Abstract
Women remain underrepresented in science and engineering, but whereas they have made steady, if slow, progress toward equal representation in most science and engineering fields, computer science stands out as a glaring exception to the trend. Despite the increasing educational attainment of women and the inroads they have made into other science fields, the representation of women among computer science majors in U.S. universities and in the information technology (IT) workforce has declined steadily since the 1980s. The authors of Gender Stratification in the IT Industry attempt to identify the mechanisms that explain this paradox.
Kenneth W. Koput and Barbara A. Gutek draw together insights from the social network, sex-role spillover and status literatures to identify the interpersonal processes that perpetuate and intensify gender stratification in the IT field. The first chapters present a thorough review of the theoretical and empirical literatures related to social capital, social networks, and how gender roles and social status influence the flow of information in networks to perpetuate status differentials. From this literature the authors derive precise propositions specifying the mechanisms that maintain significant female deficits in participation, employment, and pay in the IT industry. This is a notable contribution to the literature on gender differences in science which often attributes the observed disparities to influences such as discrimination, personal preferences, and social constraints, without specifying the micro-mechanisms by which they operate.
The empirical analysis presented in the book uses longitudinal data for five cohorts of undergraduate students enrolled in a management information sciences (MIS) program at a large university between 2001 and 2003. The data speak only to a subfield of computer science, from a single institution, and for a limited period of the life course, but they include an extensive set of measures not available in any data source that has been used to examine career progress in science and engineering. The data are collected from administrative records and multiple surveys of the students and cover the students’ last two years in the program and their job search experiences. They include rich measures of the students’ social networks, interactions with fellow MIS students, leadership experiences, responsibilities outside of their academic pursuits, sex-role attitudes, career orientations and attitudes, human capital, and job placement.
The authors aim to identify the mechanisms influencing student progress, retention, and job placement in the MIS field, and they use multiple measures to capture the multidimensionality of each outcome. To operationalize retention, for example, they analyze behavioral measures of students’ programmatic participation, including dropout and on-time completion of required courses, as well as attitudinal measures of students’ orientation, including self-reports of commitment to the field and plans to work outside of the field after graduation. The focal explanatory concepts, social capital and sex-role spillover, are also measured multidimensionally. Drawing on their unique multi-year student network data, the authors use the number of social contacts and average cohesion among contacts to operationalize social capital. Sex-role spillover is measured using a combination of attitudinal scales, reports of outside responsibilities, and leadership activities.
The results of the analysis are presented in Chapter Four and discussed extensively in Chapters Five and Six. The statistical models are necessarily sophisticated to accommodate the richness of the predictor and outcome variables and complexity of the processes the analysis aims to identify. For example, in all analyses the focal predictor variables are interacted with gender to assess how social capital and sex-role spillover operate differently for men and women and which outcomes are most affected by those disparities. Despite the complexity, the empirical results are presented in clearly organized summary tables (complete statistical results are presented in appendix tables) and described concisely in non-technical language. The authors consistently focus on the substantive importance of the statistical results, emphasizing their relevance both to the theoretical propositions they aim to test and to our understanding of the processes that influence the cultivation of an inclusive scientific labor force.
The authors find that women are more likely than men to drop out of the MIS program before graduation despite better grades and rates of progress through the program. Among the students who remain in the program, women are nearly shut out of student leadership positions, and they receive fewer job offers and lower starting salaries than men. These results are consistent with extant research—so the real value of this book is in the insights it offers into the network and social capital mechanisms that sustain these disparities. The student networks are acutely segregated by gender and they operate in gendered ways. A primary contribution provided by the study is a detailed analysis of the double-jeopardy women face in professional networks. On the one hand, connections with men are essential for women’s professional success: since men predominate at the core of the networks, connections with them provide essential access to information and legitimacy. On the other hand, connections with men heighten the experience of sex-role spillover among women and thereby jeopardize their continued network inclusion and professional standing. The results of the study illustrate the gendered dynamics of social networks as they operate to influence both the success of the MIS students during their years in the training program as well as their job-search experiences and employment outcomes when they graduate. In the concluding chapters, the authors enhance the discussion of the quantitative results with added qualitative illustrations of the interpersonal interactions that underlie the network mechanisms.
In sum, this is a clearly written presentation of a complex analysis that contributes to multiple theoretical literatures and to our understanding of the mechanisms that perpetuate the underrepresentation of women in science generally and their diminishing representation in information technology specifically. Although the study presented in Gender Stratification in the IT Industry cannot definitively identify the causes of the historical trend, the insights it provides will prompt further research and policy approaches that promise to reverse the trend.