Abstract
Numerous legal scholars and social scientists have highlighted the ways in which research has informed judicial decision making (e.g., Dunn & West, 2008; Morgan & Pullin, 2010; Moses & Marin, 2006). Because, in part, of convincing empirical research presented in several landmark cases (e.g., Grutter v. Bollinger, 2003; Parents Involved in Community Schools v. Seattle School District No. 1, 2007), the consideration of race in educational policies has been deemed permissible, albeit in limited, narrowly tailored ways. Grutter also represented an affirmation of the importance of research for social scientists whose work provided empirical evidence of the educational benefits of diversity and the importance of a “critical mass” of underrepresented students, which served as a basis for the Court’s decision. Although the Court upheld the University of Michigan Law School’s admissions policy in Grutter, the opinion of the Court also stated the expectation that race would no longer need to be considered in a generation’s time. However, as evidenced by the oral arguments of Fisher v. University of Texas, which came just 9 years after Grutter, time may be running short for race-conscious admissions.
In the Fisher oral arguments, the questions did not necessarily focus on diversity—defined as a “critical mass” of underrepresented students—as a compelling interest. Instead, many of the concerns (most often voiced by the Court’s more conservative Justices) centered around two questions: (a) Can a critical mass of racial/ethnic minority members be achieved without considering race? and (b) How will we know that a critical mass has been achieved and, thus, that the endpoint for the consideration of race has been reached? Both of these questions concern the narrow tailoring of holistic admissions policies in which race is considered as a factor (see footnote 2 in Garces, this issue, for further discussion of narrow tailoring). The intense focus on these two questions suggests that we, as social scientists, need to be more effective and strategic in our efforts to communicate findings from existing research on the effects of race-neutral policies on access and success for underrepresented populations. We should be prepared as well to conduct further inquiry into what the indicators of a critical mass are across a range of institutional contexts.
As scholar-advocates whose previous work has focused on broadening participation in science, technology, engineering, mathematics (STEM), and related fields, we argue that social scientists ought to look to the vast STEM education research literature to begin the task of empirically investigating the questions raised in the Fisher case. In the analysis that follows, we provide our answers to the two questions above, based on research and evaluation of current programs and institutional efforts aimed at broadening participation in STEM.
With regard to the first question, whether a critical mass can be achieved without considering race, “we doubt it.” There have been numerous efforts to find proxies for race such as socioeconomic status (SES) or first-generation status. Across higher education as well as within STEM, these proxies have not represented race and ethnicity. In STEM, this is evidenced by the declines in underrepresented minority student enrollment in STEM graduate degree programs that followed the implementation of race-blind admissions policies, and a failure to regain previous levels in the subsequent years (Garces, 2013; Malcom, Van Horne, Gaddy, & George, 1998).
With regard to the second question, how can one know when a critical mass has been reached, “when diversity is self-sustaining.” Decades of scholarly research and programmatic evaluations aimed at understanding the factors that contribute to diversified STEM programs indicate that sustainable diversity results from environmental changes—that is, changes in culture, curricula, and instruction; quality and quantity of supportive practices; and faculty behaviors, attitudes, and expectations)—that support the success of all students (see Chubin, DePass, & Blockus, 2009; DePass & Chubin, 2008; Fox, Sonnert, & Nikiforova, 2009; Margolis & Fisher, 2002; Maton & Hrabowski, 2004).
Learning From the Worst Case
In most research universities in the United States, the least racially and ethnically diverse classes are those on the science and engineering side of campus. In the last decade, Latinos have seen progress in many STEM fields. For example, in physics, engineering, and mathematics, three fields that are among the least diverse, Latinos increased their share of bachelor’s degrees between 2001 and 2010. In 2001, Latinos earned 4.4% of undergraduate degrees in physics, 6.8% of such degrees in engineering, and 5.5% of bachelor’s degrees in mathematics and statistics. By 2010, these figures had risen to 5.2%, 8.0%, and 6.0%, respectively (National Science Foundation, 2013). Nevertheless, Latinos remain woefully underrepresented compared to their presence within the overall or college age population. Since 2001, African American participation in these same fields has declined. In 2001, African Americans earned 3.9% of bachelor’s degrees in physics, 4.9% of bachelor’s degrees in engineering, and 7.1% of bachelor’s degrees in mathematics and statistics. By 2010, the share of bachelor’s degrees in these fields earned by African Americans had fallen to 2.9%, 4.1%, and 5.0%, respectively.
Research universities have incredible assets to support study in these fields by underrepresented minority students. However, their efforts often fall short. In physics, for example, when the contributions of minority-serving institutions are subtracted from the national degree numbers, the paucity of research universities’ contributions to physics degree production by underrepresented minorities is revealed (Mulvey & Nicholson, 2012). The challenges associated with broadening participation in STEM are shared by the University of Texas at Austin, as evidenced by data on degree production in these fields (U.S. Department of Education, 2012). Thus, the four decades of experience in developing and assessing the effectiveness of interventions to change participation levels in STEM can inform the considerations as the Court looks at the relative merits of the arguments in the Fisher case.
Race-Neutral Policies: STEM as a Natural Experiment
For more than 40 years, the American Association for the Advancement of Science (AAAS) has undertaken research, developed programs, and framed policy options to support diversity in STEM and related fields. Although gains during this period have been notable in many cases (e.g., women now representing nearly half of graduates from medical schools and PhDs in the biological sciences [when they received only about 16% of MD degrees in 1975–1976]; Jolliff, Leadley, Coakley, & Sloane [Association of American Medical Colleges], 2012), this has not been seen in all fields (e.g., low levels of diversity that remain in mechanical engineering and computer science) nor for all groups (e.g., African Americans and Latinos underrepresented in all areas of the natural sciences and engineering).
In 1997, following mounting concern about the effects of judicial rulings (e.g., Adarand Constructors v. Pena, 1995; Hopwood v. Texas, 1996), legislative referenda on graduate enrollments in STEM fields, and declines in underrepresented minority presence in medical and law schools, the AAAS in partnership with the Council of Graduate Schools undertook a study of a sample of highly selective research universities to determine if admission to STEM graduate programs was following the same downward trends being seen for professional programs. Between 1996 and 1997, a 1-year decline of 20% was seen for African American 1st-year enrollees and a more than 16% decline was seen for Hispanic students. Visits to a sample of these campuses revealed that confusion regarding legal options had made universities cautious, leading many to abandon special initiatives that had helped fuel earlier increases.
Program advocates welcomed the ruling in Grutter in 2003 because it provided more clarity about what was and was not allowed and reinforced Justice Powell’s views in Regents of the University of California v. Bakke (1978) that race could in fact be used as one of many factors in admissions. In STEM, however, issues beyond “being admitted” were critical in affecting participation and success of underrepresented minority students. We awaited guidance from the Bush Administration with respect to the actions that could be undertaken by higher education institutions in support of underrepresented minorities into STEM. The “guidance” that came related only to so-called race-neutral strategies.
Once again, the AAAS undertook a series of projects to offer further clarity to universities, given the specific needs and challenges in STEM. Researchers and attorneys undertook joint initiatives to lay out possible pathways to navigate these complex issues, helping the STEM community understand and communicate what might be programmatically effective and legally defensible. In the years following Adarand and even post-Grutter, institutions responded to the uncertainties around special programs and fear of legal challenge by withdrawing efforts that had previously been used to help them achieve diversity in STEM programs. This was done despite Justice O’Conner’s opinion in Grutter that included consideration of the use of race as a component of holistic review in order to achieve the compelling interest of diversity. The net effect of these actions was to give us the natural experiments of looking at the impact of race-neutral approaches to achieving diversity in STEM. The results of these approaches are documented in earlier work by Malcom et al. (1998) and more recent work by Garces (2013). In both instances, we note loss of ground and failure to recover.
Critical Mass: More Than Just a Number
As we note above, one of the primary questions raised by the Court in the Fisher oral arguments moved beyond the value of “critical mass” to how one might know when critical mass was reached. How much diversity is enough? What ought the unit of analysis be when considering whether critical mass has been achieved? Is it at the institutional level? Departmental? Classroom? Although the respondents’ counsel attempted to communicate that critical mass is contextually determined, more research is needed to uncover how we can know when it has been achieved within a specific context. Again, the research and evaluation literature on broadening participation in STEM fields can provide some guidance on these questions.
The research and evaluation of STEM programs demonstrate that building and sustaining a critical mass of underrepresented minority students requires actions beyond admission: a welcoming environment, supportive mentors, and the construction of a supportive research community. Where these elements have been realized, we can point to departments that have experienced considerable success in STEM diversity, such as for African Americans in chemistry at Louisiana State University (Collins, Stanley, Warner, & Watkins, 2001; Warner, 2006; Wolf, 2011), minorities in computational science at Rice University (Konisky, 2000), and minorities and women in applied physics at the University of Michigan (Kurdak, 2011). Although the field continues the long process of documenting this type of cultural change and understanding the conditions under which it occurs, there are important lessons that can be gleaned.
Conclusion
The ongoing of story of STEM, and our successes and failures to broaden participation, can act as an informative case study for the broader context of higher education. Decades of research and evaluation by the STEM education community underscore that building a critical mass matters. This work also provides clues as to what effectively builds and sustains diversity. Considering race is important to build a critical mass, but a holistic and systemic approach is necessary to ensure that this critical mass is sustained. It is up to social scientists to apply these lessons to ongoing and future cases regarding the consideration of race in higher education.