As protests by students of color across the nation bring attention to the isolating experiences and negative racial climate they face on college campuses, the U.S. Supreme Court is debating for a second time in Fisher v. University of Texas (Fisher II)1 the constitutionality of race-conscious admissions policies.2 Intended to maintain racial diversity on college campuses, the race-conscious admissions policy at issue in Fisher II involves a holistic review process that considers the identity and qualifications of each student. In practice, the policy involves knowing all aspects of an individual’s identity, including her racial or ethnic background, during the admissions process.3 Yet, despite the very limited consideration of race in this process, the policy continues to be subject to controversy and litigation.
As this Essay goes to print, the U.S. Supreme Court has heard oral argument in the case but has not yet issued a decision. Justice Scalia, who heard oral argument in the case, has also since passed away. With now only seven justices voting in the case4 and two three-justice blocks on either side of the issue,5 Justice Kennedy’s vote has added significance.6 Whereas in Fisher v. University of Texas (Fisher I)7 Justice Kennedy’s agreement with the liberal block would have only resulted in a tie, his vote in Fisher II could potentially affirm the lower court’s ruling, setting a precedent for the rest of the nation.8 This Essay considers the lessons social science research brings to the constitutional inquiry in the case, and to Justice Kennedy’s decisive vote in particular. These are important lessons to consider, as the Court’s decision will have long-term ramifications on the ability of postsecondary institutions to improve the campus racial climate for students of color and to further the educational benefits of diversity for all students.
In past cases, Justice Kennedy has acknowledged that the “enduring hope is that race should not matter; the reality is that too often it does.”9 Even with this understanding, however, Justice Kennedy remains skeptical of the limited consideration of race in admissions.10 In this Essay, I argue that, in resolving the constitutional issue in the case, Justice Kennedy’s understanding of the reality that race matters needs to fully reflect the ways race matters and the harm that takes place when race cannot be directly addressed in the admissions process, even in the limited fashion required by precedent. In Part I, I demonstrate that these considerations, which social science research informs, are relevant to Justice Kennedy’s doctrinal inquiry in the case.
The Court emphasized in Fisher I that institutions must consider race in admissions decisions in a limited manner to satisfy the narrow tailoring requirement.11 Research shows that additional restrictions on the ways in which race can be considered would inhibit colleges’ and universities’ ability to realize the educational benefits of a racially and ethnically diverse student body.12 That is because bans on race-conscious admissions practices have led to declines in student body racial diversity (Part II), can inhibit institutions’ ability to support the success of all students on campus (Part IVII), and are associated with a less welcoming campus climate for students of color and with perceptions of a lack of institutional commitment to diversity (Part IV). These factors can limit interactions among students from different racial and ethnic groups that are needed to realize the educational benefits of diversity.
Justice Kennedy, a decisive vote in Fisher II, recognized the educational benefits of student body diversity as a compelling interest in K-12 and postsecondary education, as he recognized the need to avoid racial isolation in schools.13 However, Justice Kennedy voted to strike down earlier race-conscious admissions policies and race-conscious student assignment policies on the grounds that the way race was considered by the University of Michigan in Grutter v. Bollinger14 and the school districts in Parents Involved v. Seattle School District No. 1, was not narrowly tailored and thus was unconstitutional.15 As I note below, Justice Kennedy’s doctrinal considerations in past cases suggest that research findings on the harms or benefits of racial classifications are relevant to determinations of whether the policy survives strict scrutiny.
Justice Kennedy’s doctrinal analysis, outlined in great detail in Parents Involved, involved the need to balance between two principles: liberty (personal expression free of racial categories) and equality (opportunity not denied on account of race). On the question of liberty, Kennedy stated that classifying someone on the basis of race has the potential to take away her dignity and render her a less free individual. In Parents Involved, for example, he noted that the “[r]eduction of an individual to an assigned racial identity for differential treatment is among the most pernicious actions our government can undertake.”16 Racial classifications are pernicious, he argued, because the state (in this case, the public university) “must first define what it means to be of a race,” and such labeling, which an individual is powerless to change, “is inconsistent with the dignity of individuals in our society.”17 Justice Kennedy’s view, moreover, is that the practice “can lead to corrosive discourse, where race serves not as an element of our diverse heritage but instead as a bargaining chip in the political process.”18
On the question of equality, Justice Kennedy argued that as a nation we still face racial injustice. Disagreeing with the more conservative justices on the Court, he stated in Parents Involved that:
Today we enjoy a society that is remarkable in its openness and opportunity. Yet our tradition is to . . . confront the . . . injustices that remain. This is especially true when we seek assurance that opportunity is not denied on account of race. The enduring hope is that race should not matter; the reality is that too often it does.19
Thus, Justice Kennedy said he would allow for some consideration of race in educational policies. However, rather than considering race on an individual basis, the less harmful—and constitutionally permissible—path would involve classifications that consider race in a “general way”—that is, not at the individual level.20 In Parents Involved, for example, he noted that such general approaches to addressing the problem involves (1) strategic selection of new school sites; (2) drawing attendance zones with a general recognition of neighborhood demographics; (3) allocating resources for special programs; (4) recruiting students and faculty in a targeted fashion; and (5) tracking enrollments, performance, and other statistics by race. In Justice Kennedy’s view, these measures are less harmful because they “do not rely on differential treatment based on individual classifications”21 and instead “address the problem in a general way and without treating each student in different fashion solely on the basis of a systematic, individual typing by race.”22
Embedded within Justice Kennedy’s balancing of the two constitutional principles at issue (liberty and equality) is a cost/benefit analysis: Do the costs of the racial classification (potential harm to the dignity of the individual) outweigh its benefits (furthering principles of equality)23? Questions about the costs of racial classifications, or its benefits, of course, are those that research can directly inform. With respect to the perceived harms of racial classifications, research shows that an individualized racial classification is not inherently harmful.24 As extensive work has documented, classifications based on race may be necessary to address the ways race influences students’ educational experiences.25 For example, acknowledging or affirming group membership can motivate students of color.26 For these reasons, Justice Kennedy’s underlying concern about the harm caused by individualized racial classifications is not consistent with empirical findings.
Research further shows that because individuals often rely on racial categories when making social judgments, whether they intend to or not or are aware of doing so, removing a racial classification would not eliminate the racially biased judgments individuals make.27 Indeed, Justice Ginsburg’s rationale in her dissent in Fisher I reflects these lessons of social science.28 Like Justice Kennedy, Justice Ginsburg recognizes that race matters in shaping future opportunities.29 Unlike Justice Kennedy, however, she advocates for a candid recognition of race in admissions policies,30 a recognition that can help counter the negative effects of subconscious racialized judgments. As Justice Ginsburg stated, “[Policies] that candidly disclose their consideration of race [are] preferable to those that conceal it.”31 In Justice Ginsburg’s view, which she expressed in her dissent in the 2003 case Gratz v. Bollinger, “the Constitution, properly interpreted, permits government officials to respond openly to the continuing importance of race.”32 Justice Ginsburg’s understanding of the ongoing importance of race acknowledges that “[b]ias both conscious and unconscious, reflecting traditional and unexamined habits of thought, keeps up barriers that must come down if equal opportunity and nondiscrimination are ever genuinely to become this country’s law and practice.”33 The approach Justice Ginsburg endorses thus reflects a more accurate understanding of the social construction of race and its role in the experiences of students of color.
As I explain next in Parts II–IV, eliminating a direct focus on race, either by prohibiting the consideration of race as a factor altogether or requiring the “generally” race-conscious approach that Kennedy has endorsed in past cases, can perpetuate racial inequities. Such harms to furthering principles of equality should be part of the cost/benefit analysis embedded in Kennedy’s strict scrutiny analysis in Fisher II.
Research studies demonstrate the harmful consequences of further restricting the consideration of race in postsecondary admissions decisions beyond the limits already required by Court precedent. Eight states have banned race-conscious admissions policies, six via ballot measures (California, 1996; Washington, 1998; Michigan, 2006; Nebraska, 2008; Arizona, 2010; and Oklahoma, 2012), one by executive order (Florida, 1999), and another by legislative act (New Hampshire, 2011).34 Studies of the impact these bans have had document a decline in racial diversity at selective undergraduate institutions and in graduate and professional programs, particularly in the sciences, business, medicine, and law. These declines have occurred despite sustained efforts to increase racial diversity at colleges and universities in these states. The substantial decline in student body racial diversity has made it especially difficult for postsecondary institutions to live up to the declaration in Grutter v. Bollinger that, “[i]n order to cultivate a set of leaders with legitimacy in the eyes of the citizenry, it is necessary that the path to leadership be visibly open to talented and qualified individuals of every race and ethnicity.”35 Educational quality has been compromised for all students at the institutions where these bans are in place, not only in the classroom but also in students’ increasingly limited opportunities to interact with racially diverse peers, and thus to acquire the skills they need to flourish in an increasingly diverse workforce.
A number of studies document the decline in racial diversity at selective colleges across the nation that are no longer permitted to consider race as a factor in admissions. In a study of the impact in California, Florida, Texas, and Washington State, economist Peter Hinrichs found that bans at the most selective institutions (top 50 universities listed in the 1995 US News & World Report college rankings) in these states led to a 1.74 percentage-point decline in African American enrollment, roughly a 2.03 percentage-point decline in Latino enrollment, and a .47 percentage-point decline in Native American enrollment.36 Because of the small percentage of students at these institutions who are African American (5.79 percent), Latino (7.38 percent), and Native American (.51 percent), these changes in enrollment are very large in relative terms.37 Other studies have found a similar decline in African American and Latino enrollment at the most selective institutions in these four states.38
After Proposition 209 prohibited race-conscious admissions decisions in California, racial diversity declined significantly at the University of California (UC) flagship campuses.39 From 1997 to 1998, enrollment of African American freshman at UC Berkeley declined by 53 percent, while the number of Latino enrollees fell by 45 percent.40 In the same period at UCLA, African American enrollment dropped by 38 percent and Latino enrollment by 30 percent.41 Although demographic shifts have resulted in a modest recovery in absolute numbers since that time, especially for Latinos, neither campus has regained the diversity it had in 1995. Moreover, the ban on the consideration of race as one factor among many in admissions decisions has shifted Latino and African American students to less selective colleges, which limits their educational and future career opportunities.42
In 2006, Michigan passed Proposal 2, a ballot initiative that banned race-sensitive admissions policies at public postsecondary institutions in the state.43 Despite its best efforts to maintain racial diversity, the University of Michigan, Ann Arbor, experienced a sharp decline in the enrollment of students of color after Proposal 2. From 2006 to 2012, the proportion of African American undergraduate students enrolled at the University’s Ann Arbor campus declined by about 25 percent,44 and the proportion of Latino students by 24 percent.45
Bans on race-sensitive admissions policies have also led to a substantial decline in the racial diversity of graduate programs in the science fields that are critical to continued scientific and technological advancement. The bans in four states (California, Washington, Texas, and Florida) have led to a 26 percent drop in the percentage of engineering graduate students who are Latino, African American, or Native American,46 a 19 percent decline in the natural sciences,47 a 15.7 percent decline in the social sciences,48 and an 11.8 percent drop in the humanities.49
Given the already minimal representation of students of color in these graduate fields, these declines can have significant consequences for the educational experiences of all students in the programs.50 A decline in diversity reduces the variety of perspectives available to foster innovation, tackle complex research problems, and advance scientific inquiry, particularly in fields such as engineering and the natural sciences.51 Moreover, this decline is occurring at a time when the United States urgently needs to produce one million more science, technology, engineering, and mathematics graduates to meet workforce projections and to keep America internationally competitive.52 Meeting this need is being undercut by the drop in student body diversity in the critical technological and security-sensitive science fields. The decline in student diversity can also have long-term effects on faculty diversity across all of these fields, as graduates enter the academic job market.
The United States is facing a real crisis in the health and health care of its racial and ethnic minorities. The disparities in this area are widely documented.53 The Department of Health and Human Services, for example, has shown that patients of color suffer disproportionately from numerous health conditions and are underserved in terms of quality and frequency of care.54 A substantial body of research has found that a racially and ethnically diverse medical workforce could help address these disparities by providing more positive interactions between patients and health care professionals, and greater access to health care for our increasingly diverse and underserved populations.55 Studies show, for example, that minority patients are more likely to seek care from practitioners with whom they share a common race, ethnicity, or language.56 Furthermore, racial and ethnic diversity in medical education has been found to enhance the cross-cultural learning and competencies all practitioners need to treat a diverse patient population.57 Finally, close examination of medical school graduates indicates that professionals of color are more likely than their nonminority peers to practice in minority and medically underserved communities.58
Despite these findings, bans on race-sensitive admissions at public institutions across six states (California, Washington, Texas, Florida, Michigan, and Nebraska) have led to a 17 percent drop in the percentage of Latino, African American, and Native American students enrolled in medical school, from 18.5 percent to 15.3 percent.59 A similar decline was seen in public medical schools in Texas after Hopwood v. Texas prohibited institutions from considering race as a factor in admissions.60 As noted, a racially diverse medical school student body can improve the cultural competence of all doctors. Therefore, when race cannot be considered as a factor in medical school admissions, communities of color are likely to suffer not just from the quality of health care they receive but also from its very availability, as fewer professionals of color are available to serve patients of color.
Racial diversity in business schools and other professional schools has also declined. In the pre-Proposition 209 UC graduating class of 1997, African Americans and Latinos comprised nearly 20 percent of the professional degrees granted (law, medicine, pharmacy, business, public policy, architecture, etc.) across all campuses, compared to only 10 percent in 2010.61 In 2014, at the six public business schools in the UC system, African Americans, Latinos, and American Indians combined represented only 5.3 percent of MBA students, which is less than half the average at comparable U.S. business schools.62
Barring race-conscious admissions policies not only leads to declines in racial diversity across a number of educational programs, it also inhibits an institution’s efforts to create a positive racial experience for all its students. The findings from a recent study I conducted with a fellow researcher suggest that the influence of these policies extends beyond admissions and the number of racially diverse students to efforts that are critical for supporting students of color throughout their education.63 Barring the consideration of race in admissions decisions has negative consequences on the work of administrators charged with supporting students of color once they are on campus.64
For this study, we interviewed fourteen campus-level administrators charged with implementing a diversity policy at the University of Michigan. As a flagship public institution, the University of Michigan is representative of elite public schools in other states that have been affected by laws limiting the consideration of race in admissions. We sought to understand how these administrators understood the influence Proposal 2 had on their work. Our interviews took place during 2011 and 2012, by which time Proposal 2 had been in effect for five years. Consequently, the university had time to respond to the law and to clarify any initial uncertainties regarding its legal requirements.
The administrators we interviewed described the law as (1) limiting the conversations they felt they could have around race and racism; (2) making their individual efforts to support racial diversity less visible; (3) making them feel less empowered to advocate for racial diversity; and (4) contributing to concerns that students would have a negative perception of the university’s commitment to racial diversity.65 Yet, research has demonstrated that efforts to support diversity on campus require visible, sustained support at various institutional levels66 and that individuals need to feel empowered to do their work.67 Because bans on race-sensitive admissions policies, like Proposal 2, undermine these efforts, they can have negative consequences on the experiences of students of color on campus.
In some respects, these findings are not surprising. Proposal 2 represents a color-blind perspective on policymaking and characterizes affirmative action policies as “grant[ing] preferential treatment” for racial minorities.68 By banning such “preferential treatment”, this and other color-blind practices can be interpreted as seeking to ensure equal protection for all citizens. However, this framing diminishes the significance of longstanding racial inequities and ignores existing policies that also show preference, such as admissions practices that favor the children of alumni, which generally privilege whites and disadvantage African Americans, Native Americans, and Latinos.69 Furthermore, this framing ultimately ignores the broader context of accumulated disadvantage for students of color from past and ongoing discriminatory practices that affirmative action policies were originally intended to address. The passage of Proposal 2 essentially sanctioned a color-blind perspective in Michigan’s constitution, and the findings from this study indicate that this color-blind perspective was reflected in the actions and perceptions of the individual actors in charge of its implementation. Proposal 2 made it more difficult to talk about race and about the structural factors that contribute to educational inequities in the current university climate. As a result, it effectively silenced discussions about these issues and left individuals feeling less empowered to advocate for racial diversity.
These findings are consistent with lessons of other studies that examine the consequences of “race-neutral” or color-mute language on educators’ efforts to address and counter racial inequities.70 A comprehensive study of school discourse about race and how school administrators, teachers, and students experienced a K-12 school environment in which race was not explicitly discussed, is informative.71 When analyzing the color-mute72 language imposed on schools, Pollock notes how the deracialized words we use when discussing plans for racial equality can actually keep us from discussing ways to make opportunities equal on racial terms.73 For example, goals set using generalized language may not be as achievable as they would be if targeted racial terms were used and specific populations defined. Moreover, “discarding race analysis prematurely neglects the continuing ways in which opportunities are racially unequal.”74 Indeed, as Pollock posits, “actively deleting race words from our everyday talk of social ‘problems’ can actually help increase the role race plays in those problems . . . .”75
The notion that suppressing the open discussion of race can contribute to greater inequities can be explained in part by the view that race often subconsciously shapes attitudes and behavior. Issues of race often emerge through implicit bias, such as attitudes toward particular social groups76 and other racialized psychological phenomena, such as stereotype threat.77 When these attitudes are not addressed or confronted, our actions, whether we intend to or not, can perpetuate racial inequity. This phenomenon has been documented in the K-12 context, in which race-based beliefs play out in teachers’ lower expectations for students of color78 or in a disproportionate number of disciplinary actions and special education referrals for African American boys.79 These beliefs help to reinforce inequities because race-based expectations have real implications for how students perform in schools.80
A recent major study of the decisionmaking processes used by faculty and higher education administrators in graduate school admissions documents how the harm of a color-mute framework plays out in the higher education context to create, rather than redress, racial inequities.81 This study showed that, across the graduate fields from philosophy to physics, most (nine out of ten) committees selected the first round of “highly qualified” applicants through “race-neutral” or “color-blind” methods.82 Admissions committees employing race-neutral methods did so either as required by state laws banning the consideration of race in admissions, or voluntarily.83 In the first round of application review, committees screened for very high Graduate Record Examinations (GRE) scores and very high grades. It was only after students made it past this first cut that committees evaluated applicants based on how they contributed to the diversity of the student body.84 Thus, within a color-mute framework, discussions about the validity of assessment tests and admissions standards are silenced. Also silenced is an understanding of how a focus on standardized tests as a primary (if not sole) measure of merit systematically excludes the very underrepresented groups that would further the institution’s commitment to diversity. These findings demonstrate that, even for individuals who are committed to diversity, racial bias creates a barrier to access. Unless active steps are taken to avert these actions, educators will actively perpetuate racial inequities.
Rather than reducing the focus on race through color-mute language, ensuring that students of color succeed on college and university campuses requires that institutional policies and practices (including race-conscious admissions) put a more nuanced focus on race. This is because addressing students’ differential experiences on campus requires having honest conversations around race. These discussions should address issues related to white privilege, structural racism, and race-based beliefs that manifest themselves through implicit biases.85 Conversations about race are often difficult and can generate intense and uncomfortable emotions.86 Some white individuals, for example, feel a sense of guilt or a lack of trust, and thus resist discussions of privilege and power and generally avoid the topic of race.87 This dynamic makes it difficult to sustain discussions of race and racism, even within a race-conscious context. Thus, attending to these dynamics in an environment that limits the consideration of race in admissions becomes even more critical, particularly as the racial diversity of the student body declines.
Moreover, as institutions turn to alternative measures deemed race-neutral in response to the legal requirements the Court rearticulated in Fisher I, they need to actively counter the negative consequences of such approaches. Because so-called race-neutral policies necessarily require putting a less direct focus on race, they can result in using color-mute language that the research discussed above has documented can exacerbate the very racial inequities educators want to dismantle. Thus, institutions must facilitate open discussions about race and to identify racial biases and other structural and institutional practices that contribute to racial inequity.
Failing to explicitly consider race in admissions can also allow harmful racial stereotypes to operate while eliminating any possibility of countering them. State “top ten percent plans,” which are deemed race-neutral, provide a direct example. The Texas Top Ten Percent Plan, for example, gives students in the top 10 percent of their high school class automatic admission to any public university in the state. The goal is to maintain diversity in the most competitive public universities in a race-neutral way.88 However, as the arguments the University of Texas (UT Austin) advanced in Fisher I and Fisher II highlight, because the plan’s success depends in large part on high schools being racially segregated, African American or Latino students who qualify for admission under the plan may come from high schools that are predominantly African American or Latino, or from primarily low socioeconomic backgrounds. Thus, the percent plan can have the effect of perpetuating and reinforcing racial stereotypes on campus, such as the notion that all African American and Latino students attended segregated high schools or that they all come from low-income families. Yet, African American and Latino students who attend integrated or largely white schools and could bring a range of experiences to the college campus to help counter stereotypes may not be in the top 10 percent of their class.89 Therefore, as UT Austin has argued, complementing the percent plan with a policy that allows institutions to consider other students’ personal and social experiences would help create a truly diverse environment and help break down racial stereotypes.
As I summarized in Parts II and III, bans on race-conscious policies have led to declines in the racial and ethnic diversity of the student body across numerous educational sectors and have been documented to harm institutional efforts needed to support students of color on campus. In addition to these direct harmful consequences, barring race-conscious admissions policies can have other long-term negative ramifications, such as harming the campus racial climate and undermining cross-racial engagement.
A less racially diverse student body, for example, is associated with a less welcoming campus climate90 for students of color. This, in turn, harms an institution’s ability to achieve the educational benefits of diversity because it can inhibit the interactions among students from different racial groups that are needed to realize such benefits.91 Furthermore, students who have negative racial experiences on campus are likely to feel less of a sense of belonging during the first two years of college92 and are more likely to be dissatisfied with their overall college experience.93 And how students perceive the academic and social environments on their campus could be a central factor in their decision to participate in various academic and social activities.94
This body of work supports UT Austin’s decision to reintroduce the consideration of race as a factor in its admissions policies. UT Austin did so after determining that its African American and Latino students felt isolated on campus, due in part to limited racial diversity.95 Subsequent results from the Student Experience in the Research University (SERU) survey, which investigates whether students at UT Austin feel respected on campus, support UT Austin’s initial determination.96 Responses to the survey in 2010, 2011, and 2013 combined demonstrate that students of color feel less respected at UT Austin than their white peers.97 Only 65.4 percent of African Americans and 87.7 percent of Latinos either “somewhat agree,” “agree,” or “strongly agree” that “students of my race/ethnicity are respected on this campus,” compared to 96.9 percent of whites.98
While other factors may contribute to students’ perceptions of respect on campus, responses to the same SERU survey at UC Berkeley and UCLA, which are barred from considering race in admissions, indicate that perceptions of the campus racial climate are far worse at these two institutions. Combined survey results from 2008–2014 show that only 52.1 of African American respondents at UC Berkeley and 57.2 percent at UCLA at least “somewhat agree” that students of their race are respected on campus.99 Results for Latinos at UC Berkeley (73.4 percent) and UCLA (71.2 percent) are also significantly worse than at UT Austin.100
The campus climate surveys from UT Austin and other leading research universities are consistent with the Diversity Learning Environment survey administered in recent years at a broader range of U.S. colleges and universities. The initial results from thirty-one institutions across the country, including several campuses in Texas and California, show that less campus diversity is associated with more reports of stereotyping and discrimination from African American and Latino students.101 Moreover, African American and Latino students are more likely to feel excluded from campus events and activities at institutions with little racial diversity than at institutions that are more diverse. Although African American students are not the predominant minority on any of these campuses, the data indicate that they feel more included where there is a diverse campus environment. Greater diversity on campus is also significant in reducing Latino students’ feelings of isolation.102
A more recent survey of fifty-eight institutions shows that, at the least diverse schools in the sample, about one in five (20.5 percent) African American students and one in seven (14.5 percent) Latino students had reported at least one discriminatory incident to campus authorities. These figures are likely lower than the number actually experienced, since most instances of perceived bias and discrimination go unreported.103 At the more diverse schools in the survey, significantly lower percentages of African American and Latino students reported such incidents.
Additional data from more than 8000 Latino students on eighty-two campuses show that minority students experience fewer feelings of exclusion, negative verbal comments, and offensive visual images at moderately diverse institutions (those with 21–35 percent African American and Latino enrollment) than at less diverse institutions, and that these experiences decline still further as minority enrollment increases.104 Other Diversity Learning Environment survey results confirm that African American, American Indian, and Latino students manifest much greater “racial salience” (for example, they think about their racial group membership and identity with greater frequency) than white students, which is related to the racial dynamics described above.105
The association between lower levels of diversity and a decline in perceptions of a supportive campus racial climate is thus well-documented. Indeed, a study that examined the experiences of students of color at the University of Michigan Law School following the passage of Proposal 2 in 2006 documents the relationship between laws that prohibit the consideration of race in admissions and students’ perceptions of an institution’s commitment to diversity, a factor that can facilitate cross-racial engagement. The results of this survey of 505 students (47 percent of the student body), the vast majority of whom entered the law school after Proposal 2 was in effect, indicate that almost half of the African American students (47 percent) did not perceive the campus as being supportive of diversity, whereas the vast majority of white students (84 percent) did.106 These perceptions of institutional support for diversity are important because research in other educational sectors shows that they promote student success and facilitate interactions among students from different racial groups. For example, in the undergraduate context, a perceived low level of institutional commitment to diversity is associated, in part, with higher levels of perceived hostility and discrimination, with low grades for African American students, and with a lower sense of belonging among Latino students.107 In contrast, higher perceived levels of institutional commitment to diversity are associated with perceptions of relatively low racial tension among students of color, with higher college grade point averages, and with an increase in students making it a personal goal to promote racial understanding.108
While Justice Kennedy has acknowledged that race matters in inhibiting students’ educational opportunities, in past cases he has voted to restrict the consideration of race in educational policies. This requirement is a result of Kennedy’s balancing between the constitutional principles of liberty and equality. I have argued that this balance in Fisher II should be informed by empirical findings relevant to understanding the costs and benefits of individualized racial classifications. Social science demonstrates that racial classifications are not inherently harmful—indeed, they may be required to address the ways race matters in students’ educational experiences. When race cannot be directly addressed in the admissions process, even in the limited fashion required by past cases, studies have also documented the harm to student body diversity, to institutional efforts for supporting students of color, and for the campus racial climate and cross-racial engagement. Legal reasoning should be grounded in these social realities so it does not create unnecessary complications for postsecondary institutions seeking to secure the educational benefits of diversity in a society in which race continues to shape educational opportunity. The racial bias and discrimination students of color continue to face on college campuses require that colleges and universities have all the tools at their disposal to create inclusive learning environments for their students.
. 135 S. Ct. 2888 (2015) (granting petition for writ of certiorari).
. I use the terms “race-conscious” and “race-sensitive” admissions interchangeably to describe the policy in the case and do not employ the term “affirmative action” for the reasons Justice Sotomayor explained in Schuette. See Schuette v. Coalition to Defend Affirmative Action, 134 S. Ct. 1623, 1651 n.2 (2014) (Sotomayor, J., dissenting). As Justice Sotomayor noted, the Court previously reviewed policies using quotas or point systems that accord an applicant numerical advantage because of race, or that admit students solely based on race to address the effects of historical and ongoing discrimination. Id. These policies involved “affirmative action” as historically understood, which opponents frame as granting preferential treatment to individuals. Id. Past Court cases, starting with Regents of the University of California v. Bakke, 438 U.S. 265 (1978), rendered those practices unconstitutional, and the Court has since allowed policies that instead consider race as one factor among many in admissions decisions to promote a diverse student body. See Grutter v. Bollinger, 539 U.S. 306, 334, 336–337 (2003).
. See generally Elise C. Boddie, Indignities of Colorblindness, 64 [small-caps]UCLA L. Rev. Disc.[end-small-caps] (forthcoming 2016) (in this symposium).
. As was the case in Fisher v. University of Texas (Fisher I), 133 S. Ct. 2411 (2013), Justice Kagan recused herself in light of her involvement in the case in the early stages of litigation. Justice Scalia’s passing thus reduces the number of justices voting in the case to seven.
. With Justices Sotomayor, Breyer, and Ginsburg endorsing race-conscious policies, and Chief Justice Roberts and Justices Alito and Thomas opposing them.
. Justice Kennedy was viewed as a decisive vote in Fisher I, with the possibility of his vote generating a tie that would have allowed the lower court’s decision to stand without creating precedent for the rest of the nation, or a 5-3 vote overruling the lower court’s ruling. 133 S. Ct. 2411. Given the composition of the Court at the time, therefore, few observers expected the 7-1 opinion in Fisher I that sent the case back to the Fifth Circuit for further review, leaving in place the core principles that allowed for race-conscious policies. Other analyses have described the decision in Fisher I to remand the case to the lower court as a compromise. See [small-caps]Joan Biskupic, Breaking In: The Rise of Sonia Sotomayor and the Politics of Justice[end-small-caps] 191–210 (2014). While it is possible that the justices may reach another compromise this time around, it is also possible the same voting patterns may not hold, thus reinforcing the critical role Justice Kennedy will play in the forthcoming Fisher v. University of Texas (Fisher II) opinion. 135 S. Ct. 2888.
. 133 S. Ct. 2411 (2013).
. To be sure, a vote joining the conservative block could also reverse the lower court’s ruling on the grounds that the policy is unconstitutional. It is also possible that the Court could reverse on the grounds that Abigail Fisher lacks standing. Given that the Court did not rule on standing grounds in Fisher I, it may choose not to revisit that issue in Fisher II.
. Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 787 (2007) (Kennedy, J., concurring).
. See Grutter v. Bollinger, 539 U.S. 306, 387 (2003) (Kennedy, J. dissenting).
. Criteria in this analysis include ensuring that the policy (a) does not operate as a quota; (b) is adopted after an institution’s good faith consideration of workable race-neutral alternatives; (c) involves a flexible, individualized consideration of applicants so that race, though important, is only one of a number of factors being considered; (d) does not unduly burden disfavored groups; and (e) is limited in time or includes a periodic review to assess its continued necessity. See Fisher I, 133 S. Ct. 2411, 2420; Grutter, 539 U.S. at 341–42 (majority opinion); Regents of the University of California v. Bakke, 438 U.S. 265, 318–19 (1978). In Fisher I, the Court clarified that this narrow tailoring determination is one that courts need to make themselves, with great care, while taking account of a university’s experience and expertise. 133 S. Ct. at 2420.
. I use the terms racial and ethnic diversity interchangeably.
. See Parents Involved, 551 U.S. at 783 (Kennedy, J., concurring); Grutter, 539 U.S. at 387 (Kennedy, J., dissenting).
. Grutter, 539 U.S. at 391 (Kennedy, J., dissenting).
. See Parents Involved, 551 U.S. at 787 (Kennedy, J., concurring). Parents Involved was one of two cases that addressed the constitutionality of a school district’s voluntarily adopted race-conscious student assignment policy designed to maintain racial diversity and avoid racial isolation in schools. Id. In a 4-1-4 plurality opinion, the Court struck down the plans for not being sufficiently narrowly tailored in their use of race. Id. Under the Court’s ruling, school districts still retain the ability of using more narrowly tailored race-conscious plans to further what a majority of the justices agreed was a compelling governmental interest to create racially diverse schools and avoid the harms of racial isolation that interfere with equal educational opportunities for students. Id.
. Id. at 795 (Kennedy, J., concurring).
. Id. at 797.
. Id. at 787.
. Id. at 789.
. Id. at 797.
. Id. at 788–89. While Justice Kennedy views these measures as “race-conscious,” other justices may also view them as “race neutral” because they do not explicitly reference race.
. See generally Elise C. Boddie, supra note 3, in this symposium, for why such benefits also include furthering principles of individual dignity.
. See Linda R. Tropp et al., The Use of Research in the Seattle and Jefferson County Desegregation Cases: Connecting Social Science and the Law, 7 [small-caps]Analyses Soc. Issues & Pub. Pol’y[end-small-caps] 113–14 (2007) (analyzing in detail how social science informs an evaluation of the impact of racial categorizations).
. Id. at 108–09; see also infra Part IV (arguing further that suppressing the open discussion of race can contribute to greater inequities).
. See Linda R. Tropp & Rebecca A. Bianchi, Interpreting References to Group Membership in Context: Feelings About Intergroup Contact Depending on Who Says What to Whom, 37 [small-caps]Eur. J. Soc. Psychol.[end-small-caps] 153, 165 (2007) (finding, in an experimental study, that undergraduate students of color perceive references to their racial group positively because it acknowledges specific circumstances and experiences that are often overlooked by dominant culture and that such racial acknowledgment can encourage interactions across race).
. See Tropp et al., supra note 24, at 93.
. See Fisher I, 133 S. Ct. 2411, 2432–34 (2013) (Ginsburg, J., dissenting).
. Gratz v. Bollinger, 539 U.S. 244, 299–301 (2003) (Ginsburg, J., dissenting).
. Id. at 304–305.
. Fisher I, 133 S. Ct. at 2433–34 (quoting Gratz v. Bollinger, 539 U.S. 244, 305 n.11(2003)).
. 539 U.S. at 305 n.11.
. Id. at 300–01.
. See Affirmative Action: State Action, [small-caps]Nat’l Conf. St. Legislatures[end-small-caps] (Apr. 2014), http://www.ncsl.org/research/education/affirmative-action-state-action.aspx [https://perma.cc/2G6F-JF3T].
. Grutter v. Bollinger, 539 U.S. 306, 332 (2003).
. Peter Hinrichs, The Effects of Affirmative Action Bans on College Enrollment, Educational Attainment, and the Demographic Composition of Universities, 94 [small-caps]Rev. Econ. & Stat.[end-small-caps] 712, 717 (2012).
. See Ben Backes, Do Affirmative Action Bans Lower Minority College Enrollment and Attainment? Evidence From Statewide Bans, 47 [small-caps]J. Hum. Resources[end-small-caps] 435, 447 (2012) (finding that enrollment of African American students at the most selective universities would have been 1.6 percentage points higher, and 2.9 percentage points higher for Latinos, had affirmative action not been banned in these states).
. Jerome Karabel, No Alternative: The Effects of Color-Blind Admissions in California, in[small-caps] Chilling Admissions: The Affirmative Action Crisis and the Search for Alternatives[end-small-caps], 34–37 (Gary Orfield & Edward Miller eds., 1998).
. African American enrollment dropped from 7 percent (252 in a class of 3,215) to 3.7 percent (122 in a class of 3,333). Latino and Chicano student enrollment dropped from 14.6 percent (469 in a class of 3,215) to 7.9 percent (266 in a class of 3,333). [small-caps]Univ. of Cal. Office of the President, University of California Application, Admissions and Enrollment of California Resident Freshmen for Fall 1995 Through 2014[end-small-caps] 1–2, 5 (Jan. 2015), http://www.ucop.edu/institutional-research-academic-planning/_files/factsheets/2014/flow-frosh-ca-14.pdf [https://perma.cc/5CRJ-UBW8].
. This represents a decline in the percent of African American students enrolled from 5.6 percent to 3.5 percent, and of Latino and Chicano students from 15.8 percent to 11 percent. Id. at 5.
. Eric Grodsky & Michal Kurlaender, The Demography of Higher Education in the Wake of Affirmative Action, in [small-caps]Equal Opportunity in Higher Education: The Past and Future of California’s Proposition[end-small-caps] 209, 33–34 (Eric Grodsky & Michal Kurlaender eds., 2010).
. Michigan Civil Rights Amendment, Proposal 2 (2006),[small-caps] Ballotpedia[end-small-caps], https://ballotpedia.org/Michigan_Civil_Rights_Amendment,_Proposal_2_(2006) [https://perma.cc/SR7U-MQ83].
. A drop from 6.11 percent to 4.6 percent. Univ. of Mich. Office of the Registrar, Enrollment Report for 2012, [small-caps]Deep Blue[end-small-caps] (2013), http://deepblue.lib.umich.edu/handle/2027.42/96814 [https://perma.cc/Q7KM-TXBP]; Univ. of Mich. Office of the Registrar, Enrollment Report for 2010, Deep Blue (2011), http://deepblue.lib.umich.edu/handle/2027.42/96812 [https://perma.cc/PM4V-6SH9].
. A decline from 5.1 percent to 3.85 percent. [small-caps]Univ. of Mich. Office of the Registrar[end-small-caps], Enrollment Report for 2012, [small-caps]Deep Blue[end-small-caps] (2013), http://deepblue.lib.umich.edu/handle/2027.42/96814 [https://perma.cc/Q7KM-TXBP]; [small-caps]Univ. of Mich. Office of the Registrar[end-small-caps], Enrollment Report for 2010, [small-caps]Deep Blue[end-small-caps] (2011), http://deepblue.lib.umich.edu/handle/2027.42/96812 [https://perma.cc/PM4V-6SH9].
. A decline from 6.2 percent to 4.6 percent. Liliana M. Garces, Understanding the Impact of Affirmative Action Bans in Different Graduate Fields of Study, 50 [small-caps]Am. Educ. Res. J.[end-small-caps] 251, 274 (2013).
. A drop from 7.8 percent to 6.3 percent. Id.
. A drop from 12.1 percent to 10.2 percent. Id.
. A drop from 10.2 percent to 9 percent. Id.
. See infra Part IV.
. See, e.g., [small-caps]Scott E. Page, The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies[end-small-caps] 327–28 (2007) (finding that more diverse groups consistently outperform less diverse groups in generating effective and innovative solutions to complex problems).
. [small-caps]Comm. on Underrepresented Grps. & the Expansion of the Sci. & Eng’g Workforce Pipeline et al., Expanding Underrepresented Minority Participation: America’s Science and Technology Talent at the Crossroads 34 (2011); President’s Council of Advisors on Sci. & Tech., Engage to Excel: Producing One Million Additional College Graduates With Degrees in Science, Technology, Engineering, and Mathematics [end-small-caps]1 (2012), https://www.whitehouse.gov/sites/default/files/microsites/ostp/pcast-engage-to-excel-final_2-25-12.pdf [https://perma.cc/4H7B-EZ6V].
. See, e.g., [small-caps]Comm. on Understanding & Eliminating Racial & Ethnic Disparities in Health Care, Inst. of Med. of the Nat’l Acads., Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care[end-small-caps] 114–15, 122–23 (Brian D. Smedley et al. eds., 2003).
. Id.; see also [small-caps]Agency for Healthcare Res. & Quality, U.S. Dep’t of Health & Human Servs., National Healthcare Disparities Report [end-small-caps]2011, at 2 (2012) (finding that African Americans, American Indians and Alaska Natives, and Hispanics/Latinos receive lower quality and less accessible healthcare than Whites); [small-caps]Ctrs. for Disease Control & Prevention, U.S. Dep’t Health & Hum. Servs., Health Disparities and Inequalities Report—United States,[end-small-caps] 2011, at 1–2 (2011) (reporting, among other findings, that racial and ethnic minorities are more likely to live near and suffer from the effects of air pollution, that infants born to African American women are several times more likely to die than infants born to women of other races and ethnicities, and that coronary heart disease accounts for the largest proportion of inequality in life expectancy between white and African American individuals).
. See, e.g., [small-caps]U.S. Dep’t Health & Hum. Servs., The Rationale for Diversity in the Health Professions: A Review of the Evidence[end-small-caps] 3 (2006), http://bhpr.hrsa.gov/healthworkforce/reports/diversityreviewevidence.pdf [https://perma.cc/KX6H-AZ82].
. See generally, e.g., Somnath Saha et al., Do Patients Choose Physicians of Their Own Race?, 19[small-caps] Health Aff.[end-small-caps] 76 (2000).
. See [small-caps]Comm. on Understanding & Eliminating Racial & Ethnic Disparities in Health Care[end-small-caps], supra note 53, at 203.
. Liliana M. Garces & David Mickey-Pabello, Racial Diversity in the Medical Profession: The Impact of Affirmative Action Bans on Underrepresented Student of Color Matriculation in Medical Schools, 86 [small-caps]J. Higher Educ. [end-small-caps]264, 287 (2015).
. Hopwood v. Texas, 78 F.3d 932, 962 (5th Cir. 1996), abrogated by Grutter v. Bollinger, 539 U.S. 306, 343 (2003); Mexican Am. Legal Def. & Educ. Fund (MALDEF) et al., Blend It, Don’t End It: Affirmative Action and the Texas Ten Percent Plan After Grutter and Gratz, 8 [small-caps]Harv. Latino L. Rev.[end-small-caps] 33, 36 (2005).
. See [small-caps]William C. Kidder & Patricia Gándara, Two Decades After the Affirmative Action Ban: Evaluating the University of California’s Race-Neutral Efforts[end-small-caps] 31 (2015), http://www.ets.org/Media/Research/pdf/kidder_paper.pdf [https://perma.cc/2CKU-HVEN].
. See id. at 30.
. See Liliana M. Garces & Courtney D. Cogburn, Beyond Declines in Student Body Diversity: How Campus-Level Administrators Understand a Prohibition on Race-Conscious Postsecondary Admissions Policies, 52 [small-caps]Am. Educ. Res. J.[end-small-caps] 828, 849–855 (2015).
. See id.
. See id.
. See, e.g., [small-caps]Jeffrey F. Milem et al., Making Diversity Work on Campus: A Research-Based Perspective[end-small-caps] iv (2005), https://www.aacu.org/sites/default/files/files/mei/milem_et_al.pdf [https://perma.cc/NZ5K-5U8Y].
. See generally Adrianna J. Kezar, Shared Leadership for Creating Campus Cultures That Support Students of Color, in [small-caps]Creating Campus Cultures: Fostering Success Among Racially Diverse Student Populations[end-small-caps] 150 (Samuel D. Museus & Uma M. Jayakumar eds., 2012).
. [small-caps]Mich. Const.[end-small-caps] art. I, § 26, http://www.legislature.mi.gov/(S(fgrlhdjzhwmk042bnph5zvyw))/mileg.aspx?page=GetObject&objectname=mcl-Article-I-26 [https://perma.cc/T48H-5AFB].
. See Devon W. Carbado & Cheryl I. Harris, The New Racial Preferences, 96 [small-caps]Cal. L. Rev. [end-small-caps]1139, 1194–1207 (2008) (discussing the challenges of racial “neutrality” in the admissions context).
. See [small-caps]Mica Pollock, Colormute: Race Talk Dilemmas in an American School[end-small-caps] 4–5 (2004); see also Evan P. Apfelbaum et al., In Blind Pursuit of Racial Equality?, 21 Psychol. Sci. 1587, 1591 (2010); Evan P. Apfelbaum et al., Racial Color Blindness: Emergence, Practice, and Implications, 21 Current Directions Psychol. Sci. 205, 206 (2012).
. See generally [small-caps]Pollock[end-small-caps], supra note 70.
. I employ the terms color-mute and color-blind interchangeably. They involve the purposeful silencing of race words or the active deletion of racial labels.
. See [small-caps]Pollock[end-small-caps], supra note 70, at 14.
. Id. at 216.
. Id. at 217.
. See, e.g., John F. Dovidio et al., The Nature of Contemporary Racial Prejudice: Insight From Implicit and Explicit Measures of Attitudes, in [small-caps]Attitudes: Insights From the New Implicit Measures [end-small-caps]165–86 (Richard E. Petty et al. eds., 2009); see also John F. Dovidio et al., Implicit and Explicit Prejudice and Interracial Interaction, 82[small-caps] J. Personality & Soc. Psychol.[end-small-caps] 62, 62 (2002).
. Stereotype threat is classically manifested in high-stakes test performance, and it involves the threatening experience of conforming to negative race-based stereotypes present in the larger society. See generally [small-caps]Claude M. Steele, Whistling Vivaldi: And Other Clues to How Stereotypes Affect Us [end-small-caps]134–90 (2010); Toni Schmader & Michael Johns, Converging Evidence That Stereotype Threat Reduces Working Memory Capacity, 85 [small-caps]J. Personality & Soc. Psychol. [end-small-caps]440, 449–50 (2003).
. See, e.g., John B. Diamond et al., Teachers’ Expectations and Sense of Responsibility for Student Learning: The Importance of Race, Class, and Organizational Habitus, 35[small-caps] Anthropology & Educ. Q. [end-small-caps]75, 93–95 (2004).
. See[small-caps] Amanda E. Lewis & John B. Diamond, Despite the Best Intentions: How Racial Inequality Thrives in Good Schools [end-small-caps]54–59 (2015).
. See id. at 54.
. See [small-caps]Julie R. Posselt, Inside Graduate Admissions: Merit, Diversity, and Faculty Gatekeeping[end-small-caps] 12 (2016).
. See id. at 8, 50.
. See id. at 59, 71.
. See id. at 59.
. See, e.g., Liliana M. Garces & Cynthia Gordon da Cruz, A Strategic Racial Equity Framework, 92 [small-caps]Peabody J. Educ. [end-small-caps](forthcoming Sept. 2017) (discussing the importance of naming hidden contributors to inequality to promote racial equity in education).
. See generally, e.g., [small-caps]Pollock[end-small-caps], supra note 70.
. See [small-caps]Glenn E. Singleton, More Courageous Conversations About Race[end-small-caps] 3 (2013).
. See Lindsay Daugherty et al., The Texas Ten Percent Plan’s Impact on College Enrollment, 14 [small-caps]Educ. Next[end-small-caps] 63, 63 (2014), http://educationnext.org/texas-ten-percent-plans-impact-college-enrollment [https://perma.cc/Q755-CR9L].
. This may be due to factors widely documented by research, such as tracking or differential treatment by teachers and counselors. See, e.g., Andrea Venezia & Michael W. Kirst, Inequitable Opportunities: How Current Education Systems and Policies Undermine the Chances for Student Persistence and Success in College, 19 [small-caps]Educ. Pol’y [end-small-caps]283, 287, 289 (2005) (finding inequitable tracking in Texas along racial and class lines); see also David Card & Jesse Rothstein, Racial Segregation and the Black-White Test Score Gap, 91[small-caps] J. Pub. Econ.[end-small-caps] 2158, 2158 (2007) (explaining that data on enrollment in honors courses suggest that within-school segregation increases when schools are more highly integrated); Jeannie Oakes, Keeping Track: Structuring Equality and Inequality in an Era of Accountability, 110 [small-caps]Tchrs. C. Rec.[end-small-caps] 700, 705–07 (2008) (providing a national overview of tracking based on race and class); William H. Schmidt, At the Precipice: The Story of Mathematics Education in the United States, 87[small-caps] Peabody J. Educ.[end-small-caps] 133, 140–41 (2012) (finding that in thirteen states, including Texas, a student’s opportunity to learn mathematics is greatly influenced by race and family income).
. A campus racial climate is defined as the community members’ attitudes, behaviors, and perceptions of discrimination and contact among students of different racial groups. See, e.g., Marvin W. Peterson & Melinda G. Spencer, Understanding Academic Culture and Climate, in [small-caps]Assessing Academic Climates and Cultures, 68 New Directions for Institutional Res.[end-small-caps] 3 (William G. Tierney ed., 1990).
. See, e.g., Liliana M. Garces & Uma M. Jayakumar, Dynamic Diversity: Toward a Contextual Understanding of Critical Mass, 43 [small-caps]Educ. Researcher[end-small-caps] 115, 117–21 (2014) (analyzing decades of diversity-related research showing that, for institutions to leverage the benefits of diversity, they must promote a healthy racial climate, provide a welcoming environment for all students, prevent harms due to racial isolation, diminish feelings of tokenism, and promote cross-racial interactions).
. See generally Sylvia Hurtado et al., Predicting Transition and Adjustment to College: Biomedical and Behavioral Science Aspirants’ and Minority Students’ First Year of College, 48 [small-caps]Res. Higher Educ.[end-small-caps] 841 (2007); Angela M. Locks et al., Extending Notions of Campus Climate and Diversity to Students’ Transition to College, 31 [small-caps]Rev. Higher Educ.[end-small-caps] 257 (2008).
. See Berkeley Miller & Sutee Sujitparapitaya, Campus Climate in the Twenty-First Century: Estimating Perceptions of Discrimination at a Racially Mixed Institution, 1994–2006, 145 [small-caps]New Directions Institutional Res. [end-small-caps]29 (2010).
. See generally Samuel D. Museus et al., Racial and Ethnic Minority Students’ Success in STEM Education, 36 [small-caps]ASHE Higher Educ. Rep.[end-small-caps] 1 (2011).
. See Joint Appendix to the Respondent’s Brief at 446a, Fisher v. Univ. of Texas at Austin (Fisher II), No. 14-981 (U.S. Sept. 3, 2015).
. For additional information about the SERU survey, see [small-caps]William C. Kidder, The Salience of Racial Isolation: African Americans’ and Latinos’ Perceptions of Climate and Enrollment Choices With and Without Proposition[end-small-caps] 209, at 34–37 (2012), http://civilrightsproject.ucla.edu/research/college-access/affirmative-action/the-salience-of-racial-isolation-african-americans2019-and-latinos2019-perceptions-of-climate-and-enrollment-choices-with-and-without-proposition-209/Kidder_Racial-Isolation_CRP_final_Oct2012-w-table.pdf [https://perma.cc/LDY5-MV53].
. See Brief of 823 Social Scientists as Amici Curiae in Support of Respondents at App. 1, Fisher II, No. 14-981 (U.S. Oct. 30, 2015).
. See [small-caps]Sylvia Hurtado & Adriana Ruiz, The Climate for Underrepresented Groups and Diversity on Campus [end-small-caps]2 (2012), http://heri.ucla.edu/briefs/urmbriefreport.pdf [https://perma.cc/55XF-WWRR] (cataloging an initial Diverse Learning Environments (DLE) study including 490 African American students and 3488 Latino students).
. Intergroup relations at highly diverse institutions, defined as those where underrepresented minority students represent at least 36 percent of the student body, also require attention, as increasing numbers of Latino students change the ethnic composition of previously predominantly white campus environments. Id.
. [small-caps]Sylvia Hurtado & Adriana Ruiz Alvarado, Discrimination and Bias, Underrepresentation, and Sense of Belonging on Campus [end-small-caps]2 (2015), http://www.heri.ucla.edu/PDFs/Discriminination-and-Bias-Underrepresentation-and-Sense-of-Belonging-on-Campus.pdf [https://perma.cc/U4UG-PV9J].
. See Rebecca L. Stotzer & Emily Hossellman, Hate Crimes on Campus: Racial/Ethnic Diversity and Campus Safety, 27[small-caps] J. Interpersonal Violence[end-small-caps] 644, 654–55 (2012) (finding that more diverse campuses have fewer reports of hate crimes per capita).
. See Sylvia Hurtado et al., Thinking About Race: The Salience of Racial Identity at Two- and Four-Year Colleges and the Climate for Diversity, 86 [small-caps]J. Higher Educ. [end-small-caps]127, 140–41 (2015).
. Meera E. Deo, Empirically Derived Compelling State Interests in Affirmative Action Jurisprudence, 65 Hastings L.J. 661, 691 (2014).
. See [small-caps]Milem et al.[end-small-caps], supra note 66, at 11.
. See id.