Scientists ’ Participation in University Research Centers : What are the Gender Differences ? Elizabeth Corley

University-affiliated multidisciplinary research centers have grown in importance in academia. Most research to-date has focused on these centers from an institutional perspective, with recent work only beginning to explore the ways in which such centers affect the development of academic careers. Hence, little is known about how scientists who are centeraffiliated differ from those who are not affiliated. Clearly, both selection and influence effects may be expected to operate in terms of research productivity, timing, and resources. A further puzzle is how center affiliation may differ between male and female scientists. In this study, we use a new, nationally representative dataset of scientists and engineers working in Carnegie Research Extensive universities to develop an understanding of how center-affiliated scientists differ from exclusively department-based academic scientists and engineers, and investigate the extent to which gender moderates the effects of centers. As expected, our national sample shows that women are younger, whiter, less likely to be tenured, and at a lower rank than their male colleagues. We find that women are as likely to join centers as men, and do so at a similar stage in their career. Most of the male–female differences observed in disciplinary settings are sustained in centers, but women appear to have greater research equality in them (compared to the departmental setting). In particular, men and women in centers spend the same amount of time writing grant proposals, conducting both grant-supported and unfunded research, and administering grants. This suggests that centers may constitute an institutional context in which some aspects of gender equity in science may be achieved. JEL Classificaton: C42, O32, Z13, E61 1. At the forefront of US production of scientists and engineers The development of national research infrastructure is central to the American innovation system, and the Research Extensive universities are a crucial component of that engine of growth (Crow and Bozeman, 1998). A large body of work has demonstrated that female academic scientists remain disadvantaged in their access to research resources, and the rewards that attend them such as productivity and career progression. However, this work has focused almost exclusively on scientists in traditional academic departments. A question we believe has received insufficient attention is what role the development of university-based research centers, in which 40% of academic scientists and engineers now work, plays in the career prospects and patterns of underrepresented groups. In this paper, we seek to explore how centers and departments differ in creating contexts for female academic career success. Our own earlier work leads us to expect that centers create different opportunity structures for female scientists; in this study, we use a better sampling strategy to investigate our earlier findings (Gaughan and Bozeman, 2002; Corley et al., 2003; Bozeman and Corley, 2004). We briefly recapitulate the findings about women academic scientists, bearing in mind that this special issue addresses the literature comprehensively. We focus especially on university-based research centers as new organizational contexts for scientific career development. We find there are few studies that link the direct effects of universitybased science centers on academic scientists, or that provide clues to how faculty members perform in them. There is enough work, however, to use the two literatures to generate hypotheses to evaluate men’s and women’s experiences in university-based science centers at Research Extensive universities. We then test these hypotheses using data from the 2004 Survey of Academic Researchers completed by the Research Arizona State University Georgia Institute of Technology, School of Public Policy, 685 Cherry Street NW, Atlanta, GA, 30332 E-mail: [email protected] Journal of Technology Transfer, 30: 371–381, 2005 2005 Springer Science+Business Media, Inc. Manufactured in The Netherlands. Journal : JOTT Dispatch : 29-6-2005 Pages : 11 CMS No. : h LE h TYPESET MS Code : 30404 h CP h DISK 4 4 U N C O R R E C TE D P R O O F Value Mapping Program at Georgia Tech (Barry Bozeman, PI). Finally, we draw some general conclusions about how our results may inform science policy within university-based science centers. 2. Academic women in science By the 1990s the promise of affirmative action to redress gender inequity in universities had largely been realized (Cole, 1979; NRC, 1987, 2001). Yet women remain under-represented in the senior tenured ranks, and over-represented in off-ladder positions (Barber, 1995; NRC, 2001). Early work on scientific careers evaluated how science is socially stratified by a variety of ascriptive and acquired traits (Cole andCole, 1973). This generated a stream of research that evaluated the determinants of women’s lower academic career success (Long and Fox, 1995). The university remains an employment sector that is highly sex segregated, with negative consequences for the women in it (Reskin, 1978; Bielby and Barron, 1986; Jacobs, 1996). Xie and Shauman (2003) attribute gender differences in outcomes to gender differences in personal and structural characteristics that bear on typical scientific output measures such as publication productivity and research resources. Numerous studies have been conducted that elucidate some of the individual-level mechanisms for these differential career outcomes. For example, women have differential access to resources during training (Reskin, 1978; Fox, 1995), and experience less mentoring and collaboration during their career (Long and McGinnis, 1985). Also, they tend to be less productive, which has negative impacts on career progression (Long et al., 1993). Overall, women are thought to benefit less from organizational factors that improve productivity (Allison and Long, 1990; Long and McGinnis, 1991). On the job, female academics are more aware of and sensitive to the presence of organizational constraints (Fox and Ferri, 1992). These constraints tend to reduce productivity and collaboration activity of female scientists (Fox, 1991). Given what is known about women in competitive academic science at universities, it is important to investigate how variations in the institutional context itself affect career development. Center affiliation is an important new basis of structural location in the academic structure, one we know little about in its place in the development of the academic career in general, or gender differences in particular. If center researchers are more productive, then centeraffiliates can be expected to benefit from the same contextual boost that Allison and Long (1990) found in departments. University-based research centers Since the 1980s interactions between industry and universities have become increasingly important in the development of R&D conducted at US universities (Gray et al., 2001). According to Cohen and colleagues (1994) about 70 percept of industry’s support for academic research is channeled through roughly 1100 industry–university centers that they identified in 1993. As Boardman and Bozeman (forthcoming) have noted, in the US the past three decades could be credibly called the ‘‘era of inter-institutional research collaboration.’’ Increasingly, US science funding agencies are supporting centralized, interdisciplinary research centers that represent a different institutional form from the decentralized, individual-investigator research of the past. Bozeman and Boardman (2004) use a detailed analysis of the National Science Foundation (NSF) Engineering Research Centers (ERCs) to demonstrate how the rise of university-based science centers has led to the development of a new institutional form for the execution of university-based research. The National Science Foundation’s Industry/ UniversityCooperativeResearchCenters (IUCRC) Program, which originated in 1973, is one of the oldest government initiatives focused on cooperative research across the industry and university sectors (Hetzner et al., 1989). The purpose of the NSF IUCRC program is to strengthen the relationship between industrial firms and universities, especially colleges of engineering (Adams et al., 2001). In fiscal year 2000, the NSFs contribution to the centers was about $5.2 million; however, industrial and other external support of the centers totaled about $68 million in FY 2000 (National Science Foundation, 2005). Currently there are approximately 50 IUCRCs, all of which are administered by the Engineering Education 372 Corley and Gaughan Journal : JOTT Dispatch : 29-6-2005 Pages : 11 CMS No. : h LE h TYPESET MS Code : 30404 h CP h DISK 4 4 U N C O R R E C TE D P R O O F and Centers Division at the National Science Foundation. Managers of university-based science centers face many challenges that are different from the challenges faced by traditional, academic department directors: namely, how to manage research within a new organizational structure that involves researchers from different disciplines, who bring diverse collaboration incentives and research goals to center-based research (Boardman and Bozeman, forthcoming). Even though there is some early recognition that university-based science centers represent a new institutional form for the performance of scientific research (Bozeman and Boardman, 2003), few studies have focused on exploring the effects of the new institutional form on individual scientists careers. In fact, many of the past studies on university-based science centers have concentrated on centers that explicitly involve university–industry collaborations (Jaffe, 1989; Rees, 1989; Cohen et al., 1994; Rosenberg and Nelson, 1994; Mansfield and Lee, 1996; Narin et al., 1997). Many focus on the study of industrial-university cooperative research centers (IUCRCs), which are defined as ‘‘small academic centers designed to foster technology transfer between universities and firms’’ (Adams et al., 2001, p. 73). The literature in this area, however, largely explores how IUCRCs affect organizations that participate in the centers—i.e., universities or industrial firms (for example, Blumenthal et al., 1986; Geisler et al., 1991; Mowery, 1995; Cohen et al., 1998; Santoro and Chakrabarti, 1999; Adams et al., 2001; Gray and Steenhuis, 2003)— and not on the impacts of center affiliation for individual faculty participants. Although the large majority of studies on IUCRCs focus on organization-level units of analysis, there are a few scholars who have attempted to determine (in a general way) how IUCRCs might affect individual researchers. Cohen et al. (1998) studied the impact IUCRCshave onuniversity-level research. They argued that the rewards from university research traditionally come from reputation—largely because reputation promotes faculty mobility, and mobility leads to salary increases and teaching reductions. Therefore, Cohen and colleagues argued that the rewards of academic research depend largely on the dissemination of research findings and open access to science results. Hetzner et al. (1989) also briefly mention the issue of individual faculty participation in IUCRCs. They report that faculty members expect research funding to flow from their participation in IUCRCs—especially because of the excessive proposal writing requirements placed on them as members of these centers. Additionally, Hetzner and colleagues say some faculty report that participation in centers leads to more interactions with other faculty, more support for student research, greater access to equipment, and improved consulting and research opportunities. Hetzner and colleagues conclude, therefore, that these centers appear to be supporting some of the traditional core of faculty activities. We believe that science and engineering research centers truly are a new organizational form in academe—and that researchers participating in these centers experience new challenges (and, hopefully, new advantages) as a result of their center affiliation. Prior research informs how science centers can be more effective and/or productive, but the impact that those centers have on individual researchers’ careers—especially women—is not developed. Since there is little published research in the impacts of center affiliation on individual faculty careers (in general) and female faculty (in particular), we use the intersection of the two literatures to create some new hypotheses for the research presented here. Women and centers: unanswered questions In this paper, we explore how the recent setting of university-based science centers may have differential impacts on (1) scientists in general and (2) female scientists in particular. In general, we hypothesize that university-based research centers expose scientists (of all genders, ethnic groups, and levels of seniority) to greater opportunities for research. Therefore, we believe that scientists conducting research through university-based research centers will have access to more research resources than those not participating in centers and have more time for research relative to other academic obligations. We also expect to see differences in the advantages (and disadvantages) that women and men receive from their center affiliations. In particular, we hypothesize that women scientists are more 373 Scientists’ Participation in University Research Centers Journal : JOTT Dispatch : 29-6-2005 Pages : 11 CMS No. : h LE h TYPESET MS Code : 30404 h CP h DISK 4 4 U N C O R R E C TE D P R O O F likely to be affiliated with university-based science centers, are more likely to be disadvantaged by carrying heavier service obligations because of their center participation, and in general receive fewer individual benefits of their center affiliation (such as satisfaction with job and feeling appreciated by colleagues). Stated formally, we make the following hypotheses: H1 Center-based scientists have access to more research resources than department-based scientists. H2 Center-based scientists have more time for research relative to other academic obligations. H3 Female scientists are more likely to be affiliated with university-based science centers. H2 Center-based scientists have more time for research relative to other academic obligations. H4 Female scientists are more likely to be disadvantaged by carrying heavier service obligations because of their center participation. H5 Female scientists are more likely to be disadvantaged by carrying heavier service obligations because of their center participation. 3. Research design and data collection Earlier work that we have conducted on centerbased researchers revealed no statistically significant differences between male and female researchers on likelihood of receiving grants, amount of grants, and publications (Corley et al., 2003). We believe these findings are partially the result of the research design we employed. First, only researchers at Energy Research Centers and National Science Foundation Science and Technology Centers were included in the study. Second, only 13% of the study sample was female (which was, nevertheless, representative of those centers). We believe that gender-selection dynamics had already played out in the careers of these high-achieving female scientists in especially male-dominated disciplines. In the first study, we observe them after they have joined a multidisciplinary center, and we did not collect data that would have allowed us to compare them to exclusively department-based researchers. In the current study, we solve many of the design limitations of the original center study. Rather than collect data from center-based researchers, we collect data from university-based researchers, many of whom are also center-affiliated. First, we study a representative sample of scientists and engineers in Carnegie Extensive Universities, allowing us to compare center-affiliated researchers with those who are not affiliated. Second, we extend the scope of our sample to disciplines likely to yield Center affiliations that are not exclusive to Energy Research Centers and NSF S&T Centers. Finally, we over-sample women scientists to ensure that there is sufficient sample size to understand gender dynamics—both between men and women, and within gender groups. The data for the current study come from the 2004 Survey of Academic Researchers completed by the Research Value Mapping Program (Barry Bozeman, PI). Whereas the first study administration was targeted to center-based researchers, the 2004 study administration targets the population of scientists and engineers in tenure track (or tenured) academic positions at Carnegie Extensive universities (formerly known as Research I; Carnegie, 2000). Using the Carnegie list, we retained all universities (n=150) that produced at least one Ph.D. in 2000 in at least one of 13 science and engineering disciplines. We excluded health sciences and economics from the National Science Foundation definition of science and engineering (NSF, 2000), and we subdivided engineering into five major specialties. The resulting disciplines include: biology, computer science, mathematics, physics, earth and atmospheric science, chemistry, agriculture, sociology, chemical engineering, civil engineering, electrical engineering, mechanical engineering, and materials engineering. Having delineated the target population of universities and disciplines, we then collected the names of tenure-track faculty in each university by discipline. The list of faculty was obtained from (1) the on-line university catalog, or (2) the on-line departmental website. This resulted in a sampling frame of 36,874 scientists and engineers occupying a tenure track or tenured faculty position. The target sample was for 200 men and women from 374 Corley and Gaughan Journal : JOTT Dispatch : 29-6-2005 Pages : 11 CMS No. : h LE h TYPESET MS Code : 30404 h CP h DISK 4 4 U N C O R R E C TE D P R O O F each of the 13 disciplines. Because the size of disciplines varies, as does the representation of women in each discipline, the sampling proportions varied from 0.21 (for women in biology) to 1.0 in five disciplines (e.g. the ‘‘sample’’ is actually a census of the women in the discipline). Men’s sampling proportions varied from 0.06 in biology, to 0.23 in agriculture. The final target sample (accounting for women representing fewer than 200 in the discipline) was 4916. Therefore, the 2004 Survey of Academic Researchers is unique in its ability to determine the representation of researchers in university-based research centers, and to make comparisons between those affiliated with them, and scientists who are not. In addition, the over-sampling of women means that gender differences can be evaluated by center affiliation and discipline. The survey was administered by mail, focusing in particular on the following domains of faculty activity: funding, collaboration, institutional affiliations, and career timing and transitions. The survey also obtained basic demographic information about the researchers, their research-specific motivations and values, and the perceived benefits derived from their work. In this study, we focus on research resources and benefits, and teaching and service burdens as characteristics that we hypothesize differ between center affiliates and non-affiliates, and between men and women. We obtained 1769 survey responses from scientists and engineers who were in tenure track positions, had an earned Ph.D., and who had complete information on center participation. There are 916 female respondents and 853 male respondents. There are 693 center-based researchers, and 1076 researchers who report departmental affiliations only (note, all centerbased researchers also have a departmental affiliation).