Attendance patterns and caused of primary student dropout were analyzed in Ministry of Education and Sports (MoES) commissioned reports by Nkanyike, Kasante, and Balihuta (2002) and Kasente, Nakanyike, and Balihuta (2003). The results of these studies related high dropout to a poor inspection system that lowered motivation and enthusiasm among school administrators and teachers. They suggested that retaining children in school was not as highly prized by the government as enrolling them. Parents and guardians were also lax towards absenteeism and poor performance.

Learning from Experience

103

An MoES report (August, 2004) on the development of education in Uganda noted that the transition rates from one level of education to the next level continued to be low. Of the 400,000 students who graduated from primary schools in 2003, only 50% joined post-primary educating institutions. Only 10% of the children who enrolled in primary schools ultimately reached university levels, with the transition rate for girls much lower. The report suggested that one of the main challenges facing the education system was the imbalance in gender and inadequate opportunity for the poor and persons with disabilities. The report stated the need to build awareness of gender and its impact on the education process and outcomes. It noted the need to provide teachers with the knowledge and skills required to adopt and promote gendersensitive approaches in all school activities that might have a direct impact on retention of girls in school. It described the following measures that had been taken to promote gender equal education – the development of a gender policy for education and training, training for teachers in ensuring gender equality in classrooms, afﬁrmative action, bursaries for poor but gifted girls, and encouraging the use of female role models. In addition, the report stated that the ‘‘curriculum development process is being increasingly guided by gender concern especially in its objectives and instructional materials’’ (MoES, 2004, p. 14). UPPET Initiative The adoption of UPPET in Uganda followed a similar pattern to that of UPE. With presidential elections imminent, and a large number of potential primary school graduates facing an end to their education because of the lack and cost of secondary school places, President Museveni announced in 2001 an intention to pursue free secondary education. Following this announcement, the MoES sought reports and recommendations for implementing this promise from a variety of sources. A World Bank report (Liang, 2002) produced to examine the options available to the government noted the enormous challenges the country faced in maintaining the drive towards quality universal primary education while, at the same time, responding to the increasing social and economic demand for rapid expansion of secondary education. It suggested that when considering the options for expanding access to secondary education, the government would have to decide on difﬁcult trade-offs. These would include the role of public ﬁnancing in supporting different levels (lower and upper secondary education) and types of secondary education (general, technical, vocational) as well as in ‘‘ensuring equity in provision between different population

104

JILL SPERANDIO AND ALICE KAGODA

groups (boys and girls, urban/rural, poor and more well off )’’ (Fredriksen, as quoted in Liang, 2002, p. iv). Uganda’s choice was between short-term adjustments and improvements to the existing system to rapidly expand access, or long-term reorganization of the whole of the education system, to provide a system closer to the US model. It opted for the former course. Piecemeal improvements to the existing secondary school structure began. These included giving an annual capitation grant to schools, the payment of fees of the 47,000 children in camps for people displaced as a result of the continued war with the Lord’s Army in the northern region which had targeted children in rural villages, a bursary scheme for 5,500 needy but high-ability students, the construction of 20 seed schools in sub-counties without a secondary school with plans drawn up for a further 24, and the upgrading of 196 community to government-aided status, bringing the total of such schools to 845 (Nsubuga, 2006). In November 2005, before a third presidential election campaign, President Museveni announced the government’s intention of introducing UPPET beginning in the academic year starting February 2007. UPPET included the academic 4-year secondary course concluding with the Ordinary Level examinations, and the 1-year vocational and technical education and training courses taken by students on completion of the 7-year primary school course. In an MoES (2006) document to schools outlining the mechanism for distributing funds to support the UPPET initiative, the scheme was described as having four purposes. These were to increase equitable access to secondary education and vocational training; to assure the achievement of gender parity by 2015 as required by the MDG; to enhance the sustainability of UPE; and to reduce the high costs of secondary education and vocational education. The document noted the provision of capitation grants was aimed at ‘‘assuring universal access and ensuring quality UPPET delivery’’ (MoES, 2006, p. 1). The ﬁrst cohort of students entered the UPPET institutions in February 2007. Schools were permitted to charge no more than US$40 a term to cover books and supplies, with the government providing US$378 a term for each student directly to the school, and a variable tuition fee of US$16 per student per term. The second cohort entered in February 2008, with an anticipated 160,000 students entering government schools, and 96,000 in private schools. According to MoES statements to the press, it is currently (March, 2008) undertaking a headcount of enrollments, enlisting additional private schools into the initiative, and hiring additional teachers as need dictates (Lirri, 2008; Lirri & Wossita, 2008; Wossita, 2008).

105

Learning from Experience

UPPET: A SUCCESSFUL MODEL FOR CHANGE? Gender Equality in Access Uganda chose to replicate the ‘Big Bang’ approach to the provision of UPPET that it had proved successful for UPE, focusing energy on rapidly expanding access to schooling through short-term adjustments and improvements to the existing system (Murphy et al., 2002). This was in contrast to the strategy of long-term reorganization of the whole of the education system recommended by donor agencies. Although the country’s experiment with change appears risky given the limited resources available, it does model processes of successful organizational reform described by Fullan (1993, 2001, 2008), and Gleick (1999) among other change theorists. Organizational change theory suggests that once started, change takes on a life of its own, There may initially be a period of turbulence when different actors in the process experiment and respond to the new environment created by the change initiative, a process Homer-Dixon (2000) refers to as ‘‘careening into the future’’ (p. xx). Fullan (2001) suggests that the turbulence and initial chaos of the early stages of change are ultimately replaced by a search for order. The challenge for those initiating change lies in managing this stage of the process so that its outcome, the new order, ultimately matches the goals envisaged at the start of the process. Mintzberg, Ahlstrand, and Lampel (1998) conclude that ‘‘the best way to manage change is to allow for it to happen’’ and to be ‘‘pulled by the concerns out there rather than being pushed by the concepts in here’’ (p. 33). Fullan (2001, 2008) suggests that ‘changing the culture’ – encouraging people to ﬁnd innovative solutions to problems as they appear, rather than creating the resistance to change that is the frequent result of highly prescriptive large scale restructuring of systems and organizations – is the preferred course of action when managing change. Liang (2002), considering the options available to developing SSA countries desirous of achieving the social beneﬁts that come from an educated populace, noted that slow moving, carefully planned courses of action might prove less successful that a rapid expansion of access to education. He suggested that, rather than attempting to prescribe the content and shape of secondary education, increasing access could, in itself, facilitate important change. Access to post-primary education and training could play a part in creating an environment ready to absorb change or open to new technology or critical of poor governance. He noted ‘‘at the very least diffusion of schooling to the mass of the population will itself be

106

JILL SPERANDIO AND ALICE KAGODA

part of the reason that changes in society will be adopted’’ (p. 55). Liang posited a two-way relationship where the very provision of mass schooling, or mass secondary education, will itself play a partial role in determining whether the environment is enabling of change or not. So, given that Uganda’s strategy for moving quickly to universal access has the potential for success, will the country be able to steer the change process to achieve its desired goals? Fullan (2008) suggests that success depends on there being a clear ‘moral purpose’ for the change initiative, committed and empathic leadership that champions the cause, an organizational culture open to risk taking and with an expectation of success, and the building of relationships both within the organization and with those who can provide support from outside it. Uganda appears well set with regard to all these conditions. The Ugandan government has strongly promoted the change initiative as a move towards a more equitable society. Geraldine Namirembe Bitamazire, (Minister of Education and Sports, 2006) stated in the introduction to the MoES education sector report for that year (MoES, 2006) that the ministry was ‘‘steadfastly addressing the two MDGs relating to the education and sports sector’’ (p. 10). She emphasized the ministry’s commitment to ensuring all children and speciﬁcally girls, children in difﬁcult circumstances, and those belonging to ethnic minorities, had access to completely free and compulsory education by 2015. Whereas the timing of the introduction of universal primary and secondary education may well have been prompted by political considerations connected to internal elections, the Ugandan government has shown a strong commitment to the world vision of the MGDs, including gender equality, through governmental support of women’s issues and inclusion of women in nation building since the mid1980s (Tripp & Kwesiga, 2002). Leadership of the initiative has come from President Moseveni who has demonstrated a personal commitment to achieving its goals and a strong empathic understanding of Ugandans’ driving desire for education in any form that has been demonstrated numerous times in the history of the country (Murphy et al., 2002). Uganda has also the experience gained from the successful introduction of UPE using a similar ‘Big Bang’ approach. Despite numerous studies over the years that have ascribed girls’ low enrollment in education to the indirect costs of their lost labor to their families, the removal of the direct costs of education in Uganda has seen the majority of girls enrolled. The experience and the learning provided by UPE together with the relationship building done to achieve it – gaining the buy in of government ministries, encouraging community responsibility at district and village level for the

Learning from Experience

107

provision of basic school facilities, and gaining the backing and collaboration of important donor agencies, build conﬁdence that the UPPET initiative can be steered to achieving its goals. Some current problems are likely to solve themselves as the UPPET works through its ﬁrst cycle. Others require monitoring and appropriate interventions, and here the government can draw on the experience of many successful small-scale projects designed to empower girls that have been operating successfully in Uganda and other SSA countries. This approach allows for the tailoring of interventions to particular situations – the rural girl faced with the physical difﬁculty of getting to school, or the urban girl in a large city high school who must negotiate the problems connected with large classes and unsupervised boarding. One concern likely to be resolved without further intervention is that of girls dropping out of school. Although dropout and completion rates for boys and girls are similar, they show a different pattern. Girls are most likely to ‘drop out’ as they reach puberty. Currently, girls may already be approaching their teens as they enter primary school and most children are between 14 and 17 before they enter secondary school. Parents are frequently involved in the decision that girls should leave school. Early and forced marriage is still a common strategy used by poor families to raise income for the rest of the household members. Some parents believe that marriage is an escape route out of poverty, and they encourage their teenage daughters to drop out of school so they may marry them off. Others have, in the past, simply forced their daughters’ hands by refusing to pay for their education. Currently, 16% of girls are married by the age of 18, the legal age of marriage, and 69% are sexually active by this age. Girls are now enrolling in primary school at the age of 5 or 6 and will complete secondary school by the age of 16. This should result in less pressure from parents for girls to withdraw from school at puberty. The existing government campaigns to educate girls about contraception and the dangers of contracting HIV, together with enforcement of the legal marriage age, should all contribute to a lowering of the dropout rate as girls’ progress through schools in ageappropriate classes and with peers of a similar age (Were & Nafula, 2003). The government has taken a direct role in ensuring children take advantage of their right to an education. In a recent announcement, it declared its intent to make school enrollment and attendance compulsory (Aber, Akena, & Oper, 2007) and to prosecute parents who fail to send their children to school. This will further contribute to same age classes and reduce the early withdrawal of older girls from school. The government has

108

JILL SPERANDIO AND ALICE KAGODA

also taken a stand on the issue of students repeating grades by favoring automatic promotion in primary school. Repetition rates are similar for girls and boys. However, the result of holding students back to repeat grades and thus slowing their progress through school has a far greater impact on girls because of the issues connected to puberty and marriage. Problems still remain to be addressed, however. Most of these are contextual, related to the speciﬁc characteristics of the Ugandan education system. Primary school exit scores determine which students go on to free post-primary education. The government sets the cutoff point to reﬂect the number of places in schools that are available. Students who do not obtain these grades may still access places in private schools if they are able to afford the fees. Given that girls currently have both lower primary school completion rates and lower scores on the exit examinations, the percentage of girls receiving the beneﬁt of UPPET must, for the present, be smaller than that of boys. Even with UPPET, many children have difﬁculty ﬁnding the funds needed to take up secondary school places, a situation in part connected to the historical distribution of secondary schools in Uganda. High scoring children may be offered places in schools far from home, requiring expenditure out of reach of many families. A recent newspaper story gave the following examples of the realities of life for high achieving students from poor families: Mayuge’s primary leaving exams best two candidates are too poor to afford further education in any one of Uganda’s good secondary schools. Both Arajab Lugendo, who sat at Ntinkalu Primary school and emerged best in the district with ﬁve points, and Elizabeth Akatekt of Buluba Primary school, who led among the girls in the district, are staying with their unemployed poor mothers, having lost their fathers . . . When visited by reporters, Lugendo was found preparing fresh cassava that he said had been given to him by a sympathetic neighbor because they do not have food at home. ‘A good meal is a miracle in this home and I have no cause to celebrate my victory because the results seem to at the same time mark the end of my education’ he said. He said even if the government undertook to pay his tuition fee under the USE, his mother alone could no longer afford other school expenses like stationery and uniforms. Akatekit, who is eager to continue education up to university and do medicine, said her mother has been a very caring parent and this is how she managed to excel in the exams. She said although her mother does not have money to take her to secondary school, she has started baking and selling buns so that by the time school starts, she will have saved some money to enable her to get secondary education under USE. (Siminyu, 2008)

Although the ﬁnancial problems of transitioning to secondary school affect poor children regardless of gender as these stories conﬁrm, there may be other hurdles for girls. With comparatively few secondary schools, girls

Learning from Experience

109

living too far to walk to the nearest school, or girls who are offered a place at a better school in another region of Uganda are faced with either boarding at a school, staying in a hostel close by, or living with relatives near the school. All these options have their cost, both directly in money, and indirectly in the lack of family support and personal security they pose, and may well be the deciding factor as to whether girls go on to secondary schooling. The mother of a girl of secondary school age in Kapchorwa district was asked why she and other parents in the village let their children attend what was clearly a substandard local private school. She replied ‘‘You cannot jump where you cannot reach. Most of us parents are rural people who live on less that sh10, 000 a month . . . children end up in these schools because they are poor. The rich people end up taking their children to private schools’’ (Womakuyu, 2008). Lewis and Lockheed’s (2006) examination of similar problems in many SSAs noted that scholarships have shown great promise in many settings including excluded groups in high-income countries of the OECD. In these countries, supplementary investments, engagement of parents, and other targeted initiatives have been required to overcome problems with access to existing educational opportunities. The authors suggested that other interventions such as school feeding programs, help with textbook and school uniform purchase, bussing, single sex schools, and positive discrimination in secondary school entry procedures can be used with selected groups of excluded girls with positive effects.

Equality of Outcomes with UPPET As previously noted, many countries have solved the problem of gender equal access to education, only to ﬁnd that either girls or boys fare less well in terms of academic and non-academic outcomes, such as conﬁdence building, leadership skills, and career aspirations. Uganda’s choice of a ‘Big Bang’ change model may well guarantee a rapid move to gender equality in access to post-primary education but it may fail to address another important aspect of gender bias, that of gender-differentiated outcomes. There is likely to be little evidence of how the ﬁrst cohort of girls currently entering UPPET fare in terms of the outcomes of their education until they take the exit examinations that marks course completion in January, 2010. However, studies of the causes of gendered biased school outcomes in both developed and developing countries point to the need to consider two factors security and a welcoming school environment.

110

JILL SPERANDIO AND ALICE KAGODA

Adolescence is a vulnerable time for girls. Long journeys to school, or boarding in hostels or with extended family close to the school, together with the risks presented by male teachers and classmates, make secondary schooling an unsafe environment in many countries for adolescent girls. Unwanted pregnancies affect both girls’ life-long opportunities and strain family resources and standing in the community (Ohsako, 1997; Mgalla, Boerma, & Schapink, 1998; Mirembe & Davies, 2001; Kim & Bailey, 2003; Leach, Fiscian, Kadzamira, Lemani, & Machakanja, 2003). The school environment can also be unfriendly to girls. The lack of latrines, poor conditions in hostels, lack of counseling or help with health issues, and an environment requiring a competitive approach and competition with male peers, is discouraging to girls from poor families who may lack the personal assertiveness and family support and understanding needed to persevere. A lack of female teachers and role models may exacerbate these problems, as may the lack of a critical mass of girls, where a traditionally male oriented culture has led boys to dominate in the classroom. A study of secondary school outcomes for girls in pre-UPPET secondary schools conducted by one author (Sperandio, 1998) indicated a number of ways in which both academic and non-academic outcomes were inﬂuenced by the school environment. There proved to be statistically signiﬁcant differences with regard to academic outcomes, with girls doing best in boarding schools whether single sex or coeducational. For the most able and least able girls the single sex schools enhanced academic outcomes, whereas high-ability girls did least well in coeducational day schools. These differences were explained in part by the demands made on the time of girls in day schools traveling and doing domestic chores once they reached home, time which girls in boarding schools could devote to additional studies. The more nurturing environment of single sex schools, with many women teachers and role models appeared to favor low and highability girls in particular. Conversely, high-ability girls did worse in the day schools where both the quality of teaching was lower, and levels of harassment, both in and out of school, much higher. The study also compared non-academic outcomes of secondary education for girls by examining attitudes and expectation, self-esteem, and opinions. Again, there were marked differences in the responses from girls in each school type. Single sex boarding school education, regardless of the area of the country which it was situated, appeared to foster a greater measure of self-conﬁdence and higher expectations regarding future careers that either of the other school types.

Learning from Experience

111

Other aspects of secondary schooling emerged from the study. Many of the students attending the coeducational city day schools were in fact boarding, either with extended family or in unsupervised privately operated hostels. Female teachers at these schools described the risks this presented to adolescent girls short of money and with no family support – the ‘sugar daddy syndrome’. Men would wait for girls outside the school and offer to pay school fees in exchange for sexual favors, taxi drivers would exploit girls on their journey to school, male students and teachers who would offer help with schoolwork for a similar price, and even male relatives in the extended family home would threaten to turn the girls out if they did not meet demands. Observations in coeducational secondary classrooms revealed differences in teacher attitudes and practices towards girls. Boys were more aggressive in answering questions, and the practice of calling students up to the front of the class to work math or science problems, or spell words on the chalkboard did not increase the comfort level of girls. Because of gender disparity at the time of the study in girls’ and boys’ enrollment to secondary schools, girls were in a minority in every class, and teachers at the secondary level were predominantly male. A survey of teacher attitudes regarding the ability of male and female students to learn and acquire skills indicated that both male and female teachers considered boys more intelligent and more likely to succeed. How useful are these ﬁndings as indicators of likely gender differences in outcomes in the expanded secondary education system that is now in place in Uganda? Girls and boys are now entering the secondary system in roughly equal numbers. However, girls score lower on the primary school leaving examinations, perhaps reﬂecting the interaction of class size, domestic responsibilities, and teaching methods that favor boys rather than girls. These lower scores, given the current method of selection for secondary school, suggest that boys are most likely to get their choices of places in the better government schools. Unless these schools choose to balance their gender intake classes, these better schools are likely to have more girls than boys, a situation that is not favorable to girl’s outcomes. If they do balance their classes, the girls that enter are likely to have lower achievement on the PLE than the boys, and without differentiated teaching, this could also put them at a disadvantage. Most of the government-aided single sex boarding schools are among the highest achieving schools as indicated by their public examination results at exit level examinations for the secondary school course. They accept only girls with high exit scores on the PLE. Their boarding and additional

112

JILL SPERANDIO AND ALICE KAGODA

charges place them outside the UPPET scheme, so that a high-ability but poor girl will be unable to access them. Thus, the options for most girls are now coeducational day or lower ranking boarding schools, or a participating private school. Given the large numbers of students that enrolled in the government schools in the ﬁrst year of USE, the government appears to be limiting the 2008 intake, and has brought more private schools into the scheme. For poor children whose parents cannot afford boarding fees of any kind, the option can only be a local school either government or private, and in some districts these do not exist. If parents can afford boarding, there is no guarantee that this will be on a school campus or supervised by the school, and adolescent girls immediately become more vulnerable in these situations. Completion, academic attainment, and non-academic outcomes may continue to be compromised for both boys and girls, but particularly for girls, by the current secondary school selection process (Sperandio, 2000).

Achieving Equality of Outcomes The issues for Uganda now revolve around providing gender parity in the educational outcomes from both primary and secondary education. The country has considerable experience of gender interventions. These have alternated between small-scale projects to sector-wide improvements according to the preferences of donor agencies and the recommendations of international institutions. Uganda can also draw on the experiences of developed countries such as the USA and UK in tackling girls’ underachievement in mathematics and science in the 1980s, and the research undertaken to examine discrimination in the classrooms of a number of SSA countries in the 1990s, which contributed to gender disparity in educational outcomes. Initiatives by the Forum for African Women Educationalists (FAWE) and the Female Education in Mathematics and Science in Africa Project (FEMSA) in Uganda have made important contributions to both focused attention on issues of gender equal student outcomes and culture-speciﬁc teaching (Kasente, 2003; Sutherland-Addy, 2008). Some of the ongoing gender-related issues that Uganda must face are discussed below: Girl Friendly Teaching and an Encouraging Class Environment for Girls The Uganda government has already accepted and acted upon a number of recommendations for making school environments friendlier to girls.

Learning from Experience

113

Important among these is the provision of separate latrine blocks in schools. The government is currently conducting workshops for teachers to help them provide counseling and guidance on a range of personal issues from health to relationships that are likely to be of concern to female students. Secondary schools are required to have one male and one female administrator in key positions, as well as a teacher with speciﬁc responsibility for girl students. But some aspects of creating an enabling environment for girls still need to be addressed. First, the gender composition of classes in coeducational schools needs examination to ensure that there is a critical mass of girls in each class, and that they have access to female teachers and role models. Second, there is a need to educate male students and teachers regarding the harassment of female students. The third aspect is the importance of encouraging teachers to develop classroom techniques that ensure girls get an equal opportunity to answer and discuss and showcase their achievements in the classroom. Finally, the building of self-conﬁdence and selfesteem in young women has not been a part of traditional Ugandan culture, and the ongoing education of the public regarding equal rights for women in Ugandan society must continue. Female Teachers and Role Models Secondary school teaching is currently dominated by male teachers, primarily as a result of the low numbers of girls accessing and completing secondary education and continuing on to teacher training courses (Sperandio & Kagoda, 2008). The interactions between male teachers and adolescent schoolgirls continue to be a worrying aspect of secondary schooling for many parents. Women teachers are currently taking on the burden of guidance and counseling of girls in schools without, for the most part, training in these ﬁelds. Most trained primary teachers and graduate secondary teachers are reluctant to move to rural schools away from the amenities of urban areas. For women teachers and administrators, such a move also requires considerable upheaval for the families and the problems of ﬁnding work for a male partner. Offering opportunities for local teacher training for women whose spouses work in rural areas is a solution already under consideration by the government, and one that has worked in other poor, rural countries such as the Bangladesh Rural Advancement Committee Initiative in Bangladesh (Sperandio, 2008). In the teacher training colleges, women pass ﬁnal examinations at a much lower rate than men. Fewer women teachers result in fewer women school

114

JILL SPERANDIO AND ALICE KAGODA

administrators, with women occupying only 12% of the secondary school headships in 2005 (MoES, 2005). This, in turn, leads to fewer women in educational leadership where important decisions about the form and content of education are made. The lack of women role models in the operation and direction of schools is unlikely to encourage girls to enter teaching, or consider educational leadership as a career. As new teachers are recruited to cope with the increased intake in secondary schools, attention needs to be given to the issue of gender balance on school faculties. Boarding and Hostels Considerable thought needs to be given to the issues that surround girls boarding in unsupervised hostels or with extended family members to attend secondary school. Despite the government’s progress on expanding existing schools and setting up ‘seed’ schools in areas where there were previously no secondary schools, many girls will be forced to ﬁnd a way to live closer to the schools where they have been offered a place. The conditions in these hostels, and the pressures that may be exerted on girls in extended family situations, can have a negative impact on academic outcomes. This issue was alluded to in the recent opening of a new boarding facility for girls in Gulu High School, where the representative of the sponsoring agency stated ‘‘We have put in place all the relevant facilities that a girl child may need in a dormitory for a healthier environment. This will help them excel and compete favorably with boys’’ (Eriku, 2008). Means for providing supervised boarding facilities, or for changing the selection process of schools to require them to accept students primarily from a local catchment area should be considered as a way of ensuring a girl’s entry into, and completion of, secondary schooling. Private Schools The current heavy reliance on private schools to provide places for primary school graduates raises concerns given the apparent lack of quality control in these schools, many of which have opened in response to the demand created by USE. In his survey of existing secondary school provision in 2002, Liang (2002) noted that apart from a few religiously founded elite private schools, most private schools were in rural or pre-urban areas (86%) and for the rural and the poor. Liang concluded that while there were statutory regulations for the licensing and registration of such schools, the government’s need to create secondary places had led to less rigorous policies, and there was little evidence that the schools were visited by an already overworked school inspectorate.

Learning from Experience

115

Liang (2002) also noted that a statistically signiﬁcant relationship existed between the existence of private secondary schools in an area and increased female enrollment. However, it was not clear whether private provision, in or of itself, increased access (especially for girls), or whether private providers responded to an unmet demand in certain districts (p. 58). Issues of school inspection and of the private schools ability to attract qualiﬁed teachers continue to be debated. A study of Ugandan female teachers’ career aspirations (Sperandio & Kagoda, 2008) indicated that qualiﬁed and experienced teachers would prefer to work in government or governmentaided schools, as this gave them civil service recognition as education ofﬁcers, with pension rights, job security and a guaranteed salary, all of which would be missing in the private sector. This suggests the need for government support of teachers in private schools if the distribution of trained and experienced teachers is to be more equitable. The current issue is whether these schools, which can clearly rely on the enrollment of girls from the local area and the ensuing government subsidies, will provide the same quality of education and the same educational outcomes for girls as government schools. The secondary exit examination results suggest that the top tier of these schools provide equal outcomes for boys and girls and rival the results of well-established government schools. In one such private school, whose motto is ‘‘Hard Work’’, and whose headmaster was quoted as explaining the secret to the ever-improving examination results at the schools as being ‘‘strictness on discipline, teachers’ commitment, and about all God’s blessings’’, 25 of the top 50 highest scoring students on the exit examination in 2006 were girls (Mbanga, 2006). However, the issue of the quality of education being delivered in most rural private schools is a matter of current debate in the Ugandan press, where a recent article entitled ‘‘Why quack schools exist despite government regulations’’ (Womakuyu, 2008) described the poor conditions in many of these schools and the problems connected with registration and inspection. The article also quoted local education ofﬁcials and parents perceptions of the reasons why poor quality schools were allowed to continue to function. These emphasized the lack of choice for poor parents who could not contemplate sending their children to schools in other parts of the country even when these children scored high enough on the primary leaving examination to be offered places in them. The government is clearly aware of this problem, and is taking measures to improve the situation. A MoES spokeswomen quoted in the newspaper article explained that an Education Standards Agency had been formed with

116

JILL SPERANDIO AND ALICE KAGODA

regional ofﬁces countrywide to check on school quality. The spokesman called on parents to support ministry policies and play their part by not enrolling their children in substandard schools.

CONCLUSION Uganda’s willingness to ‘take the plunge’ in the provision of 11 years of universal education has demonstrated how quickly gender parity in enrollment can be achieved. The problems involved in introducing and improving the UPE structure, and the solutions that have evolved for coping with them at both the local and national level, have given the government a wealth of experience to draw on as it moves into the second year of implementing UPPET. Recent improvements to the MoES statistics department now allows from much improved data collection and interpretation, so that developing trends can be identiﬁed and ongoing inequities monitored. Given the scale of the change initiative UPPET represents, it is inevitable that there will be some unexpected outcomes, particularly when the results of the provision of education by private schools are still a comparatively unknown variable. Issues of gender equality are prominent at all levels of Ugandan society due in large measure to a vocal and active women’s movement that has promoted the ‘girl child’ issue, and undertaken a campaign to improve all aspects of women’s lives in Uganda through legislation. As an example of this, the Ugandan Parliament is currently considering legislation to make female circumcision (genital mutilation) punishable by ﬁne or imprisonment. Although, there is general awareness of the need for equality for women in both the government and the population at large, meeting gender targets in government institutions has been slow. Women are underrepresented in all policy and decision-making positions, in school leadership, and in teaching. They are still ﬁghting the battle for equal property rights. In some areas of the country traditional gender roles are still ﬁrmly entrenched. Given the combination of geographic isolation, rural poverty and disease that consume the lives of the population on a daily basis, change is likely to be resisted and long in coming. However, creating a culture in the country that accepts gender equality as given, will remove many of the hurdles girls still face in the education system. Thus, from a gendered perspective, the challenge in Uganda is now one of the ensuring equal educational outcomes for girls and boys. Within this context, the plight of poor rural girls is particularly demanding of attention.

Learning from Experience

117

The ideal situation for these girls would be an education in a governmentsupervised single sex boarding school at the secondary level. But this is clearly not an option, so afﬁrmative action for these girls in the form of bursaries to cover boarding and travel costs, together with adjustments to the primary leaving scores girls must attain to access secondary school, should be considered. Again, the government has experience of success with such measures having undertaken a successful afﬁrmative action campaign to eliminate gender disparities in enrollment at Makerere University in Kampala. Finally, consideration should be given to dismantling the current system of secondary school selection for primary school graduates on the basis of their examination scores. This system relies on students applying to schools that are well matched to their likely primary exit examination results (and poor children and their parents are unlikely to have this information without guidance) and schools choosing which students they will take. Although, this practice continues it maintains a hierarchy of schools that favors those students with educated and ﬁnancially well-to-do parents. Allocating students to schools, to ensure that each school has a range of abilities, and an equal number of girls and boys in coeducational schools, would be a way to address this issue. Although the development of regional schools is important, it is likely to reinforce regional differences, rather than encouraging national unity, but allocating students to schools would also offer the opportunity to ensure an ethnic mix in each school, if boarding facilities could be provided. This could prove important in ensuring equal educational outcomes for poor girls from certain regions of the country. In the 1960s, Uganda had one of the largest education systems in SSA and one of the most respected for the quality of the services it offered. The system has shown remarkable resilience and an ability to adapt when tested to the limit – by civil strife, disease, and economic bankruptcy. Once again it is taking a leading role in SSA in addressing issues of social justice, and providing a population thirsty for education with the opportunity to access it. But one of these issues, that of ensuring gender parity in access to the education system, has been met. The next three years, as the ﬁrst cohort of students who have gained access to the system through both UPE and UPPET complete their secondary education, will show the level of gender parity in educational outcomes that has been achieved and the problems still to be addressed. In the mean time, Uganda’s initiatives provide a wealth of experience to guide other countries around the world committed in achieving gender equality in education sooner rather than later.

118

JILL SPERANDIO AND ALICE KAGODA

REFERENCES Aber, P., Akena, C., & Oper, D. O. (2007, December 14). Parents who fail to educate children to be punished – Govt. Monitor Online. Available at http://www.monitor.co.ug. Retrieved on February 11, 2008. Bennell, P. (1996). Rates of return to education: Does the conventional pattern prevail in subSaharan Africa? World Development, 24, 183–200. Bregman, J., & Armstrong, A. (2003). First regional conference on secondary education in Africa: Report on the proceedings. Washington, DC: SEIA Publication, Africa Region Human Development. Bruns, B., Mingat, A., & Rakotomalala, R. (2003). Achieving university primary education by 2015: A chance for every child. Washington, DC: World Bank. Castle, E. (1996). Growing up in East Africa. London: Oxford University Press. Clemens, M. (2004). The long walk to school: International education goals in historical perspective. Washington, DC: Center for Global Development. DFID. (2006). Evaluation of DFID’s policy and practice in support of gender equality and women’s empowerment, Vol. 1, Synthesis Report, Evaluation Report EV669. DFID, London. Dimmock, C. A., & Walker, A. (2005). Educational leadership: Culture and diversity. London: Sage. Elliot, C. M., & Kelly, G. P. (1980). Perspectives on the education of women in third world nations. Comparative Educational Review, 1–12. Eriku, J. (2008, February 22). Gulu High gets modern dormitory. The Monitor. Available at http://www.monitor.co.ug. Retrieved on February 20. Floro, M., & Wolf, J. (1990). Economic and social impact of girls’ education in developing countries. Washington, DC: Agency for International development. Fullan, M. (1993). Change forces: Probing the depths of educational reform. London: Falmer Press. Fullan, M. (2001). The new meaning of educational change (3rd ed.). New York: Teachers College Press. Fullan, M. (2008). The six secrets of change. London: Sage. Fuller, B., & Clark, P. (1994). Raising school effects while ignoring culture? Local conditions and the inﬂuence of classroom tools, rules and pedagogy. Review of Educational Research, 64, 119–158. Gleick, J. (1999). Faster. New York: Pantheon Books. Global Campaign for Education Report. (2008). World school report: No excuses. Available at http://www.ungel.org/resources/1612_1737.html. Retrieved on December 2008. Hansen, H. B., & Twaddle, M. (1998). Developing Uganda. London: James Curry. Herz, B., & Sperling, G. B. (2004). What works in girls’ education: Evidence and policies from the developing world. New York, NY: Council on Foreign Relations. Herz, B., Subbarao, K., Habib, M., & Raney, L. (1991). Letting girls learn: Promising approaches in primary and secondary education. Washington, DC: World Bank. Homer-Dixon, T. (2000). The ingenuity gap. Toronto: Knopf. Hyde, K. (1989). Improving women’s education in sub-Saharan Africa: A review of the literature. PHREE Background Paper Series 15. Washington, DC: World Bank. Kasente, D. (2003). Gender and education in Uganda. Background paper commissioned for the EFA global monitoring report 2003/4, Gender and education for all: The leap to equality. UNESCO, Paris, France.

Learning from Experience

119

Kasente, D., Nakanyike, M., & Balihuta, A. (2003). Improving the management of sexual maturation at primary schools: Information needs from the perspective of Ugandan school children. Kampala: Makerere Institute of Social Research, Makerere University. Kim, J., & Bailey, S. (2003). Unsafe schools: A literature review of school related gender based violence in developing countries. Arlington, VA: Development and Training Services, Inc. King, E., & Hill, A. (1993). Women’s education in developing countries: Barriers, beneﬁts, and policies. Baltimore: Johns Hopkins University Press. Leach, F., Fiscian, V., Kadzamira, E., Lemani, E., & Machakanja, P. (2003). An investigative study into the abuse of girls in African schools. London: DFID. Lee, V. E., & Lockheed, M. (1990). The effects of single-sex schooling on achievement and attitudes in Nigeria. Comparative Education Review, 34, 2. Lewin, K., & Calloids, F. (2001). Financing secondary education in developing countries: Strategies for sustainable growth. Paris: IIEP/UNESCO. Lewis, M., & Lockheed, M. (2006). Inexcusable absence: Why 60 million girls still aren’t in school and what to do about it. Washington, DC: Center for Global Development. Liang, X. (2002). Uganda post primary education sector report. Human Development Sector. Africa Region. Washington, DC: The World Bank. Lirri, E. (2008, January 19) PLE: More boys score ﬁrst grade. Monitor Online. Available at http://www.monitor.co.ug. Retrieved on February 11. Lirri, E., & Wossita, S. (2008, January 25). Senior 1 government intake up. Monitor Online. Available at http://www.monitor.co.ug. Retrieved on February 11. Lloyd, C., Mensh, B., & Hewett, P. (2000). The spread of primary schooling in sub-Saharan Africa: Implications for fertility change. Population and Development Review, 26, 483–515. Lukenge, J. (1972). Who goes to the private schools in Uganda: Patterns of social recruitment to a ‘‘Second chance’’ academic alternative. Mimeo, University of Dar Es Salaam, Tanzania. Mbabazi, J. (2007). Implementation of UPPET program and economic growth in Uganda. Paper presented at the 9th Oxford International Conference on Education and Development, 11–13 September. New College, Oxford, UK. Mbanga, J. (2006). East high school reigns. The Weekly Observer. Kampala, Uganda, p. 30. Mensh, B. S., Bruce, J., & Greene, M. S. (1998). The uncharted passage: Girls’ adolescence in the developing world. New York: The Population Council. Meyer, J. W., Ramirez, F. O., & Soysal, Y. N. (1992). World expansion of mass education, 1870–1980. Sociology of Education, 65, 128–149. Mgalla, Z., Boerma, J., & Schapink, D. (1998). Protecting school girls against sexual exploitation: A guardian program in Mwanza, Tanzania. Reproductive Health Matters, 7, 19. Mingat, A., & Tan, J. P. (1996). The full social returns to education: Estimates based on countries’ economic growth performance. Washington, DC: World Bank. Mintzberg, H., Ahlstrand, B., & Lampel, J. (1998). Strategy safari: A guided tour through the wilds of strategic management. New York: Free Press. Mirembe, R., & Davies, L. (2001). Is schooling a risk? Gender, power relations, and school culture in Uganda. Gender and Education, 13(4), 401–416. MoES. (1998). Factors inﬂuencing effectiveness in primary schools. Kampala, Uganda: Improving Education Quality Project, National Examinations board, Ministry of Education and Sports. MoES. (2004, August). The national report on the development of education in Uganda at the beginning of the 21st century. Kampala, Uganda: Education Planning Department, Ministry of Education & Sports.

120

JILL SPERANDIO AND ALICE KAGODA

MoES. (2006, October). The education sector annual performance report. Kampala, Uganda: Education Planning Department, Ministry of Education & Sports. Mugo, B. (2007). Transition to post primary education: Focus on girls. Speech given at the United Nations Girls’ Edication Initiative Regional Meeting for East and Southern Africa, 17th February, Addis Ababa, Ethiopia. Murphy, P., Bertoncino, C., & Wangi, L. (2002). Achieving universal primary education in Uganda: The ‘Big Bang’ approach. Education Notes, Human Development Newsletter, April, 2002. World Bank, Washington, DC. Nkanyike, M., Kasante, D., & Balihuta, A. (2002). Attendance patterns and causes of dropout in primary schools in Uganda. Kampala, Uganda: Ministry of Education and Sports. Nsubuga, Y. K. (2006). The case for the universal secondary education in Uganda. Available at http://www.education.go.us/USE.htm. Retrieved on November 30. Ohsako, T. (1997). Violence at school: Global issues and interventions. Paris: United Nations Educational, Scientiﬁc, and Cultural Organization-International Bureau of Education. Psacharopoulos, G., & Patrinos, A. (2004). Returns to investment in education: A further update. World Bank Policy Research Working Paper 2881, World Bank, Washington, DC. Raymond, M., & Sadoulet, E. (2003). Educational grants closing the gap in schooling attainment between poor and non-poor. Berkeley, CA: Department of Agricultural and Resource Economics, University of California. Ridell, A. (2003). The introduction of free primary education in sub-Saharan Africa. Background paper for EFA Global Monitoring Report 2003/4. The Leap to Equality. Paris: United Nations Educational, Scientiﬁc and Cultural Organization. Ridell, A., & Nyagura, L. (1991). What causes differences in achievement in Zimbabwe’s secondary schools? Policy Research and External Affairs Working Paper, WPS 705. World Bank, Washington, DC. Rihani, M. A. (2006). Keeping the promise: Five beneﬁts of girls’ secondary education. Washington, DC: Academy for Educational Development. Rose, P. (2003). Communities, gender and education: Evidence from sub-Saharan Africa. Background paper for UNESCO, Global Monitoring Report 2003. United Nations Educational. Cultural, and Scientiﬁc Organization, Paris. Siminyu, F. (2008, January 23). Mayuge’s best two might miss secondary. Monitor Online. Available at http://www.monitor.co.ug. Retrieved on February 11. Sperandio, J. (1998). Girls’ secondary education: Unintended outcomes of well-intentioned policy. Doctoral dissertation, University of Chicago, Chicago, IL. Sperandio, J. (2000). Leadership for adolescent girls: The role of secondary schools in Uganda. Gender and Development, 8(3), 57–64. Sperandio, J. (2008). Preparing women for educational leadership: Opportunities in non-formal education in Bangladesh. In: H. Sobehart (Ed.), Women leading education across the continents. New York: Rowland and Littleﬁeld Education. Sperandio, J., & Kagoda, A. M. (2008). Advancing women into educational leadership in developing countries: The case of Uganda. Advancing Women into Leadership, 24(Spring). Ssekamwa, J. C. (1997). History and development of education in Uganda. Kampala, Uganda: Fountain Press. Stash, S., & Hannum, E. (2001). Who goes to school? Educational stratiﬁcation by gender, caste and ethnicity in Nepal. Comparative Education Review, 45, 354–378.

Learning from Experience

121

Steinkamp, M., & Maehr, M. (1984). Gender differences in motivational orientations toward achievement in school science: A quantitative synthesis. American Educational Research Journal, 21, 39–59. Stromquist, N. (1989). Determinants of educational participation and achievement of women in the third world: A review of the evidence and a theoretical critique. Review of Educational Research, 59, 143–183. Sutherland-Addy, E. (2008). Gender equity in junior and senior secondary education in Sub Saharan Africa. World Bank Working Paper No. 140. African Human Development Series, World Bank, Washington, DC. Swainson, N. (1996). Redressing gender inequalities in education: A review of constraints and priorities in Malawi, Zambia and Zimbabwe. Harare: DFID. Swainson, N., Bendera, S., Gordon, R., & Kadzamira, E. (1998). Promoting girls’ education in Africa: The design and implementation of policy intervention (Serial No. 25). London, UK: Department for International Development. Tiberondwa, A. K. (1998). Missionary teachers as agents of colonialism: A study of their activities in Uganda, 1877–1927. Kampala: Fountain Publishers. Tripp, A. M., & Kwesiga, J. C. (2002). The women’s movement in Uganda: History, challenges and prospects. Kampala, Uganda: Fountain Press. UNESCO. (1999). Consortium on Secondary Education. Paris, France: UNESCO. UNESCO. (2001). UNESCO/BREDA World Bank Regional Workshop in Mauritius. Paris, France: UNESCO. UNESCO. (2006). How many children in Africa reach secondary education? UNESCO Institute for Statistics Fact Sheet. January 2006. Available at http://www.uis.unesco.org/ TEMPLATE/pdf/EducGeneral/UIS.Fact%20Sheet.2006.02SecAfr.EN.pdf. Retrieved on June 14. Were, M., & Nafula, N. N. (2003). An assessment of the impact of HIV/AIDS on economic growth: The case of Kenya. CESifo Working Paper Series No. 1034. Available at http:// ssrn.com/abstract ¼ 449241. Retrieved on January 2008. Womakuyu, F. (2008, February 19). Why quack schools exist despite government regulations. New Vision Online. Available at http://www.allAfrica.com. Retrieved on February 11. Wossita, S. (2008, January 28). Government to reintroduce double shifts in secondary schools. Monitor Online. Available at http://www.monitor.co.ug. Retrieved on February 11. Yates, L. (1993). The education of girls: Policy, research, and the question of gender. Hawthorn: The Australian Council for Educational Research Ltd.

SUGAR DADDIES AND THE DANGER OF SUGAR: CROSS-GENERATIONAL RELATIONSHIPS, HIV/AIDS, AND SECONDARY SCHOOLING IN ZAMBIA Monisha Bajaj ABSTRACT This chapter seeks to explore the nature of and motivations for crossgenerational relationships, and to examine how these relationships structure, limit and enable access to schooling for youth in Ndola (Zambia). Amidst increasing HIV infection rates and decreasing economic opportunity, youth experiences in and outside of school provide information about the impact of macro-level inﬂuences, particularly global economic trends and the HIV/AIDS pandemic, on the lives of these young women. Utilizing qualitative methods that seek to explore the lived realities of Zambian youth, this study examines perceptions of the phenomenon of ‘‘sugar daddies’’ and how they are seen to effect educational access and opportunity for young women. Although the study ﬁnds that young women are ﬁnding ways to cope with being enmeshed in a Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 123–143 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010007

123

124

MONISHA BAJAJ

context characterized by severe economic decline and an extensive HIV/ AIDS crisis, the strategy of securing a ‘‘sugar daddy’’ is one that may result in deadly infection and social isolation. Furthermore, policymakers in Zambia can and should take the opportunity to rethink austerity measures and hostility to social spending as well as the content of public health education.

Many scholars and practitioners have discussed the role of cross-generational relationships, and their often corresponding power asymmetries, in facilitating the spread of HIV/AIDS in sub-Saharan Africa (Longﬁeld, Glick, Waithaka, & Berman, 2002; Luke & Kurz, 2002; Vavrus, 2003). The material conditions framing many of these transactional sexual encounters have been linked to region-wide economic decline over the past two decades, including in Zambia (Garbus, 2003). The HIV infection rate in Zambia is 17 percent and, while many wealthy individuals are infected (UNAIDS, 2006), risky income generating activities such as prostitution and transactional sex have been identiﬁed as a primary cause of infection among the poor in Zambia, especially among poor women (de Waal & Whiteside, 2003).1 Once a middle-income country, 64 percent of Zambians now live on less than one dollar per day (UNDP, 2004). The worldwide drop in copper prices beginning in the 1970s had a ripple effect on other sectors of the economy, resulting in Zambia’s need to borrow large sums of money from international ﬁnancial institutions and accept the conditions of structural adjustment programs (SAPs), which are generally characterized by market liberalization, privatization of state-owned industries and services, and decreased government spending in the areas of health and education (Mehmet, 1997). Decreased government spending on health, education, and food subsidies, in turn, has rendered costly these previously free state provisions (Saasa, 2002). For example, the implementation of user fees for education, although abolished in 2003 at the primary level, has had an adverse effect on enrollment and completion rates in Zambia. Data appear to support this assertion: from 1985 to 1994, the ﬁrst decade of SAPs in the country, Zambia saw an overall decrease of 20 percent in the number of students completing grade seven, and a signiﬁcant drop in girls’ participation in education (Kelly, 1999). In the Copperbelt region, boosted by the prosperity of the copper industry in the 1970s and early 1980s, enrollment in primary schooling reportedly reached 100 percent; but by 1999, enrollment had dropped to just 79 percent (Kelly, 1999).

Sugar Daddies and Girls’ Schooling in Zambia

125

Although at the national level, government spending on education has decreased, resulting in a decline in both enrollment and educational quality, the public health crisis nationally has also limited the ability of Zambian families to bear the indirect costs of sending children – boys and more often, girls – to school. The HIV/AIDS pandemic has contributed to declining enrollment for young women when family resources become stretched thin and health expenditures take precedence over school fees. As a result, girls are often relied on for domestic responsibilities when a female guardian falls sick, prohibiting them from doing schoolwork or attending classes in many instances (Kelly, 1999). With school enrollment rates declining for African girls in certain countries, some young women have sought other avenues for fulﬁlling their desire for education as well as other commodities and services. Cross-generational relationships, particularly those involving a younger woman and an older ‘‘sugar daddy,’’2 have become increasingly prevalent in Zambia. Another, possibly related, danger to young women between the ages of 15 and 19 (high school age) is data which show that they are six times more likely to be HIV positive than their male counterparts (Glynn et al., 2001).3 It is important to note that cross-generational relationships have existed historically in many societies and in Zambia, these relationships overlap with longstanding cultural practices such as polygamy and widow inheritance (Gausset, 2001). Therefore, these relationships are not historically unprecedented, nor can they be reduced to economic motivations entirely. Apart from cultural practices, the scale and frequency of crossgenerational sexual relationships are unknown due to the inaccuracy of selfreports of sexual behavior and the associated social stigma (Plummer et al., 2004).4 This chapter, however, examines the economic dimension of these relationships as it pertains to educational access and opportunity in an era in which neoliberal economic policies have resulted in limited employment prospects, decreased government support for education, and heightened economic uncertainty for young women and their families (Saasa, 2002). The ‘‘sugar daddy’’ phenomenon – understood, as well, in light of historical, cultural, and social constructions of masculinity and femininity in Zambia – intersects with economic decline and HIV/AIDS by structuring, enabling, and limiting educational access and opportunity for young women in different instances. Given strong social taboos, discussions about crossgenerational relationships are often carried out in the third person; as a result, this chapter examines the perspectives of students, teachers, and parents with regard to the motivations for, nature of, and attitudes toward

126

MONISHA BAJAJ

cross-generational relationships in schools and communities in Ndola (Zambia).5 Ndola, capital of the once-prosperous Copperbelt province, was an instructive site to carry out research on educational experiences as an urban center greatly impacted by global economic trends, particularly demand for basic commodities. Ndola experienced high levels of urbanization in the mid-1900s due to the rapidly expanding copper industry (Burdette, 1988), followed by frequent return migration to villages once the copper industry began its decline in the late 1970s and early 1980s (Bond, 1982). Currently, Ndola has the highest rate of HIV infection in the country estimated at 28 percent (Glynn et al., 2001). As a result of these unique migration patterns and their outcomes, researchers have noted that the Copperbelt ‘‘becomes an especially productive site for rethinking anthropological ideas about history and modernity’’ (Ferguson, 1999, p. 25). In this social context, young women’s changing roles and the impact of social inequalities on educational opportunity are an important area of inquiry.

METHODOLOGY The information for this chapter was collected as part of a larger study on youth experiences with schooling in Ndola (Zambia) during one year of ﬁeldwork spanning 2003 and 2004. The data in this chapter come from semistructured interviews with approximately 75 students, teachers, parents, and alumni of Ndola private and government secondary schools through individual interviews and focus groups. The majority of participants lived in the Pamodzi township or around the Chifubu market, neighborhoods home to largely lower- to middle-class families. I primarily relied on qualitative methods in this study. I utilized participant observation, visited and observed classes at approximately 10 government and private schools in and around Ndola, and attended teacher trainings on occasion. Additionally, 22 high school pupils (grades 8 through 12) were identiﬁed through criterion sampling to participate in the ‘‘sibling pair cohort,’’ comprised of siblings of similar age who attended different secondary schools. Pupils in the cohort were interviewed and given a conﬁdential research notebook to complete written assignments on their attitudes toward schooling, HIV/AIDS, politics, and their futures. Additionally, each sibling cohort member was given a ‘‘free-write’’ section in his/her notebook to discuss whatever issue he or she deemed important. The notebooks were completed over a three-month period and collected in June 2004.

Sugar Daddies and Girls’ Schooling in Zambia

127

Following the ﬁeldwork period, the data were coded for signiﬁcant themes and then organized according to those responses that were representative. The information on cross-generational relationships presented below comes from the interviews, observations, and written materials collected during the research period primarily from the sibling cohort group, as well as from general interviews with other pupils, parents, secondary school graduates, and teachers. The methods utilized sought to address the following question: In what ways do youth and adult discussions of cross-generational relationships reﬂect their role in limiting, structuring, and/or enabling educational access and attainment for young women in Ndola (Zambia)? The respondents frequently referenced social attitudes and stigma around such relationships; thus, data are presented to offer a picture of the social landscape in which young women pursue or acquiesce to the advances of older men, due, in part, to their material conditions and educational aspirations. An overview of previous research on masculinities, power, and HIV/AIDS is necessary in order to situate the subsequent data within its cultural and social context.

CONSTRUCTING THE SUGAR DADDY PHENOMENON This chapter focuses on perceptions of mostly consensual (though often coerced) sex; however, the data collected in this study suggest that some of the same motivations that drive the rape of young girls also motivate crossgenerational relationships in which both parties are willing.6 The sugar daddy phenomenon in Zambia can be seen in its relationship to material inequities and as an extension of constructed masculinities (Simpson, 2005; Varga, 2001). Simpson (2005) discusses how Zambian men learn about sex and the exercise of power through aggression and sometimes violence. Through practicing and learning about their gendered roles with regard to sex, ‘‘the expression of male sexual identity was often ﬁgured as an inherently violent activity in which, in competition with other men, the conquest of women was the central element. Male ‘superiority’ had to be demonstrated’’ (p. 585). Simpson further notes that expressions of virility are also linked to multiple sexual partners as well as power over them, elements that often characterize cross-generational relationships. Although it is important to note that the social constructions of masculinity can contribute to the prevalence of crossgenerational relationships, neither all men nor all women accept these roles

128

MONISHA BAJAJ

unquestioningly. Even when women assume subservient positions, such as those induced by cross-generational relationships, a notable shift occurs from victim to agent in analyzing their actions and decisions. Scholars have increasingly acknowledged the sexual agency of African women in general (Reddy & Dunne, 2007) and young women engaged in cross-generational relationships in particular (Hunter, 2002). In his work on cross-generational relationships in South Africa, Hunter (2002) offers a distinction between young women engaging in transactional sex for subsistence and those engaging in it for consumption. The use of income generated from cross-generational relationships for increased educational opportunity does not neatly ﬁt into either category but perhaps lies somewhere along the spectrum between subsistence and consumption. Whatever the use of the income generated, the line between consent and coercion is often blurred by the ‘‘gendered material inequalities that provide a basis for such [cross-generational] transactions’’ (Hunter, 2002, p. 116). Given the complexity and diversity of cross-generational relationships, particularly with regard to consent, power, and motivation, it is useful to conceptualize what Weissman et al. (2006) have termed the ‘‘continuum of volition’’ with regard to such relationships. In analyzing the various factors that structure cross-generational relationships, the authors offer a useful framework for understanding the knotty problems of consent and coercion (Fig. 1). It is important to note that not all cases ﬁt in this schema. Crossgenerational relationships in Zambia include phenomena such as children

Voluntary Sex

Economically Driven/ Economically Rational Sex

Coerced Sex

Drivers: Emotional security, love, pleasure, social status

Material comfort, security (gifts, niceties)

Life maintenance (school fees, shoes, uniforms)

Survival needs (food, housing)

Insecurity, fear of physical or other harms

Economic Security

Fig. 1.

Reasons/Drivers for Sexual Activity along a Continuum of Volition. This Diagram has been Re-printed with the Permission of the Authors.

Sugar Daddies and Girls’ Schooling in Zambia

129

being raped because of the perceived curative powers of such sex. Here, of course, the issue of volition is irrelevant and this chapter does not address such practices (Leclerc-Madlala, 2002). In Weissman et al.’s diagram, economic rationale are but one motivation for young women entering into cross-generational relationships, and school fees are but one use of the money obtained through such, although peers and elders often assumed that economic motivations were the sole cause of young women seeking out, or acceding to the advances of, older men. With these complex constructions of gender in mind, I now examine the ways in which secondary school students encounter evidence of cross-generational relationships in their schools and communities.

CROSS-GENERATIONAL RELATIONSHIPS IN AND AROUND SCHOOLS Motivations for Cross-Generational Relationships Various motivations that fall along the spectrum between subsistence and consumption (Hunter, 2002) exist for cross-generational relationships. For young women, having a ‘‘sugar daddy,’’ or an older man with economic resources, can help offset the ﬁnancial costs of secondary schooling. Mr. Mayombo, a teacher at Umutende government secondary school in Ndola, discussed a student in his eleventh grade English class and her ‘‘sponsor’’ or sugar daddy:7 The phenomenon of sugar daddies is quite common deﬁnitely, from the primary school, secondary school, even at the tertiary level including the university. I have noticed quite a few of them; some of them are being paid for by these taxi drivers and minibus drivers. Some of them are more or less like full-time wives; when she is at school, she’s a schoolgirl. When she goes home, she doesn’t go to her parents’ home, she’s in the man’s house. But yet, that’s not a formal marriage. One girl is 18 years of age, she’s in grade 11. I have seen three of them [in my class]. I found out that their parents couldn’t afford their education, but the girls are really brilliant. In the end, they are being promised to be married to those sugar daddies. There is something very unfair in the state of affairs. These girls will be made to get married to persons who they do not want to marry simply because they have an obligation to fulﬁll. The man that I know who is with Eunice is 45 years of age and she’s 18. Parents know, [but] they have no money to pay for school. He’s a minibus driver and has two wives, legally. She . . . is being anticipated to become the third wife. In Zambia, polygamy is legal. This man has volunteered to pay her fees up to whichever level the girl will go to, as long as she is obliged to bend towards the man’s requirements. She lives with the parents, [but] she has an obligation; she can be taken at

130

MONISHA BAJAJ

anytime by the man and she is the most favored amongst the three women because she’s the youngest, and to him, she’s sweet.

Mr. Mayombo highlights the need for school fees as a factor that contributes to female secondary school students having relationships with older men. The desire for education and the lack of ﬁnancial resources for it were among the motives offered by many adults and schoolgirls for why young women engage in cross-generational sexual relationships. Young women’s motivations for cross-generational relationships were cited as primarily economic, but older men’s motives varied. As mentioned earlier in this chapter, cross-generational relationships have existed in Zambia historically, particularly given the polygamous traditions of many ethnic groups where signiﬁcant age differences between co-wives and their husband is common (Gausset, 2001). Amidst the current economic situation of decline and the prevalence of HIV/AIDS in the urban townships of Ndola, two primary reasons, aside from cultural or personal preferences, emerged in this study for older men entering into cross-generational relationships. These reasons were, ﬁrst, the perception of lower risk for contracting HIV/AIDS from younger girls and second, the belief among some men that a sexual relationship with a virgin could cure one of HIV/ AIDS.8 One parent, Mrs. Lombanya, noted the motivations for older men to engage in sexual relationships with schoolgirls, citing a colleague of her husband discussing his relationships with younger women: There was a time my husband was talking about someone and said, ‘‘That guy, I wonder why he likes going out with schoolgirls?’’ He had an opportunity to ask him and he said, ‘‘You know the young ones, they are fresh. They haven’t met any man. So meeting her, I’m assured that she’s not sick.’’ And so he’d go for a schoolgirl, thinking that ‘‘She’s all right; she’s not sick.’’ Forgetting that now he’s the one who is going to make her sick. And then this girl will go for another man. You know they can’t go for their fellow boys; the boys have nothing to offer them. What they want probably is what they can’t get from home.

Aside from the preference for ‘‘fresh’’ girls, the age differential also corresponds with the power dynamics of learned masculinities as noted by Simpson (2005). Instructively, it is important to note the increasing presence of cross-generational relationships between older women and younger men in Zambia, driven perhaps less by conventionally gendered roles than by more immediate social and economic realities. Sugar Mommies The incidence of transactional sexual relationships between older women and younger men offers an interesting counterpoint to the sugar daddy

Sugar Daddies and Girls’ Schooling in Zambia

131

phenomenon as rooted in discourses of masculinity. Participants’ accounts suggested that the phenomenon of sugar mommies was less prevalent, but nevertheless present with the reported motivations being sex and companionship. Although young men may enter into such relationships for economic reasons similar to those of their female counterparts, older women may be responding to the growing widowhood among middle-aged women due to deaths from HIV/AIDS rather than to the belief that young men might cure them of the disease. An Ndola social worker, Lucy, commented: We are seeing sugar-mommies going for small boys; those [ladies] who have money. Maybe it’s because most of the women have lost their husbands due to HIV and AIDS. Some are left with a lot of money so they can do whatever they want to lure the young boys so that they sleep with them. In turn, it’s the sugar mommy giving the small boy money. Especially the school-going and those young men who are in employment, who aren’t married and they want some extra money. Some are not in employment; they just want someone to look after them.

Although these older women may be reacting to the loss of their spouses when seeking out a young sexual partner, they, too, are placing young men at risk for contracting the virus when a woman’s partner has already passed away from AIDS. Although the phenomenon of sugar mommies merits mention, most cross-generational relationships continue to involve younger women and older men, and accordingly this chapter focuses primarily on this type of relationship. Cross-Generational Relationships in School and Society Examining the way a cross-generational relationship begins highlights further the social inequality between older men with resources and female students without them. Respondents described the types of signals and interactions that lead young women to understand that a sexual arrangement is being negotiated. In a cultural context where elders who are not related are often addressed as ‘‘mama,’’ ‘‘aunty,’’ ‘‘uncle,’’ or ‘‘bashikulu’’ (grandfather), it is important to note the differences between a non-sexual and sexual relationship between an adult and an adolescent, and how these relationships may change over time. The origins of such cross-generational relationships were discussed in detail in some student diaries. For example, Agripa, who had previously studied at a government high school in the Copperbelt town of Mufulira, discussed under a topic, which he titled ‘‘sexual abuse in school,’’ the ways that teachers and students initiated sexual relationships in school: On this page, I’m gonna holla about teenage pregnancies in schools and about how teachers manipulate the girls into having sex with them. In our government schools, it is

132

MONISHA BAJAJ

very common to hear about a teacher having an affair with a pupil. No teacher has ever been dismissed from work for ﬂirting with a pupil, more still for making her pregnant. The girl pupils are enticed with special favors like special treatment, immunity from beatings, leakages [of answer papers] during exams, promises of money or high life, etc, in exchange for a lust ﬁlled relationship because it certainly isn’t love.

Agripa’s response indicates disdain for such relationships between his female classmates and teachers. Many other respondents, both students and teachers, noted the prevalence of cross-generational relationships within secondary schools. When asked why these relationships took place, responses ranged from mere attraction and human nature to coercion and incentives offered by teachers to young women willing to become sexually involved with them. Students discussed how relationships with their teachers changed over time and how these changes were often noted by their peers. One student, Precious, noted how what seemed like a normal student–teacher relationship changed: My teacher sent me to his ofﬁce to collect some textbooks and I did so. When I reached the ofﬁce, I found that the door was locked so as I turned I saw him staring at me as if he had seen an angel or a ghost. He knew what the problem was and he opened the door for me. I hesitated to enter, but he told me that I was wasting his time. So I entered and he tried to grab my breasts and he held my hand saying that he loved me so much, but I managed to run away from the ofﬁce.

Although students reported being unsure about the nature of female students’ relationships with their male teachers, many reported the practice of teachers’ explicit solicitation of sex from girls, offering incentives, such as good grades and extra lessons, in exchange. A male student in the tenth grade noted the way that sexual resources could be exchanged for favoritism in class: The teacher is teaching and he’s just looking at the same particular girl. Then when the subject is over, they go and talk. Sometimes maybe the whole class is being beaten by the teacher, but not the girl. There was a day [when] my friend came to tell me that this is what happened in their class. They were all writing an exam. They were given the scripts to start writing. Then, the teacher started writing something very fast. When he ﬁnished, he went to a girl’s desk, he took away the paper the girl was writing, and he gave her the paper he was writing. And the girl passed.

In this student’s account, the advantages of engaging in a relationship with a teacher were noted, but some of the adverse consequences of such relationships were more clearly seen by young women who chose not to engage in them.

Sugar Daddies and Girls’ Schooling in Zambia

133

Some students were skeptical about the beneﬁts of cross-generational relationships and highlighted instead the risks. A twelfth grade student at Chilemba government high school, related the following process by which teachers and pupils become involved and the risk of acquiring AIDS when this happens: ‘‘The teachers mingle with pupils. When I was in grade ten, they found this biology teacher – he even died last year – he was caught with a grade eight girl in his ofﬁce having sex.’’ Sexual relationships between students and teachers, whatever their cause, include power asymmetries that may increase the risk involved in such encounters since teachers are one of the groups hardest hit by HIV/AIDS in Zambia. It is estimated that Zambia will have lost nearly 24,000 primary and secondary school teachers between 2000 and 2010 due to AIDS-related illnesses (Bennell, Hyde, & Swainson, 2002). Consequently, the adverse future impacts of such relationships – ranging from social isolation and abandonment to HIV infection and death – are sometimes outweighed by the immediate beneﬁts, whether real or perceived. Student accounts describe how teachers encourage young women to get into sexual relationships through promises of passing exams, and they offer examples of how such relationships begin and end to the detriment of the student. Young girls who enter into relationships with teachers are often economically disadvantaged and seek to rectify their limited access to money to obtain favors and privileges by securing these things through sex. These relationships, and the corresponding outcomes for young girls, such as unwanted pregnancy, contraction of HIV and other sexually transmitted diseases, or dropping out of school, serve as evidence of social inequality between the wealthy and the poor, as well as between older men and younger women that is perpetuated in government secondary schools (Bajaj, 2005). Student accounts also cited how cross-generational relationships begin both inside schools and beyond the school gates. Outside of school, interactions between young women and older men in public places (or ‘‘on the way’’ to them) facilitated, in many cases, the establishment of a cross-generational relationship. Mary, a twelfth grader at Chilemba High School, discussed the signals indicating a sexual relationship is beginning: You can be sent somewhere and you meet him on the way, ‘‘I’ll give you a lift, jump in the car.’’ He gives you money. You, without knowing, [think] it’s just a lift. He asks, ‘‘Where do you stay?’’ You give the correct answer, because you don’t know. He drops you where you are going. He gives you money for a drink. You go. The next day, or he even gives you two days, he’ll come back. ‘‘Hi, hi there, but don’t tell your parents about me.’’ This one is giving you money; again he gives you money. At the end, he will

134

MONISHA BAJAJ

demand for something and that’s sex. And you don’t resist because you think, ‘‘this one is giving me money and if I do this, if I sleep with him, he will start giving me more money than he does now.’’ So you fall into that trap, and then you get addicted to money. When you get addicted to money, it’s very difﬁcult to stay without money. So if that person is gone, you go for another old man because you know that old people are the ones who give money to young people. [The man could be] a manager, a person who works for a big company. It’s common wherever you go to ﬁnd a young girl and an old person. Sometimes you can even say the man [is] the father or the uncle, but that person is not. He even comes to pick [her] up at school. You think maybe that’s the father, but someone says, ‘‘No, that’s the boyfriend’’; the so-called ‘‘man-friend.’’

Exposed to cross-generational relationships in their schools and communities, young women learn the social mores surrounding them. Operating in educational and social contexts characterized by economic inequality, students learn the beneﬁts of responding positively when approached by older men offering them money. They may become ‘‘addicted to money,’’ as Mary noted, because the funds allow them to pay for school expenses and other things they need or desire, such as clothes, shoes, mobile phones, and food. Weissman et al.’s continuum of volition, presented in Fig. 1 earlier in this chapter, highlighted the complex interplay of psychological and economic factors that structure the decisions of young women engaging in such relationships, yet the material conditions of young women’s lives play a central role in their decision-making. Young girls living in the townships of Ndola reported being approached by older men and sometimes accepting their requests for sex because of the material beneﬁts that such relationships provided. For example, 12-year-old Victoria mentioned that older men had already started offering her food or money in exchange for sex. Victoria noted that female orphans may accept the sexual advances of or seek out older men given their often-precarious economic situations. Given the estimated 800,000 orphans in Zambia and their escalating numbers due to the HIV/AIDS pandemic (ZMOE, 2003), economic hardship becomes a particularly salient motivation for young women to either seek out or respond favorably to ‘‘sugar daddies.’’ Although many students noted that men propositioned schoolgirls, other students also cited the initiative of girls in seeking out such relationships. The manner in which girls sought out older men sometimes hindered their participation in school, leading to a complex situation in which crossgenerational relationships may, for different participants and in different moments, limit and/or enable educational pursuits. For example, Catherine, a tenth grade student at Umutende government school, noted, ‘‘Sometimes instead of being in class, [girls] go out and start looking for sugar daddies who can give them money to buy whatever they want to buy. I know some

Sugar Daddies and Girls’ Schooling in Zambia

135

of them. They go during school time. They go to different buildings like bars and guesthouses [where] the men are.’’ This example suggests that young girls may exert their sexual agency and decide to work the system to their immediate advantage by responding to the advances of older men or actively seeking out relationships with them to pay for their education and other things. Yet the fact that young girls encounter this situation early in adolescence, coupled with strong social taboos around parents talking with their children about sex, often results in girls experimenting with relationships where they may lack the power to negotiate safe sex. Economic hardship and being orphaned, in particular, may also drive girls to accede to the propositions of an older and more afﬂuent partner. Lucy, who is a social worker, highlighted this point, ‘‘These sugar daddies come maybe with a 50,000 kwacha note [US$10]. This man will do anything to this girl because she wants some money to sustain herself; maybe to pay for school. So the sugar daddies have taken advantage of the poverty situation.’’ Noting the contextual forces that drive such relationships and their presence as a common feature of students’ social world, it is important to explore the attitudes held by students, teachers, and parents toward crossgenerational relationships and how these attitudes are linked to discussions of HIV/AIDS.

BLAMING THE VICTIM: SOCIAL ATTITUDES TOWARD CROSS-GENERATIONAL RELATIONSHIPS Through experiences with commonplace corruption in government secondary schools, such as having to pay off a headmaster to secure a spot or being coerced to bribe teachers to receive instruction, young women are socialized to seek out ﬁnancial resources to try and level an unequal playing ﬁeld (Bajaj, 2005). For those who lack money and engage instead in crossgenerational relationships, they are the ones frequently blamed by peers and elders for violating ‘‘tradition.’’ One student commented that women are not following their cultural tradition by ‘‘wearing trousers’’ and that ‘‘girls of nowadays no longer cook like their mothers, but drink like their fathers.’’ Other respondents noted that these attitudes suggest a ﬁxed notion of culture and tradition that does not reﬂect how current economic and social realities are affecting how individuals and families are shaping decisions about their survival. A seventh grader noted in her diary the parental involvement in decisions about cross-generational relationships writing that

136

MONISHA BAJAJ

‘‘Some parents are telling their children to go out and sell their bodies in order to get some money.’’ Schools assimilate youth into the norms of an unequal society and students learn indirectly to leverage their ﬁnancial and sexual resources to advance within it. However, this does not mean that there are no social consequences for youth who engage in cross-generational relationships. Parents, teachers, students, and community members acknowledged the occurrence of crossgenerational relationships particularly in regard to the economic hardship that often fuels them; however, at the same time, they often expressed extremely negative attitudes toward those who engage in such relationships. Families frequently shun young women who are in cross-generational relationships, often referring to them as prostitutes, especially if they are found to have contracted HIV/AIDS as a result. One seventh grade student, a 12-year-old named Annie, reported: My own aunty, my mom’s younger sister has got a child. I think the dad of the child rejected her. [My aunty] doesn’t want any job. All she wants is that men come, pick her up, and pay her to go places like very big hotels, guesthouses, cinema halls, bars, and clubs. We don’t stay with her because my mom told her that she couldn’t allow any of the things she was doing in our house. So she had to move away and ﬁnd her own house. Maybe a man comes, and if [he] just has money, she’ll accept. One of the things about prostitution is that the girls are lacking education.

In Annie’s account, cross-generational relationships and prostitution are synonymous. In recent years as well, the stigma associated with commercial or transactional sex has also become associated with the risk and contraction of HIV/AIDS. On discovering that a relative is HIV positive, families may be embarrassed by the stigma that the disease holds and the implications for the family’s social standing. One eleventh grade student wrote in her diary about her best friend ﬁnding out that he and his girlfriend were HIV positive: ‘‘They decided to tell their families about their HIV status [and] they were abandoned by their loved ones.’’ She continued writing that her friend wanted to commit suicide because of his HIV status, highlighting the social stigma around it. The stigma that young women and men face, whether infected with HIV or not, inﬂuences their position in society and opportunity for mobility within it. Decisions to participate in cross-generational relationships therefore often carry the burden of heavy stigma. It is interesting to note that no respondent in this study admitted to being involved in a cross-generational relationship and most condemned them; however, several young women

Sugar Daddies and Girls’ Schooling in Zambia

137

respondents were rumored to be involved in transactional sexual relationships with older men.

DISCUSSION The preceding sections offered frameworks (Hunter, 2002; Weissman et al., 2006) on cross-generational relationships broadly to understand young people’s experiences in and around their secondary schools. Both Hunter (2002) and Weissman et al. (2006) identify the spectrum of motivations for such relationships that exists between survival and subsistence on the one hand, and consumption and material comforts on the other hand.9 The data presented in this chapter provided greater texture to the picture of the space between survival and consumption where desire and demand for education lie. In order to understand how the ‘‘sugar daddy phenomenon’’ inﬂuences young people’s educational pursuits, respondents’ perspectives were presented on (1) how and why young women and older men enter into such relationships amidst the backdrop of economic uncertainty and HIV/AIDS; (2) how cross-generational relationships occur and are viewed within schools where teachers are the older men and their students the younger women; and (3) young people’s (especially young men’s) and families’ attitudes toward such relationships, and the social consequences for young women who engage in them. In examining cross-generational relationships in and around schools, it becomes increasingly clear that the incidence of ‘‘sugar daddies’’ (and mommies) in Ndola (Zambia) must be understood as a complex phenomenon involving discourses of masculinity and the material inequities that structure the physical and discursive space occupied by young adults and the schools they attend. Although young girls (and boys, though differentially) undoubtedly exercise agency in numerous ways, these choices with regard to cross-generational relationships occur in the context of extreme inequalities and severe social stigma that limit the value of such choice. One of the myriad reasons young women enter into cross-generational relationships is to secure their school fees since parents are increasingly weighing the potential future beneﬁts of schooling against their present economic circumstances. Functionalist assumptions about the causal relationship between education and employment are being questioned throughout the global South as the ranks of the educated unemployed swell (Jeffrey et al., 2007). Without extensive personal connections, secondary school graduates in Zambia

138

MONISHA BAJAJ

(as has been noted in other countries) have limited employment prospects (Bajaj, 2005). Young women who engage in cross-generational relationships as a means of securing a better future through continued education put themselves at risk for HIV and other sexually transmitted infections and may still not be able to access the anticipated economic outcomes that initially drove their decision-making. Further research can help to elucidate the intersections among gender inequalities, education, and employment, particularly in contexts such as Zambia where the combined impact of economic decline and HIV/AIDS has adversely affected all aspects of social, political, and economic life.

Policy Implications In carrying out research on cross-generational relationships and schooling in Zambia, the question arises: What are the implications for policymakers? Steady (2005) posits in her framework for gender research in Africa that the practical aspect of research is of utmost importance for the continent; Western scholarship places a high value on theorizing and theory-building at the expense of pragmatism and relevance. For Africa, there are many economic and social problems, not least of which is its dependency on the West and its marginalization through globalization, which theory cannot solve. (p. 322)

Although theory is not rejected in an analysis of the phenomenon of sugar daddies in Ndola (Zambia), the following recommendations for practice also emerge from this chapter. First, the considerable funding being devoted to HIV/AIDS prevention programs needs to take into account social and gender inequities that fuel risky sexual behavior. Without an analysis of power asymmetries, interventions may miss the mark in reaching vulnerable populations. Given that young women have more than three times the HIV infection rates in sub-Saharan Africa as compared to their male counterparts (Fleshman, 2004), more effective policies and programs linking public health education to material well-being are sorely needed to curb the impact of this devastating disease. Although the particular nature and speciﬁc architecture of a policy platform to address economic decline and the HIV/AIDS pandemic is a highly politicized and complex venture beyond the scope of this chapter, policymakers might do well to consider measures that involve or result in job creation or a raising of base levels of income as well as the political and institutional obstacles to achieve these goals, as will be touched

Sugar Daddies and Girls’ Schooling in Zambia

139

upon later in the discussion of the resistance to social spending. Policymakers would also do well to further consider and address cultural understandings of masculinity and femininity in the development of programs related to HIV/AIDS. Young women must be included in designing and implementing such programs so that their realities are adequately considered and addressed. It may very well be that young women do not see these cross-generational relationships as problematic since they can provide income or advantages, despite negative social attitudes toward them. However, programs can focus attention in an informed manner on the power differentials of such relationships and the risk of contracting HIV for these young women. Second, the high incidence of teacher–student sexual relationships – consensual, coerced, and non-consensual – must be addressed by the Ministry of Education through more stringent sanctions for teachers engaging in such behavior. Although teachers are supposed to be dismissed for engaging in relationships with students (ZMOE, 2003), many respondents noted teacher impunity and corruption in the process of teacher discipline. Better mechanisms and more efﬁcient administration would limit the ability of teachers to initiate relationships with young women in schools by offering educational advantages. A signiﬁcant portion of cross-generational relationships, as noted by Bledsoe and Gage (1994), consists of those between teachers and their students. Better educational policy and its enforcement may serve to dissuade teachers from abusing their power and status within schools for their personal beneﬁt. Finally, it would be remiss not to mention a reconsideration of the role and use of social spending as a policy tool. Although not a comprehensive policy solution in itself, secondary school fees are an example of an additional hindrance to young women’s educational access as discussed in this chapter. Since Zambia received signiﬁcant debt relief in 2005 from international ﬁnancial institutions, policymakers have an opportunity to rethink the austerity measures vis-a`-vis social spending previously implemented as part of SAPs; such spending must consider and work to alleviate the economic hardship that drives many cross-generational relationships. In 2006, then President Mwanawasa announced that healthcare would be free as a result of monies freed up from the debt relief granted to Zambia by the G8 (BBC, 2006). In education, one productive move would be to abolish all fees for secondary education in Zambia, a move that government ofﬁcials have stated their reluctance to implement, except in the case of orphans (Times of Zambia, 2005). Recent debt relief or redirected donor aid may provide the resources necessary for making secondary education less of a

140

MONISHA BAJAJ

ﬁnancial burden for families in a time of economic decline. The demand for schooling is but one factor that motivates young women to seek out ‘‘sugar daddies.’’ Still, by emphasizing the role or place of education – albeit one with better regulations for disciplining erring teachers – alongside measures to increase job creation and base levels of income, the Zambian government may be able to rectify some of the inequalities that contribute to the prevalence of cross-generational relationships in the country today. The intersections among cross-generational relationships, schooling, and HIV/AIDS are complex and have various manifestations throughout Zambian society. Although young women are ﬁnding ways to cope with being enmeshed in a context characterized by severe economic decline and an extensive HIV/AIDS crisis, the strategy of securing a ‘‘sugar daddy’’ is one that may result in deadly infection and social isolation. By examining cross-generational relationships, HIV/AIDS, and education in Zambia, this chapter offered suggestions and preliminary recommendations for the role of educational policy in addressing economic and health crises. Further attention to the ways that young women respond to these crises can help explore possibilities for greater gender equity and educational advancement in highly impoverished and unequal contexts such as Zambia.

NOTES 1. For the purposes of this chapter, I rely upon the explanation of Kuate-Defo (2004) that offers multiple reasons for and dimensions of cross-generational relationships ‘‘Non-consensual and consensual sexual relationships between young people and older individuals are inﬂuenced by factors operating at several levels including individual, family, community, neighbourhood, province, region within a country, and by the international context of globalisation. Individuals engage in such sexual behaviours for a variety of motives including procreation, love and affection, pleasure, entertainment, conformity, recognition, competition, power, domination, submission, self-determination, stress reduction, ﬁnancial security, favours, money and presents. These motives are usually age-dependent and are greatly inﬂuenced by individual attributes, conditions, life options and opportunities’’ (p. 21). In this chapter, the link between schooling and cross-generational relationships was frequently discussed with regards to the monetary beneﬁts of such relationships. As a result, while it would be reductionist to assume that economic motivations are the sole cause of cross-generational sexual relationships, the material dimensions that do exist in many of these relationships are discussed with regard to educational experiences and opportunities. 2. The term ‘‘sugar daddy’’ refers to the older man in a cross-generational relationship where there is at least a 10-year age differential with the younger woman and where there is signiﬁcant difference in access to economic resources. The term

Sugar Daddies and Girls’ Schooling in Zambia

141

‘‘sugar mommy’’ refers to the woman in the inverse relationship between an older woman and a younger man where there is at least a 10-year differential, though empirically these relationships seem to appear with much less frequency. 3. For a more detailed discussion of the relationship between gifts and sex, see Hunter (2002), Kaufman and Stavrou (2002), and Standing (1992). Although attentive to the historical presence of inter-generational relationships, this chapter primarily deals with the economic nature of such relationships and how youth and adults perceive their inﬂuence on schooling in present day Ndola (Zambia). 4. One of the few studies that attempted to determine the scope of crossgenerational relationships found the occurrence to be relatively low. In Kisumu (Kenya), a study deﬁned a sugar daddy as a man 15 years older than his partner. The researchers studied 1,000 men aged 21–45 and found that only ﬁve percent of their sample ﬁt the deﬁnition. They did ﬁnd, however, that 84 percent of the men in the study had given money to a non-marital sexual partner in the past month and that 60 percent of the adult men had at least one adolescent partner (Luke, 2005). These ﬁndings suggest the impartiality of our knowledge with regard to the prevalence of cross-generational relationships. This chapter addresses youth and adults’ perception of the frequency of such relationships and how these are believed to be affecting educational opportunity. Hence, actual ﬁgures for the incidence of crossgenerational relationships are of less importance for discussion here. 5. Given the stigma around discussing sexual relationships, most respondents offered information about friends, students, or neighbors rather than about their own actual behavior. Information gathered through interviews suggested that even some of those respondents who condemned such relationships, reportedly engaged in them, indicating the complexity of studying such a topic. 6. For a more extensive treatment of gender violence in schools in the region, see Leach, Humphreys, and Dunne (2006). 7. All names used in this chapter are pseudonyms for the purposes of conﬁdentiality. 8. Although not discussed extensively here, the belief that sexual intercourse with a virgin cures HIV has driven an increasing number of rapes of young children in Zambia, as noted in regular news articles about the disturbing phenomenon. Research in South Africa has explored this myth in further detail (Leclerc-Madlala, 2002; Pitcher & Bowley, 2002). 9. These models acknowledge, but do not delve into details about, relationships that are at either of extreme of being purely consensual or forced/coerced with violence.

ACKNOWLEDGMENTS In addition to all the respondents who gave their time for this research, I would like to thank Belinda Chiu, Bikku Kuruvila, Frances Vavrus, David Baker, Alex Wiseman, and the anonymous reviewers for their contributions to this chapter. This research was carried out with support from the African Youth and Globalization Fellowship program of the Social Science Research Council and the American Council of Learned Societies Africa

142

MONISHA BAJAJ

Regional Advisory Panel in partnership with the Council for the Development of Social Science Research in Africa, and South Africa’s National Research Foundation.

REFERENCES Bajaj, M. I. (2005). Conceptualizing agency amidst crisis: A case study of youth responses to human values education in Zambia. Doctoral dissertation, Columbia University. BBC. (2006, April 6). Zambia overwhelmed by free health care. British Broadcasting Company. Available at http://news.bbc.co.uk/1/hi/world/africa/4883062.stm. Retrieved on 28 August 2007. Bennell, P., Hyde, K., & Swainson, N. (2002). The impact of the HIV/AIDS epidemic on the education sector in sub-Saharan Africa: A synthesis of the ﬁndings and recommendations of three country studies. Sussex, UK: Center for International Education, University of Sussex Institute of Education. Bledsoe, C., & Gage, A. (1994). The effects of education and social stratiﬁcation on marriage and the transition to parenthood in Freetown, Sierra Leone. In: C. Bledsoe & G. Pison (Eds), Nuptiality in sub-Saharan Africa (pp. 148–164). Oxford: Clarendon Press. Bond, G. (1982). Education and social stratiﬁcation in northern Zambia: The case of the Uyombe. Anthropology and Education Quarterly, 13(3), 251–267. Burdette, M. (1988). Zambia: Between two worlds. Boulder, CO: Westview Press. de Waal, A., & Whiteside, A. (2003). ‘New variant famine’: AIDS and food crisis in southern Africa. The Lancet, 362, 1234–1237. Ferguson, J. (1999). Expectations of modernity: Myths and meanings of urban life on the Zambian Copperbelt. Berkeley, Los Angeles: University of California Press. Fleshman, M. (2004). Women: The face of AIDS in Africa. Africa Renewal, 18(3), 6–8. Garbus, L. (2003). HIV/AIDS in Zambia. San Francisco, CA: University of California San Francisco. Gausset, Q. (2001). AIDS and cultural practices in Africa: The case of the Tonga (Zambia). Social Science and Medicine, 52, 509–518. Glynn, J. R., Carael, M., Auvert, B., Kahindo, M., Chege, J., Musonda, R., Kaona, F., Buve, A., & the Study Group on the Heterogeneity of HIV Epidemics in African Cities. (2001). Why do young women have a much higher prevalence of HIV than young men? A study in Kisumu, Kenya and Ndola, Zambia. AIDS, 15(S4), S51–S60. Hunter, M. (2002). The materiality of everyday sex: Thinking beyond ‘prostitution’. African Studies, 61(1), 99–120. Jeffrey, C., Jeffrey, P., & Jeffrey, R. (2007). Degrees without freedom? Education, masculinities, and unemployment in North India. Stanford, CA: Stanford University Press. Kaufman, C., & Stavrou, S. E. (2002). ‘‘Bus fare, please’’: The economics of sex and gifts among adolescents in urban South Africa. New York, NY: Population Council. Kelly, M. J. (1999). HIV/AIDS and schooling in Zambia. In: M. J. Kelly (Ed.), The origins and development of education in Zambia: From pre-colonial times to 1996. A book of notes and readings (pp. 342–343). Lusaka, Zambia: Image Publishers Limited. Kuate-Defo, B. (2004). Young people’s relationships with sugar daddies and sugar mummies: What do we know and what do we need to know? African Journal of Reproductive Health, 8(2), 13–37.

Sugar Daddies and Girls’ Schooling in Zambia

143

Leach, F., Humphreys, S., & Dunne, M. (2006). Gender violence in schools in the developing world. Gender and Education, 18(1), 75–98. Leclerc-Madlala, S. (2002). On the virgin cleansing myth: Gendered bodies, AIDS and ethnomedicine. African Journal of AIDS Research, 1(2), 87–95. Longﬁeld, K., Glick, A., Waithaka, M., & Berman, J. (2002). Cross-generational relationships in Kenya: Couples’ motivations, risk perception for STIs/HIV and condom use. Washington, DC: Population Services International. Luke, N. (2005). Confronting the ‘sugar daddy’ stereotype: Age and economic asymmetries and risky sexual behavior in urban Kenya. International Family Planning Perspectives, 31(1), 7–14. Luke, N., & Kurz, K. (2002). Cross-generational and transactional sexual relations in subSaharan Africa: Prevalence of behavior and implications for negotiating safer sexual practices. Washington, DC: International Center for Research on Women. Mehmet, O. (1997). Westernizing the Third World. London: Routledge. Pitcher, G., & Bowley, D. (2002). Infant rape in South Africa. The Lancet, 359(9303), 274–275. Plummer, M. L., Wight, D., Ross, D. A., Balira, R., Anemona, A., Todd, J., Salamba, Z., Obasi, A. I., Grosskurth, H., Changalunga, J., & Hayes, R. J. (2004). Asking semi-literate adolescents about sexual behaviour: The validity of assisted self-completion questionnaire (ASCQ) data in rural Tanzania. Tropical Medicine & International Health, 9(6), 737–754. Reddy, S., & Dunne, M. (2007). Risking it: Young heterosexual femininities in South African context of HIV/AIDS. Sexualities, 10(2), 159–172. Saasa, O. (2002). Poverty proﬁle in sub-Saharan Africa: The challenge of addressing an elusive problem. In: G. Bond & N. Gibson (Eds), Contested terrains and constructed categories: Contemporary Africa in focus (pp. 105–116). Boulder, CO: Westview Press. Simpson, A. (2005). Sons and father/boys to men in the time of AIDS: Learning masculinity in Zambia. Journal of Southern African Studies, 31(3), 569–586. Standing, H. (1992). AIDS: Conceptual and methodological issues in researching sexual behaviour in sub-Saharan Africa. Social Science and Medicine, 34(5), 475–483. Steady, F. C. (2005). An investigative framework for gender research in Africa in the new millennium. In: O. Oyewumi (Ed.), African gender studies: A reader (pp. 313–331). New York, NY: Palgrave MacMillan. Times of Zambia. (2005). State won’t abolish secondary school fees. Times of Zambia. Available at http://www.times.co.zm/news/viewnews.cgi?category ¼ 4&id ¼ 1082601682. Retrieved on 7 February 2006. UNAIDS. (2006). Report on the global AIDS epidemic 2006. New York, NY: United Nations. UNDP. (2004). Human development report 2004: Cultural liberty in today’s diverse world. New York, NY: United Nations Development Program. Varga, C. A. (2001). The forgotten ﬁfty per cent: A review of sexual and reproductive health research and programs focused on boys and young men in sub-Saharan Africa. African Journal of Reproductive Health, 5(3), 175–195. Vavrus, F. (2003). Desire and decline: Schooling amid crisis in Tanzania. New York, NY: Peter Lang. Weissman, A., Cocker, J., Sherburne, L., Powers, M. B., Lovich, R., & Mukaka, M. (2006). Cross-generational relationships: Using a ‘Continuum of Volition’ in HIV prevention work among young people. Gender & Development, 14(1), 81–94. ZMOE. (2003). HIV/AIDS guidelines for educators. Lusaka, Zambia: Zambian Ministry of Education.

LIMITS OF AND POSSIBILITIES FOR EQUALITY: AN ANALYSIS OF DISCOURSE AND PRACTICES OF GENDERED RELATIONS, ETHNIC TRADITIONS, AND POVERTY AMONG NON-MAJORITY ETHNIC GIRLS IN VIETNAM Joan DeJaeghere and Shirley J. Miske ABSTRACT This chapter examines discourses and social practices at individual, community, and institutional levels related to non-majority Vietnamese ethnic girls’ access to and participation in secondary school. This critical analysis utilizes Sen’s framework of capabilities to illustrate differences in discourse and social practice that exist around poverty, and the ways in which gendered relations and ethnic traditions are intertwined with the discourse and practices of poverty to affect girls’ choices and well-being in and through secondary education. We particularly draw on girls’ and their parents’ constructions of these issues as they negotiate and are affected by them. We argue that strategies must move beyond the discourse that Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 145–183 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010008

145

146

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

ethnic traditions and gendered relations are barriers to girls’ education to consider the inequalities and lack of capabilities that perpetuate poverty and unequal gendered relations for non-majority ethnic groups in societies.

INTRODUCTION Recent studies examining gender parity that disaggregate the data by other variables show that out-of-school girls are disproportionately members of excluded groups (Lewis & Lockheed, 2006), often including non-majority ethnic groups in many countries. Out-of-school girls tend to live in rural areas and come from low-income families (Filmer, 2000; Lewis & Lockheed, 2006; Wils, Carrol, & Barrow, 2005). Even in areas of the world usually characterized by gender parity at various levels of schooling, gender disparities become obvious when analyses are made that include ethnicity and high levels of poverty (Stromquist, 2001). In Vietnam, while primary and secondary enrollment rates approximate parity, these rates mask the disparities existing within certain ethnic groups. These ﬁndings suggest that indicators of gender parity need to be further problematized and examined before gender equality can be fully realized. Gender equality needs to be considered in combination with other issues, such as poverty and ethnic discrimination that affect overall equality in society. Stromquist (2006) argues that achieving gender equality requires equal opportunity and participation in the public sphere, as well as attention to gender differences within the private sphere, which go beyond the impacts of parity in education. Although combinations of these three characteristics – gender, non-majority ethnicity, and poverty – have been shown to affect girls’ access, participation, and achievement in schools, few studies have qualitatively examined how the nexus of these characteristics affects girls’ schooling opportunities at the secondary level, and, in turn, equality of capabilities (Sen, 1999). In the past decade, secondary education has received increased attention among governments, donor agencies, and researchers (e.g., World Bank, 2005). There is growing concern about the causes related to low transition rates from primary to secondary education, especially among girls. Opportunities for further schooling, future employment, and a family’s household income are all factors affecting participation in secondary school. No country has ever achieved universal primary enrollment without a secondary enrollment of 45 percent (Clemens, 2004), suggesting that the

Limits of and Possibilities for Equality

147

opportunity to go on to secondary school affects retention through primary school. Recent studies (e.g., Tikal, 2007) have found a signiﬁcant and previously overlooked relationship between secondary education, economic development and poverty reduction, where both economic development and poverty reduction strategies are important for young people’s participation in secondary school. At the secondary level in particular, poverty is a major decisive factor in educational participation, often disadvantaging girls (UNICEF, 2008). In his analysis of education across sub-Saharan Africa, Lewin (2005) argues that wealth is generally the most important determinant of enrollment in secondary school. ‘‘Participation at secondary level,’’ he notes, ‘‘is widely rationed by price’’ (p. 412). Among the poorest 40% of the population, the ratio of boys to girls is nearly four to one for Grade 9 participation, whereas gender is least important in explaining differences in enrollment in the top 20% (p. 19). This chapter is a critical analysis of discourse and social practices related to gender, ethnic traditions, and poverty among non-majority girls and their families in Vietnam. We attempt to illustrate in this analysis how majority and non-majority groups within Vietnamese society construct discourses and value social practices differently, and how these different constructions and values of ethnic traditions, gender, and poverty may, at times, have contradictory impacts on girls’ choices and capabilities in and through education. We also aim to illustrate how non-majority ethnic girls and their families negotiate these multiple factors affecting their capabilities. Throughout this analysis, we argue that approaches that address gender, poverty, or cultural traditions as instrumental barriers may not effectively respond to the interwoven dimensions and complex realities these girls face (Sutton, 2001). Rather we suggest approaches are needed to create capabilities for well-being, especially educational opportunities that relate to the multidimensionality of girls’ lives (Sen, 1999).

EDUCATION AND NON-MAJORITY ETHNIC GIRLS IN VIETNAM: THE PRESENT SITUATION Vietnam, the second most populous country in Southeast Asia, has 54 different ethnic groups. The Kinh ethnic group comprises the majority of the population, 53 non-Kinh ethnic groups make up 14% of the population (1999 Census; Kosonen, 2004, p. 6). Many of these non-Kinh ethnic groups live in remote and mountainous areas of Vietnam (Baulch, Chuyen,

148

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

Haughton, & Haughton, 2004). These regions are geographically disadvantaged in term of access to and opportunities for proﬁtable agriculture, resulting in subsistence farming as a primary means of living. A low level of education, malnutrition, and poverty are prevalent for many of the nonmajority ethnic groups (Kosonen, 2004; UNDP, 2002). Educational attainment is high overall in Vietnam. National education indicators suggest a high primary enrollment, with little disparity between educational attainment for boys and girls. However, these national statistics mask variations in enrollment and completion for the different ethnic groups, and for boys and girls within ethnic groups, in different regions. Primary school gender gaps appear in the poorest households (5.3%), in the Central Highlands (9.8%), and among the Chinese (6.4%) and other ethnic groups (4.6%) (based on data from 1999 for which ethnic group data are available). The primary enrollment rate is below 70% for the Ba-Na, Gia-Rai, and Xo-Dang ethnic groups in the Central Highlands and the Hmong in the Northern Mountainous areas (Baulch et al., 2004). Secondary enrollment rates are considerably lower than primary enrollment rates overall (see Table 1), and are less than 20% for most non-majority ethnic groups (e.g., the Gia-Rai, Ba-Na, Xo-Dang, Hmong, and Dao) (UNICEF, 2004, 2005). Data disaggregated by non-majority ethnic groups show greater gaps between minority and majority students, as well as between minority girls and boys (Table 2). Nearly 65% of Kinh majority are enrolled in secondary education (lower and upper), whereas only 4.5% of Hmong, 9% of Ba-Na, and less than 15% of Xo-Dang and Dao children are enrolled (see Table 2, Baulch, et al., 2004). The Kinh ethnic majority has an enrollment rate of 32% in upper secondary schools, but a 2002 study (Hirosato et al., 2002) showed that only 8% of all other ethnic groups were enrolled at that level. The gender gap for non-majority ethnic groups overall is 13.4%, with even larger gaps in the ethnic groups stated earlier. In rural areas, the gender gap averages 7.4%, and among the poorest households it is 12.4% (Desai, 2001). Table 1.

Primary and Secondary Enrollment Ratios (%). Primary

Gross enrollment rates Net enrollment rates

Secondary

Overall

Boys

Girls

Overall

Boys

Girls

93 87

96

90

75 69

76 70

74 68

Note: UNESCO, UIS 2005 Data.

149

Limits of and Possibilities for Equality

Table 2.

Primary and Secondary Enrollment Ratios (%) by 12 Different Ethnicities.

Ethnic Groups

Primary Level

Secondary Level

Gross

Net

113.6 122.6 114.5

93.4 93.7 76.3

93.5 94.5 77.3

93.4 92.9 75.3

80.6 71.0 35.9

64.8 51.7 22.5

65.5 50.4 23.8

64.0 53.1 21.2

Central highlands Gia-Rai 126.3 Ba-Na 108.9 Xo-Dang 139.3

66.4 57.8 62.2

67.6 55.0 64.7

65.1 60.4 59.3

37.1 20.0 35.2

14.9 8.9 10.1

15.2 9.0 12.7

14.5 8.9 7.1

Northern uplands Tay 135.4 Thai 135.5 Muong 133.4 Nung 136.6 Hmong 80.5 Dao 126.4

94.7 83.9 94.5 89.3 41.5 71.4

94.9 87.2 94.9 89.7 51.5 73.7

94.4 80.5 94.0 88.9 31.5 68.8

77.0 55.2 76.7 61.8 9.8 20.3

51.0 32.1 52.3 39.2 4.5 11.8

47.1 33.6 50.8 37.0 7.5 11.9

55.2 30.5 53.9 41.6 1.6 11.8

Kinh Hoa Khmer

Net (boys) Net (girls) Gross

Net

Net (boys) Net (girls)

Source: Data from Baulch, Chuyen, Haughton, and Haughton (2004). Based on 3% enumeration sample of 1999 Census.

POVERTY, ETHNIC GROUPS, AND WOMEN IN VIETNAM Vietnam has taken concerted effort to address real poverty among its population. In addition to sustained economic growth, Vietnam’s national policies have targeted extremely impoverished communities and individuals to address food poverty, health care and access to infrastructure, sanitation, electricity, and schools. The poverty rate1 nationally has declined from 58% in 1993 to 19.5% in 2004 (Vietnamese Academy of Social Sciences, 2006, p. 12). Although these rates indicate considerable progress, the reduction of poverty varies greatly by region and among ethnic groups. The Kinh and Chinese (Hoa) ethnic groups experienced the greatest reduction in poverty during this time period, from 54% to 14%, whereas the remaining 53 ethnic groups’ poverty rate declined from 86% to 61% (Vietnamese Academy of Social Sciences, 2006, p. 25). The rate of reduction is much slower, which results in an increase in the absolute difference in poverty rate between the Kinh-Hoa and other ethnic groups.

150

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

Women and girls, and particularly female-headed households in rural areas, are vulnerable to poverty, in part because of limited access to land or the labor force (Centre for International Economics, 2002). The gender wage gap is also greater in agriculture, where many ethnic groups earn their living. Women earn 62% of men, and women earn only 40% of total wages (p. 19). There are many causes of poverty among women and particularly those of non-majority ethnic groups (Centre for International Economics, 2002; Vietnamese Academy of Social Sciences, 2006), including less access to land, water, markets, and health care, and lower educational levels and participation. One of the causes of poverty for non-majority ethnic groups in contrast to the Kinh majority is a low return on education; however, they compensate by generating a higher return on land and labor (Van de Walle & Gunewardena, 2000). This means that the actual value of education is low, and, in turn, labor becomes more valuable, often leading young girls and boys to leave school. This also suggests that other barriers exist within the society, such as job discrimination, that prohibit nonmajority ethnic groups from utilizing their education to a full capacity for their well-being.

ANALYTICAL FRAMEWORK In analyzing discourse and social practices that affect girls’ lives in their pursuit of secondary education, this chapter draws on the theoretical concepts of social construction and agency. We are interested in understanding how girls and their families, in comparison to and contrast with community leaders, educators, and policy makers, construct these interwoven dimensions of their lives, and the agency they exert, or not, in developing their capabilities. We deﬁne discourse, drawing on Foucault’s (1972) work, as systems of thoughts composed of ideas, attitudes, principles, rules, and inner logics that construct individuals and their worlds. Discourse, in this analysis, goes beyond the systems of thought that guide institutions and groups in power, including governments; we also examine discourse as the systems of thought that guide individuals, particularly from speciﬁc ethnic groups. Social practices are the courses of action and practices, including government and organizational policies and community traditions, which are created in relation to these systems of thought. Among government, international organizations and communities, and within communities and ethnic groups, power exists more so for some groups and individuals than others in creating a discourse of ‘‘truth’’ and social

Limits of and Possibilities for Equality

151

practices. For example, within some ethnic communities, the local leader has greater power than children. We aim to illustrate in this analysis where institutional discourse may be more powerful than community discourse, while also showing where communities, and especially girls, use agency in the discourse and social practices. In essence, we illustrate how macro-level discourses affect local sites, and how local sites construct and react to these discourses. In examining poverty, ethnic traditions and gender as socially constructed discourse and social practices, we also acknowledge, drawing from Sen, that variations in material conditions and social opportunities exist and shape these constructions. We utilize Sen’s (1992, 1999) capability approach to understand how girls and their families choose and value different capabilities, and, thus, how they exert agency toward their well-being. We draw on the scholarship of gender relations that suggests gender is socially constructed in combination with ethnicity and social class (Hill Collins, 1990; Subrahmanian, 2002), and we draw on Vavrus’ critical analysis of ethnic traditions and gender as socially constructed by competing discourses. This critique analyzes the dominant discourses among international organizations and governments that suggest cultural traditions are impediments to ethnic groups’ and girls’ participation in education. By analyzing the interconnections among poverty, ethnic traditions, and gender relations, we illustrate that the discourses and social practices at the individual, community and societal level are often conﬂicting, and may be used to perpetuate, challenge or interrupt the lack of choices and capabilities girls have. At times, the discourse and social practices (e.g., girls are harder workers, and therefore they should work more) act to inhibit the capabilities of girls; other discourse and social practices illustrate how girls and their families make sense of, value, and exert agency over the lack of physical and social capabilities, their ethnic identity, and their gendered relations. Sen’s (1992, 1999) framework of poverty as capability deprivation, and Walker and Unterhalter’s (2007) application of Sen’s capability framework to education is useful to understand how individuals and communities have different possibilities to choose well-being in and through education. Poverty as capability deprivation2 is broader than an income or an instrumental approach to poverty, which measures poverty by access to goods, resources, or real income. Capability deprivation is a lack of freedom to pursue well-being. Sen (1992) argues in his explanation of inequality in societies that equality of ‘‘what’’ matters. Equality is more than the access to primary goods and material conditions to function in life; it is a set of capabilities with which one can choose to pursue well-being, and to convert

152

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

material conditions into well-being. Sen (1992, 1999) suggests that the relationship between low income (instrumental poverty) and low capability is variable between different communities, families, and individuals, depending on other conditions and factors, such as age, gender, social roles, and locations. For example, capabilities vary based on gender when income is distributed unequally to boys in a family (Sen, 1999, p. 88, 89). The capabilities and the conditions affecting non-majority ethnic girls’ wellbeing related to schooling are constructed and valued differently depending on social, economic, cultural, and political factors. Sen’s capability framework is not without criticism; in particular, it does not address speciﬁc capabilities for gender equality or racial or ethnic equality. Nevertheless, it provides a useful framework through which to broaden the understanding of poverty as it relates to gender and ethnic inequalities.3 Gender relations and ethnic traditions, as Sen suggests earlier, are interwoven in the capability to choose one’s well-being. Gender relations and roles are socially constructed in society by the discourses and practices of the state, the economy, and other macro- (e.g., global economic trade) and micro-processes (e.g., individual and community preservation of cultural identity) (Stromquist, 2006; Subrahmanian, 2002). As ﬂuid and socially constructed, gender relations are intertwined with economic status, or poverty, and ethnicity, or ethnic traditions. An analysis of the relationship among gender, ethnicity, and social status can help illustrate multiple oppressions affecting non-majority women and girls (Hill Collins, 1990). At the same time, Hill Collins argues that women and girls of lower social status or non-majority groups develop agency to counter oppression. Their discourse and social practices may at times reveal resistance or a sense of agency in opposition to dominant discourses. In the international development discourse, particularly around girls’ education, oppression is described instrumentally as ‘‘barriers.’’ In this analysis, we illustrate how gender relations are constructed in relationship to ethnic identity and tradition together with poverty. We critically analyze differing constructions of gender relations and how they may limit girls’ capabilities; and how girls’ discourse and social practices reﬂect possible capabilities for well-being. Less common in the development literature on gender and girls’ education is the attention to ethnic identity and tradition as social constructions, and a critical analysis of the discourses and social practices that articulate ethnic traditions (see Vavrus, 2002). The social construction of ethnic cultural traditions, Vavrus illustrates, can be critically analyzed through the lens of

Limits of and Possibilities for Equality

153

international organizations’, governments,’ and local communities’ discourses and practices. Ethnic identities and traditions, similar to gender relations, are socially constructed with their meanings and practices changing over time in relation to broader political, economic, and cultural struggles (Vavrus, 2002, p. 369). Vavrus points out that international development discourse often articulates the concept of tradition to describe ethnic or cultural impediments for non-majority ethnic groups’ participation in schooling, rather than understanding how these traditions, as historically and socially constructed, may also be challenges to colonial or majority group imposed schooling. In other words, these dominant discourses about ethnic traditions and groups can perpetuate ethnocentrism in policy and practices, and they fail to account for the complex and dynamic nature of culture and cultural change. In our analysis, we examine how the discourse of ethnic traditions has been used in national policies and in the public arena to articulate ethnic groups as ‘‘backward’’ or ‘‘not progressive’’ (la c ha^ u), and to describe gendered roles _ _ and experiences in these ethnic groups. The discourse of ‘‘backwardness’’ has become prevalent in the Doi Moi reform, in which economic and social reforms are aimed at Vietnam’s development to compete in the international economy. ‘‘Backward traditions’’ have been used to refer to modes of cultivation and farming among ethnic groups, the marriage of young boys and girls, and the cultural rituals and traditions of these ethnic families and children (UNHCR, 2002). We compare and contrast discourse at the government and international levels to discourse related to ethnic traditions at the community and individual levels, and we examine how these discourses impact social practices of non-majority ethnic girls. In taking this approach of analyzing the discourses and social practices related to poverty, ethnic traditions and gender relations at the government, community, and individual levels, we attempt to move the dialog and practice beyond two often competing camps in gender and development work, the economic and the cultural arguments (Zhang, Kao, & Hannum, 2007); or the anti-poverty and economic efﬁciency approaches on the one hand, and the gender subordination approach on the other (Moser, 1993). Rather, our analysis suggests that gender, ethnic traditions, and poverty are constructed and negotiated as interacting forces in the well-being of nonmajority ethnic girls’ lives in Vietnam. Further, we argue for approaches to move beyond instrumental interventions and explanations, for approaches that account for girls’ and their families’ agency to change these issues in their lives.

154

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

METHODOLOGY AND METHODS Sutton (2001) suggests that too often policy research on girls’ education takes an instrumental approach, conﬁned to technical ‘‘barriers’’ and ‘‘solutions’’. Less research takes a deeply interpretative or critical approach that is able to ‘‘capture the complexities of local practice and institutions and their relevance to policy formation’’ (p. 95). This study takes a critical analytical approach aimed at understanding how local people – and especially girls – construct the issues affecting their lives. In particular, we seek to understand their discourse and social practices related to poverty, gender relations, and ethnic cultural traditions. In their book, Working Method, Weis and Fine (2004) put forth an approach in which the analysis of ethnic groups is not represented as isolated and bounded; rather a critical analysis must attempt to make visible the relations within and among groups and larger sociopolitical phenomena (p. xvii). Three principles guide this analytic approach. First, this critical analysis aims to understand the lives of people within contextual understandings of economic, ethnic, and gender relations. In this study, we speciﬁcally aim to understand the social and economic conditions of poverty and gender as they relate to girls from different ethnic groups, and the constructions of ethnic relations and traditions in the wider social and economic environment of Vietnam. A second principle is that categories of social identity are real but ﬂuid as they affect people’s lives. For example, ethnic identity, social class, and gender are real categories, not to be essentialized, but to be understood in terms of how these participants, especially the girls who are not continuing on to secondary school, make sense of and enact their gender, ethnic identity, and social identity. Third, in understanding these social identity categories, variations of meaning are sought, and outliers are represented. Secondary age out-of-school nonmajority ethnic girls represent outliers in that most research and policy discourse and practice do not attend to this population. At the same time, out-of-school secondary-age girls are a majority in contrast to those who are in school, in these communities in Vietnam. Throughout the analysis section, we illustrate differences among the ethnic groups, as well as contrasts among government, community, families’ and girls’ discourses and practices. The primary researchers in this study were Vietnamese nationals (mainly Kinh majority researchers4) from the Research Center for Ethnic Minority Education (RCEME) in the Ministry of Education and Training. The RCEME researchers worked with non-majority ethnic group educators from the communities to carry out the research. We, as U.S. based

Limits of and Possibilities for Equality

155

researchers, worked collaboratively with the researchers to design, collect, and analyze data from the study. The research and analysis were conducted over a 2-year period. Visits to schools and villages in each ethnic community were generally 2 weeks in length with a team of 5 or 6 researchers. The participants included primary and secondary school age girls and boys, their parents, community leaders, village heads, school personnel, district education ofﬁcials, and provincial and district People’s Committee members. The children, parents, and community leaders are from four different ethnic groups, residing in four geographically distributed and economically disadvantaged districts in Vietnam (SCEMMAA, 2005): the Hmong (Bac Ha district, Lao Cai province), Gia-Rai (Chu Pah district, Gia Lai Province), Ba-Na (Mang Yang, Gia Lai province), and Khmer (Cau Ngang district, Tra Vinh province). Participants and the communities were selected based on purposive sampling. The communes and villages within the district were selected based on two criteria: the largest gender gap in lower secondary school enrollment and the greatest number of out-of-school girls. Out-of-school girls and boys were identiﬁed from school and community lists. In some small villages, all secondary school age (11–14) girls who were out-of-school were identiﬁed and interviewed; in other villages, girls were randomly selected from the list of out-of-school children. The selection of participants from the community, school, and district and provincial levels reﬂected nearly equal representation of males and females. The participants were chosen based on their responsibilities for education and commune development. Researchers interviewed individually 111 outof-school girls and boys; conducted 104 focus groups with approximately 700 parents, teachers, and community and education ofﬁcials; and conducted 51 classroom observations.

BACKGROUND ON KHMER, HMONG, GIA-RAI AND BA-NA ETHNIC GROUPS Participants from four ethnic groups, living in three regions of Vietnam, participated in this study: the Khmer (Tra Vinh), the Hmong (Lao Cai), the Ba-Na, and the Gia-Rai (Gia Lai) (Fig. 1). In Tra Vinh, a coastal province in the Mekong Delta, the Khmer include 30% of the population among the Kinh and Chinese majority. The Khmer are one of the largest ethnic groups in Vietnam, with a population of more than one million, and they are concentrated in southern Vietnam. Tra Vinh

156

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

Fig. 1.

Map of Vietnam and Three Provinces Included in the Study.

has one provincial town and seven districts. Two communes in Cau Ngang district were included in this study. Nearly 30% of the households in one commune and 50% in another are classiﬁed as extremely poor, and the majority of these are Khmer (SCEMMAA, 2005). The main income production is from agriculture, despite being a coastal province. Villages around these commune centers are closer than in the other communes in this study, with the furthest village being 7 km from the commune center, where the lower secondary school is located. The delta region with many ﬂooded rivers in the rainy season makes transportation difﬁcult, even with short distances. In each of the primary and secondary schools, there are Khmer teachers, although fewer female teachers than males in all schools. The enrollment rate is generally higher among the Khmer than the other ethnic groups in this study, nevertheless, the girls’ enrollment rate drops off faster than boys’ enrollment rate, particularly in the lower secondary grades.

Limits of and Possibilities for Equality

157

Lao Cai province is in northeast and northwest Vietnam, on the border with China. It is approximately 300 km away from Ha Noi, and is mountainous with different altitude levels. There is one provincial city, and eight districts. The population of Lao Cai is more than a half million, and includes 25 different ethnic groups, which comprise 70% of the population. The largest group is Hmong, with a population of nearly 1 million spread across several provinces. Agriculture and forestry contribute 78% of the income in this province. In this province, the two Lao Cai communes included in this research study were located in Bac Ha district and are extremely disadvantaged as measured by the government’s criteria (SCEMMAA). Hmong ethnic groups compose the majority of the population. One of the communes had 10% poor households; the other had 50% (SCEMMAA, 2005). In some cases, villages in these communities may be up to 30 km from the commune center and without passable roads, where the main primary and secondary school are located. In the commune with more poor households, the enrollment rate of girls to boys is less than half at every grade level, and decreases in the upper grades. In Grade 5, the enrollment rate of Hmong boys is 3.3 times that of the Hmong girls, and in Grade 9, this ratio is 5.2 times. In the second commune, which is closer to the district center, gender parity occurs in the lower grades, but in the upper grades a disparity of 2.5 times exists between boys and girls (Bac Ha District Education Ofﬁce). The Gia-Rai, whose total population is approximately 300,000 comprise 30% of the population of Gia Lai province in the Central Highlands, a mountainous region. Gia-Rai participants resided in Chupah district in the north. Although export crops are produced in this province, the Gia-Rai and Ba-Na ethnic groups primarily do small scale, subsistence farming or cultivation of forest plants. In the communes included in this study, 30% of the households are extremely poor. The communes are located approximately 20 km away from the district town. In one of the communes, a pilot bilingual project is being implemented, and there are several Gia-Rai teachers. In both of these communes, girls are consistently enrolled at a lower rate than boys, starting in Grade 1, and a lower rate remains through Grade 8, though a decrease in enrollment occurs for both boys and girls starting in Grade 6. In one of the communes at the time of the study, boys averaged 40–60% enrollment rate, whereas girls averaged 10–20% (Chupah District Education Ofﬁce). The Ba-Na ethnic group also resides in Gia Lai province, comprising 12% of the population with approximately 150,000 people. Participants in this study came from Mang Yang district, approximately 40 km from the main

158

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

provincial city. The two communes involved in this study were nearly 100% Ba-Na people, with poor households making up more than 70% of the population of these communes (SCEMMAA, 2005). The villages in the communes lie up to 25 km away from the commune center without passable roads. There was only one teacher of Ba-Na ethnicity in these schools, though non-Ba-Na primary teachers could speak the language to some extent. No secondary teachers could speak Ba-Na, which is not a written language. The enrollment rate of school age children in these two communes is approximately 1:3 and 1:4. Gender parity exists in both of these communes in the primary grades, with girls having a slightly higher enrollment rate than boys in one of them, in part due to over-aged attendance. However, in Grades 6–8, girls’ enrollment rate drops considerably compared to boys, with less than 10% in one commune and approximately 20% enrollment rate of girls (80% for boys) in the other lower secondary school (Mang Yang District Education Ofﬁce).

FINDINGS Utilizing Sen’s framework of capabilities for well-being, we organize the analysis of the discourse and social practices using these terms. Within the capabilities framework, we analyze how different groups in society construct and value ethnic traditions and gender relations. We extend Sen’s framework, and the list of capabilities that Walker (2007) and others have developed for education, by suggesting that these girls and their families also construct a discourse and try to negotiate capabilities related to a lack of cultural choice, particularly that of a mismatch in the language of teaching and learning and culturally appropriate and relevant content. Although Walker’s capability list includes the broad concept of respect and recognition of one’s race, religion, or ethnicity, our analysis proposes more speciﬁc capabilities for the freedom to choose well-being within one’s ethnic identity and traditions. These cultural capabilities are particularly relevant in societies where non-majority ethnicities have not been valued or respected.

PHYSICAL CAPABILITIES When girls, their parents, and community members utilize the discourse of poverty they articulate different meanings, values and choices within these

Limits of and Possibilities for Equality

159

ethnic communities. One of the meanings of poverty across all four groups was a lack of physical capabilities, which includes hunger, poor health, and poor family welfare. The discourses and social practices of hunger and poor family welfare are manifested differently in these four ethnic communities. For Hmong girls, hunger was not having enough rice, or other food, to eat. Most Hmong families have plots of land where they grow primarily maize, and if possible, rice. Some may also have water buffalo and chickens. A lack of sufﬁcient rice and other food is further complicated by the need for children to board at the lower secondary school because their homes and villages were often 15–25 km away from the school. As boarders, they must bring rice or money to pay for food, neither of which the poorest families have. Families’ and girls’ discourse suggests their inability to attend boarding schools because they do not have sufﬁcient rice to eat. A lack of rice is a symbol of poverty for these families in Vietnam, and in the mountainous region where these Hmong families live, it is a difﬁcult and expensive crop to grow. Household decision-making within the Hmong family, based in gendered relations and in consideration of long-term prospects for their children, ﬁrst allocates food to the boys, leaving the girls with either little food or only maize, which has a different nutritional value and is regarded as lesser valued in comparison to rice. A father of a Hmong girl exempliﬁes these conditions, ‘‘If the girls stay home or study close by the house, they can eat men men (a ﬁnely ground maize dish) with the parents. But if they go to school, they need to bring rice. But we do not have rice, and neither do we have money.’’ The lack of rice and the value given it by society and schools, creates a discourse of poverty among these families, which is enacted in practices that exclude girls from school. The government policy discourse, and the discourse among Kinh majority people, is that the Hmong practice traditional and ‘‘backward’’ agricultural practices (UNHCR, 2002), often slash and burn techniques, and that this form of agriculture is a cause of hunger and poverty. This discourse suggests that the Hmong do not grow rice because of these backward farming practices, and, in turn, they do not contribute to the well-being of their families or the development of society. In contrast, Hmong families and girls suggest that traditional forms of farming mean they farm in a manner they know and through which they are able to survive. Hmong farmers ask for additional knowledge and support to farm better, and they expect that their children would learn better agricultural techniques in school. As one Hmong father said, ‘‘I want my daughter to learn how to raise cows, goats and pigs’’ – animals that are not currently being raised in villages, but might be sold in

160

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

the market. In addition, they recognize the need to access markets, and the need for equipment and technical skills. Thus, Hmong families value instrumental goods (e.g., equipment and markets) for economic development; however, beyond instrumental needs, they lack choice in different opportunities for a healthy life. Their choices have been limited by a discourse that regards their way of life as backward with practices that try to change their culture, rather than giving them choices. They are also limited by institutional practices (e.g., those of schools) that value rice as a necessary food, and by the physical environment and social development that limits choices for food and productivity. In contrast to the Hmong girls, Ba-Na and Gia-Rai girls experience hunger that is often chronic and severe, due to a lack of availability of any food. Ba-Na and Gia-Rai girls said they often had no food before going to school, although they might have something later in the day. Access to food and cultivating crops is different for these ethnic groups than for the Hmong, in part due to the diverse geographical regions in which they live. The Ba-Na in Gia Lai Province, the Central Highlands region, live in remote areas close to forests and mountains, and are cut off from the province’s mainstream economic activities and the pervasively cultivated industrial crops such as coffee, black pepper, and cashew nuts. This means these ethnic families have little land to grow food and do not have the economic resources or availability of markets to buy food. As one Ba-Na girl explained, ‘‘I had nothing to eat in the morning. At noon there is lunch, but not enough. We eat cassava leaves for vegetables, and for dinner, we eat rice mixed with cassava.’’ Many Gia-Rai families obtain their food from forest plants. Gia-Rai girls said they were often hungry, and their families did not have enough food on a daily basis. A Gia-Rai girl explained, ‘‘We eat mi leaves (a food from the forest) . . . we have no money to buy meat or ﬁsh. We eat no breakfast, only lunch and dinner.’’ A lack of food for these girls and families is caused, in part, by a lack of land on which to grow food. The lack of cultivable land between Ba-Na and Gia-Rai families reﬂects a contested situation in the Central Highlands. Land ownership recently became an option for many of these families; however, other ethnic groups’ in-migration to this region has caused considerable conﬂict over rights to land (UNDP, 2002). Until recently, the land rights law acknowledged men as head of household and the rightful owners of land, and women in these matrilineal societies were disadvantaged. As a result of in-migration and lack of female registration for ownership, many families from these ethnic groups have moved further away from cultivable land to live in the forest

Limits of and Possibilities for Equality

161

(UNHCR, 2002). In addition to not having an equal opportunity to own or farm land, many Ba-Na and Gia-Rai families lack the knowledge and skills to take advantage of the new land ownership law, what Sen (1999) calls a deprivation of information to make decisions and the freedom to have choices. Nearly 80% of land use certiﬁcates in rural areas are still registered to the male head of household (ADB, 2002, p. 14), suggesting that Ba-Na and Gia-Rai female heads of household are not equally registering for land ownership. Government discourse reﬂected in land and anti-poverty policies suggests that these ethnic groups are choosing ‘‘traditional’’ farming (foraging) practices even though the policy allows for them to farm in more productive ways. The discourse of ‘‘traditional’’ farming is then used to suggest ethnic groups’ practices inhibit ‘‘progress’’ and are perpetuating hunger and poverty. The discourse masks a deeper analysis of the underlying causes of why harvesting of forest products continues among these ethnic groups. The tenuous political situation in this region prevents many Ba-Na or Gia-Rai families from outwardly challenging the government discourse, though families did suggest that they were being pushed from their land. Khmer families are affected by hunger because there is a lack of arable farming land in the overpopulated region of the Mekong Delta, and thus they tend to work on other landowners’ farms, or they work in the city. A Khmer girl from Nhi Truong commune, Tra Vinh Province, said she and her siblings did not have enough food at home, and were dependent on their parents, who worked far away in Long Khanh. ‘‘The family does not have enough food. Sometimes I do not eat for two to three days,’’ she said. Khmer families in these communities are dependent on the harvest of crops from other landowners, at which time they can bring food to their families, or on having a family member work so that they can buy the needed food. Khmer families, similar to the Ba-Na and Gia-Rai families, lack the capability to grow food, although the reasons for not owning land are different in these two regions. In contrast, the Hmong families have land and can grow food, but the amount is not sufﬁcient, neither is it the kind of food (i.e., rice) required to board at school. Further, they do not have access to markets, similar to the Ba-Na and Gia-Rai, to sell their crops or buy other foods. In all of these cases, the national policy discourse regards farming practices associated with non-majority ethnic groups as traditional and backward, yet these families and communities are affected by poor economic development (local economy and markets, infrastructure of roads, etc.), a lack of technical knowledge about more productive farming practices, and adverse policy effects, such as land ownership rights. Hmong

162

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

farmers discuss the need for development of roads and markets, as well as the need for further skill development. Khmer families suggest the need for land and local enterprises where they could work. Pro-poor policies are aimed at addressing these needs; however, Baulch et al. (2004) conclude that the program to improve the economic development of the poorest communities has not been successful in targeting these communes. Rather, the wealthiest communes in poor districts have received the most assistance, thus leading to divergent development. Improved access to arable land, to productive farming skills, or to markets may provide some of the instrumental goods (Sen, 1992, 1999) to alleviate poverty; but the provision of these instrumental goods, while necessary, will not be sufﬁcient to alleviate poverty. Sen argues that it is not the income or the products generated by having access to markets or skills, rather it is the importance of the freedom to exchange, to seek better lives through transactions (p. 112). By not having the freedom or the capability to choose land ownership or crops to grow, these families, and their girls, are forced into hunger. Creating conditions that allow these families choice in how to effectively utilize these goods (land, food, markets) for real opportunities would provide greater capabilities to improve their well-being (Sen, 1992, 1999). Hunger is closely related to another physical capability – poor health, particularly for the Gia-Rai families and girls. The discourse of Gia-Rai girls and their mothers is that they are often too sick to attend school or to remain in school. Illnesses include stomachaches, gastric pains, headaches, and ﬂu. One reason so many Gia-Rai women and girls have poor health may be connected to heavy workloads and nutritional deﬁciencies caused by the main vegetable in the Gia-Rai diet. Women and girls reported they eat la mi, a type of leaf that contains toxins that can cause gastric pains. Gia-Rai mothers appeared to be thin and pale, regardless of age. Few Gia-Rai families have money for medicine and treatment of any illness. The government discourse, reﬂected in anti-poverty policies and initiatives, is that health care is available in each community and additional ﬁnancial support is given to poor families through the provision of a health care card that provides fee reductions. However, social practices of Gia-Rai families, who migrate often to ﬁnd work, inhibit them from accessing health care. The government prohibits registration in another district when families migrate, thus inhibiting many Gia-Rai and some Khmer families from having a health card and access to health care (World Bank, n.d.). In addition, many poor families are not using health care services considerably

Limits of and Possibilities for Equality

163

more even if they have a health care card, in part because the fee reduction is negligible (World Bank, n.d.). The lack of family well-being, not only the hunger or health of individuals, but the overall well-being of the family unit, further compounds their lack of capabilities, and has differential impacts on the girl child. A Gia-Rai girl reﬂected the tension between family well-being and school: ‘‘My dad asked me to go to school, but I didn’t want to because there is no one to work in the house, as my mother died early.’’ The social practice of girls staying home rather than boys to care for ill family members or to support the family wellbeing reﬂects gendered relations. As one Hmong mother suggested, girls are better able to care and assist at home. A Gia-Rai father described how illness affects his family and particularly the girls. ‘‘Many families are in the same situation,’’ he stated. ‘‘I have four children attending school. It is very difﬁcult to afford clothes and school fees . . . and a mother who is always sick, and it is very difﬁcult to work. Many girls have sick family members,’’ he continued. ‘‘They are needed to help out at home.’’ This social expectation that girls provide care to family members supports the physical well-being of families and to some extent girls themselves, while also perpetuating gender inequalities that deprive girls of freedom to choose long-term well-being. Some girls try to renegotiate the demands for family well-being and their possibilities to choose well-being. For example, Ba-Na girls were seen regularly taking a younger sibling to school, which allowed for the sibling to be cared for and the girl to attend school. There are negative consequences of this care, which we discuss below with regard to girls’ work; however, alternatives for supporting individual and family well-being are sometimes sought. Care for family well-being is related to the need for household and farm labor, and paid employment, discussed in the next section. Addressing physical deprivation, such as hunger, poor health and wellbeing with instrumental approaches to poverty, such as income generation, food programs, or stipends may address some short-term needs, but they may not adequately address underlying causes that prohibit freedom to choose well-being. For these four different ethnic communities, overt and covert discrimination is one factor that deprives these ethnic groups of the freedom to choose physical capabilities for their well-being. Even laws that grant land rights or health care, when examined in relation to gender and ethnicity, still reveal discrimination in the social practices. The public discourse, in the poverty reduction plans and socio-economic development programs for ethnic minorities, aims to create equality; however, ineffective implementation and practices elude the possibilities of that equality.

164

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

ECONOMIC AND SOCIAL CAPABILITIES Families’ instrumental poverty, in terms of household income and consumption, and the physical deprivations discussed earlier, lead to a need for household, farm, and paid labor. Studies on child labor in Vietnam have found that the working hours of primary pupils from the poorest quintile are double those of pupils from the richest one. Further, the working hours of female pupils are 20–50% longer than those of males from all quintiles and this rate is 100% more for female Hmong children (ADB, 2002). Although Chant (2007) suggests that Sen’s framework does not adequately account for work and time use, differences in girls’ and boys’ work and time use can result in limiting the current and future economic and social capabilities of young girls. Sen suggests that social opportunities and social roles affect women’s capabilities differently than men’s, and here we discuss work and related factors that affect the social opportunities and roles of young women. Work among children, particularly those at the secondary-school age, has been categorized as having two possible effects on young people’s capabilities: work that is productive and liberating, and work that is harmful and deprives young people of current and future capabilities (Population Council, 2000). Liberating work can be an important part of an adolescent’s maturation, as they apply skills learned in school and elsewhere, and as they develop autonomy and a sense of self-worth. Work that deprives young people of capabilities tends not to utilize new skills, does not promote autonomy and self-worth, and may be limiting one’s current and future capabilities, including schooling, a healthy life, and future employment. Most of the work performed by girls in these ethnic groups limited their social capabilities. However, some girls, and their parents, valued and articulated a need for liberating work that could utilize skills and enhance their self-worth, as well as the worth and capacity of their families. The need for labor to support economic and social capabilities of the children themselves and their families existed in all four ethnic groups. Girls and their families construct current work obligations as limiting their capabilities, through depriving them of schooling, while also providing for family economic and social well-being. The kind of work, extent, and the ways families’ and girls’ viewed work as depriving them of capabilities varied among the ethnic groups. Ba-Na parents are often away from the home working in ﬁelds or at wage labor jobs, and girls are expected to combine domestic chores such as fetching water and caring for younger siblings while attending school.

Limits of and Possibilities for Equality

165

Gendered relations in this matrilineal ethnic group give girls and women full responsibility as heads of household, thus requiring girls to perform the majority of work. This kind of work is generally not regarded by girls or other community members as promoting capabilities for well-being among girls themselves, albeit it supported the well-being of families and communities. Some of the work Ba-Na girls do supports the ethnic traditions of the community. Girls are expected from the age of eight to prepare for cultural ceremonies, such as weddings or funerals, which may require a week or two of work, keeping them out of school. However, Ba-Na girls create social practices that enable them to attend to their work duties while also allowing them to attend school. As stated earlier, girls often carry younger siblings to school with them, caring for them while they study. Although this practice perpetuates the imbalance in girls’ responsibilities for childcare or family welfare, it is also a negotiation for girls to ﬁnd ways to continue their schooling. Few opportunities existed for Ba-Na girls to obtain paid work in their communities. An analysis of the Doi Moi economic reforms suggest that women in rural areas fare more poorly than men since the reform affects access to the labor market (Desai, 2001). Parents and community members’ discourse reﬂect the lack of economic and social opportunities. When few jobs exist in the labor market for women, girls and their families do not express much hope for these opportunities. Teachers and community leaders recounted examples of girls who stayed in school but who did not gain immediate rewards after school by acquiring a job. ‘‘An ethnic minority student who completed ninth grade could not ﬁnd any job in the commune with her certiﬁcation. She got discouraged, and then dropped out of school. To be farmers, they already know what they need to know by the ﬁfth or ninth grade. Even with education, eventually the children would end up working in the ﬁeld.’’ Another teacher noted that the lack of success for girls to ﬁnd work in the labor market discouraged others from continuing their education. ‘‘Some older students completed twelfth grade and took the exam to continue to pre-university, yet at the end, they still returned to work in the ﬁeld,’’ the teacher recalled. ‘‘The younger students said it is good to get more education, but after seeing the older students return to work in the ﬁelds, they did not want to study further.’’ In addition, some community members suggest that a lack of economic and social opportunities is multiplied when families are lower ‘‘status’’. A village head stated, ‘‘After graduation, they can only get a job if the family has money, status, or connections. But poor families like us probably only can stay home, so we can only send the children to school up to ﬁfth grade to be literate.’’

166

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

Although, the policy discourse promotes additional education (universal secondary education) as a means for economic and social development, families and communities, along with data on the labor market, suggest that this discourse is disconnected from the reality of few economic and social opportunities for these students. While some discourse in the international development literature and among educators suggests that girls and their parents in these communities may not have educational aspirations, it appears that it is not one of lower aspirations. Most girls and parents did aspire to work and to be educated. Rather, their assessment is one of the realities of the conditions and the hope they have for converting those aspirations of education and work into better well-being. Walker (2007), citing Hage (2001) states, ‘‘Hope . . . is about the sense of possibility that life can offer. Its enemy is a sense of entrapment not a sense of poverty’’ (p. 3). For the Khmer, instrumental poverty requires girls and boys to perform wage labor with their families, causing a lack of other educational and social choices and opportunities for well-being. The Khmer have relatively greater access to the formal economy and wage labor opportunities due to their geographic location. In the Mekong Delta where the Khmer live, a large industrial region has developed, requiring wage laborers to work in export industries. As one Khmer girl stated, ‘‘My family is so poor I have to work with my parents. My eldest sister already works in the city.’’ When a ﬁfth grade Khmer girl in Truong Tho commune was asked if she would continue to sixth grade, she replied, ‘‘My parents said that after ﬁfth grade, they will send me to the city to look after other people’s houses.’’ Her older sister was also working in the city. Initially, families and girls are drawn to the prospects of immediate income; they also report coming home when they can no longer ﬁnd work. This type of work can perpetuate long-term capability deprivation. Migration to cities for work may reduce extreme poverty, but these families are also found to be highly vulnerable to returning to poverty (Vietnamese Academy of Social Sciences, 2006). The social practice of migration to improve one’s economic opportunities also has cross impacts on families’ and children’s social opportunities. The government’s prohibition against issuing identity cards when these families move temporarily further reduces their opportunities to attend school. Many Khmer children reported not being able to complete a grade as they needed to provide identiﬁcation to take ﬁnal exams. The policies of residency identiﬁcation disproportionately affect the poorest families who are mobile in seeking work. For some Khmer families and children, their practices of migrating suggest that working in the city provides opportunities for economic well-being, even while it does not provide other

Limits of and Possibilities for Equality

167

capabilities for well-being, such as education and health. Economic and social opportunities for well-being do not sufﬁciently exist in their communities. Hmong girls tend to work twice the amount of boys (ADB, 2002): caring for siblings, doing household chores, collecting wood and water, and caring for buffalo. This kind of work is necessary for family and community survival, it does not provide, however, sufﬁcient opportunities for learning or using new skills or developing one’s self-worth. One of the dominant discourses that exist among Hmong parents and the community that perpetuate girls’ household and farm work is a lack of opportunities for Hmong women to work in paid labor in these communities. This discourse relates to the economic and social opportunities that do not exist for many Hmong families, and particularly women, similar to the Ba-Na and Gia-Rai. Unlike the Khmer in the Mekong Delta, Hmong young people, and particularly girls, tend not to migrate to the city, as large cities and industrial areas are much further away. Additionally, Hmong families are concerned about maintaining the autonomy and identity of Hmong people, and, thus, they do not encourage social assimilation into the majority Vietnamese society. Not only are Hmong people somewhat isolationist given a long history of forced migration (Chan, 1994), they also encounter discrimination by mainstream Vietnamese society and, as such, they are not often able to integrate without assimilation. For example, very few Hmong people have made it through the education system to become educators. In the district in which this research took place, there was one young Hmong man who had recently become a teacher. Hmong girls have even an even smaller chance than boys of obtaining work in the paid labor market. One area for developing and using new skills and in turn, creating economic and social opportunities, is the production of Hmong cultural goods for sale in nearby towns. For some parents and girls, weaving and sewing Hmong goods is a valuable form of labor; it earns money for the family and continues a cultural tradition within the community. As community members recounted, ‘‘If Hmong girls stay home, they can earn income by making textile bags for their fathers to sell for income.’’ Some young girls and fathers recounted the story of a few women in the community who now create Hmong cultural crafts and have earned an income as well as new social networks and opportunities in the community. This type of work represents possible economic and social opportunities for Hmong girls; it also suggests that Hmong girls’ freedom to choose economic and social opportunities remain limited. Parents and community members value opportunities that support ethnic cultural practices and economic

168

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

development of the community over work that supports the economic development of the country, which might represent assimilation or a loss of identity for Hmong people. In contrast, when girls were asked what kind of work they wanted to do in their future, most said they aspire to be teachers or doctors. Some parents also saw this as valuable work, as these were the two main paid work opportunities in the communities. These two kinds of jobs may be regarded as valuable within these ethnic groups because girls remain in the communities, and they support the education and health of the ethnic communities. A second discourse among many Hmong parents and communities members is that girls’ work is valued when they are young because they can contribute to their family; once they are married, they become part of the husband’s family and they no longer contribute economically or socially to their own family. This discourse has developed through gendered relations – including patriarchal norms – within the Hmong community, and through practical economic and social needs within the community. The practice of early marriage is often regarded in the girls’ education and gender literature as a negative cultural practice, and in Vietnam, it is associated with ‘‘backwardness’’ among ethnic groups, particularly the Hmong and the Ba-Na. In this analysis, we illustrate how early marriage is also a social practice, in part related to economic deprivation and the need to work. As a social practice, it limits girls’ social opportunities, whereas some community members and parents also regard it as enhancing family and community welfare and short-term economic opportunities. For many of the Hmong girls in this study, marriage occurred at a young age (before 18), although the Vietnamese marriage law requires consent before the age of 18. A Hmong girl joins her husband’s family after marriage, and thus the value of her work from adolescence onward primarily beneﬁts the husband’s family. Since this means that the costs of educating a Hmong girl will not result in economic returns for her birth family, many families decide that educating girls is not a cost-effective strategy for their immediate well-being or future livelihood. In addition to an economic rationale, this social practice is embedded in historical patterns and beliefs about gender relations between girls and boys, and thus represents not only parents’ attitudes, but also community traditions and institutions. As expressed in one community, some Hmong people believe ‘‘girls are daughters of other people, only boys should be allowed to go to school. Girls do not need to study much. Since they are others’ daughters, why should [we] invest much in them? Girls will get married and, after that, will work and stay in the husbands’ families.’’ Given the lack of alternative

Limits of and Possibilities for Equality

169

economic and social opportunities for girls in these communities, their household labor is not seen as valuable as a boy’s potential labor opportunities. Both the value of girls’ work, which diminishes over time – particularly after marriage when it beneﬁts the husband’s family only, and the value of money, gifts or food gained through having a daughter married lead many families to limit girls’ educational opportunities as they seek value in more immediate economic opportunities. As a Hmong girl who had dropped out of school conﬁded, ‘‘I did not want to marry, but my family is too poor and accepted their [the boy’s family] gifts.’’ The village head explained that the boy’s family had arranged for an early marriage mainly ‘‘to have more labor.’’ Most girls did not want to marry early and they wanted to continue in school. They resisted and yet they recognized it as a negotiation between the economic and social well-being of their families, and their own well-being. Practices of early marriage are often regarded as a result of patriarchal norms and gendered roles. However, in matrilineal ethnic groups, such as the Gia-Rai and Ba-Na, the traditions of early marriage and labor of girls are not substantively different from the patrilineal family structures and practices of the Hmong. Women, as heads of household, stay home to work and care for the family. Ba-Na and Gia-Rai girls also tended to marry early; the main difference being that girls chose their husbands. Community and cultural norms still expected them to marry in early adolescence. Once married, they took full responsibility for family well-being and household labor. While girls potentially have more choice of whom they marry in these communities, they still lack choice in their social and economic opportunities in the community. Chant (2007) refers to this as the ‘‘feminization of responsibility and/or obligation’’ rather than the feminization of poverty. Despite potentially more choices, whether in public spaces, such as the labor market, or in private spaces of the home, such as choosing one’s spouse, women have greater responsibilities and obligations that affect the conditions for well-being. Few choices exist in these ethnic communities, and even fewer choices exist for girls’ economic and social well-being. Girls are limited in their social well-being through demands for work that are often harmful, and not liberating. Opportunities for knowledge, skills, and liberating work for girls, as well as boys, could improve their economic and social well-being. For the Khmer, opportunities for meaningful and liberating work in their communities for their families may decrease the social deprivations experienced by children as they migrate. Changing the law on residency cards would also affect their choices. For the Hmong, choices for

170

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

meaningful work that supports the cultural identity and traditions of their community are important. For the Ba-Na and the Gai-rai, opportunities for meaningful work in their communities are also important. Other factors of discrimination against and mistreatment of these ethnic groups in the workforce, while not fully explored in this study, need to be further examined with regard to the limitation of their capabilities. Simply providing a secondary education may not give these non-majority ethnic girls more freedom to choose their social and economic capabilities. They have aspirations for education and work; they also aspire to contribute to their community. However when the valued knowledge, skills, and kinds of liberating employment are limited, or when discrimination is present, girls’ choices are also limited. The lack of choices tends to reinforce gendered relations and ethnic traditions of work and social opportunities. Economic and social capabilities are also related to the relevance of what these young girls learn in school, which is addressed in the next section.

CULTURAL CAPABILITIES Girls’ lack of opportunities for equality and well-being are also reﬂected in schools’ use of dominant discourses and practice in the curriculum and learning environment. Unterhalter (2007a, 2007b) and Walker (2006), following Sen’s framework, suggest that education is itself a necessary capability for equality. They also note that not all education promotes other capabilities, such as the conversion of learned skills into economic or social capabilities. Discourses about and practices of schooling affect girls’ capabilities differently from boys, and non-majority ethnic groups differently from the majority Kinh. We extend this analysis of capabilities in and through education to suggest that for non-majority ethnic groups, a lack of culturally relevant education is a particular limitation to these girls’ capabilities. The culturally relevant educational issues raised by members of these ethnic groups include language of instruction, curriculum content and pedagogy, and caring and culturally respectful relationships in schools. Girls, their parents, and some educators, particularly for Hmong, Ba-Na and Gai-rai ethnic communities, articulated the belief that the girls were not able to understand and were not learning in Vietnamese language. Mothers and fathers of Hmong, Gia-Rai and Ba-Na children remarked that it was very difﬁcult for their children to listen to the lesson in Vietnamese: ‘‘they do not understand very much.’’ Girls said, ‘‘I like school, but I do not understand, and I cannot read the lesson,’’ and ‘‘if I study without

Limits of and Possibilities for Equality

171

understanding, then I do not want to go to school.’’ In school observations in the Hmong communities, students were seen struggling with pronunciation and comprehension of concepts, with little assistance from teachers on learning the language that was not their mother tongue. The community discourse suggested that the Vietnamese language was not taught well enough for them to learn, and that they were not learning other subjects well in Vietnamese. Girls’ difﬁculty with the language in part may be that they lack exposure to Vietnamese language outside the school, whereas boys generally have more public interaction and opportunities to see and hear the language; they are also encouraged to learn and use Vietnamese so they can continue in school and participate in the labor market. In interviews with out-of-school Ba-Na and Hmong girls, a local interpreter was used because the girls could not understand or speak Vietnamese. A majority of boys could communicate in Vietnamese and very few boys needed interpreters. A rationale for the need to speak Vietnamese that parents and community members discussed was access to and participation in the labor market. National policies and documents that address poverty reduction in Vietnam also emphasize the importance of ethnic minorities to speak Vietnamese in order to participate in the labor market (Centre for International Economics, 2002). This discourse and the labor practices privilege the majority language and identity. In effect, the use of a non-majority ethnic language is regarded as a tradition to preserve and use within the communities, in part to sell their traditions to a new tourist market, but it appears not to be regarded as an ethnic tradition that can be incorporated into the larger economic and social values of the society. In contrast to learning Vietnamese, girls’ and their parents’ discourse and social practices illustrated the importance for learning and speaking their mother tongue, while also learning Vietnamese language. Community members believed strongly that teachers need to teach in the mother tongue and Vietnamese. However, in the Hmong communities, no teachers were observed using the Hmong language in the classroom. In the Gia-Rai and Ba-Na communities, a few native speakers used the mother tongue infrequently in the schools. Parents and community members valued having teachers from their own ethnic group, even though this rarely happened. They also emphasized that girls should learn certain ethnic traditions, such as weaving or dance, so they could carry on aspects of the ethnic group’s identity. While learning these traditions is important for ethnic identity preservation, it also limited their exposure to and use of Vietnamese language. Given the perceived future roles and aspirations for Hmong,

172

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

Ba-Na and Gia-Rai girls, the value of learning and speaking Vietnamese language is not high. This is especially true when it is contrasted with the need for girls to speak the mother tongue so that they can carry on and preserve the culture when other mechanisms, such as written materials, are not available. Unterhalter (2007b) points out that sometimes a community may value something that could limit capabilities, but that the more valued states are those that expand capabilities. Sen’s approach emphasizes, however, the importance of a debate and identiﬁcation among communities of what they value, and a dialog between communities and the larger society. In this case, tension exists among the value of capabilities to maintain cultural identity and the value of economic and social capabilities in the larger society. In essence, these girls’ choices are limited because speaking their native language is not valued in the social and economic spheres outside their community. The discourse in the education sector is that non-Kinh ethnic groups do not have sufﬁcient capacity in the Vietnamese language. This is reﬂected in policy debates that focus on improving the teaching of Vietnamese at earlier ages. Many educators emphasized the need to have pre-school Vietnamese teaching so that the children can learn better once they start primary school. In contrast, the discourse in policies and among educators about bilingual education, or mother tongue instruction, most often represents a transition approach, where the language is used for 2–3 years to transition students to speaking and learning fully in Vietnamese. The discourses about the language of instruction reﬂect a perspective about ethnicity and ethnic traditions in Vietnamese society. The Vietnamese constitution and education law provide ethnic minority children the right to learn in their own language in primary school, as well as to have instruction in their second language, Vietnamese. However, both in policy and practice, a model of bilingual education, in which the mother tongue and Vietnamese are languages of instruction, is not being implemented (Chantrill, Lendon, Sit, & Thanh, 2002; Kosonen, 2004), although such a model currently is being piloted with UNICEF support. Rather, teachers primarily instruct in Vietnamese and a percentage of instruction time ofﬁcially is set aside for instruction in the mother tongue. Research has found, however, that students who do not learn their mother tongue proﬁciently have greater difﬁculty learning another language well (Dutcher, 2004). Further, international development research has shown consistently that girls are more disadvantaged than boys in learning a language other than their mother tongue, often because of a lack of social interactions and access to materials (Kane, 2004).

Limits of and Possibilities for Equality

173

Although the policy discourse acknowledges ethnic identity and traditions, in practice the preservation and development of ethnic identity occurs primarily through songs and dance. The practices in schools do not support the preservation and development of ethnic identity through effective language instruction and culturally appropriate content. Further, the discourse that supports non-majority ethnic languages is not supported materially through ethnic group teachers who speak the local languages, and materials in local languages. The lack of language capability (of teachers and materials) is exacerbated by the poor economic situation in which these schools reside. Furthermore, in economically disadvantaged communities, children have little access to written material in either Vietnamese or local languages outside of school (UNDP, 2002). In addition to the language of instruction, the content of education is perceived by community members, parents and children as not reﬂecting the economic, cultural and gender needs of these communities. Therefore, many parents, community members, and some girls and boys believe that education beyond basic literacy and numeracy has little value for their immediate and future social and economic capabilities. Parents and children repeatedly suggested that curricula that includes skills and knowledge about agriculture, animal husbandry (in the case of the Hmong, Ba-Na and Gia-Rai), and marine life (Khmer) would be relevant for the opportunities in their communities. They noted that these skills are important for girls, since girls are also needed to help with farming. For the Hmong and Ba-Na, parents were concerned that their children learn cultural history and traditions. Some of these values reﬂected traditional gendered needs and roles for girls. Learning skills that provided practical knowledge and experience in cultural traditions, such as weaving or dressmaking, was considered important for creating and selling products, and also for cultural preservation and transmission. One of the challenges for the poorest families in these communities is to see beyond the immediate survival needs of the family and the kind of knowledge and skills that are important now, in contrast to the near future, if economic conditions, gender roles, and opportunities for ethnic communities would change. For example, if economic development in the local communities allowed for more employment opportunities for girls, what skills would these communities think the girls should have? One capability deprivation is the lack of information to make different decisions, and the freedom of possible opportunities and choices (Sen, 1992, 1999). Many girls hope, and their parents echoed this, to continue in school and to become a nurse or teacher. These statements reﬂect a realistic desire for

174

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

these ethnic children and families: these are the two professions in which women are employed in every community. However, the knowledge and skills that most girls currently learn, or were not learning, in school did not prepare them adequately to continue their education for these professions. Finally, relations between teachers and students and among students included discourses and practices of harassment and discrimination toward students from non-majority ethnic groups. Students often felt they were treated in a prejudicial manner. Teachers’ prejudicial comments and behaviors negatively characterize their students’ economic and social deprivation, and their ethnic heritage. These experiences were also echoed in a recent report conducted on youth (Vietnam Youth Association, Vietnam Women’s Union, & UNICEF and World Bank, 2006). Girls, particularly at the age of adolescence, were sensitive to statements made by teachers regarding their lack of clean clothes or school supplies; they perceived this to be not only a symbol of poverty, but also of their ethnicity. Hmong and Gia-Rai girls felt excluded and were disciplined for their difﬁculty in speaking the Vietnamese language. Teachers’ practices included scolding girls more often than boys for girls’ lack of language ability. One out-of-school girl observed, ‘‘Sometimes the teacher asked loudly, ‘Why don’t you write?’ [I] went outside and cried. I thought I would stop going to school so the teacher would not scold me anymore.’’ Community members expressed concern about teachers’ beliefs that Gia-Rai girls were poor learners. Teachers’ discourse about the girls is captured in this teacher’s remark: ‘‘They are slow learners and learn so badly that it would be better if they would stay home and get married.’’ Most teachers do not appear to be knowledgeable about or sensitive to the cultural challenges of learning in another language or to the conditions in which girls’ live that affect their learning. Nearly all Ba-Na girls interviewed stated they that had been punished and scolded repeatedly, and they dropped out of school as a result of this punishment. In one of the Ba-Na communities, educators and some community leaders perpetuated the social practice of punishment, stating that girls often behaved badly. Girls, however, and also their families to some extent, felt it was a discriminatory gender and ethnic practice. Only one teacher in these communes was Ba-Na, and community members expressed concern that the other teachers did not know about or understand the challenges girls faced, including learning in Vietnamese language. Social and gendered relations among boys and girls in the schools are also a concern to particularly Hmong and Ba-Na girls and their parents. In boarding communities, such as for the Hmong and Ba-Na, parents are

Limits of and Possibilities for Equality

175

worried for the safety of their girls, particularly as they approach adolescence and the possibilities of pregnancy and marriage increase. Boarding facilities in these schools are not only in poor condition, but girls often are not given privacy or separate facilities (see also Chantrill et al., 2002). Parents and girls report occasional sexual behavior, creating a belief among the community that boarding schools are inappropriate for girls. Hmong mothers are very concerned that ‘‘in the boarding school, boys and girls sleep in the same room.’’ The lack of separate bedrooms causes mothers to worry about the safety and possible sexual abuse of their daughters. One Hmong mother stated, ‘‘I only want to let my sons go to school. If girls go to school and get pregnant, I’d be very ashamed.’’ These girls’ choices and opportunities to go to school are limited by social relations in the schools, and these relations perpetuate the social practice of marriage to avoid shame from a girl’s pregnancy. Walker (2007) suggests two capabilities for education that relate to the discourse and practices of these social relations: bodily integrity and health, and emotional integrity and emotions, which include not being subjected to harassment, fear and punishment by peers or teachers, and being able to make one’s own choices about sexual relationships. A lack of bodily integrity, in these communities, includes forms of ethnic discrimination, which are prejudicial discourses and social practices deeply rooted in schools in Vietnam. In addition, the lack of choices about sexual relations, affected both by boys’ behaviors and parents’ practices, illustrates how bodily and emotional integrity are more than individual capabilities for well-being; there should also be capabilities for well-being among these communities, as individuals’ bodily integrity reﬂects on parents’ and communities’ integrity.

APPROACHES FOR ENHANCING CAPABILITIES FOR GENDER AND ETHNIC EQUALITY The literature is not devoid of strategies and approaches for improving gender equity and equality in and through education (see Chapman & Miske, 2007; Herz & Sperling, 2004; Kane, 1995; USAID, 2007; UNICEF, 2008), and emerging attention to secondary education focuses on particular strategies for adolescent girls (see DeJaeghere, 2004; UNICEF, 2008; Rihani, 2006). Two assumptions, however, permeate this literature: the ﬁrst is that additional secondary education is a key tool for poverty alleviation. While research has found higher economic returns from and growth related to secondary education (see Tikal, 2007), it is also important to consider the

176

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

other direction of this relationship: how poverty, as lack of choice for physical, social, economic, and cultural capabilities, affects the most disadvantaged girls from non-majority ethnic communities from participating in and gaining capabilities through education. The discourse and practices of the ethnic groups in this study suggests that the capabilities gained through education, such as labor market opportunities, choice for well-being in life, and greater information and agency, while important, may not reﬂect the reality of the difﬁculties to convert education as a capability into other capabilities. The lack of physical, social, and economic capabilities limited choices for well-being, and particularly the capability to be educated, whereas the lack of cultural capabilities was a great concern for well-being in and through education. A second assumption of many approaches aimed at improving gender equality is that they fail to consider how ethnic traditions and identity are interwoven with gender relations and poverty. Maslak (2005) suggests that gender and education approaches – and one could extend this to ethnicity and poverty – need to be situated within the social and economic structures, relating them to community needs and local factors and to how macrofactors manifest locally (p. 168). Strategies that aim to improve the conditions and positions related to women and girls in these communities also ought to include the conditions and position of the ethnic groups within the society. When gender roles and relations change in society, men’s roles and power are affected. Similarly, when equality among ethnic groups is sought, the Kinh, the majority group in Vietnam, will also be affected by a change in societal conditions, power, and position. Research on majority/minority group relations in society suggests that societies tend to take different approaches to these relations (See Berry, 1984, 2003). One approach is to exclude non-majority ethnic groups, materially and socially, leading to isolation and marginalization of the ethnic groups. Another approach is to assimilate the ethnic groups into a majority cultural identity and norms. A third option is for both majority and non-majority groups to seek integration, where one’s own ethnic culture and other cultures are valued. These approaches are not discrete; they may occur simultaneously in a society, with one of the approaches describing one ethnic groups’ experience, and a different approach related to another ethnic group. The present policy discourse and practices in Vietnam do not suggest an exclusionary approach, as much attention is given to ‘‘ethnic minority regions,’’ ‘‘ethnic groups and traditions,’’ and to improving the economic and social development of ethnic groups. The discourse, however, suggests an assimilation approach in which the

Limits of and Possibilities for Equality

177

development of ethnic groups is to make them more like the Kinh majority, and this has been attributed to the ‘‘progress’’ of other ethnic groups, such as the Thai (Baulch et al., 2004). Resistance to this approach is evident among these ethnic communities. At the same time, the discourse of non-majority ethnic group parents, girls, boys, and community members generally does not suggest isolation or separation from the majority group or the societal organizations in Vietnam. Hmong community members expressed interest and concern in participating in the educational, economic and social development of the country, but not at the loss of their own cultural identity. The Ba-Na and Gia-Rai groups reﬂected a greater concern for being excluded, being pushed out of their communities. The government discourse suggests these groups are seeking separation and aligning themselves with their ethnic group members in Cambodia, which poses threats to the national identity and security of Vietnam. The Khmer may be the most integrated of these groups, in part because of proximity to economic and social opportunities. Khmer children were most ﬂuent in Vietnamese language, and they tended to work and live in more diverse communities. However, Khmer community members, although less the young girls and boys than the older ones, expressed concern over their freedom to choose well-being in their own communities, particularly their Buddhist religious practices. Gender equality, including the freedom to choose to be educated, to marry or not, and to have meaningful work were all emphasized by girls, and to some extent by their parents and community members. At the same time, community members, and particularly fathers, were concerned about ethnic inequalities that limited girls’ well-being and freedom to choose. They suggested that discrimination in the labor market, and access to land, and the inability to practice their cultural identity and language were all factors that limited the well-being not only of girls, but also of families and communities. While girls, their parents, and community members support approaches that would enhance their physical, economic and social capabilities, such as income-generation activities or skill development, most community members in the four ethnic groups were keenly aware of the need for approaches to be culturally relevant and supportive of their ethnic traditions and identity. Instrumental approaches that aim to raise awareness about legal age of marriage, advocate for girls’ education, or provide additional income are not sufﬁcient. As Unterhalter (2007a) suggests, interventions and institutional building are not sufﬁcient to address the multidimensionality of social conditions and local needs; rather forms of negotiation are needed. Expanding Sen’s (1999, 2004) argument that the

178

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

state, civil society, families and international organizations all have responsibility to develop capabilities, Unterhalter states that laws, policies, and strategies offer some approaches; however, dialog, critical examination, and associated actions are equally important. One approach may utilize community enablement, or ‘‘real community participation’’ (Kane, 2004), where all community members voices are heard, and issues and solutions are identiﬁed and negotiated. This may allow for both gender and ethnic relations to be understood and altered. In such an approach, parents, children, educators, and community members – both female and male and from majority and non-majority ethnic groups – work together to identify issues, to understand what each group values, and to negotiate alternative solutions. Walker (2006, 2007), building on Sen’s work, suggests that respect, recognition and voice are all necessary for the development of capabilities. These capabilities are a process by which to choose and develop well-being. At the same time, the material conditions, or the physical and economic deprivations must also be addressed, and doing so through a process that understands how ethnic groups and girls and boys, women and men value differently these capabilities. This requires recognition of and respect for their voices and their freedom to aspire and choose capabilities for their well-being. For example, a dialog in which the different community members’ voices are heard about the policies and practices of ethnic minority language use in school may allow for greater understanding and negotiation by all stakeholders, and particularly parents and teachers, of its intent and aims. Negotiation of a language policy may also allow for it to be effectively adapted to social structures and local needs. Unterhalter (2007a) illustrates how discussions among Wajir community members in Kenya allowed for understanding and negotiation of strategies for girls’ education that ultimately resulted in greater equity for girls that ﬁt local needs and social-cultural practices and structures. Dialog and associated actions may also move the discourse and related social practices among government bodies, international organizations, and community groups beyond a single focus on girls’ education, to emphasize equality in broader terms that includes ethnic and gender relations, and a broader conceptualization of poverty, as discussed here using Sen’s approach. This study shows that a particular focus on one dimension, such as gender or poverty alone, and the dominant discourses and social practices that organizations and governments create around these dimensions is misguided and does not represent the way that families and children represent and value differently these interlocking dimensions. This chapter is

Limits of and Possibilities for Equality

179

an attempt to enable a dialog and a critical examination of the issues by bringing together the voices of girls, boys, parents, teachers, community members, and district and provincial ofﬁcials from three regions of Vietnam in order to discuss and critique the conditions affecting girls’ lives, and to suggest alternative approaches to conceptualizing and implementing effective change for equality.

NOTES 1. The poverty rate as measured by household surveys in Vietnam is based on World Bank deﬁnitions, and includes both consumption-based food and general poverty lines. In 2004, the consumption-based food rate was equivalent to VND 159,788 per capita per month. The general poverty line is derived by adding a nonfood component to the food poverty line and was equivalent to VND 173,101 per capita per month (VASS, 2006, p. 10). 2. Deprivation is used by Sen to refer to an evaluation of social structures, conditions, and opportunities that exists for any individual or group. Deprivation is not equated with ‘‘deﬁcit’’ approaches to culture. Rather, Sen’s framework examines political, economic, and social structures that do not adequately account for human diversity in its role of achieving or promoting equality. 3. Another critique that goes beyond the focus of this chapter, but is relevant to mention as a caveat, is Sen’s approach is grounded in freedom of choice and elements of democracy. Others (e.g., Unterhalter, 2007a, 2007b; Walker & Unterhalter, 2007) have described Sen’s approach as a rights-based framework, which is a basis for some of the current work on gender equality in the development today. Many countries that are not democratic states, including Vietnam, have utilized rightsbased discourse and practices to frame their development work. 4. A few researchers were of non-Kinh ethnicity, and some could speak the language of the ethnic groups in this study, particularly the Hmong.

ACKNOWLEDGMENTS This chapter is based on a study ‘‘Transition of Ethnic Minority Girls from Primary to Secondary Education’’ conducted by the Ministry of Education and Training in Vi^e t Nam and supported ﬁnancially and technically by _ UNESCO and UNICEF as an activity of the United Nations Girls’ Education Initiative (UNGEI). We offer special thanks to all the national researchers at the Research Centre for Ethnic Minority Education, Ministry of Education and Training in Vi^e t Nam. Throughout the research, we beneﬁted from their devotion _ and commitment to the commissioned study, with preparation, design, study, and analysis.

180

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

The study beneﬁted considerably from the support of Distinguished Teacher, Professor Dang Huynh Mai, former Vice Minister of the Ministry of Education and Training, and Dr. Bui Thi Ngoc Diep, Director of the Research Centre for Ethnic Minority Education, Ministry of Education and Training. We are grateful to Ms. Noala Skinner of UNICEF and Mr. Eisuke Tajima of UNESCO who guided the technical design and implementation of the study. Several education ofﬁcers, Ms. Doan Thi Dung, Mr. Nguyen Anh Ngoc in UNESCO, and Ms. Ngo Kieu Lan, Ms. Le Anh Lan and Ms. Sena Lee in UNICEF also contributed greatly to the technical and practical implementation of the study. Ms. Vibeke Jensen, UNESCO Representative and Mr. Jesper Morch UNICEF Representative provided organizational support to the study. We also thank Suzanne Miric for her contributions to a previous version of this manuscript. Last but not least, we would like to thank all the people who were involved in this study, including girls and boys, parents, teachers, and local authorities who were interviewed and contributed their life experience and viewpoints.

REFERENCES Asian Development Bank. (2002). Country brieﬁng paper: Women in Vietnam. Philippines, Manila: Asian Development Bank Regional Sustainable Development Department and Mekong Department. Baulch, B., Chuyen, T. T. K., Haughton, D., & Haughton, J. (2004). Ethnic minority development in Vietnam: A socioeconomic perspective. In: P. Glewwe, N. Agrawal & D. Dollar (Eds), Economic growth, poverty, and household welfare in Vietnam. Washington, DC: World Bank Regional and Sectoral Studies, World Bank Publications. Berry, J. W. (1984). Cultural relations in plural societies: Alternatives to segregation and their socio-psychological implications. In: M. Brewer & N. Miller (Eds), Groups in contact (pp. 11–27). New York: Academic Press. Berry, J. W. (2003). Conceptual approaches to acculturation. In: K. Chun, P. Balls-Organista & G. Marin (Eds), Acculturation: Advances in theory, measurement and applied research (pp. 17–37). Washington, DC: American Psychological Association Press. Centre for International Economics, Canberra and Sydney. (2002). Vietnam poverty analysis. Canberra: Australian Agency for International Development. Chan, S. (1994). Hmoˆng means free: Life of Hmoˆng in Laos and America. Philadelphia: Temple University Press. Chant, S. (2007). Gender, generation and poverty: Exploring the ‘feminisation of poverty’ in Africa, Asia and Latin America. Cheltenham, UK: Edward Elgar.

Limits of and Possibilities for Equality

181

Chantrill, P., Lendon, S., Sit, H. V., & Thanh, T. T. (2002). Evaluation of basic education project for ethnic minority and other disadvantaged children in Vietnam. Final Report. UNICEF Australia & AusAID. Chapman, D., & Miske, S. (2007). Promoting girls’ education in Africa: Evidence from the ﬁeld. In: M. Maslak (Ed.), The agency and structure of women’s education (pp. 87–106). Albany, NY: SUNY Press. Clemens, M. (2004). The long walk to school: International education goals in historical perspective. Washington, DC: Center for Global Development. DeJaeghere, J. G. (2004). Quality education and gender equality: Background Paper Workshop I UNESCO ICE. (Draft). New York: UNICEF. Desai, J. (2001). Vietnam through the lens of gender: Five years later-results from the second Vietnam living standard survey. Hanoi: Food and Agriculture Organization of the United Nations, Regional Ofﬁce for Asia and the Paciﬁc. Dutcher, N. (2004). Expanding educational opportunity in linguistically diverse societies (http:// www.cal.org/pubs/ford/fordreport_040501.pdf). Washington, DC: Center for Applied Linguistics. Filmer, D. (2000). The structure of social disparities in education: Gender and wealth. Policy Research Working Paper 2268. Washington, DC: World Bank. Foucault, M. (1972). The archaeology of knowledge. New York: Pantheon. Herz, B., & Sperling, G. (2004). What works in girls’ education: Evidence and policies from the developing world. New York: Council on Foreign Relations. Hill Collins, P. (1990). Black feminist theory: Knowledge, consciousness and the politics of empowerment. Cambridge, MA: Unwin Hyman. Hirosato, Y., Costin, W., Emzita, K., Takamiya, K., Tuyen, N., & Ye, Y. (2002). Report and recommendations of the president to the board of directors on a proposed loan to the Socialist Republic of Vietnam for the upper secondary education development project (No. RRP: VIE 28309): Asian Development Bank. Kane, E. (1995). Seeing for yourself: Research handbook for girls’ education in Africa. Washington, DC: World Bank. Kane, E. (2004). Girls’ education in Africa: What do we know about strategies that work? (Africa Region Human Development Working Paper Series No. 73). Washington, DC: World Bank, Africa Region. Kosonen, K. (2004). Language in education: Policy and practice in Vietnam. Hanoi: UNICEF. Lewin, K. (2005). Planning post-primary education: Taking targets to task. International Journal of Educational Development, 25(4), 408–422. Lewis, M. A., & Lockheed, M. E. (2006). Inexcusable absence: Why 60 million girls still aren’t in school and what to do about it. Washington, DC: Center for Global Development. Maslak, M. A. (2005). Re-positioning females in the international educational contexts: Theoretical frameworks, shared policies and future directions. In: D. P. Baker & A. W. Wiseman (Eds), Global trends in educational policy (pp. 145–172). Oxford: Elsevier Ltd. Moser, C. O. N. (1993). Gender planning and development: Theory, practice and training. London: Routledge. Population Council. (2000). Adolescent girls’ livelihoods. New York: Population Council. Rihani, M. A. (2006). Keeping the promise: Five beneﬁts of girls’ secondary education. Washington, DC: Academy for Educational Development (AED). Sen, A. K. (1992). Inequality reexamined. Cambridge, MA: Harvard University Press. Sen, A. K. (1999). Development as freedom. England: Oxford University Press.

182

JOAN DEJAEGHERE AND SHIRLEY J. MISKE

Sen, A. K. (2004). Elements of a theory of human rights. Philosophy and Public Affairs, 32(4), 315–356. State Committee for Ethnic Minority and Mountainous Area Affairs (SCEMMAA). (2005). Socio-economic development programme for extremely difﬁcult communes in ethnic minority and mountainous areas in the period 2006–2010. Third Draft. Hanoi: SCEMMAA. Stromquist, N. (2001). What poverty does to girls’ education: The intersection of class, gender, and ethnicity in Latin America. Compare, 31(1), 39–56. Stromquist, N. (2006). Gender education and the possibility of transformative knowledge. Compare, 36(2), 145–161. Subrahmanian, R. (2002). Gender and education: A review of issues for social policy. Paper No. 9. UNRISD, Social Policy and Development, Geneva. Sutton, M. (2001). Policy research as ethnographic refusal: The case of women’s literacy in Nepal. In: M. Sutton & B. A. U. Levinson (Eds), Policy as practice: Toward a comparative sociocultural analysis of educational policy (pp. 77–99). Westport, CT: Ablex Publishing. Tikal, J. B. G. (2007). Post-elementary education, poverty and development in India. International Journal of Educational Development, 27, 435–445. UNDP. (2002). Localizing MDGs for poverty reduction in Vietnam: Promoting ethnic minority development. Hanoi, Vietnam: UNDP, Poverty Task Force. United Nations Children’s Fund (UNICEF). (2004, November). A gender desk review in education. Hanoi: UNICEF. United Nations Children’s Fund (UNICEF). (2005). Situation of girls’ education in Vietnamadolescence. Available at http://www.unicef.org/vietnam/children_273.html. Retrieved on July 9. United Nations Children’s Fund (UNICEF). (2008). Transition to post-primary education with a special focus on girls: Examining medium-term strategies for developing post-primary education in ESAR. Nairobi: UNICEF East and Southern Africa Regional Ofﬁce. United Nations High Commissioner for Refugees (UNHCR). (2002). Vietnam: Indigenous minority groups in the Central Highlands (WriteNet Paper 05/2001). Geneva: United Nations High Commissioner for Refugees. Unterhalter, E. (2007a). Gender, schooling and global social justice. London: Routledge. Unterhalter, E. (2007b). Gender equality, education and the capability approach. In: M. Walker & E. Unterhalter (Eds), Amartya Sen’s capability approach and social justice in education (pp. 87–108). New York: Palgrave McMillan. U.S. Agency for International Development. (2007). Achieving gender equality in education: Trends in research and USAID programming (1996–2006): A synthesis report. Washington, DC: EQUATE Project, Management Systems International. Van de Walle, D., & Gunewardena, D. (2000). Sources of ethnic inequality in Vietnam. Washington, DC: World Bank Development Research Group Public Economics and Rural Development. Vavrus, F. (2002). Uncoupling the articulation between girls’ education and tradition in Tanzania. Gender and Education, 14(4), 367–389. Vietnam Youth Association, Vietnam Women’s Union, UNICEF and World Bank. (2006). Results of youth consultations (DRAFT). Hanoi: Vietnam Youth Association, Central Committee.

Limits of and Possibilities for Equality

183

Vietnamese Academy of Social Sciences. (2006). Vietnam poverty update report 2006: Poverty and poverty reduction in Vietnam 1993–2004. Hanoi: Vietnamese Academy of Social Sciences. Walker, M. (2006). Towards a capability-based theory of social justice for education policymaking. Journal of Education Policy, 21(2), 163–185. Walker, M. (2007). Selecting capabilities for gender equality in education. In: M. Walker & E. Unterhalter (Eds), Amartya Sen’s capability approach and social justice in education. New York: Palgrave McMillan. Walker, M., & Unterhalter, E. (Eds). (2007). Amartya Sen’s capability approach and social justice in education. New York: Palgrave McMillan. Weis, L., & Fine, M. (2004). Working method: Research and social justice. New York: Routledge. Wils, A., Carrol, B., & Barrow, K. (2005). Educating the world’s children: Patterns of growth and inequality. Washington, DC: Academy for Educational Development, Education Policy and Data Center. World Bank. (2005). Expanding opportunities and building competencies for young people: A new agenda for secondary education. Washington, DC: World Bank. World Bank (n.d.). Assessing the impact of Vietnam’s programs for targeted transfers to the poor using the Vietnam household living standard survey 2002. Hanoi: World Bank. Zhang, Y., Kao, G., & Hannum, E. (2007). Do mothers in rural China practice gender equality in educational aspirations for their children? Comparative Education Review, 51(2), 131–157.

WHAT MATTERS FOR CHINESE GIRLS’ BEHAVIOR AND PERFORMANCE IN SCHOOL: AN INVESTIGATION OF CO-EDUCATIONAL AND SINGLE-SEX SCHOOLING FOR GIRLS IN URBAN CHINA$ Julia Kaufman and Liqun Yin ABSTRACT In this chapter, we utilize qualitative and quantitative data from a yearlong study in four urban Chinese middle schools to investigate the learning environments for girls at these schools; the behavior and performance of girls and boys in these environments; and what factors impact that behavior and performance. This study particularly focuses on socialization through moral education and the examination system as two $

The data collection and analysis for this chapter was supported through dissertation funding from the Chiang Ching-Kuo Foundation for International Scholarly Exchange, as well as funding through a New York University Dean’s Grant.

Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 185–216 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010009

185

186

JULIA KAUFMAN AND LIQUN YIN

sources of authority motivating students’ behavior and performance in school. In the analysis, girls attending three co-educational schools are compared with girls attending one single-sex school, and outcomes for girls are also considered alongside those of boys in the co-educational institutions. Findings indicate that although moral education is particularly emphasized by teachers at the all-girls school, female misbehavior and engagement with teachers is no different for girls attending the singlesex school compared to girls in co-educational schools. Furthermore, differences in outcomes between females and males across schools transcend school-level differences for misbehavior and engagement. However, at the same time, girls at all co-educational schools report higher Chinese and English grades compared to their math and science grades, whereas all-girls school students report no such differences in grades. In regression analysis, socialization variables appear to explain more about students’ misbehavior, whereas the desire to progress to higher levels of schooling explains more about grades and engagement with teachers. That said, socialization variables including moral attitude and attachment to teachers matter more for girls’ math and science grades and their engagement with teachers as compared to boys. This research provides a rare comparative look at education for urban Chinese students and offers new insights about what matters most for girls’ behavior and performance in school.

INTRODUCTION In contrast to low school enrollment among girls in rural and less developed areas of China (Connelly & Zheng, 2003; Hannum, 2003), most girls in the most developed urban areas of China – excluding children of migrant workers – complete their ﬁrst nine years of compulsory education and go on to a high school. Many of these girls also continue on to universities. Additionally, a growing number of experimental and private schools in these urban areas provide diverse options for all students, particularly those who perform well on examinations or come from higher socioeconomic backgrounds. Less clear is whether some urban schools provide better educational outcomes for girls than others and what factors differentially impact outcomes for female and male students. Furthermore, research has not speciﬁcally examined how single-sex schools in these urban areas of China compare with co-educational institutions in terms of what they can provide to students.

What Matters for Chinese Girls’ Behavior and Performance in School

187

In this chapter, we utilize rich qualitative and quantitative data gathered over the course of a year in four junior middle schools in one of the largest urban areas in China to investigate the learning environments they provide to students, the behavior and performance of both girls and boys in these environments, and what factors impact that behavior and performance. We particularly focus on comparisons between girls attending the three co-educational schools and those attending the all-girls school. However, we also examine how girls compare with boys in co-educational institutions. This work has emerged from a larger project examining schooling and the sources of authority in the Chinese context that motivate student work in school. Speciﬁcally, Kaufman (2006) noted both moral education and the examination system as two major sources of authority in China that impact student outcomes. In this chapter, we consider how both moral education and the examination system operate as sources of authority for Chinese students and how they manifest themselves in co-educational versus singlesex environments. Furthermore, we consider how those sources of authority may differentially impact female and male students in China and what that means for student outcomes that include misbehavior, grades, and engagement with teachers in school. This research provides a rare window into what happens inside Chinese schools and classrooms and offers unique insights about what matters most for Chinese girls’ behavior and performance. Although other research in China has provided survey data comparing education for females and males across cities and provinces (Tsui & Rich, 2002; Connelly & Zheng, 2003; Hannum, 2003; Hannum & Adams, 2007; Zhang, Kao, & Hannum, 2007), we know of no other mixed method studies that compare teaching and learning among a small sample of diverse schools in one urban setting. Nor do we know of such research that speciﬁcally compares co-educational and single-sex environments for female students. Furthermore, because China’s urban areas are assumed to be centers for educational excellence and experimentation, our investigation of schools in one of the largest metropolitan areas of China allows us to note outcomes for girls in the best possible circumstances, where schools have basic resources and have the best chance to meet the idealized goals of education policies that emphasize both moral and academic education. By focusing on the sources of authority that motivate students’ work, this chapter moves away from a more narrow psychological consideration of motivation and places primary value on the social and cultural factors that may inﬂuence the work of students, as well as teachers and administrators. Although such consideration of authority and how it motivates students

188

JULIA KAUFMAN AND LIQUN YIN

work has been addressed in American studies (Hirschi, 1969; Arum, 2003; Pace, 2003), no such work has been done in China, where a strong policy emphasis on both socialization through moral education and a high-stakes examination system provides a fascinating context for research on authority. We hope that this research will spur on further studies that consider what sources of authority motivate students’ work in school across countries and cultures, as well as studies that investigate teaching and learning from within Chinese schools.

REVIEW OF RELEVANT LITERATURE Although a diverse body of sociological research considers the nature of authority and its relationship with schooling,1 this scholarship can be separated into two major streams of work with regard to the sources of legitimate authority that motivate students to behave well and work toward higher achievement in school. The ﬁrst stream draws on the work of Emile Durkheim (1961), who wrote about the important role of socialization, moral education, and discipline for integrating a child into society and thus encouraging them to embrace the moral values of that society. Sociologists of education have taken Durkheim’s theories on moral education and internalization of societal norms a step further to explore the relationship between social integration and a wide range of conventional positive behaviors internalized by successful students (Hirschi, 1969; Coleman & Hoffer, 1987; Zhou & Bankston, 1999; Arum, 2003). In this school of thought, students who feel a stronger connection to their school and their teachers and have embraced societal values will be more likely to do well in school. In contrast, a second stream of work focuses on the legitimate authority coming from the system itself and the individuals within that system. Instead of seeing authority as a set of social norms and rules that are internalized through socialization and integration into society, authority – in this perspective – emanates from people, organizations, and institutions that exert control on individuals. Much of this work can be traced to Max Weber’s (1958, 1978) original deﬁnitions of legitimate authority as traditional (authority based on established beliefs about the legitimacy of the ruler); charismatic (authority based on individual leaders that inspire strong emotional attachment and commitment); and bureaucratic (authority based on a system that exerts control through development ‘‘rational’’ rules

What Matters for Chinese Girls’ Behavior and Performance in School

189

and policies). Work at the individual level has expanded on Weber’s deﬁnitions to include ‘‘professional’’ authority or expertise that is based on an individual’s knowledge and skills, which Pace (2003) has argued is most important for considering the authority of teachers in schools. On the other end of the spectrum, research focused on the role of bureaucratic authority in schooling has considered how capitalist interests, ‘‘credentialing’’ systems, and examinations hold some degree of control over schools, students, and workers within a society (Weber, 1958; Bowles & Gintis, 1976; Collins, 1979; Eckstein & Noah, 1993; Brown, 1995). Much research has documented the hard work and discipline of Chinese students (Kessen, 1975; Unger, 1982; Tobin, Wu, & Davidson, 1991; Kipnis, 2001; Andreas, 2004). Those studies generally attribute students’ effort to examination system that pushes them to succeed or a deeply felt imperative to do well in school as internalized through moral education. However, no empirical work in China has addressed the inﬂuence of both the examination system and socialization as sources of authority inﬂuencing students’ behavior and performance in school. Gender is a particularly interesting and little studied factor connected with authority in schools. In their study of students and teachers in Australian schools, Kessler et al. (1985) have posited that the education system, as well as the academic curriculum, constructs gender for students and ‘‘actively produces women’s subordination’’ (pp. 45–46). Sociologists of education in the United States have further argued that female students are subject to more controls than males and, as an effect, misbehave less in school (Jensen & Eve, 1976; Hagan, Gillis, & Simpson, 1985; Singer & Levine, 1988). This work thus suggests that the education system both exerts authority on female students more than male students and also socializes females to behave in particular ways. However, research in China has not focused on how authority and control may inﬂuence female and male students in different ways. Beyond studies on how schooling impacts girls’ behavior and attitudes, studies have documented that single-sex schooling provides a more open and encouraging atmosphere for girls to focus on their studies and obtain higher levels of achievement, particularly with regard to girls in developing countries (Jimenez & Lockheed, 1989; Lee & Lockheed, 1990; Streitmatter, 1999). However, other research has found few or little advantages to single-sex schooling in developed countries (Marsh, 1989; Harker, 2000; Gilson, 2002). Furthermore, any studies on single-sex schooling encounter difﬁculty in knowing whether – in Rosemary Salomone’s (2003) words – ‘‘any differences in outcomes are the direct result of gender organization or

190

JULIA KAUFMAN AND LIQUN YIN

merely the effect of background differences between student populations’’ (p. 9). In China, single-sex education is advocated by Chinese educators as one path toward higher educational outcomes for female students in China, particularly with regard to math and science performance (He, 2001; Li, 2001; Yuan, 2001). However, these same educators talk less about math and science instruction at these schools and instead emphasize a curriculum ‘‘specially designed’’ for girls, which focus on languages, humanities, and moral education. He (2001), for example, writes that in the single-sex high school in Shanghai where she is a principal, ‘‘efforts are made to discover and exploit the traditional virtues of Chinese women and moral education resources of [the school].’’ Thus, although these educators write that stereotypes limit girls’ ‘‘creative potential’’ (Li, 2001), the single-sex education that they support appears to sometimes re-emphasize these stereotypes by focusing on the subjects where girls are traditionally believed to excel. In this chapter, we compare experiences and outcomes for girls and boys across four schools in a major urban area of China, including an all-girls school, and examine how particular sources of authority may impact those outcomes. This work makes a much-needed contribution to our knowledge about how different sources of authority inﬂuence the behavior and performance of Chinese students, and it especially sheds light on outcomes for girls within diverse school environments in urban China.

DATA AND METHODS This research is based on qualitative and quantitative data collected over the course of one year in four junior middle schools2 within the same large urban city in eastern China. These four schools, selected because of their diversity, can be described as follows: Number Two School,3 which is described as an ‘‘ordinary’’ or ‘‘average’’ public school by its teachers and students; the 1,000 attending students mostly live in the neighborhood around the school; International School, a relatively elite private ‘‘international’’ junior middle school that emphasizes English-language learning and operates through tuition fees paid by parents of the 1,200 students who manage to gain admission to this relatively elite school;

What Matters for Chinese Girls’ Behavior and Performance in School

191

Si Nan School, an ‘‘experimental’’ junior middle school that features advanced English instruction for a small group of 200 or so students who are talented in English and regular instruction for the larger population of about 1,000 more average students from the surrounding neighborhood who attend the school; and Girls School, a reputable all-girls junior middle school attended by approximately 1,100 girls who come from the neighborhood where the school is located and other areas throughout the city. This sample of schools offers a unique opportunity to compare girls’ experiences and outcomes in various learning environments. Qualitative data for this chapter comes from semi-structured interviews with one or two administrators, ﬁve teachers, and ten eighth-grade students at each of the four schools, as well as class observations conducted for at least 20 hours at each school. Interviews with teachers and administrators included questions about the reputation of their school, school goals and teachers’ goals for students, the qualities of good and bad students, moral education, academic learning, and school discipline. Interviews with students contained similar questions to those in teacher interviews, including questions about their opinion of their school, the qualities of good and bad students, and the discipline they receive in school. In addition, students were asked to talk about their grades, their aspirations, and their opinions about what it means to ‘‘learn’’ at school. Classroom observations took into account whether the lesson focused on the nature of the tasks that students’ were asked to do, the quality of talk among teachers and students, and misbehavior and discipline. Quantitative data was gathered through a survey administered to 800 eighth-grade students, including 200 students at each of the four schools.4 To choose which eighth-graders would be asked to complete the survey, we randomly selected classrooms (not students) where the surveys would be administered using a list of all eighth-grade homeroom classrooms supplied by each school. The focus on eighth-grade students is intended to provide more comparable data across a single grade level. The survey included questions about students’ behavior in school, their grades, their school performance, their moral attitude, their attachment to their teachers and school, their expectations for their future education, and their perceptions about their parents’ expectations for their education. Many of these questions were Mandarin translations of questions taken from largescale survey instruments including the National Longitudinal Study of 1988 and the Third International Mathematics and Science Study in 1999.

192

JULIA KAUFMAN AND LIQUN YIN

In addition, the surveys asked students to provide their sex and their family background, including each parent’s level of education and occupation.5 Appendix A in this chapter provides deﬁnitions for all survey variables used in this analysis, as well as Cronbach alphas for any composites. Through the inclusion of both qualitative and quantitative data in our analysis, we are able to offer a multi-layered depiction of teaching and learning for female and male students. Our qualitative analysis is intended to provide comparative evidence of the diverse learning environments for students in urban environments and provide a potential source of triangulation (Tashakkori & Teddlie, 1998) for data about the sources of authority that motivate students in school. In our qualitative interview analysis, codes for data initially emerged from multiple readings of interviews (Strauss & Corbin, 1998). As those themes emerged, including evidence about the goals for schooling and discipline in school, those themes were elaborated through additional codes that allowed for more detailed reporting of the data (Lincoln & Guba, 1985; Strauss & Corbin, 1998). Additionally, the qualitative analysis guided the themes and focus for our quantitative analysis. In the ﬁndings that follow, we ﬁrst map the diverse learning environments within the four schools in our study, using quantitative data on the socioeconomic disparities among schools and qualitative data on school goals and disciplinary practices. Then, we move to a more quantitative examination of student outcomes, including misbehavior, grades, and engagement with teachers, among female and male students in the schools. Following this descriptive data, we provide both correlations among the variables and multiple linear regressions that explore what sources of authority inﬂuence student outcomes, using the dependent variables of misbehavior, math and science grades, and engagement with teachers. In our regression analysis, we particularly emphasize how the independent variables of attachment to teachers, school, and moral attitudes impact outcomes alongside independent variables measuring desire for the student to be promoted to progressively higher levels of education. We use these independent variables as proxies to indicate the sources of authority that motivate student outcomes; they are intended to reﬂect students’ integration into school and society as one kind of authority versus students’ focus on getting through the bureaucratic examination system as another. Controls for school, students’ family background, and sex are also included in the regressions. Furthermore, interaction variables enable us to consider how being female and attending a single-sex school work alongside other variables to produce particular student outcomes.

What Matters for Chinese Girls’ Behavior and Performance in School

193

FINDINGS Co-Educational and Single-Sex School Environments for Girls in Urban China Socioeconomic Disparities among Schools Despite Ministry of Education policies in the 1980s that eradicated ‘‘key,’’ more reputable primary and junior middle schools, the background of students across the four schools where research was conducted – as well as the varying reputations of the schools themselves – illustrate the presence of diverse resources and opportunities among Chinese urban junior middle schools today. For example, according to our survey data, the percentage of students with at least one parent in a professional occupation is 54% at Si Nan School, 68% at Number Two School, 80% at Girls School, and 92% at International School (ANOVA, po.0001). Additionally, the percentage of students with at least one parent who is a university graduate is 23% at Si Nan, 49% at Number Two, 71% at Girls School, and 85% at International School (ANOVA, po.0001). The percentage of students at each school who enter high school should also be acknowledged, as that percentage is often tied to a school’s reputation. In this respect, Si Nan and Number Two might be regarded as less reputable because of the 50% and 60%, respectively, of their students who enter high school. However, Girls School reports 90% of its students attend high school, whereas 85% of International School’s students attend high school.6 The large socioeconomic differences among schools underscore that Girls School and International School are able to be more selective than Si Nan and Number Two. A higher number of students apply to both Girls and International School because of their good reputations, and the students with better socioeconomic resources – who receive more outside tutoring and preparation – are generally more desirable candidates for acceptance. On the other hand, Si Nan and Number Two School are largely attended by students in the surrounding neighborhoods of those schools, and these students have lower socioeconomic resources and fewer choices as to which school they can attend. Goals for Schooling When asked about the goals of their schools, administrators’ and teachers’ responses across all schools unsurprisingly reﬂected China’s Quality Education (suzhi jiaoyu) policy, in that they spoke about the ‘‘all around development’’ of their students that encompasses the three pillars of Quality Education: academic, moral, and physical education. However, at least

194

JULIA KAUFMAN AND LIQUN YIN

some teachers and administrators at every school outside of Girls School said that ‘‘high marks’’ and ‘‘examinations’’ are the most important thing for students. For example, in response to a question about his school’s goals, a Number Two School teacher explained, ‘‘For students in China and their parents, the greatest dream is for their child to go to a famous university. This makes the examination the most important thing.’’ Similarly, a Si Nan teacher noted, ‘‘The school focuses on passing rates, for example the passing examination rate. Most of the emphasis is on this. They think that all the students should enter a good high school.’’ Some educators even spoke about the conﬂicts between examination preparation and their own ideals, like one administrator at International School who said that he believed students’ moral ability was most important; however, he conceded that ‘‘at the same time, they have to be the best in the whole city [in regard to exam scores].’’ In contrast to the teachers’ responses at the co-educational schools that reﬂected the importance of academic goals, all ﬁve interviewed teachers at Girls School unanimously emphasized the importance of moral goals and those teachers did not acknowledge any potential conﬂict between those goals and students’ examination preparation. One English teacher, for example, said that ‘‘care’’ was the most important quality for a student to have. Another English teacher even spoke of valuing students’ moral abilities over their academic ones Actually, you know, not all the students the teachers like are the very good students. I mean, those students’ scores may not be very high, but actually I think they are lovely. They can help others. They are warm-hearted. And they always think about others and are considerate. I think these students are good students.

Several Girls School teachers also spoke about the school’s focus on ‘‘IACE,’’ an English-letter acronym celebrating the school’s commitment to English language learning. IACE stands for Independence, Ability, Care, and Elegance. These objectives are speciﬁcally geared toward females and indicate some ideals for female learning that are not present for males in China. Several Girls School teachers speciﬁcally said that their school provides an education especially well suited for girls. One of these teachers emphasized the importance of English language learning at the school and talked about how girls excel at such language learning. However, Girls School teachers did not talk about the girls having less aptitude or doing less well in other subjects like math and science. Additionally, Girls’ School teachers also emphasized the development the girls’ leadership qualities and

What Matters for Chinese Girls’ Behavior and Performance in School

195

self-conﬁdence, which was not something discussed by teachers in other schools. One teacher, for instance, said that girls are ‘‘brave enough’’ to speak up in class and be leaders because of the absence of boys at the school. Another teacher commented, ‘‘The girls in this school are braver and more open-minded than the girls in other schools. They can say whatever they want to say and do whatever they want to do in this school.’’ In interviews, teachers across schools were also asked to describe the qualities of a ‘‘good student.’’ As with their focus on moral goals of their schools, all Girls School teachers focused on the moral qualities of good students, including ‘‘care,’’ ‘‘honesty,’’ and ‘‘helpfulness.’’ Four of these Girls School teachers only discussed moral qualities without bringing up academic qualities of a good student. A representative example from a Girls School teacher states ‘‘I think the best quality [is to be] warmhearted. Always. And they will be kind to others and always ask, ‘What can I do for others? When I open the door, can I just let my friends or classmates go in ﬁrst?’ So this is the best quality I think.’’ Another Girls School teacher described good students’ best qualities as ‘‘honest, and kindness, and enthusiasm, and optimism.’’ Although teachers at co-educational schools also described these moral qualities as part of being a good student, those teachers talked about such moral qualities in conjunction with the importance of doing well academically and getting high marks in school examinations. A Number Two teacher provides an example Good students should be healthy in their mind, psychologically. They should be able to take care of others and their parents and be respectful to teachers and helpful to other students. And the good students should be willing to improve themselves in their study and moral education. They should not only focus their attention on the textbook or what they are learning in the classroom, but they should also actively seek more knowledge outside of the classroom. They should try to develop their overall qualities.

Some of the teachers at co-educational schools did not even bring up moral qualities and instead focused entirely on the importance of a good student having academic abilities. Interestingly, although teachers at Girls School appeared to value moral qualities and goals more than academic ones, Girls School students did not cite moral qualities of good students more often than their peers at other schools. Instead, those girls appeared to value academic goals as much as other girls across all schools. A Girls School student’s description of a good student provides an example, which states, ‘‘Well, in China, students must be smart, clever, and have great marks. And they have to listen to the teacher carefully and be honest.’’ As with other students across schools, this

196

JULIA KAUFMAN AND LIQUN YIN

student brought up academic qualities of a good student (‘‘smart, clever, have great marks’’) alongside moral qualities (‘‘be honest’’). Five students, including one from Girls School, spoke only about the academic qualities of good students and did not bring up moral qualities. Additionally, despite teachers’ emphasis on moral goals of schooling, Girls School classes did not seem to focus any less on test preparation than other schools’ classes. That is, about 25% of classes observed at each school involved preparing for tests, reviewing past tests, or practicing exercises that included multiple choice questions and ﬁll-in-the-blank work that reﬂected the kind of questions students would likely see on year-end examinations, as well as their high school entrance exam. Discipline in Schools Teachers were not often observed meting out discipline during class. However, as discussed in the next section, student misbehavior was also relatively rare during classroom observations, perhaps because the presence of a foreign researcher discouraged such misbehavior. Nevertheless, when students did engage in minor misbehaviors like talking in class or passing notes, teachers were observed reprimanding them in very subtle ways or not reprimanding them at all. For example, during 40 minutes of a Girls School class, the teacher ignored the almost continuous, occasionally loud chattering of students during her instruction. Only toward the end of the class, when the girls got very loud, did the teacher mildly said, ‘‘Pay attention.’’ Teachers at other schools might offer more quick reprimands for more minor infractions during class, but those teachers also occasionally ignored misbehavior as well. In one such example in an English class at Number Two School, a male student in the back of the room was observed talking to students around him during the teacher’s instruction, paging through a book of his own, and walking around the classroom talking to students during small-group work time. The teacher of the class consistently ignored this student and did not give the student any reprimands during or after class. This situation of a male student misbehaving in the back of a classroom occurred a few times during classroom observations at both Number Two and Si Nan. When asked about methods for disciplining students, teachers most commonly spoke about ‘‘educating’’ the student (jiaoyu) as a ﬁrst measure to discourage future misbehavior. According to these teachers, ‘‘educating’’ generally occurs after or before class, which might explain why teachers reprimanded students less during class. When such ‘‘education’’ was observed in the teachers’ room or hallway, it usually involved a teacher

What Matters for Chinese Girls’ Behavior and Performance in School

197

giving a stern lecture that required little student input beyond a few words of assent that she/he will not engage in such misbehavior again. In the words of one teacher ‘‘First teachers need to talk to the students and tell them it is wrong. That will guide them not to do that [misbehavior] any longer.’’ Teachers and administrators generally agreed that if one such lecture did not correct a student’s misbehavior, then further ‘‘education’’ from teachers and administrators would. Compared to teachers at co-educational schools, teachers at Girls School talked about providing a more open and less harsh disciplinary environment for students. One Girls School teacher, for example, said the way teachers talk to the students and the arrangement [here] is different from the other schools. Because, you know, there are all girls here. So when the teachers are talking with the students . . . they try to be warmer and softer than they would with boys.

Two other Girls School teachers also talked about not imposing too strict discipline on the students. One of those teachers talked about the school giving more freedom to both teachers and students as compared to other schools, whereas the other teacher said that she treats students less as a teacher than a ‘‘sister’’ because she will hold discussions with them rather than impose one correct answer to a question. Girls School students were almost never observed receiving reprimands either during class or lectures in between classes. However, as noted earlier, students at other schools also were rarely observed receiving discipline as well. Thus, observed discipline was not appreciably different across schools. Nevertheless, a more representative sample of student misbehavior and discipline might be captured in the next section, which compares survey ﬁndings on student outcomes including misbehavior, grades, and engagement with teachers across schools.

Comparing Student Outcomes Misbehavior and Discipline Although students were not observed committing any major misbehavior during classes, both teachers and students talked during interviews about more serious behavior infractions that had occurred within the school, including vandalism and ﬁghting. But aside from these rare incidences, teachers and students generally indicated low degree of misbehavior. In between classes, students across schools would exhibit more animation and engage in more horseplay than they did during class. As compared to their

198

JULIA KAUFMAN AND LIQUN YIN

female peers at other schools, Girls School students appeared particularly rambunctious during break times, running about the campus of the school and yelling to get each others’ attention. However, teachers across schools appeared to completely sanction students’ more rowdy activity in between classes as time for them to relax and expend some energy before they would have to sit down and quietly pay attention to teachers’ lectures. Surveys provide a broader picture of students’ misbehavior. In those surveys, students were asked to report both the estimated frequency of their misbehavior (including skipping school, cheating, and ﬁghting) and the estimated frequency of the discipline they received (including being sent to the ofﬁce and their parents receiving warnings). Averages for misbehavior and discipline are reported separately. In Fig. 1, means for frequency of misbehavior indicate that males reported more misbehavior than females 1.4

Frequency of Misbehavior

1.2 1.0 0.8 0.6 0.4 0.2 0.0 Male

Female

Number 2

Male

Female

Si Nan**

Female Girls

Male

Female ++

International

++ Significant difference between females at International School compared to females at Si Nan and Girls schools (ANOVA, Tukey HSD, p <.01) ** Significant school-level differences between males and females (independent t-test, p <.01)

Fig. 1. Means for Students’ Self-Reports of Misbehavior (Female n ¼ 545; Male n ¼ 214). Note: The Means for Misbehavior are a Sum of Frequency Skipping School and Cheating on a Test (Estimated Frequency during the Past Month) and Getting into a Physical Fight (Estimated Frequency during the Past Semester).

199

What Matters for Chinese Girls’ Behavior and Performance in School

overall, signiﬁcantly so in Si Nan School (po.01). Additionally, although there were no signiﬁcant differences in misbehavior among males across schools, females at International School misbehaved more than females at Si Nan and Girls School (po.01). Fig. 2 tells a slightly different story with regard to school discipline. As with reports of misbehavior, female means were lower than male means for reports of discipline; these differences between females and males were signiﬁcant at Number Two (po.001) and International School (po.05). Although no signiﬁcant differences emerged among females across schools, males at Number Two reported receiving more discipline than those at other schools, with a signiﬁcant difference between males at Number Two and

1.8 1.6

Frequency of Discipline

1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 Male+

Female

Number 2***

Male

Female Si Nan

Female Girls

Male

Female

International*

+ Significant difference between males at Number Two compared to males at International School (ANOVA, Tukey HSD, p <.05) *** Significant school-level differences between males and females (independent t-test, p <.001) * Significant school-level differences between males and females (independent t-test, p <.05)

Fig. 2. Means for Students’ Self-Reports of Discipline (Female n ¼ 539; Male n ¼ 215). Note: Discipline is a Sum for Frequency of Being Sent to the Ofﬁce for Misbehavior, Being Sent to the Ofﬁce for Schoolwork Problems, Parents Receiving Warning for Student’s Misbehavior, and Parents Receiving Warning for Student’s Attendance. Students Reported Frequencies for the Past Semester Only.

200

JULIA KAUFMAN AND LIQUN YIN

International School (po.05). Additionally, although teachers at Girls School indicated that they may discipline students less than teachers at other schools, no signiﬁcant differences emerged when comparing reports of discipline between females at Girls School and females at other schools. In looking at these means for discipline, one should take into account that schools may have different standards for discipline (e.g., different criteria for sending a student to the teacher’s ofﬁce or sending a warning to students’ parents). Thus, frequency of discipline in Fig. 2 may not be as clear an indicator for misbehavior as students’ reports for frequency of actual misbehavior in Fig. 1.

Grades Students’ estimates of their overall grades in each subject provide even more interesting contrasts between girls, as well as boys, across schools. Fig. 3 presents standardized averages for students’ reports of their grades in math and science compared to their grades in Chinese and English. As can be seen in Fig. 3, male students in every co-educational school reported higher math and science grades compared to English and Chinese grades, whereas 0.15 Math/Science

0.13

Chinese/English

0.11 0.09

Standardized Grade

0.07 0.05 0.03 0.01 -0.01 -0.03 -0.05 -0.07 -0.09 -0.11 -0.13 -0.15

Male** Female** Numb er 2

Male Si Nan

Female

Female only Girl

Male* Female International

** Significant difference between Chinese/English and Math/Science grades (paired t-test, p <.01) * Significant difference between Chinese/English and Math/Science grades (paired t-test, p <.05)

Fig. 3.

Means for Standardized Math/Science Grades Compared to Chinese/ English Grades (Female n ¼ 529; Male n ¼ 207).

201

What Matters for Chinese Girls’ Behavior and Performance in School

females reported higher English and Chinese grades compared to math and science grades (differences between English/Chinese and math/science grades are signiﬁcant for males in Number Two and International School and females at Number Two). An important exception to this rule of females’ higher English/Chinese grades compared to math/science grades is Girls School, where students reported about the same grades for Math/Science as Chinese/English. Potential reasons for this lack of a difference between grades of different subjects at Girls School are offered in the discussion and conclusions section following these ﬁndings. As with discipline, one should keep in mind that grades themselves vary a great deal across schools (ANOVA, po.0001). Thus, in making any grade comparisons, one should only take into account within-school differences and not differences across schools, as schools likely have different grading methods and criteria. Engagement with Teachers Fig. 4 compares females and males for the average number of times during the past month that they asked the teacher a question and engaged the 3.5

Average Number of Times

3.0 2.5 2.0 1.5 1.0 0.5 0.0 Male

Female

Number 2

Male

Female Female only

Si Nan**

Girls

Male

Female*

International

** Significant difference between male and female responses at Si Nan (t-test, p <.01) * Significant difference between females at International School and other schools (ANOVA, p <.05)

Fig. 4. Average Number of Times Students Asked the Teacher a Question and Engaged the Teacher in Debate in the Past Month (Female n ¼ 542; Male n ¼ 217).

202

JULIA KAUFMAN AND LIQUN YIN

teacher in debate.7 This outcome is included in our analysis as a possible indication of students’ reﬂection and critical thinking during class. As can be noted in Fig. 4, females engaged the teachers with questions or debate less than males at Number Two and – signiﬁcantly so – at Si Nan. However, at International School, females appeared to slightly outpace males with regard to engagement with teachers. Additionally, those females at International School engaged their teachers signiﬁcantly more than their female peers at all other schools (po.05 in Tukey HSD comparisons). No signiﬁcant differences emerge among males in cross-school comparisons. As can also be noted in Fig. 4, the average reported engagement with the teacher is fairly low across all schools, sometimes less than twice a month. These reports are on par with classroom observations, where students rarely participated in classroom discussions unless prompted by the teacher to do so. Some of these prompts involved students taking their turn to give answers to multiple-choice or ﬁll-in-the-blank worksheets they had completed for homework, although students were also sometimes asked by teachers to solve problems on the board (in mathematics classes), work with one another to practice dialogues that might later be performed in front of the class (in English class), or work together to complete worksheets or a set of questions. However, students were observed asking the teacher questions in only a handful of these situations.

Sources of Authority Impacting Student Outcomes In this ﬁnal section, we consider the sources of authority, alongside other factors, that predict three outcomes that were described in the previous section: misbehavior, grades, and teacher engagement. To make these predictions, we present multiple linear regressions on those outcomes. For the dependent variable of misbehavior, we focus solely on students’ reports of their misbehavior and do not include students’ reports of the discipline they receive, which may vary based on school and teacher standards for bad behavior and may not reﬂect actual misbehavior. For the dependent variable of grades, we include only math and science grades, rather than average total grades, because an analysis of math and science grades is more relevant to concerns about co-educational females’ lower math and science achievement relative to their English and Chinese achievement. In our regressions, we focus on two major sets of independent variables that are used as proxies to reﬂect two separate sources of authority that might inﬂuence outcomes. The ﬁrst set of variables in Model 1 of

What Matters for Chinese Girls’ Behavior and Performance in School

203

each regression examines how attachment to teachers and school, as well as moral attitude, inﬂuences outcomes. Use of these variables is meant to test the hypothesis that students who feel a stronger connection and appreciation to school and teachers, and respond in ways that reﬂect a more moral attitude, are less likely to misbehave and more likely to do well in school because those students are better socialized into the school and have internalized positive values about school. The hypothesis reﬂects Durkheimian theories and subsequent sociological research that emphasizes the importance of socialization and social support for students to do well in school. The second set of variables added in Model 2 of each regression examines how students’ expectations about progressing to higher levels of schooling, and students’ perceptions of their parents’ expectations for them to progress to higher levels of schooling, inﬂuence outcomes. These variables are meant to test the hypothesis that the education system itself is motivating students to behave well and do well in school, with the assumption that students who want to get ahead in school (or have parents who want them to get ahead in the system) are motivated by the education system and the examinations that determine students’ future within the system. The hypothesis reﬂects sociological research on the authority of credentialing and examination systems over student outcomes. In addition to the sets of variables measuring different sources of authority that motivate students’ behavior and performance in school, we have included independent variables to consider the impact of being a female at a co-educational school versus Girls’ School, as well as variables representing students’ attendance at Number Two and International School (Si Nan School is used as the reference for the school variables). Additional independent variables are included to control for parents’ education and whether either parent is in a professional occupation.8 Finally, interactions between all independent variables with being female, female at a coeducational school, and female at Girls School were tested for their impact on student outcomes. When any of those interactions were signiﬁcant, regressions include a Model 4 with those signiﬁcant coefﬁcients for the interactions alongside all other independent variables. As mentioned previously, Appendix A contains deﬁnitions for all the variables used the regression models; Appendix B in this chapter includes means and standard deviations for these variables as well. The correlation Table 1 indicates that several independent variables used the regressions are correlated with one another. Particularly, the socialization variables are highly correlated with one another, as are the expectation

204

Table 1.

Correlations for Independent Variables Used in Regressions. 1

3

4

5

6

7

8

9

10

11

Moral attitude 1.00 Teacher attachment .32** 1.00 Liking school .26** .46** 1.00 Students’ education expectations .07 .10** .16** 1.00 Parents’ education expectations .07 .05 .11** .62* 1.00 Co-education females .09** .03 .10** .09* .14** 1.00 Girls School females .00 .02 .06 .04 .03 .54** 1.00 Number Two student .08* .17** .19** .06 .03 .08* .32** 1.00 International student .03 .03 .17** .24* .21** .13** .33** .33** 1.00 Si Nan student .10** .11** .03 .22 .21** .32** .34** .34** .35** 1.00 Parents’ education .01 .05 .15** .36** .32** .13** .18** .11** .31** .37** 1.00 Professional parents .02 .02 .07 .22** .20** .08* .11** .10** .27** .28** .41***

*po.05; **po.01; ***po.001.

JULIA KAUFMAN AND LIQUN YIN

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

2

205

What Matters for Chinese Girls’ Behavior and Performance in School

variables correlated with one another. However, the set of socialization variables is not so correlated with the set of expectation variables, and any relationship between these sets will also be captured by the inclusion of multiple models within the regression. The regressions will also shed more light on the strength of various other signiﬁcant correlations that have been observed among variables. Factors Inﬂuencing Misbehavior In the ﬁrst regression Table 2, all the independent variables taken together explain just 14% of the variance in students’ misbehavior. As can be noted in Table 2, higher teacher attachment is the most signiﬁcant factor negatively inﬂuencing misbehavior. Thus, students who reported more attachment to their teachers tended to report lower misbehavior. Furthermore, higher moral attitude also has a negative impact on misbehavior. The signiﬁcance for teacher attachment and moral attitude suggest that some socialization and integration into a school limits misbehavior. Additionally, as suggested in the prior descriptive comparisons of female and male misbehavior, being female – both in the co-education and single-sex settings – is also a large predictor of misbehavior. Finally, students with parents who have attained a higher level of education reported signiﬁcantly less misbehavior, whereas – by contrast – students with professional parents reported signiﬁcantly more misbehavior. No interactions were signiﬁcant among the independent Table 2.

Unstandardized Betas and (Standard Errors) for Linear Regressions on Misbehavior.

Moral attitude Teacher attachment Liking school Student education expectation Parents’ education expectations Co-ed females Girls School females Number Two School International School Parents’ education Professional parents r2 *po.05; **po.01; ***po.001.

Model 1

Model 2

Model 3

.11** (.04) .11*** (.03) .15 (.12)

.11** (.04) .11*** (.03) .16 (.12) .06 (.10) .01 (.09)

.09* (.04) .12*** (.03) .18 (.12) .08 (.11) .04 (.09) .50** (.15) .49* (.22) .12 (.17) .32 (.19) .14** (.05) .32* (.16) .14

.08

.08

206

JULIA KAUFMAN AND LIQUN YIN

variables and being female or being at Girls School. Thus, those interactions are not reported in the full model.

Factors Inﬂuencing Math and Science Grades The next set of regressions focuses on what impacts students’ average selfreported grades in math and science. The moral attitude and teacher attachment variables that had some impact on misbehavior have little to no effect on math and science grades (this is also true of the impact of moral attitude and teacher attachment on all grades). Yet, liking school is signiﬁcantly and positively connected with math and science grades. Nonetheless, students’ expectations for themselves – and students’ reports of their parents’ expectations – are one of the most signiﬁcant factors inﬂuencing reports of higher math and science grades (Table 3). Unsurprisingly, the school variables also have an impact on math and science grades, as students at some schools reported generally higher grades than students at other schools. Students of parents in professional occupations are also more likely to have higher math and science grades. Additionally, although teacher attachment is not a signiﬁcant predictor for math and science grades among all students, it predicts signiﬁcantly more Table 3.

Unstandardized Betas and (Standard Errors) for Linear Regressions on Math and Science Achievement.

Moral attitude Teacher attachment Liking school Student education expectation Parents’ education expectations Co-ed females Girls School females Number Two School International School Parents’ education Professional parents Female Teacher attachment Girls school Expectations r2 *po.05; **po.01; ***po.001.

Model 1

Model 2

Model 3

Model 4

.02 (.03) .02 (.02) .23* (.09)

.03 (.03) .02 (.02) .15 (.09) .45*** (.07) .13* (.07)

.02 (.03) .01 (.02) .26** (.08) .52*** (.07) .15* (.06) .20 (.10) .80*** (.15) .08 (.12) .91*** (.13) .05 (.08) .23* (.11)

.03

.18

.02 (.03) .07 (.04) .25** (.08) .62*** (.08) .31* (.06) 1.51** (.51) 1.14 (.68) .15 (.12) .97*** (.13) .04 (.08) .22* (.11) .11** (.04) .29* (.13) .31

.29

207

What Matters for Chinese Girls’ Behavior and Performance in School

Table 4.

Unstandardized Betas and (Standard Errors) for Linear Regressions on Engagement with Teacher.

Moral attitude Teacher attachment Liking school Student education expectation Parents’ education expectations Co-ed females Girls School females Number Two School International School Parents’ education Professional parents Female Moral attitude r2

Model 1

Model 2

Model 3

Model 4

.02 (.07) .07 (.06) .49* (.22)

.01 (.07) .07 (.06) .39 (.22) .67*** (.19) .06 (.17)

.01 (.07) .05 (.06) .32 (.22) .62** (.19) .01 (.17) .56* (.27) .91* (.40) .36 (.32) .04 (.35) .09 (.21) .27 (.28)

.25* (.11) .04 (.06) .35 (.22) .64** (.19) .03 (.17) 4.35*** (1.24) 4.61*** (1.24) .32 (.31) .05 (.35) .09 (.21) .19 (.28) .42** (.13) .10

.02

.07

.08

*po.05; **po.01; ***po.001.

about female achievement than male achievement. Higher expectations at Girls School also have a slightly negative effect on math and science grades. Factors Inﬂuencing Engagement with Teachers As with misbehavior, being female – both at co-education and single-sex schools – is a predictor of lower teacher engagement. However, unlike misbehavior, students’ education expectations are a high predictor of teacher engagement. Additionally, although moral attitude is a negative predictor for teacher engagement among students, moral attitude has a positive inﬂuence for female engagement. Thus, as with the ﬁnding that higher teacher attachment inﬂuences females’ math and science grades more so than males, higher moral attitude is connected to females’ engagement with teachers more so than males’ engagement with teachers. Unlike with the other variables, family background does not predict students’ engagement with their teachers in any signiﬁcant way (Table 4).

DISCUSSION AND CONCLUSIONS This chapter investigates how diverse schools in urban China educate female and male students, particularly focusing on comparisons among one singlesex school for girls and three co-educational schools. To that end, we have

208

JULIA KAUFMAN AND LIQUN YIN

presented qualitative and quantitative data with regard to the schools’ learning environments, outcomes for females and males in cross-school comparisons, and the sources of authority that impact those outcomes. Qualitative data from teachers indicates that moral education is emphasized more by teachers in the single-sex school environment compared to the coeducational schools, both with regard to teachers’ descriptions of school goals and what it means to be a good student. This ﬁnding suggests that moral education may have more force in terms of inﬂuencing behavior and work in the single-sex school compared to other schools. Additionally, Girls School teachers spoke of disciplining students less harshly than teachers at other schools. However, our qualitative data from classroom observations and interviews with students themselves did not demonstrate that moral education is a more important motivator of Girls School student behavior and performance compared with students in other schools. Nor did our regression analysis based upon students’ survey data demonstrate that moral education or any of our other ‘‘socialization’’ variables motivate outcomes for Girls School students in any way that is signiﬁcantly different from females or other students in co-educational schools. Thus, although Girls School teachers who we interviewed may value moral education more than teachers at other schools, girls’ behavior and performance in school is not necessarily connected with that moral education, at least as measured by the variables included in our analysis. That said, the differences between female and male students across schools are signiﬁcant, both in t-tests comparing female and male outcomes within schools and regression analysis examining the independent variables that inﬂuence students’ behavior and performance. Across schools, reports of misbehavior and discipline in school are lower for females than males, and being a girl at either a co-educational or single-sex school was a very signiﬁcant factor inﬂuencing misbehavior in the regression analysis (po.001). Similarly, reports of engagement with teachers are generally lower for females than males, with being female at a co-education or singlesex school again as a signiﬁcant factor inﬂuencing teacher engagement in the regression analysis (po.05). This data is not necessarily surprising, given evidence in the United States that girls report less misbehavior than boys (Jensen & Eve, 1976), and it indicates that differences among female and males students across schools transcends any school-level differences, even when taking the single-sex school into account. Although descriptive data from the survey suggested that girls at International School may misbehave and engage their teachers more than females at other schools, that ﬁnding

What Matters for Chinese Girls’ Behavior and Performance in School

209

was not borne out in the regression analysis when we controlled for other factors including students’ aspirations and family background. Alongside the differences between females and males with regard to misbehavior and engagement, our descriptive survey data also demonstrated little to no difference between math/science and Chinese/English grades at Girls School compared to the signiﬁcantly lower math/science grades compared to Chinese/English grades among females at the other schools (and lower Chinese/English grades compared to math/science grades among males across co-educational schools). The regressions on math/science grades did not shed further light on what might be happening at Girls School to encourage the absence of such a difference. A small clue might be the small negative interaction between Girls School and expectations, thus suggesting that Girls School students may strive for high math and science grades for reasons beyond the high expectations that drive other students. However, those reasons are difﬁcult to ascertain beyond the potentially lower degree of gender bias at Girls School that could give girls the conﬁdence to work hard and do well in all subjects. This ﬁnding that Girls School females do equally well in math/science and Chinese/English is particularly interesting given the aforementioned evidence that Girls School caters to the ‘‘special needs’’ of girls and would therefore be expected to introduce stronger gender biases than coeducational schools, which does not appear to be the case. At the same time, we must also take into account that Girls School administration has the freedom to select at least some of the girls who attend that school. The selection of students might take into account whether students already receive high math and science grades. So while Girls School students report more balanced grades across subjects, they may do so for multiple reasons beyond what is measured in this analysis. In addition to examining differences in female and male outcomes across schools, we investigated what sources of authority motivate those outcomes. Speciﬁcally, we considered how much socialization and integration into the school (as exempliﬁed by moral attitudes, teacher attachment, and liking school) inﬂuenced student outcomes as compared to the desire to progress to higher levels of education. Across schools, for all students, attachment to teachers and moral attitude does negatively impact misbehavior, which is a conclusion that has already been drawn by sociologists of education doing work in the United States (Hirschi, 1969; Coleman & Hoffer, 1987; Arum, 2003; Bankston, 2004). However, as mentioned previously, females seem to misbehave less than males for reasons beyond socialization or other factors tested in these models.

210

JULIA KAUFMAN AND LIQUN YIN

Unlike what American sociological research suggests, socialization variables have less impact on achievement and engagement variables compared to students’ desire to move ahead to higher levels of schooling, even when school-level variables are added to the analysis. Although ‘‘liking school’’ was one socialization variable that had signiﬁcant effects on math and science grades (po.01) and positive, but non-signiﬁcant – effects on teacher engagement, those effects are likely bi-directional, in that students who do well in school and engage with their teachers probably like school more than those who do not. Thus, the performance of students across schools appears to be driven – in large part – by their desire to continue on to high school and university. These effects of a desire to get ahead in school are not conﬂated that much with socialization variables (i.e., the desire to get ahead is not closely correlated with teacher attachment or moral attitude, and therefore does not seem to reﬂect the same source of authority over student outcomes). This connection between Chinese students’ performance and their desire to get ahead might tell us something about how the education system itself operates as a legitimate authority motivating students to work toward higher levels of schooling. Interestingly, although socialization variables had less inﬂuence on achievement and engagement among all students, teacher attachment had a more signiﬁcant effect on math and science grades for girls than boys; and – similarly – moral attitude also has a more signiﬁcant effect on engagement with teachers for girls than boys. Thus, attention to socialization in Chinese schooling may encourage better outcomes for girls in both math and science achievement and engagement with teachers. That said, we should reiterate that the focus on moral education that seemed to be present at Girls School did not appear to impact outcomes for girls at that school. Therefore, teachers’ intentions to emphasize moral education in school may not – by themselves – impact student outcomes. As noted in our ﬁndings, International and Girls School both have more ability to select which students will attend their schools, which suggests prior differences among students who sit in schools as opposed to differences in teaching and learning within the schools. We added family background variables to our regression analysis in an attempt to control for this selectivity among some schools. Furthermore, our data indicates that family background does inﬂuence outcomes in interesting ways, in that students reporting higher parents’ education misbehaved less, whereas students of parents in a professional occupation misbehaved more. Additionally, students of professional parents also reported higher mathematics and science grades. However, despite our efforts to control for school selectivity

What Matters for Chinese Girls’ Behavior and Performance in School

211

through family background, we still cannot be certain that ﬁndings are not an artifact of schools’ selection processes. Finally, the reader should keep in mind that any positive or negative relationships between dependent variables and the sets of socialization variables and expectation variables might indicate culturally relevant elements of authority within Chinese education beyond any sociological conceptions of authority. For example, remnants of Confucianism within Chinese culture might encourage students to respond afﬁrmatively about their teachers and school beyond what might be predicted through sociological theories suggesting that socialization into the norms and values of any culture will produce positive attitudes toward education. Likewise, the examination system also likely derives its authority from age-old examination practices in China and their enduring connection with success, which is not considered as part of sociological theories on the control of examination and credentialing systems. However, such cultural inﬂuences conﬁrm rather than negate the use of these variables, as they might demonstrate how culture works alongside sociological predictions to tell us something about what inﬂuences student outcomes. Particularly, the greater inﬂuence of educational expectations on performance in the Chinese middle schools in this analysis, compared with existing data from American schools, suggests that the examination system holds sway over Chinese students in a way that cannot be reﬂected among students who do not exist in such a system. Continued research comparing female and male students’ behavior and learning across multiple settings will shed more light on how to provide the best educational outcomes for all students. Speciﬁcally, research comparing outcomes for girls and the factors that inﬂuence those outcomes in at least two settings (e.g., cultural/country comparisons or rural–urban comparisons) could further inform how examination systems and socialization in various contexts drive girls’ behavior and performance in school. Additionally, more data is needed to fully understand how single-sex and co-educational school environments inﬂuence girls across China and beyond. Because many of these schools have diverse missions and goals, such research in multiple settings might demonstrate what learning environments work best to educate all students.

NOTES 1. For a recent review of various theories and research related to authority, see Pace and Hemmings (2007).

212

JULIA KAUFMAN AND LIQUN YIN

2. In China, junior middle school (chuzhong) follows the ﬁrst six years of primary school and typically includes three grade levels: Junior One, Junior Two, and Junior Three (chuyi, chuer, chusan). The junior middle schools where I did my research also included a Pre-Junior (yubei chuyi) class that would otherwise be the sixth and last year of students’ primary school education. On graduation from junior middle school, most students hope to attend high school, or what is also called senior middle school (gaozhong). 3. For the purposes of conﬁdentiality, names for schools are pseudonyms. 4. The survey was additionally pilot-tested with 50 students at each of two schools prior to broad survey dissemination. Results from the pilot test were used to revise the survey for unclear or confusing words, phrases, and items. 5. On the basis of this occupational data, each parent was coded as ‘‘professional’’ if at least one parent worked in an occupation that ﬁt into the category of administration, management, military work, professional work, teaching, owning a business, or protective service work. Parents engaged in other kinds of work, including factory and labor work, were coded as ‘‘non-professional.’’ The coding, as well as interpretations of what deﬁnes ‘‘professional,’’ may have resulted in a higher percentage of students with parents in professional occupations than is actually present in each school. 6. The percentages of students who attend high school are based on reports from administrators at each school. 7. In the survey, the Chinese translation for engaging the teacher in debate is, ‘‘ ,’’ or engaging the teacher in debate about an academic question. 8. The variables – ‘‘parents’ education’’ and ‘‘professional parents’’ – takes into account the higher of either mother’s or father’s education/occupation. Occupational data was coded from students’ write-in responses in regard to their mother’s and father’s occupation, where parents were coded as ‘‘professional’’ if at least one parent worked in an occupation that ﬁt into the category of administration, management, military work, professional work, teaching, owning a business, or protective service work. The coding, as well as interpretations of what deﬁnes ‘‘professional,’’ may have resulted in a higher percentage of students with parents in professional occupations than is actually present in each school.

ACKNOWLEDGMENT The authors wish to thank Richard Arum, Philip Hosay, and Joanna WaleyCohen for their suggestions and academic guidance, as well as Gu Daxi and Frank Tang for providing the support and direction that enabled data collection for this chapter. Additionally, Julia Kaufman wishes to thank the Chiang Qing-kuo Foundation for International Scholarly Exchange for support through a dissertation writing fellowship, as well as New York University for additional ﬁnancial support through a University Dean’s Grant. Finally, this research could not have been done without the gracious cooperation of many administrators, teachers, and students at the schools where data was collected.

What Matters for Chinese Girls’ Behavior and Performance in School

213

REFERENCES Andreas, J. (2004). Leveling the little pagoda: The impact of college examinations, and their elimination, on rural education in China. Comparative Education Review, 48(1), 1–47. Arum, R. (2003). Judging school discipline: The crisis of moral authority in American schools. Cambridge, MA: Harvard University Press. Bankston, C. L., III. (2004). Social capital, cultural values, immigration, and academic achievement: The host country context and contradictory consequences. Sociology of Education, 77(2), 176–179. Bowles, S., & Gintis, S. (1976). Schooling in capitalist America: Educational reform and the contradictions of economic life. New York: Basic Books. Brown, D. K. (1995). Degrees of control: A sociology of educational expansion and occupational credentialism. New York: Teachers College Press. Coleman, J. S., & Hoffer, T. (1987). Public and private high schools: The impact of communities. New York: Basic Books. Collins, R. (1979). The credential society: A historical sociology of education and stratiﬁcation. New York: Academic Press. Connelly, R., & Zheng, Z. (2003). Determinants of school enrollment and completion of 10 to 18 year olds in China. Economics of Education Review, 22(4), 379–388. Durkheim, E. (1961). Moral education: A study in the theory and application of the sociology of education. New York: The Free Press. Eckstein, M. A., & Noah, H. J. (1993). Secondary school examinations: International perspectives on policies and practice. New Haven: Yale University Press. Gilson, J. (2002). Single-gender or coeducation for middle-school girls: Does it make a difference in math? In: A. Datnow & L. Hubbard (Eds), Gender in policy and practice: Perspectives on single-sex and coeducational schooling (pp. 227–242). New York: Routledge-Falmer. Hagan, J., Gillis, A. R., & Simpson, J. (1985). The class structure of gender and delinquency: Toward a power-control theory of common delinquent behavior. The American Journal of Sociology, 90(6), 1151–1178. Hannum, E. (2003). Poverty and basic education in rural China: Villages, households, and girls’ and boys’ enrollment. Comparative Education Review, 47(2), 141–159. Hannum, E., & Adams, J. (2007). Girls in Gansu, China: Expectations and aspirations for secondary schooling. In: M. Lewis & M. Lockheed (Eds), Social exclusion, gender, and education: Case studies from the developing world (pp. 71–98). Washington, DC: Center for Global Development. Harker, R. (2000). Achievement, gender and the single-sex/coed debate. British Journal of Sociology of Education, 21(2), 203–218. He, Y. (2001). Moral education and teaching and learning at the Shanghai No. 3 Girls Middle School. Chinese Education and Society, 34(1), 39–53. Hirschi, T. (1969). Causes of delinquency. Berkeley, CA: University of California Press. Jensen, G. J., & Eve, R. (1976). Sex differences in delinquency. Criminology, 13(4), 427–448. Jimenez, E., & Lockheed, M. E. (1989). Enhancing girls’ learning through single-sex education: Evidence and policy conundrum. Educational Evaluation and Policy Analysis, 11(2), 117–142. Kaufman, J. (2006). Authority and education in the People’s Republic of China. Doctoral dissertation, New York University. Dissertation Abstracts International, DAI-A67, 05, 288pp.

214

JULIA KAUFMAN AND LIQUN YIN

Kessen, W. (Ed.) (1975). Childhood in China. New Haven, CT: Yale University Press. Kessler, S., Ashenden, D. J., Connell, W., & Doussett, G. W. (1985). Gender relations in secondary schooling. Sociology of Education, 58(1), 34–48. Kipnis, A. (2001). The disturbing educational discipline of ‘‘peasants’’. The China Journal, 46, 1–24. Lee, V. E., & Lockheed, M. E. (1990). The effects of single-sex schooling on achievement and attitudes in Nigeria. Comparative Education Review, 34(2), 209–231. Li, Y. (2001). Beijing Huaxia Girls School. Chinese Education & Society, 34(1), 22–36. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Thousand Oaks, CA: Sage. Marsh, H. W. (1989). Effects of attending single-sex and co-educational high schools on achievement, attitudes, behaviors, and sex differences. Journal of Educational Psychology, 81(1), 70–85. Pace, J. L. (2003). Managing the dilemmas of professional and bureaucratic authority in a high school English class. Sociology of Education, 76(1), 37–52. Pace, J. L., & Hemmings, A. (2007). Understanding authority in classrooms: A review of theory, ideology, and research. Review of Educational Research, 77(1), 4–27. Salomone, R. C. (2003). Same, different, equal: Rethinking single-sex schooling. New Haven, CT: Yale University Press. Singer, S. I., & Levine, M. (1988). Power-control theory, gender, and delinquency: A partial replication with additional evidence on the effects of peers. Criminology, 26(4), 627–647. Strauss, A., & Corbin, J. M. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory. London: Sage. Streitmatter, J. (1999). For girls only: Making a case for single-sex schooling. Albany, NY: SUNY Press. Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Combining qualitative and quantitative approaches. Thousand Oaks, CA: Sage. Tobin, J., Wu, D., & Davidson, D. (1991). Preschool in three cultures: Japan, China and the United States. New Haven, CT: Yale University Press. Tsui, M., & Rich, L. (2002). The only child and educational opportunity for girls in urban China. Gender and Society, 16(1), 74–92. Unger, J. (1982). Education under Mao: Class and competition in Canton Schools, 1960–1980. New York: Columbia University Press. Weber, M. (1958). Bureaucracy. In: H. H. Gerth & C. W. Mills (Eds), From Max Weber: Essays in sociology (pp. 196–244). New York: Oxford University Press. Weber, M. (1978). Economy and society: An outline of interpretive sociology, Volumes I and II. Berkeley, CA: University of California Press. Yuan, Q. (2001). Restore schools for girls, optimize personality education, and improve the attributes of females. Chinese Education and Society, 34(1), 8–21. Zhang, Y., Kao, G., & Hannum, E. (2007). Do mothers in rural China practice gender equity in educational aspirations for children? Comparative Education Review, 51(2), 131–157. Zhou, M., & Bankston, C. L., III. (1999). Growing up American: How Vietnamese children adapt to life in the United States. New York: Russell Sage Foundation.

What Matters for Chinese Girls’ Behavior and Performance in School

215

APPENDIX A. VARIABLE DEFINITIONS Misbehavior

Math and science grades Teacher engagement

Moral attitude

Teacher attachment

Like school Students’ education expectations Parents’ education expectations Parents’ education

Professional parents

Composite variable taking into account how often during the past month student has skipped class or cheated on a test, and how often during the past semester student has been sent to ofﬁce for misbehavior, gotten in a ﬁght, or had attendance or misbehavior warning sent to parents. Higher indicates greater frequency of bad behavior (a ¼ .60). Average of student’s grades in math and science; grades coded 1–5 with 5 ¼ A; 4 ¼ B; 3 ¼ C; 2 ¼ D; 1 ¼ F. Composite variable taking into account how often during the past month student asked the teacher a question and debated with teachers. Higher indicates greater frequency of behavior (a ¼ .62). Composite variable taking into account student’s agreement with the following statements: ‘‘most students who get into trouble shouldn’t be blamed for the things they have done’’; ‘‘it is alright to break school rules if you can get away with it,’’ and ‘‘the student who leaves his/her bag unattended is about as much to blame for its theft as the student who steals it.’’ Higher indicates greater degree of disagreement (and thus more ‘‘moral’’ attitude) (a ¼ .60). Composite variable of student’s agreement with the following statements: ‘‘teachers are good’’; ‘‘teachers are interested in students’’; ‘‘teachers really listen to what I say’’; and ‘‘I respect teachers at this school’’ Each statement coded 1–4 with higher indicating greater agreement (a ¼ .78). Students’ responses coded as 1 ¼ Dislike school; 2 ¼ Like and dislike school equally; 3 ¼ Like school. Coded 1–4 with 1 ¼ at least graduate high school; 2 ¼ at least attend college or vocational school; 3 ¼ at least graduate college; 4 ¼ at least attend graduate school. Coded 1–4 with 1 ¼ at least graduate high school; 2 ¼ at least attend college or vocational school; 3 ¼ at least graduate college; 4 ¼ at least attend graduate school. Coded 1–3 with 1 ¼ at least graduated middle school; 2 ¼ at least graduated high school; and 3 ¼ graduated from university. Father’s education level unless father’s is less than mother’s. Coded 0/1 with 1 ¼ professional occupation and 2 ¼ nonprofessional occupation. Father’s occupation level unless father is non-professional, whereas mother is professional.

216

JULIA KAUFMAN AND LIQUN YIN

APPENDIX B. MEANS AND STANDARD DEVIATIONS FOR DEPENDENT AND INDEPENDENT VARIABLES USED IN REGRESSIONS Variables

All N

Misbehavior*** Math and science grades** Teacher engagement Moral attitude* Teacher attachment Like school Students’ education expectations Parents’ education expectations** Parents’ education Professional parents

769 750 762 768 757 765 756

Males

Mean (SD) .84 3.46 2.22 9.02 11.73 2.22 3.37

(1.97) (1.06) (2.53) (1.79) (2.30) (.59) (.76)

N 218 209 217 217 214 217 213

Females

Mean (SD) 1.51 3.62 2.58 8.75 11.79 2.27 3.46

(2.64) (1.01) (2.79) (2.09) (2.50) (.62) (.73)

N 548 538 542 548 540 545 541

Mean (SD) .57 3.39 2.08 9.14 11.71 2.20 3.34

(1.56) (1.07) (2.41) (1.63) (2.21) (.576) (.776)

639

3.40 (.87)

177

3.58 (.75)

460

3.33 (.902)

716 653

2.50 (.62) .76 (.43)

199 181

2.48 (.66) .75 (.43)

514 468

2.51 (.600) .77 (.423)

Note: Independent t-tests indicated some signiﬁcant differences between means of female and male students. Variables where we found signiﬁcant differences between females and males are noted, with ***po.001; **po.01; *po.05.

GENDER GAP AND WOMEN’S PARTICIPATION IN HIGHER EDUCATION: VIEWS FROM JAPAN, MONGOLIA, AND INDIA Yoshiko Nozaki, Rima Aranha, Rachel Fix Dominguez and Yuri Nakajima INTRODUCTION One of the most signiﬁcant worldwide transformations in education over the past several decades has been the drastic increase in women’s access to colleges and universities. Research suggests that the trend of the narrowing gender gap in higher education is remarkable (particularly, among the industrialized nations), and sometimes it involves an interesting phenomenon – women outnumbering men, in what some scholars refer to as a ‘‘reverse gender gap’’ (Goldin, Katz, & Kuziemko, 2006; Woodﬁeld & EarlNovell, 2006; King, 2006; Mortenson, 1999). This higher education gender gap trend is consistent with a general global trend of narrowing gender gaps in education in recent decades. The data – at least, analysis of statistical data from countries around the world – support the contention that the disparity between men and women, at all levels of education and in terms of both academic achievement and enrollment rates, is not as dramatic as it once

Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 217–254 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010010

217

218

YOSHIKO NOZAKI ET AL.

was (Arnot, David, & Weiner, 1999; United Nations Children’s Fund, 2005). However, there is a dissonance: the narrowing gender gap in education does not necessarily mean that gender inequalities in various spheres of society are narrowing simultaneously.1 Indeed, reports and studies often mention both narrowing and persistent gender gaps in education as well as in other societal spheres such as economy and politics (e.g., United Nations International Children’s Emergency Fund, UNICEF, 2005), and studies in the area of higher education are no exception (e.g., Shavit, Arum, & Gamoran, 2007). Does this dissonance appear in higher education systems and societies in Asia, where the systems have grown and societies have industrialized as rapidly as ever? Since there have been few studies that have pursued this question from an international comparative perspective, it seems to be worth examining the gender gap in higher education in Asia from that perspective. This chapter presents comparative country case studies of three Asian nations – Japan, Mongolia, and India – concerning gender gaps in their higher education enrollments, along with a review analysis of relevant literature. Speciﬁcally, we are concerned with the following questions: How has women’s higher education enrollment changed over time in the contexts of higher education system development, transition, and/or expansion in each of these three countries? How are men and women distributed across the higher education system of each country? What types of institutions (e.g., two year or four year institutions) have a larger share of women’s enrollment? Which ﬁelds of study are predominantly female and which are predominantly male? If there are patterns of gender disparities, what are their meaning(s) in relation to women’s paid work and the socio-historical contexts of marriage and families? There have been two main theories for the gender disparities in tertiary education generally, and for the choices of ﬁeld of study in particular, between men and women.2 One theory emphasizes the preferences and choices women make for their participation in economy, and the other stresses women’s socialization (or identity formations) inﬂuenced by cultural values and norms, societal pressures and expectations. The former theory assumes that women have total human agency to make ‘‘free choices,’’ and the latter often ends up depicting women as ‘‘victims’’ of social and historical oppressions (Arnot, 2002). We would argue, however, that this is a false dichotomy, because, indeed, whatever choices women make, their decisions are inevitably inﬂuenced and limited by societal – be it economic, cultural, or otherwise – conditions for women. Moreover, implicit in the

Gender Gap and Women’s Participation in Higher Education

219

dichotomy is a binary opposition of ‘‘economy versus culture,’’ which ignores that any cultural activity – say, marriage – has an economic, or class, dimension. In this chapter, we would like to begin to explore – but by no means conclusively – a possible third explanation for why women’s choices in education and career are often more difﬁcult, or non-straightforward, than men’s, which may result in helping reproduce, or maintain, gender disparities. We would argue that women’s decision-makings turn on two axes: one involves jobs and labor market outcomes, and the other involves marriage and family, and that both are in part economic and class – and perhaps caste in the Indian case – decisions women make for their conditions of existence (for further discussion, see Nozaki, forthcoming; also see Hara & Seiyama, 1999).

A Methodological Note Although a good deal of literature exists on the subject of gender gap and higher education enrollment in the United States, it is often conﬁned within the limited perspective of U.S. higher education institutions (e.g., King, 2006; Jacob, 2002), or economically developed countries (e.g., VincentLancrin, 2008). International studies offer, therefore, an important opportunity to examine, and gain perspectives from, higher education gender gaps in non-U.S. contexts. Those existing international, or non-U.S., studies on gender and higher education also tend to focus on (relatively) speciﬁc topics within a single national context. For example, chapters in edited volumes such as Kelly and Slaughter (1991), Conway and Bourque (1993), and Lie, Malik, and Harris (1994) contain very rich material and discussions, but lack an encompassing framework and comparable units of analysis to make tight cross-study comparisons. Bringing a comparative framework into country case studies is a vital ﬁrst step toward generating ﬁndings comparable (and generalizable) across countries. We have used a collaborative (team) approach to the development of international comparative country case studies.3 As a methodological concept, our attempt is similar to that of some quantitative sociologists (e.g., Shavit et al., 2007); however, our study employs a comparative history/ sociology approach with descriptive statistics (Neuman, 2006). Further, although Shavit et al. (2007) use nationally representative data, we use internationally comparable statistical data, available from international organizations such as United Nations Educational, Scientiﬁc, and Cultural

220

YOSHIKO NOZAKI ET AL.

Organization (UNESCO) Institute for Statistics and the World Bank (and these data are checked against each country’s national data, whenever possible). Our research design also acknowledges the fundamental importance of historical and contextual interpretations of statistical data; such interpretations are sometimes (perhaps inevitably) glossed over when studies conduct a comparative quantitative study of many countries (e.g., Bradley, 2000; Charles & Bradley, 2002). The Japanese, Mongolian, and Indian country case studies herein each examine common threads of thematic issues, and these particular countries were selected for inclusion based on the variety or diversity they provide and the availability of both data and researchers who have familiarity with each country’s education system (for further discussion on the methodological issues, see Nozaki, forthcoming).

THE CASE OF JAPAN Historical Development of Japanese Higher Education Japan’s postsecondary education began with the country’s modernization efforts in the late 19th century. In 1877, the Japanese government established the nation’s ﬁrst university, Tokyo University (renamed as Imperial University in 1918, now the University of Tokyo), with four departments of law, science, literature, and medicine (University of Tokyo, 2008). Later it founded four other imperial universities (Narita, 1978).4 These imperial universities, which admitted very few female students, were to produce high-ranking civil servants, modeled after western countries such as Britain, the United States, and Germany (Amano, 1979).5 There were also other types of institutions, such as higher schools (or schools of higher learning), technical schools, and teacher’s colleges (some of these were private).6 In the late 1910s, in addition to the 5 imperial universities, the government accredited 10 national colleges and 20 private universities, all of them being predominantly male institutions. When World War II ended, Japan had a complex, streamlined system of postsecondary education, with more than 500 institutions (also see Amano, 1979). After the war, the Japanese state reformed the higher education system, ﬁrst by creating a new category of co-educational, four-year institutions (not differentiating universities from colleges), with the imperial universities and other postsecondary institutions such as higher schools. Since the national universities in the pre-surrender period were extremely privileged, elite institutions, the state established at least one national university (the so

Gender Gap and Women’s Participation in Higher Education

221

called ‘‘local-national’’ university) in each prefecture from the viewpoint of equal access to higher education among various provinces (Narita, 1978). The state also created a category of two-year institutions, or junior colleges, in 1950, which soon became dominated by women. Furthermore, in 1962, the state established a category of colleges of technology (ﬁve-year institutions of grades 10 to 14), the last two years of which are regarded as tertiary education. In 1976, many technical schools were classiﬁed (and so accredited) as ‘‘specialized training schools.’’7

Recent Expansion of Higher Education System As shown by Fig. 1, student enrollment in tertiary education increased from approximately 1.8 million in 1970 to approximately 4 million by 2005. In recent decades, there has been an expansion of higher education student enrollment in the early 1990s, especially due to the university reform, which aimed at deregulation, leading to the increase in the number of private institutions (Morozumi, 2005; also see Amano & Poole, 2005). However, currently, the student enrolment is stagnant, and the higher education institutions are in the middle of seeking new directions (e.g., Eades, Goodman, & Hada, 2005). As of 2006, there were 176 public universities8 and 568 private universities. Additionally, there were 468 junior colleges and 64 colleges of technology (Japanese Ministry of Education, Culture, Sports, and Technology (hereafter Japanese MOE), n.d.[a]; also see ‘‘Japan,’’ n.d.).

Fig. 1. Tertiary Education Enrollment in Japan. Source: The Data Accessed through EdStats Query (World Bank Group, 2008).

222

YOSHIKO NOZAKI ET AL.

The Japanese tertiary education system has remained hierarchical and complexly diversiﬁed with fairly clearly ranked institutions. In spite of the effort for regional equality, national universities have been divided into two classes: an upper group composed of former imperial universities and some higher schools from the old system, which could offer graduate education through the PhD level, and a lower group composed of education universities and the newly formed local–national universities (e.g., Amano, 1979). Prestigious private universities have been ranked just below the most prestigious national universities, but above the local–national universities. Junior colleges and colleges of technology, public and private, have usually been ranked below the four-year institutions, and below are the specialized training schools. The universities and colleges, including junior colleges, have focused on awarding academic credentials, as Japanese industries and businesses have preferred in-house training of new recruits (e.g., Rohlen, 1984; for a critique of Japanese higher education, see McVeigh, 2002). Although the University of Tokyo and several national universities are more prominent and prestigious than elite private institutions, in terms of numbers, the Japanese higher education system is dominated by the private sector. As of 2006, 76 percent of four-year institutions were private and 90 percent of junior colleges were private. 95 percent of colleges of technologies were public, but they constituted a small number (Japanese MOE, n.d.[b]). In addition, nine out of ten specialized training schools are private (Ishida, 2007; Nagasawa, 2005). Many private institutions have to meet the state accreditation requirements in matters such as availability and size of physical facilities, personnel qualiﬁcations, and curricula; they are also eligible to receive state funding, which often results in the strengthening of state control.

Gender Gaps in Student Enrollment Even though its compulsory education ends at 9th grade, Japan has achieved universal education both at the primary and secondary levels (Tables 1 and 2). As shown in Table 2, the country’s secondary education net enrollment rate was 98.7 in 2005 with a gender parity index of 1.00.9 At the level of tertiary education, however, men have outnumbered women for decades.10 Fig. 1 shows that the gap in sheer student enrollment between men and women has constantly narrowed since 1970, but that it has not reached the gender parity point yet. The time series data of the Gross Enrollment Ratio (GER) reveals more details: the gap between men’s GER

223

Gender Gap and Women’s Participation in Higher Education

Table 1.

Japan Mongolia India

Primary Education Net Enrollment Rate (NER) and Gross Enrollment Ratio (GER) (2005). NER/GER (in 2005)

Total %

Male %

Female %

Gender Gap

Gender Parity Index

NER GER NER GER NER GER

99.8 99.9 86.9 96.8 88.5 114.6

99.7 99.7 85.9 95.8 90.1 115.9

100.0 100.0 88.0 97.9 86.8 113.2

0.3 0.3 2.1 2.1 3.3 2.7

1.00 1.00 1.02 1.02 0.96 0.98

Source: The data accessed through Data Centre (UNESCO Institute for Statistics, 2008).

Table 2.

Japan Mongolia India

Secondary Education Net Enrollment Rate (NER) and Gross Enrollment Ratio (GER) (2005). NER/GER

Total %

Male %

Female %

Gender Gap

Gender Parity Index

NER in 2006a GER in 2005 NER in 2005 GER in 2005 NER in 2005 or 2006 GER in 2005

98.7 101.6 83.1 90.7 Data unavailable 54.0

98.5 101.5 77.9 85.2 Data unavailable 59.0

98.9 101.6 88.5 96.4 Data unavailable 48.6

0.4 0.1 10.6 11.2 Data unavailable 10.4

1.00 1.00 1.14 1.13 Data unavailable 0.82

Source: The data accessed through Data Centre (UNESCO Institute for Statistics, 2008). a Data in 2006 are presented because those in 2005 are unavailable.

and women’s GER widened in the 1970s and 1980s, but had signiﬁcantly narrowed by 2005, with a gap of 20.1 in 1980, 12.9 in 1990, and 6.6 in 2005 (Table 3).11 However, there are some critical gaps. First, Japanese women still tend not to pursue graduate degrees. As of 2005, 40.2 percent of undergraduate students (including those in junior colleges) were women, 30.0 percent of Master course students were women, and 29.7 percent of Doctoral course students were women (Japanese MOE, n.d.[c]). Although no speciﬁc research has been done to understand this tendency, governmental reports ﬁnd that women researchers face difﬁculties staying and advancing their careers because of their responsibility on child rearing and taking care of

224

YOSHIKO NOZAKI ET AL.

Table 3.

Japan Total % Male % Female % Gender gap Mongolia Total % Male % Female % Gender gap India Total % Male % Female % Gender gap

Teritary Education Gross Enrollment Ratio (1970–2005). 1970

1975

1980

1985

1990

1995

2000

2005

17.6 25.3 9.9 15.4

26.3 35.3 17.2 18.1

30.5 40.4 20.3 20.1

27.8 35.5 19.8 15.7

30.7 37 24.1 12.9

41.7 45.6 37.6 8

47.4 51 43.6 7.4

55.3 58.5 51.9 6.6

5.5 No data No data N/A

7.3 7 7.5 0.5

21.8 16.9 26.7 9.8

21.6 15.9 27.6 11.7

14.3 10 18.8 8.8

15.2 9.1 21.4 12.3

28.8 20.7 37.1 16.4

43.3 33.1 53.7 20.6

5.1 7.5 2.5 5.0

5.2 7.4 2.9 4.5

6.0 8.1 3.8 4.3

6.2 7.9 4.3 3.6

6.6 8.1 5.0 3.1

9.6 11.5 7.6 3.9

11.0 12.8 9.1 3.7

4.9 7.4 2.2 5.2

Source: The data accessed through Edstats Query (World Bank Group, 2008).

aging parents (Japanese Gender Equality Bureau, Cabinet Ofﬁce, 2008). One may, thus, suspect that female students, anticipating these difﬁculties, decide not pursue academic and research careers – not to mention an increasing number of reported sexual harassment incidents in Japanese universities (e.g., Murakami, 2003; Science Council of Japan, 2005). Second, gender segregation is taking place between the different types of – or differently ranked – Japanese tertiary education institutions to a great extent (e.g., Matsui, 1997). Although approximately two-thirds of male students go on to four-year universities, a signiﬁcant number of female students go on to two-year junior colleges (though the portion of female students going on to universities has been increasing in recent years). For instance, in 2004, female students constituted 39 percent of all students of four-year universities, whereas students were almost exclusively female in junior colleges (which are predominantly private institutions) in that same time period (Japanese MOE, n.d.[d]). In addition, only 18 percent of students in colleges of technology were women in 2003, whereas specialized training schools were attended by an almost equal number of men and women (Ishida, 2007). The former schools are predominantly public, prestigious (as the vocational education institutions), and offer an academic degree (and credits earned there are fully transferable to other universities).

Gender Gap and Women’s Participation in Higher Education

225

In other words, in the technology- and vocation-related education, women are concentrated in the lower strata.

Gender Distribution by Fields of Study As shown in Table 4, in 2005, approximately 1.9 million women were enrolled in the higher education institutions in Japan, constituting 45.9 percent of its entire student body. In terms of sheer numbers of students attending, the ﬁeld of Humanities and Arts drew the largest number of women, with approximately 440,000 women enrolled. The second was the ﬁeld of Social Sciences, Business, and Law, with 400,000 female students, and the third was the ﬁeld of Health and Welfare, with 290,000 female students.12 Examining women’s enrollment share in each ﬁeld of study, we ﬁnd that the Japanese higher education system exhibits a strong pattern of gender segregation by ﬁeld. In 2005, the range of variations in women’s enrollment share by ﬁelds of study was 66.4 percentage points (78.3 11.9), indicating that some highly female-dominated ﬁelds existed along with some highly male-dominated ﬁelds. Speciﬁcally, in 2005, the most women-dominated ﬁeld was that of Service, in which 78.3 percent of the enrollees were women (see Table 4). A unique discipline of ‘‘home economics’’ is included in this ﬁeld, and, indeed, according to the Japanese government data, more than 90 percent of students have been women (National Women’s Education Center, Japan, n.d.; also see Nozaki, Fix Dominguez, & Nakajima, 2008). In the ﬁeld of Education, women also constituted 70.9 percent of its student body and, in the ﬁeld of Humanities and Arts, 67.0 percent of its student body. Women’s enrollment share in the Japanese higher education as a whole was 45.9 percent in the same year, so we can calculate each ﬁeld’s deviation scores in terms of women’s enrollment share. The ﬁeld of Service had 32.4 percentage points more women (78.3 45.9), the ﬁeld of Education had 25.0 percentage points, and the ﬁeld of Humanities and Arts had 21.1 percentage points more women than the higher education system as a whole. Two ﬁelds of study in the Japanese tertiary education system stood out as greatly male dominated: the ﬁeld of Engineering, Manufacturing, and Construction and the ﬁeld of Science. The former had 34.0 percentage point fewer women (11.9 45.9) and the latter had 20.7 percentage points fewer women than the higher education system as a whole.

226

Table 4.

Women’s Enrollment by Fields of Study (2005).

Japan

India

Total enrollment

Women’s enrollment

Women’s Share (%)

Total enrollment

Women’s enrollment

Women’s Share (%)

Total enrollment

Women’s enrollment

Women’s Share (%)

4038302

1853130

45.9

123824

76049

61.4

11777296

4641576

39.4

289063 653330 1158144

204968 437925 400189

70.9 67.0 34.6

12126 15926 47290

9340 11502 30540

77.0 72.2 64.6

155192 4241507 1587285

68049 1873324 578488

43.8 44.2 36.4

118704 668526

29911 79468

25.2 11.9

8262 20117

3919 8253

47.4 41.0

1689504 696609

671223 165402

39.7 23.7

87214 478656 275846 308819

34607 294422 215965 155675

39.7 61.5 78.3 50.4

3854 9585 6214 450

2324 7750 2120 301

60.3 80.9 34.1 66.9

0 256748 0 3150451

0 89052 0 1196038

N/A 34.7 N/A 38.0

Source: The data accessed through Data Centre (UNESCO Institute for Statistics, 2008).

YOSHIKO NOZAKI ET AL.

All ﬁelds combined (total enrollment) Education Humanities and arts Social sciences, business, and law Science Engineering, manufacturing, and construction Agriculture Health and welfare Service Unspeciﬁed

Mongolia

Gender Gap and Women’s Participation in Higher Education

227

Japanese researchers have argued that unequal gender distribution by the ﬁelds of study in higher education are a result of earlier genderdifferentiated socialization. Nakanishi (1992) points out that high schools have a ‘‘gender track’’ based on the view of gender roles, and it is expressed by their schools’ histories, policies, curriculum, and practices, some of which encourage female students to be good wives and mothers, and some of which encourage women to be labor force in certain occupations. In her view, female students are socialized into this ‘‘gender track,’’ and choose their majors in college and their future careers based on the values and norms permeating in the track. Some studies demonstrate the gendered hidden curriculum at the elementary and secondary education levels, and sexism in school textbooks, customs, teachers’ behavior, and school policies (e.g., Kameda, 1995; Kimura, 1997; Miyazaki, 1991; Mori, 1989; Ujihara, 1996). The gender track then continues at the level of tertiary education (and, indeed, in the workplace), by institutional arrangements, higher education curriculum, or women’s choices (e.g., McVeigh, 1997). Although the theory of socialization explains the way society’s cultural values, norms, and expectations shape women’s views on themselves, it neglects taking into account women as possessing human agency. Women are not completely passive to be molded into the roles the dominant culture offers, but they often actively read societal gender dynamics, seek and evaluate alternative ideas and thoughts about women’s roles, and make decisions about their life styles and careers. In other words, we have to assume that, when making decision on their postsecondary education, including their choices not to enter, women are aware of various social relations surrounding them, and they make decisions based on their desires within their conditions of existence. Two domains of social relations are important here: labor market and marriage.

Career, Marriage, and Educational Attainment among Japanese Women In Japan, there seems to be a (major) disconnection between women’s higher education attainment and their occupational outcomes and destinations (e.g., Amano, 1986). In particular, the increase in the number of women with tertiary education does not necessarily correspond to an increase in the employment rate of women (e.g., Manabe, 1997). Rather, the higher is women’s educational attainment, the lower the rate of their being in employment (e.g., Hara & Seiyama, 1999; Brinton, 1993).

228

YOSHIKO NOZAKI ET AL.

The disconnection is, in part, because of gender discriminatory labor market practices. College-educated women in Japan try to get a job on their graduation; however, there they face gendered hiring practices. For example, Ishida (1998) ﬁnds that, although male university graduates and high-school graduates of highly competitive academic schools have particularly greater chances of employment in large ﬁrms, young women’s chances do not seem to be closely related to education.13 If they get a job, they tend to continue to work after marriage, but most often quit after having their ﬁrst child, and try to re-enter the labor market as part-time or non-regular workers after childrearing (Imada, 1985). At this re-entry into labor market, the effects of academic attainment disappear (Manabe, 1997). In other words, the effects of educational attainment for women are high only when (and if) they continue their career after the marriage and child bearing. The disconnection is also because of gender role division in the institution of marriage. Women with higher academic attainment tend to get married to men with higher academic attainment. In this case, husbands are likely to work in high status jobs, well-paid with a good beneﬁt package, but they have long-working hours, expecting their wives to deal with family matters, including parenting and caring (aging) parents and in-laws. It is a condition upon which often women have to discontinue their work; however, because their husbands are well-paid, the women can maintain their upper and middle class status, and this is important to their identity as a good mother and wife of upper/middle class family (also see Hara & Seiyama, 1999). It appears that gendered conditions and practices in the labor market and in the institution of marriage affect Japanese women’s identities and choices in higher education, helping, in part, maintain gendered social, cultural, and economic relations. For example, women enrolled in liberal arts and professional education programs often have the end goal of entering femaledominated occupations such as elementary school teachers, nurses, nursery teachers, and nutritionists, in which they have better chances of getting hired (Amano, 1986). Amano (1988) has argued that as women attend higher education institutions, they become more culturally conformist. Speciﬁcally, in the job-hunting phase, female university students’ motivation for and conﬁdence about job performance after graduation is lower than male counterparts (Rodo Seisaku Kenkyu Kenshu Kiko, 2006). Yoshihara (1995) argues that the women recognize and accept the cultural norm of femininity (e.g., ‘‘women be women’’), resulting in the differentiation in women’s career patterns. A recent report by Organization for Economic Co-operation and Development (OECD) (2008) points out that Japanese women’s underemployment

Gender Gap and Women’s Participation in Higher Education

229

is a considerable waste of valuable human resources, and that the nation needs to address it urgently as it faces a rapidly aging population. The literature reviewed here suggests, however, that solving the problem may not be easy because the disconnection between Japanese women’s higher education attainment and underemployment involves multitudes of social (and economic) issues that need to be dealt with simultaneously.

THE CASE OF MONGOLIA Mongolian Higher Education: Soviet Era Development and the Recent Transition and Expansion In 1991, Mongolia was the ﬁrst Central Asian nation to transition from a command to a market economy, which led to dramatic changes to the structure of higher education in the country (e.g., Weidman et al., 1998; Weidman, 2002). The transition was not an easy experience for the nation in general, and for women in particular (e.g., Robinson & Solongo, 2000). Although it remains to be seen how this transition has changed gender, class, and other social and cultural dynamics across Mongolia, it is important to examine the impact of transition on gender and higher education to date. Certainly, Mongolia already had a well-structured, if limited, higher education system before the transition (e.g., Weidman et al., 1998; Weidman, 2002). The ﬁrst university in Mongolia, Choybalsan University (previously known as Mongolian State University and now as the National University of Mongolia, hereafter NUM), was founded in 1942 with three departments – education, medicine, and veterinary medicine – in which both men and women were given the opportunity for higher education equally (Robinson & Solongo, 2000). The language of instruction was Russian and most faculty members were Russian. Over time, the university expanded and split into different institutes – engineering, sciences and mathematics, social sciences, economics, and philology. By the mid-1980s, over 90 percent of the faculty was Mongolian, many of whom had studied in Russia. Much of the instruction was still in Russian, reﬂecting the lack of Mongolian language texts in advanced and specialized ﬁelds (Worden & Savada, 1989). There were eight branches (or institutes) making up the state-run university system before the transition, including the Institute of Medicine, the Institute of Agriculture, the Institute of Economics, the State Pedagogical Institute, the

230

YOSHIKO NOZAKI ET AL.

Polytechnic Institute, the Institute of Russian Language, and the Institute of Physical Culture. After the transition, the new conceptual framework of higher education was shaped by the World Bank and it focused on rationalization of public expenditure, reallocation of resources, and mobilization of new resources (Government of Mongolia, 2003) Under this framework, the Mongolian state allowed private higher education institutions to open their doors in 1991 and initiated a student loan scheme and introduced tuition fees in 1992. Furthermore, in 1993, the state eliminated student stipends for higher education, raised student-to-faculty ratios, sought out new sources of funding for higher education institutions, including donations from international organizations and contributions from individuals, and demanddriven educational programs such as business management and foreign language. Additionally, in 1998, the state created the National Council for Higher Education Accreditation (Mongolia, n.d.). Since the transition, there have been many new private higher education institutions created, though the quality of education offered by these private institutions needs to be examined in greater depth. Additionally, English replaced Russian as the ofﬁcial second language in Mongolia in 2004 (Brooke, 2005). Currently there are approximately 185 higher education institutions in Mongolia. About 22 percent of these institutions are public, another 74 percent private, and the remaining 4 percent satellite branches of foreign universities (Mongolia, n.d.). The NUM is still the most prominent and prestigious higher education institution in the country. There are several institutions types, including four-year universities, two-year colleges, and vocational schools. Data disaggregated by institutional type (e.g., four-year universities versus two-year colleges) among public and private institutions are not currently available. Since the introduction of student tuition, Mongolian higher education enrollments have increasingly been driven by student demand, resulting in both diversiﬁcation of programs and a market-oriented approach to higher education (Weidman et al., 1998). In 2005, there were approximately 124,000 students in higher education institutions in Mongolia (see Fig. 2 and Table 3). The GER of higher education was 7.3 percent in 1975 and 21.6 percent in 1985. Following the transition, this number decreased in the early 1990s (hit the bottom of 10.6 percent in 1992), but rebounded in the late 1990s to rise dramatically by 2005 to 43.3 percent. GER indicates the growth of higher education in the country over the past several decades, especially its enormous expansion in the second half of the 1990s and in the 2000s. As discussed below, the

Gender Gap and Women’s Participation in Higher Education

231

Fig. 2. Tertiary Education Enrollment in Mongolia. Source: The Data Accessed through EdStats Query (World Bank Group, 2008).

Mongolian GER also shows a growing reverse gender gap between men’s and women’s tertiary education attendance over time.

Gender Gaps in Student Enrollment Mongolian education’s GER data disaggregated by gender indicates the presence of a reverse gender gap at every level of the education system (see Tables 1–3). Though research on education and the reverse gender gap in Mongolia has, until recently, focused primarily on the primary and secondary levels (UNICEF, 2004; UNICEF-Mongolia, 2007), there is a wider GER gap at the tertiary level than at the primary or secondary levels (Fix Dominguez, 2008). This means that Mongolian women have pursued higher education qualiﬁcations and degrees more than their male counterparts, which stands in contrast to other two countries in this study, and, indeed, to many other Asian nations (Robinson & Solongo, 2000; Shabaya & Konadu-Agyemang, 2004). Although Table 3 shows that the reverse gender gap in Mongolian higher education dates back to the late 1970s (see also Fig. 2), in the period the higher education decline, transition reform, and drastic expansion of the 1990s and 2000s, there was a growing reverse gender gap in GER. Both male and female GERs decreased signiﬁcantly before the transition, but female

232

YOSHIKO NOZAKI ET AL.

GER rebounded quickly in the late 1990s. Speciﬁcally, although only 10 percent of all eligible male students attended higher education institutions in 1990, 18.8 percent of female eligible students attended these institutions. Five years later, in 1995, 9.1 percent of eligible males were at universities, whereas 21.4 percent of females were enrolled in higher education. By 2005, there was a 24.0 percentage point gain in the GER of male tertiary students (bringing it to 33.1 percent), whereas women’s gain was 32.3 percentage points. In other words, more than half (53.7 percent) of all eligible female students were enrolled at the tertiary level. However, this growing GER gender gap might be somewhat misleading. Fig. 3 displays the time series data of women’s share in tertiary education enrollment of three countries, and Mongolian women’s enrollment share has fallen from a peak of 70.0 percent in 1995, to 63.8 percent in 2000, and to 61.4 percent in 2005. In other words, since the late 1990s, the number of male students attending tertiary institutions has rebounded and, indeed, male enrollment share has grown almost every year. In terms of raw numbers, there are now more than twice as many men attending higher education institutions as compared to the early 1990s (see also Fig. 2). So although a reverse gender gap does still exist at the tertiary level, Mongolian men appear to be availing themselves of higher educational opportunities as well, and this is an interesting angle for a further investigation.

Fig. 3. Women’s Enrollment Share in Tertiary Education. Source: The Data Accessed through EdStats Query (World Bank Group, 2008).

Gender Gap and Women’s Participation in Higher Education

233

Gender Inequality by Fields of Study Mongolia’s reverse gender gap phenomenon does not mean that the distribution of men and women are equal by ﬁelds of study in its tertiary education system. In 2005, approximately 76,000 women were enrolled in Mongolian higher education institutions (see Table 4). In terms of sheer numbers of female student enrollment, the ﬁeld of Social Sciences, Business, and Law was the largest, with approximately 30,000 women enrolled. The second was the ﬁeld of Humanities and Arts, with approximately 12,000 female students, and the third was the ﬁeld of Education, with approximately 9,000 female students. It is critical to examine women’s enrollment share in order for us to understand the gender dynamics by ﬁelds of study. In Mongolia, women dominated ﬁve out of the eight ﬁelds of study in 2005. Although the Mongolian higher education system shows signs of gender segregation by ﬁelds of study, it is not to the extent Japan does. The range of variations in female enrollment share in each ﬁeld was 46.8 percentage points (80.9 34.1), which is lower than Japan’s but higher than India’s. The most femaledominated ﬁeld was that of Health and Welfare, in which women constituted 80.9 percent of its student body. Women also constituted 77.0 percent of attendees in the ﬁeld of Education. Women’s enrollment share in the entire Mongolian higher education student population was higher than other countries (61.4 percent), so looking at deviation scores from this population average is important in terms of comparison. The ﬁeld of Health and Welfare had a deviation score of 19.5 percentage points (80.9 61.4) and the ﬁeld of Education had a deviation score of 15.6 percentage points. Two ﬁelds of study that stood out as greatly male-dominated were the ﬁeld of Service (including personal services, transport services, environmental protection, and security services)14 and the ﬁeld of Engineering, Manufacturing, and Construction. The former had 27.3 percentage points fewer women (34.1 61.4) and the latter had 20.4 percentage points fewer women (41.0 61.4) than the higher education system as a whole. Actual gender segregation by ﬁelds of study at local levels may differ somewhat from that suggested by the national-level statistics discussed above, depending on the types of institutions. For example, the data available on men and women’s participation in different ﬁelds at the NUM, the nation’s top university, slightly deviate from the national statistics.15 In the 2003–2004 academic year, 61 percent of NUM’s 10,017 students were female, which is about the national level indicated by UNESCO data.

234

YOSHIKO NOZAKI ET AL.

However, only 29 percent of undergraduate students in NUM’s engineering, including computer and electronic engineering, were female, below 41.0 percent the national level statistics would suggest for the ﬁeld of Engineering, Manufacturing, and Construction. In the same school year at NUM, a little over half (53 percent) of undergraduate students enrolled in mathematics and statistics programs were female and a little less than half (47 percent) of students in physics were female. These numbers are more or less consistent with the national level statistical data for the ﬁelds of Science. On the other hand, in the foreign language translation program, women constituted 87 percent of undergraduates, the number being higher than that of the UNESCO data for the ﬁeld of Humanities and Arts (NUM, 2004; also see Fix Dominguez, 2008). However, we must note that NUM is the most prestigious and highly competitive institution with best programs in the nation. It appears that gender segregation by ﬁelds of study may be more pronounced at NUM than at the national level, and this raises a concern that should be further investigated.

Women’s Work, Marriage, Family, and Higher Education in the Transition How might we understand the gendered dimensions of the Mongolian higher education system in the social and historical contexts, including the recent period of transition? Some inquiry suggests that Mongolia’s widening of reverse gender gap in higher education over the past decade can in part be explained by the increasing dependence on traditional agriculture as a form of employment during transition (Mongolia, n.d.) – the point being that males ﬁnd employment in the agricultural sector without higher education degrees, but females cannot. Although more research on the reasons behind the growing reverse gender gap in Mongolian higher education is needed, it is arguably the case that there has been a renegotiation of gender roles in the society, including the institutions of marriage and family.16 Silova and Magno (2004), drawing on social policy discourse from the Soviet era, point out that pre-transition Soviet countries ‘‘spent comparatively large amounts of state budgets on social services to provide the necessary support for women to actively participate in the labor market’’ (p. 418). However, the policy effectively ignored the social and cultural dimensions of the issue – that is, women did household chores in addition to their (required) paid-work (e.g., Deloach & Hoffman, 2002). The pre-transition Socialist Mongolian government was no exception. It looked for equal participation of girls and women in education and work.

Gender Gap and Women’s Participation in Higher Education

235

However, the fact that the state provided kindergartens and day care facilities both in cities and rural communities, as well as support for elderly and disabled care in order to enable women to work outside the home, indicates that these obligations were still regarded as women’s responsibility. The state, indeed, perpetuated traditional gender roles at home. For example, women were rewarded with medals, subsidies, annual vacations, discounts in childcare fees, and other beneﬁts from the Mongolian People’s Republic for having ﬁve or more children.17 In other words, gender roles in the institutions of marriage and family remained more or less intact during the Soviet period. In this context, education was key to providing women with greater career opportunities (and guaranteed jobs by the state), which in turn proved their loyalty and commitment to the Socialist government (for further discussion, see Fix Dominguez, forthcoming). The advent of an open-market economy and higher education expansion in Mongolia arguably opened up more opportunities for girls’ and women’s schooling. However, females might have had to take these educational opportunities to increase their value in the labor market. This is in part because females do not have the same promise of economic security as males, who more often take jobs in the more male-dominated areas of cattle rearing and livestock care. In the process of privatization, assets, including livestock, were registered in the name of heads of household, which were predominantly (90 percent) male (Robinson & Solongo, 2000). Jobs in the state sector (e.g., those in education, healthcare, and social services) have diminished, though it has remained the largest employer in the nation.18 State jobs largely require higher education credentials, as do many private employers in the country (though such white-collar jobs are increasingly difﬁcult to obtain). Although many young women now report dissatisfaction with their job opportunities on completion of a higher education degree, noting that many women must take low-paid, service-industry positions (waiting tables, working in shops) after graduation, a higher education degree still allows them to have access to better jobs (Fix Dominguez, forthcoming). An interesting twist to the reverse gender gap in Mongolian higher education is that marriage and family are still critical institutions, as an (economic) condition of living and source of identity, for Mongolian women, and that those women with tertiary education attainment appear to experience some difﬁculty in ﬁnding husbands with equivalent education (Fix Dominguez, forthcoming). Gender segregation by ﬁelds of study suggests that women may be overrepresented in historically feminized areas, and underrepresented in historically masculinized areas. It is quite possible

236

YOSHIKO NOZAKI ET AL.

that ‘‘the underlying cause of gender-based disparities is gender-based stigma . . . and discrimination . . . emanating from childhood through cultural traditions and social norms followed in the home and the community’’ (UNICEF-Mongolia, 2007, p. 37). However, it also appears to be the case that the ways the society, including women themselves, evaluate their achievements have two axes: one involves their own occupation and career (and education and values associated with these) and the other involves their marriage and family (and education and values associated with these), and that the cause of gender-based disparities, including gender gaps in higher education, is a result of Mongolian women dealing with these two axes through complex balancing and weighing acts. Hence, we can argue that the educational gains made by girls and women do not necessarily signify that there is greater gender equity in Mongolia now, nor do they necessarily support the contention that women are better prepared to participate in the local economy. We call for a critical enquiry that is concerned with larger societal issues – such as family, marriage, and the economy – that inform women’s decisions in the ﬁeld of higher education in Mongolia. This will also help us understand the long-term effects of the transition on gender disparities in higher education and Mongolian society.

THE CASE OF INDIA Higher Education in India: Colonial Origin and Postcolonial Development The Western system of higher education was introduced in India in the mid1880s, during which period the ﬁrst three universities were established in Calcutta, Bombay, and Madras. The development of India’s higher education system was greatly impacted by British imperialism and colonialism, since all indigenous forms of academic scholarship were disavowed in favor of a British style of educational system and thought. After India’s independence in 1947, an additional 18 universities were opened (India, n.d.). Though these universities were coeducational, few women were a part of the higher education system – in 1950–1951, only 40,000 women (or 14 women for every 100 men) were enrolled in university (Chanana, 1993). Since independence, although the governance and organizational structure of India’s higher educational system, set in the mid-19th century,

Gender Gap and Women’s Participation in Higher Education

237

remained more or less the same up to the late 1980s (Altbach, 1987), there have been some milestones. For example, although the federal (central/ union) and state governments have remained responsible for overall planning of higher education, the University Grants Commission (UGC) was established in 1956 to take responsibility for funding, coordinating, monitoring, and maintaining the universities (India, n.d.). Under the control of UGC, all universities are public (run either by state or central government or some combination thereof, or ‘‘deemed universities,’’ sanctioned by the central government on recommendation of the UGC), whereas colleges (all afﬁliated to a public university) may be either public or private (India, n.d.).

Higher Education Expansion Since 1990s India’s higher education system entered a dramatic expansion phase in the 1990s. Some scholars (e.g., Prakash, 2007) attribute this to an increased demand for higher education and participation of the private sector, particularly in technical and professional education. Since the ﬁrst Five Year Plan in 1950,19 the public expenditure on higher education has increased by a whopping 550 times (Prakash, 2007, p. 3255).20 There has been a 13-fold increase in the number of universities and a 26-fold increase in the number of colleges. India now has 17,973 higher education institutions (Government of India, 2007; Chitnis, 1999; India, n.d.). Of these, 1.5 percent are central or state universities, 0.6 percent are deemed universities, and 97.9 percent are colleges. The number of enrollment has also grown from approximately 2.5 million in 1970 to approximately 12 million today (see also Fig. 4). It is crucial to understand the classed and caste-based nature of the recent developments in the Indian higher education system – especially, some of the country’s so called ‘‘self-ﬁnancing’’ private colleges. Although the 1992 judicial judgment banned capitation fee,21 a 1993 judgment paved the way for the growth of the very same type of private colleges under the name of self-ﬁnancing colleges. Today such colleges outnumber public institutions. For instance, in 2001, the state of Andhra Pradesh had 303 self-ﬁnancing medical colleges, compared to 25 government medical colleges (Gupta, 2004; Tilak, 2002). Rekha Kaul (1993), in her study in the state of Karnataka, concludes that the capitation fee phenomenon must be understood as the interplay of caste, class, and power. The colleges are elite enterprises, primarily interested in proﬁt. The dominant, economically

238

YOSHIKO NOZAKI ET AL.

Fig. 4. Tertiary Education Enrollment in India. Source: The Data Accessed through EdStats Query (World Bank Group, 2008).

well-off castes view these colleges as an avenue for building their own professional cadre, thereby leaving lower castes behind. Within such dynamics, educational institutions in India offer three levels of higher education degrees – Bachelors, Masters, and Doctoral. Bachelor degree courses in arts, commerce, and science takes three years, dentistry and engineering four years, and medicine ﬁve years. Masters and doctoral degrees are of two- and ﬁve-year durations, respectively (Government of India, n.d.). These degrees are ﬁnanced by the state (fully funded in public educational institutions and partially funded in private colleges).22 Also and at the same time, despite the vast and ‘‘uncontrollable’’ (Altbach, 1993) expansion of India’s higher education sector, and the relative speed of degree acquisition (typically, a Bachelor’s degree takes 3–4 years to complete, a Master’s degree 1–2 years, and doctoral study 3 years), the country educates only somewhere between 6 and 10 percent of its relevant age-cohorts in the higher educational arena (Altbach, 2005). The worst hit sections are groups that have been historically marginalized – most notably the ‘‘scheduled caste’’ (SC) and ‘‘scheduled tribe’’ (ST) groups. Of the total enrollment in higher education in 1993–1994, only 8.0 percent and 4.3 percent comprised of SCs and STs, respectively (Pinto, 2004). However, other historically marginalized groups under-represented in higher education include ‘‘other backward castes’’ (OBCs),23 religious minorities, and most relevant to the present study, women.

Gender Gap and Women’s Participation in Higher Education

239

Gender Gaps in Student Enrollment India’s enrollment rate for primary education has increased in recent years (and indeed now can compare with other countries with high marks); however, India still faces serious challenges in secondary education enrollment. Indian school enrollment data disaggregated by gender indicates the presence of a gender gap favoring men at every education level, though it is more pronounced at the secondary education level. Gender gaps in GER for primary education was 2.7 (with a male GER 115.9 percent and a female GER 113.2 percent), for secondary education was 10.4 (with a male GER 59.0 percent and a female GER 48.6 percent), and for tertiary education was 3.7 in 2005 (with a male GER 12.8 percent and a female GER 9.1 percent) (see Tables 1–3). Indian women’s enrollment share in tertiary education has signiﬁcantly increased since independence. The share was 10.9 percent in the 1950–1951 school year (Chanana, 2004). It increased to 21.3 percent in 1970 and to 39.4 percent in 2005 (see also Fig. 3). In the 2006–2007 school year, approximately 4.6 million women were enrolled in the higher education system, which constituted 40.4 percent of the total enrollment (Government of India, 2007). This means that, although a small percentage of female population in India who enter primary school makes it to university, as a result of the differentials in the enrollment and dropout of students at different levels of school education, this small segment makes up over 40 percent of the total enrollment rate at university level.24 We now encounter twice as many female students than we did in the 1970s, though this does not mean that gender gaps in higher education have disappeared. Table 3 contains the time series data of the GER of men and women in Indian tertiary education between 1970 and 2005. There has been a persistent GER gender gap (though, interestingly, it has been narrower than Japan’s in these years). There is a trend – the gap was narrowing from 1970s to the mid-1990s (5.2 in 1970, 4.5 in 1980, and 3.1 in 1995), but it has actually slightly grown since then (3.7 in 2005).25 Given the fact that India’s higher education has expanded enormously since 1990s and now with many more women and men enrolled overall, it is perhaps disappointing to see that the disparity between women’s and men’s attendance, as indicated by GERs, appears to be either persistent or, in fact, widening. Certainly, recent Indian higher education expansion needs to be examined through the lens of gender and its intersections with class and caste (Aranha, Fix Dominguez, & Nozaki, 2008; Mazumdar, 1987).

240

YOSHIKO NOZAKI ET AL.

As Chitnis (1989) points out, although the gender gap narrows as one goes up the education ladder, higher education in India has primarily been for middle and upper middle class women, who make up only 3 percent of the total female population. Thus, the gender gap widens among those who fail to have access to the education (Mazumdar, 1987). Therefore, Chanana (1993) urges us to carefully examine the meanings and intersections of gender with class, caste, and geographical region. For example, Akora (2000) ﬁnds that, although males of all groups, including SC and ST groups, are represented in much greater numbers in professional (as opposed to general/academic oriented) higher education, there is a great disparity for SC and ST students. Does this apply to women? It is important to ﬁnd out patterns of women’s representation across India’s higher education system.

Gender Segregation by Fields of Study In 2005, approximately 4.6 million women were enrolled in the Indian higher education system, constituting 39.4 percent of its entire student body (see also Fig. 4 and Table 4). In terms of sheer numbers of female student enrollment, the ﬁeld of Humanities and Arts was the largest, with approximately 1.9 million women enrolled. The second was the ﬁeld of Science, with approximately 0.7 million female students, and the third was the ﬁeld of Social Sciences, Business, and Law, with approximately 0.6 million female students. An examination of women’s enrollment share for each ﬁeld reveals some interesting features of gender dynamics operating in Indian higher education. As shown in Table 4, in 2005, women dominated no ﬁeld of study – in other words, in every ﬁeld of study, men were the majority. However, women’s representation is proportionate and even in most ﬁelds of study (also see Velkoff, 1998). In other words, women are minority in every ﬁeld, but there are lesser signs of gender segregation by the ﬁelds of study, compared to Japan and Mongolia. The range of variations in female enrollment share by ﬁelds of study was only 20.5 percentage points (44.2 23.7). Speciﬁcally, in the same year, the ﬁeld of study that had the largest female enrollment share was Humanities and Arts, in which women constituted 44.2 percent of its student body. The second was the ﬁeld of Education, in which women constituted 43.8 percent of its student body. Since women’s enrollment share in the entire Indian higher education was only 39.4 percent (the lowest among the countries examined here), it is important to look at deviation scores of each ﬁeld, in order to make a comparison. The ﬁeld of Humanities

Gender Gap and Women’s Participation in Higher Education

241

and Arts had only 4.8 percentage points more women (44.2 39.4) and the ﬁeld of Education had 4.4 percentage points more women than the higher education system as a whole. Interestingly, only one ﬁeld of study stood out as the ﬁeld more male dominated than others – the ﬁeld of Engineering, Manufacturing, and Construction, in which female students constituted 23.7 percent of its student body. However, though India’s higher education female enrollment share was low, the ﬁeld had 15.7 percentage point fewer women (23.7 39.4) than the higher education system as a whole. In India, historically, the largest percentage of women has chosen the ﬁelds of Humanities and Arts. Various studies (for instance, see, Chanana, 2000; Chitnis, 1989), based on statistics, interviews, and surveys, report that studying in the ﬁeld of Humanities and Arts was viewed as ‘‘womanly’’ and ‘‘easy’’; women were encouraged to get a humanities degree so as to occupy their time before they get married. Also, the studies report that the women studied often mentioned that they were not permitted to take up careers, unless compelled by circumstances to earn a living; their possible career ambitions were kept masked and in abeyance until they were married and had obtained the ‘‘approval’’ of their in-laws.

Marriage, Work, and Higher Education Attainment There has been some seminal inquiry into the gendered experiences of women within higher educational institutions, family, and marriage in India. For instance, in their work with middle and upper-class, urban, college-going women, scholars such as Blumberg and Dwaraki (1980), Seymour (1994), and Mies (1980) argue that the patriarchal system continues to relegate them to roles in the domestic sphere. They were asked to give highest priority to marriage and the well-being of their families, accept the authority of their husbands, and then think about their education and professional career. Mazumdar (1987) also stresses that the entrance into higher education for those women who had been historically deprived the opportunity (because of purdah and social seclusion) does not necessarily indicate gender equity. She argues that their communities educate women for other social gains – to ﬁnd a suitable groom, and to push them into the labor market in the face of increasing living costs. Drawing from Liddle and Joshi’s (1986) ethnography, Mukhopadhyay and Seymour (1994) assert that, ‘‘male supremacy is no longer maintained through a ban on women’s education, but through controlling the kind, quantity, and purpose of the education’’ (Liddle and Joshi, as cited in Mukhopadhyay & Seymour, 1994, p. 21). Thus,

242

YOSHIKO NOZAKI ET AL.

Mukhopadhyay and Seymour (1994) conclude that ‘‘the degree of systemic change in Indian society that educating women may produce, however, is as of yet unknown’’ (p. 21). Accordingly, Mazumdar (1987) calls for a reexamination of the higher educational system, wherein the system’s gendered nature, objectives, organizational structure and curriculum are closely examined and there is a shift away from an oversimpliﬁed emphasis on women’s entry therein. Chanana (2000) also reminds us to continuously examine the nuanced classed and caste nature of women’s enrollment in speciﬁc disciplines and faculties in the Indian higher education, so as to further our understanding with regard to processes of gendering. For instance, among the already limited female cohort attending university, most women come from upper and middle class backgrounds and it seems that a degree in a ‘‘womanly’’ ﬁeld (such as Humanities and Arts) has had, in essence, cultural commodity of ornamental value they bring into their marriage market (Chanana, 2000). So when the higher education sector further expands, diversiﬁes, and attracts more women, we are urged to pay more close attention to how, in what ways, and to what end do women, both from upper and lower class/caste backgrounds, attend university. Furthermore, within the current context of globalization and the increasing privatization of the Indian economy, we must be even more concerned with critically exploring women’s choices of ﬁelds of study and job market decisions, and what this means to issues of caste, class, and other social relations such as marriage and family. Hence, Heward (1999) asserts that the pre-occupation with the narrowly conceived issue of closing gender-gaps in college enrollments fails to address larger questions of feminist research: Do these enrollment rates shed light on the nature of gender formations and politics in colleges today? What are the experiences and perspectives of Indian women in higher education within the context of the rise of conservative nationalist tides and globalization of the Indian economy? How do these experiences inform their identities? Statistical data examined in the present chapter cannot directly answer these questions, but they provide broader contexts for studies of qualitative nature (Aranha, forthcoming), in which we can examine Indian women’s voices, experiences, and perspectives.

CONCLUDING THOUGHTS Although research has suggested that gender gaps in education have been narrowed in recent years, such research has primarily been concerned with primary and secondary education (and often focused on ‘‘developing’’

Gender Gap and Women’s Participation in Higher Education

243

countries). The speciﬁc trends of gender gaps in higher education enrollment in Asian nations have been somewhat understudied, especially through international comparative approaches. This chapter has examined the higher education enrollment trends of three Asian countries – Japan, Mongolia, and India – using the notion of the gender gap. Although our three country case studies have by no means exhausted every aspect of the issues involved, we brieﬂy discuss some of the ﬁndings, implications, and questions for further studies. In terms of the GER of each country, even though the country has achieved universal secondary education for both boys and girls, Japan still has a small gender gap. This raises a question of why Japanese women do not attempt to attain the same levels of tertiary education as their male counterparts. In Mongolia, there has been a reverse gender gap phenomenon throughout the entire education system, but it is more salient at the higher education level (at least in terms of its GER). Although scholars may be urged to examine men’s decisions (e.g., leaving education to take jobs in the livestock business), if we are interested in the meaning of reverse gap from a gender perspective, we should make further inquiry into why women tend to keep their educational credentials high (or higher than men). In India, the number of women participating in higher education has increased greatly; however, the gender gap in GER remains very persistent. This is in part because the population has grown, and in part because girls’ participation in and completion of secondary education is still laden with obstacles. Our analysis of the gender composition of each ﬁeld of study reveals that there are signs of gender segregation by ﬁelds of study in all three countries, though to different degrees. The country where gender segregation is most pronounced is Japan, seconded by Mongolia. In India, where the percentage of women enrolled in higher education is still small, the gender distribution by ﬁelds of study is proportionate except for certain historically femaledominated ﬁelds of study (such as the Humanities and Arts). As far as we can assume that there is a connection between women’s choices for ﬁelds of study in higher education and their (anticipated) choices in the labor and marriage markets, it is reasonable to suggest that women’s decision-making concerning higher education attainment generally, and ﬁeld of study in particular, turn on two axes: one involves labor market outcomes, and the other involves marriage and family. These two axes are in greater conﬂict for women than for men (for whom the axis of jobs and careers is more dominant) (for further discussion, see Nozaki, forthcoming; also see Hara & Seiyama, 1999). Insofar as gender relations and politics in higher education have an impact on, and are impacted by, societal norms and practices, we can postulate a relationship between a narrowing gender gap and greater gender

244

YOSHIKO NOZAKI ET AL.

equality. However, it is problematic if the shrinking gender gap in higher education is used to deﬂect attention from unequal gender relations. For example, as discussed in the section on Mongolia, the notion of a reverse gender gap is problematic because it may suggest that the country is close to achieving gender equality. In fact, our research suggests that such an assumption can be faulty, and calls for us to once again examine the connections of Mongolian women’s aspiration for and attainment in higher education with their (economic) conditions of living and their gender identities and politics. The conceptualization of a reverse gender gap may obscure the socio-historical context of Mongolian society, and keep researchers from questioning the larger societal arrangements of gendered institutions and markets of economic opportunities, marriage, and family. Finally, this chapter has shown some of the limitations of using the notion of the higher education gender gap as a lens for assessing the degree of equal, or unequal, gender relations socially, culturally, and structurally. It is important to examine career choices and trajectories, marriage opportunities, and other impacts related to women’s graduation from higher education institutions. However, unless and until we expand the scope of the empirical data, both quantitative and qualitative, to understand the relationships between gender and higher education on the one hand and between and among women’s higher education attainment, the labor market, and the institutions of marriage and family on the other hand, we will not have a robust picture of the perpetuation of unequal gender relations and gender identity conﬂict by and through colleges and universities. Moreover, societal gender inequalities may or may not be perpetuated directly by colleges and universities; it could also happen through policy, discursive practices of governments, and multilateral organizations. We have drawn insights from the ﬁelds of comparative education, Asian studies, and feminist research, as it is critical to have perspectives broader than traditional higher education concerns. In other words, an interdisciplinary study is essential to the growing scholarship on gender and comparative higher education.

NOTES 1. We would like to acknowledge the ongoing debates over the terms equality, equity, and parity in education. Some scholars argue that ‘‘equality’’ refers to the numerical distribution of goods or services, whereas ‘‘equity’’ refers to judgments concerning the fairness or justice of that distribution (Bronfenbrenner, 1973). In this

Gender Gap and Women’s Participation in Higher Education

245

sense, ‘‘parity’’ is more closely aligned with equality than equity; UNESCO deﬁnes gender parity in education as access to school enrollment for equal numbers of males and females (UNESCO, 2003). In this chapter, we, in principle, follow these distinctions. 2. There can be another explanation to suggest overt gender discrimination in the higher education; however, this may not be the strongest explanation today, as the legal frameworks and university policies have been set to ensure equal treatments for men and women. 3. In addition to the three nations examined here, the team has conducted preliminary studies of several countries, including China, Mexico, Argentine, Latvia, and Israel (see, e.g., chapters in Nozaki & Fix, 2008). 4. These additional universities were established in Kyoto, Sendai, Fukuoka, and Sapporo. 5. Although these institutions ﬁrst employed western scholars as professors, who taught classes in western languages, they were eventually replaced with Japanese qualiﬁed to teach at these levels of higher education (Amano, 1979). 6. Eligibility requirements for these institutions varied. Higher schools provided general education with high standards; however, they mainly functioned as preparatory schools for university entrance (Narita, 1978). Technical schools offered advanced training in certain areas such as law, medicine, foreign language, commerce, agriculture, and engineering (Amano, 1979; Ishida, 2007), and teachers’ colleges offered teacher training. These technical schools and teachers’ colleges were ranked below universities (Ishida, 2007) or as a part of secondary education (Hara, 2000). 7. Before this change, some technical schools of the old system had already been merged into universities or accredited as junior colleges (Ishida, 2007). 8. In addition to 87 national universities, public universities included 89 prefectural/city universities, run by municipal governments. 9. GER is the number of students enrolled in a level of education, regardless of age, as expressed as a percentage of the eligible ofﬁcial school-age population in the relevant age group for that level in a given school year. Net Enrollment Rate (NER) is the number of students enrolled in a level of education who belong in the relevant age group, as a percentage of the population in that age group. GER is widely used to show the general level of participation in a given level of education. It indicates the capacity of the education system to enroll students of a particular age group. It is used as a substitute indicator to net enrollment rate when data on enrollment by single years of age are not available. Furthermore, it can also be a complementary indicator to NER, by indicating the extent of over-aged and under-aged enrollment. As such, it indicates the capacity of each level of the education system. NER provides a better indicator of a school system’s efﬁciency. 10. Scholars in Japan have discussed the issue of gender and higher education (e.g., Amano, 1988; Fujimura-Fanselow, 1985). However, fewer studies have utilized internationally comparative data. 11. The Japanese government, which has fairly tightly regulated tertiary education over the years, has made a clear distinction between non-degree granting postsecondary institutions (i.e., specialized training schools) and degree-granting institutions (i.e., universities, colleges, junior colleges, and colleges of technology), and it has focused on the ‘‘advancement rate’’ of relevant age cohorts, because students of non-traditional college ages have been rare. According to governmental

246

YOSHIKO NOZAKI ET AL.

statistics, in terms of the students of relevant age cohorts entering postsecondary education as a whole, Japan has achieved gender parity since the 1980s and has observed a slight reverse gender gap since the late 1980s to the mid-2000s (Nozaki, Fix, & Nakajima, 2008). In terms of the students advancing to degree-granting institutions, the nation is very close to the parity point (53.6 percent for males and 51.0 percent for females, as of 2006) (Japanese MOE, n.d.[d]). Perhaps because of these widely available governmental statistics, few Japanese researchers have discussed the gender gaps in GER. 12. To compare distribution of women across various ﬁelds of study in three countries, we examine the data collected and organized by UNESCO Institute for Statistics (UIS). The UIS uses International Standard Classiﬁcation of Education (ISCE) (UNESCO, 2006) that classify various ﬁelds of study into eight (large) ﬁelds: ‘‘Education,’’ ‘‘Humanities and Arts,’’ ‘‘Social Sciences, Business, and Law,’’ ‘‘Science,’’ ‘‘Engineering, Manufacturing, and Construction,’’ ‘‘Agriculture,’’ ‘‘Health and Welfare,’’ and ‘‘Services’’ (so each ﬁeld consists of a number of subﬁelds). In addition to these eight ﬁelds, there is a category ‘‘Unspeciﬁed,’’ created for data collection purpose. 13. Ishida (1998) notes that getting a full-time/permanent job in a large Japanese ﬁrm constitutes a considerable socio-economic advantage. 14. The ﬁeld of Service includes several professions that has historically been male-dominated in Mongolia, such as military service, law enforcement, land management, and air trafﬁc control. More research is needed to understand the gender dynamics at work in the Service ﬁeld. 15. This is a collection of data on ﬁelds of study at the NUM. Data on ﬁelds of study at NUM does not include areas such as education and nursing because these are applied ﬁelds that are taught at the Medical and Pedagogical Universities of Mongolia (MPUMs). NUM and the MPUMs are part of the same public, governmental system but are separate entities (also see Fix Dominguez, forthcoming). 16. Kenway and Kelly (2002) discuss similar issues in the context of Australian vocational education, with notions of ‘‘de-traditianlization’’ and ‘‘re-traditionalization’’ of gender roles. 17. These incentives were particularly important for nomadic women, who were not technically working outside the home, but could still quantify their contribution to the Socialist government in terms of their maternal role (Fix Dominguez, forthcoming). 18. More people worked in the state sector than the private sector (51.7 percent) and women constituted 55.8 percent of those employed in the state sector and 46.2 percent of those in the private sector at the end of 1990s (Women’s Information and Research Centre, 1998, cited in Robinson & Solongo, 2000). 19. The Indian Planning Commission (IPC), which includes the Prime Minister and a Deputy Chairman who has the rank of a Cabinet minister, bases its economic decision-making on Five-Year Plans, which include economic dimensions and strategies, sectoral policies and programs (including higher education), and state plans. These plans are developed, executed, and monitored by the IPC. The Tenth Five-Year Plan was completed in March 2007.

Gender Gap and Women’s Participation in Higher Education

247

20. The increase was from a modest level of 171.5 million rupees in 1950–1955 to 95,620 million rupees in 2004–2005 (budget estimates). Rising inﬂation, however, makes this increase an illusion. To get a realistic picture, one may have to look at trends in public expenditure adjusted for inﬂation. After adjusting public expenditure both on higher and technical education for inﬂation with national income deﬂators, the annual growth rate turns out to be just 5.4 percent and 5.2 percent, respectively (Prakash, 2007). 21. The 1992 judgment stated that a capitation fee (fees that include tuition and a compulsory contribution toward a ﬁxed capital fund of the college) was ‘‘patently unreasonable, unfair and unjust.’’ 22. Though there has been a sizeable increase in the overall ﬁnancial expenditure in India’s Five-Year plans for the education sector, as a proportion of the national income, it is less than 4 percent. Within this expenditure, higher education seems to be a neglected sector in India. In the Tenth Five-Year Plan (2002–2007), it has a mere share of 10 percent of the total education fund (Higher education, 2005). 23. OBCs are castes in the Indian social system, which are situated above the Untouchables but below the forward castes (the ‘‘twice born,’’ Brahmins, Kshatriyas [warriors], and Vaishyas [merchants]) and the intermediate castes (mostly peasant proprietors and even dominant castes). They form the bulk of the Shudras – the fourth category (varna) of the classical Hindu social arrangement. The OBCs, whose professional activity is often as ﬁeld-workers or artisans, represent about half of the Indian population, but they have occupied a subaltern position so far (Jaffrelot, 2000). 24. Chitnis (1989) calculates, based on her research, that barely 2 percent of the female population in India who enter primary school makes it to university. 25. The gap was 3.8 in 2006.

ACKNOWLEDGMENTS The authors are grateful to the Baldy Center for Law and Social Policy, University at Buffalo (State University of New York), for its support to this study at the critical stage of manuscript preparation. We thank the reviewers of this chapter for their useful suggestions. We also thank all the participants in both Gender Equality and Higher Education (GEHE) Project and the Tuesday-Saturday (TS) Group for their contributions to the project and their insights shared at the TS group meetings.

REFERENCES Akora, A. (2000). The increasing costs of higher education to parents and students in India: A policy study on cost recovery and access. Doctoral dissertation, State University of New York at Buffalo. Dissertation Abstracts International, 9967784.

248

YOSHIKO NOZAKI ET AL.

Altbach, P. G. (1987). Higher education in the Third World: Themes and variations. New York: Advent Books. Altbach, P. G. (1993). The dilemma of change in Indian higher education. In: P. G. Altbach & S. Chitnis (Eds), Higher education reform in India: Experiences and perspectives (pp. 13–37). New Delhi, India: Sage. Altbach, P. G. (2005, April 2). Higher education in India. The Hindu. Available at http:// www.hindu.com/2005/04/12/stories/2005041204141000.htm. Retrieved on April 17, 2005. Amano, I. (1979). Continuity and change in the structure of Japanese higher education. In: W. K. Cummings, I. Amano & K. Kitamura (Eds), Changes in the Japanese university (pp. 10–39). New York: Praeger Publishers. Amano, I., & Poole, G. S. (2005). The Japanese university in crisis. Higher Education, 50, 685–711. Amano, M. (Ed.) (1986). Joshi koto kyoiku no zahyo [Coordinate of higher education for women]. Tokyo: Kakiuchi-Shuppan Press. Amano, M. (1988). ‘Sei (jenda) to kyoiku’ kenkyu no gendaiteki kadai: Kakusareta ‘ryoiki’ no jizoku [Contemporary issues in the study of gender and education: Persistence of hidden ﬁelds]. Shakaigaku Hyoron, 39(3), 266–283. Aranha, R. (forthcoming). (Re)inventing nation: Hindutva, gender, and student politics in India. Unpublished doctoral dissertation, State University of New York at Buffalo. Aranha, R., Fix Dominguez, R., & Nozaki, Y. (2008). Gender and higher education: The Indian case. In: Y. Nozaki & R. Fix Dominguez (Eds), Gender equality and higher education in Asia and beyond (pp. 29–38). Buffalo, NY: State University of New York at Buffalo Graduate School of Education & State University of New York Press. Arnot, M. (2002). Reproducing gender? Essays on educational theory and feminist politics. New York: Routledge-Falmer. Arnot, M., David, M., & Weiner, G. (1999). Closing the gender gap: Postwar education and social change. Cambridge, UK: Polity Press. Blumberg, R. L., & Dwaraki, L. (1980). India’s educated women: Options and constraints. New Delhi, India: Hindustan Publishing Corporation. Bradley, K. (2000). The incorporation of women into higher education: Paradoxical outcomes? Sociology of Education, 73, 1–18. Brinton, M. C. (1993). Women and the economic miracle: Gender and work in postwar Japan. Berkley, CA: University of California Press. Bronfenbrenner, M. (1973). Equality and equity. Annals of the American Academy of Political and Social Science, 409, 5–25. Brooke, J. (2005, February 15). For Mongolians, E is for English, F is for future. New York Times. Available at http://www.nytimes.com. Retrieved on August 31, 2007. Chanana, K. (1993). Accessing higher education: The dilemma of schooling women, minorities, scheduled castes and scheduled tribes in contemporary India. Higher Education, 26(1), 69–92. Chanana, K. (2000, March 18). Treading the hallowed halls: Women in higher education in India. Economic and Political Weekly, 1012–1022. Chanana, K. (2004). Gender and disciplinary choices: Women in higher education in India. Available at http://portal.unesco.org/education/en/ev.php-URL_ID ¼ 36528& URL_DO ¼ DO_TOPIC&URL_SECTION ¼ 201.html. Retrieved on January 22, 2008. Charles, M., & Bradley, K. (2002). Equal but separate? A cross-national study of sex segregation in higher education. American Sociological Review, 67, 573–599.

Gender Gap and Women’s Participation in Higher Education

249

Chitnis, S. (1989). India. In: G. P. Kelly (Ed.), International handbook of women’s education (pp. 135–162). New York: Greenwood Press. Chitnis, S. (1999). The transformation of an imperial colony into an advanced nation: India in comparative perspective. In: P. G. Altbach & P. McGill Peterson (Eds), Higher education in the 21st century: Global challenge and national response (IIE research report No. 29). Annapolis Junction, MD: Institute of International Education. Conway, J. K., & Bourque, S. C. (Eds). (1993). The politics of women’s education: Perspectives from Asia, Africa, and Latin America. Ann Arbor: The University of Michigan Press. Deloach, S. B., & Hoffman, A. L. (2002). Russia’s second shift: Is housework hurting women’s wages? Atlantic Economic Journal, 30(4), 422–432. Eades, J. S., Goodman, R., & Hada, Y. (Eds). (2005). The ‘big bang’ in Japanese higher education: The 2004 reforms and the dynamics of change. Melbourne, Australia: Trans Paciﬁc Press. Fix Dominguez, R. (2008). Gender and higher education in Mongolia: The reverse gender gap and the de- and re-traditionalization of gender. In: Y. Nozaki & R. Fix Dominguez (Eds), Gender equality and higher education in Asia and beyond (pp. 39–46). Buffalo, NY: State University of New York at Buffalo Graduate School of Education & State University of New York Press. Fix Dominguez, R. (forthcoming). Higher education in transition: Gender, urban/rural issues and the changing role of the Mongolian state. Unpublished doctoral dissertation, State University of New York at Buffalo. Fujimura-Fanselow, K. (1985). Women’s participation in higher education in Japan. Comparative Education Review, 29(4), 471–489. Goldin, C., Katz, L. F., & Kuziemko, I. (2006). The homecoming of American college women: The reversal of the college gender gap. Working Paper Series, 12139. National Bureau of Economic Research. Government of India. (2007). Annual report: 2006–2007. Available at http://www.education. nic.in/secondary.htm. Retrieved on September 2, 2007. Government of India. (n.d.). Information. Available at http://www.education.nic.in. Retrieved on April 20, 2005. Government of Mongolia. (2003). Economic growth support and poverty reduction strategy. Available at http://povlibrary.worldbank.org/library/country/117/?offset ¼ 10. Retrieved on August 31, 2007. Gupta, A. (2004, Spring). Divided government and private higher education growth in India. International Higher Education. Available at http://www.bc.edu/bc_org/avp/soe/cihe/ newsletter/subject_index.htm. Retrieved on April 10, 2005. Hara, J. (2000). Kindai sangyo shakai nihon no kaiso shisutemu [Stratiﬁcation system in modern industrial society Japan]. In: J. Hara (Ed.), Kindaika to shakai kaiso: Vol. 1. Nihon no kaiso shisutemu (pp. 3–43). Tokyo: Daigaku Shuppankai. Hara, J., & Seiyama, K. (1999). Shakai kaiso: Yutakasa no naka no fubyodo [Social stratiﬁcation: Inequality in wealth]. Tokyo: Daigaku Shuppankai. Heward, C. (1999). Introduction: The discourses of gender, education, and development. In: C. Heward & S. Bunwaree (Eds), Gender, education, and development: Beyond access to empowerment (pp. 1–14). New York: St. Martin’s Press. Higher education: Rethink required. (2005). Available at http://www.ﬁnancialexpress.com/ news/Higher-education-rethink-required/119316/3. Retrieved on August 27, 2008.

250

YOSHIKO NOZAKI ET AL.

Imada, S. (1985). Josei no shokugyo keireki to kyoiku tassei: Raihu hisutori apurochi kara [Women’s work-career and educational achievement: Seen from the approach of lifehistory]. Kyoiku Shakaigaku Kenkyu, 76, 50–64. India. (n.d.). Available at http://www.gse.buffalo.edu/org/IntHigherEdFinance/. Retrieved on September 3, 2008, from State University of New York at Buffalo, The International Comparative Higher Education Finance and Accessibility Project. Ishida, H. (1998). Educational credentials and labour-market entry outcomes in Japan. In: Y. Shavit & W. Muller (Eds), From school to work: A comparative study of educational qualiﬁcations and occupational destinations (pp. 287–335). Oxford, UK: Clarendon Press. Ishida, H. (2007). Japan: Educational expansion and inequality in access to higher education. In: Y. Shavit, R. Arum & A. Gamoran (Eds), Stratiﬁcation in higher education: A comparative study (pp. 63–86). CA: Stanford University Press. Jacob, B. A. (2002). Where the boys aren’t: Non-cognitive skills, returns to school and the gender gap in higher education. Working Paper Series, 8964. National Bureau of Economic Research. Jaffrelot, C. (2000). The rise of the other backward classes in the Hindi belt. The Journal of Asian Studies, 59(1), 86–108. Japan. (n.d.). Available at http://www.gse.buffalo.edu/org/IntHigherEdFinance/. Retrieved on September 3, 2008, from State University of New York at Buffalo, The International Comparative Higher Education Finance and Accessibility Project. Japanese Gender Equality Bureau, Cabinet Ofﬁce. (2008). Danjo kyodo sankaku hakusho: Heisei 20 nen ban [Annual report on the state of the formation of a gender equal society and policies to be implemented to promote the formation of a gender equal society]. Available at http://www.gender.go.jp/whitepaper/h20/gaiyou/html/honpen/index.html. Retrieved on July 31, 2008. Japanese Ministry of Education, Culture, Sports, Science, and Technology. (n.d.[a]). Gakko kihon chosa heisei 18 nendo: Sokatsu [2006 basic statistical data on schools in 2006: Overview]. Available at http://www.mext.go.jp/b_menu/toukei/001/06121219/004.htm. Retrieved on July 30, 2008. Japanese Ministry of Education, Culture, Sports, Science, and Technology. (n.d.[b]). Shiritsu gakko no shinko [Promoting the development of private schools]. Available at http:// www.mext.go.jp/a_menu/koutou/shinkou/main5_a3.htm. Retrieved on January 3, 2008. Japanese Ministry of Education, Culture, Sports, Science, and Technology. (n.d.[c]). Students by ﬁeld of study (University)(Master’s Courses)(Doctor’s Courses). Available at http:// www.mext.go.jp/english/statist/index11.htm. Retrieved on January 6, 2008. Japanese Ministry of Education, Culture, Sports, Science, and Technology. (n.d.[d]). Enrollment and advancement rate, 1948 to 2005 (Continued). Available at http:// www.mext.go.jp/english/statist/index01.htm. Retrieved on January 6, 2008. Kameda, A. (1995). Sexism and gender stereotyping in schools. In: K. Fujimura-Fanselow & A. Kameda (Eds), Japanese women: New feminist perspectives on the past, present, and future (pp. 107–124). New York: The Feminist Press at the City University of New York. Kaul, R. (1993). Caste, class, and education: Politics of the capitation fee phenomenon in Karnataka. New Delhi, India: Sage. Kelly, G. P., & Slaughter, S. (1991). Women’s higher education in comparative perspective. Dordrecht, The Netherlands: Kluwer Academic Publishers. Kenway, J., & Kelly, P. (2002). Local/global labor markets and the restructuring of gender, schooling and work. In: N. Stromquist & K. Monkman (Eds), Globalization and

Gender Gap and Women’s Participation in Higher Education

251

education: Integration and contestation across cultures (pp. 173–195). Lanham, MD: Rowan and Littleﬁeld. Kimura, R. (1997). Kyoshitsu ni okeru jenda keisei [Constructing gender in the classroom]. Kyoiku Shakaigaku Kenkyu, 61, 39–53. King, J. E. (2006). Gender equity in higher education: 2006. Washington, DC: American Council on Education, Center for Policy Analysis. Available at http://store.acenet.edu/showMultiple Items.aspx?session ¼ DD30F605046346789. Retrieved on September 3, 2008. Liddle, J., & Joshi, R. (1986). Daughters of independence: Gender, caste and class in India. New Delhi, India: Kali for Women. Lie, S. S., Malik, L., & Harris, D. (Eds). (1994). The gender gap in higher education. London: Kogan Page. Manabe, R. (1997). Josei no kyoiku tassei to shokugyo tassei [Women’s job achievement and educational achievement]. Kyoiku Shakaigaku Kenkyu, 60, 83–98. Matsui, M. (1997). Tandai wa dokoeyuku [Where are junior colleges going?]. Tokyo: Keiso Shobo. Mazumdar, V. (1987). Education, development, and women’s liberation: Contemporary debates in India. In: R. Ghosh & M. Zachariah (Eds), Education and the process of change (pp. 198–211). New Delhi, India: Sage. McVeigh, B. J. (1997). Life in a Japanese women’s college: Learning to be ladylike. New York: Routledge. McVeigh, B. J. (2002). Japanese higher education as myth. Armonk, NY: M. E. Sharpe. Mies, M. (1980). Indian women and patriarchy: Conﬂicts and dilemmas of students and working women. New Delhi, India: Concept. Miyazaki, A. (1991). Gakko ni okeru ‘‘Sei yakuwari no shakaika’’ saiko: Kyoshi ni yoru seibetsu kategori shiyo wo tegakari toshite [Reconsideration of ‘‘gender role socialization’’: Sex categorization]. Kyoiku Shakaigaku Kenkyu, 48, 105–123. Mongolia. (n.d.). Available at http://www.gse.buffalo.edu/org/IntHigherEdFinance/. Retrieved on September 3, 2008, from State University of New York at Buffalo, The International Comparative Higher Education Finance and Accessibility Project. Mori, S. (1989). Sei yakuwari no gakushu to shitsuke koui [Learning gender role and discipline]. In: S. Shibano (Ed.), Shitsuke no shakaigaku: Shakaika to shakai tosei (pp. 155–171). Tokyo: Sekai Shisho Sha. Morozumi, A. (2005). Koto kyoiku no shijoka to nihon no shiritsu daigaku [The marketization of higher education and private universities in Japan]. Crump Working Paper Series, 8. Available at http://daikei.p.u-tokyo.ac.jp/index.php?Publications. Retrieved on July 30, 2008, from University of Tokyo, Center for Research of University Management and Policy. Mortenson, T. G. (1999). Where are the boys? The growing gender gap in higher education. The College Board Review, 188, 8–17. Mukhopadhyay, C., & Seymour, S. (1994). Introduction and theoretical overview. In: C. Mukhopadhyay & S. Seymour (Eds), Women, education, and family structure in India (pp. 1–33). Boulder, CO: Westview Press. Murakami, E. (2003). Kenkyusha yosei katei ni okeru kenryokusei to sei sabetsu [The power and the gender discrimination in the training process of academics]. Gijutsu Manejimento Kenkyu, 2(1), 14–26. Nagasawa, M. (2005, Summer). Gender stratiﬁcation in Japanese private higher education. International Higher Education, 40. Available at http://www.bc.edu/bc_org/avp/soe/cihe/ newsletter/Number40/p10_Nagasawa.htm. Retrieved on January 1, 2008.

252

YOSHIKO NOZAKI ET AL.

Nakanishi, Y. (1992). Jenda torakku: Sei yakuwari kan ni motozuku shinro bunka mekanizumu ni kansuru kosatsu [Gender track: Analysis on the mechanism of tracking based on the views of gender role]. Kyoiku Shakaigaku Kenkyu, 53, 131–154. Narita, K. (1978). Systems of higher education: Japan. New York: International Council for Educational Development. National University of Mongolia (NUM). (2004). Suraltsagchid: Mergejil, angi, kurseer, 2003–2004 [Students: Profession, class, course, 2003–2004]. Ulaanbaatar, Mongolia: Mongol Ulsiin Ikh Sorguul. National Women’s Education Center, Japan. (n.d.). Gender statistics database: Students by ﬁeld of study and sex (university undergraduate course). Available at http:// winet.nwec.jp/cgi-bin/toukei/load/bin/tk_sql.cgi?bunya ¼ 06&hno ¼ 0&rfrom ¼ 1&rto ¼ 0&fopt ¼ 1. Retrieved on February 23, 2008. Neuman, W. L. (2006). Social research methods: Qualitative and quantitative approaches (6th ed.). Boston: Pearson Education, Inc. Nozaki, Y. (forthcoming). A comparative examination of economy, higher education enrollments, and labor market outcomes through a gender lens. Comparative Minds, Critical Visions: Working Paper Series at the Center for Comparative and Global Studies in Education, 4. Nozaki, Y., & Fix Dominguez, R. (Eds). (2008). Gender equality and higher education in Asia and beyond. Buffalo, NY: State University of New York at Buffalo Graduate School of Education & State University of New York Press. Nozaki, Y., Fix Dominguez, R., & Nakajima, Y. (2008). Gender parity, the reverse gap, and higher education: Key issues from Japan and the United States. In: Y. Nozaki & R. Fix Dominguez (Eds), Gender equality and higher education in Asia and beyond (pp. 8–21). Buffalo, NY: State University of New York at Buffalo Graduate School of Education & State University of New York Press. Organization for Economic Co-operation and Development. (2008). OECD Employment outlook 2008. Available at http://www.oecd.org/document/46/0,3343,en_2649_33927_ 40401454_1_1_1_37457,00.html. Retrieved on September 3, 2008. Pinto, A. (2004). Higher education, ideology and politics. Bangalore, India: Books for Change. Prakash, V. (2007, August 4). Trends in growth and ﬁnancing of higher education in India. Economic and Political Weekly, 3249–3258. Robinson, B., & Solongo, A. (2000). The gender dimension of economic transition in Mongolia. In: F. Nixson, B. Suvd, P. Luvsandorj & B. Walters (Eds), The Mongolian economy: A manual of applied economics for a country in transition (pp. 231–255). Northampton, MA: Edward Elger Publishing. Rodo Seisaku Kenkyu Kenshu Kiko. (2006). Daigakusei no shushoku boshu katsudo to jittai chosa kekka II [Results of the survey on job hunting, recruitment and adoption of university students]. JILPT Research Series No. 17. Tokyo: Rodo Seisaku Kenkyu Kenshu Kiko. Rohlen, T. P. (1984). Seishin-kyoiku in a Japanese bank: A description of methods and consideration of some underlying concepts. Anthropology and Educational Quarterly, 15(1), 17–28. Science Council of Japan. (2005). Josei Kenkyusha no Kanten kara Mita Daigakuin Kyoiku no Mondaiten [Issues on graduate education from the viewpoint of women’s researchers]. Available at http://www.scj.go.jp/ja/info/kohyo/2005.html. Retrieved on August 2, 2008.

Gender Gap and Women’s Participation in Higher Education

253

Seymour, S. (1994). College women’s aspirations: A challenge to the patrifocal family system? In: C. Mukhopadhyay & S. Seymour (Eds), Women, education, and family structure in India (pp. 213–233). Boulder, CO: Westview Press. Shabaya, J., & Konadu-Agyemang, K. (2004). Unequal access, unequal participation: Some spatial and socio-economic dimensions of the gender gap in education in Africa with special reference to Ghana, Zimbabwe and Kenya. Compare: A Journal of Comparative Education, 34(4), 395–424. Shavit, Y., Arum, R., & Gamoran, A. (Eds). (2007). Stratiﬁcation in higher education: A comparative study. CA: Stanford University Press. Silova, I., & Magno, C. (2004). Gender equity unmasked: Democracy, gender, and education in Central/Southeastern Europe and the Former Soviet Union. Comparative Education Review, 48(4), 417–442. Tilak, J. (2002, Fall). Privatization of higher education in India. International Higher Education. Available at http://www.bc.edu/bc_org/avp/soe/cihe/newsletter/News29/Newslet29. html. Retrieved on August 27, 2008. Ujihara, Y. (1996). Chugakko ni okeru danjo byodo to sei sabetsu no sakuso: Futatsu no ‘‘Kakureta karikyuramu’’ reberu kara [The complexity of gender equality and sexism in junior high school: Two "hidden curriculum" levels]. Kyoiku Shakaigaku Kenkyu, 58, 29–45. United Nations Children’s Fund. (2005). Progress for children: A report card on gender parity and primary education. New York: UNICEF. United Nations Educational, Scientiﬁc, and Cultural Organization. (2003). What is gender parity? Available at http://portal.unesco.org/education/en/ev.php-URL_ID ¼ 14090 &URL_DO ¼ DO_TOPIC&URL_SECTION ¼ 201.html. Retrieved on January 25, 2008. United Nations Educational, Scientiﬁc, and Cultural Organization. (2006). ISCED 1997: International standard classiﬁcation of education (Re-editon). Available at http:// www.uis.unesco.org/ev.php?ID ¼ 3813_201&ID2 ¼ DO_TOPIC. Retrieved on August 30, 2008, from UNESCO Institute for Statistics. United Nations Educational, Scientiﬁc, and Cultural Organization (UNESCO) Institute for Statistics. (2008). Data Centre [Data ﬁle]. Available at http://stats.uis.unesco.org/. Retrieved from UNESCO Institute for Statistics. United Nations International Children’s Emergency Fund. (2004). The state of the world’s children 2004. New York: UNICEF. Available at http://www.unicef.org/sowc04/. Retrieved on August 2, 2008. United Nations International Children’s Emergency Fund. (2005). Progress for children: A report card on gender parity and primary education. New York: UNICEF. United Nations International Children’s Emergency Fund-Mongolia. (2007). Situation analysis of children and women in Mongolia. Ulaanbaatar, Mongolia: UNICEF-Mongolia. Available at http://www.unicef.org/mongolia/. Retrieved on August 2, 2008. University of Tokyo. (2008). History. Available at http://www.u-tokyo.ac.jp/gen03/b03_02_e. html. Retrieved on July 19, 2008. Velkoff, V. (1998). Women’s education in India (Series WID 98–1, 1998). The U.S. Bureau of the Census, International Programs Center. Available at http://www.census.gov/ipc/www/ publist.html#PUB. Retrieved on September 4, 2008.

254

YOSHIKO NOZAKI ET AL.

Vincent-Lancrin, S. (2008). The reversal of gender inequalities in higher education: An on-going trend. In: Center for Educational Research and Innovation (Ed.) Higher education to 2030, Volume 1: Demography (pp. 265–298). Paris: OECD Publishing. Weidman, J. C. (2002). Developing the Mongolian education sector strategy 2000–2005: Reﬂections of a consultant for the Asian development bank. Current Issues in Comparative Education, 3(2), 99–108. Weidman, J. C., Bat-Erdene, R., Yeager, J. L., Sukhbaatar, J., Jargalmaa, T., & Davaa, S. (1998). Mongolian higher education in transition: Planning and responding under conditions of rapid change. Tertium Comparationis, 4(2), 75–90. Woodﬁeld, R., & Earl-Novell, S. (2006). An assessment of the extent to which subject variation between the Arts and Sciences in relation to the award of a ﬁrst class degree can explain the ‘‘gender gap’’ in U.K. universities. British Journal of Sociology of Education, 27(3), 355–372. Worden, R. L., & Savada, A. M. (1989). Mongolia: A country study. Washington, DC: GPO for the Library of Congress. World Bank Group. (2008). EdStats Query [Data ﬁle]. Available at http://www.worldbank.org/ Yoshihara, K. (1995). Joshi daigakusei ni okeru shokugyo sentaku no mekanizumu: josei nai bunka no yoin toshiteno josei sei [Mechanism of career choice among woman students in universities: Femininity as the factor for differentiation among women]. Kyoiku Shakaigaku Kenkyu, 57, 107–124.

RE-GENDERED EDUCATION AND SOCIETY IN THE NEWLY INDEPENDENT STATES (NIS) OF CENTRAL ASIA Alan J. DeYoung and Elizabeth A. Constantine Women in the former Soviet states of Central Asia (CA) – Kyrgyzstan, Kazakhstan, Tajikistan, Turkmenistan and Uzbekistan – have been losing substantial ground in terms of gender equality. Emerging gender inequities in education are an essential part of this process, as well as contributing factors to its long-term consequences. Overall school quality, too, is in serious decline – which affects life chances for both girls and boys. During the Soviet period, education was an instrument for social and cultural reforms. When the Soviets came to power, they systematically utilized educational institutions in an attempt to undermine traditional patterns of social life in the region and to change the status of women. The ‘‘emancipation’’ of CA women arguably had more to do with the implementation of the Soviet political and economic project than it did with constituting an act of altruism (Corcoran-Nantes, 2005, p. 38). Nevertheless, women’s equal rights and responsibilities were enshrined in the constitution of the USSR and in national legislation, whereby the government proclaimed that men and women were to be equal in all aspects of economic, social and political life. The ﬁrst Soviet education law in 1918 established the legal right to education at public expense, and thus began Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 255–299 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010011

255

256

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

CA society’s introduction to the diffusion of secular values, scientiﬁctechnological knowledge and socialist economic principles. The Soviet state sought to alter the position and status of women to conform to new ideological, political and economic agendas. Its underlying ideology and applied policies regarding women attempted to redeﬁne ‘‘womanhood,’’ resulting in new gender role deﬁnitions and expectations. The objective was to enable (or compel) women to participate in the political, cultural and material possibilities of Soviet society. Soviet educational policy was dedicated to creating the new socialist person (both men and women) whose position in life would be determined by hard work, commitment to objectives of the state, and skills needed to advance the cause of socialism (Bronfenbrenner, 1970; Inkeles & Bauer, 1957; Kerr, 1993). In this regard, schools as well as adult literacy programs were particularly critical in Soviet CA since they both aimed to alter family, economic and religious traditions that bound women to household labor and to paternalist patterns of social control (Gleason, 1997; Johnson, 2004; Roy, 2000). Pedagogical goals of the state proclaimed universal standards of quality and equality, although regional and demographic factors mitigated against actual equality in the provision of educational opportunities (Dobson, 1977; Jacoby, 1974; Northrup, 2007). Nevertheless, literacy rates in the Republics of Uzbekistan, Kyrgyzstan, Kazakhstan, Tajikistan and Turkmenistan increased exponentially from the 1920s to the 1980s, with almost universal participation in and graduation from secondary schools by the time of Perestroika (Holmes, Read, & Voskresenskaya, 1995). The social, political, economic and educational situations of women in CA today, however, are in uneasy transition as a result of the Soviet demise. Women are experiencing the return of the same paternalistic cultural patterns, which were the target of Soviet pedagogy earlier in the 20th century (Johnson, 2004; Rywkin, 1990). Although Soviet laws and institutions ofﬁcially imposed gender equality as social policy, in many communities signiﬁcant ‘‘traditional’’ nomadic and Islamic family, cultural networks and practices always remained visible and informally operational – particularly in rural communities (Northrup, 2007; Olcott, 1987). Sometimes, though, even the traditional networks that supported families in difﬁcult periods of the former USSR have been severely weakened during the transition period (Kuehnast, 1997; Kuehnast & Dudwick, 2004). Most new CA republics are dedicated to redeﬁning or overturning Soviet regulations and social aims, calling often to a return to former customs and practices (Gleason, 1997; Smith, 2001). A variety of such customs and practices undermine equity provisions formerly guaranteed under Soviet

Newly Independent States

257

rule. With several important exceptions, customary laws and traditions throughout the CA maintain that men are leaders and authority ﬁgures, whereas women should be considered as responsible for nurturing children at home and focusing on domestic rather than or in addition to economic activities in the private sector (Liczek, 2005; Werner, 2003). With regard to women, the ‘‘ideal’’ Soviet woman drove a tractor or got a PhD. Since 1991, the mostly male leaders of the CA seem to welcome the new old model of womanhood: a young bride surrounded by relatives running an extended household. Rising sentiments of nationalism, long periods of economic crises and out-migration for greater economic possibilities have negatively affected women and their experiences and access to education and public life in most CA communities. From the 1990s, under the banners of becoming ‘‘democratic’’ and of ‘‘joining the market economy,’’ the status of women as well as school quality and equality has declined (Weidman et al., 2004). Most CA political leaders have given lip service to gender equity and school quality, but for all intent and purposes, they have abandoned both the socialist program with regard to women, and have allowed their public school education systems to collapse. At the same time, returning national ideologies that reconstruct the image and role of women as subservient to those of men is often combined with the deteriorating economic possibilities for them in declining state sectors (Roy, 2000). Many of the professional positions where women used to work are disappearing or have lost signiﬁcant prestige. Schoolteachers, medical doctors and university professors – and most other skilled professions dominated by women during Soviet times – are now paid salaries that often put them below national poverty levels. Each of the ﬁve former Soviet states, which are now independent, have ﬂowery language in their new constitutions guaranteeing access to free, quality education; but the quality and quantity of formal schooling for most children and families today in CA is both worse than it was during Soviet times, and worse than the education provided even today in many other former republics of the USSR (UNICEF, 2007). During the transition period, most of these newly independent states have also seen the complete collapse of their rural infrastructures, where the majority of children in all ﬁve republics live (McMann, 2007). Schools in such places slowly crumble; teachers migrate to bigger cities or to other former Soviet republics for high-paying jobs. Since there are no longer clear connections between public schools and industrial or professional work, some older children and their families are now choosing other options.

258

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Increasing numbers of parents are keeping their children out of school to help them work in the ﬁelds in the agricultural communities, or to assign them as herdsmen for family livestock in mountain pastures, or ‘‘in jailoo,’’ when the weather is favorable (Keller, 2007). Many CA families have by now reorganized their economies to the extended household level, where the earnings of children are often considered as family resources. Meanwhile, remittances from out-migrants of countries like Tajikistan and Kyrgyzstan are now ofﬁcially recognized by their governments as important national income streams (World Bank, 2006). We now turn to consider more deeply the trajectory and current situation of gender and education in CA.

GOALS, METHODOLOGIES AND STRUCTURE OF THIS CHAPTER The ‘‘gender problem’’ emerging today in CA as it relates to and involves education actually has long history, and was a target of serious social and political reform during Soviet times. We are interested in describing the problematic emergence; subsequent decline; and current difﬁculties, policies and practices connected to gender equality CA – with a particular focus on education and higher education. There are important historical writings on this topic, as well as contemporary statistical description of the issues. We undertake to illustrate brieﬂy and describe work in both areas to begin this writing. Yet, this chapter is as much interested in the experiences and understandings of gender, education and lived culture as it is in what the history books say and how the statistics read. Our historical and conceptual discussions and generalizations are thus used primarily as scene setters for our later ethnographic accounts. We begin our analysis with a very brief synopsis of Soviet pedagogical and gender policy as it was applied to CA, followed by an overview of the social, political and economic challenges, which was faced by most CA countries during early years of the ‘‘transition’’ to democracy and the market economy. We then turn brieﬂy to how gender equality issues are playing out in the region, and the role education seems to be having in this process. In part two of this chapter, we turn to survey and ethnographic data from several countries in the region, undertaken by each of the authors in different time periods and in different republics over the past 15 years. The observations, surveys and interviews compiled in these accounts are necessary to better document how the statistics actually come to impact the lived experience of CA women today, as well as to suggest why schools

Newly Independent States

259

were considered to be such an important cultural change agents during Soviet times. Available statistical data show and our ethnographic accounts suggest a general decline in gender equality and educational opportunities throughout the region, but the decline has been more pronounced in several of the ﬁve republics involved. Unfortunately, data provided by national governments on educational provision and opportunities are often inconsistent and contradictory. Thus, we base many of the comparisons between countries and over time by utilizing several important international data sets on education and gender. But, since much of these data, too, come from national data bases, they are also problematic, especially in the case of Turkmenistan and Uzbekistan. The primary ethnographic accounts presented in this chapter come from ﬁeldwork in Uzbekistan and Kyrgyzstan. The focus of this work centered on gender in Uzbekistan; and on general educational decline and the occupational status of teachers in Kyrgyzstan. An additional discussion is based on ethnographic interviews with several visiting Tajik university professors at different stages in their careers. The topics of those interviews involved the role and status of women in Tajik families today, particularly how perceptions of women’s place in the household limit opportunities for ﬁnishing secondary schooling, enrolling in the universities, or entering the labor market without the permission of male family heads. In both the Kyrgyz and Tajik cases, the consideration of education and occupation is also critical since the improved status of women during Soviet times was partly accomplished by signiﬁcant professional careers in the service sector – primarily teaching – which is now rapidly eroding.

HISTORICAL BACKGROUND: SOVIET LEGACY, EDUCATION AND THE ‘‘WOMAN’S QUESTION’’ By the time of their independence in 1991, the ﬁve CA Republics had inherited from the Former Soviet Union an educational system that had been continuously formed and reformed for over seven decades of the Soviet power. The status of CA women in newly independent states was also dramatically different by the end of the century, since Soviet policy speciﬁcally targeted the elimination of female subordination in society, what they described as the problem of ‘‘women of the East.’’

260

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

When the Soviet Union was formed, there were few educational opportunities in CA, and illiteracy was the norm (Johnson, 2004). Furthermore, nomadic tribes like the Kazakhs, Kyrgyz and Turkmen had no written language. The few educational opportunities in the region – whether Russian schools for children of the political elites or Muslim religious instruction – were reserved almost exclusively for boys, although there were growing number of ‘‘new method’’ schools established by Jadids that also promoted education for women (Khalid, 1998; Roy, 2000). Before the Soviets, some Turkestani women were tutored in homes by educated women called Otins, who imparted Islamic teachings, social norms and literature in order to make them more ‘‘marriageable’’ (Northrup, 2001). The Jadid movement – to be found more typically in the settled or agricultural regions of Turkestan – also competed with the new compulsory Soviet educational model in the second and third decades of the 20th century. This movement combined efforts to ‘‘modernize’’ both instructional and curricular educational practices and taught such subjects as history and geography, whereas also developing more child-centered instructional strategies – yet retaining core Islamic values and practices (Johnson, 2004). Compulsory education for children and adults was central in the campaign to create a new vision of socialist economic, cultural and social life. In 1918, the Central Executive Committee for the Turkestan Republic ratiﬁed the statute ‘‘On the Organization of Public Education in Turkestan,’’ which called for universal free instruction in the student’s native language, and the secularization of schools. Turkestan at this time had not formally been divided into the ﬁve separate political entities. That would happen over the next decades, and each republic would adopt similar educational policies. In 1930, for example, the Central Committee of the Communist Party of Uzbekistan adopted the statute ‘‘On the Introduction of Universal Compulsory Education for Children and Adolescents,’’ which mandated four years of compulsory school attendance for all children (Holmes et al., 1995). The Bolsheviks launched massive literacy campaigns from the 1920s until the 1940s; and in some areas until the end of the 1950s. Evening and likbez (liquidation of illiteracy) schools were established all over CA. The aim was to educate everyone, regardless of age, gender or ethnicity. This movement was particularly focused on adults, and the belief was that the road to industrialization and social progress and away from provincialism and tradition was through the printed word. These words, of course, were created by the Bolsheviks themselves.

Newly Independent States

261

Special attention during the campaign was paid to women, whose literacy rates were extremely low. Literacy rates for rural females in 1926 were claimed to be only 9.1% in Kazakhstan, 4.6% in Kyrgyzstan, 0.3% in Tajikistan, 1.1% in Turkmenistan and 1.0% in Uzbekistan. By 1939, literacy rates for women even in rural Kazakhstan were 63.3%, 60.5% in Kyrgyzstan, 63.3% in Tajikistan, 53.4% in Turkmenistan and 57% in Uzbekistan. Similar rates for urban women were higher than those of women in rural areas, but totals for men were higher at every level in both rural and urban places (Grenoble, 2003, p. 156). Parallel with the literacy campaigns was the formation of Soviet schooling, providing free, secular and compulsory uniﬁed education for boys and girls. It gradually grew in scope and duration over the decades, although the ability of the USSR to deliver secondary education in the most rural regions of the country and in the furthest ﬂung republics was partially compromised in the 1940s due to events of Second World War. During the 1950s and 1960s, years of compulsory schooling in the USSR were extended to seven and then to eight years. Between the late 1970s and mid-1980s, a transition toward compulsory and universal secondary education had been accomplished. By this time, complete (10 and later 11 years of education) or incomplete (8 and later 9 years of education) – with further vocational or specialized professional training – was available to all (Holmes et al., 1995). Higher education was still very competitive and available only to a small percentage of citizens in CA, as it was in the rest of the USSR (World Bank, 2002). For the Soviets, the social and educational task of creating the new ‘‘socialist person’’ in CA seemed more problematic than in Russia and its more westernized states. They understood that Soviet pedagogy was up against long-standing and complicated socio-economic, political and cultural factors. Building a system of education was more difﬁcult, for example, in the primarily nomadic regions of what are now Kazakhstan, Kyrgyzstan and Turkmenistan during period of forced settlement in the 1930s. Creating schools and forcing children to attend during periods of enforced collectivization in the 1930s and severe famine was also traumatic (Olcott, 1987). It was also more difﬁcult because the formal alphabet changed several times, from Arabic to Latin, and then to Cyrillic – all before the Second World War (Grenoble, 2003; Korth, 2005). Language of instruction was and continues to be a huge instructional issue. Russian ﬂuency eventually became an earmark for preferential treatment in the quest for social mobility. Unfortunately, school provision as well as instructional competence of teachers has historically been part of the ‘‘center’’ versus the ‘‘periphery’’ issue in CA, before as it is now (Korth, 2005).

262

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Another critical difﬁculty for the Soviets in CA was that the secular, universal and applied pedagogies and system of schools they tried to create ran counter to pastoralism and paternalism, cultural orientations which give less status and autonomy to young people and to women, in particular. Historically, CA societies operated in accordance with customary law (adat) reﬂected in strong communal relationships and in the system of clans, and religious law (sharia). These practices in CA both presented and preserved a lower status of women in the society, limiting women’s life mostly to functions within her family and household responsibilities. And, Irina Liczek elaborates how norms of adat historically applied to Turkmen women As adat does not recognize equality between men and women, the latter’s lives were more strictly regulated from early childhood until marriage and thence through their whole family life. The existing literature that details the ancient Turkmen nomadic customs suggests that women were subjected to underage marriages for the economic beneﬁt of their family; were married upon the receipt of kalym (bride-price) from the groom’s family, which created an absolute economic dependence on her husband; had no right to initiate divorce; lacked property rights; and had an isolated existence predominated by household chores, giving birth, and rearing a large number of children. Adat also allowed polygamy for men and inheritance of the widow by male relatives of the late husband’s family. (Liczek, 2005)

Women’s emancipation from rules of adat was an essential part of the Soviet policy of building socialism, and it was tackled in Soviet pedagogy through both curricular and instructional strategies. Parallel to education policies dedicated to creating the new socialist person, new laws defending women’s rights were implemented during the ﬁrst years of Soviet rule, and various institutions and councils were established at different community and political levels to enhance the political, public and economic power of women. These included the Zhenotdely (Women’s Departments within the Communist Party) and the Zhensovety (Women’s Councils within the Soviet authority structures). In his seminal work The Surrogate Proletariat, Gregory Massell observed that early Soviet policies on women in Russia were intended to legitimize a revolution in the relations between the sexes that was already underway, whereas in CA, Soviet policies were ﬁrst and foremost designed to induce a revolution: In Russia the relevant policies were initiated immediately after the October Revolution, and were not terminated (in their special form) until the mid-1930’s. In Central Asia, the policies discussed here were initiated much later (in the 1920’s) and were by and large terminated-and for very different reasons-by 1929. These variations alone make it very

Newly Independent States

263

clear that we can distinguish between the functions of the Russian and Central Asian policy-models. In Central Asia, the Soviet action-scheme in this realm came to be far more specialized than in Russia, far broader in scope, far more self-conscious and autonomous. Indeed, it is fair to say that in Russia, female emancipation was a secondary issue, a by-product of broader revolutionary concerns. In Central Asia, it was, for a time, a primary issue par excellence, a lever for social destruction and reconstruction, an important catalyst for generating the revolutionary process itself. (Massell, 1974, p. xxix)

In executing their plan to emancipate women, the Bolsheviks took the condition of CA women into account and adjusted their programs and policies accordingly. The Bolsheviks were determined to bring modernity to the ‘‘backward’’ women of CA, whom they regarded as perhaps the most oppressed group in the former Empire: women of the East were victimized by patriarchal and feudal customs, and possessed ‘‘low cultural standards.’’ Many also lived under the thumb of the Muslim clergy, and hidden behind veils (Massell, 1974, p. 74). From the Bolshevik perspective, the obvious way to release women from their misery, and bring them into the modern era, was to destroy the religious beliefs and traditions that kept them enslaved. In 1917, the new government proclaimed that men and women were equal in all aspects of social, economic and political life. That same year Bolshevik decrees on ‘‘The Dissolution of Marriage’’ and ‘‘Civil Marriage, Children, and the Civil Registry’’ were announced in Turkestan. The Family Code of 1918 gave women legal equality with men, recognized civil marriage, legalized divorce, and gave all children equal rights whether they were born in wedlock or not. Women legally controlled their earnings and property after marriage, could live apart from their husbands, and were promised access to abortion. The Family Code reﬂected the core Bolshevik belief that the state was a better guardian of the individual’s rights than the family. In 1919, the state opened civil registry ofﬁces, or ZAGs, to record marriages and births; henceforth, traditional marriages were no longer recognized as legal. In 1921, the Central Executive Committee of Soviets of the Turkestan ASSR adopted a decree abolishing the payment of bride money and raising the age of marriage from 16 to 18. In 1924, the Turkestan Central Executive Committee decreed polygamy and forced marriage to be criminal acts, where one article in the Criminal Code stated that any attempt on the person or dignity of a woman, in connection with her emancipation from bondage, was punishable by a three-year prison term. As well, the denial of a woman’s right to engage in work outside of the household or attend school or public meetings was punishable by up to one year of

264

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

imprisonment; and it was made illegal to mistreat or insult a woman or use force to keep her in seclusion (Aminova, 1971; Northrup, 2001). In 1926, the Soviet government stepped up its campaign to emancipate women, launching the so-called hujum (‘‘attack’’ in the Turkic languages) throughout CA. This socio-political campaign, organized under the auspices of the Communist Party, was a direct attack on traditional CA lifestyles and sought a quick end to the practices of seclusion, kalym or bride price and veiling. Soviets widely proclaimed their success with achieving gender equality by the middle of the twentieth century. In 1963, a representative of the former USSR from Turkmenistan went before the United Nations and lectured the West on its failure to deliver to women the sorts equality then guaranteed under law and in practice throughout the USSR. She admonished those assembled that ‘‘decisive revolutionary measures are required of national governments in order to change the former slavish position of women without which social and economic progress of the country is impossible’’ (Liczek, 2005, p. 576). To be fair, some contemporary scholars argue that the formal efforts to acculturate CA women into the norms of the USSR showed mixed results until at least the generation of women who matured in the 1960s (Morozova, 2004; Northrup, 2007). Morozova further claims that many traditional Muslim gender values remained underground almost permanently to the present day; and scholars like Greg Gleason introduce contemporary students to the region by arguing that vestiges of the cultural norms and practices countered during Soviet times are clear and present today To outsiders, Central Asian political practices often seem idiosyncratic and subject to mysterious protocols, secretive and mutual understandings, and subtle but powerful local political idioms. Central Asian traditions of patriarchy, popular submissiveness, deference to authority and to elders, and weak democratic institutions would seem to impel Central Asian societies toward an authoritarian future . . . Perhaps the most visible aspect of the public culture of these countries is the great importance associated with hurmat, the idea of ‘‘deference’’ or ‘‘respect.’’ . . . Hurmat begins in the family. Personal life is family life in Central Asian societies: property is communal, (food) is shared, elders are given deference without question, and women are subordinated. Authority is personalized. (Gleason, 1997, p. 38)

International organizations working in the region now acknowledge that both the system of education and status of the women throughout the USSR and in CA were truly signiﬁcant achievements of the Soviet period. As a result of Soviet policy ‘‘at the beginning of the last decade a comparison of the UNDP human development index (HDI) with its gender-related counterpart (GDI) showed that the countries of Eastern Europe and Central

Newly Independent States

265

Asia had a relative advantage in terms of gender equality compared to countries at a similar level of GDP outside the region’’ (World Bank, 2002).

EDUCATION, ECONOMIC DEVELOPMENT AND GENDER IN CENTRAL ASIA During the past 50 years, human capital theory has become part of the economic development policy of many developing and developed nations (Black, 1999; Fagerlind & Saha, 1989). Meanwhile, the belief in formal schooling as the avenue to individual upward social mobility has itself become a primary legitimizing ideology in many modern societies (Meyer, Ramirez, Frank, & Shofer, 2007; Brint, 2006). During this same period, prominent international human rights organizations have also espoused, and facilitated educational access and equal educational opportunity as fundamental human rights (Spring, 2001; Ramirez & Wotipka, 2001). In each of these understandings, beliefs and calls, gender equity is seen as critical either to provide more human capital resources in activities related to scientiﬁc and technological growth of particular national economies; to improve the social and economic situations of women within their societies; and to ﬁnally deliver human rights and opportunities to women and various ethnic and racial minorities which have often been denied within most patriarchal societies and cultures (Fraser, 1999; Kelly, 1992). Almost all Western scholars interested in women’s rights and equal opportunity concur that there are numerous social and cultural factors involved in the problem of gender equity, which are related to both educational and occupational access. Among them are how gender is constructed within the household, and the extent to which local religious teachings and practices and customs affect beliefs about how women should be treated. Both of these variables are central to the dynamics of gender, education and society in CA. Stephen Brint, for example, maintains that patriarchy typically privileges male prerogative and power in the households and cultures throughout the world, which is linked to both economic and social opportunity. The main question is whether women operate within (strong) patriarchal or less patriarchal contexts (Brint, 2006). For CA, it is clear that the movement is from less patriarchal contexts to more patriarchal ones. Brint also argues that the larger context for gender equality and improved life chances for women is coming to involve post-industrial labor markets where knowledge and communication industries have positive effects on

266

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

gender equity. Nelly Stromquist (2002) counters that contemporary forces of globalization are having the opposite effect with regard to equity and equality. Meanwhile, the ‘‘transitional’’ economy in several CA republics is in fact de-industrializing, not moving into a post-industrial era. Such developments are undercutting many of the gains made previously. International agencies are now explicitly concerned with issues of gender equity in the region as a result of contemporary economic and social development. They typically use gender-sensitive approaches in their own projects (Asian Development Bank, 2005, 2006a, 2006b; World Bank, 2006). The fall of the Soviet Union led to a political, economic and social vacuum in the CA states that left education and women’s place in education in turmoil. Islam, which took a muted role under the Soviets in CA, has regained prominence as has a renewed interest in the construction of national identities (Roy, 2000). Educational reforms that started under Soviet rule during Glasnost and Perestroika came to a halt even in the Russian Federation when the Soviet Union collapsed. Both educational funding and the pedagogical leadership, previously orchestrated by Moscow, vanished, as did many Russian scholars and administrators who repatriated back to Russia from CA (OSI, 2003; UNESCO, 2006; Webber, 2000). All CA states faced varying levels of economic decline in the years following the Soviet Union’s demise (Anderson, Pomfret, & Usseinova, 2004; Gleason, 2003). The 1990s saw a signiﬁcant civil war in Tajikistan. All ﬁve republics experienced double-digit inﬂation as their new currencies were created. Uzbekistan and Kazakhstan faced ecological destruction as a result of over-irrigation and excessive use of pesticides in cotton-growing regions near the Aral Sea (Micklin, 1992). Kyrgyzstan joined the WTO, but found it had little to trade since the Soviet era manufacturing economy collapsed, as did the previous industrial economies of Tajikistan and Kazakhstan. Both Kazakhstan and Turkmenistan, however, began to experience growing demand for oil and natural gas by the late 1990s, but the beneﬁts of these new enterprises had only marginal utility during the ﬁrst 10 years of independence. Comparative economic ﬁgures suggest that only Kazakhstan currently has an economy where the ofﬁcial unemployment rate is less than double digits and where per-capita income is anywhere near what it was two decades earlier. More than half of all citizens of both Tajikistan and Uzbekistan in 2004 lived on the equivalent of less than two dollars a day, and all of the CA republics had external development debts (except Tajikistan) of billions of US dollars (Olcott, 2005). The ‘‘transition’’ to democracy and a market economy has thus had an uneven and painful impact on most of CA, and certainly on education and

Newly Independent States

267

higher education (UNESCO, 2003/2004). Public preschool has virtually disappeared throughout all ﬁve republics (Weidman et al., 2004); school facilities crumbled and also went unrepaired all through the region, since shrinking educational resources were targeted at teacher salaries (UNESCO, 2006); low wages and inﬂation led to double-digit losses among the ranks of teachers who often went to work in the bazaar selling commodities and reselling manufactured goods from China and Turkey (Niyozov, 2004; DeYoung, Reeves, & Valyayeva, 2006). Such problems were confounded by others. For example, each of the new republics sought to re-discover and re-deﬁne their lost national identities, and most of them re-instituted their titular national languages and histories, only to ﬁnd they were hard to teach without textbooks on either (Bahry, 2005; DeYoung et al., 2006; Marat, 2007; Osmonov, 2003). Ethnic Russians whose families had lived in CA for decades and who were overrepresented among the ranks of school teachers and university professors left in droves during the 1990s, fearing there would be no place for them in the nation if and when nationalistic political parties or leaders came into power, as was clearly the case in Turkmenistan. Internal migration from the villages to the cities in each republic also led to underenrollment and rural school demise even as cities increasingly had often to provide double shifts to accommodate the new wave of internal immigrants (OSI, 2003; UNESCO, 2006) With the collapse of the command economy, clear avenues from school to work also disappeared, which has affected student enrollment, achievement and completion (Pitt & Pavola, 2001). Post-secondary vocational and technical education institutions lost popularity and utility, and many closed (Mertaugh, 2004). Many formerly public schools began to seek under-the-table payments from parents which have led to increasing inequality among schools, which often use these funds to attract the best teachers of those who remain (DeYoung et al., 2006). Poorer rural parents sometimes now keep their children home from school in order to work in agriculture or in tending family livestock; and schools and universities throughout the region have become increasingly corrupt as their quality has declined (Silova, Johnson, & Heyneman, 2007). The situation for girls and young women amid all these economic and social issues has become exacerbated. Migration separated many families and economic uncertainty revived Islamic attitudes especially with regard to women. At the same time, career opportunities for women also decreased as available jobs went to men instead of women, and women took on more work in the home partially as a result of men migrating for jobs (Burke, 2001; UNESCO, 2006). This was especially true in Tajikistan and Kyrgyzstan.

268

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Although women have been provided fewer economic and career opportunities the past two decades, most governmental policies throughout CA have ofﬁcially proclaimed equal access and opportunity. The government of Uzbekistan’s ofﬁcial stance toward women, for example, has been that of equality. The Uzbek Constitution of 1992 guarantees all citizens equal rights regardless of gender. The penetration of these laws is minimal, however, since it conﬂicts with the government’s proclamation of the importance of the traditional role of women in Uzbek society as ‘‘the touchstone for its cultural heritage’’ (Human Rights Watch, 2001, p. 9). Similarly, Tajik and Turkmen women have been promised full participation in their new republics, but bear the brunt of economic decline with limited career opportunities and increased burden in home life. Re-emerging Islamic norms there also ﬁnd fewer women allowed to be out on their own in rural places, and ﬁnd more of them wearing headscarves out of fear of being harassed in the street (Falkingham, 2000; Liczek, 2005). Kyrgyzstan and Kazakhstan have encountered somewhat less of an Islamic revival than their neighbors, partly due to a history of pastoralism and nomadism where Islam historically always competed with shamanism, Zoroastrian and other religious traditions; and where tribal and clan allegiances remain as central as religious ones (Kuehnast & Dudwick, 2004; Montgomery, 2007). As Roberts (2007) and Montgomery (2007) argue, the meaning and practice of Islam are both intensely negotiated in many if not most CA communities, and certainly in these two countries. In the Kyrgyz and Kazakhs cases, nationalistic leaders now claim that in their cultures and historical epics, women occupied leadership roles in both combat and in the household, and suggest that their women have rarely been veiled. Kyrgyz and Kazakh women have also historically enjoyed prominent public positions, including ministry appointments and leadership roles in universities and hospitals. Currently, the new constitutions of CA republics guarantee access to free quality education. Schools, however, no longer receive the level of state economic investment as they did during the Soviet period. The lack of state support has led to a critical shortage of teachers, and lowered educational standards (Mertaugh, 2004; Niyazov, 2004; UNESCO, 2006). Funding education in CA has been largely ‘‘decentralized’’ to local and regional governments, which lack the resource base for most social services (Heyneman, 1998). Such developments are in accord with making society more ‘‘democratic’’; but in the short term, this has had severely negative consequences. For example, nursery schools once enrolled the vast majority of pre-school children in the USSR, simultaneously enabling generations of

Newly Independent States

269

young women with families to enter the workplace. When the Former Soviet Union (FSU) had a ‘‘command’’ economy, factories and state farms could be ‘‘commanded’’ to work with and support nursery schools and secondary education. With the collapse of these industries and farms throughout CA, however, almost none of these economic institutions survived, leaving all local governments and educational institutions to solve the problem of budgets and expenditures, which has been extremely difﬁcult to achieve (UNESCO, 2006). Moscow had previously been the site for all education policy and planning during Soviet times, and in fact ofﬁcially discouraged national pedagogical and independent school improvement initiatives. With the collapse of the FSU, every regional education ministry has thus had to create school leadership, administration, and policy formulation activities related to goal setting, organization at the community level and outcomes evaluation. They had to create these specializations without adequate budgets, and the myriad of international aid agencies funding school improvement efforts in all ﬁve CA states have dedicated millions of dollars to budgeting and educational planning and assessment activities (UNESCO, 2006; Mertaugh, 2004). ‘‘Market forces’’ have also had a negative impact in the region related to equality and access, since schools have increasingly become unequal. Wealthier parents and communities have become able to either substantially subsidize ‘‘public’’ education for their children and later to ofﬁcially enroll them in private secondary schools and prestigious private or international universities. Meanwhile, most parents send their children to public schools of poor quality and if possible, place them later in a new generation of universities exhibiting marginal quality (Heyneman, 2004; DeYoung et al., 2006; Keller, 2007). Inequalities in quality of life are now more pronounced in demographic differences since the fall of the Soviet state. Schools in rural CA are in need of substantial repair; and many teachers leave for better paying opportunities in major cities or other countries. International donor agencies have devoted huge sums of money in every country but Turkmenistan just for building maintenance, heating system repair, school furniture and textbooks. Often these projects are targeted speciﬁcally at rural schools (Mertaugh, 2004; Weidman et al., 2004). As this all suggests, the opportunity for social mobility through formal education is becoming more problematic for girls and women with the general decline in school access and quality (Weidman et al., 2004). Meanwhile, disputes are increasingly rekindled today about appropriate occupational futures for girls, since pathways to ofﬁcially sanctioned employment once overseen by Soviet authorities are no longer present. And, women with jobs today, like everyone else, ﬁnd their wages put

270

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

them below the poverty level (Corcoran-Nantes, 2005). Increasingly teaching is now viewed as an occupation of last resort (DeYoung et al., 2006; Niyozov, 2004). Every CA republic reports a lack of qualiﬁed secondary school teachers, and even those in pedagogical institutes hope or expect not to actually have to enter teaching for their living.

STATISTICAL ‘‘SNAPSHOTS’’ OF THE REGION SINCE 1990 In the post-Soviet period, journalistic sources report women have had varied negative experiences in secondary and higher education. For example, after Turkmenistan gained independence in 1991, both the number of years of compulsory education and the number of days in the school year were shortened. Russians and the Russian language – the language many teachers and most university professors were trained in and utilized – were for most purposes removed from use in schools and universities. One prominent critic of the former president argued to the press in 2004 Children are still using textbooks and supplies left over from the Soviet era, and all schools outside the capital are closed from September 1 through November 1 so that students can help with the cotton harvest. In recent years, (former president Niyazov’s) regime has reduced the duration of education from 10 to nine years, and has discontinued subjects deemed unnecessary, including foreign languages, art and physical education. In 2001, Niyazov replaced much of the traditional school curriculum with his spiritual treatise, the Rukhnama. The move effectively transformed the education system into an instrument for political indoctrination. In the opening chapters of his book, Niyazov explains that the Rukhnama contains ‘‘the total of the Turkmen mind, customs and traditions, intentions, doings and ideals.’’ (EurasiaNet, 2004)

Also in Turkmenistan and parts of Uzbekistan, education – particularly university education for women – has decreased as traditional attitudes were claimed to be on the return (Sara, 2006). A 2004 report noted a shift in education and attitudes have affected women the most ‘‘Many Uzbeks cling to a traditional view, in which success for a woman is based on a good marriage, rather than on a good education and a career’’ (Islamov, 2004). This publication also reported that in 2003, women constituted only 30% of students in Uzbekistan and only about 25% of those enrolled in Tajik universities. In the Ferghana Valley, women report they were unable to attend school because their families felt it improper for a young woman to leave home for a university education unattended (Neweurasia, 2007). Tajikistan’s general school enrollments have also struggled in large part due

Newly Independent States

271

to civil strife and ongoing humanitarian crises there. In addition, some argue that Soviet gender equity norms always had less impact on Islamic traditions there than in other republics (Harris, 2004). The struggle between the Islam practiced in the household versus remaining secular aims of the state has an additional paradox for girls and women who want and are allowed to further their schooling. Women coming from Muslim backgrounds in Uzbekistan can and are sometimes expelled for wearing outward symbols of Islam such as the headscarf. Meanwhile, young men can face the same consequence if they wear beards. This trend can further discourage families from sending females to school for fear they might be required to unveil or face the humiliation of expulsion. In 1999, Human Rights Watch reported that 28 students had been expelled from Uzbek schools and universities in a two-year period for wearing veils or beards. Students were also coerced into removing their veils or beards and reported harassment of themselves and their families by authorities (Human Rights Watch, 1999). Perhaps the quickest way to get a snapshot of gender equity and economic and educational development is available through the World Economic Forum (WEF), which provides summary indicators on all ﬁve of the CA republics, but with some speciﬁcity (in 2007) on only four: Tajikistan, Uzbekistan, Kyrgyzstan and Kazakhstan. The WEF is ‘‘an independent, international organization incorporated as a Swiss not-for-proﬁt foundation . . . striving towards a world-class corporate governance system where values are as important a basis as rules . . . (Which believes) that economic progress without social development is not sustainable, while social development without economic progress is not feasible’’ (WEF, n.d., p. 1). Kazakhstan ranked highest on their gender gap index (among the former Soviet CA republics) at number 32 (out of 128). The overall index is based on subindexes in economic participation and opportunity, educational attainment, health and survival, and political empowerment (Global Gender Gap Report, 2007). Kazakhstan scored highest on economic participation indicators, with women and men about equally represented in the labor force, although earning signiﬁcantly less than men. Female salaries approached $6,000 per year, compared to just over $9,000 for men. Yet women were signiﬁcantly overrepresented among professional and technical workers. Education statistics, however, were not as glowing as economic participation rates. Primary and secondary education enrollments had fallen to about 90% for both genders in 2007, although women exceeded men by about 20% in tertiary education: 62% versus 44% (WEF, 2007, p. 91). On the WEF gender education attainment subindex, Kazakhstan ranked 65th.

272

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Uzbekistan’s overall gender gap score was second highest among the ﬁve CA republics, ranking 41st. This seems partly explained by better economic possibilities there than in any other country but Kazakhstan. At the same time, women’s salaries in Uzbekistan were only about $1,400 per year, and only 60% for comparable work done by men. The WEF ranked the gender gap in education on that subindex at only 86, but did not provide actual gender differences or ratios in primary or secondary education. They found overall primary enrollments to have dropped to 81% by 2007, which is a far cry from Soviet times. Secondary enrollments were at 91%, still lower than during the Soviet era. Higher education had fallen to 14% for women and 17% for men. The correlation between increased education participation and completion rates and social and economic possibilities seems seriously eroding at this time in Uzbekistan (WEF, 2007, p. 155). Kyrgyzstan ranked 70th on the WEF overall gender gap index, mostly due to one of the lowest equity scores worldwide in the arena of political empowerment. They ranked 118 out of 128 on this subindex. Working women’s annual incomes were reported at $1,422 compared to $2,464 for men, a ratio of 60%. Educational enrollments were highest for Kyrgyzstan of all four republics with complete data in this analysis. The WEF reported enrollments for boys at 87% and for girls 86% in primary schools; 80% for boys and 81% for girls in secondary schools; and 37% for boys and 46% for girls in universities. Kyrgyzstan thus ranked highest of any country in this last category, a signiﬁcant advantage for women compared to men in higher education; and 43rd worldwide in overall gender educational equity (WEF, 2007, p. 95). Tajikistan was the lowest ranked CA republic (not counting Turkmenistan) on the WEF Gender Gap Index, 79th. Women earned less than $900 there compared to $1,530 for men, or just 57% as much. And, of course, any household trying to survive on either of these salaries would have to have other income sources. The education gender gap subindex ranking for Tajikistan was 107, and is primarily based on the increasing disappearance of girls and women from secondary schools and universities. Primary enrollments are in the high 90th percentile for both genders, but fell by 2007 to 73% for girls and 86% for boys in secondary schools; and to only 9% for girls and 26% for boys in tertiary education (WEF, 2007, p. 143). There is a growing pressure to keep maturing young women out of institutions also populated by men in Tajikistan, as is suggested in several of the interviews ahead. Turkmenistan, under its previous president, apparently paid little attention to gender equity issues, failing to even respond to inquiries from the WEF. In fact, Turkmenistan had signed the international Convention of

Newly Independent States

273

all Forms of Discrimination against Women on December 20, 1996, and agreed to work with various international agencies [UNDP, UNICEF, the Ofﬁce of the United Nations High Commissioner for Human Rights (OHCHR)] on its implementation. With regard to measuring assessing gender equity, however, they reported The country still lacks precise and reliable statistical data disaggregated by sex in the report and the written responses, which makes it difﬁcult to assess accurately the actual situation of women in regard to all areas covered by the Convention and whether direct or indirect forms of discrimination exist . . . The Committee is (also) concerned that it did not receive sufﬁcient information about the status of the Convention in the domestic legal system . . . (And) about the apparent limited understanding in the State party of the concept of formal and substantive equality contained in the Convention, and its prohibition of direct and indirect discrimination against women. (NIEW, 2008, pp. 3–4)

Turkmenistan’s pursuit of equity in social, political and economic equality also drew a number of serious criticisms regarding equity in the above report. The NIEW complained that there was insufﬁcient information on labor markets and women’s participation within it; little data on unemployment rates, anecdotes about discrimination against women in hiring processes, pay gaps and disproportional loss of jobs for women in the healthcare and education sectors. They also claimed that The labour code of Turkmenistan is overly protective of women as mothers and restricts women’s economic opportunities in a number of areas. This may create obstacles to women’s participation in the labor market, in particular in the private sector, and perpetuate gender role stereotypes. (NIEW, 2008, p. 5)

International agencies working in the region include, among others, UNESCO, UNICEF, the Asian Development Bank (ADB), the World Bank (WB), OECD and the Open Society Institute (OSI). Gender equity, equality and improved school quality are among the explicit aims of most international and donor organizations now operating in CA, but both of these aims are increasingly difﬁcult to achieve in much of the region. Importantly, very few of the statistical comparisons for CA republics involving education include outcome data. That is, what students actually learn in schools is not available within the statistical ﬁndings of international agencies, and the creation of testing and accountability schemes within each CA country are also on the ‘‘to do’’ list of many donor groups. Monitoring the United Nation’s Education for All program, UNICEF ‘‘found a marked increase in disparities in the quantity and quality of education’’ throughout Eastern Europe and the Commonwealth of Independent States (CEE/CIS) since the onset of the transition (UNICEF, 2007, p. 9),

274

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

‘‘with systems in the Caucasus and Central Asia suffering far more than those in Central and Eastern Europe’’ (UNICEF, 2007, p. 19). It also found that enrollment rates in upper secondary education were still below 50% in Kyrgyzstan, Tajikistan and Turkmenistan (UNICEF, 2007, p. 11). With regard to the UN’s Millennium Development Goals (MDG), most CEE/CIS nations were on track to achieve MDG #2 (universal primary education completion by 2015), but two CA countries – Kyrgyzstan and Tajikistan – were judged ‘‘in danger’’ of not meeting this goal, and Tajikistan was judged ‘‘unlikely’’ to achieve this (UNICEF, 2007, p. 10). UNICEF also found that Tajikistan was ‘‘in major trouble with MDG #3 (elimination of gender disparity at all levels of education by 2015)’’ (UNICEF, 2007, p. 11). Although education expenditures, facilities and program quality are generally understood to have declined since independence, literacy rates (as opposed to school enrollments) have remained high in all ﬁve republics (above 98% in 2007), with UNESCO’s Gender Parity Index (GPI) as 1.00 in Kazakhstan and Tajikistan, and 0.99 in Kyrgyzstan and Uzbekistan (UNESCO, 2008).

ETHNOGRAPHIC EVIDENCE FROM UZBEKISTAN, KYRGYZSTAN AND TAJIKISTAN Current statistics indicate that gender parity is in retreat in much of CA, economic indicators are mostly in decline, and school quality has deteriorated. Meanwhile, national leaders in all ﬁve former CA republics are involved with redeﬁning national identities that are either not Soviet, or only partly Soviet. Many of these nation-building identity narratives are refocusing on households and families and clans as central, rather than the collectivist life dedicated to advancing the cause of socialist labor promulgated under the Soviets. In the current era, the role of women is more typically portrayed as supportive of and subservient to ends and means chosen by men. Meanwhile, ‘‘protecting’’ women from the world of men and of work has become quite difﬁcult to maintain when local economies have collapsed. Many women are now forced into the market as sellers and resellers, primarily in Tajikistan and Kyrgyzstan. Others have had to out-migrate with their husbands or alone to Russia (or now, Kazakhstan) in order to remit funds to extended families back home. What emerges in many cases are struggles between traditions and generations and

Newly Independent States

275

even within households over appropriate gender relations. These are further complicated by demographic issues (i.e., rural versus urban), and by social class. In the cities and among elites, participation rates and opportunities for women often still echo at least some of the previous opportunities of the former USSR. This is usually not the case in the countryside. Each of the three cases mentioned later gives voice to one or more of these conﬂicts and dilemmas.

EDUCATION AND GENDER IN UZBEKISTAN The ethnographic work in Uzbekistan took several forms: participant observation, and formal and informal interviews – frequently involving oral histories. Fieldwork took place in O’rta Shahar and Katta Qishloq. The primary method of ﬁeldwork was participant observation, involving two different but related families; one urban and one rural. The focus was on the everyday practices of the women because, as the German anthropologist Erika Friedl notes, in everyday practices gender attributes come in ‘‘assemblages’’ (Friedl, 1994). Under the tutelage of an Uzbek ethnographer, a guided questionnaire was developed in which women were asked about their personal history, educational background, work and salary/wage history, living arrangements, household duties and daily routines, type of marriage, childbearing history, religious background, traditional practices and language use, and about their mothers and grandmothers. In addition to very speciﬁc questions, a series of open-ended questions were asked of women to interpret their experiences under Soviet rule were incorporated. The survey was developed with the aim of learning the views of Uzbek women from within, speciﬁcally the evaluative views they have as women, mothers, wives, daughters and workers. The women interviewed ranged in age from 16 to 92. To protect the identities of the informants, pseudonyms are used for cities and individuals mentioned later. The O’rta Shahar oblast takes up 1 percent of the total land of Uzbekistan but is home to 9 percent of the Uzbek population. Almost two million people live in the region, the vast majority of them in one of the 92 villages in the district. Eighty-ﬁve percent of the inhabitants are Uzbek. The industrial and cultural center of the district, and of the Ferghana Valley, is the city of O’rta Shahar. O’rta Shahar city has a population of over 300,000 people and as of 1993 had over 150 state enterprises and factories. The Soviet government claimed an achieved a literacy rate among women of 100 percent in this oblast by 1990, which may have been slightly exaggerated.

276

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

But by mid-1990s, many rural women were very well-educated, especially those women aged 50 years and younger. With the exception of women who were of school age before or during World War II, all of my respondents had at least 10 years of education. Most had even more. Respondents who were born in the 1920s and 1930s, all expressed disappointment at their own lack of education, indicating quite clearly their belief that more education would have won them better jobs. Many of the women in this category had spent most of their working lives in agricultural production. One of them, a woman who spent 40 years milking cows and working as a ketmonchi (ﬁeld worker), gave a response typical of this group ‘‘Oh, I wanted to study but what could I do? The conditions of the time were such that it was impossible. My family could not have survived unless I worked.’’ Another woman, who was born in 1933, did succeed in getting an education long after the War ended. She attended evening school in the 1950s because she was determined not to spend her life in the ﬁelds. In the 43–55 age cohort, women were equally divided on the question ‘‘Did you want to continue your studies?’’ It would seem that those women who wanted to continue did, and those who did not, quit. The usual reason for wanting to stop after secondary school was the desire to marry. In the 29–42 and 16–28 age cohorts, 75 percent of respondents wanted to continue their education and most were able to do so. One woman expressed disappointment that she had received her teaching degree then gotten married off and never used it. Again, many women wanted to continue to study, but also wanted to marry – which would mean they had to withdraw from school. A couple of women in each age group did not wanted to continue their education after secondary school, but reported that their families forced them to undertake university studies anyway. One woman, a wealthy bank director (now retired) from Andijon, said that her parents forced her to study in Tashkent when all she wanted was to get married. She was the only child of a banker and her father expected her to follow his career. She did ultimately marry (a groom of her own choice), became quite wealthy, and is a leading member of the Andijon elite. At the time of the survey, Umida, a respondent in her early 20s, also claimed that education was forced on her. Umida was enrolled in the Andijon Pedagogical Institute, though she rarely attended classes. She claimed that her mother-in-law insisted she get an education even if she never has to work. Her mother-in-law had a pragmatic, even pessimistic view of education; a college degree ensures a better livelihood if all else (marriage, business) fails. She believes that an education is a woman’s insurance against poverty. Umida does not share this point of view. She believes that men are solely responsible for earning money and that women

Newly Independent States

277

should not have to work. Fortunately for Umida she is able to get her degree without actually earning it: in independent Uzbekistan it is commonplace (as it was to a lesser extent in Soviet times) for students to purchase passing grades. Umida had her husband buy her grades with money she borrowed from her friends. Her husband was happy to pay the money as long as it came from her personally, and not from the family budget. With rare exceptions, all respondents expressed the belief that education is equally important for boys and girls. When asked ‘‘What is the biggest challenge facing girls now?’’ the majority said ‘‘getting an education.’’ Women who had girls in primary school were vocal about the schools’ poor quality. Most blamed the general decline in state funding and teaching morale that they felt began under Gorbachev (or even Brezhnev), though one woman blamed the attitude of the girls ‘‘Now girls just want to wear pretty clothes. We had uniforms. Girls should study, but [apparently] now it’s enough for them to look pretty.’’ Many stated that girls had fewer opportunities in independent Uzbekistan, and, given the current economy, might never have the same opportunity for higher education they had enjoyed. Three-quarters of the respondents studied only in Uzbek language schools; the rest studied in both Uzbek and Russian. The 16–28 age cohort had the greatest representation in Russian schools, 7 women out of 24 respondents. An even greater percentage of that cohort’s children study in Russian schools. With the exception of nine women, all respondents said that the Russian language was important for children. One woman said ‘‘I sent all my children to Russian schools. My husband and I suffered a lot because we didn’t know Russian. We didn’t get promoted and had to work in the ﬁelds. I don’t want that for my children.’’ Another said ‘‘of course children need Russian, it is our second mother tongue.’’ To a couple of women, the importance of Russian stemmed from Uzbekistan’s new status as an independent country. Several noted that Russian is a world language, whereas Uzbek is not. ‘‘How would Uzbekistan survive in the world if not for Russian,’’ they asked rhetorically. ‘‘Nobody knows Uzbek, but everyone knows Russian.’’ Some associated knowing Russian with the opportunity to travel, which may be a perceptive assessment of Uzbekistan’s current place in the world, or a holdover from Soviet times when the chance to travel did indeed depend on one’s ability to speak Russian, or both. Uzbek families who could afford it used to travel widely throughout the Soviet Union. Many of the informants, especially urban ones, complained that the dissolution of the Soviet Union stripped them of the opportunity to share their ‘‘country’’ with their children.

278

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Another respondent noted that, with Uzbek boys now doing their military service in Uzbekistan, they were losing their opportunity to travel and learn Russian. The Red Army traditionally sent conscripts outside their home republics in order to broaden their knowledge of the USSR and, hopefully weakens local loyalties in favor of a wider, Soviet one. For this reason, she felt, boys at least needed to study Russian. Those women who indicated that Russian was not important usually said nothing more than ‘‘Uzbekistan is an independent country now, what do we need Russian for?’’ without giving any indication of having thought the matter through in any detail. This attitude was shared by Feruza’s husband, ironically the Russian teacher at the Oiim Uzbek-language secondary school. Feruza and Qamaradin had many heated discussion about their daughters’ future schooling. Feruza had been educated in the Russian schools in Oiim and later at the Andijon Music Conservatory in the Russian-language track. She valued her education and enjoyed speaking Russian. She wanted her daughters to have the same opportunities she had had. Her husband was adamant that their two girls should go to Uzbek schools, though his reasoning, in typical fashion, did not extend beyond the bare observation that Uzbekistan is an independent country and whose ofﬁcial language is Uzbek. Mohira, one of the few informants born before Soviet rule in Uzbekistan, felt very strongly that her life and the lives of her own children had been improved by developments in Uzbekistan after 1917. I never went to school until much later. My mother and father never studied. We had nothing. When the Russians came we had food for the ﬁrst time and we earned good money for our work. Our problems disappeared. The war years were difﬁcult. Especially for us [Mohira and her second husband], because we had small children. But we worked like everyone else. And then things got very good again. After the war we never needed anything again. Today things are difﬁcult, though we still have houses and food. I get my pension. It is not bad like it was before [the Revolution].

Yulduz lived across the street from my hosts in Katta Qishloq. She was legendary in the village for her reproductive achievements. When asked if there were any Hero Mothers in the neighborhood I was immediately led to her. Yulduz was born in 1924 in Katta Qishloq. In 1970, at the age of 46, she gave birth to her youngest child, a girl. Yulduz was pregnant 21 times in her life. She carried 18 children to term, 15 of whom survive today. One of the 18 died within one month of its birth, two died of natural causes later in life. She experienced three miscarriages, but only stopped ‘‘because God stopped giving [her] children.’’ Yulduz was widowed and lived in a large hovli with

Newly Independent States

279

two of her sons and their wives and children. At the age of 71 she was energetic, opinionated and active in the lives of her children and grandchildren. She was passionate about Indian ﬁlms and war movies and kept her television on as long as there was programming. ‘‘I won’t talk to you if I’m watching a movie,’’ she told me the ﬁrst time we met. She loved the music and dance of Indian movies; as for war movies she said ‘‘I lost my brother in the war. I look for him in the movies. I know he’s not there but someday maybe I’ll see him anyway.’’ Yulduz never knew her mother, who died from an undetermined illness at the age of 25. Her mother was born in the 1890s in Katta Qishloq and married Yulduz’s father at the age of 13. Her mother had ﬁve children, of whom three were still alive in 1995. Yulduz’s father, Salahadin, died at the age of 97 in 1987. He was, according to Yulduz, a katta odam (big man) in Katta Qishloq. He was a Communist Party member and worked ﬁrst as a sovkhoz rais, and later as a member of the Party’s regional executive committee. Although Yulduz described her life as ‘‘blessed,’’ her early life was not without some difﬁculties. In the early years of Bolshevik rule, there was a pan-Turkic resistance movement to Soviet rule. This Basmachis were considered bandits and outlaws rather than an ofﬁcial political movement by Soviet authorities, and episodes of violence between contending forces occurred throughout CA in the 1920s (Roy, 2000). As Yulduz explains We had money so that was not a problem. My grandmother was a famous seamstress and my grandfather was a stone mason. He built the mosque here and others too. Before I was born my father was shot in the arm by the basmachis. The basmachis were terrible. They fought with everybody and if they got mad they shot you. Later my father was jailed because of religion. He was a religious man as was my grandfather. When he got out of jail he started to work on the kolkhoz and joined the Party. He had a long and successful career with the Party. After he got out of jail things were ﬁne. Then the war came. We had so many difﬁculties during the war. It was terrible. Today people complain about this and that, but it is nothing. Look at me. I have no husband and I live without problems. After the war things were really good.

Yulduz had also attended school in Katta Qishloq. After she ﬁnished high school she continued her studies for six months and received a teaching certiﬁcate. ‘‘I studied to be a teacher, but never worked. I planned on working but never had the opportunity. I was married at ﬁfteen. It was 1939 and the war was beginning. My husband left for the war and I immediately went to work in the ﬁelds. We all did. He came back and I had my ﬁrst child in 1944. After that I had lots of children and didn’t have time to work.

280

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

I didn’t need to work. My husband had a good job and good position. We had money so there was no reason. My job was to raise my children.’’ Mohira was a poor woman whose personal circumstances allowed (or even obliged) her to take advantage of Bolshevik/Soviet, policies affecting women. Raised in a ‘‘traditional’’ manner (not formally educated, married very young), her apparent inability to have children forced her to look around for non-traditional means of self-support. Following the dissolution of her marriage, Mohira returned home and immediately began working in the ﬁelds. She never thought she would get married again and would live anywhere but her parents’ home until, in an episode that might have easily come from a Soviet feature ﬁlm, encouragement from her brigade leader opened her eyes to the new possibilities that awaited her. Would Mohira have made the same decisions (to get an education and pursue better employment) had she succeeded in having children? It is of course impossible to know. What is clear from Mohira’s story is that, while taking advantage of Soviet gender and social mobility programs often motivated by the need for economic security, Uzbek women reaped other beneﬁts as well. Mohira gained personal fulﬁllment and status from her work and her (limited) education. And she has instilled the values of education and work in her daughter and granddaughter, as we will see later. Minovar describes her early life as one of poverty but happiness ‘‘We never had anything, but at the same time we did not know that. My mother stayed home and took care of us. She taught us, read to us, and encouraged us to study hard.’’ At a young age Minovar was taught the Arabic alphabet by her mother and father. She attended school, beginning at the age of seven, in Katta Qishloq, and remembered I enjoyed school but my sister did not. For some reason it was very easy for me to read and write, and difﬁcult for her. My sister eventually dropped out of school and worked in the ﬁelds. I did not want that. I wanted to continue and was not really sure how that would be possible. During the war nobody had money and we were always poor. And I knew that I would have to marry some day. But my parents encouraged education. My father was a learned man and my mother was literate. She was very religious and could recite from the Koran. She told me and my sister to study hard because education was important.

Then Minovar found an easy solution to her desire to continue her education. When I was near the end of my studies, some Party activists came to our school to tell us about the opportunities available to people who join the Communist Party. I really didn’t know what they were talking about exactly, but I understood that there would be an opportunity to continue my studies.

Newly Independent States

281

She was asked if opportunity was what made her want to join the Party. I don’t know. I just wanted to study more and I wanted to see something else besides the village. I knew I’d get married and this seemed like a good opportunity to do something different. It was exciting. My family thought it would be a good opportunity for me and I think they thought I might earn money. I went to O’rta Shahar and studied at the Pedinstitut [Pedagogical Institute]. I met my future husband. He was from Katta Qishloq too. We worked together for over a year. I liked him and was happy when my parents agreed to arrange the match. I was nineteen and he was twenty-six. We married and lived in O’rta Shahar. We joined the Party. I took a job as a teacher and my husband was sent to Moscow for more education. At that time I had two small children and stayed in O’rta Shahar. It was very difﬁcult. I was lonely and had no one to help me with my children. But I understood that as a wife of someone with immense responsibility it was my job to make sure that all was well at home. My husband did not need the stress . . . When my husband worked in the Party it was hard. They made many demands on him and we had to move whenever he got a promotion or changed jobs. We never had a choice. We lived in another part of Uzbekistan. I was happy when we ﬁnally returned to O’rta Shahar though I miss some of the friends we made [in the other place]. Once back here [O’rta Shahar] I became the director of a school. I enjoyed that work. My teachers and students worked very hard. Things were really wonderful when we came back [to O’rta Shahar]. We had coffee. It was cheap. We had every kind of food. It was all cheap.

Zamira believed that the Russian school would provide them with a better education and ultimately provide them with better opportunities and more personal satisfaction. She said that her Russian-language education had been a wonderful experience I remember the ﬁrst time I heard Mozart played on a piano. I knew that I had to learn to play it. Someone told me that it was ‘‘Russian’’ music and that I shouldn’t learn it. But I didn’t care. And I also knew that it was not Russian music. Mozart was German. My mother, at my request, signed me up for music lessons. The Russian girls had studied since they were six or seven. I was ten. I practiced until my ﬁngers hurt. I loved it and I loved going to the conservatory in O’rta Shahar. I didn’t really like the Russian girls because they looked down on me, but they respected my ability to play the piano.

Zamira had also enjoyed the independence that studying in O’rta Shahar gave her. I loved living in the city. I shared an apartment with a group of my classmates. We had a shower. It was so great. I worked and earned money to pay for my dowry. I was independent. I want my girls to have the same experience. I don’t want them to marry until they’ve had a chance to live on their own and earn their own money. I want them to pay for their dowry. I want them to choose whether to come back to the village or stay in the city.

Zamira was asked why she had come back to the village. She said that she enjoyed village life because it was easy. She had a good job teaching piano and little stress at home. And her family was nearby. She liked living near

282

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

her mother and grandmother, and her school friends. She also liked not having to worry about ﬁnding money for food or other necessities. She even liked her husband ‘‘[He’s] no intellectual; I used to worry that I would get bored [with him]. But he respects me and lets me do what I want. He takes the girls to and from school every day and helps me cook when I’m busy with other work. He is a good man; he does not drink, works hard and loves our daughters. Could it get any better?’’ Zamira was concerned that her daughters would not have the choices today that she had had. Like everyone else in the village, she believed that the state-run schools were broke and unofﬁcially taking money from students. One of the middle schools in Katta Qishloq charged the equivalent of 50 cents per student per month to cover the cost of keeping its crumbling Soviet-era building habitable. Zamira claimed that her husband and his brother did not value education as much as she did. ‘‘Only my husband continued his education. The others went to work. I don’t think they think it is necessary.’’ But Zamira also made it clear that gendered opportunities for her were not a real issue. Referring to her brother- and sister-in-law Timur and Malika, she said ‘‘I don’t think they’ll pay for [any of their children] to go to school. I think they don’t want to spend the money. The boys can work with Timur in the store.’’ As for the girls, she thought that no matter where they went to school, they would learn Russian. She thought it necessary for Uzbeks to know foreign languages, and she did not rule out the possibility that Russian might become useful in the future. ‘‘I don’t know what will happen in the future. Uzbekistan is a little country and Russia is big. Why not learn Russian?’’ Zamira was inﬂuenced by the choices her grandmother and mother had made. Uzbekistan’s economy is so dominated by cotton that, in early autumn, schools are closed and students bussed to the cotton ﬁelds for several days or even weeks. Negotiating state needs and demands involving agriculture was and remains an important dynamic now as it was for Zamira’s parents and grandparents. But educational opportunities provided by the Soviet Union did allow for some escape from the ﬁelds, and Zamira’s grandmother was very pleased about this After my grandmother was divorced she learned to read and write. She didn’t learn much Russian but she knew enough for her work. My mother went to the Russian school because my grandmother wanted her to get a good job. She didn’t want my mother to work in the ﬁelds. My mother met my father at school. They sent all of us to Russian school and they told us to study. We never had to work in the ﬁelds except to pick cotton. That was nothing. Everyone had to do it. But other than that I never worked in the ﬁelds. I don’t even work in the garden.

Newly Independent States

283

EDUCATION AND GENDER IN KYRGYZSTAN Kyrgyzstan is more mountainous, has fewer large cities, and its economy was historically more pastoral than agricultural than Uzbekistan. Yet many of the themes discussed and remembered in the Uzbek case have parallels in the contemporary Kyrgyz republic. Formal schooling provided or compelled during Soviet times remains highly valued in Kyrgyzstan, even if the utility of advanced education appears to have less direct relevance now than when it led directly to a professional occupation in the planned economy. The data presented earlier in this chapter revealed that secondary education enrollment in Kyrgyzstan remains near 90%, and parents who can send their children to the university work hard to do just that (Drummond & DeYoung, 2004). University enrollments in Kyrgyzstan are the highest among any of the new CA republics. Many parents also want their children to study Russian in school, although there is greater debate about how much emphasis should be given to this (DeYoung, 2008a; Korth, 2005). Meanwhile, this is really only a choice in the cities, since native Russian speakers are mostly now gone from the village (DeYoung, 2006; DeYoung et al., 2006). Russian is still valued by many families for its perceived economic and social value, as it is in the Uzbek case. And for many poorer Kyrgiz (and Tajiks), Russian is also valued for its utility in employment, since many outmigrate. Remittances from jobs held by CA citizens now working in other republics represent a signiﬁcant portion of the household income for thousands of families (World Bank, 2006). Nevertheless, most Kyrgyz families want their boys and girls to get as much education as possible; even though parents understand that it is personal connections that are required to secure virtually any professional employment (Kuehnast & Dudwick, 2004). Becoming a ‘‘cultured person’’ by going to the university was a goal partly unto itself during Soviet times (DeYoung, 2008b). The urban–rural divide is now an important problem in CA, as opportunities for education and social mobility are less available in the village (McMann, 2007; World Bank, 2008). Rural roads are in serious disrepair in much of CA, and running water is now a luxury in the village. The marginally adequately shops stocked with basic and affordable foodstuffs during the Soviet era are now mostly gone, and electricity outages also frequent in many places. The burdens of such developments affect women disproportionately, since they still essentially run CA households. A good education today is understood as one way for rural Kyrgyz to get out of the village and head for an oblast center or better yet, the capital. We heard much testimony to this fact in our rural school

284

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

ﬁeldwork in 2004 and 2005 in several rural communities in the Kyrgyz republic (DeYoung et al., 2006). A primary research technique involved ethnographic interviewing (Spradley, 1979). Gulshat, a 23-year-old English teacher from the Kyrgyz raion center of At-Bashy in the Naryn oblast is an ethnic Kyrgyz, originally from the place where she now teaches. At-Bashy is the regional (raion) center of one of Kyrgyzstan’s more remote areas – Naryn – 5 h away from the national capital of Bishkek over rugged mountain terrain. In Soviet times, teachers working in high-mountain rural settlements that were considered vital outposts for agricultural and livestock production areas were substantially subsidized. Economic, structural, administrative and cultural opportunities in such remote places were provided, even if relocating to capital cities were often the dream of upwardly aspiring citizens. Gulshat never knew Soviet times, and living now in the raion center of a remote province of the country was not her goal. She hoped desperately to escape her place of birth in order to move to Bishkek. Gulshat had decided, as many young women do, that moving from the periphery to the center was possible as a bilingual (in her case trilingual) specialist. Without oil and gas reserves of surrounding CA states and with no industrial base any longer, working with or for international organizations and businesses is understood as a viable employment seeking strategy now. Many students in Kyrgyzstan today undertake language-training specializations offered by national universities under the pretext that they will then teach in national schools. However, those who become adept in a second or third language believe they can quickly move away from school teaching which pays $30–50 per month, and into a job working with an international ﬁrm with wages six to ten times that high. Gulshat could not afford to enter a foreign language specialization offered at most universities by herself, but she was able to enter the English language program for teachers, a government subsidized program. She had not understood that her scholarship required her to remain in her village as a teacher for two years after she ﬁnished coursework. She could not obtain her diploma otherwise. When we met and interviewed her, Gulshat had become very ﬂuent in three languages (Kyrgyz, Russian and English), and had no long-term intention of remaining a teacher. She planned to move as soon as possible to the capital of Bishkek when her two years of obligatory teaching in At-Bashy was up. Gulshat lived at home with her parents in the village, but chafed at the idea that she would remain here. She wanted to experience life in the city and make enough money to live comfortably. We did not press

Newly Independent States

285

Gulshat on matters of gender in our interviews of 2005, since rural versus urban was the focal point of our study. But she did move into gender issues both indirectly and directly over the course of our interviews. Since secondary schools were a primary outpost for developing gender equity during Soviet times, and since teachers were a primary group leading the ﬁght for ‘‘modernity’’ and equality, the struggles Gulshat faced as a teacher in her village are suggestive of the re-gendered situation in rural Kyrgyzstan. Gulshat equated educational success as an avenue to become ‘‘modern.’’ She had attended one of the only Russian language schools in her small town as a girl, and had come to appreciate the advantages and freedoms promised at least to those who could make it to Bishkek. At the same time, she was critical of those in the city whom she thought did not honor their own language and traditions. She opined, Our Kyrgyz mentality suggests that if you live in the village, you are not modern enough; you are old fashioned. (After being at the regional university in the city of Naryn), I do not believe in that. ‘‘Naryn is all mountains everywhere; and herds of cows are everywhere; and we are wandering between and among them.’’ Bishkek inhabitants – all of them – think that way. Some even do not know where our town is. I think I am quite modern, because I know English really well. Just Kyrgyz is not enough for me . . . To be considered to be modern, in our culture, you have to know no fewer than three languages, to be intellectually developed, and to understand the situation of the country.

Gulshat’s parents did not mind that she wanted to go to the city and create a future there. Girls will be leaving their birth family eventually anyway to join the family of their future husband, and, to ﬁnd a good husband in the city who can support you and raise grandchildren is usually a good thing. But the real problem for Gulshat was that she might be kidnapped by a local boy (and his family), thus squashing any chance for moving to the city to become independent. Bride kidnapping is either a returning tradition, or a new accommodation to ﬁscal problems in Kyrgyzstan that avoid some of the extreme costs of arranged marriages. Usually, romantically linked boys and girls agree for her to be whisked away to his family’s home, the details to be sorted out later. In some cases, though, women are taken on the street by young men they have not even met if and when the family of the boys thinks it is time for him to get married but he has not found a suitable candidate himself (Amsler & Kleinbach, 1999). The case of bride kidnapping in Kyrgyzstan underscores the inconsistency of government laws and social practice in much of Kyrgyzstan speciﬁcally as it applies to gender. Although there is a law in the books that forbids this practice, in reality the local norms which have re-emerged since the fall of

286

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Soviet power are opposed to using the civil courts to resolve what are considered inter-family affairs. Who young men and women marry is considered a family affair in rural Kyrgyzstan. And many families believe calamity will befall them if parents attempt to rescue a daughter even from a forced kidnapping. Gulshat had already ‘‘lost’’ one sister to a forced kidnapping, and her parents had done nothing about it. The national implications for sending single women into the countryside in Kyrgyzstan with regard to the ‘‘profession’’ of teaching is also profound, and another reason why it is increasingly impossible to attract female teachers to the more remote parts of Kyrgyzstan. The implications for Gulshat are best expressed in her own words: If I stay here for many years, I think I will be kidnapped. I do not want to be dependant on my husband or my husband’s family. I want to be really independent; just to get a job. I want to live life in such way that I’ll never regret about my past . . . Kidnapping is our old (tradition), and becoming stronger and stronger each year. And my parents tell me that I need to ﬁnd a good guy for myself in order not to be kidnapped by someone. But if I am kidnapped, my parents would decide what to do. This is a Kyrgyz tradition. No one can break it. My parents do not want to be accused (of breaking this tradition), and I do not want them to be ashamed of me, either. I do not want to put them into an awkward situation. If I am kidnapped, I will probably agree (with my parent’s decision). I do not know. I cannot imagine this. (But) my family is considered to be really honorable in our village. I do not want them shamed.

The dilemmas facing women in Kyrgyzstan who might want to be more cosmopolitan and ‘‘modern’’ in a rural and traditional setting in Kyrgyzstan is also illustrated in the biography of a more experienced teacher in the same school as Gulshat who had returned to At-Bashy from the regional city of Talas, after a failed marriage. Even as a teacher raised and educated in Soviet times, she believed it was her fate that led her back to the village. We would argue it is the result of the traditional constraints of her gender in a patriarchal society. She was at the same time both a divorcee – never a good situation in CA – and the only unattached daughter of her father. Living in a Kyrgyz village means that there are virtually no institutions to deal with the elderly or the incapacitated. These are family matters, and Gulzhan is part of the family. This has put her in a very distressing situation, but one from which she saw no escape I got married to a Talas guy, but after living there with him for six years – to use the common phrase – our characters were not compatible. I had to get divorced. After this divorce, I needed to raise my daughter, and I could have lived separately. But when I arrived here, in At-Bashy, my mom had just died. She was sick, and died at the age of 49.

Newly Independent States

287

And we have our father at home. One married brother lives separately, and I have two other brothers: one is not married, and the other has been handicapped from the birth. After me, there is my sister. She is married and has her own life, living with her husband. So, I am here with three men, and with my daughter, as if being the entire labor force. I cook for them, and wash clothes for them. In essence I am a free servant. To say honestly, I am tired, but I am tied. The only reason that I am not leaving the house is that I have my brother who is sick. He is sick from his childhood. He is old enough (35), but he requires permanent care. He can (only) walk; put on his cloths; and eat. I am chained to him to take care of him. That’s it.

A ﬁnal fascinating case where the gendered interplay of tradition and professionalism in Kyrgyzstan played out in 2005 in At-Bashy comes from the biography of Burulsun, formerly the school director at Kazybek School. She lives a very traditional Kyrgyz life, although she is only partly Kyrgyz. She milks the two family cows daily after school, and in 2005 helped her husband with feeding and moving their 20-plus herd of sheep in the morning and in the evening. In the winter, the livestock lived behind her house; in the Spring and Summer, they are driven to the high mountain pastures – or jailoo as it is called in Kyrgyzstan. She had a large vegetable garden, ran her household, cooked and raised her three sons. Two were away at Bishkek universities in 2005. A third son was working at the television transmitter tower in the nearby mountains. The understanding was that like most Kyrgyz boys, the younger son would eventually take over the household and care for both Burulson and her husband when they reached retirement age. Burulsun was not only the School Director at Kazybek School, she was also a former Komsomol leader in Soviet times and in 2005 a facilitator of innovative teaching techniques provided by the Soros Foundation, USAID and several English language teaching programs sponsored by other international donors. But Burulsun participated fully in 2005 in several other ‘‘Eastern’’ traditions that she certainly did not learn in her Soviet pedagogy training or at English language courses offered by foreign organizations after independence. For one thing, Burulsun’s neighbors often came to her to predict their fortunes. She was considered to have special shamanistic powers in this regard. Also, the logic of bride kidnapping as understood on the ‘‘receiving end’’ of the tradition was revealed to us in Burulsun’s home. Technically, Burulsun was not kidnapped herself, because her husband’s family and hers did partially arrange their marriage years back. But Burulsun’s youngest son who worked at the transmitter station had kidnapped his wife, the daughter of a family friend. This kidnapping was understood as mutually satisfactory, and was later sanctiﬁed by a local Imam, even though bride kidnapping is not ofﬁcially recognized as a

288

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Muslim tradition. The new daughter-in-law was from a single parent household, and that parent was currently unemployed. Marrying into Burulsun’s family was a key to security for the young woman, and as part of the agreement, Burlsun and her husband agreed to pay for the young woman’s continuing college education in the accounting faculty at Naryn University. This would have to be done by ‘‘correspondence,’’ as the new couple immediately began to have children of their own. They lived in the home of his parents, and the young woman was also partly responsible for running the household of her parents-in-law and raising their young daughter in 2005. Burulsun and her husband were quite satisﬁed with the new domestic relationship, and talked with us freely about the possibility of ‘‘kidnapping’’ another young woman for her oldest son. She had discouraged him from a romance he reported was having with a young woman from a distant part of the country, arguing that he should ‘‘choose’’ a potential mate from a good family closer to home. And, she told him to be careful not to marry a divorced woman or one from a broken home. This would be the sign of a poor gene pool. Burulsun was somewhat eager for all of their sons to be married before she retired from her school directorship, as wedding presents to the family from friends, relatives and co-workers would more than offset the huge cost of throwing a traditional several day wedding party for them. The value of presents (e.g., horses, sheep, furniture or money) would be expected to be greater the more important the family is of any son who gets married. And, being a school director in Kyrgyzstan is considered to be an esteemed and powerful position.

EDUCATION, PROFESSION AND GENDER IN TAJIKISTAN In much of Tajikistan, the possibilities of gender equality are even further compromised than they are in Kazakhstan and Kyrgyzstan. Education – and especially education for girls in Tajikistan – is even more closely associated not so much to future careers and getting a better job, but mostly to the notion of whether it is good or bad for marriage. And, the answer to this question appears deeply rooted in certain family beliefs as expressed by the men in the family. Fathers and husbands in this case ultimately deﬁne the level of education a girl is going to get – whether it is primary, secondary or higher education. To work or not to work after graduating from the

Newly Independent States

289

school or university will also be decided for a girl by her husband and his family, especially by her father-in-law. And how a particular family will decide on both of these matters will, in turn, be inﬂuenced by a number of factors, reﬂecting a societal split along rural versus urban, elite versus nonelite, region of the country and religious differences. Arranged marriages are also prevalent in Tajikistan, where a bride and a groom might see each other for the ﬁrst time only during a wedding. With a mean age of marriage for woman of 21 years old, and a fertility rate of 3.7, ‘‘education,’’ ‘‘profession’’ and ‘‘job’’ can only be understood by considering how particular families control the connections. Our insight into issues of education, gender and culture come from extensive interviews with several visiting Tajik university faculty members who are from diverse backgrounds there. They are from two different regions of the country (east versus west), different Muslim denominations (Ismaili versus Sunni), different ages (an unmarried junior professor versus an established teacher and administrator with a family of her own) and from different social classes: elite versus non-elite. They talked with us about their personal perceptions on status of woman and role of education, as well as what they tried to identify as ‘‘traditional’’ or ‘‘cultural’’ patterns that are very common within their republic. Their family (and extended family) biographies suggest that the value of education in Tajikistan and especially for the girls is often determined by the anticipated and assumed role in the family hierarchy. Consider the story of Garib, whose family’s fortunes are unfortunately not unusual in post-Soviet Tajikistan. Notice also the implicit belief in the value of higher education that often goes monetarily unrewarded now in her country. Even so, professional possibilities in many families, including Garib’s take a back seat to needs of elders and male elders in particular. And Garib comes from a professional family; her father is a teacher as well as an educational administrator at the raion (district) level. Her mother is also a teacher I have two brothers and two sisters. My sisters are older than me; and they are married. (One) elder sister is working with her husband in Russia, in order to earn money. She has higher education, but she is not working in accordance to her specialization. She (was)a teacher of literature in Tajik language. And my second sister, she is married as well. She and her husband are living with her father-in-law because he is alone; a widower; in a village about 20 km from Khorog. And she and her husband – they are both Candidates of Science (equivalent of Ph.D.s) in physical sciences. (But), both of them are jobless. There is a vacancy in a Khorog State University (where her sister could work), but she is not allowed to work by her father-in-law, (even though) her husband cannot ﬁnd a job. (She is not allowed to work) because her father-in-law would be left alone . . . When we get married, our fathers-in-law, our parents-in-law, make the decisions for us to work or not.

290

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

While Garib was talking about her older sister, our other informant, Naima, says the same beliefs about women’s professional versus care-taking roles is visible today among students in their early 20s in the classes she teaches. Her mostly male students still believe it is important to keep wives or potential wives out of the labor force where they might come into contact with other men. Still now, among our young people, opinions did not change. Some (of my students) – when they are married – they want their young wives to sit at home. There are (still) such students; we had this discussion in my class (in 2007). They ﬁnished Russian (language) school, but their opinions are still as before the Soviet times. I asked, ‘‘when you marry, will you allow your wife to study?’’ They answered, ‘‘maybe I will allow them study, (but) only to study at home, or at correspondence courses – they would (have) to transfer to correspondence courses (after marriage).

Naima also asked After graduating, will you allow your wife to work somewhere, to make her career?’’ They answered, ‘‘No.’’ Some of them – they are against it. Some of them said, ‘‘why should my wife work? . . . and, if some men will speak with my wife and somebody will tell me ‘I saw your wife with a man’, what will it be?’’ . . . (One student said), I do not allow my wife to work if there will be a man, if she will be in contact with men. If she (were to) work at some kindergarten or school, where there are very few men, then I will allow to work.’’

Naima tried to explain this tradition of gender inequality as it exists in her life and in her community with reference to custom and tradition that has existed for many generations, seemingly not erased even during Soviet times, and still alive and well It is a tradition a tradition from ancient times. Men are considered as the owner of his wife also. And he bought his wife . . . A young girl comes to another family (her new husband’s), where there will be many hosts or ‘‘owners:’’ her mother-in-law; father-inlaw; sister-in-law; and her husband. For example, in some of our families, when (a wife) wants to go somewhere, (even) to visit her own parents, she must ask her father-in-law; then mother-in-law; and then she would be told: ‘‘ask your husband.’’ She should pass through three levels in order to go somewhere.

The equal status of women in contemporary Tajikistan, though, is not only under threat from ‘‘tradition’’ or ‘‘Islam,’’ but is also paradoxically constructed these days as an unwanted consequence of the ‘‘market economy,’’ in ways suggested by Stromquist (2002). Many men in the country have no work, and women are increasingly forced into selling and trading in the marketplace for their families to survive. Since much of this trading involves border crossing in search of goods to bring back for resale, there is always a good possibility of theft and extortion at national borders

Newly Independent States

291

and on the open road. Women are thus increasingly thrust into the market since they are less likely to be harassed along the route. This may increase their value within their families, but at a high cost; or so argues Naima, a senior university faculty member studying international diplomacy in the US during 2008 Family attitudes towards men and women (are) changing, because most of the men in our country are unemployed, they have no job. And women are made to make their business and to earn their living. Because the women have privileges (advantages): They will not be checked (searched) at the borders; they often go abroad in order to make business (and) it is easier for them to go than for the men. They will be robbed. The money will be taken away from the men on the road . . . Before, men did not allow a woman to go somewhere, since men were considered to be the owner (of women), and only with the allowance of a man could we do something. But now it is changing. Nowadays the women’s voice is becoming louder than men’s in the families. Yes, then women earns money; it changes the position when a woman makes business and (keeps) house. (Meanwhile), things are becoming more rude, because sometimes when you pass a market – you hear such (foul) words from women that you cannot tell the difference between men and women. They use all the words and expressions, yes, used by men . . . That is why I do not like it when women are sitting (selling) in the market. I do not like such women, and do not want to get in contact with such women. They will not think when they speak, (and) they speak all they want.

Even though women have lower status in most Tajik households, and they sometimes are not allowed to work in the specializations they are trained in even if there is a job, still, education and higher education are often highly valued – just as we heard in Kyrgyzstan. Often obtaining a university diploma is the real objective, even if families have to pay bribes and other fees under the table for their daughters to get one. Garib discusses this matter further, also suggesting that her minority Islamic community supports education for equity in the Pamir region of Tajikistan With regard to higher education, I think it is a kind of tradition in our society; it is very modern to get higher education, whether you are good or not. For example, as I see my neighbors, some of them are even not able to take their exams themselves, but . . . they pay for that; their parents, they kind of made them to go to the university; they make money in order for their children to get the education . . . Yes, their parents try their best for their children – to get an education whether the child is interested or not. Maybe because of getting job . . . or because of becoming independent . . . Parents try to give education to their daughters. I think that the main motive for this is that – as Aga Khan said – as we are followers of Aga Khan – as he said, ‘‘if you have three children in the family: two boys and one girl – you have to give education, you have to educate, ﬁrst the girl, and after that the boys. Because she is a future mother, and she brings up her children, and she is the future of new generation – (everything) depends upon her modern(ity).

292

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Naima, who is also a mother and grandmother (but a Sunni), argues that most interests in equal education and status for women comes to her region as a function of former Soviet inﬂuence and now social class before the Soviet formation, education was only accessible to men; just men studied at universities; but now it is somehow changed. (Even so), the girls from remote villages, they do not study. There are some (village) girls (in the university), but only by presidential quota. They get to study at the universities if they have some privilege; otherwise . . . they get married very early, just after the ﬁnishing secondary school. But each educated, intelligent family wants their children to be educated also, and at the center of our city, in the large cities – most of the parents – they want their children to enter the university; they study at the university. But still, the minority is girls, and majority is boys in our universities.

Garib talked further about the city versus village divide in gender and in education in other parts of Tajikistan. She described an interaction she had with a young wife who had moved to the capital from the countryside recently I kind of witnessed one girl in Dushanbe (the capital) who was a neighbor of my aunt. She was married, and 25 years old – but she had not gone to school at all; she went just till the third grade. I asked her why, and she said in her region if a girl goes to school and especially high school, no one will marry her: she is kind of spoiled; everyone will think she is spoiled. (This) girl she was always wearing a hajib . . . She always said ‘‘I am afraid to go outside, to the street.’’ I asked her why. She said ‘‘maybe some of my relatives will see me and it is kind of sin for them . . . And then I asked her ‘‘why are you afraid to go out’’ and she said that her husband was in Russia and he told her not to go out (of the house), never. He said ‘‘if someone sees you, I will divorce you.’’

Naima, also talked about the importance of caste and education, where it is assumed that education is much lower now among low ranking families without the control and compulsion better organized during Soviet times In our country . . . if – not my daughter, but someone else’s daughter – falls in love with a boy from remote village, parents will not allow them to get married; because maybe of their position. This family is educated, and that family is from remote village, they are not educated, they are collective farmers; because of their position, because of their caste. Caste – yes. For example, ours is considered a very high caste, and we cannot – our caste cannot give their daughters or marry their sons to a lower cast . . . Now, my ﬁrst daughter-in-law, I married my son to our family relative, a far-removed relative, but a relative. My son saw her at another relatives wedding and liked her . . . And he asked me if I could marry this girl. And I said that is all right. They are relatives, we know them; they will not speak such rude words, to relatives, they will give respect . . . But my daughter is married to someone who is not my caste. Such a story! Many boys desired to marry my daughter, but my third brother’s friend had a great desire. And my brother – who helped me and sponsored me and everything – he asked me: will you give your daughter to my friend? But they are nice family, though they are not our caste; this

Newly Independent States

293

son-in-law is very modern; he spent three years in the USA (and) is very educated, now I am satisﬁed with (them).

Although we did not speciﬁcally ask Naima about her overall thoughts on independence and national development in Tajikistan since the former Soviet times, it is pretty clear that gender equality is at least in the short term a casualty of this process in her country. We conclude this section of the chapter with her implicit observations on marriage, education and equality I have four brothers and one sister . . . My brothers and sisters, they are all educated; my parents were such nice people; they educated all of us . . . My husband and I, we are distant relatives by caste. He lived in Kyrgyzstan, in one of the regions in Kyrgyzstan, and sought (our family) from a far distance, but he is Tajik, yes. And he sought us in Khojand in order to marry me. I had not seen my husband in my life. I saw him ﬁrst during our civil (marriage) registration. He wanted to see; to meet me ﬁrst – to get acquainted, but I did not want to. It was after graduation (from the university) in the former Soviet Union that girls used to marry, but now it has become a tradition – a habit – to marry girls just after (secondary) school, or in the ﬁrst or the second year of university. If you marry a girl in the third or fourth year (of the university), she will already be too old; there will be no desire for an old girl. From one side, it was very nice. I was 21 when I married. I graduated from the university, and I was working at a medical college. And we lived happily (then).

SUMMARY OBSERVATIONS In her study of Arab women, Nadia Hijab tells her readers that ‘‘[a] society in transition is one where traditions are still alive, and where many different kinds of social systems are in force (Hijab, 1988, p. 164).’’ As we hope the reader has observed in our historical and statistical overview, there are signiﬁcant economic, social and cultural differences within the CA region. This makes it very difﬁcult to easily summarize the entire situation. Yet, the lives of the women ethnographically portrayed here in the three countries of Uzbekistan, Kyrgyzstan and Uzbekistan testify to Hijab’s claim. The women we have given voice to lived and yet live in the contexts of the Soviet state – and what remains of this state – as well as multiple CA traditions. The forces that impel them and the choices they have to make reﬂected and reﬂect their need to balance them all. Although some fully embraced the opportunities provided to women by the Soviet state, none of them completely abandoned the traditions that scholars have traditionally labeled ‘‘Central Asian.’’ The lives of these women illustrate the hybridization of identities – Soviet and CA – that is the most important result of Soviet rule in CA where women and female gender identity are concerned.

294

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

For all of its other inadequacies and failings, the former Soviet Union once had a systematic approach to gender equity policy that was at the time both ‘‘symbolic and sociostructural’’; and much of this policy was intentionally located in the education sector. It was here where emergent interests of building socialism met and wrestled with the weight and dynamics of several CA cultures and traditions. Following the dissolution of the Soviet Union, leaders of the ﬁve CA republics have mostly abandoned the socialist program though not with regard to every sphere of national life. This abandonment of course is different and uneven in the different republics. Uzbekistan, for example, retains key elements of its command economy, whereas Kazakhstan does exactly the opposite. And each of the other republics have experienced over a decade of instability or decline. More obvious steps moving away from the Soviet ideal are taking place in social and cultural areas. In Uzbekistan, one of the more fully elaborated and vigorously promulgated components of this ideology is an imagined pre-revolutionary past in which the restriction of women to the private sphere supposedly enriched the lives of women and the entire nation. Similar patterns are visible in each of the other republics. Although we must take care not to ennoble Soviet rule in CA – a system founded, after all, on physical terror and ideological monopolization – we must also take care not to ignore how Soviet rule changed the lives of CA women, and how the lives of women under Soviet rule compare with the lives of CA women in the present, post-Soviet period. For one thing, a substantial majority of CA women believe that the Bolshevik Revolution brought in conditions for women that were vastly better than they had enjoyed previously. In contrast to what we would expect to ﬁnd, this belief becomes stronger as one moves from the cities into the countryside, where millions of CA women still live. But more importantly, what we know about CA women’s lives today suggests that their perception of their Soviet experiences is quite closely grounded in actuality. Female literacy, which reached levels approaching 100 percent under communist rule, is now slowly falling as a result of increased political emphasis on traditional roles for women, early marriage ages for women and lower salaries for women compared to men in all republics. Domestic violence, also not unknown before 1991, has become a more serious problem, whereas new threats, like sex slavery, have appeared alongside it. The question today is whether the present period will take its place in a trajectory of constant improvement in which Soviet rule represented a way station toward full participation in modern society, or whether it will come to be seen as a golden age for women, a chance that the country’s post-Soviet rulers allowed to slip away. The verdict is of course still out on this matter.

295

Newly Independent States

REFERENCES Aminova, R. (1971). The October revolution and women’s liberation in Uzbekistan. Moscow: Nauka. Amsler, S., & Kleinbach, R. (1999). Bride kidnapping in the Kyrgyz republic. International Journal of Central Asian Studies, 4, 185–216. Anderson, K., Pomfret, R., & Usseinova, N. (2004). Education in Central Asia during the transition to a market economy. In: S. Heyneman & A. DeYoung (Eds), Challenges for education in Central Asia (pp. 131–152). Greenwich, CT: Information Age Publishing. Asian Development Bank. (2005). The Kyrgyz republic: A gendered transition – Soviet legacies and new risks. Manila, Philippines. Asian Development Bank. (2006a). Gender assessment synthesis report. Mainstreaming gender in poverty reduction strategies in four Central Asia republics (Azerbaijan, Kazakhstan, The Kyrgyz Republic, Tajikistan). Available at http://adb.org/Documents/Reports/ Country-Gender-Assessments/cga-synthesis,pdf Asian Development Bank. (2006b). Mainstreaming gender in poverty reduction strategies; Tajikistan – Country gender assessment. Manila, Philippines. Bahry, S. (2005). Traveling policy and local spaces in the republic of Tajikistan: A comparison of the attitudes of Tajikistan and the World Bank of the provision of textbooks. European Educational Research Journal, 4(1), 60–78. Black, J. (1999). Development in theory and practice (2nd ed). Boulder, CO: Westview Press. Brint, S. (2006). Schools and societies (2nd ed). Stanford, CA: Stanford University Press. Bronfenbrenner, U. (1970). Two worlds of childhood: U.S. and U.S.S.R. New York: Russell Sage Foundation. Burke, J. (2001). Women likely to suffer most in Central Asia’s turmoil. Women’s eNews, September 31, 2001. Available at http://www.womensenews.org/article.cfm/dyn/aid/671/ context/cover/. Retrieved on 6/6/2007. Corcoran-Nantes, Y. (2005). Lost voices: Central Asian women confronting transition. London: Zed Books. DeYoung, A. J. (2006). Problems and trends in education in Central Asia since 1990: The case of general secondary education in Kyrgyzstan. Central Asian Survey, 25(4), 499–514. DeYoung, A. J. (2008a). National language, national identity and the university: Contextualizing the debate at two rural Kyrgyz secondary schools. Journal of Central Asian Studies, (in press). DeYoung, A. J. (2008b). Conceptualizing paradoxes of post-Soviet education in Kyrgyzstan. Nationalities Papers, 36(4), 641–657. DeYoung, A. J., Reeves, M., & Valyayeva, G. K. (2006). Surviving the transition? Schools and schooling in the Kyrgyz republic since independence. Greenwich, CT: Information Age Publishing. Dobson, R. (1977). Social status and inequality of access to higher education in the USSR. In: J. Karabel & A. Halsey (Eds), Power and ideology in education (pp. 232–253). New York: Oxford University Press. Drummond, T., & DeYoung, A. J. (2004). Perspectives and problems in education reform in Kyrgyzstan: The case of national scholarship testing. In: S. Heyneman & A. DeYoung (Eds), Challenges for education in Central Asia (pp. 225–242). Greenwich, CT: Information Age Publishing. EurasiaNet. (2004). Turkmenistan’s education system in downward spiral. Available at http:// www.eurasianet.org/departments/rights/articles/eav050504.shtml

296

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Fagerlind, I., & Saha, L. (1989). Education and national development (2nd ed). Oxford: Pergamon Press. Falkingham, J. (2000). Women in Tajikistan: Country gender assessments. Manila, Philippines: Asian Development Bank. Fraser, A. (1999). Becoming human: The origins and development of women’s human rights. Human Rights Quarterly, 853(1999), 853–906. Friedl, E. (1994). Notes from the village: On the ethnographic construction of women in Iran. In: F. Go¨c- ek & S. Balaghi (Eds), Reconstruction gender in the Middle East. New York: Columbia University Press. Gleason, G. (1997). The Central Asian states. Boulder, CO: Westview Press. Gleason, G. (2003). Markets and politics: Structural reform and political change in Central Asia. London: Routledge. Global Gender Gap Report. (2007). Country proﬁles. World Economic Forum, Geneva, Switzerland. Available at http://www.weforum.org/pdf/gendergap/report2007.pdf. Retrieved on 3/15/2008. Grenoble, L. (2003). Language policy in the Soviet Union. Norwell, MA: Springer. Harris, C. (2004). Control and subversion: Gender relations in Tajikistan. London: Pluto Press. Heyneman, S. (1998). The transition from party/state to open democracy: The role of education. International Journal of Educational Development, 1, 21–40. Heyneman, S. (2004). Education and corruption. International Journal of Education Development, 4(6), 638–648. Hijab, N. (1988). Womanpower: The Arab debate on women at work. Cambridge, UK: Cambridge University Press. Holmes, B., Read, G., & Voskresenskaya, N. (1995). Russian education: Tradition and transition. New York: Garland Publishing. Human Rights Watch. (1999). Class dismissed: Discriminatory expulsions of Muslim students (Vol. 11, no. 12). HRC Country Report, Uzbekistan, October 1999. Human Rights Watch. (2001). Sacriﬁcing women to save the family? Domestic violence in Uzbekistan. Available at http://www.hrw.org/reports/2001/uzbekistan/ Inkeles, A., & Bauer, R. (1957). The Soviet citizen. London: Oxford University. Islamov, E. (2004). ‘‘Uzbekistan’s corruption-ridden educational system seen as source of frustration.’’ Eurasianet.org, April 29. Jacoby, S. (1974). Inside Soviet schools. New York: Hill and Wang. Johnson, M. (2004). The legacy of Russian and Soviet education and the shaping of ethnic, religious, and national identities in Central Asia. In: S. Heyneman & A. DeYoung (Eds), Challenges for education in Central Asia (pp. 21–36). Greenwich, CT: Information Age Publishing. Keller, S. (2007). Going to school in Uzbekistan. In: J. Sahadeo & R. Zanca (Eds), Everyday life in Central Asia: Past and present (pp. 248–265). Bloomington: University of Indiana Press. Kelly, G. (1992). Education, women and change. In: R. Arnove, P. Altbach & G. Kelly (Eds), Emergent issues in education (pp. 267–282). Albany, NY: SUNY Press. Kerr, S. (1993). USSR. In: P. W. Cookson, A. R. Sadovnik & S. F. Semel (Eds), International handbook of educational reform (pp. 473–493). New York: Greenwood Press. Khalid, A. (1998). The politics of Muslim cultural reform: Jadidism in Central Asia. Berkeley: University of California Press. Korth, B. (2005). Language attitudes towards Kyrgyz and Russian discourse, education and policy in post-Soviet Kyrgyzstan. Berne: Peter Lang.

Newly Independent States

297

Kuehnast, K. (1997), Let the stone lie where it has fallen: Dilemmas of gender and generation in post-Soviet Kyrgyzstan. Ph.D. dissertation, The University of Minnesota, Minneapolis. Kuehnast, K., & Dudwick, N. (2004). Better a hundred friends than a hundred rubles? Social networks in transition – The Kyrgyz Republic. Washington, DC: World Bank. Liczek, I. (2005). Cultural parameters of gender policy making in contemporary Turkmenistan. Comparative Studies of South Asia, Africa and the Middle East. Duke University Press, 25(3): 567–583. Marat, E. (2007). State-propagated narratives about a national defender in Central Asian States. Journal of Power Institutions in post-Soviet Societies, Available at http:// www.pipss.org/document545.html. Retrieved on August 25, 2008. Massell, G. (1974). The surrogate proletariat. Princeton, N.J: Princeton University Press. McMann, K. (2007). The shrinking of the welfare state: Central Asians’ assessment of Soviet and post-Soviet governance. In: J. Sahadeo & R. Zanca (Eds), Everyday life in Central Asia: Past and present (pp. 233–247). Bloomington: University of Indiana Press. Mertaugh, M. (2004). Education in Central Asia, with particular reference to the Kyrgyz Republic. In: S. Heyneman & A. DeYoung (Eds), Challenges for education in Central Asia (pp. 153–180). Greenwich, CT: Information Age Publishing. Meyer, J., Ramirez, F., Frank, D., & Shofer, E. (2007). Higher education as an institution. In: P. Gumport (Ed.), Sociology of Higher Education (pp. 187–221). Baltimore: Johns Hopkins Press. Micklin, P. (1992). Water management in Soviet Central Asia: Problems and prospects. In: J. M. Stewart (Ed.), The Soviet environment: Problems, policies and politics (pp. 88–114). Cambridge: Cambridge University Press. Montgomery, D. (2007). Namaz, Wishing Trees, and Vodka: The diversity of everyday religious life in Central Asia. In: J. Sahadeo & R. Zenca (Eds), Everyday life in Central Asia (pp. 355–370). Bloomington: University of Indiana Press. Morozova, I. (2004). Legal systems and political regimes in post-socialist Central Asia. International Institute for Asian Studies (IIAS) Newsletter (34), July. Available at http:// www.iias.nl/nl/34/IIAS_NL34_16.pdf. Retrieved on 2/09/2009. Neweurasia.net. (2007) Being a girl in a post-Soviet state. April 3, 2007. Available at http:// uzbekistan.neweurasia.net/?p ¼ 164 NIEW (Nam Institute for the Empowerment of Women). (2008). Turkmenistan. Available at http://www.niew.gov.my/project/TURKMENISTAN.pdf. Retrieved on 2/10/2009. Niyozov, S. (2004). The effects of the collapse of the USSR on teachers’ lives and work in Tajikistan. In: S. Heyneman & A. DeYoung (Eds), Challenges for education in Central Asia (pp. 37–64). Greenwich, CT: Information Age Publishing. Northrup, D. (2001). Subaltern dialogues: Subversion and resistance in Soviet Uzbek family law. Slavic review, 60(1), 119. Northrup, D. (2007). The limits of liberation: Gender, revolution and the veil in everyday life in Soviet Uzbekistan. In: J. Sahadeo & R. Zanca (Eds), Everyday life in Central Asia: Past and present (pp. 89–102). Bloomington: University of Indiana Press. Olcott, M. (1987). The Kazakhs. Stanford, CA: Hoover Institution Press. Olcott, M. (2005). Central Asia’s second chance. Washington, DC: Carnegie Endowment for International Peace. OSI. (2003). Educational development in Kyrgyzstan, Tajikistan and Uzbekistan: Challenges and ways forward. Available at http://www.osi-edu.net/esp

298

ALAN J. DEYOUNG AND ELIZABETH A. CONSTANTINE

Osmonov, O. (2003). History of Kyrgyzstan: From the middle of the 19th century – Until the present. History textbook for the 11th grade, as approved by the Ministry of Education and Culture of the Kyrgyz Republic, Bishkek, KG. Pitt, J., & Pavlova, M. (2001). Pedagogy in transition: From labor training to humanistic technology education in Russia. In: S. Webber & I. Liikanen (Eds), Education and civic culture in post-communist countries (pp. 231–247). London: Palgrave. Ramirez, F., & Wotipka, C. M. (2001). Slowly but surely: The global expansion of women’s participation in science and engineering ﬁelds of study, 1972–1992. Sociology of Education, 74, 231–251. Roberts, S. (2007). Everyday negotiations of Islam in Central Asia: Practicing religion in the Uyghur neighborhood of Zarya Vostoka in Almaty, Kazakhstan. In: J. Sahadeo & R. Zanca (Eds), Everyday life in Central Asia: Past and present (pp. 339–354). Bloomington: University of Indiana Press. Roy, O. (2000). The new Central Asia: The creation of nations. London: Tauris. Rywkin, M. (1990). Moscow’s Muslim challenge (2nd ed.). Armonk, NY: M.E. Sharpe. Sara, M. H. (2006). Resurgence of Islam in Central Asia: A case study of Uzbekistan. Strategic Studies, The Institute of Strategic Studies. Available at http://www.issi.org.pk/journal/ 2006_ﬁles/no_2/article/a7.htm. Retrieved on June 13th, 2008. Silova, I., Johnson, M., & Heyneman, S. (2007). Education and the crisis of social cohesion in Azerbaijan and Central Asia. Comparative education review, 51(2), 159–180. Smith, J. (2001). National schools and national identity in and after the Soviet Union. In: S. Webber & I. Liikanen (Eds), Education and civic culture in post-communist countries (pp. 55–73). New York: Palgrave. Spradley, J. (1979). The ethnographic interview. New York: Holt, Rinehart and Winston. Spring, J. (2001). Globalization and educational rights. Mahwah, NJ: Lawrence Erlbaum Associates. Stromquist, N. (2002). Education in a globalized world. Oxford: Rowan and Littleﬁeld Publishers. UNESCO. (2003/2004). EFA Global Monitoring Report. Gender and education for all: The leap to equality. Available at http://portal.unesco.org/education/en/ev.php-URL_ ID ¼ 24284&URL_DO ¼ DO_TOPIC&URL_SECTION ¼ 201.html UNESCO. (2006). Literacy for life; Background paper prepared for the Education for All Global Monitoring Report, 2006. Available at http://unesdoc.unesco.org/images/0014/ 001459/145934e.pdf UNESCO. (2008). Institute for Statistics; UIS Statistics in Brief: Education in Kyrgyzstan; Kazakhstan; Uzbekistan; Tajikistan; Turkmenistan. Available at http://stats.uis. unesco.org/unesco/TableViewer/document.aspx?ReportId ¼ 121&IF_Language ¼ eng& BR_Country ¼ 4160. Kyrgyzstan ¼ 4160; Kazakhstan ¼ 4020; Uzbekistan ¼ 8060; Tajikistan ¼ 7610; Turkmenistan ¼ 7940. Retrieved on 1/7/2008. UNICEF. (2007). Education for some more than others? Geneva, SW. Available at http:// www.unicef.org/media/ﬁles/Regional_Education_Study_-pdf Webber, S. (2000). School, reform and society in the New Russia. New York: St. Martin’s Press. Weidman, J., Chapman, D., Cohen, M., & Lelei, M. (2004). Access to education in ﬁve newly independent states of Central Asia and Mongolia. In: S. Heyneman & A. DeYoung (Eds), Challenges for education in Central Asia (pp. 181–197). Greenwich, CT: Information Age Publishing.

Newly Independent States

299

Werner, C. (2003). Between family and market: Women and the new silk road in post-Soviet Kazakhstan. In: G. Clark (Ed.), Gender at work in economic life (pp. 115–124). Lanham, MD: Altamira Press. World Bank. (2002). Gender in Transition. Washington, DC, May 21. World Bank. (2006). Migration and remittances: Eastern Europe and the former Soviet Union. Available at http://siteresources.worldbank.org/INTECA/Resources/ 257896-1167856389505/Migration_FullReport.pdf. Retrieved on 6/6/2008. World Bank. (2008). Rural development (regional web proﬁles). Available at http:// web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/ECAEXT/EXTECAREG TOPRURDEV/0,,contentMDK:20530216BmenuPK:1893744BpagePK:34004173Bpi PK:34003707BtheSitePK:573582,00.html World Economic Forum. (2007). The Global Gender Gap Report 2007, Geneva. Available at http://www.niew.gov.my/project/TURKMENISTAN.pdf. Retrieved on 8/26/2008. World Economic Forum (WEF). (n.d.). Available at http://www.weforum.org/en/about/ Our%20Organization/index.htm. Retrieved on 8/26/2008.

ISSUES OF GENDER, EQUALITY, EDUCATION, AND NATIONAL DEVELOPMENT IN THE UNITED ARAB EMIRATES Daniel Kirk and Diane Napier INTRODUCTION In recent years, interest in educational issues in the Muslim world has grown rapidly. This interest runs parallel with a media-led exploration of all things Muslim and the ideas that are fundamental to the tenets of Islam. This trend in examining the educational issues that exist in countries that are predominantly Islamic in history, culture, and belief, is part of a wider awareness of the educational elements of a globalized system of commerce, communication, education, and modernization. The United Arab Emirates (UAE) has embraced many facets of globalization, striving to become a regional power and a new ﬁnancial and commercial hub in the Middle East and a high-tech center in a globally oriented society. Along with other Arab nations, the UAE has recognized the strategic role played by education in national development and modernization. Although Arab countries cover an expansive geographical area, and are diverse in terms of traditions, they share many education-related issues and challenges, including issues of access to educational provision for minority

Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 301–331 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010012

301

302

DANIEL KIRK AND DIANE NAPIER

groups and females. Along with the diversity and multifaceted aspects of the Arab World, as Kirdar (2006) noted it is necessary to dispel the notion of the ‘Arab woman’ as a subjugated, powerless person (p. 191). Together with the sharp differences that exist between the various countries in the region, there exist common elements, such as language, cultural heritage, and the adherence to Islam. These ingredients are often used to talk about Arab nations, and their people, in a very homogenous and limiting way, while in fact each nation is distinctive with its own set of internal contextual factors and circumstances shaping the details of development. The UAE is undergoing rapid change in all sectors. Attempts to keep pace with the growing demands of the economy and the national workforce have led to rapid growth in educational provision (at primary, secondary, and to a lesser extent higher education levels), yet access is not equally distributed by gender, and the roles of religion and culture in the local society also inﬂuence female perceptions of and participation in education. In the context of the theme of this volume, the case of the UAE is an interesting one in that recent developments in its education sector reﬂect many of the issues being examined worldwide with regard to education and gender, within larger considerations of national development and equity, as well as the current situation in educational and societal development in the region of the Middle East. Consequently, in this chapter we frame our treatment of the UAE case within the global and regional contexts and the dominant trends therein, and in appropriate historical context. The case of the UAE is signiﬁcant, given its dichotomous relationship between Western systems of capitalism and commerce, and traditional Islamic society and norms. This leads to an education system that has stated goals to become ‘‘world class’’ and competitive in the global arena, one that also aims to preserve desirable aspects of traditional society and that operates within a cultural system of gender inequity in terms of educational access to career paths for young women. The choices made by the federal government of the UAE to allow Western higher education institutions and private grade schools into the country to serve expatriate groups have complicated the issue. Many of these schools are coeducational and they are becoming more attractive to wealthy nationals who want their daughters to be educated along Western lines. This is generating conﬂict and debate in leadership circles, the wider society, and in particular among the religious elite, placing female educational access at the heart of the discussions about educational and national development in the UAE. The UAE shares educational development challenges with many or most developing countries in which globalization inﬂuences and economic imperatives are shaping dramatic

Issues of Gender, Equality, Education, and National Development in UAE

303

changes in education with regard to its roles and functions. Worldwide, teacher education programs are buffeted by an array of such inﬂuences and pressures to conform to global standards and evaluation systems, within developing countries’ larger agendas for national development, as noted by Gal (2005). Given these interwoven dimensions in the case of the UAE, it is necessary to consider how they emerge as processes, realities, and outcomes regarding gender, equity, education, and overall development. The focus of this chapter is to provide an examination of these interlocking issues in the rapidly changing hybridized education model operating in the UAE, and to consider the central questions of whether (or not) gender equality in education is really being achieved in the UAE, if so, how, and whether or not current policies and realities in the UAE suggest hope for gender equality in the future. Because educational participation and opportunities for females are at the cutting edge of change and debate in the UAE, we do focus primarily on considerations for females regarding the overriding questions related to gender equity and change, although we do not ignore considerations related to males who have been the dominant presence in education in the UAE. We concur with the arguments made by Findlow (2000) and Kirdar (2006, 2007) that new conceptualizations related to girls and women are needed in the context of the UAE and other Islamic states, that research on the experiences and perceptions of females shows tensions between adherence to traditional female roles in the society and new roles linked to education and career development. The insights and examples we offer in this chapter reinforce these notions of partially shifting roles and increased participation, but persistent adherence to societal norms that perpetuate situations which appear as inequities to cultural outsiders. In the process, we contemplate some issues of policy versus practice, in that modernization and Westernization thrusts in UAE teacher education programs might not necessarily translate into substantive change in schools and in teachers’ practices. Initially, we present the theoretical perspectives that we ﬁnd useful as lenses through which to examine the issues in the UAE with regard to the global context as well as the regional and local contexts, given the fact that so many of the dilemmas and challenges in the UAE have counterparts elsewhere in the world and in the region. Thereafter, we provide an historical overview of education and development in the UAE as a post-colonial state and as a wealthy oil-producing state, within the regional context. We follow this with a brief overview of the country’s national development goals including Emiritization, and how educational expansion with equitable access across

304

DANIEL KIRK AND DIANE NAPIER

gender lines forms part of the new self-actualization goal for the country. Contained within the latter are many issues including gender inequality, gender segregation, single-sex schooling, the role of religion in statesponsored schooling, trends in schooling and their impact on development needs in higher education, and the conundrum of weighing these against internal (emic) standards as opposed to against external (etic, and Western) standards. We contemplate the processes of developing education in the UAE for speciﬁc purposes, including the manner in which females are viewed in particular ways with regard to appropriate career paths (mainly as teachers) and to the question of what emic views surface regarding desirable traits of ‘‘the new Arab woman’’ previously considered downtrodden. We cite examples from the ongoing empirical research to highlight some perspectives voiced by female pre-service teachers who are enrolled in teacher education programs that were imported, practically intact in form, from the United States, to train teachers for schools that remain traditional, Islamic, and nonWestern. We touch on curriculum and language issues as they relate to these issues. We end the chapter with some critical considerations of gender-related implications of educational development linked to societal development, and of the challenges facing the country as it strives to modernize, to afford equitable participation, to resolve internal dilemmas and contradictions such as in training an indigenous workforce yet having an abundance of resources for hiring expatriate labor, and to maintain a judicious balance of modernity and traditionalism. It is too early to form any deﬁnitive judgment regarding the consequences of the very young educational programs under implementation in the UAE, particularly for those in teacher education that are discussed here. However, the case of the UAE offers compelling glimpses into a range of very complex challenges with regard to education, development, and gender equity/inequity that will warrant long-term study for full understanding. To provide some general context for our discussion, in the following two sections, we offer a summary overview of general features of education and Islam in historical context, and of the general features of education for females in Islamic societies such as the UAE.

EDUCATION AND ISLAM: THE HISTORICAL CONTEXT In all of its varieties across the Muslim world, Islam has had a long history of educational development and the promotion of learning opportunities.

Issues of Gender, Equality, Education, and National Development in UAE

305

Muslim society places a high level of esteem on the ability to read and memorize the Quran, and it is an ideal that every Muslim is expected to master the Arabic language with sufﬁcient proﬁciency to be able to understand the messages held within the holy text. It is this historical and ideological imperative which places education at the forefront of Islam. Promotion of education and learning was a major factor contributing to the spread of Islam worldwide, inspiring believers to acquire knowledge and carry out scientiﬁc investigation (Daun, Arjmand, & Walford, 2004). In the global history of educational development, Muslim scholars made notable contributions in ﬁelds such as medicine, astronomy, mathematics, geography, history, and the sciences. Under the patronage of Islam, monitored and supported by regional and religious leaders, the Arabic language and literature developed, ﬂourished, and left an indelible mark on the great civilizations and artistic endeavors of the world. During the 15th and 16th centuries European scholars began to show an interest in the intellectual realm of the Muslim world. Professorships of Arabic were established at the universities of Oxford, Cambridge, Salamanca, Rome, and Bologna. Muslim scholars contributed to the academic and intellectual awakening in Western Europe. In fact, Islam and Christianity progressed alongside each other for many centuries. However, following the Renaissance, and continuing through the Industrial Revolution and the formation of strong Western nation-states, the Muslim world lost ground in terms of development in most areas, including education. As described by Singh (2006), the prominence of Islam declined as Islamic civilizations clung to antiquated systems and decayed institutions, cut off from external contact, while the Christian West rose in intellectual prominence. This had an impact on the development of formal education systems in Islamic states, particularly those on the Arabian Peninsula, where formalized nation-states did not begin to develop until late into the 19th century. Schooling was traditionally carried out by Ulama (Islamic scholars) under the patronage of local mosques, and it was usually limited to the male offspring of the head tribesmen or local dignitaries. Education was valued and, as mentioned earlier, was viewed as a way for the teachings of the Quran to be understood and passed on. Yet outside of the established Arab nations, such as Egypt, Syria, and Iraq, schooling was an informal and localized practice, resulting in extremely high rates of illiteracy, especially among girls (Reagan, 2005). As we enter the 21st century, education across the Middle East and Arab world is in a state of disarray or ﬂux (Christina, Mehran, & Mir, 2003), beset by tensions between traditionalism and modernity, but with educational development and delivery high on the agenda of many developing countries in the region.

306

DANIEL KIRK AND DIANE NAPIER

FEMALE EDUCATION IN MUSLIM SOCIETIES Historically, girls in Muslim societies have had less opportunity to take part in formal education than those afforded to their male peers. Educational opportunities for women are also limited by the conservative nature of Arab society, which favors certain activities and professions for the female members of society (Azzam, Abu Nasr, & Lorﬁng, 1985). The low percentage of female students generated issues of gender inequality and inequity, but these need to be examined in relation to the unique contextual settings found in Islamic society. The role of Islam in this process is important, as Islam is not only a spiritual religion and code, but is also the basis for the legal, social, and moral codes of behavior within the social group (Kirdar, 2006). As Kirdar notes, due to various interpretations of Islam, and differing levels of adherence to the Quran, leads to the position of women in the Muslim world must be regarded as neither static nor uniform. Correspondingly, educational opportunities for females are complex and varied. Moreover, as Kirdar argues, there is also the question of purpose for which Muslim women are being educated: are they being educated to meet the needs of the modern economy or because it is their fundamental human right to (also) be educated? In the case of the UAE, this question is a compelling one, perhaps most compelling for national leaders, and the women of the UAE themselves, to answer. Across the Muslim, and more speciﬁcally the Arab world girls rarely receive the same educational opportunities and experiences as their male relatives and friends. Those who do follow an educational route into further and higher education can be found entering the very ﬁelds of study that reinforce gender divisions within the society, namely the ﬁelds of teaching, nursing, and the humanities (Davies, 1996; Kirdar, 2007). These particular professions and ﬁelds are viewed as being of a lower status than the so-called hard sciences, engineering, business, and medicine. As a result of these perceptions, certain professions and ﬁelds become segregated by gender. Female enrollment remains concentrated in the humanities and education ﬁelds, while women were largely absent in ﬁelds that are crucial to national development goals, such as agriculture and engineering. Simply stated by Shaw (2006), gender matters in the Muslim world. Although girls do better academically when compared to boys, their employment opportunities are still limited in keeping with the cultural and religious beliefs of the patriarchal society. Structural, attitudinal, and cultural barriers stand in the way of equitable participation of women at the higher levels of education, mainly secondary

Issues of Gender, Equality, Education, and National Development in UAE

307

and post-secondary education. Often the decision to enter higher education is made by the family, following traditional patriarchal patterns of male dominance within the family unit. The underrepresentation of women in higher education, and later in the workforce, is not a result of any deﬁciency in academic ability. Rather, it was a symptom of cultural heritage and social structures. In many Arab countries, and in particular the Gulf states, high levels of welfare payments and social service provision for nationals are ﬁnanced through oil revenues, making it possible for women in such countries to remain in the traditional position of housewife regardless of how highly educated they are. This system of public beneﬁts, although beneﬁcial to the citizens and necessary for the ruling elite to remain in power, does little to encourage women to take a more active part in the development of a nationalized workforce. Nonetheless, the gender divisions in UAE society are becoming less rigid in some respects as women begin to penetrate the barriers of educational and professional participation in the modernizing society.

THEORETICAL PERSPECTIVES FOR EXAMINING EDUCATIONAL DEVELOPMENT, MODERNIZATION, BORROWING, AND INTERNAL–EXTERNAL INFLUENCES IN THE UAE Following the consideration of education and Islam, and female education in Muslim societies, earlier, we turn to an overview of several theoretical models that are useful for understanding the special considerations of the UAE case, in global and regional context and in terms of external versus internal dimensions and forces. We employ an eclectic theoretical perspective in this chapter, since several complementary theoretical models explaining reform, borrowing, and implementation processes facilitate understanding of the apparent contradictions and the different dimensions inherent in the UAE case. First, because the UAE, and the countries from which it is borrowing educational policies AND programs are largely top-down systems, it is necessary to consider how policies and practices are ﬁrst, borrowed or imported by the UAE from other countries; second, how reforms and programs in both are implemented in largely top-down fashion such as in the policies of No Child Left Behind (NCLB) in the United States, albeit a country with a largely decentralized system; third, what the motives and

308

DANIEL KIRK AND DIANE NAPIER

processes of implementation are; and fourth, how these are then adopted, modiﬁed, and variously implemented down to the local level. Embodied in most global–local models, and in schemes of thinking about educational policy borrowing, are the underlying ideas of structural functionalism as the framework of centralized systems (and of persistent top-down policy implementation), as well as notions of how schooling worldwide came to assume a set of standardized forms based on the Napoleonic model, how Western-style democracy inﬂuenced the path of educational modernization in many or most developing countries, and how countries (including the UAE) readily engaged in borrowing or importation of educational ideas and practices from other countries regardless of their having very different contexts and histories (see, for instance, Kelly & Altbach, 1982; McGinn & Cummings, 1997; Ramirez & Boli, 1987). Much research has been conducted on the peculiar needs of developing countries seeking both to modernize and to become competitive in the global arena as well as to preserve desirable traits of traditional society and culture, and their proclivity for importing educational programs and reform packages (as opposed to developing internal, indigenous, and contextually relevant equivalents) (see Brook Napier, 2005; Jansen, 2004; Phillips & Ochs, 2004). The UAE might be considered one of these countries. Differences between the oil-rich states in the Gulf (the UAE, Saudi Arabia, and Kuwait) and the non-oil-rich states (Egypt, Jordan, Qatar) make for interesting comparisons in terms of the choices made by these various states when resources were abundant or not, and in terms of decisions made to develop and modernize education systems as a result. The pressures of globalization and the emergence of a global system of education, complete with priority subjects, standards and testing, accountability systems, and ﬁerce competition amidst cross-national comparisons of achievement, add to the need for understanding how a country like the UAE has made decisions with regard to educational development as it participates in an increasingly convergent world. Developments in the UAE in this regard illustrate aspects of the World Systems Theory with regards to trends in educational policy, summarized by Baker and Wiseman (2005). Related to this, externalization theory seeks to explain the processes of borrowing, and the tensions between globalizing forces and local traditions, identities, and needs within a given country. This scheme of thinking (as portrayed by Steiner-Khamsi, 2002 and Phillips, 2004) assists one’s considerations of how a ‘‘peculiar set of circumstances’’ in a given country such as the UAE can determine whether a self-referential (internal) approach, or an externalization (borrowing) approach prevails with regard to educational development,

Issues of Gender, Equality, Education, and National Development in UAE

309

or that some combination of the two might emerge. The UAE might be seen to illustrate the issue of a ‘‘peculiar set of circumstances’’ in which oil revenues, a culture of privilege for UAE nationals, and desire for rapid modernization inﬂuenced by Western-educated elites, translated into the adoption of a hybrid model of education in which many or most of the new ingredients were imported. Further, Phillips’ (2004) articulation of several stages of the borrowing process (cross-national attraction, decision, implementation – whether mediated, resisted, or supported locally – and full indigenization or internalization) is useful to considering the degree to which borrowed or imported programs in the UAE (such as the teacher education programs imported from the United States, discussed later in this chapter) are likely to be indigenized and internalized or by what means they might be only partially implemented. A complementary model is that of the global to local continuum and the dialectic of the global and the local as articulated by Arnove and Torres (2003) and others, in which global systems of ideas and imperatives are adopted at the national level, then implemented in some form at lower levels such as provincial or regional, and further they might penetrate to the local and classroom levels with a wide array of modiﬁcations, processes, and possibilities between the two extremes. Similarly, World Culture Theory is a largely anthropological perspective that seeks to explain the same scheme but with an emphasis on culture and transmission, and also on the processes of ‘‘creolization’’ and ‘‘recreolization’’ that can occur as global, external, imported programs and policies are enacted at the national level and then at lower levels in a given country. The work of Anderson-Levitt, Bartlett, and Brook Napier (see Anderson-Levitt, 2003) illustrated these processes as they applied to several countries such as Brazil and South Africa, with the principles applicable to the case of the UAE in terms of how imported programs interplay with local inﬂuences and forces on various levels. For instance, by the time that imported ideas for training of new UAE teachers reach these pre-service teachers, and when these new teachers perhaps take the ideas out into local schools, what are the outcomes and issues, given that the teachers will be teaching students in traditionally focused classrooms, in Arabic not English, and when the local cultural context is markedly different to that for which the teacher training programs were developed in the United States? Here, the tensions between educational borrowing and internal contextual factors are manifold. Finally, in examining educational development in the UAE, considerations of educational empowerment of Arab women are demanded in this discussion. As noted by Kirdar (2007, p. 49) it is simplistic to consider Arab

310

DANIEL KIRK AND DIANE NAPIER

women as simply downtrodden and inferior, or to engage in binary thinking by considering their empowerment and development as either Western or Eastern. We recognize the danger of considering that inequitable access to (all) professional career paths for Arab women in a country like the UAE might well be a slanted notion seen through Western eyes. Therefore, we acknowledge that there are different interpretations of the image of girl or woman in this context and that the possible roles of empowered, educated Arab women in the case of the UAE might better be viewed through a different lens, that of a new form of modern female, emerging from special circumstances of traditionalism and modernity within the societal context, as described by Kirdar as a ‘‘ﬁrst generation professional, Arab, Muslim women.’’ We also acknowledge what has been argued by scholars such as Findlow (2007) whose research focused on UAE women students and graduates, that the story is complex and that women’s and girls’ roles in a modernized UAE state are multifaceted, not uniform. This ﬁts with the notion explained by Steiner-Khamsi (2002) that educational borrowing (such as of Western teacher education programs by universities in the UAE) might well be evident at the policy and program level, but not at the level of discourse (among the students themselves, for instance). Similarly, it has been shown that there is something of a mismatch between women’s limited opportunities and female students’ achievements, as noted by Mazawi (2007, p. 77 citing numerous scholars’ work in the Gulf states). The gendered inequality in segregated institutions is juxtaposed with increasing visibility of females in the system, and with new opportunities for empowerment and participation. As noted by Mazawi (2007, p. 86, citing Salloum, 2003, p. 104), ‘‘women’s academic opportunities are ‘assembled’ by policymakers, faculty members, and consultants in a network of private international higher education institutions and at the center of the debate stands the political construction of the ‘‘ ‘ideal-typical’ modern-yetauthentic-Gulf-woman who represents a cultural project of modernity and renaissance endorsed and promoted by a benevolent state.’’ Therefore it is even more necessary to consider, in the case of the UAE, how female preservice teachers and other professional women ﬁt into the larger scheme of development from their own perspectives and from the (sometimes apparently contradictory, perhaps not always) perspectives voiced at the policy level. For illustration, we offer some accounts by university students of how they see themselves as well as their hopes and aspirations, in terms of these ironies and contradictions. Clearly, our view of women and girls in the complex terrains of schooling and higher education in the UAE has to take into account such notions that point to very different constructions of an

Issues of Gender, Equality, Education, and National Development in UAE

311

average pre-service teacher, or of a teacher, or of a female student in the UAE in comparison to those in countries elsewhere without the equivalent complexities of context. We recognize that it is important to avoid judgment about internal cultural practices that do not mesh with non-Islamic norms and practices. In the discussion to follow, we present an overview of the historical context for the UAE to demonstrate the extreme youthfulness of the new state, and the rapid pace of contemporary growth built on a long historical and cultural heritage shared with neighboring states.

THE UNITED ARAB EMIRATES: THE REGIONAL CONTEXT The Gulf Cooperation Council (GCC) (Bahrain, Qatar, Saudi Arabia, UAE, Oman, and Kuwait), was formed in 1981 to foster closer collaboration and ties between its member states, and within them the historically linked tribal/ family groups that had been artiﬁcially separated by the creation of colonial states. The member countries have a shared cultural, religious, historical legacy (Arab identity and culture, Islam, colonial legacy) as well as common economic and political interests. Among them, Saudi Arabia, the UAE, and Qatar are oil-resource rich, which makes them somewhat different to most developing countries in that they have signiﬁcant revenues for capacity building and development. Among the aims of the GCC are the improvement of educational provision and access at all levels in the member states, through regional collaboration, promotion of Gulf Arab educational opportunities, and in general the promotion of national and regional development, modernization, and global competitiveness. In the Gulf region, the countries of Egypt, Jordan, and Kuwait spearheaded the development of higher education, given their relatively established development infrastructures. University education was linked to and provided for in response to human resource development (HRD) needs. In Egypt and Jordan, non-oil rich countries, the focus was heavily on capacity building and development of human resources, and higher education included cultural, moral, and religious aims in programs. In contrast, the oil-rich nations of Saudi Arabia, Qatar, and the UAE sought to develop higher education to reduce their reliance on expatriate personnel, and to foster cultural transmission and modernization. As has occurred in most countries, the development of higher education in this region occurred in response to national development goals as well as to the rapid expansion of primary and secondary education provision.

312

DANIEL KIRK AND DIANE NAPIER

Another ingredient was that of post-colonial development, in which these modern states were inﬂuenced by their respective colonial legacies and ties to Britain, France in particular. ‘‘Modern’’ Western-style education only came into being in the post-colonial era, from the 1950s onwards, and as secular education in contrast to the ‘‘mutawwa’’ or religious schools. Educational development in the region is heavily impacted by the ‘‘ideological battle . . . of balancing traditional values and the contemporary scene’’ (Mina, 2006, p. 223; Grifﬁn, 2006, p. 22). Development of higher education was gradual, and then it accelerated in recent decades. Acute shortages of indigenous educated, skilled workers plague each of the countries, while individual country differences are also noted, for instance, in the large population and high unemployment rate of Egypt compared to much smaller population of Jordan and the UAE. In the UAE, because of the shortage of skilled and qualiﬁed nationals and the availability of revenues to pay for importing workers, there is heavy dependence on expatriate or foreign workers in all sectors, so too in education. Teacher education poses a signiﬁcant challenge for each of these states since it is relatively underdeveloped. The need to develop an indigenous, qualiﬁed teaching force is an important goal in the UAE and several other member states. The UAE is a relatively new nation, formed in December 1971, through the union of seven existing sheikhdoms; Abu Dhabi, Dubai, Sharjah, Umm al-Quwain, Ajman, Fujairah, and Ras al-Khaimah. The country is predominantly desert, an arid area of land bordered by Saudi Arabia, Oman, Qatar, and both the Arabian Gulf and Gulf of Oman. Until relatively recent times, the emirates were characterized by long-standing disputes, feuds, and rebellion, coupled with ﬁghting between the various tribal groups. The impetus for setting aside intertribal feuding, and for uniting the sheikhdoms was the discovery and commercial development of proﬁtable oil ﬁelds, and the realization by several sheikhs of the dangers inherent in small, separate, oil-rich emirates attempting to survive in a volatile region of the world, thus attempting to keep the ‘‘bullies at bay’’ (Al-Kaabi, 2005). Since its formation as a modern state, the UAE has developed a system of governance that reﬂects the teachings and interpretation of Islam to which the dominant group, Sunni Muslims, adhere. The scope and inﬂuence of religion underpins all aspects of society in the UAE, and social policy is based on religious doctrine (Findlow, 2000). Although religion plays a central role in the development of the UAE, it is rarely as overt an inﬂuence as can be seen in other Arab states, such as Saudi Arabia, where religion and governance are consciously made visible and interlinked. In the UAE

Issues of Gender, Equality, Education, and National Development in UAE

313

religion is formally institutionalized, pervasive in the fabric of society and everyday life in the country, and ingrained in Emirati cultural identity. The exploitation of oil in the UAE fueled a rapid period of growth and modernization in the country. The UAE, traditionally dependent on ﬁshing, pearl diving, trading, and wadi agriculture, was transformed from a sleepy desert backwater to a complex, modern, consumer economy supported by developed infrastructure and communications systems. The country embarked on developing a diversiﬁed economy and the major cities of Abu Dhabi, Dubai, and Sharjah developed extensive links with international trade as part of a larger plan to position the UAE as a new ﬁnancial and commercial hub serving the wider Middle East region. Despite efforts to diversify the economy, oil remains its mainstay and, like wealth, the oil is not evenly distributed between the seven emirates (Taifour, 1994). Despite this imbalance, the federation remained ﬁrmly resolute to develop nationally, with the richer emirates subsidizing the poorer areas of the country. Under the presidency of the late Sheikh Zayed bin Sultan al-Nahayan, who was the president of the UAE from its formation in 1971 until his death in 2004, the ruling elite balanced traditional elements of the various tribal groups, although developing the country to suit the needs of national citizens and establishing its position and inﬂuence in both regional and international organizations (Peck, 1986). The fast-paced economic expansion of the country has beneﬁted the citizens of the UAE, but growth has been reliant on imported foreign workers who have the necessary skills and were attracted by the high salaries and beneﬁts that expatriates could demand. The result of this practice means that the UAE relies almost entirely on expatriate labor in all sectors of the economy, with the exception of government sector positions, which actively recruits nationals. UAE nationals, therefore, make up only 10% of the workforce in the country (Al-Kaabi, 2005). The most recent census data, from 2005, shows that although the growth in population of the UAE topped 4 million, the number of UAE nationals consist of only 20% of those living in the country (Abdullah, 2007). This situation, in which Emiratis are outnumbered twice over by foreigners, once in the general population and again in the workforce, is an ongoing concern for the leaders of the country as well as the citizens themselves. There is a perception among Emiratis that too much of the nation’s present and future remains in the hands of foreigners. The government introduced a campaign to attempt to attract larger numbers of Emiratis to enter the workforce and, therefore, play a more prominent role in the machinations of the state (Shouly, 1995). The aim of this program was

314

DANIEL KIRK AND DIANE NAPIER

to educate and train Emiratis to make them more competitive in the job market and to allow a greater proportion of nationals entry into the economy. This effort placed a tremendous strain on the national education system to produce a growing number of nationals who are ready and willing to enter the workforce, including larger numbers of national teachers into the nation’s schools (Al-Kaabi, 2005). An argument that is often made, and that relates to the teacher situation in the UAE, is that one of the key elements to the development of an indigenous workforce is the nationalization of teachers and the production of a teaching force that is dominated by national citizens (Findlow, 2000). Teachers hold a special place in the development and building of societies and this is a central element of nation building and HRD. Efforts to attract more nationals into the teaching profession has become a priority of government and is driving various incentives and promotions to market teaching as an attractive career, although this has had little success to date.

EDUCATION IN THE UAE Since the founding of the UAE, a little over 25 years ago, education was seen as a fundamental element of the nation’s development plans, with the recognition that investment in HRD would help secure the future workforce needs of the national economy (Al-Kaabi, 2005), as well as support the development of a coherent and ‘‘new’’ society, formed from the old tribal groupings. Education was viewed by the new leadership as playing a key role in the development aims of the country, alongside establishing a new national identity and coherent society. Alongside the perceptions related to education that the government hold, public education occupies an important role and position among the citizenry of most nations (Tischer & Wideen, 1990). In fact, as Weber (2007) argues, referencing McGovern, education has a role in the political and social development of a society, stating that Politically, education aids in generating a sense of national identity among people living in a country . . . the diffusion of certain knowledge through a formal education is believed . . . to intensify the modernization or development of traditional societies. (McGovern, 1999, p. 8)

Initially, teachers were recruited from Arab Muslim countries to ﬁll the need for a teaching force in the rapidly expanding school system (Findlow, 2000). The education system continued to evolve and develop and there is now a system of schooling that is universal and free to all citizens from

Issues of Gender, Equality, Education, and National Development in UAE

315

kindergarten through to university. Compared to other countries, the education system of the UAE is a relatively new one that has seen rapid investment and growth over the past two decades. The education sector is dichotomous, as it consists of two distinct groups of schools: public and private. The public sector is funded by the central government, with some discretionary local funding often provided by the individual emirates where the school is located. Public schools are open to all UAE national citizens, and the structure and curricula is based on Islamic and Arabic principles (Gaad, Arif, & Scott, 2006). The public sector is administered by the central government and each school is separated by gender, with either different schools for boys and girls, or separation through school design and regulation. With very few exceptions, public schools are not open to the children of expatriates in the country. This creates a contrast between a wholly national student body and a teaching force reliant on expatriate faculty. Another factor in the public/private school contrast is that government schools are primarily Arabic language medium, although English is taught to a limited extent in high school. The decision to promote and mandate English language instruction in all schools would have been met with some resistance in the UAE, as the government moved away from a system of schooling based on the Quran. Findlow sees the expansion of education in Arab countries, and the Gulf in particular, as being characterized by Tension between imitation of, and resistance against, Imperial models . . . [as while reforming] its own education system governments have had to decide whether to follow the inherited colonial model, to opt for the nearest Arab cultural inﬂuence . . . or attempt to build a more or less indigenous framework. (2000, p. 5)

The UAE decided to draw upon the Egyptian system as a model for development, and this was apparent in the curricula and stafﬁng of public schools and the dominance of Egyptians in the UAE Ministry of Education (Findlow, 2000). However, the UAE government put pressure on the Ministry to modernize and reform, especially in light of the governmental aim to Emiratize the national workforce. This reliance on expatriate faculty in the national schools has been an issue since the formation of government schools. Since the 1950s, the UAE has been reliant of expatriate faculty for schools, with a large proportion of the teaching staff at all levels drawn from Egypt, where there is a surplus of teachers. The recruitment of teachers was, originally, focused on other Arab Muslim countries, as the teachers were accessible, relatively inexpensive, and culturally close to Emirati society (Findlow, 2000). In more recent times,

316

DANIEL KIRK AND DIANE NAPIER

however, large numbers of faculty from North America, Britain, Australia, Canada, and New Zealand have been recruited to staff the growing higher and tertiary level sector. The government does not see any conﬂicts of interest with the recruiting of such faculty and their national development goals, since they view such staff as temporary, but necessary to assist with the Emiritization process.

RAPID GROWTH IN UAE EDUCATIONAL PROVISION The UAE has seen dramatic and rapid growth in the numbers of students enrolling in the secondary level schools (Fig. 1). This rise in school numbers is a result of increased spending on education as well as the recognition that literacy rates in the country were not comparable to worldwide averages. This rise led to an increase in the need for university and higher education places, a need that could not be met with the only university in the country, United Arab Emirates University (UAEU), prior to the opening of alternative institutions. This growth in student numbers allowed the development of higher education to gain ofﬁcial approval and institutions, both private and public, began to appear in the country.

Fig. 1.

UAE Secondary School Graduates.

Issues of Gender, Equality, Education, and National Development in UAE

317

In 1978, the UAE government approved the formation of the UAEU with a mission to ‘‘meet the educational and cultural needs of the UAE society by providing programs and service of the highest quality’’ (UAEU, www.uae. ac.ae). The UAEU, located in Al Ain, set out to replicate older, more established institutions such as those found in Egypt and Kuwait, but with the needs of the national population ﬁrmly in mind. The UAEU, following cultural norms, became a segregated campus, with male and female students studying separately, often in identical buildings, and sharing a faculty. As the need for university places grew, the federal government, the Supreme Council, began to explore the formation of other national institutions. This development ran parallel with several rulers of emirates becoming interested in having higher education institutions under their control in individual emirates. This led to the system that is in place today, with federal institutions (e.g., UAE University and Higher Colleges of Technology (HCT)), emirate-funded institutions (e.g., American University of Sharjah, Ajman University of Science and Technology) and private, foreign institutions opening campus operations under the patronage of federal or emirate government (e.g., British University in Dubai, George Mason University (Ras al-Khaimah) and Sorbonne University (Abu Dhabi). The federal government also created a Ministry of Higher education, which oversees and accredits all of the institutions, both private and public. The private sector, alongside government schools, developed to provide education for non-national children, and these schools are generally funded through the fees paid by parents. Private schools opened to meet the cultural, religious, and educational needs of certain national groups or sectors of the UAE society. The private sector is growing at a very fast pace, with more schools opening each year, and the number will overtake that of public schools within the next few years. Many of the private schools are based on foreign national systems, with a very large number of schools offering British, Indian, and American curricula. Although these schools were founded to meet the needs of a particular group, they now tend to have a very multicultural student body, with many nationalities in one school. For example, Jumeirah College, a British curriculum school in Dubai, was founded to offer private education that followed the national curriculum of England and Wales. The school is staffed with mainly British-trained teachers and all teachers and administrators must have experience of teaching in the British education system. Although the admission policy favors the children of British expatriates living in the Gulf, student enrollments include over 40 nationalities, and entrance is gained through a competitive entrance examination, along with registration fees and, at times,

318

DANIEL KIRK AND DIANE NAPIER

interviews. According to the (lead) author’s interactions and conversations with faculty and administrators while teaching at Jumeirah College, this college (as is the case with many of the established British and American curricula schools in the Gulf region), is favored by wealthy national Emirati families who see it to be offering Western-style education that affords more opportunities than the national colleges. Such families are willing to forego fee-free state education and to pay the high fees to gain entry to the foreign system.

HIGHER AND TERTIARY EDUCATION IN THE UAE Unlike Lebanon and Egypt, where higher education has a long and established history, the existence of higher and tertiary education in the UAE is a relatively new concept. Higher education in the UAE dates back to 1977, six years after the formation of the state, when the UAEU was opened as the ﬁrst national university in the country (Touq, 1992). The UAEU opened to students in the academic year 1977/1978 with four colleges and it has continued to grow and play a leading role in the higher education development of the country. Over the past couple of decades the higher education sector has seen rapid growth, both in government and private institutions. With regard to participation by female students, the UAE experience of rapid expansion of women in higher education is part of a worldwide trend, one of the most signiﬁcant developments in recent years in higher education in developing countries (Stromquist, 2003, p. 190). In 1988, the government-owned and -operated HCT opened to offer three-year post-secondary courses, along with a four-year Bachelor of Education degree to aspiring national teachers. HCT is an English medium institution and provides mainly vocational training, largely based on a Canadian model. HCT has campuses in all of the emirates and are segregated by gender, with the result that all pre-service teachers are female. Zayed University (ZU), located on two campuses in Dubai and Abu Dhabi, offers a liberal arts style education to an exclusively female student body and is based on a hybrid of a United Kingdom and Australian curriculum. Faculty members are recruited primarily from institutions in the United Kingdom and Australia, with a small number of Arab nationals who have higher degrees from Europe or Australia. The higher education sector has been allowed to open up to overseas and private organizations, and this has resulted in a recent growth spurt in degree awarding institutions in the UAE. In 1997 the ruler of Sharjah opened the

Issues of Gender, Equality, Education, and National Development in UAE

319

American University of Sharjah, an overtly American-style institution that offers Bachelor’s and Master’s level degrees based on a US curriculum. The faculty includes many nationalities, the majority being American citizens, with one of the main recruitment criteria being that all faculties must have received their terminal degree from an American institution and/or have experience teaching at a North American college or university. This means that among the faculty are Arab nationals, Europeans, Indian, and Asian faculty, alongside North American colleagues. All of the individual emirates are expanding their educational provision and opening up the education sector to both public and private institutions. Large sums of money are being invested and areas of land developed to house the wide range of institutions that are both operating, and planned, in the country. The ruler of Sharjah demonstrated a commitment to education with a project to house several higher and tertiary level institutions in an area of the Emirate known as University City. This area, on the outskirts of the city of Sharjah, is tailor-made to house the six institutions that are set into the manicured, park-like grounds, including the University of Sharjah, a government-funded, Arabic medium and gender-segregated college and the American University of Sharjah, which is coeducational with English as the language of instruction. Dubai, the emirate that borders Sharjah to the south, is also following Sharjah by establishing two educational zones in the city. Knowledge Village will house the growing number of private and foreign institutions setting up in Dubai, and Academic City is the new home to ZU and HCT in Dubai, along with private and government schools and colleges.

ISSUES OF GENDER IN EDUCATION IN THE UAE The notion of teaching among nationals in the UAE is one that is heavily inﬂuenced by gender issues. Teaching is viewed among nationals as a role for women, one that often does not acknowledge any professional or educated status. As Davies (1996) points out, there has been little research on gender issues in teacher education in developing countries, yet gender plays an important role in teacher education. In turn, this has implications for a country’s development. The traditional role of women in Emirati society, guided, in part, by the teachings of Islam and the Quran is often viewed as the source for guidelines as to gender roles within the family and wider society. Within traditional familial structures, women are responsible for the upbringing of children, alongside the management of the home and

320

DANIEL KIRK AND DIANE NAPIER

the passing on of domestic knowledge to female children. This role is continued within the wider societal organization, with women taking on roles in the ‘‘caring’’ sectors of society, such as education and nursing. The education that women are traditionally involved in is separate from the teaching that takes place in the mosques, where a male Imam is responsible for Quranic teaching to the young males of the society. With the roles split along gender lines, it is easy to see how teaching in formal schools is viewed as a role for women. With entry into teacher education programs in the UAE linked to the role, status, prestige, salary, career, and availability of other jobs, many male undergraduates choose courses that are in keeping with the traditional gender divisions of the culture. Alongside the traditional and religious aspects of gender roles in the UAE, teaching as a profession is also being inﬂuenced by questions of power and authority. While the teaching profession in the UAE consists of women in the vast majority, the senior positions in schools and within the Ministry of Education are held predominantly by men (Al-Kaabi, 2005). In part, the educational and professional opportunities available to women have been limited by the traditional and conservative attitudes of the society of the UAE, even though the UAE government’s commitment to liberalization has created many more opportunities for women (Talhami, 2004). Classroom teaching is viewed as an acceptable profession for women by a society which, in the words of Azzam et al. (1985) ‘‘deems only certain activities appropriate for women to pursue outside the traditional roles of marriage and childbearing’’ (Azzam et al., 1985, p. 5). This situation leads to a wider questioning of the way gender is treated within education and the wider society. The social agenda, set in the UAE by a patriarchal system, sets teaching within the framework of social change in the country and the role of schools and teachers is central to the success, or failure, of any such change. Although the government of the UAE is attempting to promote teaching to Emiratis, the profession is still viewed by those setting policy as a feminized one. In fact, female student choice of degree courses are still inﬂuenced to some extent by a rather narrow range of employment options that are considered culturally appropriate for women to pursue (Shaw, 1997). This cultural norm also excludes the vast majority of women from studying abroad opportunities unless they are accompanied by a male family member, in direct contrast to male students who are often encouraged to seek degrees in Western institutions. As noted earlier, issues of gender in education generally, and teaching in particular, are bound up with questions relating to power and authority and, in the case of the UAE. Furthermore, the power and authority lies with

Issues of Gender, Equality, Education, and National Development in UAE

321

central government, the Supreme Council, all of whom are male. Despite the patriarchal nature of the UAE governance, the status and promotions of women’s rights has been a public policy initiative of government over the past 20 years. Linked to education, this has had the effect of a rapid rise in female enrollment in school and university institutions, though women are still usually taught separately from their male peers (Shaw, 1997). Within the state school system of the UAE, female participation has steadily risen since the formalization of education by the federal government. In fact, as reported by UNICEF (2008), primary school enrolment in 2006 stood at 82% for girls compared to 85% for boys, yet by the time students enter secondary school, the situation is reversed with an female enrollment of 66% compared to 62% for male students (UNICEF, 2008). From the very beginning of formal education, women and girls were granted equal access and opportunity to the school system (Findlow, 2007). This right was built into the constitution of the UAE, written in 1971, and assured access to free education for all of the citizenry. This practice ﬁltered up into higher education, with the founding of the UAEU allowing women to gain a university education without the need to go overseas, thereby appeasing cultural norms that were wary of single, unaccompanied women leaving the family unit. Although access to university was seen as a step toward equity within education, women were still guided toward the ﬁelds of teaching and social science, areas that were viewed as appropriate for the future mothers and wives of the country. Coupled with this was the gendersegregated campus of the UAEU, which had two separate campuses, one for female and the other for male students. This segregation, and the placing of certain faculties on a speciﬁc campus (e.g., the College of Education exists only at the women’s campus) highlights the remaining inequality based on gender that exist in the education system. Such segregation, and the ‘‘feminizing’’ of certain academic subject areas, is rooted in the cultural practices and norms of the country and region, rather than the educational system itself. Although feminization of ﬁelds such as teaching is also prevalent in many Western countries, the cultural and religious traditions of the Gulf exacerbate the gender disparity in schools and in university campuses as the federal government of the UAE, a male-dominated and patriarchal group, administers the education system, the various elements are inextricably tied up in cultural practices. The awareness of the need to educate the women of the UAE was further emphasized with the formation of an all-female university, ZU, which opened with campuses in Abu Dhabi and Dubai. This institution, staffed by mainly Western expatriate faculty and administrators, developed a mission

322

DANIEL KIRK AND DIANE NAPIER

focused on cutting-edge academic practice and ﬁelds of study, such as information technology, business, visual arts, and education. The university is viewed as a place where female students can study and learn, free from social and cultural interactions with male students, which creates a space for expression and collaboration. However, like the UAEU, Emirati women at the university are also subject to the reinforcement of cultural norms and practices, with social science and education programs taking precedence over the hard sciences and engineering. This binary of modernization and traditionalism is highlighted through the ZU organized and sponsored conference ‘‘Women as Global Leaders’’ (Zayed University, 2008), held in Dubai in March 2008. This highly publicized and lauded conference (media awareness included international advertising in The Economist and on BBC World), attracted high proﬁle female politicians, business leaders, academics, and scientists from around the world, along with many governmental and non-governmental organizations, to discuss the role of women in the global arena. ZU students attended, participated in panel discussion, assisted with the organization and presented papers, yet the irony of many of the themes was not lost on many. Although such a conference should be applauded and the issues tackled taken seriously on a global level, many felt that until women in the UAE, and wider Gulf region, were given ample opportunities to become leaders in their ﬁelds and within the society, that the discourse of women as ‘‘Global Leaders’’ sounded a little too removed from the reality of the role of women in the region. Policy and program level rhetoric might be awash in accounts of women’s new roles in UAE society and in participation. However, as we suggested earlier, in acknowledging the usefulness of theoretical explanations of policy borrowing processes, there might be disconnect between this ofﬁcial rhetoric and the discourse to be heard on a more personal level. In the following section, we offer some glimpses into the perceptions of women who are participants in the rapidly changing UAE educational landscape but who express feelings of tension between the alleged goals and purposes of their training and their own inner feelings and aspirations. Here, it is interesting to ponder the degree to which such rapidly conceived, largely imported programs in teacher education might (or might not) be fully indigenized and implemented at some point in the future, and also to ponder the question of whether or not such UAE women will ultimately experience gender equity or equality. Finally, we are obliged to return to the question of whose opinion counts: will insider views such as these presented below prevail, or will they be lost in the higher level debate over progress in UAE educational development – or in outsider views of the progress record in the UAE?

Issues of Gender, Equality, Education, and National Development in UAE

323

VOICES FROM THE CLASSROOM The current position of females in the education system in the UAE presents them with opportunities alongside challenges in terms of gaining equity and promoting women as agents of change and beneﬁt to the wider society and cultural norms. Women in education often ﬁnd that they are positioned between two social institutions, on the one hand, the family unit is a privileged and important element of society, and on the other hand, is the place of women within the socially provided educational system. These women require skill and an understanding of their place to navigate through the socio-cultural spaces in which they exist at different times. One area where this cultural navigation is very evident is in the teacher education programs offered within government institutions. Research on the perceptions of student teachers in the UAE revealed aspects of these cultural navigation processes, and the tensions between traditional roles and new roles linked to education and career opportunities. As part of a larger qualitative comparative study of teacher education issues in the United States, the United Kingdom, and the UAE, Kirk (2008) interviewed eight women enrolled as teacher candidates in the UAE during the 2006–2007 academic year, with a view to examining the position of these women within the wider educational framework of the UAE, the Gulf, and global structures. The purpose of the research was to investigate the views of pre-service teachers and the occurrences that affected their perceptions of their teacher education courses and experiences and how their university experience related to their personal and professional socialization into the role of ‘‘teacher.’’ The participants, chosen at random from a cohort of preservice teachers, were undertaking the teaching practice phase of their program. Semi-structured conversational interviews using an interview protocol of guiding questions on these issues, aimed at allowing the women to talk about their experiences while maintaining a focus on the pre-service education program. The voices of these student teachers shed light on the challenges and opportunities that exist within the teacher education programs of the UAE educational system. In addition to interviews, data and insights came from curriculum and policy document analysis, classroom observations, and informal meetings with faculty and administrators. The ﬁeldwork, conducted over two visits, each of two weeks in December 2006 and May 2007. When examined as an example of the female experience in the education system of the UAE, student teachers are an interesting case as they not only reﬂect on their role as participants, but also their part as professional

324

DANIEL KIRK AND DIANE NAPIER

members of the educational community, the future teachers who will become part of the process of education in the country. During interviews with these students, it became clear that they debated and discussed those issues related to gender inequality within their university experiences within small friendship groups, yet such discussions were absent in a more public form of discourse within the institution. This dichotomy is evident in the words of one student: Girls do not get to go over there [the male campus which houses the engineering and business school] to study, we have to stay on our campus and study . . . We are viewed as teachers and teachers are always women in our country . . . Why not woman [sic] be businessman or engineer or politician? . . . I like teachers but it is not want I want to do for job, I want to be a good mother but also, maybe, study to be an architect or designer . . . But this is not easy here, men decide. (Female student, government university, UAE, 2007)

The unfairness is often viewed in tangible and physical terms. The student in the quote above mentioned the prohibitive situation that did not allow the female students to go to the male campus (and vice versa) in terms that identiﬁed an exclusion and difference between the two settings. Alongside the physical separation, which is an element of the cultural space in which the institutions exist, the student concluded that men are the decision makers in the structural and cultural situations that are created. There are cultural practices that also play a role in the experiences of student teachers in the UAE. Many of the women who were interviewed expressed a desire to attend a different university program to the education route that they were completing. One student noted this clearly: I do not like to be teacher [sic] . . . I like business and want to be business person . . . My father tell me to be teacher in UAE [sic] . . . I want to go to university in America and look at how to do business . . . The teacher is not a good job, and I do not like it. My father makes me come to university here at NGU [my use of pseudonym] and makes me be teacher. I like so much many of the girls here who are my friends but I do not want to be teacher . . . I know UAE needs teachers but not me! (Female student, government university, UAE, 2007)

Evidence in the study suggested that, at least in part, some patriarchal dominance in the culture is shifting. This shift was also noted by Findlow (2007) and Kirdar (2007). Women are now attending university in greater numbers, yet the choice of institution and course of study is often not made by the female student. Education is culturally viewed as an appropriate course of study for women in the UAE, as it is thought of by conservative elements of society as a logical step in the preparation of girls to be new mothers and wives. Many of the girls who were not involved in the decision-making

Issues of Gender, Equality, Education, and National Development in UAE

325

regarding their course of study had little interest in entering the workforce as teacher on graduation, despite the high salaries offered to Emirati teachers by the federal government. There was also an undercurrent of dissent in some of the voices from the women regarding their status and role within the society and the education system. The inability to make their own choices regarding university programs, among other things, highlighted the gendered cultural differences, particularly in relation to the idea of ‘‘ownership,’’ where the female is the property of her father ﬁrst, and husband later. Burn (2005) reported these realities, and they can also be discerned in this quote from our more recent empirical research: I would like to do a different course, maybe choose to go to a different place and study . . . My family choose [the university] for me and said I was to be teacher . . . teacher is OK, but not for me, I think I maybe like to be owner of business, of fashion business . . . My father, he knows what is good for me and our family, but sometimes I would like to decide what to do at university. (Female student, government university, UAE, 2007)

As the student in the quote above points out, girls in the UAE are often removed from the decision-making process that inﬂuences many aspects of their daily lives. The absence of the power and right to make personal choices related to their education and career options does little to empower the individual and can, in certain cases, leave the woman with a perception that her rights and agency are curtailed by the patriarchal system of the society. The loss of personal agency and freedom to make life choices leads individuals to feel dehumanized and impotent in their own course of events. Several of the female student teachers that were interviewed felt this way and recognized that the culture in which they lived contained religious and moral elements that formed the basis of the cultural norms (Kirk, 2008).

THE FUTURE OF EDUCATION IN THE UAE: PROSPECTS FOR GENDER EQUALITY? The development of the education sector in the UAE can be characterized as showing continued rapid growth in education provision and choice for students in the country along with a public policy commitment to attempt to Emiratize the teaching and education sectors. Education in the UAE, as part of government policy, is undergoing a careful reexamination and evaluation in order to improve its efﬁciency, effectiveness, and adaptability to rapidly changing global conditions (Touq, 1992), and this can be seen in formal policy

326

DANIEL KIRK AND DIANE NAPIER

such as Vision2020, which aims to have 90% of all government teachers drawn from national citizens by the year 2020. Reform efforts, driven by central government, cover all aspects of the educational system including teacher education, and have sparked a national debate regarding the path education should take in the country, as discussed at the ﬁrst ever School Reform Conference held in Dubai in 2007 (Kirk, 2007). Despite the drive for a more indigenous teaching force, the UAE will continue to depend on expatriate teachers in the coming years, despite its generous spending on education in general and its emphasis on the education of women (Talhami, 2004). The UAE holds a powerful position within the region of the Gulf, and wider Middle East, as it is economically strong and drives much of the growth and development in the region. This is mirrored in the education systems that are developing in the country, and the inﬂux of foreign investment, interest, and institutions concerned with educational provision. Along with the rise in the number of educational institutions and opportunities comes awareness that the future success and development goals of the nation can only be met if proper consideration is given to the educational experiences afforded to national women. Gender equality in the education sector is problematic in the UAE, as there are two competing, and often contradictory, sides of the equation. Firstly, there are the access and enrollment issues, which demonstrate that the UAE education structure allows women equal access to educational opportunities, from elementary school through to university. The creation of an all-female higher education institution, ZU, and the stated HRD aims that promote the growth in female participation in the national workforce, demonstrate that the policy initiatives of the federal government are aimed in the direction of promoting gender equality and parity, at the policy level, and in addressing the ‘‘peculiar set of circumstances’’ that make up the internal needs of the UAE. The promotion of female participation in education, however, does not adequately deal with the cultural factors, the second side of the equation, that are often a bar gender equality in the UAE. The patriarchal system that is prevalent in the UAE, and wider Gulf region, stymies the freedom of educational choice for many young women. The examples of female students’ views presented previously illustrate these tensions. Although access to higher education for females has expanded considerably, opening new opportunities to females for participation and career development, the options available are often restricted by the choices, wants, and beliefs and of the patriarchal familial structure. Although we should applaud the rapid rise in female student enrollment in all levels of education in the UAE, this

Issues of Gender, Equality, Education, and National Development in UAE

327

must be muted by a realization that a signiﬁcant number of girls in education are completing courses and programs that they themselves did not choose. As Steiner-Khamsi (2002) noted, one needs to consider this level of discourse, among the actual female participants in the country, whose views on the prospects for true gender equity and full participation are signiﬁcant. We agree with Kirdar (2006) who raises a thorny question when she asks whether the Gulf States can progress without using one half of their population to ﬁll important niche positions in both the technical and professional domains . . . the root cause impeding female progress appears to be a high religious content within education, and the apparent conﬂict between traditional Islamic values and the needs of modern economic development. (quoted by Grifﬁn, 2006, p. 17)

Kirdar (2006, pp. 207–208) characterized the larger issue with regard to women and education in the UAE and other Gulf states as ‘‘heavy but inappropriate investment in education in the Gulf States, with a lack of emphasis on primary education and with women seen as a resource only in times of emergencies.’’ The situation in the UAE fares well when viewed against enrollment rates of many regional neighbors, yet we cannot ignore the juxtaposition of increased and promoted access to education for girls against the often-limited choices they have regarding their course of study. Furthermore, indigenization of the workforce and of the professional teaching force is another long-term consideration with implications for gender equity in the UAE.

A LOOK AHEAD: CONCLUDING THOUGHTS The UAE educational system and the issues relating to gender, equality, and national development reﬂect many of the wider global and regional forces that have an impact upon the governance and path of education in a single country. Along with many other countries in the region, the UAE is a case of educational hybridity, characterized by the importation of foreign systems and models, coupled with a traditional and local education tradition based on an Islamic model of cultural transmission. This situation raises questions regarding the equity of access to education, the suitability of imported systems, the role of education and national development, and the affect of a hybrid system of education on societal and cultural development. The role of women in the wider development aims of the UAE is inextricably linked to their access and equality in the education system. Although it is often easy to legislate and in theory to provide wider access for

328

DANIEL KIRK AND DIANE NAPIER

women in education and the workforce, it is harder to overcome the cultural norms and practices that limit the roles of women and the choices they are able to make. One of the issues related to understanding change in the UAE is that the country that has been neglected in educational literature, with few empirical studies related to educational development and gender issues in education. As noted by Stromquist (2003, p. 198) there is a dearth of qualitative research focusing on the education of women in developing countries, yet such research is sorely needed to uncover the realities and the ‘‘spaces of rupture of dominant gender norms and representations,’’ and to document the instances in which agency is beginning to take place. There is a need for a long-term study of the UAE education systems to ascertain the nature of the implementation, gender, equity, and development issues, to critically examine the policies and practice, to determine the ingredients of success and/or failure, and to document and report the intended and unintended outcomes. As Phillips (2004) argued, full implementation of borrowed, or partially borrowed, programs of education might be difﬁcult implant successfully into a signiﬁcantly different societal context. It will be interesting to follow the developments in education in the UAE and to note any evidence of whether or not current, largely imported teacher education programs do indeed become indigenized and whether or not they are implemented fully. We will also need to ponder the question of whether such programs successfully prepare the new generation of UAE teachers to meet the needs of their radically changing society and of the schools and students they will serve. Scholars following the dynamics of educational change in the UAE are faced with a challenge in understanding cultural practices and societal norms in the UAE that shape gender roles and education-related inequities, and that are quite likely to continue doing so in the future despite modernization and Westernization inﬂuences.

REFERENCES Abdullah, A. (2007, April 14). UAE’s demographic imbalance. Gulf News, p. 8. Al-Kaabi, A. (2005). Factors inﬂuencing teacher attrition in the United Arab Emirates. Unpublished doctoral dissertation. University of Pittsburgh. Anderson-Levitt, K. M. (Ed.) (2003). Local meanings, global schooling: Anthropology and World Culture Theory. New York: Palgrave, Macmillan. Arnove, R. F., & Torres, C. A. (Eds). (2003). Comparative education: The dialectic of the global and the local. Oxford: Rowman and Littleﬁeld.

Issues of Gender, Equality, Education, and National Development in UAE

329

Azzam, H., Abu Nasr, J., & Lorﬁng, I. (1985). An overview of Arab women in population, employment and economic development. In: J. Abu Nasr, F. Nabil, H. Khoury & T. Azzam (Eds), Women, employment and development in the Arab world. Berlin: Mouton. Baker, D., & Wiseman, A. (Eds). (2005). Global trends in educational policy. Oxford: Elsevier Science. Brook Napier, D. (2005). Implementing educational transformation policies: Investigating issues of ideal versus real in developing countries. In: D. Baker & A. Wiseman (Eds), International perspectives on education and society (Vol. 6, pp. 59–98). Amsterdam: Elsevier Science. Burn, S. M. (2005). Women across cultures. A global perspective. New York: McGraw-Hill. Christina, R., Mehran, G., & Mir, S. (2003). Education in the Middle East. Challenges and opportunities. In: R. Arnove & A. Torres (Eds), Comparative education. The dialectic of the global and local. Maryland, VA: Rowman & Litleﬁeld. Daun, H., Arjmand, R., & Walford, G. (2004). Muslims and education in a global context. In: H. Daun & G. Walford (Eds), Educational strategies among Muslims in the context of globalization (pp. 5–36). Leiden, The Netherlands: Koninklijke Brill. Davies, L. (1996). Teacher education, gender and development. In: C. Brock (Ed.), Global perspectives on teacher education (pp. 47–64). Oxford: Triangle Books. Findlow, S. (2000). The United Arab Emirates: Nationalism and Arab-Islamic identity. Abu Dhabi: Emirates Centre for Strategic Studies and Research. Findlow, S. (2007). Women, higher education and social transformation in the Arab Gulf. In: C. Brock & L. Levers (Eds), Aspects of education in the Middle East and North Africa (pp. 57–76). Oxford: Symposium Books. Gaad, E., Arif, M., & Scott, F. (2006). Systems analysis of the UAE education system. International Journal of Educational Management, 20(4), 291–303. Gal, D. G. (2005). Global perspectives for teacher education. In: D. Baker & A. Wiseman (Eds), International perspectives on education and society (Vol. 6, pp. 259–288). Amsterdam: Elsevier Science. Grifﬁn, R. (Ed.) (2006). Education in the Muslim world: Different perspectives. Oxford: Symposium Books. Jansen, J. D. (2004). Importing outcomes based education into South Africa: Policy borrowing in a post-communist world. In: D. Phillips & K. Ochs (Eds), Educational policy borrowing: Historical perspectives (pp. 199–220). Oxford: Symposium Books. Kelly, G. P., & Altbach, P. G. (1982). Introduction: Approaches and perspectives. In: P. G. Altbach, R. F. Arnove & G. P. Kelly (Eds), Comparative education. New York: Macmillan. Kirdar, S. (2006). The development of women’s education in the Arab world. In: R. Grifﬁn (Ed.), Education in the Muslim world: Different perspectives (pp. 191–210). Oxford: Symposium Books. Kirdar, S. (2007). The impact of educational empowerment on women in the Arab world. In: C. Brock & L. Levers (Eds), Aspects of education in the Middle East and North Africa (pp. 39–56). Oxford: Symposium Books. Kirk, D. (2007, April). Global education for a local population: An international study of pre-service teachers’ perceptions of professional training in the United States, United Kingdom and United Arab Emirates. A paper presented at the conference School Reform: Challenges and Aspirations. United Arab Emirates University, Dubai, UAE.

330

DANIEL KIRK AND DIANE NAPIER

Kirk, D. (2008). Local voices, global issues: A comparative study of the perceptions student teachers hold in the United Arab Emirates, Unites States of America and United Kingdom. Unpublished doctoral dissertation. Athens, GA: University of Georgia. Mazawi, A. (2007). Besieging the king’s tower? En/gender academic opportunities in the Gulf Arab states. In: C. Brock & L. Levers (Eds), Aspects of education in the Middle East and North Africa (pp. 77–98). Oxford: Symposium Books. McGinn, N. F., & Cummings, W. K. (1997). Introduction. In: W. K. Cummings & N. F. McGinn (Eds), International handbook of education and development: Preparing schools, students, and nations for the twenty-ﬁrst century. New York: Elsevier/Pergamon. McGovern, S. (1999). Education, modern development and indigenous knowledge: An analysis of academic knowledge production. New York: Garland. Mina, F. M. (2006). Teacher education in Egypt: Visions and reality. In: R. Grifﬁn (Ed.), Education in the Muslim world: Different perspectives (pp. 323–338). Oxford: Symposium Books. Peck, M. C. (1986). The United Arab Emirates: A venture in unity. Colorado: Westview Press. Phillips, D. (2004). Toward a theory of policy attraction in education. In: G. Steiner-Khamsi (Ed.), The global politics of educational borrowing and lending (pp. 54–69). New York: Teachers College Press. Phillips, D., & Ochs, K. (Eds). (2004). Educational policy borrowing. Historical perspectives. Oxford: Symposium Books. Ramirez, F., & Boli, J. (1987). The political construction of mass schooling: European origins and worldwide institutionalization. Sociology of Education, 60(1), 2–17. Reagan, T. (2005). Non-western educational traditions. New Jersey: Lawrence Erlbaum. Shaw, K. (1997). Higher education in the Gulf. Problems and prospects. Exeter: University of Exeter Press. Shaw, K. (2006). Muslim education in the Gulf states and Saudi Arabia: Selected issues. In: R. Grifﬁn (Ed.), Education in the Muslim world: Different perspectives (pp. 41–54). Oxford: Symposium Books. Shouly, E. (1995). 16,000 national apply for jobs. Gulf News, 19 June, p. 18. Singh, S. P. (2006). The interface between Islam and the West in politics: Intellectual awakening and education in the middle ages. In: R. Grifﬁn (Ed.), Education in the Muslim world: Different perspectives (pp. 25–40). Oxford: Symposium Books. Steiner-Khamsi, G. (2002). Re-framing educational borrowing as a policy strategy. In: M. Caruso & H. E. Tenorth (Eds), Internationalisierung/Internationalisation. Semantik und Bildungssystem in Vergleichender Perspektive/Comparing educational systems and semantics (pp. 91–107). Frankfurt am Main: Lang. Stromquist, N. P. (2003). Women’s education in the twenty ﬁrst-century: Balance and prospects. In: R. F. Arnove & C. A. Torres (Eds), Comparative education: The dialectic of the global and the local (pp. 176–203). Oxford: Rowman and Littleﬁeld. Taifour, M. (1994). UAE gearing up for the 21st century. Gulf News, 3 January, p. 18. Talhami, G. (2004). Women, education and development in the Arab Gulf countries. Abu Dhabi: Emirates Center for Strategic Studies and Research. Tischer, R. P., & Wideen, M. F. (Eds). (1990). Research in teacher education: International perspectives. London: Falmer Press. Touq, M. (1992). United Arab Emirates. In: W. Wickremasinghe (Ed.), Handbook of world education. Houston, TX: American Collegiate Service.

Issues of Gender, Equality, Education, and National Development in UAE

331

UNICEF. (2008). At a glance: United Arab Emirates. UNICEF website, available at http:// unicef.org/infobycountry/uae_statistics.html#46. Retrieved on 17 April, 2008. Weber, E. (2007). Globalization, ‘‘Glocal’’ development, and teachers’ work: A research agenda. Review of Educational Research, 77(3), 279–309. Zayed University. (2008). Women as global leaders conference. Zayed University website, available at www.zu.ac.ae. Retrieved on 2 April, 2008.

GENDER SEGREGATION IN STUDENT CAREER ASPIRATIONS IN NORWEGIAN SECONDARY SCHOOLS Lihong Huang INTRODUCTION Along with the other Nordic welfare states, Norway has achieved relative gender parity as measured by the Gender Gap Index of the United Nations (Hausmann, Tyson, & Zahidi, 2006) and is therefore looked upon by many as a model in minimising gender gaps in the society. With a 47 per cent share of the active labour force in the country, Norwegian women have had a high level of labour market participation since the late 1980s. In addition, Norway ranks among the top countries in the world in terms of offering women and men equal access to education at all levels, equal access to leadership positions in the workplace and in politics and generous parental leave beneﬁts. Although gender parity in education at all levels as well as in labour market participation is a reality in Norway, there are signiﬁcant gender differences vis-a`-vis in career aspiration among students and the Norwegian labour market is characterized by gender segregation (Foss, 2005) which results in a gender gap in pay where women earn less than men.

Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 333–356 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010013

333

334

LIHONG HUANG

It seems that, in the Norwegian case, gender parity in education and in labour market participation does not necessarily result in equality between men and women in the society. As education is the most important factor inﬂuencing occupational attainment both for men and women, it is essential to understand the effect of schooling process to an individual’s career development in order to understand the root of gender inequality in the labour market and in a society. In this chapter, the Norwegian upper secondary school system and different dimensions of student career aspirations are investigated. Using data from a national questionnaire survey conducted in secondary schools in Norway in 2002, this chapter offers some insights into how Norwegian young people in the upper secondary schools choose their future career paths by (1) describing the pattern of values in student career aspirations between female and male students; (2) exploring the relationships between student career aspirations, their home socio-economic background and their academic achievement, using linear structure modelling techniques; and (3) explaining the statistical results and concluding at the end.

BACKGROUND OF THE STUDY Although children typically start school at the age of six in Norway, they do not need to make their vocational choice until they have completed 10 years of compulsory education at about the age of 16. The vast majority of Norwegian children (98 per cent) progress to upper secondary school after completing compulsory school. Upon entry into upper secondary schools, students make a choice between an academic education path, which leads to university entrance, and a vocational education path leading to certain vocational competences and labour market entry or further vocational training at the tertiary level. Although lifetime incomes do not vary signiﬁcantly between young people in Norway who pursue different education paths, their choices upon entry to upper secondary school lead them to different occupations, where certain class cultures have developed to distinguish them from each other in the society (Skogen, 1996, 1998; Huang, 2007). In Norway’s rather ﬂatly structured society, there are some salient features of cultural fractions within the middle class and among different occupational groups. For example, research shows that young people from families in the ‘humanistic/social intermediate strata’ of the middle class participate more in environmental youth organisations and spend more

Gender Segregation in Student Career Aspirations

335

out-of-school time on artistic/cultural activities, than those from families in the ‘technical/economic intermediate strata’ of the same class (Skogen, 1996, 1998). Moreover, these so-called class cultures are able to reproduce generations directly and via the Norwegian education system. Recent sociological research in Norway (Skogen, 1998) has found a distinct relationship between parents’ class position and student recruitment into different educational programmes at upper secondary schools, which channels young people with different class background into separate sectors of the labour market. The same research found that boys are more likely to select educational paths that follow in their fathers’ occupational footsteps than girls do. Moreover, children from families with working class parents are likely to enter the labour market without completing upper secondary education. It is evident in the Norwegian research that social origin has an impact on individuals’ educational choice, occupational success (Hansen, 1995; Skogen, 1998) and eventual social mobility (Ringdal, 2004). More women than men now study at the tertiary levels of education in Norway. In fact, females account for 65 per cent of those pursuing a Bachelor degree, 50 per cent of those pursuing a Master degree but less than half of those engaged in doctoral studies (Mastekaasa, 2005). Similar patterns are found in many other post-industrial countries (Charles & Bradley, 2002). In the Norwegian labour market, gender differences are still visible as more women work in the public sector and are often employed in typical ‘female’ occupations as teachers, nurses, cleaners and administrative assistants. Men, however, are more likely to work in the private sector and work in typical male occupations as, for example, trades people, manual workers, drivers and engineers. This gender-segmented labour markets is one reason Norway has a persistent gender gap in pay as women’s monthly earnings amount to 84 per cent of men’s earnings (Statistics Norway, 2006). This is the result of gendered labour divisions in occupations, ﬁrms and positions as well as the tradition for male-dominated jobs in Norway for payment of higher salaries (Barth, Røed, & Torp, 2002). It is said that in Norway, as well as in other industrial countries, gender ‘vertical segregation’ or vertical inequality (i.e., access to certain levels of education, certain occupations and leadership positions in companies) has decreased substantially while gender ‘horizontal segregation’ (i.e., gendertypical study programmes in higher education and gender-typical job sectors in the labour market) has become more prominent (Blackstone, Browne, Brooks, & Jarman, 2002; Bradley, 2006; Charles & Grusky, 2004; Ellingsæter, 1999). Recent research on gender in the labour market has focused on explaining and understanding why horizontal segregation

336

LIHONG HUANG

persists and has even increased in education and in the labour market despite a substantial decrease in vertical segregation, which has nearly disappeared in many industrial countries. As educational attainment is found to be the most consequential factor affecting the occupational attainment of both women and men, surpassing even the inﬂuence of home background (Shavit & Blossfeld, 1993; Joy, 2006), factors explaining gender segregation in education are found to echo those seeking to explain labour market gender segregation (Blackstone et al., 2002; Charles, 1992; Charles & Grusky, 2004; Hanson, Schaub, & Baker, 1996). Explanatory factors at the contextual level include organisational trends and processes (Kaufman, 2002; Robinson, Taylor, Tomaskovic-Devey, Zimmer, & Irvin, 2005; Nagy, Trautwein, Baumert, Ko¨ller, & Garrett, 2006), social and political structures of opportunity (Charles & Bradley, 2002; Bradley, 2006; Baker & Jones, 1993) and norms, ideologies and culture (Charles & Bradley, 2002; Buller & Hoggart, 2004). Explanatory factors at the individual level include family origin (Opheim, 2007), parental role models (Hansen, 1993; Dryler, 1998), parents’ expectations and encouragement (Jocobs, Chhin, & Bleeker, 2006; Baker & Jones, 1993), individual cognitive and affective complexity (Correll, 2001; Jonsson, 1999), individual academic achievement (Ayalon, 2003; Shapka, Domene, & Keating, 2006) and career aspiration during earlier school years (Gottfredson, 1981, 2002; Rainey & Borders, 1997; Daehlen, 2005; Meece, 2006; Miller & Hayward, 2006). Research (Jacobs, 1995; Levine & Zimmerman, 1995) suggests that the level of segregation by gender vis-a`-vis individual career aspirations reﬂect the level of gender segregation in the labour market. Therefore, in order to better understand gender segregation in the labour market it is necessary to further explore student career aspiration at secondary school. These issues are addressed, theoretically and empirically, in the following sections of the chapter.

THEORIES AND PREVIOUS RESEARCH ON CAREER ASPIRATION Research has found that occupational aspiration inﬂuences individuals’ pursuit of educational and occupational opportunities (Lent, Brown, & Hackett, 1994; Schoon & Parsons, 2002; Webb, Lubinski, & Benbow, 2002). In social psychology, career aspiration is viewed as a reﬂection of selfefﬁcacy and important mediators of motivation and career development.

Gender Segregation in Student Career Aspirations

337

The pioneer of vocational psychology, Holland (1965, p. 2), states: ‘The choice of an occupation is an expressive act which reﬂects the person’s motivation, knowledge, personality and ability’. Developmental theory in vocational psychology (Super, 1990; Super, Savickas, & Super, 1996) suggests that career aspiration can change over time with the development and implementation of an occupational self-concept. Career choice, in the context of development, is primarily an expression of the social self, followed by the psychological self. When choosing a career path, an individual tends to base their decision on his or her social self (i.e., social criteria such as gender stereotypes and occupational status) rather than the psychological self (i.e., personal characteristics such as interests and personality type) (Gottfredson, 1981, 2002). Career aspiration tends to become increasingly stable as adolescents mature to young adulthood. In fact, adolescents with aspiration–expectation discrepancies tend to consistently change their aspirations in line with their expectations following a decision-making process of circumscription and compromise (Gottfredson, 1981, 2002; Armstrong & Crombie, 2000). Circumscription is the process of eliminating unacceptable occupations from the range of career possibilities (Gottfredson, 1981, 2002). When a person ﬁnally makes an occupational decision, he or she does so after deeming the chosen occupation the most acceptable among all options. Compromise is a choosing strategy used in the circumscription process. When considering a preferred occupational choice, individuals may encounter barriers that would inhibit them from successfully achieving that goal. When individuals abandon their initial preferences for less desirable yet more achievable alternatives, the result is compromise. During adolescence, as the individual’s understanding of different occupations becomes more realistic and complex, compromise may become necessary if the individual perceives their original aspirations as unrealistic or unattainable (Gati, 1993; Gottfredson, 2002). According to the principle of compromise in career aspiration (Gottfredson, 2002), if the individual is under a low level of pressure to make a compromise, then personal interests will be the most protected aspect of a career choice, followed by prestige and gender-typical values. If the individual is engaged in a moderate level of compromise, prestige value in a career choice become the most protected, followed by personal interests and gender-type. If the individual is faced with major compromise, then gender-type will be the most important aspect to preserve, followed by prestige and then interests. However, career aspiration tends to become increasingly stable as adolescents mature to young adulthood.

338

LIHONG HUANG

Many factors are found to inﬂuence the formation and development of career aspiration, which is moderately correlated with personality, interests and many other psychological and sociological variables. Studies have found that the development of the career aspiration process is explained by a combination of socio-economic background variables, personal psychological factors and sociological or environmental inﬂuences. More speciﬁcally, the vast body of research on this subject since the 1960s has found that socioeconomic background, child–parent relationship, parenting style, anxiety, pressure or encouragement and family size are but a few of the determinants of career aspiration (Rehberg & Westby, 1967; Vignoli, Croity-Belz, Chapeland, de Fillipis, & Garcia, 2005; Jocobs et al., 2006). Other factors such as gender (Davey & Stoppard, 1993; Frome, Alfeld, Eccles, & Barber, 2006; Schoon, Martin, & Ross, 2007), ethnicity (Carter, 1999), types of institution (Brint & Karabel, 1989), college environment (Smart & Thompson, 2001; Sax & Bryant, 2006), ﬁeld of study (Nauta, Epperson, & Kahn, 1998) and cost and ﬁnancing methods of study (Huang, 2005) are also found to have an impact on student career aspiration. Moreover, career aspiration is shown to reﬂect the effects of bias and discrimination, social attitudes, cultural expectations and stereotypes based on gender, race and/or socio-economic status (Hotchkiss & Borow, 1996; Perna, 2000). Fig. 1 illustrates a conceptual framework, built upon theories and previous research, which also serves as an analytical framework guiding the premises investigated in this chapter. The framework provides a series of analytical hypotheses tested by Norwegian student data. As illustrated in Fig. 1, four blocks of factors are hypothesised to impact student career aspiration at upper secondary school (i.e., home background, personal characteristics, community environment and school effect). The links of

Home background

Personal characteristic

School effect

Career aspiration

Community environment

Fig. 1.

The Conceptual and Analytical Framework.

Gender Segregation in Student Career Aspirations

339

home background, personal characteristics, community environment, with school effect are well established in previous educational research (Buru-Bellat, 2004; Cabrera & La Nasa, 2001; Choy, Horn, Nun˜ez, & Chen, 2000; Erikson & Jonsson, 1998; Hu, 2003; Huang, 2005, 2007; Huse´n, 1989; OECD, 2001), and the hypothesised effects of these background and school factors on student career aspiration are based on theories and previous research discussed earlier in this chapter. It is hypothesised that background variables (i.e., home background, personal characteristics and community environment) not only have a direct impact on student career aspiration but also an indirect impact on career aspiration via their inﬂuence on student schooling experiences.

DATA AND METHODS The data used in this study are from a national survey on secondary school students (including lower and upper secondary schools) in Norway in 2002, called Young in Norway 2002 (Ung i Norge, 2002). This was a questionnaire survey administrated in schools nationwide during class time in February 2002. Exactly 12,000 students from 73 schools (47 lower secondary schools and 26 upper secondary schools) throughout Norway were randomly selected and given two hours in school to answer the questionnaire. The response rate was 92.3 per cent and the complete dataset entails 11,406 responses from students ranging in ages from 13 to 19. The 5,564 upper secondary school students aged from 16 to 19, 52 per cent of whom are female, are the focus of this chapter. The data includes information on student home background, student relation to parents and friends, schooling satisfaction and adjustment, educational planning and occupational aspiration, problematic behaviour, body and self-image, organisational participation, sports and leisure activities, use of mobile telephone, use of computer and Internet, etc. Fig. 2 is the operational model corresponding to the conceptual framework of this study shown in Fig. 1. It presents hypothesised links between measures of the 10 latent variables tested by the data. Home background includes two latent variables: home social status and home economic status. Home social status is built around parental (both father’s and mother’s) educational attainment. Home economic status is measured by two items asking students’ perceptions of their home economic situation over the past years and their home economic situation compared with other families in their neighbourhoods. Personal characteristics are student age and ethnicity

340

X1 X2 X3

LIHONG HUANG

Study subject Home social status

Y6

Y1 Prestige career aspiration

Home economic status

Intrinsic career aspiration

X4 X5

Personal characteristics

X6

Home community

X7

Ethnicity

KEYS: X1 Father’s educational attainment X2 Mother’s educational attainment X3 Perception of home economic situation through past years X4 Perception of home economic situation comparing with neighbourhood families X5 Student age X6 Home location in rural/urban areas X7 Ethnicity Y1 ‘1’ as academic path, ‘0’ as vocational path Y2 Norwegian test score

Y7 Y8 Y9 Y10 Y11

Y2 Academic achievement

Y3 Y4

Perceived social barrier

Y5

Y3 Y4 Y5 Y6 Y7 Y8

English test score Mathematics test score Sum score of ten items shown in Table 3 Jobs that give prestige and high status Jobs that give good salary Jobs that give good possibilities to become leaders Y9 Jobs that allows me using my own special talent Y10 Jobs that allow me using my fantasy and creativity Y11 Jobs that are creative and idea-rich

Fig. 2. Path Diagram for a Hypothesised Model of Linking Student’s Home SocioEconomic Background, Student Personal Characteristics, Community Environment and Ethnicity with Student Study Subject, Academic Achievement and Perceived Social Barrier, and Student Prestige and Intrinsic Career Aspiration.

(‘1’ as both parents are Norwegian, ‘2’ as one of the parents is an immigrant, ‘3’ as both parents are immigrants). Community environment is measured by the location of the home (from ‘1’ as remote village, to ‘6’ as large urban area). Three latent variables represent school effect in the hypothesised model since schooling process not only includes student learning choices and achievement but also form student attitudes. Study subject is a dummy variable with ‘1’ as an academic path leading to university education and ‘0’

Gender Segregation in Student Career Aspirations

341

as a vocational path leading to occupational certiﬁcates and labour force entry. Student achievement is a construct built around the test scores of three subjects (i.e., Norwegian, Mathematics and English, respectively ranged from lowest score of ‘0’ to highest score of ‘6’). Student perception of social barriers to career development is the sum score of 10 response items (as discussed later in the chapter and shown in Tables 3 and 4) asking students’ perceptions of the importance of parental educational attainment, family economic situation, growing up in a rural or urban area, being Norwegian or an immigrant and the impact of being female or male on their career development. Two value orientations of student career aspiration are tested in the model (i.e., prestige career aspiration and intrinsic career aspiration), as they present low and moderate levels, respectively, of compromise according to Gottfredson’s (2002) compromise principle. They are measured by students’ responses on a Likert scale to the question asking about the most important things they consider to have in their future jobs (as shown in Table 2). The approach used to analyse the data in this chapter follows two steps. The ﬁrst step presents descriptions of background characteristics and career aspirations of the students, comparing the female and male student groups. At the second step, the relationships between student background, educational path choice, academic achievement and future career aspirations are explored, using linear structural equation techniques (LISREL) to test the hypothetical model presented in Fig. 2 with female and male student data. The model testing at the second step is able to illustrate not only the contributions of various factors in predicting the outcome variable (i.e., student career aspiration) but also unique information about the direct and indirect paths of reliable inﬂuence (Mueller, 1996). The reporting procedure of model test results follows criteria established in previous social science research. According to a classiﬁcation of standardised regression weights (Desjardins, 2003) in social science research using population sample survey data, a regression weight over 0.30 is considered a very strong effect, from 0.20 to 0.30 is a strong effect, from 0.10 to 0.20 is a moderate effect and below 0.10 is a weak effect. The decision to accept or reject a hypothesised structural model is taken with reference to the ﬁt statistics. Chi-square (w2) is most frequently cited as a measure of the overall goodness of ﬁt of the model to the data (Jo¨reskog & So¨rbom, 1993). The Root Mean Square Residual (RMR) represents the average deviation of the predicted from the actual correlation matrix. The Good-of-ﬁt Index (GFI) indicates the proportion of the joint amount of data variance and covariance that can be explained by the tested model. The common rule for an acceptable ﬁt of a model is an

342

LIHONG HUANG

RMR below 0.05 points, with AGFI (Adjusted Good-of-ﬁt Index) and GFI exceeding 0.90 (Hoyle & Panter, 1995; Tuijnman & Keeves, 1997).

RESULTS Approximately 50 per cent of Norwegian students take academic courses (or the academic path) at upper secondary school leading to university entrance in the future while the other half take a vocational path leading to occupations in sectors such as construction, heavy industry including oil, gas and chemicals, information technology, forestry, retail and social services, health care, music and others. However, there is gender disparity between the two educational paths as there are more female students in academic courses and more male students in vocational courses. Overall, 60 per cent of Norwegian upper secondary students plan to pursue tertiary education while 20 per cent of the students would exit school immediately if there was a job offer (i.e., they responded ‘yes’ to the question ‘if you get job now, would you rather work than continue schooling?’). Although there are some minor differences in home background factors between female and male students where male students appear to have slightly better conditions, gender difference is more evident in student career choice as shown in Table 1. Almost half of the male students do not plan to continue to tertiary education while 65 per cent of the female students plan to do so. Among the 20 per cent ‘dropout-risk’ students, there are more males than females. All the home background differences between female and male students presented in Table 1, though very modest, are statistically signiﬁcant at the 0.05 level while gender differences in career choice such as study path, higher education plan and possible dropout are signiﬁcant at the 0.01 level. The academic achievement of female students on three subjects (Norwegian, Mathematics and English) is slightly higher than male students and the mean difference between genders is signiﬁcant at the 0.05 level. Table 2 presents the means and standard deviations of students’ responses to the question asking about the most important things they consider to have in their future jobs. The responses are on a 1–5 point Likert scale with ‘1’ denoting ‘not important at all’, ‘5’ denoting ‘very important’ and ‘3’ as neutral. The pattern of student responses can be grouped into roughly four categories of value orientations. The data in the table reveal that jobs with intrinsic values have the highest mean scores in both groups, followed by the social meaning and safety values of the jobs. Gender differences are evident and statistically signiﬁcant at the 0.01 level in utility values where female

Comparing Characteristics of Female and Male Students in Upper Secondary Schools (Per Cent).

Female students: Number of cases 2,869 Male students: Number of cases 2,668

Parents with Tertiary and Above Education*

Both Parents Working*

Both Parents are Immigrants*

Parents Own the Living Place*

Study in Vocational Courses or Path**

Plan to go to Higher Education**

Possible Mean Sum Dropout if Score of Tests Get a Job** in Three Subjects*

46.3

70.2

5.9

93.4

45.9

65.4

15.0

11.3

47.5

73.3

5.4

94.9

54.9

51.8

25.6

10.8

Gender Segregation in Student Career Aspirations

Table 1.

Notes: *indicates mean difference between genders signiﬁcant at 0.05 level; **indicates mean difference between genders signiﬁcant at 0.01.

343

344

Table 2.

What is the Most Important Consideration in Your Future Jobs?

Category/Value Orientation

Utility/prestige money, status

Intrinsic/attainment

Social/humanity

Safe/security

Response Items

that that that that that that that that that that that that that

give prestige and high status** give good salary** give good possibilities to become leaders** allows me using my own special talent** allow me using my fantasy and creativity** are creative and idea-rich** are socially meaningful** I can do something for the others** I can work with people** do not have much difﬁcult things to learn I can have a lot free time** are not too stressful are least possible to become unemployed

Note: **indicates mean difference between genders signiﬁcant at 0.01 level.

Male Students

Mean

Standard deviation

Mean

Standard deviation

2.8 3.7 2.7 4.2 3.6 3.8 3.3 3.9 3.9 2.6 3.0 3.4 4.2

1.1 0.9 1.2 0.8 1.1 1.0 1.0 0.9 1.1 1.0 0.9 1.0 0.9

3.2 4.0 3.1 4.0 3.4 3.6 3.1 3.3 3.3 2.6 3.3 3.4 4.1

1.1 0.9 1.2 0.9 1.2 1.1 1.1 1.1 1.2 1.1 1.0 1.1 0.9

LIHONG HUANG

Jobs Jobs Jobs Jobs Jobs Jobs Jobs Jobs Jobs Jobs Jobs Jobs Jobs

Female Students

345

Gender Segregation in Student Career Aspirations

students score relatively lower than male students, in intrinsic and social values where female students have higher means than male students, and in humanity values where female students again have higher means than male students. However, gender differences in security values of future jobs are not signiﬁcant in this case. Following the survey question on future job considerations, there are further questions asking students about the importance of parental educational attainment, family economic situation, growing up in a rural or urban area, being Norwegian or an immigrant and being female or male to their future job or to any person’s future job. The responses are on a 0–4 point Likert scale with ‘0’ denoting ‘no importance at all’, ‘1’ denoting ‘almost no importance’, ‘2’ denoting ‘little importance’, ‘3’ denoting ‘quite important’ and ‘4’ denoting ‘very important’. Table 3 presents means, medians and modes of students’ responses to the questions. The mean scores of the responses to all questions in Table 3 are all close to the ‘little importance’ side of the Likert scale. Parental education level is considered the least important in determining student career aspiration while ethnicity is the most signiﬁcant among those background factors. Interestingly, male students scored 0.1 points higher than female students on many items, especially those related to the importance of immigrant status. Nevertheless, the results show that students in Norway generally perceive that their future careers are inﬂuenced, to a modest degree, by their socioeconomic, ethnic background and gender. Further exploration of these Table 3. How Important are the Following Items for Your Future Job or for Any Person’s Future Job? Response Items

Whether Whether Whether Whether Whether Whether Whether Whether Whether Whether

your parents have long or short education** your parents have much or little money** your parents are immigrants or born in Norway** you are a boy or a girl you grow up in big city or a small village** that person’s parents have long or short education** that person’s parents have much or little money that person’s parents are immigrants or born in Norway* that person is a boy or a girl** that person grows up in big city or a small village

Female Students

Male Students

Mn

Md

Mo

Mn

Md

Mo

0.9 1.6 1.8 1.6 1.6 1.2 1.8 2.0 1.7 1.7

0 2 2 2 2 1 2 2 2 2

0 2 2 2 2 0 2 2 2 2

1.1 1.7 2.1 1.6 1.7 1.3 1.8 2.1 1.6 1.7

1 2 2 2 2 1 2 2 2 2

0 2 2 2 2 0 2 2 2 2

Notes: Mn, Mean; Md, Median; Mo, Mode; *indicates mean difference between genders signiﬁcant at 0.05 level; **indicates mean difference between genders signiﬁcant at 0.01 level.

346

LIHONG HUANG

factors is facilitated by the creation of a new variable, using the sum of the scores of the response items in Table 3. The new variable technically assigns each student a real score of perceived social barrier based on the aforementioned social factors and on how much each of those factors is perceived to be important to any person’s career development. Table 4 presents the descriptive statistics of this new variable, which can be called ‘Perceived social barrier’ as it captures student awareness of social barriers in their career path. It has to be noted that less than 4 per cent of the total student group score is ‘0’. The additional information presented in Table 4 further conﬁrms the conclusion based on the data presented in Table 3 – that there is modest but signiﬁcant awareness of social barriers deriving from individuals’ background factors. Male students seem to have, to a certain extent, a higher level of awareness of these factors/barriers than female students and the gender difference, in this case, is statistically signiﬁcant at the 0.01 level. At this point, statistically signiﬁcant differences are observed between female and male students in all aspects investigated (i.e., home background, academic achievement, attitudes and career aspiration). Table 5 presents the statistical results of a linear structural model, as hypothesised in Fig. 2, linking student home social status (constructed by father’s and mother’s educational attainments), home economic status (i.e., student perception of their home economic situation), student age, ethnic background, home community, study subject, academic achievement and perceived social barriers to their future career. The dependent variables

Table 4. Descriptive Statistics of Student Perception of Social Barriers to Their Career Development by Total, Female and Male Student Groups. Statistics Number of valid cases Number of missing cases Mean Median Mode Minimum Maximum 25th percentile 50th percentile 75th percentile

Total

Female Student Group

Male Student Group

5,017 547 16.2 16 20 0 40 10 16 22

2,624 245 15.7 16 14 0 40 10 16 21

2,374 294 16.6 17 20 0 40 10 17 22

347

Gender Segregation in Student Career Aspirations

Table 5. Standardised Maximum Likelihood Regression Weights of Direct Effects of Five Independent Variables on Three Intermediating Variables and Two Dependent Variables, by Female and Male Student Groups. Intermediating Variables Study subject Boy Independent variables Home social status Home economic status Student age Ethnicity Home community

0.43 0.07 0.09 – –

Girl

0.45 0.11 0.09 0.15 0.19

Dependent Variables

Academic achievement

Perceived social barrier

Boy

Boy

Girl

0.19 – – 0.05 0.07

0.27 – 0.10 – 0.08

0.38 – – 0.07 0.09

Girl

0.45 – – – 0.13

Intermediating variables Study subject Academic achievement Perceived social barrier R2

0.22

0.24

0.15

0.20

0.05

0.09

Prestige aspiration Boy

Girl

Intrinsic aspiration Boy

Girl

0.16 0.18 0.11 0.09 0.08

– 0.14 – 0.29 –

0.09 – 0.07 0.12 0.05

0.15 – 0.08 0.13 0.08

0.12 0.07 0.23

0.19 0.08 0.19

– 0.06 0.11

0.32 0.09 0.06

0.13

0.15

0.07

0.13

Note: – indicates an effect not signiﬁcant at the 0.05 level. Fit statistics for the models: Boy: w2 /df: 9.04; RMR: 0.04; GFI: 0.96; AGFI: 0.94. Girl: w2 /df: 8.93; RMR: 0.04; GFI: 0.96; AGFI: 0.94.

in the model are student career aspirations of utility (prestige) values and intrinsic values. The results in Table 5 indicate that home social status has a very strong positive effect on both female and male choice of academic path in upper secondary education and on their academic achievement. Home social status also has a rather moderate positive effect on student perception of social barriers to their future career. The opposite is also true and high levels of parental education therefore have strong negative effects on student choice of vocational paths at upper secondary school. In fact, student likelihood of choosing a vocational path at upper secondary school decreases as parental education level increases. The positive association between home social status and student perception of social barriers also points to the fact that the higher the level of parental educational attainment, the more aware students are of social barriers to career development, especially among female students. However, home economic status has very weak positive effects on student choice of academic paths at upper secondary school and no statistically signiﬁcant effect on either achievement

348

LIHONG HUANG

or student perception of social barriers. The weak impact of home economic status could be due to the poor measures of this variable as student perceptions of their home economic situation might not be the optimal measure of home economic status. Home background factors work rather differently for female and male career aspiration as home social status has a direct negative effect on boys’ prestige aspirations and no effect for that of girls. At the same time, this factor is slightly more positive for girls’ intrinsic aspiration than that of boys. Home economic status is weakly positive for prestige aspiration among both genders and has no statistically signiﬁcant effect on student intrinsic aspiration. Student age has a very weak positive effect on both student choice of study subject and intrinsic aspiration of both female and male students while it has a weak positive effect on girls’ perceptions of social barriers and a weak negative effect on prestige aspiration among boys. Immigrant background appears to exert a more positive impact on girls as it has a weak positive effect on girls’, but not boys’, choice of study subject and a moderate positive effect on girls’ prestige aspirations compared with the weak effect for that of boys. Moreover, immigrant background weakly promotes male, but not female academic achievement and perceived social barriers. Home community is negatively associated with choice of academic study paths and student achievement, especially for girls. This implies that students, and especially females from rural areas, are slightly more likely to choose vocational paths in upper secondary school and achieve slightly better results than students from urban areas. The choice of academic path is moderately positively associated with student prestige aspiration among both gender groups while it is strongly negative for girls’ intrinsic aspiration and has no effect on that of boys. This implies that girls’ choice of academic study path is against their intrinsic values of future careers while girls who choose vocational paths have in fact more closely followed their intrinsic career aspirations. Academic achievement has rather equal weak effects on both female and male student career aspiration. Perceived social barriers are moderately positively associated with prestige aspiration for both boys and girls and have weak positive effects on intrinsic aspiration for both genders. The results in Table 5 show that student career aspiration is indeed inﬂuenced by home socio-economic and ethnic background, by where they live. Moreover, the effects of these factors are, to some extent, differentiated by gender. To achieve a better understanding of the subtle mechanisms of these inﬂuences, Table 6 presents the same model testing results as Table 5 but with additional information designed to measure the total and indirect

Standardised Maximum Likelihood Regression Weights for the Total, Direct and Indirect Effects of Eight Predictors on Student Career Aspiration, by Female and Male Student Groups. Dependent Variables Prestige aspiration Boy Total

Direct

Independent variables Home social Home economic status Student age Ethnicity Community

0.10 0.19 0.10 0.10 0.10

0.16 0.18 0.11 0.09 0.08

Intermediating variables Study subject Achievement Social barrier

0.12 0.07 0.23

0.12 0.07 0.23

R2

0.13

Intrinsic aspiration Girl

Indirect

Total

Direct

0.06 0.01 0.01 0.01 0.02

0.03 0.12 – 0.28 0.05

– 0.14 – 0.29 –

# # #

0.19 0.08 0.19

0.19 0.08 0.19

Boy Indirect

Girl

Total

Direct

Indirect

0.03 0.01 – 0.02 0.05

0.13 – 0.07 0.13 0.05

0.09 – 0.07 0.12 0.05

# # #

– 0.06 0.11

– 0.06 0.11

0.15

0.07

Total

Direct

Indirect

0.04 – – 0.01 –

0.06 0.03 0.06 0.08 0.14

0.15 – 0.08 0.13 0.08

0.09 0.03 0.02 0.03 0.05

# # #

0.32 0.09 0.06

0.32 0.09 0.06

# # #

Gender Segregation in Student Career Aspirations

Table 6.

0.13

Note: –, an effect not signiﬁcant at the 0.05 level; #, a parameter not estimated.

349

350

LIHONG HUANG

effects of the background variables and intermediating variables on career aspiration. It shows that background factors not only have direct effects on student career aspiration but also exert indirect effects via their inﬂuence on student choice of study path, academic achievement and student perception of social barriers, which in turn have a direct impact on student career aspiration. Gender differences are evident in every link between the predicting factors and student career aspiration but are most striking vis-a`-vis ethnicity and study subject where female students appear to be more affected. The statistical results suggest that the subgroup data ﬁt the conceptual model proposed in Fig. 1 and Fig. 2 well as all ﬁt statistics of the models met the conventional criteria (see notes in Table 5). Overall, the variances explained by the models are substantial. In fact, background variables explain 23 per cent of the variance of student choice of study subject at upper secondary school and approximately 20 per cent of the variance of student achievement. However, the effect of student perception of social barriers to their career development is minimally explained by the background variables. Approximately 14 per cent of the variance of prestige aspiration is explained for both groups while the model explains 13 per cent of female students’ intrinsic aspiration but only 7 per cent of male students’ intrinsic application.

CONCLUSIONS Students in Norwegian upper secondary schools at the beginning of 21st century are highly interested in jobs with intrinsic values. This is especially true for female students. The intrinsic value of student career aspiration reﬂects a low level of pressure for compromise in their career development according to Gottfredson’s theory of circumspection and compromise (Gottfredson, 2002). However, other career values such as utility and social values are also highly sought after, with male students more interested in the utility values of their future jobs. To some extent, and especially so among males, students are aware of potential barriers to their career development due to their socio-economic background but they do not know that background factors exerted inﬂuence on their career development long before their entrance into upper secondary school and that these inﬂuences remain long after they enter working life. As presented in the conceptual framework of this chapter, the statistical results conﬁrm that student choice of study path at upper secondary school, academic achievement and attitudes towards future career are inﬂuenced by

Gender Segregation in Student Career Aspirations

351

student home socio-economic and ethnic background as well as by place of residence. Parental educational attainment, perceived by the students to be the least important factor in their career development, has a very strong inﬂuence on student choice of study path, which is the strongest determinant of student career and exerts a strong inﬂuence on student career aspiration. Interestingly, home social status and home economic status have different impacts on student career choice and aspiration. However, these effects should be viewed with caution as measures of home economic status lack accuracy in this case. Female students in Norwegian upper secondary schools achieve better results and are more likely to aspire to higher education than male students. Nevertheless, academic achievement has similar weak effects on career aspiration for both genders. However, female students who choose academic paths appear to give up chasing the intrinsic values of their career aspirations and to be realistic vis-a`-vis the utility values of their career. This effect is especially pronounced among girls with immigrant backgrounds. Awareness of the social barriers to their career development makes both female and male students seemingly more practical in pursuing the utility values of future jobs. Student choice of study subject at upper secondary school turns out to be the most consequential factor in gender disparity in student career aspiration. We do not know the input of the student psychological states at the point of choosing a study path upon entrance to upper secondary school but we do know that this choice does not work well for some students, especially girls. The strong inﬂuence home social status exerts on chosen study path makes it clear that parents are one of the most important factors in the decision-making process. This implies that the inﬂuence of parents is the key factor shaping career aspiration and eventual career choice as it, in fact, represents the opportunity structure as well as class culture in Norwegian society. The results of this study clearly suggest that home background has a strong inﬂuence on gender differences in students’ career choice by introducing the opportunity structure of the society. The Norwegian educational system also contributes to this gender gap through the stratiﬁcation function of a study path choice at upper secondary school. In both cases, female students seem to be at disadvantage. However, student age has very little to do with career choice at this stage. Therefore, some implications for educational policy change could be: (1) postpone study path choice to a later age or a later stage of education instead of age 16 upon entrance to upper secondary school so that students are better prepared to make their choice; or (2) schools provide good career counselling to students

352

LIHONG HUANG

before they make study path choice and expand vocational subjects at vocational path while opening the university to students who choose either path at their upper secondary education; or (3) more incentives for females to enter traditional male professions and vice versa for males entering traditional female professions in order to gradually change the opportunity structure of the Norwegian labour market. The situation in schools reﬂects the reality in society. Gender segregation in career choice and in working life is a result of a much more profound and complex process, which involves actors from every level in the structure of an education system and in the society as a whole. This chapter captures only a small piece of these processes. Moreover, as student career aspiration is only partially explained by the factors investigated in this chapter, further research is needed to include other factors such as parents’ attitudes and their actual input on the process of their children’s career decision-making. It is also important to investigate how other factors in a student’s life such as teachers, friends, media and Internet lead them to internalise the opportunity structure of society and realise it in their career aspirations and career choices.

ACKNOWLEDGMENT The chapter presents some of the results of the ‘‘Reproduction of Social Inequality in Schools’’ project, conducted at NOVA – Norwegian Social Research. Save the Children Norway funds the project. The author would like to express her gratitude to four anonymous reviewers for their valuable comments on earlier drafts.

REFERENCES Armstrong, P. I., & Crombie, G. (2000). Compromises in adolescents’ occupational aspirations and expectations from grades 8 to 10. Journal of Vocational Behavior, 56, 82–98. Ayalon, H. (2003). Women and men go to university: Mathematical background and gender differences in choice of ﬁeld in higher education. Sex Roles, 48(5/6), 277–290. Baker, D. P., & Jones, D. P. (1993). Creating gender equality: Cross-national gender stratiﬁcation and mathematical performance. Sociology of Education, 66, 91–103. Barth, E., Røed, M., & Torp, H. (2002). Towards a closing of the gender pay gap: A comparative study of three occupations in six European countries. Oslo: Institute for Social Research. Blackstone, R. M., Browne, J., Brooks, B., & Jarman, J. (2002). Explaining gender segregation. British Journal of Sociology, 53(4), 153–536.

Gender Segregation in Student Career Aspirations

353

Bradley, K. (2006). Cultural coexistence: Gender egalitarianism and difference in higher education. In: D. P. Baker & A. W. Wiseman (Eds), The impact of comparative education research on institutional theory (pp. 75–94). Oxford: JAI Press. Brint, S., & Karabel, J. (1989). The diverted dream: Community colleges and the promise of educational opportunity in America, 1900–1985. New York: Oxford University Press. Buller, H., & Hoggart, K. (Eds). (2004). Women in the European countryside. Wiltshire: Ashgate Publishing Limited. Buru-Bellat, M. (2004). Social inequality at school and educational policy. Fundamentals of educational planning. Paris: IIEP, UNESCO. Cabrera, A. F., & La Nasa, S. M. (2001). On the path to college: Three critical tasks facing America’s disadvantaged. Research in Higher Education, 42(2), 119–149. Carter, D. F. (1999). The impact of institutional choice and environments on African-American and White students’ degree expectations. Research in Higher Education, 40(1), 17–41. Charles, M. (1992). Cross-national variation in occupational sex segregation. American Sociological Review, 57, 483–502. Charles, M., & Bradley, K. (2002). Equal but separate? A cross-national study of sex segregation in higher education. American Sociological Review, 67(4), 573–599. Charles, M., & Grusky, D. B. (2004). Sex segregation in comparative perspective. Palo Alto, CA: Stanford University Press. Choy, S. P., Horn, L. J., Nun˜ez, A., & Chen, X. (2000). Transition to college: What helps at-risk students and students whose parents did not attend college. In: A. F. Cabrera & S. M. La Nasa (Eds), Understanding the college choice of disadvantaged students (pp. 45–63). San Francisco: Jossey-Bass. Correll, S. (2001). Gender and the career choice process: The role of biased self-assessments. American Journal of Sociology, 106, 1691–1730. Daehlen, M. (2005). Change in job values during education. Journal of Education and Work, 18(4), 385–400. Davey, F. H., & Stoppard, J. M. (1993). Some factors affecting the occupational expectations of female adolescents. Journal of Vocational Behavior, 43, 235–250. Desjardins, R. (2003). Determinants of economic and social outcomes from a life-wide learning perspective in Canada. Education Economics, 11(1), 11–38. Dryler, H. (1998). Parental role models, gender, and educational choice. British Journal of Sociology, 49, 375–398. Ellingsæter, A. L. (1999). Gender mainstreaming and employment policy: Norway report. Rapport 99:11. Oslo: Institute for Social Research. Erikson, R., & Jonsson, J. O. (1998). Social origin as an interest-bearing asset: Family background and labour-market rewards among employees in Sweden. Stockholm: Swedish Institute for Social Research. Foss, L. (2005). The Norwegian paradox: World champion in gender equality – Loser in female management recruitment. In: A. M. Fuglseth & I. A. Kleppe (Eds), Anthology for Kjell Grønhaug in celebration of his 70th birthday (pp. 237–258). Bergen: Fagbokforlaget. Frome, P. M., Alfeld, C. J., Eccles, J. S., & Barber, B. L. (2006). Why don’t they want a maledominated job? An investigation of young women who changed their occupational aspirations. Educational Research and Evaluation, 12(4), 359–372. Gati, I. (1993). Career compromises. Journal of Counselling Psychology, 40, 416–424. Gottfredson, L. S. (1981). Circumspection and compromise: A developmental theory of occupational aspirations. Journal of Counseling Psychology Monographs, 28, 545–579.

354

LIHONG HUANG

Gottfredson, L. S. (2002). Gottfredson’s theory of circumscription and compromise. In: D. Brown (Ed.), Career choice and development (4th ed., pp. 85–148). San Francisco, CA: Jossey-Bass. Hansen, M. N. (1993). Sex segregation in higher education: Inﬂuence of parents’ education and social background on students’ choice of ﬁeld (in Norwegian). Tidsskrift for smfunnsforskning, 34, 3–29. Hansen, M. N. (1995). Sex segregation and the welfare state. Report 95: 2. Oslo: Institute for Social Research. Hanson, S. L., Schaub, M., & Baker, D. P. (1996). Gender stratiﬁcation in the science pipeline: A comparative analysis of seven countries. Gender and Society, 10, 271–290. Hausmann, R., Tyson, L. D., & Zahidi, S. (2006). The global gender gap report 2006. Geneva: World Economic Forum. Available at http://www.weforum.org/pdf/gendergap/ report2006.pdf Holland, J. L. (1965). Manual for the vocational preference inventory. Palo Alto, CA: Consulting Psychologists Press. Hotchkiss, L., & Borow, H. (1996). Sociological perspectives on work and career development. In: D. Brown, & L. Brooks (Eds), Career choice and development (3rd ed., pp. 281–336). San Francisco: Jossey-Bass. Hoyle, R. H., & Panter, A. T. (1995). Writing about structural equations models. In: R. H. Hoyle (Ed.), Structural equation modelling: Concepts, issues, and applications. London: Sage. Hu, S. P. (2003). Educational aspirations and postsecondary access and choices: Students in urban, suburban, and rural schools compared. Educational Policy Analysis Archives, 11(14). Available at http://epaa.asu.edu/epaa Huang, L. H. (2005). Elitism and equality in Chinese higher education: Studies of student socioeconomic background, investment in education, and career aspirations. Stockholm: Institute of International Education, Stockholm University. Huang, L. H. (2007). The contribution of home background to student inequality in secondary schools in Norway. In: D. M. Mcinerney, S. Van Etten & M. Dowson (Eds), Standards in education: Research on sociocultural inﬂuences on motivation and learning (pp. 331–345). Greenwich, CT: Information Age Press. Huse´n, T. (1989). Higher education and social stratiﬁcation: An international comparative study. Paris: IIEP, UNESCO. Jacobs, J. A. (1995). Introduction. In: J. A. Jacobs (Ed.), Gender inequality at work (pp. 1–22). London: Sage. Jocobs, J. E., Chhin, C. S., & Bleeker, M. M. (2006). Enduring links: Parents’ expectations and their young adult children’s gender-typed occupational choices. Educational Research and Evaluation, 12(4), 395–407. Jonsson, J. O. (1999). Explaining gender differences in educational choice: An empirical assessment of a rational choice model. European Sociological Review, 15, 391–404. Jo¨reskog, K. G., & So¨rbom, D. (1993). LISREL 8: Users’ reference guide. Chicago: Scientiﬁc Software. Joy, L. (2006). Occupational differences between recent male and female college graduates. Economics of Education Review, 25(2), 221–231. Kaufman, R. L. (2002). Assessing alternative perspectives on race and sex employment segregation. American Sociological Review, 67, 547–572.

Gender Segregation in Student Career Aspirations

355

Lent, R. W., Brown, S. D., & Hackett, G. (1994). Towards a unifying cognitive theory of career and academic interest, choice and performance (Monograph). Journal of Vocational Behavior, 45, 79–122. Levine, P. B., & Zimmerman, D. (1995). A comparison of the sex-type of occupational aspirations and subsequent achievement. Work and Occupations, 22, 73–84. Mastekaasa, A. (2005). Gender differences in educational attainment: The case of doctoral degrees in Norway. British Journal of Sociology of Education, 26, 375–394. Meece, J. L. (2006). Introduction: Trends in women’s employment in the early 21st century. Educational Research and Evaluation, 12(4), 297–303. Miller, L., & Hayward, R. (2006). New jobs, old occupational stereotypes: Gender and jobs in the new economy. Journal of Education and Work, 19(1), 67–93. Mueller, R. O. (1996). Basic principles of structural equation modelling: An introduction to LISREL and EQS. New York: Springer. Nagy, G., Trautwein, U., Baumert, J., Ko¨ller, O., & Garrett, J. (2006). Gender and course selection in upper secondary education: Effects of academic self-concept and intrinsic value. Educational Research and Evaluation, 12(4), 323–345. Nauta, M. M., Epperson, D. L., & Kahn, J. H. (1998). A multiple-groups analysis of predictors of higher level career aspirations among women in mathematics, science and engineering majors. Journal of Counseling Psychology, 45(5), 483–496. OECD. (2001). Knowledge and skills for life: First results from PISA 2000. Paris: OECD. Opheim, V. (2007). Equal opportunities? The effect of social background on transition from education to work among graduates in Norway. Journal of Education and Work, 20(3), 255–282. Perna, L. W. (2000). Differences in the decision to attend college among African American, Hispanics, and Whites. Journal of Higher Education, 71(2), 117–141. Rainey, L. M., & Borders, L. D. (1997). Inﬂuential factors in career orientation and career aspiration of early adolescent girls. Journal of Counseling Psychology, 44, 160–172. Rehberg, R. A., & Westby, D. L. (1967). Parental encouragement, occupation, education and family size: Artifactual or independent determinants of adolescent educational expectations? Social Forces, 45(3), 263–374. Ringdal, K. (2004). Social mobility in Norway 1973–1995. In: R. Breen (Ed.), Social mobility in Europe. Oxford: Oxford University Press. Robinson, C. L., Taylor, T., Tomaskovic-Devey, D., Zimmer, C., & Irvin, JR. M. W. (2005). Study race or ethnic and sex segregation at the establishment level: Methodological issues and substantive opportunities using EEO-1 reports. Work and Occupation, 32(1), 5–38. Sax, L. J., & Bryant, A. N. (2006). The impact of college on sex-atypical career choices of men and women. Journal of Vocational Behavior, 68(1), 52–63. Schoon, I., & Parsons, S. (2002). Teenage aspirations for future careers and occupational outcomes. Journal of Vocational Behavior, 60, 262–288. Schoon, I., Martin, P., & Ross, A. (2007). Career transitions in times of social change: His and her story. Journal of Vocational Behavior, 70(1), 78–96. Shapka, J. D., Domene, J. F., & Keating, D. P. (2006). Trajectories of career aspirations through adolescence and young adulthood: Early math achievement as a critical ﬁlter. Education Research and Evaluation, 12(4), 347–358. Shavit, Y., & Blossfeld, H. (1993). Persistent inequality: Changing educational attainment in thirteen countries. Boulder, CO: Westview Press.

356

LIHONG HUANG

Skogen, K. (1996). Young environmentalists: Post-modern identities or middle-class culture? The Sociological Review, 44(3), 452–473. Skogen, K. (1998). A touch of class – The persistence of class cultures among Norwegian youth. Young, 6(2), 15–37. Smart, J. C., & Thompson, M. D. (2001). The environmental identity scale and differentiation among environmental models in Holland’s theory. Journal of Vocational Behavior, 58, 436–452. Statistics Norway. (2006). Women and men in Norway: What the ﬁgures say. Statistics Norway. Available at http://www.ssb.no/ola_kari_en Super, D. E. (1990). A life-span, life-space approach to career development. In: D. Brown & L. Brooks (Eds), Career choice and development: Applying contemporary theories to practice (2nd ed., pp. 197–261). San Francisco: Jossey-Bass. Super, D. E., Savickas, M. L., & Super, C. M. (1996). The life-span, life space approach to careers. In: D. Brown & L. Brooks (Eds), Career choice and development: Applying contemporary theories to practice (3rd ed., pp. 121–178). San Francisco: Jossey-Bass. Tuijnman, A. C., & Keeves, J. P. (1997). Path analysis and linear structural relations analysis. In: J. P. Keeves (Ed.), Educational research methodology and measurement: An international handbook (2nd ed.). Oxford: Pergamon Press and Elsevier Science. Vignoli, E., Croity-Belz, S., Chapeland, V., de Fillipis, A., & Garcia, M. (2005). Career exploration in adolescents: The role of anxiety, attachment, and parenting style. Journal of Vocational Behavior, 67(2), 153–168. Webb, R. M., Lubinski, D., & Benbow, C. P. (2002). Mathematically facile adolescents with math–science aspirations: New perspectives on their educational and vocational development. Journal of Educational Psychology, 94, 68–81.

GENDER DIFFERENCES IN POLITICAL EFFICACY AND ATTITUDES TOWARD WOMEN’S RIGHTS AS INFLUENCED BY NATIONAL AND SCHOOL CONTEXTS: ANALYSIS FROM THE IEA CIVIC EDUCATION STUDY$ Carolyn Barber and Judith Torney-Purta Debates on the extent, nature, and sources of gender differences in political socialization have gone on for at least four decades (as reﬂected in the reviews by Alozie, Simon, & Merrill, 2003; Hooghe & Stolle, 2004; and Sapiro & Conover, 2001). Overall, these debates have focused on examining the differences in educational resources, the role of women in the family, and the number of women in the elite levels of politics as these factors may contribute to females in general being less politically and civically engaged. Many of the studies that support a gendered view of political life consider

$

A previous version of this analysis was presented at the 2006 International Society for Political Psychology annual conference in Barcelona (Spain).

Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 357–394 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010014

357

358

CAROLYN BARBER AND JUDITH TORNEY-PURTA

differences between men and women in their knowledge of current political information, in their expectations of running for political ofﬁce, or in their interest in politics. Smaller gender gaps are found when the focus of politics shifts from national- to local-level politics; however, they still exist in favor of males (Delli Carpini & Keeter, 2000). As Kennedy (2006) has acknowledged, much of this research on gendered differences in political engagement has focused on one country at a time, though some cross-national studies do exist. More recently, however, crossnational research has not only focused on the extent to which such gender gaps appear cross-nationally, but has also used country- and individual-level characteristics to explain both the magnitude and direction of such gaps in civic-related outcomes. Research on political knowledge gaps, drawing on data from 23 countries participating in the Comparative Study of Electoral Systems, found males scoring higher on this outcome than females (Gro¨nlund & Milner, 2006). This was measured by knowledge of how political institutions work and of the ideologies held by leading political ﬁgures and parties. Other research on adult differences in attitudes has drawn on data collected in the World Values Survey (WVS) and other large-scale research projects. Inglehart and Norris (2000), for example, used the WVS to explore the gender gap in political ideology and behavior. They found that women expressed an interest in voting for more liberal candidates than did men in postindustrial countries. The opposite was true in post-communist and developing societies, where women preferred the more conservative candidates. Women favoring more liberal candidates in postindustrial countries could be partially explained by differences in social status. These factors, however, did not have the same inﬂuence on the gender gap in political ideology in post-communist and developing countries. This is an example of how interactions between gender and other factors may differ in different countries. What do cross-national patterns observed among adults mean for interpreting the development of political and civic attitudes and norms of engagement among young people internationally? The purpose of this chapter is to explore gender gaps among adolescents, who are in the process of developing such attitudes and norms, and to do so in various national contexts. We will look at the inﬂuence of the position of women in society as well as at the characteristics of the educational context to explain observed gaps between male and female students in two civic-related attitudes. In order to do this, we will consider data collected on 14-year-old students and their schools in 28 countries surveyed in the International Association for the Evaluation of Educational Achievement (IEA) Civic Education Study of 1999.

Analysis from the IEA Civic Education Study

359

PREVIOUS RESEARCH ON GENDER AND CIVIC ENGAGEMENT Theories and research on gender and civic engagement have changed dramatically since studies were conducted 50 years ago. Over time, deﬁnitions of political socialization, knowledge, and engagement have all evolved, and with these developments come differences in how we view male and female political and civic engagement. Early Studies The ﬁrst empirical studies focusing on the development of political and civic knowledge and attitudes among children and adolescents, often referred to as studies of political socialization, were conducted in the 1960s and the early 1970s. At this time, there were gender gaps in many of the most traditional areas of political socialization. Male students received higher scores on tests of civic knowledge, were more likely than female students to plan on participating in political activities, and were more politically selfefﬁcacious (Hess & Torney, 2005[1967]; Torney, Oppenheim, & Farnen, 1975). To contrast, females were more supportive of women’s political and economic rights. Political socialization studies at this time focused both on students within the United States (Hess & Torney, 2005[1967]) as well as students across several countries (Torney et al., 1975). In the ﬁrst IEA study on civic education, conducted in eight countries, gender gaps in knowledge were found in four of the participating countries, whereas differences in political attitudes were found in support for women’s rights and some other attitudinal items such as perception of political conﬂict (Torney et al., 1975). During the intervening decades, research and theory would reﬂect similar patterns in political socialization favoring males (as described in a literature review by Alozie et al., 2003). However, the issue gained more prominence in postindustrial countries in the 1970s with the advent of the women’s movement. With this shift came new ways of thinking in the legal and academic spheres about what these gender gaps meant, whether they needed to be addressed, and (if so) what the best process for eliminating them would be (Sapiro & Conover, 2001, among others). With this new focus on gender equality also came new thinking about the source of observed gender gaps in political socialization. Most important for the study of adolescents’ civic engagement was the growing assumption that differences between males and females come in part from differences in cultural and social aspects of

360

CAROLYN BARBER AND JUDITH TORNEY-PURTA

human development, rather than solely from differences in the attainment of resources such as income and education as adults (see Norris (2007) in her discussion of feminist challenges to theories of political behavior). In other words, ﬁndings from early studies such as those of Hess and Torney (2005[1967]) and Torney et al. (1975) suggested that up through the 1970s females were being socialized at a young age to expect to be less involved in the political realm as adults. The process by which that socialization takes place (and how it could be changed to foster greater equality), however, has not been carefully explored. This is not to say that differences in the attainment of education and other resources did not play a role in understanding gender gaps in political socialization observed during this time. A longitudinal study of adolescents and young adults conducted from 1965 to 1971 found gender gaps in expectations of political activity favoring males (Paulsen, 1991). However, more in-depth analysis revealed that these gender gaps existed primarily among students who did not choose to pursue further education. In interpreting this ﬁnding, it is important to remember the extent to which university campuses in the United States served as a center of political action during this period. Male and female university students also participated in political activism in many other countries (Barnes & Kaase, 1979). Recent surveys of women who graduated from the University of Michigan during these times of political activism reinforce the conclusion that protest participation by late adolescents and young adults is a critical experience in developing political awareness (Zucker & Stewart, 2007). In short, the effect of higher education on political socialization may be understood both as providing access to important resources and serving as a context for political activism by individuals of both genders. Another perspective focuses not on the ways in which female students may be ‘‘inferior’’ to male students in civic engagement, but rather on ways in which females’ civic engagement may be unique. For example, Conover (1988) developed a cognitive-affective model connecting political thinking to social group membership. In other words, individuals are thought to be more sympathetic to the group to which they belong, and this sensitivity to a particular group in turn inﬂuences political thinking. Using this model, Torney et al.’s (1975) ﬁnding that females show greater support for women’s rights can be interpreted as sensitivity to social justice issues among a power-minority group. This sensitivity in and of itself, even if not an important aspect of political socialization studies in the 1960s and 1970s, is a deﬁning characteristic of women’s political development deserving further research.

Analysis from the IEA Civic Education Study

361

Recent Studies Recently, the study of political, social, and civic engagement has expanded to consider gender gaps in more nuanced and multidimensional ways. Among adults, Atkeson and Rapoport (2003) examined gender differences in the willingness to express political attitudes in the National Election Survey from 1952 to 2000, and found gender gaps of consistent size, with females less likely to volunteer opinions about candidates and more likely to choose ‘‘Don’t Know’’ options. Data collected on adolescents in the 1990s and 2000s, however, revealed gendered patterns of participation that are complex. In some respects, male students still appeared more civically engaged. For example, political discussions have been found in recent studies to be dominated by young men. Studying young participants in a model United Nations program, Rosenthal, Jones, and Rosenthal (2003) found that many more male students took part in discussions than did female students. Moreover, students participating in this program perceived politics as an inherently male domain. This gendered pattern was inﬂuenced by the gender of who was leading the group discussions more than by the proportion of females in a given group. This view of politics as a male domain among adolescents, however, may not hold consistently across national contexts. A study of a nonrepresentative sample of junior high-school adolescents in China, Japan, Mexico, and the United States revealed that although perceptions of interest in politics were similarly low in all four countries among male students, they varied greatly among females (Mayer & Schmidt, 2004). Greater proportions of female students in Mexico and China, two developing countries, agreed with a statement that ‘‘girls are more interested in politics’’ than they did in the United States and Japan. However, in every country except for China, the gender gap in the perception of females as interested in politics can be explained away as other individual characteristics such as age, athome political discussion, family structure, trust in the government, and individualistic versus collectivistic orientation are controlled for using multiple regression analyses. At the same time, other recent studies have also suggested that females may be catching up to, and even surpassing, males in civic engagement. A study of U.S. students participating in the Kids Voting USA program found that female students surpassed male students in many dimensions of political engagement (Alozie et al., 2003). These dimensions included participating in political discussions in school and at home, consuming political media (on television, radio, and in the newspaper), expecting to vote, and viewing

362

CAROLYN BARBER AND JUDITH TORNEY-PURTA

voting as important. Although these gender differences began in childhood, Alozie and colleagues found that they continued through adolescence at almost the same level. Moreover, the gap in an overall political orientation composite (based on these and other indicators of political activity) only widens as additional contextual variables are controlled for (such as race/ ethnicity and family subscription to a newspaper). In comparing Alozie et al.’s study to those conducted by Rosenthal et al. (2003) and Mayer and Schmidt (2004), however, it is necessary to note that this study did not consider whether these students perceived these forms of political activity to be more appropriate for males or females. Interest in politics and discussion of political issues are not the only conventional civic- related outcomes for which mixed results have been found. Studies of civic knowledge similarly found some results favoring males and some favoring females. Historically, the National Assessment of Educational Progress (NAEP) Civics Report Card, conducted in the United States, has found higher levels of civic knowledge among male students (Niemi & Junn, 1998). However, in more recent NAEP Civics female students performed higher on average than at grades 8 and 12, although a greater proportion of male students scored at the ‘‘advanced’’ level than did female students (Lutkus, Weiss, Campbell, Mazzeo, & Lazer, 1999). Although subsequent sections of Lutkus et al.’s ﬁndings address the context of civic education, these results are not disaggregated by gender. Therefore, it is difﬁcult to tell in these basic reports whether schools treat civic education as an inherently more ‘‘male’’ or ‘‘female’’ domain. Moreover, the extent to which these differences observed in U.S. students reﬂect more global trends in civic knowledge cannot be ascertained from this single study. Although cross-national research on civic knowledge among adults1 has favored males (e.g., Gro¨nlund & Milner, 2006), additional information is needed on how adolescents’ civic knowledge develops in a variety of contexts. Further, research on sixth-grade students by Rettinger (1993) suggests that the content of questions may explain gender differences, as female students tended to answer more questions on social justice issues correctly, whereas male students answered more questions about history (including military history) and government correctly. As Kneedler (1988, cited in Rettinger, 1993) noted, the advantage of male students in these areas may be due to the disproportionate number of male adults who are visible actors in accounts of history and descriptions of government. Recently, research on political socialization and civic engagement has expanded to include community-based forms of participation in addition to conventional electoral participation. Although differences in community

Analysis from the IEA Civic Education Study

363

participation among adult men and women appear to be in the type of participation rather than its frequency (Norris, 2007), research on adolescents tells a different story. Although conventional engagement (political knowledge and discussion) has traditionally been thought to be the domain of males, research on community-based forms of participation has identiﬁed female adolescents as more participatory, regardless of whether that volunteerism involves political or non-political activities. Some of this research has focused on one country (e.g., Rosenthal, Feiring, & Lewis, 1998, in the United States); however, other research on non-representative samples has conﬁrmed that this gender gap exists in several countries. Flanagan, Bowes, Jonsson, Csapo, and Sheblanova (1998), for example, found this gap in Australia, Bulgaria, Hungary, Russia, and the United States, but not in Sweden or the Czech Republic. Although particular contexts of community service in Sweden and the Czech Republic may have contributed to this lack of a gender gap, analysis of Flanagan and colleagues’ data did not suggest that these gaps correspond to overall lower expectations of community service in these countries. The consideration of community-based forms of civic participation presents challenges to many traditional conceptualizations of gender and political socialization, which can be described as adopting a deﬁcit model in which females lack political resources and power (Hooghe & Stolle, 2004). Additional research, therefore, has attempted to explain why females and males are drawn to different types of participation. An interview study of children and adolescents by Rettinger (1993) found that students view traditional politics as a ‘‘top-down’’ institution in which few have control over many. Rettinger then argued that this view of politics lies in opposition to females’ personal beliefs on how problem solving should occur, which values coming together as a community and caring for others. Rettinger’s study provides a possible explanation for the different preferences of male and female students.

GENDER DIFFERENCES AND THE IEA CIVIC EDUCATION STUDY The largest study in which it is possible to examine gender differences in political socialization is the IEA Civic Education (CIVED) Study of 1999. This study, conducted by the same organization that sponsored the Civic Education Study cited previously (Torney et al., 1975), assessed the civic knowledge and engagement of nationally representative samples of

364

CAROLYN BARBER AND JUDITH TORNEY-PURTA

14-year-old students in 28 countries.2 This enabled researchers to take a comparative perspective on civic education, interpreting cross-national differences among students in terms of variations in their social, political, and educational contexts. In fact, a series of qualitative case studies, developed during the ﬁrst phase of the CIVED study, was instrumental in providing researchers the background needed to interpret the differences found in the survey (Torney-Purta, Amadeo, & Schwille, 1999). In addition to its comparative perspective and national representativeness, the CIVED study also expanded previous research on political socialization in several other important ways. First, the CIVED study broadly conceptualized both civic knowledge and participation. Its test on civic knowledge did not focus on speciﬁc current events covered in the news media (on which males often do especially well), as several of the adult studies have (e.g., Delli Carpini & Keeter, 2000). Rather, it assessed students’ knowledge of key concepts of democracy as well as their skills in interpreting political material such as election leaﬂets and cartoons. This content knowledge and skill set is thought to be more fundamental to an understanding of politics and to participation than is the identiﬁcation of current political ﬁgures (who vary in their visibility according to the political structure of different countries). Similarly, civic participation itself was considered in a variety of ways. The CIVED study surveyed students on their intent to run for political ofﬁce or to vote, but it also asked about their intent to volunteer in their communities and to participate in protest activities. Given the extent to which tests of current events and surveys of intent to run for political ofﬁce have tended to favor males (Delli Carpini & Keeter, 2000; Rosenthal et al., 2003; Gro¨nlund & Milner, 2006), the expanded conceptualization of both civic knowledge and skills could have important implications for assessing gender gaps in civic development. In the CIVED study, the concept of ‘‘civic engagement’’ was formally expanded to consider whether an individual possesses certain attitudes related to participation in such activities. As a result, the major reports of the CIVED study (Torney-Purta, Lehmann, Oswald, & Schulz, 2001; Amadeo, TorneyPurta, Lehmann, Husfeldt, & Nikolova, 2002) not only summarized students’ civic knowledge and expected participation across countries, but also their beliefs in norms of citizenship (i.e., what sorts of behaviors a ‘‘good citizen’’ engages in), attitudes toward their country and government, and their support for extending political rights to minority groups such as immigrants and women. This expanded conceptualization of what it means to be a ‘‘civically engaged’’ citizen provides additional outcomes to consider as they reﬂect differences between male and female students (Hooghe & Stolle, 2004).

Analysis from the IEA Civic Education Study

365

Gender Gaps in Knowledge, Attitudes, and Behaviors The assessment of gender differences in civic knowledge, behaviors, and attitudes was a key focus in the report on the CIVED study of 14-year-olds (Torney-Purta et al., 2001) and in additional reports on the development of new measures of civic engagement (Husfeldt, Barber, & Torney-Purta, 2005). Torney-Purta and colleagues and Husfeldt and colleagues found systematic gender gaps across countries in some measures of civic engagement, but not in others. One important outcome in which systematic gender differences were not observed was civic knowledge. Unlike tests of civic knowledge focusing on current events, which usually favor males (Delli Carpini & Keeter, 2000), CIVED’s test of democratic concepts and interpretive skills did not yield noticeable gender differences. According to the major report on the study of 14-year-olds, a statistically signiﬁcant gender gap was only observed in one of the 28 countries (Slovenia), and that difference favored female students over males (Torney-Purta et al., 2001). Although these results contrast with studies of adult knowledge, they mirror the ﬁndings of other large-scale studies of adolescents’ civic knowledge, including the ﬁrst Civic Education Study (Torney et al., 1975) and the NAEP Civics Studies in the United States (Lutkus et al., 1999). Like other studies, however, the IEA Civic Education Study did ﬁnd more systematic signiﬁcant gender gaps favoring males in several other outcomes. Male students had stronger expectations of future participation in conventional political activities (such as running for political ofﬁce) than did females in nine of the 28 participating countries (Torney-Purta et al., 2001). These nine countries, diverse in their geographic location and recent political history, were Belgium (French), Cyprus, Estonia, Greece, Hong Kong, Italy, Lithuania, Romania, and Switzerland. Even more systematic differences were found when comparing male and female students’ expectations of participating in illegal protest activities in the future (e.g., spray-painting protest slogans), present in 26 of the 28 countries (including the United States, as discussed by Hooghe & Stolle, 2004). Extending the conceptualization of civic engagement to attitudes and dispositions also revealed other signiﬁcant differences favoring males. In 23 of the 28 countries, male students reported having higher self-efﬁcacy for participating in political discussions than females. Only in Belgium (French), Bulgaria, Chile, Colombia, and the United States did male and female students report similar levels of efﬁcacy (Husfeldt et al., 2005). Additional analyses identiﬁed other civic-related measures in which female students scored higher. Expanding the deﬁnition of ‘‘civic

366

CAROLYN BARBER AND JUDITH TORNEY-PURTA

participation’’ to include work in the community revealed that females were signiﬁcantly more likely to expect to participate in such activities than males (Husfeldt et al., 2005). Signiﬁcant gender differences favoring females were found in every country except Bulgaria, Romania, and the Russian Federation. Additionally, analysis of U.S. CIVED data by Hooghe and Stolle (2004) found that greater proportions of females also planned to vote, to gain information on candidates in elections, and to protest peacefully. Further expanding the conceptualization of ‘‘civic engagement’’ overall to include attitudes and dispositions as well as knowledge and participation, the CIVED study revealed that females have systematically more positive attitudes toward the inclusion of women in the political process than do males. As previously discussed, one may expect to ﬁnd systematic gender differences favoring females in support for women’s rights (Conover, 1988). Indeed, there were substantial gender differences in this attitude favoring female students across all 28 countries (Torney-Purta et al., 2001). These ﬁndings from 14-year-olds participating in the CIVED study mirror ﬁndings of adults from these same countries in the WVS in two respects. First, the country rank order in support for women’s political rights is very similar in CIVED and WVS, for the 18 overlapping countries. The top ﬁve countries in support for women’s rights in WVS all had scores on the parallel CIVED measure that are signiﬁcantly above the international mean; likewise, the bottom ﬁve countries in WVS all have averages on the parallel CIVED measure that are signiﬁcantly below the international mean (Inglehart & Norris, 2003). Second, females surveyed in WVS also have signiﬁcantly higher support for women’s rights, a ﬁnding that holds in postindustrial, industrial, and agrarian societies. This gender gap in attitudes was also seen in support for political rights of immigrants in the CIVED study (23 out of 28 countries) and ethnic minorities (25 out of 28 countries) (Husfeldt et al., 2005; Torney-Purta et al., 2001), suggesting a similarity between attitudes toward women’s rights and attitudes toward the support of rights for other diverse groups (also supported by Kane, 2000). These ﬁndings from 14-year-olds participating in the CIVED study mirror ﬁndings of adults from these same countries in the WVS, suggesting that this gender gap in social justice issues other than women’s rights is a societal phenomenon rather than a developmental one (Inglehart & Norris, 2000). Additional analysis of CIVED data by Kennedy (2006) reminds us, however, that these distributions overlap, with substantial groups of males highly supportive of women’s and minorities’ rights. This suggests that other personal and contextual factors need to be considered in addition to gender.

Analysis from the IEA Civic Education Study

367

In summary, the expanded conceptualization of civic engagement provided by the CIVED study and supported by other studies of young adults (Zukin, Keeter, Andolina, Jenkins, & Delli Carpini, 2006) reveals a more complex picture of gender differences than has been available from previous research. An expanded consideration of civic participation and of attitudes as a type of civic engagement revealed that whereas male students were more conﬁdent in political participation and more likely to engage in conventional and illegal activities, female students were more likely to plan to participate in their communities and to extend political rights to minorities. Although these differences are observed cross-nationally, differences in the magnitude of these gender gaps in various countries and contexts deserve to be the topic of more complex analysis.

Explaining Observed Gender Gaps What accounts for the observed gender differences in the CIVED outcomes? Researchers have attempted to explain these differences with varying degrees of methodological sophistication. In commenting broadly on the observed differences in the major CIVED report, Torney-Purta et al. (2001) noticed that gender differences were more pronounced in older democracies (e.g., the United States) and less pronounced in the newer democracies (e.g., Russia). One explanation for this pattern may be that adult women tended to be more socially conservative than men in post-communist countries (when compared to longer-established democracies), and thus less likely to endorse women’s rights and other social justice attitudes (Inglehart & Norris, 2000; Malak-Minkiewicz, 2007). Patterns observed in the CIVED study, then, can be interpreted as evidence that 14-year-old students are absorbing the gendered norms of political engagement in present in their cultures. However, only a few authors have gone beyond documenting and brieﬂy commenting about gender differences. The relationship of percent of women in national parliaments to adolescents’ support for women’s rights in different countries was noted in both of the IEA Civic Education Studies (Torney et al., 1975; TorneyPurta et al., 2001). This line of research was carried further by Campbell and Wolbrecht (2006), who focused on the predictive power of one variable (the visibility of female politicians) in explaining why gender differences in CIVED’s measures of political participation were less pronounced in certain areas both within a given country and across several countries. Looking within the United States, Campbell and Wolbrecht found that the relation

368

CAROLYN BARBER AND JUDITH TORNEY-PURTA

between females’ expectations of political participation and the presence of viable female candidates in their political districts was mediated by participation in political discussion, particularly with parents. Looking across countries, the researchers similarly found that countries with more female parliament members had higher average expectations of participation among females, and in these countries females also reported discussing political issues more often among friends (Wolbrecht & Campbell, 2007). Campbell and Wolbrecht’s research has shown how more distal inﬂuences (e.g., visible female political presence) and more proximal processes in family and peer contexts (e.g., discussion with parents) inﬂuence female’s political engagement. However, in explaining differences between male and female students, it is also important to look at other proximal contextual inﬂuences. In addition to its comparative focus, one of the main goals of the CIVED study was to examine the extent to which schools foster the civic development of students (Torney-Purta et al., 2001; Torney-Purta, 2002). The potential of the CIVED study to assess the school’s role in reproducing or reducing gender inequalities has been largely overlooked by researchers. As access to education in secondary school and beyond serves as an important pathway to citizenship, particularly for women, the understanding of this context is potentially very important (Paulsen, 1991; Stromquist, 2006). Moreover, participation in classes in which social issues are explicitly discussed (especially as related to diversity and equality) has been shown to increase students’ identiﬁcation with, and support for, these issues (Case, 2007; Torney-Purta, Barber, & Wilkenfeld, 2007). Therefore, the discussion of these issues within the educational context is potentially as important for understanding gendered differences in civic outcomes as for educational attainment in itself.

THE CURRENT STUDY The purpose of the current analysis is to analyze the CIVED study’s wideranging survey data collected from nationally representative samples crossnationally to understand observed gender gaps in two civic-related outcomes. This study focuses on the two outcomes that show systematic gender gaps cross-nationally: support for women’s rights (favoring females in all 28 countries) and internal political efﬁcacy (favoring males in 23 of the 28 countries). There are several conceptual reasons why these two outcomes are important. Adolescents’ support for social justice, including support for political and economic gender equality, has been an important aspect of

Analysis from the IEA Civic Education Study

369

civic engagement since the ﬁrst Civic Education Study (Torney et al., 1975) and in other studies of political attitudes. The expanded consideration of attitudes and dispositions as outcomes of civic education that began with the 1999 CIVED study further strengthens the case for considering support for social justice issues. If the next generation is to work toward gender equality, both males and females must believe that these issues are important. It is well documented that female adults, as well as adolescents, express greater support for women’s rights (Conover, 1988). Jennings (2006) has recently argued, based on longitudinal and cross-generational data that these gender gaps are larger in the current generation of young adults than in the generations of their parents and grandparents. More speciﬁcally there is a growing ambivalence or resistance to gender equality among the ‘‘postfeminist-movement’’ generation of young men. Jennings cites other data from a variety of surveys pointing in the same direction. However, local school contexts may also be important to consider, especially for females in countries with relatively low levels of overall support for gender equality. Social cognitive theory posits that internal self-efﬁcacy serves as a means of self-motivation, and that individuals who believe that they can do something are more likely to do it than those who do not (Armingeon, 2007; Bandura, 1997; Barber, Torney-Purta, Homana, & Wilkenfeld, 2007). Norris (2007) described internal political efﬁcacy as one of the psychological orientations that is ‘‘most associated with political participation’’ (p. 733). She and other authors refer to documented gender differences favoring males (see also the classic study by Verba, Burns, & Scholzman, 1997). Systematic gender gaps in efﬁcacy as it relates to political discussions are likely to translate into systematic gender gaps in actual participation in political discussions (Rosenthal et al., 2003), which is an important arena for civic engagement. Fernandez-Ballesteros, Diez-Nicolas, Caprara, Barbaranelli, and Bandura (2002) additionally found a gender difference favoring males in sense of collective efﬁcacy in promoting social change among Spanish adults. In addition to considering these two important civic-related outcomes in a more nuanced way, this analysis also furthers our understanding of gender gaps through the use of complex statistical methodologies for analysis. It uses multilevel modeling techniques to allow characteristics of schools and countries to predict the magnitude of observed gender gaps in countries and school systems that differ in systematic ways. Unlike studies focusing on gender gaps that examine country-level indicators only (e.g., analysis of WVS data by Tesch-Ro¨mer, Motel-Klingebiel, & Tomasik, 2008) or on individual-level indicators only across several countries (e.g., analysis of

370

CAROLYN BARBER AND JUDITH TORNEY-PURTA

data from the International Social Justice project by Jasso & Wegener, 1999), a multilevel model takes into account variability in attitudes and contexts both within and between countries. In doing so, we can not only control for individual background, but also consider the more proximal inﬂuences of school context along with more distal country-level contextual factors. Norris (2007) points out that European studies (in contrast to those conducted in the United States), are much more likely to highlight the structural characteristics of formal institutions in generating a gender gap (in contrast to the U.S., where the explanation more frequently invokes gender role socialization as an individual characteristic rather than looking at the institutional context). In this sense, this current analysis of the CIVED study allows us draw on both of these traditions, highlighting the role of formal institutions while also considering individual correlates of gender socialization. Speciﬁcally, this cross-national analysis of the IEA CIVED data attempts to answer the following research questions: 1. Are gender gaps in support for women’s rights and in internal political efﬁcacy more or less pronounced in schools with certain characteristics? 2. Are gender gaps in these two civic-related outcomes more or less pronounced in countries with certain characteristics? As we address these research questions, we keep in mind several theoretical assumptions that have evolved over the course of studying gender and political socialization. First, we assume that gender differences in political attitudes and involvement can be traced in large part to social and cultural differences in how male and female students are socialized. Preadult political socialization as an explanation for adult political behavior has gone in and out of fashion over the past several decades. Most researchers at present give credence to this process, though there is tendency to prefer terms that recognize this as bi-directional and dynamic (rather than top-down and static) and conceptualizations that do not focus on individual differences alone but consider the social construction process. At both the country and the school levels, we can look at how differences in institutional practices and context moderate gender gaps in political socialization, thus explaining observed gender differences in political attitudes and behavior (i.e., a focus on ‘‘demand-side factors,’’ as described by Norris, 2007). Individual characteristics such as home educational resources (i.e., some of Norris’s ‘‘supply-side explanations’’) may be considered as statistical controls when considering gender gaps, our focus is on how the context of the nation broadly, and of schools more

Analysis from the IEA Civic Education Study

371

speciﬁcally, shape development among male and female adolescents. In particular opportunity structures are of importance. Second, we deﬁne civic engagement as consisting of more than knowledge of, and engagement in, conventional forms of political participation. Our study focuses on two of the possible attitudinal modes of civic engagement; however, our consideration of these two outcomes is situated in the broader deﬁnition of civic engagement that guided the CIVED study (Torney-Purta, 2002).

DATA AND METHODOLOGY About the IEA Civic Education Study In the early 1990s, IEA was asked by its member countries to organize a second study of civic education (following the study summarized in Torney et al., 1975) and in 1993 planning began for this research on civic knowledge, attitudes, and expected behaviors. The ﬁrst phase of the CIVED study (1994–1998) involved extensive collection of qualitative data (as structured national case studies found in Torney-Purta et al., 1999). Civic education did not have a designated curriculum even in many wellestablished democracies, and several emerging democracies in the postcommunist region planned to participate. The case studies were the basis for a consensus process to develop content speciﬁcations for a test of civic knowledge and a survey of political attitudes and civic behavior. These case studies also provided contextual information for interpreting the survey data collected in 1999–2000. Further analysis of the case study data within and across the 24 participating countries is reported in Steiner-Khamsi, Torney-Purta, and Schwille (2002). The second phase of the IEA Civic Education Study began in 1997. International and national coordinators constructed items and piloted an instrument that would be suitable for early and late adolescents (approximately 14 and 17 years of age, respectively). The students were administered a 38-item test of civic knowledge and skills in interpreting political information, with the late adolescents’ test including several more difﬁcult items. The instrument also included a survey of civic attitudes and behaviors, which was substantially the same for the two age groups. Nationally representative samples of students in the modal grade for 14-year-olds were tested in 1999 (results summarized in Torney-Purta et al., 2001), whereas older adolescents were tested in 2000 (Amadeo et al., 2002). Altogether, the

372

CAROLYN BARBER AND JUDITH TORNEY-PURTA

study of 14-year-olds, analyzed here, included data from 93,882 students in 4,136 schools from a total of 28 countries. Secondary Analysis of the CIVED Survey Data Since the major reports of primary analysis (Amadeo et al., 2002; TorneyPurta et al., 2001) and the technical report (Schulz & Sibberns, 2004) were issued in the early 2000s, researchers have been conducting secondary analysis using CIVED data in order to address more speciﬁc research questions regarding civic engagement. Although collected in 1999, the data continue to be valuable for several reasons. First, the survey items, many adapted from those used in adult studies of political attitudes, provide a breadth of information gathered from nationally representative samples of students in 28 countries. The rigor of the sampling and scaling procedures used means that the CIVED study can provide extensive, high-quality data on the nature of civic and political engagement cross-nationally. Such data are time- and labor-intensive to collect, and data from a single large-scale, international study are often analyzed for many years after data collection has been completed (as been the case for collections of data from the WVS, for example). Second, the CIVED participants who were 14-year-olds in 1999 are young adults beginning to ﬁnd their places in the adult political realm in 2008. In this sense, the 1999 CIVED data provide unique information on the development of this generation of young adult citizens. This is, for example, the generation of Chilean students who organized protests about educational quality in 2006 (Franklin, 2006) or who worked in campaign mobilization in the U.S. presidential elections of 2008 (Harvard University Institute of Politics, 2008). Third, the nested structure of the data (i.e., students within schools within countries) allows for the use of multilevel modeling techniques, which can address the role of both national and educational context in shaping political development (a research direction noted positively by Jennings (2007) in his review of political socialization research). Scaling the CIVED Data The data collection and the development of composite measures from the survey data also extended over several years. Thirteen scales assessing civic knowledge and engagement were developed and presented in the major reports of the CIVED study (Schulz & Sibberns, 2004; Torney-Purta et al., 2001). Subsequently, researchers at Civic Education Data and Researcher Services (CEDARS) at the University of Maryland developed eight additional scales as part of an extensive program of secondary analysis of

Analysis from the IEA Civic Education Study

373

the CIVED data (Husfeldt et al., 2005). All of the scales were developed using conﬁrmatory factor analysis techniques to test the hypothesized unidimensionality of each of the scales, and all involved the use of item response theory (IRT) techniques in the scale creation itself. IRT scaling, based on modern measurement theory, allows estimation of missing responses and facilitates comparisons between groups across multiple scales (see Schulz & Sibberns (2004) for a discussion of the application of IRT to the CIVED study).

Examination of Gender Gaps Using IEA Data Outcomes Two of the IRT scales developed by IEA and CEDARS researchers served as the major outcomes in our current analysis. The ﬁrst was the scale of Support for Women’s Political and Economic Rights. This scale, several items from which were originally developed for the ﬁrst IEA Civic Education Study testing in 1971, assesses the extent to which students agreed with a variety of statements about whether men and women have (or should have) the same rights. An example of an item positively loaded on this scale reads ‘‘Women should have the same rights as men in every way.’’ An example of an item negatively loaded on this scale reads ‘‘Men are more qualiﬁed to be political leaders than are women.’’ The psychometric properties of this scale are summarized by Schulz and Sibberns (2004), and national scores on this scale are reported in the major CIVED report on 14-year-olds (Torney-Purta et al., 2001). The second outcome variable considered in this analysis is labeled in CIVED reports as Internal Political Efﬁcacy (Husfeldt et al., 2005). More speciﬁcally, this scale assesses students’ self-efﬁcacy in participating in political discussions. An example of an item in this scale reads ‘‘When political issues or problems are being discussed, I usually have something to say.’’ The psychometric properties of this scale, which was developed by researchers as part of the CEDARS project, and the average national scores can be found in a report by Husfeldt et al. (2005). Predictor Variables Characteristics of students, schools, and countries were considered as predictors of support for women’s rights and internal political discussion efﬁcacy. With lack of a clear multilevel theoretical model for explaining gender gaps favoring men or women, we selected variables at each level that

374

CAROLYN BARBER AND JUDITH TORNEY-PURTA

previous theory and research have found to be important in understanding gender differences in civic engagement generally. These characteristics are summarized in Table 1. Looking ﬁrst at student-level variables, the most important predictor was the students’ gender. In addition, we chose two other student-level variables that were important to consider as statistical controls due to their relationships to the two outcomes of interest. One student-level control was the overall score on the CIVED civic knowledge test (including both items on democratic concepts and skills in interpreting political material). The other was the number of books that students reported at home, which serves as the most commonly used proxy for home socioeconomic resources in IEA studies (Mullis, Martin, Gonzalez, & Chrostowski, 2004; TorneyPurta et al., 2001). Variables at the school level served as proximal-level predictors of gender gaps in these two attitudes. All school-level variables were aggregated from the student level, and included the proportion of females in the school, the average expectations of further education among students in a school, the proportion of students who agreed that women are discriminated against in society (more speciﬁcally, in opportunities for employment and secondary education), and the average perceptions of an open classroom climate for discussion. The proportion of females in a school served as a control variable. Because we chose outcomes based on their substantial gender gaps, we wanted to control for the possibility that average attitude levels in the school were due to higher or lower proportions of students of a particular gender. Our focus on expectations of further education was based on theory and research that uses differences in educational resources to explain gender gaps in males and females (Stromquist, 2006). Although at age 14 many students are too young to have made concrete plans for higher education, it is possible that local expectations about continuing education past secondary school may differ in the strength with which they inﬂuence male and female attitudes. The consideration of the proportion of students in a school who perceived gender inequality in educational contexts and in employment contexts was meant to measure the general perception of ‘‘male privilege’’ within the school (Case, 2007). The perception of discrimination measure is also somewhat similar to a measure used by Jennings (2006) concerning the inﬂuence that each gender has on American life (too much inﬂuence, about the right amount, or too little inﬂuence). Finally, given the importance of discussion in fostering the expectations of conventional participation among females (Campbell & Wolbrecht, 2006; Wolbrecht & Campbell, 2007), we considered the openness of the classroom climates for discussion.

375

Analysis from the IEA Civic Education Study

Table 1.

Description and Unweighted Descriptive Statistics of Student, School, and Country-Level Variables.

Variable and Description Outcome Variables Support for women’s rights Internal political efﬁcacy Student Variables Total civic knowledge Home literacy

Gender School Variables Proportion of girls Standardized average class climate

Local expectations for further education

Percent perceiving school inequality

Percent perceiving employment inequality Country Variables Standardized GDI score National expectations of further education

Description

Attitudinal IRT scale: see Torney-Purta et al. (2001) and Schulz and Sibberns (2004) for details Attitudinal IRT scale: see Husfeldt et al. (2005) for details

Mean

SD

9.99

1.98

10.04

2.00

Total civic knowledge IRT scale: see Torney-Purta et al. (2001) and Schulz and Sibberns (2004) for details Number of books in the home: student report on a scale of 0–6 (0 ¼ 0 books; 6 ¼ more than 200 books) Student self-report: 0 ¼ male; 1 ¼ female

100.61

20.35

3.24

1.33

0.51

0.50

Proportion of students in the school sample self-identifying as female Open classroom climate for discussion IRT scale (see Torney-Purta et al., 2001; Schulz & Sibberns 2004), aggregated to the school level. Standardized at the school level across all schools in the analytic sample to aid in interpretation Student report of expectations of further education on a scale of 1–7 (1 ¼ 0 years; 7 ¼ more than 10 years). These reports are aggregated to the school level and standardized across all schools in the analytic sample to aid in interpretation Proportion of students in the school sample who ‘‘agree’’ or ‘‘strongly agree’’ that ‘‘girls have fewer chances than boys to get a high-school education in this country.’’ Proportion of students in the school sample who ‘‘agree’’ or ‘‘strongly agree’’ that ‘‘women have fewer chances than men to get a good job in this country.’’

0.50

0.10

0.00

1.00

0.00

1.00

0.13

0.12

0.31

0.15

0.00

1.00

0.00

1.00

GDI Score, standardized across countries in the sample Student report of expectations of further education. These reports are aggregated to the country level and standardized across all countries

376

CAROLYN BARBER AND JUDITH TORNEY-PURTA

Finally, variables at the national level, like school-level variables, were considered both as general predictors of attitudes and as predictors of the extent of gender gaps. Given that there are only 28 countries in the CIVED sample, we limited our consideration of country-level variables to two. The ﬁrst variable was the gender development index (GDI; United Nations Development Programme, 2000).3 This is a version of the Human Development Index corrected for gender equality. Higher values on the GDI indicate a higher quality of life for women in the country relative to women in other countries; lower values indicate the opposite. The second variable considered the average expectation of further education among students in the entire country, aggregated from the student level. Each of these variables was re-standardized across countries for ease in interpreting differences among the 28 countries in the CIVED sample. Standardized scores on these variables in each country and their rankings can be found in Table 2. Statistical Analysis We conducted statistical analysis for this three-level model using Hierarchical Linear Modeling software (HLM; Raudenbush, Bryk, Cheong, & Congdon, 2004). In addition to taking into account the nesting of students within schools within countries in this data set, HLM allows for the use of variables at each level of analysis to be considered as predictors of a studentlevel outcome. As a result, this methodology is very appropriate for this cross-national study of gender gaps. In-depth information about the threelevel model used in this analysis can be found in the appendix. Because we used the school-level variables to model differences in attitudes between the male and female students in a given school, this model assumes that all schools have both male and female students. Schools serving only male or only female students cannot be analyzed using this model. Therefore, these schools were removed from the analysis. The resulting analytic sample includes 76,071 students from 3,383 schools in 28 countries. We employed normalized population weights at level 1 such that the within-country sample in each country was representative of 14-year-olds attending co-educational schools in the country. At the same time, we weighted each of the 28 countries equally in analysis, so that countries with larger sample sizes or with larger populations did not contribute disproportionally to analysis.

RESULTS Before adding additional predictors at the three levels, we observed the average magnitude of the gender gaps and their variances. These

377

Analysis from the IEA Civic Education Study

Table 2. Country

Australia Belgium (French) Bulgaria Chile Colombia Cyprus Czech Republic Denmark England Estonia Finland Germany Greece Hong Kong (SAR) Hungary Italy Latvia Lithuania Norway Poland Portugal Romania Russian Federation Slovakia Slovenia Sweden Switzerland United States

Summary of Country-Level Variables and Rankings. GDI Score

Country Expectations of Further Education

Z-score

Rank

Z-score

Rank

1.28 1.11 1.42 0.75 1.59 0.43 0.24 0.94 0.94 0.91 0.94 0.77 0.26 0.10 0.75 0.77 1.42 1.08 1.28 0.75 0.10 1.42 1.42 0.58 0.10 1.11 0.94 1.28

1.00 4.00 24.00 19.00 28.00 12.00 17.00 6.00 6.00 22.00 6.00 10.00 13.00 14.00 19.00 10.00 24.00 23.00 1.00 19.00 14.00 24.00 24.00 18.00 14.00 4.00 6.00 1.00

0.22 0.69 0.12 1.09 0.54 0.04 0.38 0.23 1.37 0.79 0.54 1.57 1.03 1.00 1.14 0.31 0.13 0.59 0.43 2.01 0.99 1.23 0.21 1.43 0.04 0.48 2.02 1.60

16.00 8.00 13.00 5.00 9.00 15.00 18.00 17.00 26.00 7.00 21.00 27.00 24.00 6.00 4.00 10.00 12.00 22.00 19.00 1.00 23.00 25.00 11.00 3.00 14.00 20.00 28.00 2.00

unconditional models of gender gaps in support for women’s rights and internal political efﬁcacy provide a baseline by which our complete, fully conditional models can be judged (Table 3). Before other variables were added, females across all 28 countries in the CIVED study had on average a score on the support for women’s rights scale that was 1.27 points higher than males. With a standard deviation of 2, this amounts to a .64 SD advantage for females over males. However, as seen in the ‘‘Random effects’’ portion of Table 3, the actual size of this gap varied across locations. In addition to signiﬁcant variance components for the intercepts (r0 and u00), which indicate that the average support for women’s rights varied across schools and countries, there were signiﬁcant variance

378

CAROLYN BARBER AND JUDITH TORNEY-PURTA

Table 3. Summary of Unconditional Models of Gender Differences in Support for Women’s Rights and Internal Political Efﬁcacy. Fixed Effects

Support for Women’s Rights Coefﬁcient

Intercept Gender

9.98 1.27**

Random Effects

Error (Level-1) variance (e) Level-2 intercept variance (r0) Level-2 gender variance (r3) Level-3 intercept variance (u00) Level-3 gender variance (u30)

Standard error 0.10 0.08

Internal Political Efﬁcacy Coefﬁcient 9.99 0.58**

Standard error 0.09 0.04

Variance Component

Variance Component

2.93 0.27** 0.31** 0.29** 0.19**

3.50 0.17** 0.31** 0.23** 0.04**

Note: Variables in bold are centered on their group mean. **po.01.

components associated with the gender gaps at both the school level (r3) and the country level (u30). Each of these variance components was signiﬁcantly larger than 0; therefore, the use of both school- and country-level variables to reduce this amount of variance in the magnitude of gender gaps in support for women’s rights was warranted. The unconditional model of the gender gap in internal political efﬁcacy reveals that, on average, male students scored .58 points higher on the Internal Political Efﬁcacy scale. Given that this scale, like the women’s rights scale, has a SD of 2, this amounts to a .33 SD advantage of males over females in terms of internal political efﬁcacy. Here also there were signiﬁcance variance components at both the school and country levels, indicating that the magnitude of these differences varies across contexts. Therefore, the consideration of school- and country-level variables to model the gender gap in internal political efﬁcacy was warranted.

Modeling the Gender Gap in Support for Women’s Rights We ﬁrst examined covariates of, and gender gaps in, support for women’s political rights, which favored female students. The summary of this HLM analysis can be found in Table 4. According to Table 4, civic knowledge and home literary resources were signiﬁcant covariates of support for women’s

379

Analysis from the IEA Civic Education Study

Table 4.

Summary of Hierarchical Linear Model for Support for Women’s Rights.

Fixed Effects Average support for women’s rights score (g000) Standardized GDI score (g001) Proportion girls (b01) Standardized average class climate (b02) Local expectations for further education (b03) Average local expectations (c030) Standardized GDI score (g031) Percent perceiving school inequality (b04) Percent perceiving employment inequality (b05) Total civic knowledge (p1) Number of books in the home (p2) Gender (p3) Average gender gap (c300) Standardized GDI score (g301) Standardized average class climate (b31) Local expectations for further education (b32) Average local expectations (c320) Standardized GDI score (g321) Random Effects Error (Level-1) variance (e) Level-2 intercept variance (r0) Level-2 gender variance (r3) Level-3 intercept variance (u00) Level-3 gender variance (u30)

Coefﬁcient

Standard Error

10.01 0.36** 0.88** 0.07**

0.06 0.08 0.11 0.01

0.03* 0.04* 1.08** 0.27** 0.03** 0.06**

0.02 0.02 0.11 0.09 0.00 0.01

1.28** 0.36** 0.07**

0.05 0.04 0.02

0.10** 0.08**

0.03 0.03

Variance Component (% Reduction) 2.69 0.11** 0.26** 0.10** 0.07**

(8) (59) (16) (66) (63)

Note: Variables in italics are centered on their grand mean; variables in bold are centered on their group mean. All other variables are left uncentered. All variables except for gender have their error variances ﬁxed to zero. *po.05; **po.01.

rights (p1 and p2, respectively). At the school level, support for women’s rights was higher in schools with more females and in schools where students more strongly agree that their classroom climate is open to discussion (b01 and b02). The relation of local expectations of further education to average women’s rights support was more variable (analysis of b03). Being in a school where many students expected to continue their education (local expectations of further education) was associated with higher support for

380

CAROLYN BARBER AND JUDITH TORNEY-PURTA

women’s political rights (g030); this relationship was especially strong in high-GDI countries (g031). Moreover, being in a school where many students perceive gender inequality in education (b04) and in the workplace (b05) was associated with lower support for women’s rights. Only one signiﬁcant country-level covariate was found: support for women’s rights was higher in countries with higher GDI scores (g001). Only a few of these school- and country-level variables, however, can serve to explain why the gender gap in support for women’s rights was of a certain size in certain contexts (analysis of p3). Across countries, the gender gap for women’s rights was signiﬁcantly larger in high-GDI countries, due to greater support among females in these countries (g301). Within countries, experience of an open classroom climate, where respect is shown for everyone’s opinions, was associated with smaller gender gaps. This was due to higher levels of support from male students (b31). In other words, females across countries supported women’s rights regardless of the climate for discussion of issues in the classroom, whereas males who experienced the open discussion of diverse opinions (including those of their female peers) had greater support for gender equality. At the same time, the gender gap in support for women’s rights overall was larger in schools where the expectations for further education were stronger (analysis of b32). Local expectations for further education, a positive predictor of support for women’s overall, was associated with signiﬁcantly higher levels of support especially among females. In other words, female students who are in schools, where many students expect to continue their education, more strongly support equal rights for women than do females who do not experience these expectations in their schools. Male students do not similarly adjust their support for women’s rights in the presence of strong local support for education in their schools. Analysis conducted at the country level, however, suggests that this effect was not uniform across all countries. A cross-level interaction term (g321) indicates that being in a high-GDI country negated the effect of local expectations of further education on the size of the gender gap. In other words, females in high-GDI countries had especially high support for women’s rights regardless of educational expectations in their local schools. In relatively lower-GDI countries (such as the post-communist and Latin American countries) these expectations of education at the school level appeared to inﬂuence young women’s belief in the importance of political and economic rights for women. Overall, only 16% of the variance in school-level gender gaps in support for women’s rights is explained by the variables considered here. However,

Analysis from the IEA Civic Education Study

381

consideration of the GDI and its interaction with school-level expectations of further education, along with the rest of the multilevel model, explained 63% of the variability in country-level gender differences. However, there is still a signiﬁcant amount of variance left to be explained.

Modeling the Gender Gap in Internal Political Efﬁcacy We report a summary of HLM analysis for internal political efﬁcacy in Table 5. Civic knowledge and home literary resources were positive and signiﬁcant covariates of internal political efﬁcacy at the student level (p01 and p02, respectively), whereas average classroom climate and the proportions of students perceiving educational and employment inequality in the school were signiﬁcant and positive at the school level (b02, b04, and b05). In other words, the average internal political efﬁcacy in a school was higher when students on average perceived their classes to be more open to discussion, and when greater numbers of students perceived gender inequality in the education system. At the same time, internal political efﬁcacy was lower when the proportion of girls in the school was higher (b01). Average report of expectations for further education among students in the school was not signiﬁcantly related to internal political efﬁcacy (b03). We were able to partially explain the observed gender gap in internal political efﬁcacy (analysis of p3) using several variables at the school and student levels. Within schools, the gender gap was larger in environments in which more employment inequality (but not educational inequality) was perceived (b31). Female students had similar levels of internal political efﬁcacy regardless of school context; however, male students had greater political efﬁcacy in contexts with more widespread perceptions of employment inequality. Finally, at the country level, internal political efﬁcacy was lower in countries with higher GDI scores (g002). There was no signiﬁcant main effect of national expectations of further education in internal political efﬁcacy (g001); however, we retained this variable because of a signiﬁcant interaction between GDI and average further education. This interaction term (g003), which is statistically signiﬁcant, suggests that students in highGDI countries had higher levels of efﬁcacy if they lived in countries that also had strong countrywide expectations for further education. The GDI score and national expectations of further education, along with their interaction, also inﬂuenced the magnitude of the gender gap in internal political efﬁcacy (g301, g302, and g303, respectively). The gender gap in efﬁcacy was larger in

382

CAROLYN BARBER AND JUDITH TORNEY-PURTA

Table 5.

Summary of Hierarchical Linear Model for Internal Political Efﬁcacy.

Fixed Effects

Coefﬁcient

Standard Error

Average internal political efﬁcacy score (g000) National expectations for further education (g001) Standardized GDI score (g002) National further education GDI interaction (g003) Proportion girls (b01) Average class climate (b02) Average further education in school (b03) Percent perceiving school inequality (b04) Percent perceiving employment inequality (b05) Total civic knowledge (p1) Number of books in the home (p2)

10.02 0.07 0.30** 0.13* 0.46* 0.13** 0.02 1.43** 0.24* 0.01** 0.10**

0.07 0.05 0.06 0.05 0.19 0.01 0.01 0.18 0.10 0.00 0.01

Gender (p3) Average gender gap (c300) National expectations for further education (g301) Standardized GDI score (g302) National further education GDI interaction (g303) Percent perceiving employment inequality (b31)

0.57** 0.06* 0.13** 0.06* 0.25*

0.03 0.03 0.02 0.02 0.11

Random Effects Error (Level-1) variance (e) Level-2 intercept variance (r0) Level-2 gender variance (r3) Level-3 intercept variance (u00) Level-3 gender variance (u30)

Variance Component (% Reduction) 3.43 0.14** 0.30** 0.12** 0.01**

(2) (18) (3) (48) (86)

Note: Variables in italics are centered on their grand mean; variables in bold are centered on their group mean. All variables except for gender have their error variances ﬁxed to zero. *po.05; **po.01.

countries where female opportunities were more developed. Women in highGDI countries had lower efﬁcacy than did women in low-GDI countries, whereas for men levels of efﬁcacy were more similar across national contexts. At the same time, the gender gap in efﬁcacy was smaller in countries where expectations of postsecondary education were widespread, due to somewhat higher efﬁcacy among females overall. Moreover, the interaction term between GDI score and countrywide expectations of further education indicates that high educational expectations were especially effective in narrowing the gender gap in high-GDI countries.

Analysis from the IEA Civic Education Study

383

In other words, high expectations of further education canceled out the negative effect of living in a high-GDI country on female students’ selfefﬁcacy. According to this model, gender gaps in political efﬁcacy are most pronounced in countries like England, Germany, and Switzerland (which have high-GDI scores but low expectations of education), and least pronounced in Poland and Colombia (which have high expectations of further education despite low-GDI scores). Overall, with these variables, we could only explain 4% of the variance among students, 18% of the variance in average efﬁcacy differences among schools, and 3% of the variance in the magnitude of gender gaps among schools. However, we also explained 48% of the variance of average efﬁcacy levels across countries and 86% of the variance in the magnitude of countrylevel gender gaps.

DISCUSSION Overall, the purpose of this study was to examine predictors of the gender gaps in two important civic-related outcomes: one in which gender differences overwhelmingly favor females (support for women’s rights) and another in which gender differences overwhelmingly favor males (internal political efﬁcacy). The results of these analyses showed that both proximal and distal social contexts shape gender gaps in each of these areas in speciﬁc ways. Although some of the results suggest how schools can inﬂuence gendered civic development similarly in diverse national contexts, others imply that this association varies depending on opportunity structures available in a given country. Further, although these results provide insights into the relationship between education and gender in civic engagement cross-nationally, they also raise questions requiring further multimethod research and exploration.

School-Level Perceptions of Gender Inequality and ‘‘Male Privilege’’ For each of the two outcomes, there was one school-level variable that predicted the magnitude of gender gaps similarly across all countries in the CIVED study. Stronger average perceptions of an open classroom climate for discussion were associated with narrower gender gaps in support for women’s rights due to higher support among male students in schools with open climates. In contrast, being in a school where students believed there

384

CAROLYN BARBER AND JUDITH TORNEY-PURTA

was employment inequality in society was associated with larger gender gaps in internal political efﬁcacy. This was due to higher efﬁcacy among male students in these schools. Looking ﬁrst at the predictor of the gap in internal political efﬁcacy, it may initially seem surprising that the gender gaps in efﬁcacy are larger in schools in which inequality is more widely perceived. Research by Case (2007) among older adolescents the United States suggests that increasing the awareness of ‘‘male privilege’’ in schools is accompanied by a greater support for afﬁrmative action programs and greater identiﬁcation with feminist issues. However, an important distinction between the measure of perception of employment inequality considered here and the study conducted by Case is the extent to which ‘‘male privilege’’ awareness is discussed through the formal curriculum. In the current study, students did not indicate why they believed that employment inequality existed for women, indicating only that they agreed that it did exist. Without a formal curriculum to guide thinking about such issues, male students may view this inequality operating in their favor, increasing their own efﬁcacy for civic participation while having little effect on the efﬁcacy of women. Further research on the inﬂuence of perceived male privilege on gender gaps in political outcomes should consider whether this topic is raised in the context of formal instruction on this topic. The contrast between Case’s ﬁnding and the CIVED analysis also suggests that it is unwise to generalize about the meaning of male privilege or gender inequality and ways to address it across countries with different histories and traditions regarding women’s political participation. Contrast this relationship between perception of employment inequality and gender gaps in internal efﬁcacy to the relation between perception of an open classroom climate and gender gaps in support for women’s rights. The measure of open classroom climate for discussion speciﬁcally prompts students to consider the classes that they take in general, and in their social studies and other civic-related classes in particular (Schulz & Sibberns, 2004). Our analysis was limited to schools in which both male and female students were surveyed. As a result the items in this scale require these students to think about classroom environments in which both male and female students are present and able to participate. In order for a school to have high average openness for discussion rating, this openness must be perceived by both male and female students. This means that the male students in these schools with very open climates are surrounded by female students who are willing to speak their mind on controversial issues. This form of intergroup contact among males and females seems to raise male

Analysis from the IEA Civic Education Study

385

students’ support for women’s rights. Further and smaller-scale research should focus on the gendered dynamics of political discussion in typical classroom settings in more nuanced ways, providing more insight into this phenomenon (Hahn, 1998). Together, these two ﬁndings have important implications for how schools in varied national contexts can increase gender equality in attitudes. First, interpretation of these ﬁndings in light of previous research suggests that explicit instruction in social justice issues and concerning ‘‘male privilege’’ is important for reducing gender inequalities. Adolescents may be aware of inequalities in the absence of formal instruction, but may not think that such issues are of particular importance. In fact, male students may actually view such inequalities as beneﬁcial to them and female students may become resigned to living in unequal social and political settings. Second, explicit instruction must be accompanied by a climate in which all students feel able to voice their own opinions on political and social topics. In general, discussion in the family has been shown in previous studies to relate positively to the development of civic engagement among females (Campbell & Wolbrecht, 2006; Wolbrecht & Campbell, 2007). This study has demonstrated that open discussion in the school context can have a positive inﬂuence on male attitudes toward females’ participation by exposing them to the opinions of their female peers.

Expectations of Education and Female Opportunities Other components of the models of support for women’s rights and internal political efﬁcacy suggest that effective ways to reduce gender gaps may differ across countries. A common ﬁnding in each of the models is a signiﬁcant interaction between expectations of further education and the quality of life for females (as measured by GDI) in a country. The exact nature of this interaction differed somewhat between these two outcomes; the interaction occurred with local expectations of further education in support for women’s rights and with national expectations of further education in internal political efﬁcacy. However, it existed for both outcomes. This interaction between GDI and support for women’s rights suggests that when young women live in countries with few economic opportunities (low GDI), attending schools where there are strong expectation for further education increases their support of political and economic rights for themselves. Remember that support for women’s rights is lower overall in these low-GDI countries. Although our purpose in this study was to identify

386

CAROLYN BARBER AND JUDITH TORNEY-PURTA

characteristics of school and country that narrowed the gender gap, perhaps the more pressing issue in these countries is how to mobilize young women to believe in their own rights, which may be followed by action to claim those rights. These countries in the CIVED are largely the post-communist countries, which have a complex history in this area of women’s rights (Malak-Minkiewicz, 2007). These ﬁndings also support the views of Stromquist (2006) and others, who emphasize further education as a step on the path to active citizenship among women. In countries where societywide opportunities and expectations may not be very high for women, school-wide beliefs that further education is necessary, important, and attainable can serve to mobilize girls’ support for the rights of all women, and by extension, for themselves. Further multi-method research within these low-GDI countries looking at the relationship of support for women’s rights to eventual civic engagement among females should be high on the list of further research (as should studies of the validity of this inference for countries in regions of the world where CIVED did not test, for example, Africa). At the same time, another line of research should consider other ways to increase support for women’s rights among males in these low-GDI countries, so that the next generation of citizens values social justice issues more equally regardless of their own gender. In contrast, in examining this interaction as it relates to internal political efﬁcacy, these data revealed that female adolescents’ levels of self-efﬁcacy were the lowest (and the gender gap in efﬁcacy widest) when opportunities for women in society were not accompanied by high expectations for education. In these countries, there may be relatively more opportunities for women in comparison to lower-GDI countries, but not similarly high expectations of further education that would give women the skills necessary to feel conﬁdent in fully participating in civic life. As was the case when analyzing support for women’s rights the ultimate message is still that education is a necessary path to fully participating in the civic and political life of a nation. Why would these interactions between further education expectations and GDI as they relate to civic-related outcomes not be seen to a similar extent among adolescent males? This question is worth further exploration. Some of the early theories about gender differences in political socialization provide interesting perspectives but no deﬁnitive answers. Perhaps males are expected to participate in politics regardless of their educational background (as was suggested by Paulsen, 1991). For females to become equally engaged and equally conﬁdent in their own abilities, they need greater levels of social support through educational structures. Encouraging female students in

Analysis from the IEA Civic Education Study

387

particular to aspire to higher levels of educational attainment then becomes a key way to foster gender equality in the civic and political arena.

Limitations and Further Areas of Research We believe that this multilevel analysis has furthered understanding of gender gaps in these two important political attitudes. However, there are several limitations to the current study that should be addressed in further research. For example, our decision to model gender gaps at the school level meant removing single-gender schools from the analysis. Therefore, the results cannot be generalized to students attending single-gender schools. The ways in which these unique school contexts foster the political and civic engagement of male and female students is worthy of further study. Further, the large majority of our 28 countries are located in Europe. The ﬁndings therefore cannot be generalized to entire regions of the world from which data were not collected, many of which continue to have some of the most traditional gender dynamics (Inglehart & Norris, 2000, 2003). The examination of local contexts, which foster more equitable civic development among male and female students, may be especially interesting in these countries. One additional limitation in interpreting these results is that the CIVED data were collected in 1999. Similar analyses should be conducted with more recently collected data in order to determine the extent to which sociohistorical changes account for the differences between high-GDI and low-GDI countries. One of the major focuses of the 1999 CIVED study was the examination of post-communist countries, which are some of the lowestGDI countries in the study and had recently transitioned to democratic systems of government (Torney-Purta et al., 1999, 2001). Now that these democracies have been in existence for over 15 years, researchers should revisit these issues to see what has changed. The International Civics and Citizenship Study (ICCS), to be conducted in 2009–2010 by IEA, may be useful for this purpose, at least for those measures that are repeated. Finally, our analysis of school and student characteristics in this analysis is limited to the variables collected by the IEA CIVED study. Although this study examined many dimensions of adolescents’ civic knowledge and engagement (indeed, many more than earlier studies), there are still some potential explanations for these gender gaps that were missing. One potentially important variable not addressed by the CIVED study is the relation of gender identity, rather than gender itself, in political engagement.

388

CAROLYN BARBER AND JUDITH TORNEY-PURTA

Although an adolescent may be female, she may not necessarily feel that she identiﬁes strongly with the roles and attitudes that are expected of a woman (Barber, 2008). Therefore, differences in gender identity and sex-role orientation might be an important psychological component of gender differences in political engagement. Sex-role orientation may be especially useful in further understanding gender gaps in internal political efﬁcacy. Sex-role orientation has been shown to mediate gender gaps in writing selfefﬁcacy (e.g., Pajares & Valiante, 2001); further research could look to determine whether it has a similar effect for self-efﬁcacy in political discussions.

CONCLUSION Overall, this study has demonstrated the extent to which contexts of education, both within schools and at the national level, inﬂuence gender differences in two civic-related attitudes. Overall, the gender gap is smaller when students attend schools in which women’s political participation is valued and where associated issues are openly discussed, and when educational expectations are sufﬁciently high to meet the demands of civic participation. At the same time, this analysis suggests that more needs to be done to encourage the consideration of social justice issues among male and female students alike, especially in countries where opportunities for women are more limited. Although students can work within the educational structures currently in place in order to become engaged citizens, we must continue to strive to improve educational experiences and opportunities to meet young people’s developmental needs, and to tailor these institutions to their particular national contexts.

NOTES 1. Just because two measures are called ‘‘civic or political knowledge’’ does not necessarily mean they cover the same content areas or mean the same thing. One important difference between studies of adults (where differences favoring males are often found) and those of adolescents (where they are less frequently found) is the type of question asked. Adults are generally questioned about facts related to electoral politics, party ideology, and current events, whereas adolescents generally are asked to answer more conceptually oriented questions, often related to the social studies or history topics that they study in school.

Analysis from the IEA Civic Education Study

389

2. The CIVED study also surveyed upper-secondary students (i.e., students in their last year of secondary education) in 16 countries (Amadeo, Torney-Purta, Lehmann, Husfeldt, & Nikolova, 2002). However, this analysis focuses only on results from the survey of 14-year-olds. 3. We examined GDI scores from 2000 because we are analyzing data collected in late 1999. It is important to remember, however, that GDI scores (and subsequent country rankings) change from year to year, and may not reﬂect current levels of gender (in)equality in a country.

ACKNOWLEDGMENT The authors would like to thank Wendy Klandl Richardson and Conrad Mueller for their assistance with this manuscript.

REFERENCES Alozie, N. O., Simon, J., & Merrill, B. D. (2003). Gender and political orientation in childhood. The Social Science Journal, 40, 1–18. Amadeo, J.-A., Torney-Purta, J., Lehmann, R., Husfeldt, V., & Nikolova, R. (2002). Civic knowledge and engagement: An IEA study of upper secondary students in sixteen countries. Amsterdam: IEA. Armingeon, K. (2007). Political participation and associational involvement. In: J. W. van Deth, J. Montero & A. Westholm (Eds), Citizenship and involvement in European Democracies (pp. 358–383). London: Routledge. Atkeson, L. R., & Rapoport, R. (2003). The more things change the more they stay the same: Examining gender differences in political attitude expression, 1952–2000. Public Opinion Quarterly, 67, 4495–4521. Bandura, A. (1997). Self-efﬁcacy: The exercise of control. New York: W.H. Freeman and Company. Barber, C. (2008). Gender identity. In: E. Anderman & L. Anderman (Eds), The psychology of classroom learning. Detroit: MacMillan Press. Barber, C., Torney-Purta, J., Homana, G., & Wilkenfeld, B. (2007). Communities of practice that foster the development of political efﬁcacy in three Nordic countries. Paper presented at the American Educational Research Association annual conference, Chicago, IL. Barnes, S. H., & Kaase, M. (1979). Political action: Mass participation in ﬁve western democracies. Beverly Hills, CA: Sage. Campbell, D. E., & Wolbrecht, C. (2006). See Jane run: Women politicians as role models for adolescents. The Journal of Politics, 68(2), 233–247. Case, K. A. (2007). Raising male privilege awareness and reducing sexism: An evaluation of diversity courses. Psychology of Women Quarterly, 31, 426–435.

390

CAROLYN BARBER AND JUDITH TORNEY-PURTA

Conover, P. J. (1988). The role of social groups in political thinking. British Journal of Political Science, 18(1), 51–76. Delli Carpini, M. X., & Keeter, S. (2000). Gender and political knowledge. In: S. TollesonRinehart & J. J. Josephson (Eds), Gender and American politics: Women, men, and the political process (pp. 74–90). Armonk: M. E. Sharpe. Fernandez-Ballesteros, R., Diez-Nicolas, J., Caprara, G., Barbaranelli, C., & Bandura, A. (2002). Determinants and structural relation of personal efﬁcacy to collective efﬁcacy. Applied Psychology: An International Review, 51, 107–125. Flanagan, C. A., Bowes, J. M., Jonsson, B., Csapo, B., & Sheblanova, E. (1998). Ties that bind: Correlates of adolescents’ civic commitments in seven countries. Journal of Social Issues, 54(3), 457–475. Franklin, J. (2006, June 7). Protests paralyse Chile’s education system. The Guardian, p. 14. Gro¨nlund, K., & Milner, H. (2006). The determinants of political knowledge in comparative perspective. Scandinavian Political Studies, 29(4), 386–406. Hahn, C. (1998). Becoming political: Comparative perspectives on civic education. Albany: SUNY Press. Harvard University Institute of Politics. (2008). Executive summary: The 14th Biannual Youth Survey on Politics and Public Service. Available at http://www.iop.harvard.edu/ Research-Publications/Polling/Spring-2008-Survey/Executive-Summary. Retrieved on September 3, 2008. Hess, R. D., & Torney, J. V. (2005). The development of political attitudes in children. New Brunswick, NJ: Transaction Press. (Original work published in 1967). Hooghe, M., & Stolle, D. (2004). Good girls go to the polling booth, bad boys go everywhere: Gender differences in anticipated political participation among American fourteen-yearolds. Women and Politics, 26(3/4), 1–23. Husfeldt, V., Barber, C., & Torney-Purta, J. (2005). Students’ social attitudes and expected political participation: New scales in the enhanced database of the IEA Civic Education Study. (CEDARS Report). College Park, MD: University of Maryland Department of Human Development. Inglehart, R., & Norris, P. (2000). The developmental theory of the gender gap: Women’s and men’s voting behavior in global perspective. International Political Science Review, 21(4), 441–463. Inglehart, R., & Norris, P. (2003). Rising tide: Gender equality and cultural change around the world. New York: Cambridge University Press. Jasso, G., & Wegener, B. (1999). Gender and country differences in the sense of justice: Justice evaluation, gender earnings gap, and earnings functions in thirteen countries. International Journal of Comparative Sociology, 40, 94–115. Jennings, M. K. (2006). The gender gap in attitudes and beliefs about the place of women in American political life: A longitudinal, cross-generational analysis. Politics and Gender, 2, 193–219. Jennings, M. K. (2007). Political socialization. In: R. J. Dalton & H.-D. Klingemann (Eds), The Oxford handbook of political behavior (pp. 29–44). New York: Oxford University Press. Kane, E. (2000). Racial and ethnic variations in gender-related attitudes. Annual Review of Sociology, 26, 419–439. Kennedy, K. (2006). The gendered nature of students’ attitudes toward minority groups: Implications for teacher education. Citizenship Teaching and Learning, 2(1), 55–64.

Analysis from the IEA Civic Education Study

391

Lutkus, A. D., Weiss, A. R., Campbell, J. R., Mazzeo, J., & Lazer, S. (1999). NAEP 1998 Civics report card for the nation. Washington, DC: National Center for Education Statistics. Malak-Minkiewicz, B. (2007). Civic education in times of change: The post-Communist countries. Citizenship Teaching and Learning, 3(2), 58–70. Mayer, J. D., & Schmidt, H. M. (2004). Gendered political socialization in four contexts: Political interest and values among junior high school students in China, Japan, Mexico, and the United States. The Social Science Journal, 41, 393–407. Mullis, I. V. S., Martin, M. O., Gonzalez, E. J., & Chrostowski, S. J. (2004). TIMSS 2003 international mathematics report: Findings from IEA’s Trends in International Mathematics and Science Study at fourth and eighth grades. Chestnut Hill, MA: TIMSS and PIRLS International Study Center, Lynch School of Education, Boston College. Niemi, R., & Junn, J. (1998). Civic education: What makes students learn. New Haven: Yale University Press. Norris, P. (2007). New feminist challenges to the study of political engagement. In: R. J. Dalton & H.-D. Klingemann (Eds), The Oxford handbook of political behavior (pp. 724–743). New York: Oxford University Press. Pajares, F., & Valiante, G. (2001). Gender differences in writing motivation and achievement of middle school students: A function of gender orientation? Contemporary Educational Psychology, 26, 366–381. Paulsen, R. (1991). Education, social class, and participation in collective action. Sociology of Education, 64(2), 96–110. Raudenbush, S., Bryk, A., Cheong, Y., & Congdon, R. (2004). HLM 6: Hierarchical linear and nonlinear models. Lincolnwood, IL: Scientiﬁc Software International. Rettinger, V. (1993). The moral and political consciousness of preadolescents. Moral Education Forum, 18(4), 2–35. Rosenthal, C., Jones, J., & Rosenthal, J. (2003). Gendered discourse in the political behavior of adolescents. Political Research Quarterly, 56(1), 97–104. Rosenthal, S., Feiring, C., & Lewis, M. (1998). Political volunteering from late adolescence to young adulthood: Patterns and predictors. Journal of Social Issues, 54(3), 477–493. Sapiro, V., & Conover, P. (2001). Gender equality in the public mind. Women and Politics, 22(1), 1–35. Schulz, W., & Sibberns, H. (2004). IEA Civic Education Study technical report. Amsterdam: IEA. Steiner-Khamsi, G., Torney-Purta, J., & Schwille, J. (2002). New paradigms and recurring paradoxes in education for citizenship. New York: Elsevier Press. Stromquist, N. P. (2006). Women’s rights to adult education as a means to citizenship. International Journal of Educational Development, 26, 140–152. Tesch-Ro¨mer, C., Motel-Klingebiel, A., & Tomasik, M. J. (2008). Gender differences in subjective well-being: Comparing societies with respect to gender equality. Social Indicators Research, 85, 329–349. Torney, J. V., Oppenheim, A. N., & Farnen, R. F. (1975). Civic education in ten countries: An empirical study. New York: Wiley. Torney-Purta, J. (2002). The school’s role in developing civic engagement: A study of adolescents in twenty-eight countries. Applied Developmental Science, 6(4), 203–212. Torney-Purta, J., Barber, C., & Wilkenfeld, B. (2007). Latino adolescents’ civic development in the United States: Research results from the IEA Civic Education Study. Journal of Youth and Adolescence, 36(3), 111–126.

392

CAROLYN BARBER AND JUDITH TORNEY-PURTA

Torney-Purta, J., Lehmann, R., Oswald, H., & Schulz, W. (2001). Citizenship and education in twenty-eight countries: Civic knowledge and engagement at age fourteen. Amsterdam: IEA. Torney-Purta, J., Schwille, J., & Amadeo, J. (1999). Civic education across countries: Twentyfour national case studies from the IEA Civic Education Project. Amsterdam: IEA. United Nations Development Programme. (2000). Human development report. New York: Oxford University Press. Verba, S., Burns, N., & Schlozman, K. L. (1997). Knowing and caring about politics: Gender and political engagement. Journal of Politics, 59, 1051–1072. Wolbrecht, C., & Campbell, D. E. (2007). Leading by example: Female members of parliament as political role models. American Journal of Political Science, 51(4), 921–939. Zucker, A., & Stewart, A. (2007). Growing up and growing older: Feminism as a context for women’s lives. Psychology of Women Quarterly, 31, 137–145. Zukin, C., Keeter, S., Andolina, M., Jenkins, K., & Delli Carpini, M. X. (2006). A new engagement? Political participation, civic life, and the changing American citizen. Oxford: Oxford University Press.

APPENDIX. DETAILS ABOUT THE MULTILEVEL MODEL In this multilevel model, a student’s attitude can be modeled as shown in Eq. (A1) Y ¼ p0 þ p1 ðTOTCGMLEÞ þ p2 ðHOMELITÞ þ p3 ðGENDERÞ þ e (A1) where the p parameters are coefﬁcients for each term in the equations and e is the random error among students. In other words, the attitude, be it support for women’s rights or internal political efﬁcacy, consists of some average value (i.e., the intercept) plus an amount associated with a student’s civic knowledge, an amount associated with level of home literacy resources, an amount associated with gender, plus individual error. Because civic knowledge and home literacy resources are considered only as statistical controls, we centered these variables on their grand mean in this analysis (indicated by the italicized variable names). The average value of an attitude, however, differs depending on the speciﬁc country and school in which the students go to school. Thus, this intercept can be further broken down as indicated in Eq. (A2) p0 ¼ g000 þ

p X 1

g00p ðX p Þ þ

n X

b0n ðX n Þ þ u00 þ r0

(A2)

1

In this model, the average attitude, or intercept, is a function of an overall average attitude level across students in all schools in all countries (g000),

393

Analysis from the IEA Civic Education Study

plus some amount (g) associated with each of p country-level predictors, plus some amount (b) associated with each of n school-level predictors, plus error. In Eq. (A2), the error associated with an individual attitude can come from one of two sources: country-level variability (u00) and school-level variability (r0). As previously described, country-level variables considered for these models included GDI scores and average expectations of further education among students in the country. School-level variables considered for these variables include the proportion of girls in the school, the average perception of an open classroom climate for discussion, the average expectations of further education among students in a school, and the proportions of students perceiving gender inequality in schools and in the workplace. Still more central to our current analysis, however, is how the size of the gender coefﬁcient (i.e., the gender gap) varies across schools and countries. Similar to the intercept, the size of the gender gap can be further considered as indicated in Eq. (A3) p3 ¼ g300 þ

p X 1

g30p ðX p Þ þ

n X

b3n ðX n Þ þ u30 þ r3

(A3)

1

The interpretation of Eq. (A3) is similar to the interpretation of the equation for the intercept. In each case, the goal of selecting country- and school-level variables is to reduce the amount of random error at each level of these two levels (i.e., to reduce the variance associated with the u and the r terms in Eqs. (A2) and (A3)). In other words, we are attempting to use characteristics of schools and countries, after controlling for individual students’ civic knowledge levels and home literacy resources, to explain differences not only in average attitude levels, but also in the sizes of gender gaps observed both between countries and among schools within countries. Finally, it is possible that certain school-level predictors of intercepts or gender gaps may be stronger or weaker in certain countries. HLM allows for the statistical analysis of the variability of the school-level coefﬁcients, as well as the student coefﬁcients. If signiﬁcant variation in a coefﬁcient exists, it can be modeled using country-level variables, as indicated in Eq. (A4) bn ¼ g3n0 þ g3np ðX p Þ

(A4)

where n refers to the number of the school-level predictor that needs additional modeling, p to the country-level predictor that is considered in

394

CAROLYN BARBER AND JUDITH TORNEY-PURTA

that role, and 3 to gender, which is the third student-level predictor considered. These cross-level interactions add additional complexity to these already-complex models; therefore, they will only be considered if their inclusion helps to further explain differences in gender gaps. In other words, cross-level interactions between school and country variables were not included if they only modeled the intercept.

SHIFTING GENDER EFFECTS: OPPORTUNITY STRUCTURES, INSTITUTIONALIZED MASS SCHOOLING, AND CROSS-NATIONAL ACHIEVEMENT IN MATHEMATICS Alexander W. Wiseman, David P. Baker, Catherine Riegle-Crumb and Francisco O. Ramirez ABSTRACT Prior research shows that stratiﬁcation of future adult opportunities inﬂuences stratiﬁcation in the academic performance of students. This perspective is used to generate hypotheses regarding the sources of crossnational gender differences in mathematics performance. These hypotheses are tested using multivariate and multilevel analyses of adult opportunities for women and cross-national differences in mathematics performance by gender. This future opportunity perspective is expanded to take into account the historical incorporation of women in modern nation-states through institutionalized mass schooling emphasizing egalitarian ideals. Results indicate a cross-national shift in the direction Gender, Equality and Education from International and Comparative Perspectives International Perspectives on Education and Society, Volume 10, 395–422 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1479-3679/doi:10.1108/S1479-3679(2009)0000010015

395

396

ALEXANDER W. WISEMAN ET AL.

of less gender inequality in overall school mathematics performance. However, gender inequality is more evident in the advanced 12th grade mathematics. The results of a more specialized analysis of the advanced 12th grade mathematics are compared with the earlier ﬁndings regarding mathematics performance.

The inﬂuence of gender on a wide range of education-related outcomes has become a major topic in theory and research in comparative social stratiﬁcation (Charles & Bradley, 2002; Charles & Grusky, 2004; Jacobs, 1996; Ridgeway & Correll, 2004). An older yet persistent theme in the stratiﬁcation literature is that future opportunity structures are consequential in shaping present educational expectations and behaviors (Stinchcombe, 1964). Empirical evidence in the studies of stratiﬁcation support this reasoning, but many of these studies are limited to locally contextualized situations and ignore the importance of schooling as an institution to this process. Relatively few studies have cross-nationally considered how the opportunity structure–educational outcome argument ﬁts within the global phenomena of the incorporation of women into modern nation-states and institutionalized mass schooling. Our research shows how these shifting global effects of gender impact the relationship between opportunity structures and mathematics achievement, speciﬁcally. The ﬁrst part of this chapter reviews this literature, paying special attention to studies that emphasize the inﬂuence of gender-stratiﬁed future opportunities on important educational outcomes leading to future occupations. Next, we turn to the literature on the incorporation of women and the role of institutionalized mass schooling in promoting this process of incorporation. These literatures offer different logics or accounts of gender. The opportunity structure perspective focuses on cross-national variability in gender outcomes due to cross-national differences while the incorporation argument stresses commonalities in gender-related trends, reﬂecting the inﬂuence of common transnational norms or standards. The logic of both types of arguments are empirically examined in the theoretically challenging context of historical declines in gender effects on academic outcomes overall, yet with persistent gender differences remaining in many nations. The third part of the chapter describes an empirical analysis incorporating these ideas about the changing world of gender and education and what causes persistent cross-national gender differences in academic performance. Lastly, the chapter reﬂects on

Shifting Gender Effects

397

the overall results within a broader perspective that deals with both stratiﬁcation and institutional perspectives on gender inequality.

1. OPPORTUNITY STRUCTURE AND GENDER EFFECTS The structure of opportunities perspective has been used to examine the causes of cross-national gender differences in academic performance in previous studies (Baker & Jones, 1993; Hanson, Schaub, & Baker, 1996; Riegle-Crumb, 2000). The basic argument is that if students and their parents perceive that important future educational and occupational opportunities are linked to current academic performance, their attempts to improve their performance will generally intensify. If, however, the future opportunity is stratiﬁed in such a way as to not present clear opportunities to certain categories of students, then these students and their families may not invest as much in current performances. For example, if males are afforded the possibility of future educational and occupational opportunities as a function of their mathematical performance in school, they may try harder, teachers may more strongly encourage them, and parents and peers may more consistently support their commitment to the study of mathematics. Conversely, to the degree that future educational and occupational opportunities are stratiﬁed by gender, favoring males, in this case, female students faced with the prospects of lesser opportunities tied to mathematical skill and achievement may tend to perceive these subjects as less important for their future. Moreover, their teachers, parents, and friends are likely to reinforce these behaviors associated with a sense of weak links between current mathematics performance and future adult status. This simple yet compelling line of reasoning has received substantial empirical support in studies of stratiﬁcation by race and socio-economic status within more speciﬁc contexts (Bourdieu, 1977; Kerckhoff, 1977; McClelland, 1990; Mickelson, 1990; Ogbu, 1979). Moreover, there is a large body of literature on gender-related socialization that argues that the gender-related experiences of adult women in their families, jobs, and societies clearly inﬂuence the experiences of younger generations (Eccles, 1986; Gerris, Dekovic´, & Jansens, 1997; Marini & Brinton, 1984). This perspective has been applied to make sense of cross-national variation in gender differences in mathematics performance. Baker and Jones (1993) show that cross-national variations in gender stratiﬁcation in higher

398

ALEXANDER W. WISEMAN ET AL.

education and the labor market correlate with cross-national differences in 8th grade mathematics performance in the early 1980s. There are also parallel associations between future opportunity structures and crossnational gender differences in parental encouragement to learn mathematics (Baker & Jones, 1993; Baker & LeTendre, 2005; Jones, 1989). In either human capital formation or social credentialing terms (Collins, 1979) student engagement in schooling is greater if school achievement is perceived to be connected to future desirable outcomes. And, the perceptions of female students are likely to be more positive when in fact women’s share of future educational and occupational opportunities is greater. Given these prior ﬁndings, girls’ achievement in mathematics may be especially vulnerable to future role expectations for several reasons. First of all, societal messages concerning the appropriate roles for and competencies of women are likely to reinforce academic stereotypes of mathematics as a masculine domain. This in turn may discourage women and lead them to direct their abilities and efforts away from mastering mathematics. Much research guided by expectations theory emphasizes the interaction dynamics that would lead to this ﬁrst point (Ridgeway, 1997). Additionally, performance in school mathematics serves as a prerequisite to many future educational paths towards careers in modern labor sectors that are increasingly based on science and technology (Stage & Maple, 1996). If girls expect limited opportunities in such a labor force, this will likely lead to lower investment in the ‘‘gatekeeper’’ academic subject of mathematics (Stevenson, Schiller, & Schneider, 1994). Finally, when opportunities for women’s careers are related to domestic and traditionally feminine careers, female students may see little reason to work hard in challenging and competitive mathematics. The opportunity structure argument is important in and of itself. However, as it stands, the argument more or less leaves out an active role for schooling as an institution that shapes gender equity policies and practices crossnationally. Heretofore the imagery has been one of a passive school context, which provides the setting for the connection of future opportunities and current performance. Similarly, the argument has not been expanded to connect opportunity structures to larger historical trends in social stratiﬁcation. The contours of gender stratiﬁcation of opportunity structures do not remain stationary; rather, they change over time, often globally, and mostly in the direction of greater egalitarian standards (Ramirez & Wotipka, 2001). A useful extension of the opportunity structure argument is to consider how it ﬁts within the longer term and larger scale incorporation of women into

Shifting Gender Effects

399

modern nation-states and how institutionalized mass schooling has facilitated this process.

2. INCORPORATION OF WOMEN AND THE ROLE OF MASS SCHOOLING In the midst of centuries of stratiﬁcation and inequality of the life roles of females and males there has emerged a countertrend in the ways modern nation-states have developed citizenship roles and their impact on gender stratiﬁcation (Arie`s & Be´jin, 1985; Chang, 2000). These changes have been conceptualized in world polity theory as outcomes of long-term processes that have resulted in nation-states embracing similar human capital and gender equality premises to enact their appropriate nation-state identity (Meyer, Boli, Thomas, & Ramirez, 1997). Human capital imagery evokes the potential economic contributions that women can make to national economies if women are adequately trained to enter into productive jobs and if allocation to these jobs is based on achievement instead of ascriptive criteria, such as gender (Chang, 2000). The underlying idea is that the incorporation of women as productive citizens is in the national economic interest and that every rational nation-state should foster the human capital development of its girls and women (Schultz, 1982). A second underlying idea directly stresses egalitarian ideologies that are built on the establishment of progress and justice as national goals (Fiala & Gordon-Langford, 1987), the political incorporation of women as voting citizens (Ramirez, Soysal, & Shanahan, 1997), and the rise of women’s issues in national and international organizations and agendas (Berkovitch, 1999; Huber & Stephens, 2000). If the human capital imagery links women’s incorporation as productive citizens to rational economic national interest, the egalitarian emphasis links women’s incorporation as participating citizens to respectable political national identity. Both of these emphases and the transformations they imply are clearest in the state-sponsored mass schooling systems now virtually common to all countries (Fuller & Rubinson, 1992). Modern schooling is expected to play an active role in both the creation of universal criteria for human capital development and the promotion of gender equality as a matter of social justice. This expectation surfaces in both national and international policy with an increase in data gathering and assessment to identify and implicitly chastise national laggards with respect to gender equality. This is an

400

ALEXANDER W. WISEMAN ET AL.

especially important undertaking in those domains more directly linked to economic productivity and national economic growth, domains such as the technical and scientiﬁc ﬁelds of endeavor and the educational paths that lead to them. The inﬂuence of multilateral agencies in promoting educational opportunity for girls and women is of particular note in this regard. Along with UNDP, UNICEF, and UNESCO, the World Bank co-sponsored the important Jomtien Conference of 1990 where representatives from a large number of nations met to discuss plans for education within the developing world. Foremost among all the issues was education for females. The declaration stemming from the conference, Education for All (EFA or the Jomtien Declaration, 1990), stated the following about the importance of education for girls and women in the worldwide effort to provide basic education for all: ‘‘the most urgent policy is to ensure access to, and improve the quality of, education for girls and women.’’ This international attention to the education of females comes precisely at a time during which the female to male enrollment ratio has increased at the primary, second, and tertiary levels throughout much of the world. The broad goal of educating females is more or less widely accepted throughout the world, with only a few notable national exceptions. Now the debate has shifted to monitoring the ﬁner aspects of educational gender parity. An example of this is the report on gender mainstreaming in the science policies of the European Union (ETAN Report on Women and Science, 2000) in which cross-national differences in gender inequalities throughout the life course are identiﬁed and assessed. Throughout this report the rationale for eradicating these differences involved the goals of progress and justice we have just discussed. It is punctuated with key phrases and ideals about human capital, efﬁciency, economic growth, and traditional obstacles to women’s access to and participation in valued curricula, schools, advanced training, jobs, and career ladders in the name of sensible human resource management. Graphic and tabular displays throughout this report make clear which countries are both underusing human resources and failing to comply with women’s human rights standards. In the education sector, more sophistication in sources of international data and a policy shift from an emphasis on quantity to one on quality has led to an explosion of data on cross-national comparison in educational process and its achievements. This movement has culminated with the IEA’s Trends in International Mathematics and Science Studies (TIMSS), which we analyze later, and the OECD’s Programme for International Student Assessment (PISA). Both TIMSS and PISA are cross-national investigations

Shifting Gender Effects

401

unparalleled in scope, visibility, cost, and both with plans for more such data collection well into the future. Not surprisingly, given what we described earlier, a key policy issue to emerge from all of this cross-national data on school quality and educational outcomes is gender and differences in academic outcomes, particularly (Baker & LeTendre, 2005; Wiseman & Baker, 2007).

3. EMPIRICAL CASE OF GENDER AND MATHEMATICS 3.1. Empirical Context of Gender Differences in Mathematics Cross-Nationally The third part of this chapter investigates the degree to which this crossnational variation in opportunity structure is associated with gender differences in academic outcomes. Even though there are powerful international movements working towards gender equality in education, the empirical situation across nations is complicated – there is evidence of historical decline in inequality yet with persistent inequality remaining cross-nationally. This picture is similar to a number of other sectors that have recently opened up to more gender equity worldwide. A related example is the case of an historical trend towards gender equality in enrollments in higher education, yet with persistent cross-national variation in gender inequality in speciﬁc subjects of earned degrees (Charles & Bradley, 2002). Over the past four decades there has been a decline in gender differences in mathematics achievement. Table 1 shows the national gender differences on the 8th grade mathematics tests in every IEA mathematics study conducted between 1964 and 2003.1 Overall, the percentage of countries with signiﬁcant gender differences in mathematics achievement dropped dramatically between 1964 and 1999 from 27% to 11%, respectively. However, within this trend of declining gender differences there are still persistent cross-national differences in the degree to which gender associates with achievement. As shown in Table 1, the exceptions to this overall decline in statistically signiﬁcant gender differences are the Second International Mathematics Study (SIMS) in 1982 and the TIMSS in 2003 when the percentage of countries with signiﬁcant gender differences in mathematics achievement was 58% and 39%, respectively. But, these two testing years are years in

FIMS 1964a Superior performance of girls

SIMS 1982b 0%

67%

Equal performance

USA Sweden Australia

2 8 14

Sweden USA Japan

France Finland England

16 18 19

Scotland Canada Nigeria England Hong Kong

21%

Superior performance of girls

2.07* 1.63* 1.26* 1.10*

None

42%

0.48 0.14 0.04 0.15 0.28 0.45 0.46 0.50

Equal performance

TIMSS-R 1999d 0%

Superior performance of girls

TIMSS 2003e 0%

None

79%

Equal performance

Thailand Australia Belgium (Flemish)

9.0 5.0 4.0

Philippines Jordan New Zealand

Canada Cyprus Singapore Lithuania Russian Federation Hungary Colombia Iceland Sweden Germany Romania England Latvia

4.0 3.0 2.0 1.0 1.0 0.0 2.0 2.0 2.0 3.0 3.0 4.0 4.0

Malaysia Romania Belgium (Flemish) Cyprus Thailand Hong Kong Bulgaria Macedonia Russian Federation Slovenia Australia Singapore Turkey

89%

Superior performance of girls

Bahrain Jordan Cyprus Philippines Armenia Republic of Moldova Singapore Macedonia Serbia

33* 27* 16* 13* 10* 10* 10* 9* 7*

Equal performance

61%

Iran Malaysia Palestinian National Authority 5.0 Chinese Taipei 5.0 Latvia 4.0 Lithuania 4.0 Scotland 4.0 Romania 2.0 Botswana 0.0 New Zealand 0.0 Norway 1.0 Russian Federation 1.0 Slovenia 2.0 Estonia 2.0 Hong Kong 2.0 Egypt

15.0 7.0 7.0

20%

9 8 8 7 6 5 5 4 3 3 3 3 3 2 2 1

ALEXANDER W. WISEMAN ET AL.

Equal performance

Superior performance of girls

Thailand Finland Hungary Belgium (French)

None

TIMSS 1995c

402

Table 1. National Gender Differences (Boys’ Minus Girls’ Mean) on the 8th Grade Mathematics Test in IEA Mathematics Studies (1964–2003).

Superior performance of boys

33%

Netherlands Japan Belgium

22* 25* 38*

Superior performance of boys Canada Israel New Zealand Swaziland Luxembourg Netherlands France

37%

0.78* 1.05* 1.09* 1.40* 1.60* 1.77* 2.84*

Superior performance of boys

Japan Spain Portugal Greece Iran Denmark Republic of Korea Israel

4.0 4.0 5.0 5.0 6.0 6.0 8.0 8.0 8.0 9.0 11.0 11.0 14.0 16.0 20.0

21%

9.0* 10.0* 11.0* 12.0* 13.0* 17.0* 17.0* 29.0*

Canada Finland Lithuania Republic of Moldova Chinese Taipei Indonesia Republic of Korea Latvia Netherlands Slovak Republic Hungary USA Japan Chile Italy South Africa Morocco England Superior performance of boys

Israel Czech Republic Iran Tunisia

11%

16.0* 17.0* 24.0* 25.0*

Indonesia England Slovak Republic Bulgaria Sweden Japan South Africa Republic of Korea Netherlands Israel Saudi Arabia Australia

Superior performance of boys

Italy USA Hungary Lebanon Belgium (Flemish) Morocco Chile Ghana Tunisia

1 0 0 1 1 3 3 5 7 8 10 13

20%

6* 6* 7* 10* 11* 12* 15* 17* 24*

403

*po.05. a Huse´n (1967, p. 240). Germany, Israel, and Scotland are not included in the original results. b Baker and Jones (1993, p. 96). c Mullis, Martin, Fierros, Goldberg, and Stemler (2000a, p. 12). d Mullis et al. (2000b, p. 50). e Mullis, Martin, Gonzalez, and Chrostowski (2004, p. 48).

3.0 3.0 3.0 3.0 4.0 5.0 5.0 5.0 5.0 5.0 6.0 7.0 8.0 9.0 9.0 16.0 17.0 19.0

Shifting Gender Effects

Norway Slovak Republic Switzerland USA Belgium (French) France Austria Netherlands Slovenia New Zealand Czech Republic South Africa Ireland Scotland Hong Kong

404

ALEXANDER W. WISEMAN ET AL.

which there were also a considerable number of countries where girls outperformed boys. So, there were 21% of the SIMS 1982 countries and 20% of the TIMSS 2003 countries where girls’ math achievement scores were signiﬁcantly higher than boys. Thus, with the exception of girls’ advantage in mathematics achievement, the results of these IEA studies suggest that gender differences in achievement favoring boys have declined over the past four decades.2 There is corroborating evidence of this trend from national assessments of mathematics achievement over time in speciﬁc nations as well. For example, in the United States, national assessments of 8th grade mathematics reported no gender differences in 1990, 1992, and 1996; also there is evidence that a male-dominated difference in 12th grade since the early 1990s signiﬁcantly declined in absolute size by 1996 (McGraw, Lubienski, & Strutchens, 2006).

3.2. Hypotheses Even while nation-states face a common international institutional inﬂuence in the direction of greater gender parity, nation-states continue to vary in terms of gender and mathematics performance, as well as to the degree to which internal opportunity structures are stratiﬁed by gender. The latter may reﬂect historical legacies or the extent to which some nation-states more successfully buffer themselves from external inﬂuences. Following the logic described earlier, the main hypothesis (H1) is that cross-national variation in gender stratiﬁcation in future opportunity structures, such as higher education and the labor force, will inﬂuence gender differences in mathematics achievement in school. To test this hypothesis, we examine the effect of gender stratiﬁcation of opportunities in two major arenas of modern society, chosen because they stem from the dual logic of national human capital development and international norms of social justice behind the increasing incorporation of women. The ﬁrst is structural and is indicated by the participation of females and males in the education system, science and engineering higher education, and the labor force. The arena of education and labor market are at the core of a national logic to develop human capital through greater incorporation of women. For the student, performance in academic mathematics is directly tied to some of the best future educational and occupational opportunities. The second arena is political power and social justice for women, and is indicated by women and men holding positions of national leadership and national legal policy about

405

Shifting Gender Effects

women’s ability to exercise control over their own reproduction. For the student, visible women with political power and reproductive rights represent overall levels of gender stratiﬁcation in society and the likelihood of gender egalitarianism in their future adult lives (Riegle-Crumb, 2000). A reasonable conditioning of this basic hypothesis is that the future opportunity–current performance relationship should be more evident among the segments of the opportunity structure that are the most gender-stratiﬁed, and less so among segments in which gender equity is rapidly becoming normative internationally (H2). For example, the crossnational literature indicates that while higher education in general has undergone de-stratiﬁcation in all developed nations over time (Bradley & Ramirez, 1996), a higher level of stratiﬁcation continues to prevail in high status curricular areas (Bradley, 2000, 2006). A similar pattern characterizes the labor market, with evidence of progressive de-stratiﬁcation in some areas but continued gender segregation in the more prestigious occupational sectors (Charles & Grusky, 2004; Shu & Marini, 1998). There are two ways in which we can examine this conditioning hypothesis. First is to predict that the relationship between gender and mathematics achievement will be more inﬂuenced by cross-national variation in the still highly gender-stratiﬁed sector of science and engineering higher education than by cross-national variation in the less gender-stratiﬁed sector of overall enrollment in higher education (H2a). And second is to predict that cross-national variation in gender stratiﬁcation of elite future opportunities, such as science and engineering training and access to upper level civil service, will have a more pronounced effect on gender and mathematics achievement among academically elite students than on general ability students (H2b).

3.3. Data, Measures, Models The TIMSS provides cross-national estimates of gender differences in mathematics performance of U.S.-equivalent age/grades of 4th, 8th, and 12th grade students, as well as related variables described later. Additional measures of the gender stratiﬁcation of opportunity structure were collected from cross-national data sources of the United Nations, Organisation for Economic Co-operation and Development (OECD), and the US National Center for Educational Statistics (NCES) reports among others

406

ALEXANDER W. WISEMAN ET AL.

(Matheson et al., 1996; OECD, 1995, 1997; UNESCO, 1997, 1998; UNICEF, 1996; United Nations, 1994; World Bank, 1996a, 1996b). The TIMSS was conducted during the 1994–1995 school year with 45 nations participating. These nations collected sufﬁcient data for analysis here.3 Each nation used a two-stage sampling design. The ﬁrst stage consisted of a probability-proportionate-to-size sample of schools selected from a sampling frame of all schools in that nation enrolling most of the students in the targeted grade level. The second stage sampled up to two mathematics classrooms per school with an equal probability of selection, and all students in these classrooms were included in the study. Mathematics and science achievement tests as well as background questionnaires for students, teachers, and principals were designed to be comparable across nations. The study developed sampling weights to adjust for disproportional sampling of subgroups and non-response (Gonzalez & Smith, 1997). This sampling procedure yielded four samples of students in each of the countries analyzed here: the 8th and 12th grade students taking mathematics, but not advanced courses, referred to here as ‘‘12th grade general,’’ and the 12th grade students in advanced mathematics courses, or ‘‘12th grade advanced.’’4 3.3.1. Measures at the Student Level Estimated student mathematics score is each student’s TIMSS score on an achievement test designed to capture a range of mathematics skills appropriate to the curriculum at the targeted grade level and is the dependent variable (8th grade mathematics mean ¼ 519.51, SD ¼ 100.29; 12th grade general mathematics mean ¼ 500.46, SD ¼ 102.97; 12th grade advanced mathematics mean ¼ 494.76, SD ¼ 101.54).5 Gender is dummy-coded (1 ¼ female, 0 ¼ male) (8th grade mean ¼ 0.50, SD ¼ 0.50; 12th grade mean ¼ 0.49, SD ¼ 0.50). Family socio-economic status is indicated by ﬁrst, mother’s highest education level which ranges from 1 (ﬁnished primary school) to 6 (ﬁnished university) (8th grade mean ¼ 3.06, SD ¼ 1.67; 12th grade mean ¼ 3.70, SD ¼ 1.61); and second, number of books in the home which ranges from 1 (0–10 books) to 5 (W200 books) (8th grade mean ¼ 3.37, SD ¼ 1.26; 12th grade mean ¼ 3.90, SD ¼ 1.12). 3.3.2. Measures at the National Level Women’s status in education is indicated by (1) the percentage of females participating in all levels of education averaged for each year of the 1990s

Shifting Gender Effects

407

(mean ¼ 49.48, SD ¼ 7.01) and (2) the percentage of females in university majoring in the sciences or engineering in 1992 (mean ¼ 24.45, SD ¼ 8.50). Women’s status in the labor force is indicated by the percentage of females in the economically active labor force in 1995 (mean ¼ 41.33, SD ¼ 5.53). Women’s status in governmental and family policy is indicated by (1) the percentage of females holding ministerial level government positions in 1995 (mean ¼ 10.86, SD ¼ 9.19) and (2) a dummy-coded variable indicating nations with legal abortion availability (1 ¼ yes, 0 ¼ no) in 1995 (mean ¼ 0.63, SD ¼ 0.49). National representation of female students in the 12th grade advanced mathematics, indicated by the national percentage of females in advanced mathematics sample minus the national percentage female in the 12th grade, positive values equal female overrepresentation and negative values equal underrepresentation (mean ¼ .03, SD ¼ .07). Level of national economic development is the Gross Domestic Product (GDP) per capita in 1995 expressed in international dollars (mean ¼ 11,655.74, SD ¼ 56,30.38).

3.3.3. Models To examine evidence relative to the hypotheses, a hierarchical linear regression (HLM) modeling strategy estimates the multilevel and contextual effects of gender stratiﬁcation on student achievement. HLM is especially appropriate in the context of nested data; here this is applied to studentlevel effects nested within national contexts (Bryk & Raudenbush, 1992; Raudenbush & Bryk, 2002; Willms & Raudenbush, 1989). For the 8th and 12th grade general and advanced samples, separate models are estimated for each indicator of female status of future opportunity and in each case the dependent variable is the student’s mathematics score. The objective here is to estimate the effect of cross-national variation in gender stratiﬁcation of opportunity on size of the student-level gender gap in mathematics. This is done as follows. First an equation is estimated of the inﬂuence of a student’s gender and, for control of family background effects, two indicators of the student’s family SES level: Yij ¼ b0j þ b1j ðFemaleÞij þ b2j ðBooksÞij þ b3j ðMother0 s educationÞij þ eij (1) where Yij is the achievement score of the ith student within country j, b0j the estimate of the adjusted mean achievement score for country j, and eij the student-level residual. By assumption, E(eij) ¼ 0 and Var(eij) ¼ s2.

408

ALEXANDER W. WISEMAN ET AL.

For the national level equations, the coefﬁcients identiﬁed in the student-level equation (b0j . . . b3j) are modeled as a function of national characteristics. The gender stratiﬁcation of opportunity variables is the effect of substantive importance here and is entered as predictor variables in separate models one at a time. Additionally, the level of national economic development (g02) is included as a national level control. In this analysis, estimates for non-centered indicators of gender stratiﬁcation (g01) predict countries’ adjusted mean achievement scores (b0j): b0j ¼ g00 þ g01 ðGender Stratification IndicatorÞ0j þ g02 ðNational Economic DevelopmentÞ0j þ u0j

ð2Þ

where g00 is an estimate of the adjusted international mean achievement score, and u0j the nation level residual. By assumption, E(u0j) ¼ 0 and Var(u0j) ¼ t00. We also used indicators of gender stratiﬁcation of opportunity structure (g11) to measure each country’s adjusted mean of student’s gender (b1j) as a predictor of student achievement b1j ¼ g10 þ g11 ðGender Stratification IndicatorÞ0j

(3)

where g10 is roughly an adjusted average gender gap across countries. The error term in this equation is constrained at zero. As in Eq. 2, the same, non-centered, indicators of gender stratiﬁcation were entered into this equation and estimated independently of one another (although the same indicator was simultaneously entered into both Eqs. 2 and 3 each time). The relationship hypothesized about gender stratiﬁcation of national opportunity structure is identiﬁed in Eq. 3, where indicators of gender stratiﬁcation are independently modeled on each nation’s adjusted mean of student’s gender as a predictor of student achievement. The coefﬁcients g11 and g12 yield the effects of gender stratiﬁcation on the gender gap within each country (b1j). These coefﬁcients estimate the degree to which the effects of students’ gender on student achievement vary crossnationally according to national variation in gender stratiﬁcation of future opportunity.

409

Shifting Gender Effects

3.4. Results As shown in Tables 2, 3, and 4 for the 8th and 12th grade general and advanced samples respectively, the same equation is estimated with the only difference being the indicator of gender stratiﬁcation inﬂuence on the relationship between gender and mathematics achievement.6 The effects of the same ﬁve indicators of national gender stratiﬁcation of future opportunities are estimated in the 8th and 12th grade general mathematics equations, and only the indicators of gender stratiﬁcation of science and engineering training and holding high political position are estimated in the 12th grade advanced equations. And one additional equation is estimated for the 12th grade advanced samples. In all of the models the coefﬁcients in the fourth row indicate that being female lowers a student’s mathematics score relative to males after controlling for other student-level factors. This indicates the overall level of persistent gender effects on mathematics performance among students in all of these nations. The coefﬁcients in the ﬁrst row test the main future-opportunity current-performance hypothesis (H1) by indicating how various measures of gender stratiﬁcation of opportunity structures modify the relationship between gender and mathematics performance.7 A positive coefﬁcient (i.e., increase in the female slope) indicates that living in a nation with more gender equality in a particular social arena weakens the male-dominated relationship between gender and mathematics achievement. Or in other words, a statistically signiﬁcant positive coefﬁcient implies that in nations where there is more gender equity in future opportunities, current mathematics performances become more equal between male and female students. In considering these results, it should be kept in mind that these models are very conservative in that they control for family SES at the student level and GDP at the national level; two prominent variables that are associated with much of the between-student achievement variation and between-nations achievement variation respectively. It is relatively rare in these kinds of models to ﬁnd robust effects of national level characteristics on individual schooling processes, and this speaks to the underlying dynamics of the main hypothesis. As shown in Table 2, greater gender equity in education and the labor market lessens the effect of gender on mathematics achievement among the 8th graders. More gender equity in a nation’s opportunities in the overall educational arena and in the study of science and engineering in higher education, where mathematics is an important gatekeeper subject, increase the

HLM Estimates of Effects of Gender Stratiﬁcation of Opportunity and Status on the Relationship Between Student Gender and 8th Grade Mathematics Achievement. Indicators of Gender Stratiﬁcationa

Dependent Variable: 8th Grade Mathematics Achievement

Women’s structural roles in education and labor force

Fixed effects Gender stratiﬁcation indicator on female slope National level controls Gross domestic product per capita 1995

Student level controls Female Mother’s highest education level Number of books in student’s home Intercept Random effects Level 2 Variance, u0j Level 1 Variance, rij

*po.05, **po.01, ***po.001. a Country N ¼ 32, Student N ¼ 89,103. b Standard errors are in parentheses.

Percentage of female in total education (1990s)

Percentage of female in science and engineering higher education (1992)

Percentage of female in labor force (1995)

0.61*** (0.08)b

0.42*** (0.09)

0.57*** (0.12)

3.27E-03* (1.44E-03) 2.67* (1.05)

2.35E-03 (1.64E-03) 3.19* (1.22)

4.18E-03* (1.61E-03) 0.59 (1.72)

35.86*** (4.19) 8.64*** (0.21) 14.38*** (0.27) 546.26*** (55.40)

15.51*** (2.29) 8.69*** (0.23) 14.52*** (0.29) 501.59*** (37.22)

29.81*** (4.95) 8.58*** (0.21) 15.12*** (0.26) 382.86*** (78.19)

1692.72 6808.82

1766.31 6974.38

2327.21 6896.64

Societal egalitarianism in politics and reproductive rights Percentage of female in ministerial level of government (1995)

0.04 (0.07)

Abortion availability (1995)

1.81 (1.24)

4.43E-03* (1.88E-03) 1.35 (1.18)

3.34E-03* (1.52E-03) 40.63* (17.80)

5.11*** (0.89) 8.54*** (0.21) 15.17*** (0.26) 414.05*** (20.19)

6.72*** (1.04) 8.53*** (0.21) 15.16*** (0.26) 384.97*** (22.19)

2411.38 6915.62

2106.76 6915.47

ALEXANDER W. WISEMAN ET AL.

Gender stratiﬁcation indicator on intercept

410

Table 2.

HLM Estimates of Effects of Gender Stratiﬁcation of Opportunity and Status on the Relationship Between Student Gender and 12th Grade General Mathematics Achievement. Indicators of Gender Stratiﬁcationa

Dependent Variable: 12th Grade Mathematics Achievement

Women’s structural roles in education and labor force Percentage of female in total education (1990s)

Fixed effects Gender stratiﬁcation indicator on female slope National level controls Gross domestic product 1995 Gender stratiﬁcation indicator on intercept Student level controls Female Number of books in student’s home Mother’s highest education level Intercept Random effects Level 2 Variance, u0j Level 1 Variance, rij

po.10, *po.05, **po.01, ***po.001. Country N ¼ 22, Student N ¼ 48,612. b Standard errors are in parentheses. a

0.01*** (1.47E-03) 0.04 (1.30)

0.68*** (0.17) 4.44E-03** (1.30E-03) 0.23 (1.03)

Percentage of female in labor force (1995)

0.34 (0.25) 0.01** (1.59E-03) 3.01 (2.42)

Percentage of female in ministerial level of government (1995)

0.87*** (0.08) 2.35E-03 (1.95E-03) 2.32* (1.01)

Abortion availability (1995)

4.70** (1.79) 4.82E-03** (1.48E-03) 23.90 (15.70)

63.67*** (7.21) 18.67*** (0.41) 8.37*** (0.30) 354.99*** (72.99)

21.88*** (4.00) 19.00*** (0.44) 8.38*** (0.32) 369.87*** (30.02)

49.02*** (11.17) 18.40*** (0.39) 9.04*** (0.29) 227.83þ (112.45)

22.01*** (1.39) 18.57*** (0.38) 9.00*** (0.28) 361.37*** (20.50)

37.68*** (1.56) 18.37*** (0.38) 8.87*** (0.28) 346.57*** (23.95)

951.98 7175.79

538.84 7279.60

1245.01 7146.84453

1156.07661 7121.6691

1157.11896 7113.21409

411

þ

0.54*** (0.14)b

Percentage of female in science and engineering higher education (1992)

Societal egalitarianism in politics and reproductive rights

Shifting Gender Effects

Table 3.

412

ALEXANDER W. WISEMAN ET AL.

Table 4. HLM Estimates of Effects of Gender Stratiﬁcation of Opportunity and Status on the Relationship Between Student Gender and 12th Grade Advanced Mathematics Achievement. Dependent Variable: 12th Grade Advanced Mathematics Achievement

Fixed effects Gender stratiﬁcation indicator on female slope National level controls Gross domestic product 1995 Gender stratiﬁcation indicator on intercept Student level controls Female Number of books in student’s home Mother’s highest education level Intercept Random effects Level 2 Variance, u0j Level 1 Variance, rij

Indicators of Gender Stratiﬁcationa Percentage of female Percentage of female in science and in ministerial level engineering higher of government education (1995) (1992)

1.82*** (0.28) 5.08E-04 (2.32E-03) 1.26 (1.77)

0.42** (0.14) 1.19E-03 (2.53E-03) 0.03 (1.24)

Representation of females in advanced mathematics (% advanced minus % total) (1995)

0.86*** (0.22)b 8.94E-04 (1.78E-03) 1.52 (1.34)

79.10*** (6.93) 10.64*** (0.77) 6.77*** (0.49) 416.28*** (57.44)

41.94*** (2.35) 12.56*** (0.71) 7.33*** (0.46) 457.75*** (27.81)

37.66*** (1.34) 12.81*** (0.70) 7.41*** (0.45) 445.17*** (25.70)

1330.88 7790.13

1545.43 7842.06

1285.45 7843.70

*po.05, **po.01, ***po.001. a Country N ¼ 17, Student N ¼ 19,736. b Standard errors are in parentheses.

female slope, and hence reduce gender effects on current performance in mathematics. And the same is true of cross-national variation in gender equity in the overall labor market. For the 8th graders neither gender stratiﬁcation of political power, or variation in reproductive rights have an inﬂuence. As shown in Table 3, gender effects on mathematics among the 12th graders taking general mathematics are lessened in nations with more gender equity in all levels of education and in nations with a more liberal reproductive rights policy. But there is no effect among these students of gender stratiﬁcation of national labor markets, and there are negative effects

Shifting Gender Effects

413

of more gender equity in science and engineering training in higher education and high proﬁle political power. As shown in Table 4, gender effects on mathematics among the 12th graders taking advanced mathematics decline (1) in nations with more females participating in science and engineering training in higher education and (2) in nations with females in the upper echelons of government. In analyses not reported here, gender effects on performance among these advanced students was not inﬂuenced by stratiﬁcation of the overall education system, or the overall labor market, or by more liberal policy on reproduction rights.8 To illustrate the future-opportunity current-performance effect, consider, for example, nations like Iran, Israel, and South Korea where there is pronounced male advantages at the 8th grade level in mathematics. These nations also tend to have higher gender stratiﬁcation of future opportunities, as reﬂected by lower percentages of females in the labor force and less females studying advanced science and engineering. At the same time, nations like Sweden and Iceland have no gender differences in mathematics and also have very low levels of gender stratiﬁcation of the structural arena of education and labor market. The United States, with no national gender difference in the 8th grade and a male advantage in the 12th grade general mathematics, falls among nations that are above the world average in incorporating females into advanced education and the labor market. The main future-opportunity current-performance hypothesis is supported by these results, although not perfectly. The hypothesis receives more support from the 8th and 12th grade advanced samples than from the 12th general samples, where there are two results running counter to the main hypothesis. There is more inﬂuence in the direction predicted of indicators of structural opportunities in education and the labor market than in the arena of political power and policies towards reproductive rights of women. Also, the differences in what opportunities inﬂuence gender effects on mathematics performance between the two 12th grade samples is telling and supports the conditioning hypothesis (H2b). If the future-opportunity current-performance hypothesis is correct, gender effects among advanced mathematics students should be inﬂuenced by national gender stratiﬁcation of more elite opportunities than more general opportunities. And this is what the results show. Students, and all the micro processes around them, are sensitive to the opportunities that ﬁt their abilities. Finally, as to the other conditioning hypothesis (H2a), namely that cross-national variation in arenas with overall more persistent gender stratiﬁcation would have the largest inﬂuence on gender effects on achievement, although there is not overwhelming support for it, it is the case for the larger signiﬁcant inﬂuence

414

ALEXANDER W. WISEMAN ET AL.

of science and engineering on the 12th grade advanced compared to no inﬂuence of gender stratiﬁcation on the overall education system.9 When in many of the participating TIMSS nations, policy discussions were made about the implications of bivariate gender effects, much was made of the large male-dominated effects in the 12th grade, particularly among advanced mathematics students. The argument was, if even among the mathematically talented 12th grade students such uniform gender effects are found, then it must be because of a universal micro effect of gender, not a macro-socio logical one. For example, these kinds of results have in the past been cited to support claims of endogenous physical, hormonal, and even brain differences between men and women (e.g., Benbow & Stanley, 1980). The results earlier suggest a countering social factor; namely, that within the overall trend towards greater incorporation of women, gender equity of opportunity structures does lessen the persistent effect of a student’s gender on their mathematics performance. But what about the large male dominated differences among the 12th grade advanced mathematics students in all the nations, even though some nations have considerable gender parity in elite opportunities? One ﬁnal analysis suggests a plausible explanation that is in line with the international rationale for greater gender incorporation for human capital development and social justice. The ﬁnal model estimated for the 12th grade advanced samples examines the effect of cross-national variation in the national representation of females in advanced mathematics. Even among the 17 nations collecting data from advanced mathematics students, there is substantial cross-national variation in female representation in the advanced mathematics courses. For example, gender stratiﬁcation of advanced mathematics varies from a nation like Germany with a statistical overrepresentation of female students at 57% of the enrollment, to a nation like the United States with gender parity in advanced mathematics across the nation, and ﬁnally to nations like Greece, Israel, and Switzerland with low female representation in the 12th grade advanced mathematics. All of the reasons for these differences are beyond the analysis here, but this variable proves important in explaining the prevalence of male-dominated gender differences among advanced mathematics students. The third equation in Table 4 shows that overrepresentation of females in advanced 12th mathematics lowers, not raises, the female slope (H3b).10 Among advanced mathematics students, overrepresentation in female enrollment unintentionally creates a less selective pool of females relative to males and increases average gender differences among advanced students. Statistical overrepresentation and a larger and unintentionally less talented

Shifting Gender Effects

415

pool of females in advanced mathematics is likely to be a by-product of the worldwide interest in incorporating females into science and technology. This effect is a substantial cross-national force in the creation of national gender differences among these students. To illustrate just how substantial, a bivariate regression with nations as units of analysis indicates that the gender ratio in advanced mathematics enrollment accounts for about 70% of the cross-national variation in mean gender differences in advanced mathematics. What was widely celebrated as a ‘‘natural’’ gender difference among talented students is actually an unintentional by-product of attempts to incorporate females in advanced training.

4. A CROSS-NATIONAL PERSPECTIVE ON GENDER EFFECTS ON MATHEMATICS PERFORMANCE Taken as a whole, these analyses support the perspective that there are corresponding links between the stratiﬁcation of relevant future opportunities and social stratiﬁcation of individual performances leading to those opportunities. The main thrust of our ﬁndings is in line with a growing literature on comparative social stratiﬁcation (Charles & Bradley, 2002; Charles & Grusky, 2004; England, 1982; Jacobs & Lim, 1995), as well as the application of an opportunity perspective on cross-national differences in gender effects on mathematics and science achievement and attainment (Baker & Jones, 1993; Hanson et al., 1996; Riegle-Crumb, 2000). These ﬁndings help to validate this leading sociological perspective on gender differences and academic performance found in earlier studies that used less extensive cross-national data than here, as well as statistical models that could not isolate this effect to the degree that is done here. The national context of gender stratiﬁcation of opportunities shapes numerous consequential calculations about effort, appropriateness, and resources devoted to the study of mathematics in school on the part of students and their families. In nations where gender clearly stratiﬁes future educational and labor market opportunities, these calculations may also become stratiﬁed to a degree by gender and ultimately lead to different achievement levels by gender. The effects are most likely subtle, but nevertheless pervasive, across the socialization processes known to shape behavioral patterns between social categories of individual actors. The exact social psychological mechanisms that forge this link are beyond the scope of this chapter, and

416

ALEXANDER W. WISEMAN ET AL.

there is a useful literature on how these might work (e.g., Baker & Jones, 1993; Ogbu, 1979). Similarly there is evidence that certain opportunity structures work for certain groups of students and not others, which further illustrates the strength of this argument. For example, indicators of increased gender equality among future opportunities targeted by elite, highly able students, such as science and engineering training in higher education as well as upper level governmental positions, reduce gender differences in these nations among advanced mathematics students. However, these effects do not apply to gender differences among students in the lower mathematics courses. For these latter students, more general gender equality reﬂected in education overall has a more salient opportunity effect. The opportunity structure argument also works for 8th grade gender differences, but since these samples mix all types of students together (high and low mathematics aptitudes), there most likely is an offsetting mix of sensitivity to gender stratiﬁcation of elite versus more general opportunities. The clearest illustration among these data of the potential inﬂuence of mass education on gender differences in achievement beyond a future opportunity argument is the unexpected effect of overrepresentation of females in advanced 12th grade mathematics on increasing gender differences. Although the causes of why particular students end up in advanced mathematics classes are not discernable from the TIMSS data, one can assume that one cause for female overrepresentation are policies encouraging greater mathematics participation by females. It is not surprising that a consequence of greater gender equality in education is greater interest in advanced mathematics among female students. As these analyses show, however, with greater emphasis on female participation also comes the unintentional likelihood of less selective female cohorts relative to males and hence a lower average performance compared to male students, who as a group may remain more selective. While it can be argued that this is a minor problem relative to the greater good of more female representation in advanced mathematics, the reporting of these sizable gender differences in advanced mathematics performance by public domestic and international agencies, as was the case in the wake of the initial release of the TIMSS data, can further gender stereotypes. These stereotypes have included the image that females do not perform well in advanced mathematics as well as a broader image of performance parity in more elementary mathematics and some ‘‘innate’’ male advantage in advanced mathematics, which was a persistent public belief stemming from high proﬁle earlier studies (e.g., Benbow & Stanley, 1980, 1982, 1983a, 1983b).

Shifting Gender Effects

417

The effects described here suggest that some caution should be taken in accepting this idea of gender mathematics differences increasing over school careers. Considering the concurrent inﬂuences of mass education and opportunity structures available to female students yields a number of other hypotheses to be tested. Including the broader sociological contexts of mass education and over-arching trends towards greater incorporation of women enriches the future opportunity argument for considering how social context inﬂuences gender effects, as well as other inequality producing social categories (i.e., race). Without arguments about the role of schooling and the larger process of gender stratiﬁcation, questions about historical shifts in gender differences and unintended effects arising from the sources of gender stratiﬁcation are difﬁcult to explore. Mathematics in school is an important ‘‘gate-keeper’’ skill for students’ long-term educational and career changes. Understanding how social factors, such as gender, inﬂuence performance links patterns of social stratiﬁcation to individual differences in performance. Recently, biological arguments about why gender has an inﬂuence on mathematical ability have been reasserted. Similarly, there is an extensive literature that examines very micro-social psychological processes as an explanation of gender effects (for an early example, see Eccles, 1986). Neither argument, however, can be adequately expanded to account for the existence of cross-national variation in gender differences in mathematics performance. Unless one is to make the unsubstantiated assumption that the very basic biological and social psychological processes vary across modern nations, these perspectives do not go far enough to explain cross-national variation in gender differences. The ‘‘future opportunity–current performance’’ argument sheds some light on cross-national gender differences by linking variation in gender stratiﬁcation of relevant opportunities to gender effects on mathematics performance across nations. These results add to a crossnational perspective on the creation of gender differences about which much has been speculated. Adding this component, however, does not rule out signiﬁcant social psychological processes. More immediate social contexts than national educational systems, labor markets, and governments are indeed signiﬁcant. Similarly, it must be admitted that biological factors have some inﬂuence on social behavior. Rather the approach investigated here enriches the study of gender and performance. In addition to biological and psychological dimensions, gender is an extremely salient social characteristic, and hence it is a very complicated social variable with clear implications for social behavior at a number of levels of the social system. For example, there are

418

ALEXANDER W. WISEMAN ET AL.

important ﬁndings about gender and school performance originating from biological and social psychological inﬂuences that do not require theory at more aggregated levels of the social system (e.g., Simmons & Blyth, 1987). What the future-opportunity current-performance argument does, though, is to incorporate gender stratiﬁcation at national levels into the analysis of gender and mathematics performance. Salient cross-national variation in gender differences, in both size and direction, has been evident throughout 40 years of international testing. Consequently, national gender stratiﬁcation of future educational and occupational opportunities provides some understanding about the origins of cross-national differences. Lastly, the analysis here expands the future opportunity argument by adding historical and institutional components. Without these the argument is more limited to a steady-state relationship between national levels of gender stratiﬁcation and performance patterns. This is unrealistic given recent increases in female incorporation into modern society and what that means for reduction of inequalities sector by sector. Similarly, gender and stratiﬁcation are developed through institutional arrangements. The mass school system, and its role in advancing both egalitarian ideology and speciﬁc organizational processes, plays an important part in this process.

NOTES 1. These historical cross-national trends in gender differences in mathematics are taken from the First and Second International Mathematics Studies (FIMS in 1964 and SIMS in 1982) as well as the TIMSS in 1995, TIMSS-Repeat in 1999 and the TIMSS in 2003. Each of these studies was conducted by the International Association for the Evaluation of Educational Achievement (IEA) (Huse´n, 1967; Travers & Westbury, 1989). 2. There are too few national cases to estimate a ﬁx-effect model that examines the change in female status on the declining size of gender differences across these time periods. 3. Beginning with the 2003 study, the acronym ‘‘TIMSS’’ now stands for Trends in International Mathematics and Science Study. In TIMSS 1995 (used here), 41 nations collected usable data for 8th grade analysis, 22 nations for the 12th grade general, and 17 nations for the 12th advanced. 4. For replication purposes for other parts of the study, TIMSS also included 3rd and 7th grade samples but with the same testing materials as the upper adjacent target grade. Preliminary analyses of these samples supported the general conclusions here. 5. IRT (Item Response Theory) scale scores were estimated for each student based on an imputed plausible value analysis undertaken by the TIMSS study. Five plausible values per student were estimated (Gonzalez & Smith, 1997).

419

Shifting Gender Effects

After extensive preliminary analysis, we found no substantial differences among results using these values; therefore, the ﬁrst plausible value is used here. 6. The 4th grade assessments presented too modest an amount of overall crossnational gender differences to undertake an opportunity structure analysis. 7. Various models in which more than one indicator of gender stratiﬁcation yielded essentially the same results were conducted. 8. Analyses available from authors upon request. 9. Results showing non-signiﬁcant coefﬁcient for gender stratiﬁcation of overall education with 12th grade advanced available upon request from authors. 10. A student level analysis of likelihood of enrolling in advanced mathematics is not possible with the TIMSS data for a number of reasons such as the absence of information on 12th grade students not taking mathematics.

REFERENCES Arie`s, P., & Be´jin, A. (1985). Western sexuality: Practice and precept in past and present times. New York: Blackwell. Baker, D. P., & Jones, D. P. (1993). Creating gender equality: Cross-national gender stratiﬁcation and mathematical performance. Sociology of Education, 66(April), 91–103. Baker, D. P., & LeTendre, G. K. (2005). National differences, global similarities: Current and future world institutional trends in schooling. Stanford: Stanford University Press. Benbow, C. P., & Stanley, J. C. (1980). Sex differences in mathematical ability: Fact or artifact? Science, 210(4475), 1262–1264. Benbow, C. P., & Stanley, J. C. (1982). Consequences in high school and college of sex differences in mathematical reasoning ability: A longitudinal perspective. American Educational Research Journal, 19(4), 598–622. Benbow, C. P., & Stanley, J. C. (1983a). Differential course-taking hypothesis revisted. American Educational Research Journal, 20(4), 469–473. Benbow, C. P., & Stanley, J. C. (1983b). Sex differences in mathematical reasoning ability: More facts. Science, 222(4627), 1029–1031. Berkovitch, N. (1999). From motherhood to citizenship: Women’s rights and international organizations. Baltimore: The Johns Hopkins University Press. Bourdieu, P. (1977). Cultural reproduction and social reproduction. In: J. Karabel & A. H. Halsey (Eds), Power and ideology in education (pp. 487–510). New York: Oxford University Press. Bradley, K. (2000). The incorporation of women into higher education: Paradoxical outcomes? Sociology of Education, 73, 1–18. Bradley, K. (2006). Cultural coexistence: Gender egalitarianism and difference in higher education. In: D. P. Baker & A. W. Wiseman (Eds), The impact of comparative education research on institutional theory (Vol. 7, pp. 77–96). Oxford: Elsevier. Bradley, K., & Ramirez, F. O. (1996). World polity and gender parity: Women’s share of higher education, 1965–85. Research in Sociology of Education and Socialization, 11, 63–91. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data analysis methods. Newbury Park, NJ: Sage. Chang, M. L. (2000). The evolution of sex segregation regimes. American Journal of Sociology, 105(6), 1658–1701.

420

ALEXANDER W. WISEMAN ET AL.

Charles, M., & Bradley, K. (2002). Equal but separate? A cross-national study of sex segregation in higher education. American Sociological Review, 67(4), 573–600. Charles, M., & Grusky, D. B. (2004). Occupational ghettos: The worldwide segregation of women and men. Stanford, CA: Stanford University Press. Collins, R. (1979). The credential society: An historical sociology of education and stratiﬁcation. New York: Academic Press. Eccles, J. S. (1986). Gender-roles and women’s achievement. Educational Researcher, 15(6), 15–19. England, P. (1982). The failure of human capital theory to explain occupational sex segregation. Journal of Human Resources, 17, 358–370. ETAN Report on Women and Science. (2000). Science policies in the European Union: Promoting excellence through mainstreaming gender equality. Brussels, Belgium: European Commission. Fiala, R., & Gordon-Langford, A. (1987). Educational ideology and the world educational revolution, 1950–1970. Comparative Education Review, 31, 315–332. Fuller, B., & Rubinson, R. (Eds). (1992). The political construction of education: The state, economic change, and school expansion. New York: Praeger. Gerris, J. R. M., Dekovic´, M., & Jansens, J. M. A. M. (1997). The relationship between social class and childrearing behaviors: Parents’ perspective taking and value orientations. Journal of Marriage and the Family, 59(4), 834–847. Gonzalez, E. J., & Smith, T. A. (Eds). (1997). User guide for the TIMSS international database. Chestnut Hill, MA: International Study Center, Boston College, International Association for the Evaluation of Educational Achievement. Hanson, S. L., Schaub, M., & Baker, D. P. (1996). Gender stratiﬁcation in the science pipeline: A comparative analysis of seven countries. Gender and Society, 10(3), 271–290. Huber, E., & Stephens, J. D. (2000). Partisan governance, women’s employment, and the social democratic service state. American Sociological Review, 65(3), 323–342. Huse´n, T. (Ed.) (1967). International study of achievement in mathematics: A comparison of twelve countries. Stockholm: Almquist & Wiskell. Jacobs, J. A. (1996). Gender inequality and higher education. Annual Review of Sociology, 22, 153–185. Jacobs, J. A., & Lim, S. T. (1995). Trends in occupational and industrial sex segregation in 56 countries, 1960-1980. In: J. A. Jacobs (Ed.), Gender inequality at work (pp. 259–293). Thousand Oaks, CA: Sage. Jones, D. P. (1989). Gender differences in mathematics achievement: A cross-national analysis. Washington, DC: Catholic University of America. Kerckhoff, A. C. (1977). Race and social status differences in the explanation of educational ambition. Social Forces, 55(3), 701–714. Marini, M. M., & Brinton, M. (1984). Sex-typing in occupational socialization. In: B. F. Reskin (Ed.), Sex segregation in the workplace: Trends, explanations, remedies (pp. 191–232). Washington, DC: National Academy Press. Matheson, N., Salganik, L. H., Phelps, R. P., Perie, M., Alsalam, N., & Smith, T. M. (1996). Education indicators: An international perspective, NCES 96-003. Washington, DC: National Center for Education Statistics, U.S. Department of Education. McClelland, K. (1990). Cumulative disadvantage among the highly ambitious. Sociology of Education, 63(2), 102–121.

Shifting Gender Effects

421

McGraw, R., Lubienski, S. T., & Strutchens, M. E. (2006). A closer look at gender in NAEP mathematics achievement and affect data: Intersections with achievement, race/ ethnicity, and socioeconomic status. Journal for Research in Mathematics Education, 37(2), 129–150. Meyer, J. W., Boli, J., Thomas, G. M., & Ramirez, F. O. (1997). World society and the nationstate. American Journal of Sociology, 103(1), 144–181. Mickelson, R. A. (1990). The attitude-achievement paradox among black adolescents. Sociology of Education, 63(1), 44–61. Mullis, I. V. S, Martin, M. O., Fierros, E. G., Goldberg, A. L., & Stemler, S. E. (2000a). Gender differences in achievement: IEA’s Third International Mathematics and Science Study (TIMSS). Chestnut Hill, MA: TIMSS International Study Center, Boston College. Mullis, I. V. S, Martin, M. O., Gonzalez, E. J., Gregory, K. D., Garden, R. A., O’Connor, K. M., Chrostowski, S. J., & Smith, T. A. (2000b). TIMSS 1999 International Mathematics Report. Chestnut Hill, MA: The International Study Center, Boston College. Mullis, I. V. S, Martin, M. O., Gonzalez, E. J., & Chrostowski, S. J. (2004). TIMSS 2003 International Mathematics Report. Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College. OECD. (1995). Education at a glance, OECD indicators (3rd ed.). Paris: Organisation for Economic Co-operation and Development. OECD. (1997). Education at a glance, OECD indicators. Paris: Organisation for Economic Co-operation and Development. Ogbu, J. U. (1979). Social stratiﬁcation and the socialization of competence. Anthropology and Education Quarterly, 10(1), 3–20. Ramirez, F. O., Soysal, Y. N., & Shanahan, S. (1997). The changing logic of political citizenship: Cross-national acquisition of women’s suffrage rights, 1890 to 1990. American Sociological Review, 62(October), 735–745. Ramirez, F. O., & Wotipka, C. M. (2001). Slowly but surely? The global expansion of women’s participation in science and engineering ﬁelds of study, 1972–92. Sociology of Education, 74(3), 231–251. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage. Ridgeway, C. L. (1997). Interaction and the conservation of gender inequality: Considering employment. American Sociological Review, 62(2), 218–235. Ridgeway, C. L., & Correll, S. J. (2004). Unpacking the gender system: A theoretical perspective on gender beliefs and social relations. Gender and Society, 18(4), 510–531. Riegle-Crumb, C. C. (2000). International gender inequity in math and science education: The importance of gender stratiﬁcation across generations. Chicago, IL: University of Chicago. Schultz, T. W. (1982). Investing in people: The economics of population quality. Berkeley, CA: University of California Press. Shu, X., & Marini, M. M. (1998). Gender-related change in occupational aspirations. Sociology of Education, 71(1), 43–67. Simmons, R. G., & Blyth, D. A. (1987). Moving into adolescence: The impact of pubertal change and school context. New York: A. de Gruyter. Stage, F. K., & Maple, S. A. (1996). Incompatible goals: Narratives of graduate women in the mathematics pipeline. American Educational Research Journal, 33(1), 23–51. Stevenson, D. L., Schiller, K. S., & Schneider, B. (1994). Sequences of opportunities for learning. Sociology of Education, 67(3), 184–198.

422

ALEXANDER W. WISEMAN ET AL.

Stinchcombe, A. L. (1964). Rebellion in a high school. Chicago: Quadrangle Books. Travers, K. J., & Westbury, I. (Eds). (1989). The IEA study of mathematics I: Analysis of mathematics curricula. Oxford: Pergamon Press. UNESCO. (1997). Statistical yearbook. Paris: UNESCO. UNESCO. (1998). World education report. Paris: UNESCO. UNICEF. (1996). The state of the world’s children. Available at http://www.unicef.org United Nations. (1994). Women’s indicators and statistics (WISTAT) database. New York: United Nations. Willms, J. D., & Raudenbush, S. W. (1989). A longitudinal hierarchical linear model for estimating school effects and their stability. Journal of Educational Measurement, 26, 209–232. Wiseman, A. W., & Baker, D. P. (2007). Educational achievements in international context. In: B. J. Bank (Ed.), Gender and education: An encyclopedia (pp. 407–414). Westport, CT: Greenwood Press. World Bank. (1996a). Selected world development indicators. Washington, DC: World Bank. World Bank. (1996b). World development indicators [CD-ROM]. Washington, DC: International Bank for Reconstruction and Development.

AUTHOR INDEX Bahry, S. 267 Bailey, S. 110 Baker, D.P. 336, 397–398, 401, 403, 415–416 Balihuta, A. 102 Balira, R. 125 Bandura, A. 369 Bankston, C.L. 188, 209 Barbaranelli, C. 369 Barber, B.L. 338 Barber, C. 365–366, 368, 373, 375, 388 Barnes, S.H. 360 Barrow, K. 146 Barth, E. 335 Bat-Erdene, R. 229–230 Bauer, R. 256 Baulch, B. 148–149, 162, 177 Baumert, J. 336 Benbow, C.P. 336, 414, 416 Bendera, S. 95 Beneria, L. 24 Bennell, P. 93, 133 Berkovitch, N. 399 Berman, J. 124 Berry, J.W. 176 Biraimah, K.C. 40–41, 52–53, 59 Be´jin, A. 399 Black, J. 265 Blackstone, R.M. 335–336 Blanc, A.K. 11 Blau, P.M. 4 Bledsoe, C. 139 Bleeker, M.M. 336, 338 Bloch, M. 41, 59 Bloom, D. 10–11

Abdullah, A. 313 Abu Nasr, J. 306, 320 Adams, J. 187 Ahlstrand, B. 105 Akora, A. 240 Alderman, H. 2, 24 Alfeld, C.J. 338 Al-Kaabi, A. 312–314, 320 Alozie, N.O. 357, 359, 361 Alsalam, N. 406 Altbach, P.G. 237–238, 308 Amadeo, J.-A. 364, 371–372, 387, 389 Amano, I. 220–222, 245 Amano, M. 228 Aminova, R. 264 Amsler, S. 285 Anderson, K. 266 Anderson-Levitt, K.M. 41, 59 Andolina, M. 367 Andreas, J. 189 Anemona, A. 125 Aranha, R. 239, 242 Arif, M. 315 Arie`s, P. 399 Arjmand, R., 305 Armingeon, K. 369 Armstrong, A. 94–95, 97 Armstrong, P.I. 337 Arnot, M. 41, 59, 218 Arum, R. 188, 209 Ashenden, D.J. 189 Atkeson, L.R. 361 Auvert, B. 125–126 Ayalon, H. 336 Azzam, H. 306, 320 423

424 Blossfeld, H. 336 Blumberg, R.L. 241 Blyth, D.A. 418 Boerma, J. 110 Boli, J. 308, 399 Bond, G. 126 Borders, L.D. 336 Boserup, E. 7 Bourdieu, P. 397 Bowes, J.M. 363 Bowles, S. 189 Bowley, D. 141 Bradburn, N.M. 48–49 Bradley, K. 220, 335–336, 396, 401, 405, 415 Bregman, J. 94–95, 97 Brint, S. 4, 265, 338 Brinton, M.C. 227, 397 Bronfenbrenner, M. 244 Bronfenbrenner, U. 256 Brook Napier, D. 308 Brooke, J. 230 Brooks, B. 335–336 Brown, D.K. 189 Brown, S.D. 336 Browne, J. 335–336 Bruce, J. 96 Bruns, B. 92 Bryant, A.N. 338 Bryk, A.S. 376, 407 Buchmann, C. 11 Burdette, M. 126 Burke, J. 267 Burn, S.M. 325 Burns, N. 369 Buru-Bellat, M. 339 Butler, J. 3 Buve, A. 125–126 Cabrera, A.F. 339 Calloids, F. 93 Campbell, D.E. 367–368, 374, 385 Campbell, J.R. 362, 365

AUTHOR INDEX Caprara, G. 369 Carael, M. 125–126 Carrol, B. 146 Carter, D.F. 338 Case, K.A. 368, 374, 384 Castle, E. 99 Chan, S. 167 Chanana, K. 236, 239–242 Chang, M.L. 399 Changalunga, J. 125 Chant, S. 164, 169 Chantrill, P. 172, 175 Chapeland, V. 338 Chapman, D. 175 Charles, M. 220, 335–336, 396, 401, 405, 415 Chege, J. 125–126 Chen, X. 339 Cheong, Y. 376 Chhin, C.S. 336, 338 Chitnis, S. 237, 240–241, 247 Choy, S.P. 339 Christina, R. 305 Chrostowski, S.J. 374, 403 Chuyen, T.T.K. 148–149, 162, 177 Clark, P. 96 Clemens, M. 92, 146 Cocker, J. 128, 137 Coleman, J.S. 188, 209 Collins, R. 189, 398 Congdon, R. 376 Connell, W. 189 Connelly, R. 186–187 Conover, P.J. 357, 359–360, 366, 369 Corbin, J.M. 192 Corcoran-Nantes, Y. 255, 270 Correll, S.J. 336, 396 Costin, W. 148 Croity-Belz, S. 338 Crombie, G. 337 Csapo, B. 363 Csapo, M. 11 Cummings, W.K. 308

Author Index Daehlen, M. 336 Daun, H. 305 Davaa, S. 229–230 Davey, F.H. 338 David, M. 218 Davidson, D. 189 Davies, L. 110, 306, 319 Davison, J. 40–41, 60 de Fillipis, A. 338 de Waal, A. 124 Debourou, D.M. 45 DeJaeghere, J.G. 175 Dekovic´, M. 397 Delli Carpini, M.X. 358, 364–365, 367 Deloach, S.B. 234 DeRose, L.F. 2, 19 Desai, J. 148, 165 Desjardins, R. 341 DeYoung, A.J. 267, 269–270, 283–284 Diaw, C. 8 Diez-Nicolas, J. 369 Dimmock, C.A. 91 Dobson, R. 256 Domene, J.F. 336 Doussett, G.W. 189 Drummond, T. 283 Dryler, H. 336 Dudwick, N. 256, 268, 283 Duncan, O.D. 4 Dunne, M. 128, 141 Durkheim, E. 188 Dutcher, N. 172 Dwaraki, L. 241 Earl-Novell, S. 217 Eccles, J.S. 338, 397, 417 Eckstein, M.A. 189 Ellingsæter, A.L. 335 Elliot, C.M. 95 Eloundou-Enyegue, P.M. 10–11, 24 Emzita, K. 148 England, P. 415 Epperson, D.L. 338

425 Erikson, R. 339 Ernst, P. 60 Ethell, R. 60 Eve, R. 189, 208 Fagerlind, I. 265 Falkingham, J. 268 Farnen, R.F. 359–360, 363, 365, 367, 369, 371 Feiring, C. 363 Ferguson, J. 126 Fernandez-Ballesteros, R. 369 Fiala, R. 399 Fierros, E.G. 403 Filmer, D. 11, 146 Findlow, S. 303, 310, 312, 314–315, 321, 324 Fine, M. 154 Fink, B. 60 Fiscian, V. 110 Fix Dominguez, R. 225, 231, 234–235, 239, 246 Flanagan, C.A. 363 Fleshman, M. 138 Flores, B.B. 60 Floro, M. 39, 95 Foss, L. 333 Foucault, M. 150 Frank, A.G. 6 Frank, D. 265 Fraser, A. 265 Fraser, N. 3, 8 Friedl, E. 275 Frome, P.M. 338 Fujimura-Fanselow, K. 245 Fullan, M. 105–106 Fuller, B. 96 Gaad, E. 315 Gage, A. 139 Gage-Brandon, A.J. 11 Gal, D.G. 303 Garbus, L. 124

426 Garcia, M. 338 Garden, R.A. 403 Garrett, J. 336 Gati, I. 337 Gausset, Q. 125, 130 Gerris, J.R.M. 397 Gillis, A.R. 189 Gilson, J. 189 Gintis, S. 189 Giroux, S. 10 Gleason, G. 256, 264, 266 Gleick, J. 105 Glick, A. 124 Glynn, J.R. 125–126 Goldberg, A.L. 403 Goldin, C. 217 Gonzalez, E.J. 374, 403 Gordon, R. 95 Gordon-Langford, A. 399 Gottfredson, L.S. 336–337, 341, 350 Graham, S. 60 Greene, M.S. 96 Gregory, K.D. 403 Grenoble, L. 261 Gro¨nlund, K. 358, 362, 364 Grosskurth, H. 125 Grown, C. 2 Grusky, D.B. 335–336, 396, 405, 415 Guba, E.G. 192 Gunewardena, D. 150 Gupta, A. 237 Habib, M. 95 Hackett, G. 336 Hagan, J. 189 Hannum, E. 96, 153, 186–187 Hansen, H.B. 100 Hansen, M.N. 335–336 Hanson, S.L. 336, 397, 415 Hara, J. 219, 227–228, 243, 245 Harker, R. 189 Harris, C. 271 Harris, K.R. 60

AUTHOR INDEX Haughton, D. 148–149, 162, 177 Haughton, J. 148–149, 162, 177 Hausmann, R. 333 Hayes, R.J. 125 Hayward, R. 336 He, Y. 190 Hemmings, A. 211 Herz, B. 11, 95, 175 Heward, C. 242 Hewett, P.C. 2, 19, 97 Heyneman, S. 267–269 Hijab, N. 293 Hill, A. 95 Hill Collins, P. 151–152 Hirosato, Y. 148 Hirschi, T. 188, 209 Hoffer, T. 188, 209 Hoffman, A.L. 234 Holland, J.L. 337 Holmes, B. 256, 260–261 Homer-Dixon, T. 105 Honneth, A. 3 Hooghe, M. 357, 363–366 Horn, L.J. 339 Hotchkiss, L. 338 Hoyle, R.H. 342 Hu, S.P. 339 Huang, L.H. 334, 338–339 Huber, E. 399 Humphreys, S. 141 Hunter, M. 128–129, 137, 141 Husfeldt, V. 364–366, 371–373, 375, 389 Huse´n, T. 339 Hyde, K. 40–41, 95, 133 Ilon, L. 11 Imada, S. 228 Inglehart, R. 358, 366–367, 387 Inkeles, A. 256 Irvin, JR.M.W. 336 Ishida, H. 222, 224, 228, 245–246 Islamov, E. 270

Author Index Jacob, B.A. 219 Jacobs, J.A. 336, 396, 415 Jacoby, S. 256 Jaffrelot, C. 247 Jansen, J.D. 308 Jansens, J.M.A.M. 397 Jargalmaa, T. 229–230 Jarman, J. 335–336 Jasso, G. 370 Jeffrey, C. 137 Jeffrey, P. 137 Jeffrey, R. 137 Jenkins, K. 367 Jennings, M.K. 369, 372, 374 Jensen, G.J. 189, 208 Jimenez, E. 189 Jocobs, J.E. 336, 338 Johnson, M. 267 Jones, D.P. 336, 397–398, 403, 415–416 Jones, G.W. 2, 8, 10, 22 Jones, J. 361–362, 364, 369 Jonsson, B. 363 Jonsson, J.O. 336, 339 Joshi, R. 241 Joy, L. 336 Jo¨reskog, K.G. 341 Junn, J. 362 Kaase, M. 360 Kadzamira, E. 95, 110 Kagoda, A.M. 113, 115 Kahindo, M. 125–126 Kahn, J.H. 338 Kameda, A. 227 Kandiwa, V. 11 Kane, E. 172, 175, 178, 366 Kanyuka, M. 40–41, 60 Kao, G. 153, 187 Kaona, F. 125–126 Karabel, J. 338 Kasente, D. 102 Katz, L.F., 217 Kaufman, C. 141

427 Kaufman, J. 187 Kaufman, R.L. 336 Kaul, R. 237 Keating, D.P. 336 Keeter, S. 358, 364–365, 367 Keeves, J.P. 342 Keller, S. 258, 269 Kelly, G.P. 95, 219, 265, 308 Kelly, M.J. 7, 124–125 Kelly, P. 246 Kennedy, K. 358, 366 Kenway, J. 246 Kerckhoff, A.C. 397 Kerr, S. 256 Kessler, S. 189 Khalid, A. 260 Kim, J. 2, 24, 110 Kimura, R. 227 King, J.E. 95, 217, 219 Kipnis, A. 189 Kirdar, S. 302–303, 306, 309, 324, 327 Kirk, D. 323, 325–326 Kleinbach, R. 285 Ko¨ller, O. 336 Knodel, J. 2, 8, 10, 22 Konadu-Agyemang, K. 231 Korth, B. 261, 283 Kosonen, K. 147–148, 172 Kravdal, O. 2, 19 Kremer, M. 24 Kuate-Defo, B. 140 Kuehnast, K. 256, 268, 283 Kurz, K. 124 Kuziemko, I. 217 Kwesiga, J.C. 106 La Nasa, S.M. 339 Lampel, J. 105 Lazer, S. 362, 365 Leach, F. 110, 141 Leclerc-Madlala, S. 129, 141 Lee, V.E. 95, 189

428 Lehmann, R. 364–368, 371–375, 387, 389 Lemani, E. 110 Lendon, S. 172, 175 Lent, R.W. 336 LeTendre, G.K. 398, 401 Levine, M. 189 Levine, P.B. 336 Lewin, K. 93, 147 Lewis, M.A. 23, 96, 109, 146, 363 Li, Y. 190 Liang, X. 93, 102–105, 114–115 Lichter, D. 10 Liczek, I. 257, 262, 264, 268 Liddle, J. 241 Lim, S.T. 415 Lincoln, Y.S. 192 Lloyd, C.B. 2, 11, 19, 25, 41, 59, 97 Lockheed, M.E. 23, 95–96, 109, 146, 189 Lockwood, M. 8 Longﬁeld, K. 124 Lorﬁng, I. 306, 320 Lovich, R. 128, 137 Lubienski, S.T. 404 Lubinski, D. 336 Luke, N. 124, 141 Lutkus, A.D. 362, 365 MacArthur, C. 60 Machakanja, P. 110 Maehr, M. 97 Magno, C. 234 Malak-Minkiewicz, B. 367, 386 Malhotra, A. 2 Manabe, R. 227–228 Maple, S.A. 398 Maralani, V. 25 Marat, E. 267 Marini, M.M. 397, 405 Marsh, H.W. 189 Marteleto, L. 11 Martin, M.O. 374, 403

AUTHOR INDEX Martin, P. 338 Maslak, M.A. 176 Massell, G. 263 Mastekaasa, A. 335 Matheson, N. 406 Matsui, M. 224 Mayer, J.D. 361–362 Mazawi, A. 310 Mazumdar, V. 239–242 Mazzeo, J. 362, 365 McClelland, K. 397 McGinn, N.F. 308 McGovern, S. 314 McGraw, R. 404 McLaughlin, M. 60 McMann, K. 257, 283 McMeniman, M. 60 McVeigh, B.J. 222, 227 Meece, J.L. 336 Mehmet, O. 124 Mehran, G. 305 Mensch, B. 25, 41, 59 Mensh, B.S. 96–97 Merrill, B.D. 357, 359, 361 Mertaugh, M. 267–269 Meyer, J.W. 92, 265, 399 Mgalla, Z. 110 Mickelson, R.A. 397 Micklin, P. 266 Mies, M. 241 Miguel, E., 24 Miller, L. 336 Milner, H. 358, 362, 364 Mina, F.M. 312 Mingat, A. 92–93 Mintzberg, H. 105 Mir, S. 305 Mirembe, R. 110 Miske, S. 53, 175 Miyazaki, A. 227 Montgomery, D. 268 Mori, S. 227 Morozova, I 264

Author Index Morozumi, A. 221 Mortenson, T.G. 217 Moser, C.O.N. 153 Motel-Klingebiel, A. 369 Moulton, J. 39 Mueller, R.O. 341 Mukaka, M. 128, 137 Mukhopadhyay, C. 241–242 Mullis, I.V.S. 374, 403 Mundy, K. 39 Murakami, E. 224 Musonda, R. 125–126 Nagasawa, M. 222 Nagy, G. 336 Nakajima, Y. 225, 246 Nakanishi, Y. 227 Nakanyike, M. 102 Narita, K. 220–221, 245 Nauta, M.M. 338 Neuman, W.L. 219 Niemi, R. 362 Nikolova, R. 364, 371–372, 389 Niyozov, S. 267 Nkanyike, M. 102 Noah, H.J. 189 Norris, P. 358, 360, 363, 366–367, 369–370, 387 Northrup, D. 256, 264 Nozaki, Y. 219–220, 225, 239, 243, 246 Nun˜ez, A. 339 Obasi, A.I. 125 O’Connor, K.M. 403 O’Gara, C. 40 Ogbu, J.U. 397, 416 Ohsako, T. 110 Olcott, M. 256, 261, 266 Opheim, V. 336 Oppenheim, A.N. 359–360, 363, 365, 367, 369, 371 Orazem, P. 2, 24

429 Osmonov, O. 267 Oswald, H. 364–368, 371–375, 387 Pace, J.L. 188–189, 211 Pajares, M.F. 60, 388 Pande, R. 2 Panter, A.T. 342 Parish, W. 11 Parker, R.A. 48 Parsons, S. 336 Paulsen, R. 360, 368, 386 Peck, M.C. 313 Perie, M. 406 Perna, L.W. 338 Peterson, R.A. 48 Phelps, R.P. 406 Phillips, D. 308–309, 328 Pinto, A. 238 Pitcher, G. 141 Pitt, J. 267 Plummer, M.L. 125 Pomfret, R., 266 Poole, G.S. 221 Powers, M.B. 128, 137 Prakash, V. 237, 247 Rainey, L.M. 336 Rakotomalala, R. 92 Ramirez, F.O. 92, 265, 308, 398–399, 405 Raney, L. 95 Rapoport, R. 361 Rathgeber, E.M. 8 Raudenbush, S.W. 376, 407 Raymond, M. 96 Rea, L.M. 48 Read, G. 256, 260–261 Reagan, T. 305 Røed, M. 335 Reddy, S. 128 Reeves, M. 267, 269–270, 283–284 Rehberg, R.A. 338 Rettinger, V. 362–363

430 Rich, L. 187 Richardson, V. 60 Ridgeway, C.L. 396, 398 Riegle-Crumb, C.C. 397, 405, 415 Rihani, M.A. 96–97, 175 Ringdal, K. 335 Roberts, S. 268 Robinson, B. 229, 231, 235, 246 Robinson, C.L. 336 Rohlen, T.P. 222 Rosenthal, C. 361–362, 364, 369 Rosenthal, J. 361–362, 364, 369 Rosenthal, S. 363 Ross, D.A. 125, 338 Roy, O. 257, 260, 266, 279 Rywkin, M. 256 Saasa, O. 124–125 Sadoulet, E. 96 Saha, L. 265 Sakabe, Y. 11 Salamba, Z. 125 Salganik, L.H. 406 Salomone, R.C. 189 Sapiro, V. 357, 359 Sara, M.H. 270 Savada, A.M. 229 Savickas, M.L., 337 Sax, L.J. 338 Schapink, D. 110 Schaub, M. 336, 397, 415 Schiller, K.S. 398 Schlozman, K.L. 369 Schmidt, H.M. 361–362 Schneider, B. 398 Schoon, I. 336, 338 Schultz, T.W. 399 Schulz, W. 364–368, 371–375, 384, 387 Schwille, J. 364, 371, 387 Scott, F. 315 Sen, A.K. 146–147, 151–152, 161–162, 173, 177 Seymour, S. 241–242

AUTHOR INDEX Shabaya, J. 231 Shanahan, S. 399 Shapka, J.D. 336 Shavit, Y. 336 Shaw, K. 306, 320–321 Sheblanova, E. 363 Sherburne, L. 128, 137 Shofer, E. 265 Shouly, E. 313 Shu, X. 405 Sibberns, H. 372–373, 375, 384 Silova, I. 234, 267 Simmons, R.G. 418 Simon, J. 357, 359, 361 Simpson, A. 127, 130 Simpson, J. 189 Singer, S.I. 189 Singh, S.P. 305 Sit, H.V. 172, 175 Skogen, K. 334–335 Slaughter, S. 219 Smart, J.C. 338 Smith, J. 256 Smith, T.A. 403 Smith, T.M. 406 Solongo, A. 229, 231, 235, 246 Soumare´, A. 41, 59 Soysal, Y.N. 92, 399 Sperandio, J. 112–113, 115 Sperling, G. 175 Spradley, J. 284 Spring, J. 265 So¨rbom, D. 341 Ssekamwa, J.C. 99–100 Stage, F.K. 398 Standing, H. 141 Stanley, J.C. 414, 416 Stash, S. 96 Stavrou, S.E. 141 Steady, F.C. 138 Steiner-Khamsi, G. 308, 310, 327, 371 Steinkamp, M. 97

Author Index Stemler, S.E. 403 Stephens, J.D. 399 Stevenson, D.L. 398 Stewart, A. 360 Stinchcombe, A.L. 396 Stolle, D. 357, 363–366 Stoppard, J.M. 338 Strauss, A. 192 Streitmatter, J. 189 Stromquist, N.P. 10, 97, 146, 152, 266, 290, 318, 328, 368, 374, 386 Strutchens, M.E. 404 Subbarao, K. 95 Subrahmanian, R. 2, 24, 151–152 Sudman, S. 48–49 Sukhbaatar, J. 229–230 Super, C.M. 337 Super, D.E. 337 Sutton, M. 147, 154 Swainson, N. 95, 133 Taifour, M. 313 Takamiya, K., 148 Talhami, G. 320, 326 Tan, J.P. 93 Tashakkori, A. 192 Taylor, T. 336 Teddlie, C. 192 Tesch-Ro¨mer, C. 369 Thanh, T.T. 172, 175 Thomas, G.M. 399 Thompson, M.D. 338 Thornton, R. 24 Tiberondwa, A.K. 98 Tietjen, K. 40, 52 Tikal, J.B.G. 147, 175 Tilak, J. 237 Tipps, D. 7 Tobin, J. 189 Tobin, K. 60 Todd, J. 125 Tomasik, M.J. 369

431 Tomaskovic-Devey, D. 336 Torney, J.V. 359–360, 363, 365, 367, 369, 371 Torney-Purta, J. 364–368, 371–375, 387, 389 Torp, H. 335 Touq, M. 318, 325 Trautwein, U. 336 Tripp, A.M. 106 Tsui, M. 187 Tuijnman, A.C. 342 Tuyen, K., 148 Twaddle, M. 100 Tyson, L.D., 333 Ujihara, Y. 227 Unger, J. 189 Unterhalter, E. 170, 172, 177–179 Usseinova, N. 266 Valiante, G. 388 Valyayeva, G.K. 267, 269–270, 283–284 Van de Walle, D. 150 Varga, C.A. 127 Vavrus, F. 124, 152–153 Velkoff, V. 240 Verba, S. 369 Vignoli, E. 338 Vincent-Lancrin, S. 219 Voskresenskaya, N. 256, 260–261 Waithaka, M. 124 Walford, G. 305 Walker, A. 91 Walker, M. 158, 170, 178 Webb, R.M. 336 Webber, S. 266 Weber, E. 314 Weber, M. 188–189 Wegener, B. 370 Weidman, J.C. 229–230, 257 Weiner, G. 218

432 Weis, L. 154 Weiss, A.R. 362, 365 Weissman, A. 128, 137 Werner, C. 257 Westby, D.L. 338 Whitehead, A. 8 Whiteside, A. 124 Wight, D. 125 Wilkenfeld, B. 368 Willis, R. 11 Willms, J.D. 407 Wils, A. 146 Wilson, W.J. 3, 8 Wiseman, A.W. 401 Wolbrecht, C. 367–368, 374, 385 Wolf, J. 39, 95 Wollstonecraft, M. 7 Woodﬁeld, R. 217 Worden, R.L. 229

AUTHOR INDEX Wotipka, C.M. 265, 398 Wu, D. 189 Yates, L. 97 Ye, Y. 148 Yeager, J.L. 229–230 Yoshihara, K. 228 Young, I.M. 3 Yuan, Q. 190 Zahidi, S. 333 Zhang, Y. 153, 187 Zheng, Z. 186–187 Zhou, M. 188 Zimmer, C. 336 Zimmerman, D. 336 Zucker, A. 360 Zukin, C. 367

SUBJECT INDEX Academic achievement, 99, 217, 334, 336, 340–342, 346–348, 350–351 Academic capability, 48, 50–52, 54–55, 59 Academic performance, 50–52, 54–55, 58, 395–397, 415 Africa, 1–3, 9, 11, 16, 22, 24, 39, 49, 93–95, 97, 112, 124, 128, 138, 141, 147, 309, 386, 403 AIDS, 16, 93, 98, 100, 123–127, 130–131, 133–140, 314 Ajman University of Science and Technology, 317 American University of Sharjah, 317, 319 Arab, 293, 301–302, 304–307, 309–312, 314–316, 318–319 Arab Education, 311 Arab woman, 302, 304 Arabian Gulf, 312 Asian Development Bank, 266, 273 Australia, 316, 318, 363, 377, 402–403 Authority, 186–190, 192, 202–203, 208–211, 241, 257, 262, 264, 320, 402

Benin, 24, 39–40, 42, 44–48, 60–61 Boarding school, 98–100, 110–112, 117, 159, 175 Bolsheviks, 260, 263 Brazil, 309 British imperialism, 236 British system, 97–98, 236, 317 Bulgaria, 363, 365–366, 377, 402–403 Burkina Faso, 16 Cameroon, 16, 19 Canada, 316, 402–403 Capabilities approach, 151 Capability framework, 151–152 Career, 109, 114–115, 219, 227–228, 235–236, 241, 244, 267–268, 270, 276, 279, 290, 302–304, 310, 314, 320, 323, 325–326, 333–343, 345–352, 400, 417 Career aspiration, 109, 115, 333–341, 343, 345–352 Central Asia (CA), 229, 255–274, 279, 283–284, 286, 293–294 Chile, 365, 377, 403 China, 157, 185–190, 193–195, 207, 211–212, 245, 267, 361 China, urban, 186–187, 190, 193, 207 Chinese education, 211 Civic Education Study (CIVED), 357–359, 361, 363–377, 379, 381, 383–387, 389 Civic engagement, 359–362, 364–367, 369, 371–372, 374, 383, 385–387 Co-educational institutions, 186–187

Bahrain, 311, 402 Behavior, 25, 42, 48–49, 51, 54–55, 60, 98, 125, 138–139, 141, 175, 185–187, 189–191, 193, 195, 197, 199, 201–203, 205, 207–209, 211, 227, 306, 358, 360, 370–371, 417 Belgium (French), 40, 48, 53, 61, 365, 377, 402–403 Beliefs, 39, 46, 49–54, 60, 168, 174, 188, 263, 265, 288, 290, 306, 326, 363–364, 386

433

434 Co-educational school, 186–187, 194–195, 197, 200, 203, 207–209, 211, 376 Co-educational schooling, 99, 113, 117, 186, 376 Colombia, 365, 377, 383, 402 Compulsory education, 106, 186, 222, 260, 270, 334 Conference of African Ministers of Education (COMEDAF), 94 Convention of all Forms of Discrimination against Women, 272–273 Convergence, 1–6, 9, 11–16, 18, 20–23 Corruption, 135, 139 Cross-Generational Relationships, 123–141 Cross-national achievement, 395 Curriculum, formal, 41, 384 Curriculum, hidden, 41, 227 Cyprus, 365, 377, 402 Czech Republic, 363, 377, 403 Democracy, 44, 179, 258, 266, 308, 364 Demographic Health Survey (DHS), 15 Denmark, 377, 403 Dependency theory, 6 Discipline, 53, 61, 115, 139, 188–189, 191–192, 196–202, 208, 225 Eastern Europe, 264, 273–274 Education for All (EFA), 92, 273, 321, 400 Educational access, 123, 125, 127, 139, 265, 302 Educational attainment, 4, 148, 227–228, 271, 336, 339–341, 345–347, 351, 368, 387 Educational borrowing, 309–310 Egalitarian ideologies, 399 Egalitarian standards, 398 Egypt, 305, 308, 311–312, 315, 317–318, 402 Emancipation, 255, 262–263

SUBJECT INDEX Emiritization, 303, 316 Employment, 2, 7, 11, 16, 94, 97, 125, 131, 137–138, 146, 163–164, 170, 173, 227–228, 234, 269, 280, 283–284, 306, 320, 374–375, 379, 381–382, 384 Engagement with teachers, factors inﬂuencing, 186, 187, 192, 197, 201, 202, 207, 208, 210 England, 317, 377, 383, 402–403, 415 English instruction, 191 English language, 194, 284, 287, 315 Enrollment rates, 19–20, 45–46, 125, 146, 148, 217, 242, 274, 327 Equity in the Classroom (EIC), 40, 46, 48–51, 54–60 Estonia, 365, 377, 402 Ethnic communities, 151, 159, 163, 168–170, 173, 176–177 Ethnic groups, 130, 146–151, 153–157, 160–161, 163–164, 168–170, 172, 174, 176–179 Ethnic identities, 153 Ethnic traditions, 145–147, 151–153, 158, 165, 170–172, 176–177 European Union, 400 Examinations (or testing), 90, 98–99, 104, 108–109, 111, 113, 186, 189, 194–196, 203 Family and women in higher education, 234, 242, 307, 318 Feminist theory, 8 Feminization, 24, 169, 321 Fertility, 11, 16, 22, 289 Finland, 377, 402–403 Formal education, 90–91, 93, 269, 305–306, 314, 321 France, 312, 402–403 Gender and Development (GAD), 4, 8, 153 Gender classiﬁcation, 41–42, 59 Gender codes, 41

Subject Index Gender convergence, 2–4, 6, 9, 11–12, 14, 18, 20, 22 Gender development index (GDI), 264, 375–377, 379–383, 385–387, 389 Gender differentiation, 98 Gender effects, 395–397, 399, 401, 403, 405, 407, 409, 411–415, 417 Gender equality, 7, 22, 39, 89–91, 103, 105–106, 109, 116–117, 146, 152, 176–177, 179, 224, 244, 255–256, 258–259, 264–265, 288, 293, 303, 325–326, 359, 368–369, 376, 380, 385, 387, 399, 401, 409, 416 Gender equity, 2–3, 11, 23, 39, 46, 51, 60, 94, 140, 175, 236, 241, 257, 265–266, 271–273, 285, 294, 303–304, 322, 327, 398, 401, 405, 409, 412–414 Gender framing, 41–42, 59 Gender gap, 1–3, 9–12, 14, 16, 19–22, 24, 96, 102, 148, 155, 217–219, 221–225, 227, 229, 231–237, 239–246, 271–272, 333, 335, 351, 358–361, 363–370, 373–374, 376–388, 407–408 Gender Gap Index, 271–272, 333 Gender in Muslim world, 264, 306 Gender parity, 91, 96, 104, 112, 116–117, 146, 157–158, 222–223, 245–246, 274, 333–334, 400, 404, 414 Gender problem, 258 Gender recontextualization, 42, 59 Gender relations, 7, 41, 151–154, 158, 168, 176, 178, 243–244, 275 Gender segregation, 224–225, 233–235, 240, 243, 304, 333, 335–337, 339, 341, 343, 345, 347, 349, 351–352, 405 Gender segregation by ﬁeld of study, 218, 243, 338 Germany, 220, 377, 383, 402–403, 414 Ghana, 16, 19, 403 Girls’ Education, 10, 40–41, 44–45, 49–50, 53–55, 58–59, 95–96, 99, 146, 152, 154, 168–169, 177–179

435 Global Campaign for Education (GCE), 96 Globalization, 11–12, 94, 138, 242, 266, 301–302, 308 Grades, factors inﬂuencing, 200–202, 206–207, 209–210, 215–216 Greece, 365, 377, 403, 414 Gross domestic product (GDP), 9, 265, 407, 409–412 Guinea, 24, 41 Gulf Cooperation Council (GCC), 311 Gulf region, 311, 318, 322, 326 Headscarf, 271 Health care, 149–150, 162–163, 342 Hierarchical linear modeling (HLM), 376, 378, 381, 407, 410–412 Higher education, 93, 190, 217–223, 225, 227–247, 258, 261, 267, 270, 272, 277, 288–289, 291, 302, 304, 306–307, 310–312, 316–318, 321, 326, 335, 342–343, 351, 360, 374, 401, 404–405, 409–413, 416 Higher education expansion in Asia, 218, 243 HIV, 98, 107, 123–127, 130–131, 133–141 HIV/AIDS, 98, 123–127, 130–131, 133–140 Home background, 336, 338–339, 342, 346, 348, 351 Hong Kong, 365, 377, 402–403 Hostels, 110–111, 114 Human capital development, 399, 404, 414 Human capital theory, 265 Human development index (HDI), 98, 264, 376 Human resource development (HRD), 311, 314, 326 Human rights, 93, 265, 268, 271, 273, 400 Hungary, 363, 377, 402–403

436 IEA (International Association for the Evaluation of Educational Achievement), 357–359, 361, 363, 365, 367, 369–371, 373–375, 377, 379, 381, 383, 385, 387, 400–402, 404, 418 Illiteracy, 260, 305 Incorporation of women, 395–396, 398–399, 404, 414, 417 India, 217–218, 223–224, 226, 233, 236–241, 243, 247 Information communication technology (ICT), 93–94 Infrastructure, 5, 17, 45, 149, 161, 313 International development, 95, 152–153, 166, 172 Iran, 402–403, 413 IRT (Item Response Theory), 373, 375, 418 Islam, 266, 268, 271, 290, 301–302, 304–307, 311–312, 319 Islamic, 256, 260, 267–268, 271, 291, 301–306, 311, 315, 327 Islamic revival, 268 Israel, 245, 403, 413–414 Italy, 365, 377, 403 Japan, 217–218, 220–228, 233, 239–240, 243, 245–246, 361, 402–403 Jordan, 308, 311–312, 402 Kazakhstan, 255–256, 261, 266, 268, 271–272, 274, 288, 294 Kenya, 16, 24, 141, 178 Kidnapping, 285–288 Knowledge-based economy, 93 Korea, South, 413 Kuwait, 308, 311, 317 Kyrgyzstan, 255–256, 258–259, 261, 266–268, 271–272, 274, 283–288, 291, 293 Labor market, 5, 11, 95, 165–167, 169, 171, 176–177, 219, 227–228, 234–235,

SUBJECT INDEX 241, 243–244, 259, 265, 273, 333–336, 352, 398, 404–405, 409, 412–413, 415, 417 Labor market and women in higher education, 333–336 Language (of instruction), 48, 53, 97, 158, 170–174, 177–179, 190, 194, 229–230, 234, 245, 257, 260–261, 270, 275, 277–278, 281, 284–285, 287, 289–290, 302, 304–305, 315, 319 Language learning, 190, 194 Language, foreign, 230, 234, 245, 270, 282, 284 Latvia, 245, 377, 402–403 Lebanon, 318, 403 LISREL, 341 Literacy, 39, 91, 101, 173, 256, 260–261, 274–275, 294, 316, 375 Lithuania, 365, 377, 402–403 Madagascar, 16–17, 19 Makerere University, 117 Market economy, 229, 235, 257–258, 266, 290 Marriage, 16, 24, 45, 102, 107–108, 129, 153, 168–169, 175, 177, 218–219, 227–228, 234–236, 241–244, 262–263, 270, 275–276, 280, 286–290, 293–294, 320 Marriage and women in higher education, 234–236, 241–244 Marriage, arranged, 285, 289 Marxism, 44 Mass schooling, 106, 395–396, 399 Math and science performance, 190 Mauritius, 93–94 Meritocracy, 4–5, 9 Mexico, 245, 361 Middle East, 301–302, 305, 313, 326 Middle schools, 185, 187, 190, 193, 211–212, 282 Migration, 126, 160, 166–167, 257, 267

437

Subject Index Millennium Development Goals (MDGs), 24, 89, 96, 106, 274 Millennium Development Project, 9 Misbehavior, 186–187, 191–192, 196–200, 202, 205–209 Misbehavior, factors inﬂuencing, 205–209 Missionary, missionaries, 98 Modernization, 4–7, 9, 25, 220, 301, 303, 307–309, 311, 313–314, 322, 328 Mongolia, 217–218, 223–224, 226, 229–231, 233–236, 240, 243–244, 246 Moral education, 185–191, 195, 208, 210 Muslim, 260, 263–264, 271, 288–289, 301, 304–307, 310, 314–315 Mutawwa (religious schools), 312 National Assessment of Educational Progress (NAEP), 362, 365 National policy-making, 91 National University of Mongolia (NUM), 229–230, 233–234, 246 Nationalism, 91, 257 Neoliberal Economic Policies, 125 New Zealand, 316, 402–403 Niger, 16 No Child Left Behind (NCLB), 307 North America, 316, 319 Norway, 333–335, 339, 345, 352, 377, 402–403 Occupational aspiration, 336, 339 Occupational attainment, 5, 334, 336 OECD (Organisation for Economic Co-operation and Development), 94, 109, 228, 273, 339, 400, 405–406 Oman, 311–312 Open Society Institute (OSI), 266–267, 273 Opportunity structure, 351–352, 371, 383, 395–398, 401, 404–405, 408–409, 414, 416–417, 419

Orphan, 98, 134–135, 139 Out-of-school girls, 146, 155 Parent-Teacher Associations (PTAs), 100 Patriarchy, patriarchal norms, 168–169, 264–265 Pedagogy, 39, 41, 43, 45–47, 49, 51, 53, 55, 57–60, 93, 170, 256, 261–262, 287 PISA (Programme for International Student Assessment), 400 Poland, 377, 383 Political activity, 360, 362 Political attitudes, 359, 361, 369–372, 387 Political efﬁcacy, 357, 368–370, 373, 375, 377–378, 381–386, 388 Political socialization, 357, 359–360, 362–364, 370, 372, 386 Polygamy, 125, 129, 262–263 Portugal, 377, 403 Post-colonial development, 312 Post-colonial education, 39 Post-colonial state, 9, 303 Poverty, 6, 24, 93–95, 101–102, 107, 116, 135, 145–154, 158–159, 161–164, 166, 169, 171, 174–176, 178–179, 257, 270, 276, 280 Pregnancy, 24, 45, 133, 175 Pregnant, 45, 132, 175, 278 Principal Components, 49 Principal components analysis, 49 Private schools, 104, 108–109, 112, 114–116, 126, 186, 317 Prostitution, 124, 136 Punishment, 174–175 Qatar, 308, 311–312 Quran, 305–306, 315, 319 Quranic, 320 Reverse gender gap, 217, 231–235, 243–244, 246 Romania, 365–366, 377, 402

438 Rural, 9, 17, 22, 40, 44–45, 91, 97, 101–102, 104, 107, 109, 113–116, 146, 148, 150, 161, 165, 186, 211, 235, 256–257, 261, 267–269, 275–276, 283–286, 289, 340–341, 345, 348 Russia, 229, 261–263, 266, 274, 282, 289, 292, 363, 367 Russian Federation, 266, 366, 377, 402 Russian schools, 260, 277–278 Saudi Arabia, 308, 311–312, 403 School environment, 40, 90, 96, 101, 109–110, 112, 190, 193, 208, 211 Schooling inequality, 12, 22 Second International Mathematics Study (SIMS), 401–402, 404, 418 Secondary education in Africa (SEIA), 94 Secondary school students, 91, 129–130, 339 Secondary Schooling, 100, 109–111, 113–114, 123, 129, 259 Senegal, 92 Sex, economically rational, 128 Sex, coerced, 127–128 Sex, consensual, 140 Sex, transactional, 124, 128, 130, 136–137 Sex, voluntary, 128 Sexual activity, 128 Sexual behavior, 125, 138, 175 Sexual favors, 111 Sexual relationships, 125, 130–131, 133, 137, 139–141, 175 Sexual resources, 132, 136 Single-sex education, 190 Single-sex school, 185–186, 189, 192–193, 207–208, 304 Single-sex schooling, 185, 189, 304 Slovakia, 377 Slovenia, 365, 377, 402–403 Social construction, 23, 125, 127, 150, 152, 370

SUBJECT INDEX Social reproduction, 4–5, 9 Socialism, 256, 262, 294 Socialization, 185–186, 188–189, 203, 205, 208–211, 218, 227, 323, 357, 359–360, 362–364, 370, 372, 386, 397, 415 Socioeconomic resources, 193, 374 Socioeconomic status (SES), 1, 3, 5, 7, 9–17, 19–25, 407, 409 South Africa, 24, 128, 141, 309, 403 Soviet era, 229, 234, 266, 270, 272, 283 Soviet period, 235, 255, 264, 268, 270, 294 Soviet schooling, 261 Soviet Union, 259–260, 266, 269, 277, 282, 293–294 Structural adjustment policies (SAPs), 124, 139 Structural adjustment programs (SAPs), 16, 124, 139 Structural equation model, 341, 346 Student-Teacher Relationships, 132 Sub-Saharan Africa (SSA), 1–3, 11, 16, 22, 24, 39–40, 59, 93–94, 96–97, 105, 107, 112, 117, 124, 138, 147 Sugar Daddies, 123, 125, 127, 129, 131, 133–135, 137–140 Sugar Mommies, 130–131 Survey Design, 48 Sweden, 363, 377, 402–403, 413 Swedish International Development Cooperation Agency (SIDA), 95 Switzerland, 365, 377, 383, 403, 414 Tajikistan, 255–256, 258, 261, 266–267, 270–272, 274, 288–293 Tanzania, 16 Teacher Education, 303–304, 309–310, 312, 319–320, 322–323, 326, 328 Teachers, 39–61, 93, 99–104, 110–111, 113–115, 125–127, 131–133, 135–137, 139–140, 155–158, 165, 168, 171–175, 178–180, 186–198, 200–203, 205, 207–212, 227–228, 245, 257, 259,

439

Subject Index 261, 267–270, 281, 284–286, 303–304, 309–310, 314–315, 317–318, 320, 323–326, 328, 335, 352, 397, 406 Tertiary education, 92, 218, 221–222, 224–225, 227, 231–233, 235, 238–239, 243, 245, 271–272, 318, 342 Trends in International Mathematics and Science Study (TIMSS), 191, 400–402, 404–406, 414, 416, 418–419 Turkmenistan, 255–256, 259, 261, 264, 266–267, 269–270, 272–274 Uganda, 16, 19, 89–91, 94, 97–109, 111–112, 116–117 UNDP, 124, 148, 160, 173, 264, 273, 400 UNESCO, 9, 89, 92–93, 96–97, 148, 179–180, 220, 223, 226, 233–234, 245–246, 266–269, 273–274, 400, 406 UNICEF, 2, 95, 147–148, 172, 174–175, 179–180, 218, 231, 236, 257, 273–274, 321, 400, 406 United Arab Emirates (UAE), 301–328 United Arab Emirates University (UAEU), 316–318, 321–322 United Kingdom, 318, 323 United States, 189, 208–209, 219–220, 304, 307, 309, 323, 359–363, 365, 367, 370, 377, 384, 404, 413–414 Universal post-primary education and training (UPPET), 89–91, 102–105, 107–110, 112, 116–117 Universal primary education (UPE), 89–92, 95–96, 101, 103–106, 116–117, 274 Universal secondary education (USE), 5, 7, 10, 13, 15, 46, 49–51, 54–55, 57–59, 91–93, 97, 103, 108, 112, 114, 128–129, 139, 151, 161, 164, 166, 170–171, 178, 192, 203, 211, 219, 243, 261, 264, 266–267, 270, 275, 286, 291, 324, 339, 369, 372–373, 376, 378

University of Tokyo, 220, 222 User fees, 124 USSR, 255–257, 261, 264, 268, 275, 278 Uzbekistan, 255–256, 259–261, 266, 268, 270–272, 274–275, 277–278, 281–283, 293–294 Vietnam, 10, 145–149, 153–157, 159, 164, 168, 171, 174–177, 179 Vision 2020 (policy), 1–4, 6–8, 10–13, 18, 20, 22, 24–25, 45, 59, 61, 91, 99–100, 103, 116, 138–140, 150, 153–154, 159, 161, 166, 172–173, 176, 178, 188, 193, 234, 244, 256, 258–259, 262–265, 269, 294, 303, 308, 310, 312, 317, 320–323, 325–326, 351, 399–401, 404, 407, 412–414 Vulnerable groups, 24 Western-style education, 312, 318 Widow, 125, 262 Widowed, 278 Women in Development (WID), 4, 6–9 Women’s enrollment, 218, 225–226, 232–233, 239–240, 242 Women’s rights, 98, 262, 265, 321, 357, 359–360, 366–370, 373, 375, 377–380, 383–386 World Bank (WB), 6, 9, 11, 45, 93–94, 103, 146, 162–163, 174, 179, 220–221, 224, 230–232, 238, 258, 261, 265–266, 273, 283, 400, 406 World Economic Forum (WEF), 94, 271–272 World Values Survey (WVS), 358, 366, 369, 372 Zambia, 16, 123–131, 133–135, 137–141 Zayed University, 318, 322

International Perspectives on Diversity and Inclusive Education

Gender, Equality and Education from International and Comparative Perspectives (International Perspectives on Education and Society, Vol. 10)

International Perspectives on Diversity and Inclusive Education

Education And Gender Equality

Educational Leadership: Global Contexts and International Comparisons (International Perspectives on Education and Society, Vol. 11)

Early Childhood Care and Education: International Perspectives

Education and the Law: International Perspectives

The Worldwide Transformation of Higher Education (International Perspectives on Education and Society, vol. 9)

Inequality in Education: Comparative and International Perspectives (CERC Studies in Comparative Education)

International Perspectives on Citizenship, Education and Religious Diversity

International Relations, Volume 3 (International Perspectives on Higher Education Research)

Education for All, Volume 8: Global Promises, National Challenges (International Perspectives on Education and Society)

Sociolinguistics and Language Education (New Perspectives on Language and Education)

Rethinking rape law: international and comparative perspectives

The Impact of Comparative Education Research on Institutional Theory, Volume 7 (International Perspectives on Education and Society)

Comparative and International Research In Education

Global Trends in Educational Policy (International Perspectives on Education and Society, Vol. 6)

Power, Prestige and Bilingualism: International Perspectives on Elite Bilingual Education (Bilingual Education and Bilingualism, Volume 35)

Creating Classroom Communities of Learning: International Case Studies and Perspectives (New Perspectives on Language and Education)

Second Language Teacher Education: International Perspectives

Multicultural Education: Issues and Perspectives

Second Language Teacher Education: International Perspectives

Second Language Teacher Education: International Perspectives

Perspectives on Participation and Inclusion: Engaging Education

Researching Education: Perspectives and Techniques

Linguistic Perspectives on Language and Education

Linguistic Perspectives on Language and Education

Doctoral Education in Nursing: International Perspectives

Child Perspectives and Childrens Perspectives in Theory and Practice (International perspectives on early childhood education and development)

Feminism And Social Justice In Education: International Perspectives

Religion, Education, and Adolescence: International and Empirical Perspectives (University of Wales - Religion, Education, and Culture)

Contextualizing Inclusive Education: Evaluating Old and New International Perspectives

Gender, Equality and Education from International and Comparative Perspectives (International Perspectives on Education and Society, Vol. 10)

International Perspectives on Diversity and Inclusive Education

Education And Gender Equality

Educational Leadership: Global Contexts and International Comparisons (International Perspectives on Education and Society, Vol. 11)

Early Childhood Care and Education: International Perspectives

Education and the Law: International Perspectives

The Worldwide Transformation of Higher Education (International Perspectives on Education and Society, vol. 9)

Inequality in Education: Comparative and International Perspectives (CERC Studies in Comparative Education)

International Perspectives on Citizenship, Education and Religious Diversity

International Relations, Volume 3 (International Perspectives on Higher Education Research)

Education for All, Volume 8: Global Promises, National Challenges (International Perspectives on Education and Society)

Sociolinguistics and Language Education (New Perspectives on Language and Education)

Rethinking rape law: international and comparative perspectives

The Impact of Comparative Education Research on Institutional Theory, Volume 7 (International Perspectives on Education and Society)

Comparative and International Research In Education

Global Trends in Educational Policy (International Perspectives on Education and Society, Vol. 6)

Power, Prestige and Bilingualism: International Perspectives on Elite Bilingual Education (Bilingual Education and Bilingualism, Volume 35)

Creating Classroom Communities of Learning: International Case Studies and Perspectives (New Perspectives on Language and Education)

Second Language Teacher Education: International Perspectives

Multicultural Education: Issues and Perspectives

Second Language Teacher Education: International Perspectives

Second Language Teacher Education: International Perspectives

Perspectives on Participation and Inclusion: Engaging Education

Researching Education: Perspectives and Techniques

Linguistic Perspectives on Language and Education

Linguistic Perspectives on Language and Education

Doctoral Education in Nursing: International Perspectives

Child Perspectives and Childrens Perspectives in Theory and Practice (International perspectives on early childhood education and development)

Feminism And Social Justice In Education: International Perspectives

Religion, Education, and Adolescence: International and Empirical Perspectives (University of Wales - Religion, Education, and Culture)

Contextualizing Inclusive Education: Evaluating Old and New International Perspectives

Recommend Documents