The University of Michigan
Information on Admissions Lawsuits

News Releases & Articles | Statements by University Leaders & Others | Press Kits, Photos & Media Contacts
¡En Español! | E-mail Sign-up | Archived Documents | U-M News Service | U-M Gateway

Supporting Research Court Filings Legal Overview FAQs What's New Admissions Lawsuits Home Page

Evidence for the Educational Benefits of Diversity in Higher Education:

Response to the Continuing Critique by the National Association of Scholars of the Expert Witness Report of Patricia Gurin in Gratz, et al. v Bollinger, et al. and Grutter v Bollinger, et al.

By Patricia Gurin, Nancy Cantor Distinguished University Professor, Emerita
University of Michigan

May 20, 2003

The National Association of Scholars (NAS) and other amici for the plaintiffs have repeatedly criticized my work on the educational benefits of experience with diversity, originally reported in Compelling Interest and most recently reported in two peer-reviewed articles, one in the Harvard Educational Review, the other in the Journal of Social Issues.

I am focusing here on the NAS amicus brief to the Supreme Court and an addendum to that brief authored by Thomas E. Wood, Executive Director of the California Association of Scholars, and Malcolm J. Sherman, Associate Professor of Mathematics and Statistics, State University of New York at Albany. Wood and Sherman are also the authors of an earlier critique that NAS used in earlier amicus briefs. Both critiques by Wood and Sherman and the NAS amicus brief can be found at , , and .

The NAS amicus brief and the addendum by Wood and Sherman again argue 1) that I have not proven that structural diversity (percent minority students on a campus) has educational benefits and that evidence showing that students benefit from actual experiences with diverse peers is irrelevant; 2) that my measures of educational outcomes are flawed and show a liberal bias; 3) that the size of the effects of diversity experience are trivially small; and 4) that racial preferences and diversity have negative effects on students.

Once again I feel compelled to respond to each of the central charges. And once again it is important to point out that none of these criticisms was raised during my deposition in these cases, nor did any expert witness step forward to counter my testimony in court. The NAS raised some of these issues in a general way in an amicus brief filed in the summer of 2000, which I answered in a supplemental expert report in January, 2001. However, it was only after the opinion by Judge Duggan in the Gratz et al. v Bollinger, et al. case, 122 F. Supp. 2d 811 (E.D. Mich. 2000), that the NAS released the original Wood-Sherman Report. Since then the NAS has continued to play an inappropriate role in these cases by introducing new arguments on appeal that were not presented by the parties or raised by the court.

None of these critiques — the earlier Wood and Sherman critique, the NAS amicus brief to the Supreme Court, and the addendum to that brief — have refuted the testimony I provided in the Gratz or the Grutter cases.

Three important independent assessments of my methods, analyses, and conclusions have been carried out and submitted to the Court.1 All agree that my report is scientifically valid.

“In sum, the Gurin Report is sound research that employs credible and widely accepted methodologies to conclude that student body diversity leads to a wide range of positive educational outcomes.” (p 13, amicus brief of the American Educational Research Association et al.)

“In sum, the methodology of the Gurin Report is consistent with scientific standards for research. Moreover, Dr. Gurin’s research confirms numerous studies brought to the Court’s attention by other amici and respondents which demonstrate that diversity experiences in college and graduate school are associated with a variety of important educational objectives. . . “ (p. 23, amicus brief of the American Psychological Association)

“Based upon our review of Gurin’s and Wood & Sherman’s analyses for this report, we believe that Gurin’s approach is scientifically sound, and that these and other attacks in the Wood and Sherman Report are misplaced and unwarranted. Through careful statistical analysis, Gurin has shown that racial (structural) diversity operating through students’ diversity experiences in college produces educational benefits to the students involved and to the larger society.” (Thomas and Shavelson, p. 9)

Meaning of Diversity: The Structural Diversity Controversy

The heart of the continuing criticism by the NAS is that I have not shown that percent minority on a campus has direct impact on student outcomes. This criticism has been addressed in my supplemental testimony of January 11, 2001 (pages 1-5), website response to the original Wood and Sherman (pages 2-16), and website response to Learner and Nagai (page 1). 2

These critics have used various approaches in their claim that my work has not shown any effect of diversity: 1) charges that diversity experience measures were proxies for percent minority, and that I should have used percent minority itself to examine its direct impact on student outcomes, that is an association with outcomes after controlling for diversity experiences; 2) charges that I deliberately focused on experience with diversity because I knew that percent minority has no direct impact; and 3) new charges in the addendum to the NAS Amicus Brief to the Supreme Court that my theory requires that I find evidence of a statistical interaction between percent minority, diversity experience, and educational outcomes.

These critics simply keep repeating the same argument, namely that only percent minority is relevant to this case, that diversity experiences are irrelevant to this case, and thus that they have refuted my testimony. I never argued that percent minority should have direct impact on educational outcomes. I never argued that the effect of experience with diversity (in classrooms and in the informal campus environment) should be greater on campuses with the largest percentages of minority students (the statistical interaction they now argue is required by my theory).

I have always argued that the presence of diverse peers

  • provides the possibility of having actual experience with diversity (you can’t have experience with diversity without diversity), and that

  • positive educational outcomes result from having actual experience with diverse peers that the presence of diverse peers promotes.

I have shown in the analyses of the national data from CIRP that the probability of having experience with diversity is greater where the percentage of minority students is greater. Chang, Astin, and Kim, using a different CIRP data set, substantiate that diversity experiences are more widespread where structural diversity is greater.

I do assume that affirmative action increases the probability that students will have experience with diverse peers or at least that affirmative action, combined with other efforts can have that effect. This is trivial: in a segregated school, interracial relationships cannot take place so that the benefit of actual experience with diversity cannot be realized. Since Stephen Raudenbush’s supplemental testimony of July 13, 2002, and February 24, 2001, shows that something approaching segregation would result from “race-blind admissions,” this is not a hard argument to make.

Previous Responses to the Structural Diversity Controversy

I first addressed the structural diversity controversy in my supplemental expert report of January 11, 2001, from which I quote:

“The NAS brief contends that the causal variables (e.g., participating in a racial/cultural awareness workshop and discussing racial issues) used in my work are not adequate proxies for racial diversity…. This NAS criticism seems to miss the point I make explicitly in my report that: Structural diversity is essential but, by itself, usually not sufficient to produce substantial benefits; in addition to being together on the same campus, students from diverse backgrounds must also learn about each other in the courses that they take and in informal interaction outside of the classroom. For new learning to occur, institutions of higher education have to make appropriate use of structural diversity (italics added).”

I also pointed out that I was clear in the original report that my view of how structural diversity operates is substantiated by other research in higher education.

“The conclusion that racial diversity of a campus operates through the experiences that students have on the campus is supported not only by my own work but also by research contained in Appendix B of my report (See Gurin Expert Rep., Appendix B). The NAS brief ignores this research entirely. Many years of social psychological research on racial contact shows that interracial contact by itself does not have much effect on people unless the contact provides opportunities for them to get to know each other well across racial lines. The critical importance of actual experience for student development is also noted by Pascarella and Terenzini, who reviewed 2,600 studies on the impact of college. In this preeminent book on college impact, (How College Affects Students: Findings and Insights from Twenty Years of Research), they conclude that structural features generally have “an indirect rather than a direct influence on students, their effect being mediated through the institution’s general environment” (Pascarella and Terenzeni, 1991).”

The validity of my approach is further supported by Thomas and Shavelson (supra, at 9), who note that the NAS’s argument about structural diversity “flies in the face of a large body of social science research showing that institutional variables (such as structural diversity) have their effects on individual level variables (such as positive educational outcomes) through other mechanisms….This approach is firmly established in research regarding effects on higher education outcomes, in addition to the psychology sources cited by Gurin.”

What Justice Powell Meant by Diversity

At the basis of this continuing controversy between the NAS and me and my colleagues is the critical question: What did Justice Powell mean by diversity? I addressed this in my original response to Wood and Sherman from which I quote:

“The interpretation of what Justice Powell meant in providing the diversity rationale for the use of race as one of many factors in university admissions is, of course, up to the courts. I believe, nonetheless, that Justice Powell understood that actual interaction with diverse peers is the vehicle by which diversity benefits students.

"Wood and Sherman cite Justice Powell's decisive statement as support for their claim that diversity means only percentage of minority students:
‘The atmosphere of 'speculation, experiment, and creation' — so essential to the quality of higher education — is widely believed to be promoted by a diverse student body. As the Court noted in Keyishian, it is not too much to say that the 'nation's future depends upon leaders trained through wide exposure' to the ideas and mores of students as diverse as this Nation of many peoples.’ Regents of the University of California v. Bakke, 438 U.S. 265, 312-13 (1978).
"Wood and Sherman leave out the reference to a crucial footnote in this passage that appears between 'diverse student body' and 'as the court.' Footnote 48, on page 312 (See 438 U.S. at 312, n. 48.), reads as follows:
‘The president of Princeton University has described some of the benefits derived from a diverse student body: '(A) great deal of learning occurs informally. It occurs through interactions among students of both sexes; of different races, religions, and backgrounds; who come from cities and rural areas, from various states and countries; who have a wide variety of interests, talents, and perspectives; and who are able, directly or indirectly to learn from their differences and to stimulate one another to reexamine even their most deeply held assumptions about themselves and their world. As a wise graduate of ours observed in commenting on this aspect of the educational process, 'People do not learn very much when they are surrounded only by the likes of themselves.'

‘In the nature of things, it is hard to know how, and when, and even if, this informal 'learning through diversity' actually occurs. It does not occur for everyone. For many, however, the unplanned, casual encounters with roommates, fellow sufferers in an organic chemistry class, student workers in the library, teammates on a basketball squad, or other participants in class affairs or student government can be subtle and yet powerful sources of improved understanding and personal growth.’ Bowen, Admissions and the Relevance of Race, Princeton Alumni Weekly 7, 9 (Sept. 26, 1977).
"It is clear from this passage that Justice Powell understood that actual interaction with diverse peers is precisely how campus diversity has its effects. My expert testimony's focus on interaction with diverse peers in what we call diversity experiences follows the logic of Justice Powell as to how a diverse student body improves understanding and personal growth.”

Statistical Interactions: Wood and Sherman’s Conceptual and Statistical Misinterpretations

Conceptual Misinterpretation. In addition to restating their previous arguments about structural diversity, the current Wood and Sherman addendum to the NAS amicus brief to the Supreme Court contains one major new argument, based on their own analysis of CIRP data (the national data on multiple institutions that I used in my analyses). They now contend that my theory demands that I look at the statistical interaction of structural diversity and diversity experiences. In the preface summarizing their current addendum, Wood and Sherman state:

“The University of Michigan expert witness, Professor Patricia Gurin, has contended that the putative benefits of racial preferences in admissions are realized through the increased (italics added) effectiveness of diversity activities ….If Gurin’s theory were correct, there would be interactions between the percent of minorities and her diversity experience variables. We used our on-site access at HERI-UCLA to show that such interactions are either totally absent (i.e., not statistically significant) or are too small to be of any practical significance.”

As they later note, “The interaction terms (which Gurin omits) are crucial for her analysis since she needs to show that the campus experience variables… are more effective (italics added) at higher levels of minority enrollment.”

These statements of Wood and Sherman are incorrect.

  • Nowhere in my original expert report or subsequent statements have I contended that the impact of diversity experiences are more effective (or greater) in institutions where there are larger proportions of minority students.

  • My theory does not predict that diversity experiences would be more effective (or greater) in institutions where there are larger proportions of minority students.

  • The summary effect of diversity experiences in an institution is the product of a (the magnitude of the effect on those who have diversity experiences) and b (the number of students who have them).
  • Thus, the proportion of minorities on a campus is important because it determines how many students (both white and students of color) have the opportunity to have relationships with diverse peers and thus benefit from experience with diversity. When there are only a small proportion of the institution’s students who are minorities, just a few students have the benefit of actual experience with diversity

  • That is why structural diversity is important. It is not important because it increases the magnitude of the effect on those who have diversity experiences.

The analyses testing for statistical interactions that comprise the major new material that Wood and Sherman offer in the addendum to the NAS amicus brief to the Supreme Court are therefore irrelevant as a test of my theory of the educational benefits of diversity experiences.

The amicus brief of the American Educational Research Association, the Association of American Colleges and Universities, and the American Association for Higher Education supports my view.

“Professor Gurin has never argued that structural diversity should have a direct impact on educational outcomes; nor has she ever argued that the effect of experience with diversity should be greater on campuses with the largest percentages of minorities. Instead, she has consistently argued that the presence of diverse peers provides the possibility of having actual experiences with diversity. . . Under Gurin’s model, structural diversity affects the number of students who will have the diverse experiences and gain educational benefits, not the magnitude of the effects. The NAS analyses are thus irrelevant to Gurin’s basic theory.” (p 9, 11)

Statistical Flaws. Even if Wood and Sherman’s argument were relevant, it is not possible to judge the adequacy of the analyses and interpretations of the findings they present in the addendum (pages 15-26). Their textual descriptions of their analyses are so incomplete that it is impossible to identify the specific statistical methods employed in the analyses. Wood and Sherman indicate only that they replicated many of my regressions, adding some additional controls and statistical interaction terms of diversity experience variables and percent minorities in the institution. The results presented fail to conform to basic standards for presenting statistical analyses. That is important because such standards provide readers with important information critical for judging the adequacy of the work being presented. Such standards are basic to all scientific inquiries, and are embedded into standard approaches to peer review.

From what they present, we conclude that they entered the diversity experience variables, the % minorities, and interaction terms in one regression, and then they present findings on the significance of both the interaction effects and the main effects (the effects of diversity experience and % minorities separately). It appears that their conclusions about statistical interaction and main effects are drawn from the same regression. They seem not to recognize the fact that when interactions are in a model, the main effects — that is, the separate effects of the variables making up the interactions — are uninterpretable. It is not just that the main effects are uninterpretable when the interactions are statistically significant. Even if the interactions are non-significant, interpretation of the main effects is meaningless when the interactions are in the model (as long as the interaction terms and the main effect terms are correlated as they undoubtedly are in this case). What they should have done is the following: a) if the interactions were significant, they should have graphed and reported the association between diversity experiences and outcomes at various levels of percent minority; b) if the interactions were not significant, they should have removed the interactions from the model, redone the analysis, and reported the main effects. From what Wood and Sherman present in the addendum to the NAS amicus brief to the Supreme Court, it appears that they did not follow these correct procedures.

In conclusion, even if they were correct that my theory calls for statistical interactions (which it does not), what they have provided is grossly inadequate.

There are two other examples of the carelessness and inadequacy in the presentation of their findings. In one of their results (page 18), they indicate that attending a private college is simultaneously a positive (.068) and negative (-.088) predictor of the outcome they studied. In another example of carelessness, among the 14 tables they present as replicating the regressions in my report, they present “race no longer a problem” as an outcome measure. This measure asks students for their agreements/disagreements with the statement that “race is no longer a problem.” Wood and Sherman comment (page 21) that this “is another example of prejudicial wording," and that “those who designed the HERI questionnaires were building their own perspectives into the wording of the items.” While I might argue whether or not the wording is prejudicial, I agree that responses to this question reflect a political and social point of view that would question its use as an educational outcome in my analysis. For that reason I never used this variable as an outcome in any of our regression analyses. I don’t know why they chose to undertake and present this analysis but it is not a “replication” of one of my analyses. To present it as such is inexcusably careless and misleading.

Criticisms of The Educational Outcomes in the Gurin Report

All of the critiques on the plaintiffs’ side have criticized the concepts and measures of educational outcomes in the Gurin Report. In the NAS amicus brief to the Supreme Court, I am criticized again for rejecting traditional measures of outcomes (such as grades, GRE scores) in favor of subjective ones. They further contend that my measures of democracy outcomes are politically biased in a liberal direction and are irrelevant educationally. I responded to these criticisms many times before, namely in my supplemental testimony of January 11, 2001, response to Wood and Sherman, and response to Lerner and Nagai.

Subjective Self-Reports

The Wood and Sherman addendum to the NAS amicus brief to the Supreme Court again criticizes subjective self-reports. I addressed this criticism in my earlier supplemental testimony of January 11 and in my response to the original Wood and Sherman critique, from which I quote.

“Wood and Sherman further criticize my use of self-report measures of student outcomes. Contrary to their assertions, self-assessments are credible and widely accepted methods for measuring learning. As part of the recent national concern with accountability and the need for assessment of educational outcomes, in 1991 the National Center for Education Statistics (NCES) began hosting a series of workshops to examine the feasibility of creating measures of college student learning similar to the NAEP. A result of these workshops, as well as the deliberations of other higher education groups, has been the recommendation that, given the difficulties and time that will be required to develop national assessment instruments, alternative measures should be used as proxies for the proposed national assessment. Student self-reports have been particularly strongly proposed as appropriate proxies. For example, NCES contracted with the National Center for Higher Education Management Systems (NCHEMS) to review the research on a variety of possible indicators of college outcomes. One of their conclusions was that self-report data on academic development and experiences have moderate to high potential as proxies for a national test, and as possible indicators for decision-making in higher education (Ewell et al., 1993).

"The recommendation to use self-ratings for the assessment of college student impact is based on a long history of research. In their review of over 2600 studies on the impact of college on students, Pascarella & Terenzini (1991) review and integrate the research on the acquisition of specific academic skills as well as more general cognitive competencies. They review the studies that used self-reports as well as those that used standardized tests of skills and competencies, and of gains in these skills and competencies. They use these studies interchangeably as mutual supports for the conclusions they draw, showing that they do not believe that self-reports and standardized tests should be distinguished as totally different types of assessments. They justify their use of self-assessments by the research they reviewed that indicates that self-reports of learning outcomes are positively correlated with standard tests of achievement (just as I did on pages 5 and 6 of Appendix C of my Expert Report).

"Further evidence for the convergence of self-reports and standardized test measures of learning outcomes comes from the fact that both methods have tended to yield similar results when related to student entrance characteristics and to student college experiences (Pascarella & Terenzini, 1991; Ewell and Jones, 1993). A recent study (Anaya, 1999) should be particularly noted because it comes from analyses conducted on a sub-sample of students who had taken the GRE drawn from the same cohort of CIRP that was used in the Gurin Report. Anaya examines what conclusions would be drawn using GRE, GPA, and student-reported growth as measures of learning. The results show that similar substantive conclusions can be made about the relationships of college experiences to learning outcomes, regardless of which measure of learning was used.”

So-Called Hard Measures of Achievement

Wood and Sherman in the Addendum to the NAS amicus brief again reiterate their earlier criticisms that I failed to use hard measures of achievement. I have responded to this criticism on several previous occasions and quote here from my supplemental testimony of January 11, 2001.

“The NAS brief questions why my report does not use ‘accepted measures of academic achievement.’ One reason is that measures such as grades, drop-out, admission to graduate school, and performance on standardized tests do not measure active, complex thinking or intellectual engagement, the outcomes that I predict that diversity ought to affect and which are valued effects of education. Specific knowledge will fade over time. Engagement in learning and active thinking will not. The other reason is that grades and GRE (or other graduate/professional school tests) themselves have psychometric problems.

"There are two problems with utilizing college grades. First, the course grade assigned to a student rarely takes into account the student’s knowledge before taking the course. Thus, grades are not measures of gain or growth in knowledge. Second, grades are not standardized and thus their meaning varies greatly from institution to institution, course to course, major to major. (Footnote: Because of such limitations, Pascarella and Terenzini explicitly do not include studies of GPA outcomes in the chapter devoted to ‘Verbal, Quantitative, and Subject Matter Competence’ outcomes.) Of course, grades are useful for some purposes, especially for an assessment of how much a student knows, whether from increased learning or from prior knowledge, at a given point in time and in a particular setting. They should not be held out as a perfect, or even better, measure of learning outcomes, and certainly not for the learning outcomes of active, complex thinking and intellectual engagement.

"There are also two problems with using GRE scores as measures of learning during college. First, only a fraction of college graduates take the GRE (or other graduate/professional entrance exams). Second, evidence is clear that the major determinant of performance on the GRE is performance on the SAT. The relationship between SAT at the beginning of college and GRE at the end of college is so large that variation in GRE after screening out the influence of SAT is relatively small. The impact of the four-year college experience could explain only the portion that is left after taking account of the SAT, and this trivializes the impact of college."

Democracy Outcomes

In the NAS amicus brief to the Supreme Court, the outcome and measure of beliefs about diversity and divisiveness are singled out as showing a liberal political perspective. The NAS is wrong on three counts.

  • First, all of the Likert-type (agree-disagree) statements about divisiveness and diversity are written to express the common criticism of critics of diversity, namely that a focus on diversity (or difference) creates division among students and is inimical to the unity needed for democracy. Thus, the questions are presented from a conservative, not a liberal, perspective. For example, students are asked to react to such statements as, “The University’s commitment to diversity fosters more intergroup division than understanding,” and “The University’s emphasis on diversity means that I can’t talk honestly about ethnic, racial, and gender issues.”

  • Second, students can agree or disagree with these statements. We scored the items for rejection of the belief that there is an inevitable conflict between diversity and democracy. Leaders of states all over the world that are either democratic or aspire to democracy have to make democracy work despite ethnic and racial heterogeneity of their citizens. Of course, we could just as easily scored it for the belief that democracy and diversity are incompatible. The results would have been exactly the same, merely with opposite signs for the relationships showing that students with the most experience with diversity rejected that belief.

  • Third, the NAS is wrong as well when they suggest that I ignore or leave out the students who accept the idea that diversity is divisive. The students who are scored low on non-divisiveness are in all of our analyses.

In addition to the charge that the measures show a liberal bias, the NAS has previously charged, and again charges, that the measures of democracy outcomes are irrelevant educational outcomes. I have responded to this charge several times before and quote here from my response to the original Wood and Sherman critique:

“Benjamin Barber, a political scientist who has written widely on democracy and higher education, stresses that all traditional political theory — liberal, republic, and democratic — have viewed citizens as created, not born. He asks the question: "Does a university have a civic mission? Of course, for it is (his emphasis) a civic mission. The cultivation of free community — of civility itself" (Barber, 1998).

"But how does diversity foster civic preparedness? It plays a role in two critically important theories discussed in my Expert Report: Aristotle's theory of democracy that is built on difference rather than on similarity, and Piaget's theory of moral development. Both emphasize the following conditions: the presence of diverse others who bring multiple, and sometimes conflicting, perspectives; discussion among peers who are equals; and discussion under rules of civil discourse.

"Higher education must prepare students today to be leaders in an incredibly and increasingly diverse society. This is exactly what the General Motors amicus brief contends that major corporations are also looking to the University of Michigan to do:
'Diversity in academic institutions is essential to teaching students the human relations and analytic skills they need to thrive and lead in the work environments of the twenty-first century. These skills include the abilities to work well with colleagues and subordinates from diverse backgrounds; to view issues from multiple perspectives; and to anticipate and respond with sensitivity to the needs and cultural differences of highly diverse customers, colleagues, employees, and global business partners.' [See brief of General Motors Corp. as Amicus Curiae, Gratz v. Bollinger et al., No. 97-75231, at 2, (6th Circuit 2000)]

The rationale for including democracy outcomes is therefore about preparing students for a future as leaders of democracy and of our major economic/other societal institutions. It is the mission of higher education to:

  • educate students to take the perspectives of other people;

  • help them grasp that difference can be compatible rather than antithetical to democracy and to perceive commonalities as well as differences between social groups in the United States and in the world;
  • provide opportunities for leadership on the campus that will help them learn how to be citizens and leaders of organizations in a diverse society;

  • expect them to play leadership roles after college.

The amicus brief presented to the Supreme Court by three of the major higher education national organizations (AERA, ACE, AAHE) concur with this assessment. “Indeed, it is arguably the central mission of higher education to educate and produce society’s future leaders, and measuring the civic preparedness of students should be highly relevant to an inquiry about the educational benefits that result from a diverse student body.” (p 12)

Size of Effects

I have made three sets of responses to the NAS critique about the size of effects between diversity experiences and educational outcomes: 1) the effects using single items of outcomes are very consistent; 2) the effects using multiple-item indicators of outcomes show (as would be expected) considerably higher effects; and 3) modest effects spread over a large population are socially important. I quote here from my response to the original critique by Wood and Sherman.

Consistency of Effects

"The effects between any single measure of diversity experience and any single measure of educational outcomes would be expected to be small. Using single-item measures makes it possible to assess level of consistency of statistically reliable effects across many measures. The level of consistency that we found is striking. Wood and Sherman attempt to trivialize the importance of consistency of findings."

In the national study for white students, who constitute approximately 90 percent of the sample, I found that of the possible relationships between experiences with diversity and learning outcomes, 62 percent were statistically reliable; of the possible relationships between diversity and democracy outcomes, 85 percent were statistically reliable. As I said in my original Expert Report, this is an impressive level of consistency across measures in this field. Any one effect of a single measure of diversity experience on a single measure of educational outcomes is bound to be small, especially after instituting various control measures. But using these single-item measures, the results show very high consistency.

The very large literature on college impact supports the importance of consistency of effects since no single college experience “will be an important determinant of change for all students. A majority of important changes that occur during college are probably the cumulative result of a set of interrelated experiences sustained over an extended period of time. Consequently, research that focuses on the impact of a single or isolated experience, a characteristic of most investigations of within-college influence, is unlikely to yield strong effects." (Pascarella and Terenzini, p 610)

Analyses Using Multiple-Item Measures

In my response to the NAS Amicus brief of summer 2000, I presented evidence that subsequent analyses of both the CIRP and Michigan databases in which we formed multiple-item indices of both diversity experiences and of educational outcomes show, as would be expected, larger effects. In my response to Wood and Sherman, I then provided tables of the regressions showing the effects with the multiple-item indices. They knew these when they continued to argue about the size of effects using single items, including the NAS brief to the Supreme Court and the Wood and Sherman addendum to that brief. The authors of the brief and the addendum also had access to an article published in the Harvard Educational Review (HER) 2002 fall issue when they wrote to the Supreme Court. The regressions presented in that prestigious educational journal use multiple-item measures and add Asian American students to the white, African American, and Latino(a) students included in the original tables. The regressions in the published article show coefficients (effects) that range from .057 to .337 in the national study and from .082 to .361 in the Michigan Student Study. In these analyses the regression coefficients represent the effects of diversity experiences, controlling for pre-college-entrance scores on the same outcomes, other pre-college personal characteristics of the students, and in the case of the national study measures of institutional characteristics. In the HER article I and my colleagues stated that:

“The size of these effects is commonly viewed in social science as highly consequential for policy, especially when outcomes and predictors are likely to be measured with substantial random error, as they typically are in studies of college impact. It is widely known that the kinds of processes and outcomes of interest here are difficult to measure with high precision and that measurement error diminishes effect size. Given that the dependent variables in the CIRP analyses were multiple-item scales with calculated reliability estimates, we replicated the analyses for each of the racial/ethnic groups in the national study after applying the standard attenuation correction. In each instance, the results were consistent with those presented here, but with larger regression coefficients and an enhanced level of explained variance. For example, the coefficients and degree of predictability associated with the White student analyses were roughly one-third larger in the attenuation-corrected analyses.” (p 358).

Social Contribution of Modest Effects over Numbers of People

The NAS repeatedly ignores the crucially important fact that modest effect sizes are of large policy importance when large numbers of persons are exposed to those effect sizes. In health policy, for example, very small reductions in risk of disease, when enjoyed by entire populations, are viewed as major contributions to public health. Thus, even if the benefits of diversity experiences are modest in size, the fact that many thousands of Michigan students enjoy those benefits over many years would constitute a significant contribution of the University to the larger society.

The amicus brief of the American Psychological Association to the Supreme Court makes the same point, in addition to highlighting the larger effects that resulted when I used multiple-item rather than single-item measures.

“The NAS critique also fails to recognize that although effect size is widely used in social and behavioral research as a measure of the strength of a relationship, ‘a large effect size is not the only way to demonstrate that an effect is important (Epstein, 1983).’ The key point is that the importance of the findings depends upon the context in which they are made. Small effect sizes may have enormous impact over time.

"For example, today, physicians routinely advise men over a certain age to take an aspirin each day to prevent heart attacks. In the original study that led to this advice, the effect size between taking an aspirin and having a heart attack was .0011 — which resulted in almost twice as many heart attacks in the placebo group as in the aspirin group. The effect was considered so important that the study was discontinued early because it was determined that aspirin could no longer ethically be withheld from the placebo-control group.

"Moreover, in higher education, involving some 15 million college students plus additional graduate students every year, even small benefits can yield substantial rewards.” (pp 20-21)

The Negative Impact of Racial Preferences and Diversity

The NAS amicus brief to the Supreme Court, and the Wood & Sherman addendum to that brief, claim more prominently than any previous critique (excepting the earlier Lerner & Nagai critique) that I do not consider possible negative effects of diversity and affirmative action. They present the argument that there will be increased tension and conflict, increased hostility from those whites who believe that affirmative action is simply reverse discrimination, increased stereotypes by whites interacting with less “adequate” minority students, and increased self-doubts and alienation among minority students.

The accusation that I did not consider the possibility of negative effects of diversity is not true. I specifically noted on page 118 of the Gurin Report: “In the public discourse and controversy over the increasing diversity on our college campuses, critics claiming that diversity has had unfortunate consequences on college campuses have pointed to the supposedly negative nature of interracial interaction on diverse campuses. As I detail in Appendix E, the data from the Michigan Student Study clearly disprove this contention…. White students particularly come from segregated backgrounds but the amount of their contact with students of color increases at Michigan. Moreover, the quality of these interactions is predominantly positive, involving the sharing of academic, social, and personal experiences… the type of cooperative and personal relationships that I have argued produce learning and such democracy outcomes as interracial understanding and perspective taking. In general, this also happens for students of color at Michigan, as detailed in Appendix E.”

I also addressed the possible negative effects of diversity in our response to the critique by Lerner and Nagai.

The Importance of Systematic Rather than Anecdotal Evidence

Despite years of controversy about affirmative action and its impact, and the abundance of anecdotal evidence brought to this debate, there is little systematic, quantitative evidence on the nature of intergroup relationships on college campuses in the United States. Have affirmative action and multicultural activities within and outside of the classroom produced the hostile effect on intergroup relationships that critics claim have subverted the purposes of the proponents of affirmative action? These are important questions that need to be answered by systematic research, and not by assertions and anecdotes from either the proponents or the critics of affirmative action.

A number of questions in the surveys of the Michigan Student Study address these issues. And as we will see, the students’ responses to these questions do not support the critics’ views of the negative state of intergroup relationships on such campuses, at least not on the Michigan campus.

In general, most white students as well as students of color feel that the University of Michigan’s commitment to diversity has had a positive impact on their college experiences, and that their relationships with other racial/ethnic groups on campus have been predominantly positive. About 90 percent of the Asian American, Latino, and white students, and about 80 percent of the African American students agreed with the statement that “My relationships with students from different racial/ethnic groups at the University have been positive.”

Most compelling are the students’ responses to a question that asked them to describe their relationships with the racial/ethnic group they interacted with the most on the Michigan campus. The question presented a list of positive and negative intergroup interactions (e.g., “studied together;” “shared our personal feelings and problems;” “had guarded, cautious interactions”) and asked students to what extent they had these types of interactions with students in this group.

The quality of intergroup interactions that emerged from the responses to this question was positive. From white students’ perspectives, approximately two-fifths to one half have had close relationships (studied together, shared personal feelings and problems) quite a bit or a great deal with Asian American and Latino(a) students. Approximately 15 percent to 30percent of them have had these kinds of relationships with African American students this frequently. Most impressive is how little negative interaction or hostility has been present in these relationships. Only 7 percent of the white students report “tense, somewhat hostile interactions with Latino students; only 4 percent report tension and hostility in relationships with African American students; only 1 percent report this with Asian American students. Similarly, only 1 percent said they had “guarded, cautious interactions” this often with Latino students. The comparable percents for their relationships with African American and Asian American students are three and two percentages.

The quality of relationships reported by students of color with white students are also positive, and very few report having negative interactions as often as “a great deal” or “quite a bit.” The one exception is that more African American students than students in other groups report tension and hostility (15 percent) and cautious, guarded relationships (23 percent) with white students. But even these somewhat larger percentages pale in comparison with the rhetoric of the media and of the critics of diversity about how extensive negative interactions are on college campuses.

Why Do Many People Believe that Racial Tension is Widespread on College Campuses?

Given this evidence about the quality of interracial relationships on the Michigan campus, why do so many people believe that our campuses are rife with racial conflict? Responses to another question in the Michigan Student Study further illuminate the question of why the negative view prevails so widely and over time. A question on “racial climate” presented students with a list of phrases and asked them the extent to which each of them was descriptive of the University of Michigan campus. Over one-quarter (27 percent) of the white students responded that there was “quite a bit” or “a great deal” of "interracial tension on campus.” Although this result about perception of racial tension actually shows only a quarter of white students holding the view that such tension exists at Michigan, this is considerably more than the four to seven percent of white students, as we noted above, who actually had “tense, somewhat hostile interactions” in their own relationships with students of color. The important conclusion that we draw from the Michigan Student Study results is that perception of racial climate does not tell an accurate story about students’ own interracial relationships. Their own personal relationships across race and ethnicity are overwhelmingly positive, and very few involve tension, hostility, or guardedness.

The Problem with Perceptions of Climate

When people report about climate — a campus climate, an industrial climate, a national climate — they are subject to pluralistic ignorance. Pluralistic ignorance is a set of beliefs that are shared by many people but do not reflect their individual experiences. Pluralistic ignorance about campus racial climate is based largely in media depictions of campus climate. Newspapers, magazines, and television give inordinate attention to specific racial incidents — graffiti on walls, uncivil and racially-framed comments, even physical harassment. Such incidents have taken place on various campuses, including the University of Michigan, but they hardly represent the quality of student relationships on a daily basis. It appears that the fairly common belief that there are widespread racial tensions on our college campuses, and especially on those campuses with affirmative action policies, are overgeneralizations from highly publicized, ugly racial incidents that have occurred on some campuses, rather than reflections of the students’ personal day-to-day experiences with interracial relationships.

In addition to the arguments and repetitions of anecdotal evidence about racial tension and hostility, Wood and Sherman present some new data derived from their analyses of the CIRP data in the addendum to the NAS Amicus Brief to the Supreme Court. These analyses show a relationship between selectivity of the school (which they argue is a proxy for affirmative action) and student perceptions of racial conflict in the school. One should not be surprised that the perception of racial tension would be more widespread on campuses where commitment to racial/ethnic diversity is strong since students surely talk more about the meaning of race in America on such campuses. On campuses where there are few students of color and where there is little emphasis on increasing diversity, there would also be less perception of racial tension. Homogeneous environments do not bring issues of race to the forefront. However, all around the world societies are being threatened by racial and ethnic cleavages. Surely, college students today need to learn how to work, live, and be leaders in a diverse society lest our own heterogeneous society and democracy become as hopelessly divided as many in the world are today.

Attention to race and ethnicity does not cause division. Instead, inattention to race and ethnicity in heterogeneous societies where different groups sometimes have different interests causes societal breakdown.

Learning to Deal with Conflict

I have never suggested that conflict is unimportant. Indeed, in the original report, and in a soon-to-be published article in the Journal of Social Issues (Gurin, et al, in press), I described a curricular program at the University of Michigan, The Program on Intergroup Relations, that has a goal of teaching students about theories of conflict and skills for negotiating conflict. In both the Gurin Report and peer-reviewed article, I showed that The Program on Intergroup Relations had a statistically significant effect on student views of conflict, helping them understand that conflict is a normal aspect of social life and can be handled rather than avoided and allowed to fester. This is an important study on the impact of diversity experience on attitudes toward conflict and on other democracy outcomes because it had two unusual methodological strengths: a control group for the students who participated in The Program on Intergroup Relations; and, measures taken both at the beginning and end of the college experience. These two features made it possible to be fairly sure about the causal impact of the Program.

Learning about conflict and how to deal with it are two of the many reasons that many corporations, medical schools and law schools, and former military leaders have entered these cases supporting the University of Michigan. They know that corporate leaders, military leaders, and professionals who work with diverse clients both here and abroad need to know how to deal with cultural differences, including cultural conflict.


Neither the NAS nor Wood and Sherman have refuted my testimony on the educational benefits of experience with diversity.


1       One is by two eminent statisticians at Stanford University: Ewart A.C. Thomas & Richard J. Shavelson, Analysis of Report of Wood & Sherman, Addendum to the National Association of Scholars Amicus Brief (analysis by the Stanford Institute for Higher Education Research). A second is in the Amicus Brief of three leading higher education associations in the United States, the American Educational Research Association, the Association of American Colleges and Universities, and the American Association for Higher Education (.pdf) as AMICI CURIAE in Support of Respondents, Gratz v. Bollinger, et al., No. 02-241 (2003). A third is in the amicus brief of the leading professional organization of psychologists in the United States, the American Psychological Association (.pdf), Gratz and Grutter v. Bollinger et al., No 02-241 & 02-516 (2003).

2       I noted this issue in response to Lerner and Nagai, but referred to the extended treatment in the response to Wood and Sherman.


Brief of the American Educational Research Association, et al. as Amici Curiae in Support of Respondents, Gratz v. Bollinger et al., No. 02-516 (2003).

Brief Amicus Curiae of the American Psychological Association in Support of Respondents, Grutter and Gratz v. Bollinger, et al., Nos. 02-241 & 02-516 (2003).

Anaya, Guadalupe (1999). College Impact on Student Learning: Comparing the Use of Self-Reported Gains, Standardized Test Scores, and College Grades. Research in Higher Education, 40 (5), 499-526.

Barber, Benjamin (1998). A Passion for Democracy. Princeton: Princeton University Press.

Bowen, William. Admissions and the Relevance of Race. Princeton Alumni Weekly 7, 9 (September 26, 1977)

Epstein, Seymour. Aggregation and Beyond: Some Basic Issues on the Prediction of Behavior, Journal of Personality 360 (1983)

Ewell, Peter T. and Jones, Dennis, P. (1993). Actions Matter: The Case for Indirect Measures in Assessing Higher Education's Progress on the National Educational Goals. The Journal of General Education, 42. 123-148

Brief of General Motors Corp. as Amicus Curiae, Gratz v. Bollinger et al., No. 97-75231, at 2 (6th Cir. 2000)

Gurin, Patricia. The Compelling Need for Diversity in Higher Education

Gurin, Patricia. Evidence for the Educational Benefits of Diversity in Higher Education: Response to the Wood & Sherman Critique by the National Association of Scholars of the Expert Witness Report of Patricia Gurin in Gratz, et al. v Bollinger, et al. and Grutter v Bollinger,et al., May 30, 2001

Gurin, Patricia. Evidence for the Educational Benefits of Diversity in Higher Education: An Addendum (Response to Lerner & Nagai critique), July 2, 2001

Gurin, Patricia, Dey, Eric, Hurtado, Sylvia, and Gurin, Gerald. Diversity and Higher Education: Theory and Impact on Educational Outcomes. (2002) Harvard Educational Review, 72, 3, 332-366

Gurin, Patricia, Nagda, Ratnesh, and Lopez, Gretchen. The Benefits of Diversity in Education for Democratic Citizenship. Gurin, P., Nagda, R., Lopez, G. In press. Journal of Social Issues.

Lerner, Robert, and Nagai, Althea K. A Critique of the Expert Report of Patricia Gurin in Gratz v. Bollinger, (pdf) available at

Pascarella, Ernest T. and Terenzini, Patrick. T. (1991) How College Affects Students" Findings and Insights from Twenty Years of Research. San Francisco: Jossey-Bass

Raudenbush, Stephen K. Expert report on the likely racial composition in various social contexts on campus under alternative admissions policies, (pdf) July 13, 2000

Raudenbush, Stephen K. Expert report on the probability of admissions of underrpresented minority applicants if race was not used as a factor in admissions decisions, (2.2M .pdf)* February 24, 2000

Regents of the University of California v. Bakke, 438 U.S. 265 (1978)

Thomas, Ewart A.C. and Shavelson, Richard J. Analysis of Report of Wood and Sherman, Addendum to the National Association of Scholars Amicus Brief (analysis by the Sanford Institute for Higher Education Research), available at

Wood, Thomas E. and Sherman, Malcolm J. Is Campus Racial Diversity Correlated with Educational Benefits? April 4, 2001, available at

Wood, Thomas E. and Sherman, Malcolm J. Supplement to Race and Higher Education (.pdf), January 2003, available at

Top of page

Questions? Comments? Please send e-mail to
Site last updated: September 5, 2012.   Copyright © 1997–2013 Regents of the University of Michigan.