Download this article in MS Word. (118K)


The Benefits of Diversity in Education for Democratic Citizenship

Patricia Gurin

University of Michigan

Biren (Ratnesh) A. Nagda

University of Washington

Gretchen E. Lopez

Colgate University

In press, Journal of Social Issues

January, 2003

*Correspondence for this article should be directed to: Patricia Gurin, Department of Psychology, University of Michigan, Ann Arbor, MI 48109; (734) 936-1875; [], Biren (Ratnesh) A. Nagda, School of Social Work, University of Washington, 4101 15th Avenue NE, Seattle, WA 98105; (206) 616-9083; [], Gretchen E. Lopez, Violence Prevention Project, School of Education, Syracuse University, 372 Huntington Hall, Syracuse, NY 13244; (315) 443-4555; [].


The social science statement in Brown v. Board of Education stressed that desegregation would benefit both African American and White children. Eventually, it was recognized that integration, rather than mere desegregation, was important for benefits to be realized. A parallel argument is made in the legal cases concerning affirmative action in higher education: educational benefits of diversity depend on curricular and co-curricular experience with diverse peers, not merely on their co-existence in the same institution (Gurin, 1999, Gurin, Dey, Hurtado, & Gurin, 2002). Positive benefits of diversity were demonstrated in a study comparing students in a curricular diversity program with students in a matched control group (n=174), and in a longitudinal survey of University of Michigan students (n=1670).

The Benefits of Diversity in Education for Democratic Citizenship

The controversies that have surrounded the Supreme Court ruling in Brown v. Board of Education of Topeka, Kansas (Zirkel & Cantor, this issue) apply as well to current debates about the educational value of racial and ethnic diversity, and the importance of diversity in defending affirmative action in higher education. One of the controversies concerns the difference between racial desegregation and racial integration, or the difference between mere contact and actual interaction between students of different racial backgrounds (Pettigrew, 1998). In current debates about the educational role of diversity, some argue that the mere presence on campus of students from varied racial backgrounds must be shown to directly foster educational benefits (Wood & Sherman, 2001). This argument mirrors the early assertion that mere contact of racially diverse students through school desegregation would be beneficial to all students. Eventually it became clear, however, that mere contact through desegregation was not sufficient to produce educational benefits (Zirkel & Cantor, this issue). Just as Allport (1954) had theorized, contact needed to occur under certain conditions — where there was equality in status, existence of common goals, and intimacy of interaction if it was to have positive effects. Educators needed to create a racially integrated learning environment that went far beyond simply putting diverse students together in the same classroom.

These conditions that make intergroup contact positive also help determine now when racial and ethnic diversity has educational benefits. As Orfield (2001) recently summarized in regard to K-12 public education, there is strong evidence of “instructional techniques that increase both the academic and human relations benefits of interracial schooling” (p. 9). Higher education institutions as well need to create curricular and co-curricular opportunities for students to experience genuine racial integration — to interact in meaningful ways and to learn from each other — if diversity is to have a positive educational impact. The presence of diverse students on a campus is a necessary but certainly not sufficient condition for diversity to work in a positive manner. In this article we stress the importance of actual experiences with diversity through cross-racial interaction in classrooms, intergroup dialogues that bring students from diverse backgrounds together to discuss racial issues, and participation in multicultural campus events.

A second controversy that arose from Brown v. Board of Education concerns what kind of benefits may stem from racial integration in education. Many different outcomes have been studied in the fifty years since the Brown decision; many are analyzed in this volume. We focus on preparation for citizenship, which we argue is an important outcome of experience with racial and ethnic diversity just as it was seen as an important aspect of personal development at the time of Brown v. Board of Education (Clark & Clark, 1947; Deutscher & Chein, 1948). We argue that experiences with diversity educate and prepare citizens for a multicultural democracy.

We analyze the impact of curricular and co-curricular experience with racial and ethnic diversity on democratic sentiments and citizenship activities in two field studies: a quasi-experimental study comparing undergraduate participants in a curricular diversity program with a matched control group (n=87 in each group), and a longitudinal survey of University of Michigan students (n=1670).

Democratic Education and Diversity

How do diversity experiences affect the process of learning to become citizens? We contend that students who interact with diverse students in classrooms and in the broad campus environment will be more motivated and better able to participate in a heterogeneous and complex society. The congeniality of democracy and diversity, however, is not self-evident. Neither representational nor participatory conceptions of democracy deal with the issues raised by multicultural educators, namely the cultural dimensions of citizenship and the central tension of modern social life — the tension between unity and diversity (Parker, 1996, p. 104). Critics of multicultural education worry that a focus on identities based on race, ethnicity, gender, class or other social categorizations are inimical to the unity needed for democracy. Critics of democratic citizenship education that ignores these small publics in an exclusive emphasis on a single unity worry that young people will be ill-prepared to be citizens and leaders of an increasingly ethnically and racially diverse nation.

This tension between diversity and unity, however politically charged it is in contemporary United States, is not new. Saxonhouse in Fear of Diversity (1992) describes how pre-Socratic playwrights, Plato, and Aristotle dealt with the fear that “differences bring on chaos and thus demand that the world be put into an orderly pattern.” Plato, Saxonhouse says, envisioned a city in which unity and harmony would be based on the shared characteristics of a homogeneous citizenry (though even he warned against striving for too much unity). It was Aristotle, Saxonhouse (1992, p. 235) argues, who was able to overcome the fear and welcome the diverse. “Aristotle embraces diversity as the others had not.” Aristotle, according to Saxonhouse, saw the city as made up of parts — families, owners, lovers — that would have different and often conflicting ideas about the good and the bad, the just and unjust. She concludes that anyone interested in politics must study, analyze, and incorporate those parts (Saxonhouse, 1992, p. 235). Pitkin and Shumer (1982) stress that what makes democracy work, in Aristotle’s political theory, are two elements that bring those parts and multiple perspectives into political discussion: equality among citizens who are peers (admittedly only free men at the time, not women and not slaves), and relationships that are governed by freedom and discussion under rules of civil discourse. In this framework, multiplicity of perspectives and discourse over conflict, rather than homogeneity and a single, unified perspective, help democracy thrive (Pitkin & Shumer, 1982).

Sociologist Coser (1975) emphasizes similar conditions in a theory of complex social structures. Complex social structures are social situations that are not familiar to us and are often quite discrepant with our past lives. Complex social structures are composed of many rather than a few people who have different, even contradictory, expectations of us. She argues that unfamiliarity, discrepancy, multiplicity, and potential conflict in the complex social structure require people to pay attention to the social situation and challenge them to think or act in new ways. People develop what Coser calls an outward orientation. She showed that people who function in complex social structures develop a deeper understanding of the social world and are better able to function as effective citizens.

Many cognitive developmental theories also emphasize discontinuity and discrepancy. Cognitive growth is fostered when individuals encounter experiences and demands that they cannot completely understand or meet, and thus must work to comprehend and master the new (or at least not completely familiar) and discontinuous demands. Piaget (1971,1975/1985) calls this optimal learning situation one of disequilibrium. Drawing on these theories, Ruble (1994), a developmental psychologist, theorizes that cognitive growth (and other developmental changes) will be stimulated by developmental transitions, such as going to college or taking a new job. Transitions are significant moments for development because they put individuals into new situations involving uncertainty and requiring new knowledge.

The University of Michigan’s racial and ethnic composition presents discrepancy and discontinuity from the pre-college backgrounds of most of its students. At the time (during the 1990s) that the research reported here was conducted, approximately 90 percent of the White students and 50 percent of the African American students attending the university had grown up in neighborhoods and attended high schools that were racially and ethnically homogenous (Gurin, G. 1992). Because of its discrepancy from their past experiences, racial and ethnic diversity offers students at the University of Michigan (and many other institutions that draw largely from racially/ethnically segregated locations) an opportunity for cognitive growth and preparation for citizenship.

Democracy and Diversity at Work: The Intergroup Relations Program

One such program at the University of Michigan is the Intergroup Relations Program (IGR). It offers a curricular program for first-year students that incorporates five conditions these theories suggest are important for making diversity and democracy compatible: the presence of diverse others; discontinuity from pre-college experiences; equality among peers; discussion under rules of civil discourse; and normalization and negotiation of conflict.

Program participants in the study presented here came from diverse backgrounds. Slightly over a quarter were students of color; a third were men; and, thirty percent grew up in states other than Michigan. For nearly all of the students, this amount of diversity was quite discrepant with their pre-college backgrounds. The design of the first course that students take in the program, in addition to lectures, readings, and papers, includes participation in intergroup dialogues. These groups bring together students from two different identity groups that have had a history of disagreement over group-relevant experiences and policy issues (Zuñiga, Nagda & Sevig, 2002). The groups are led by two trained co-facilitators, usually upper-division or graduate students. These groups are comprised of between twelve to fourteen students with roughly an equal number of students from each of two identity groups. Examples include people of color and White people; women and men; African Americans and Jews; gay men, lesbians, bisexuals and heterosexuals; Anglos and Latinos. Students indicate demographic information about themselves and in which intergroup dialogue they would like to participate. Program coordinators assign students to specific groups based on their choices as well as keeping the groups balanced. For seven weeks, these groups engage in weekly two-hour discussions about policy issues that could divide the groups and individuals within the groups.

In the beginning of the groups, students commit themselves to clear ground rules for civil discourse to guide their discussion. They engage with each other in a truly public way that is needed for a diverse democracy to work. Barber (1989) defines public talk as entailing listening no less than speaking; affective as well as cognitive work; drawing people into the world of participation and action; and expressing ideas publicly rather than merely holding them privately. In these intergroup dialogues, students examine commonalities and differences between and within groups. They learn neither to ignore group differences, which some students do in the service of individualism or color-blindness, nor to privilege differences as an end in themselves. They read about and discuss theories of conflict and its impact on intergroup relationships. They engage in intergroup communication processes and practice skills to negotiate conflicts. They identify collaborative actions that the two groups could take by forming an intergroup alliance or coalition, though they do not actually carry out the action (see Zuñiga, et al., 2002).


We hypothesized that participation in this multicultural program would help students learn sentiments and skills that will be needed in a plural democracy. Specifically, we predicted that first-year students who took the initial course in the Intergroup Relations Program, compared to a matched sample of non-participants, as seniors would show greater: perspective-taking; understanding that difference need not be divisive; perception of commonalities in values between their own and other groups; mutuality in learning about their own and other groups; interest in politics; participation in campus politics; commitment to civic participation after college; and acceptance of conflict as a normal part of social life.

Study 1: The IGR Study


This is a longitudinal field study in which two groups of students were surveyed at time of entrance to the University, and surveyed again at the end of the term when the participants took the initial course, and four years later in their senior year. The two groups of students are those who elected the first course in the IGR Program, and a control sample of non-participants matched one for one on gender, race/ethnicity, in-state v. out-of-state pre-college residency, and campus residency. This means that an in-state, African American female participant living in a particular residence hall was matched with an in-state, African American female non-participant in that same residence hall. The control students were drawn from a larger, comprehensive study of the class that entered the University of Michigan in 1990 (the Michigan Study; see Gurin, G, 1992). All of the course participants were also part of the Michigan Study sample. Thus, both the participants and control students had baseline measures that enabled us to control for self-selection in several analyses below. Altogether 174 students, 87 participants and 87 non-participants, were in the first-year study. In the senior year, students were mailed two questionnaires, one from the IGR program and the second from the Michigan Study. Eighty one percent of the sample (140 students) completed at least one of the surveys in their senior year; 70 percent (122 students) completed both senior year surveys. The data analyzed here come primarily from the two senior year surveys, with some responses from the entrance survey used as controls for self-selection.


Perspective-taking was measured with four items (Davis, 1983). An example is “I find it difficult to see things from the ‘other person’s’ point of view.” The response scale ranges from 1 (very much like me) to 5 (not at all like me). This was measured at entrance and four years later. (Cronbach’s a pre-test = .62, post-test = .68; M=3.80, SD=.70)

Non-divisiveness of difference was measured with four items written for the Michigan Student Study to assess how divisive students perceive the emphasis on diversity at the University of Michigan. An example is: “The University’s emphasis on diversity fosters more intergroup division than understanding.” The response scale ranges from 1 (strongly agree) to 5 (strongly disagree). (Cronbach’s a: post-test only = .83; M=2.61, SD=.64).

Perception of commonalities in values across groups was measured specifically for the Michigan Student Study, and was described in the questionnaire as: “People often feel that some groups in our society share many common values, while other groups have few common values. For each of the groups listed below (African Americans, Asian Americans, Hispanics/Latinos, Native Americans, and White Americans), please indicate how their values and your group’s values are similar or different.” The index summing across these judgments of commonality with groups other than one’s own ranges from 1 (much more different than similar) to 4 (much more similar than different). Commonalities in values were measured at entrance and four years later (Cronbach’s a: pre-test =.84, post-test =.86; M=2.60, SD=.74).

Mutuality in learning about own and other groups was measured by agreement/disagreement with statements about one’s own group, and with statements about groups other than one’s own. These statements were positioned at different places in the questionnaire so that students would consider their own and other groups as independently as possible. The response scale for each statement ranges from 1 (strongly disagree) to 4 (strongly agree). The statements about one’s own groups include: “Since coming to college, I have gained greater knowledge of my racial/ethnic group’s contributions to American society” (M=2.47, SD=.86), and “I have thought more about my memberships in different groups” (M=3.10, SD=.69). The statements about other groups are: “Since coming to college, I have enjoyed learning about the experiences and perspectives of other groups” (M=3.38, SD=.59), and ”I have learned a great deal about other racial/ethnic groups and their contributions to American society” (M=2.86, SD=.73). These items were analyzed separately.

Acceptance of conflict as a normal part of social life was measured by asking students to evaluate conflict on eight statements. Factor analysis revealed two factors, a positive and a negative evaluation factor. An example of positive evaluation is: “Conflict and disagreements in classroom discussion enrich the learning process.” An example of negative evaluation is: “The best thing is to avoid conflict.” The response scale ranges from 1 (strongly agree) to 4 (strongly disagree). High scores represent high positive and high negative evaluations of conflict. (Cronbach’s a for the positive index: post-test only = .70, M=3.21, SD=.42; a for the negative index: post-test only = .64, M=1.92, SD=.49).

Interest in politics was measured by agreement/disagreement with four statements that indicate low interest such as: “I do not enjoy getting into discussions about political issues,” and “I do not try hard to keep up with current events.” The response scale for each statement ranges from 1 (strongly agree) to 7 (strongly disagree). High scores indicated high interest in politics. (Cronbach’s a = .67, M = 5.11, SD =1.18).

Participation in campus politics was measured by asking seniors how involved they had been during their years in college in “campus political activities.” The response scale ranges from 1 (not at all involved) to 4 (substantially involved), (M=1.24, SD=.58).

Participation in community service was measured by asking seniors how involved they had been during college in “community services activities on campus or off-campus, such activities as Big Brother/Big Sister, Project SERVE”. The response scale ranges from 1 (not at all involved) to 4 (substantially involved), (M = 2.36, SD=1.10).

Commitment to post-college civic participation was measured by asking seniors how important the following activities would be after college: “influencing the political structure;” “helping my group or community;” “helping to promote racial/ethnic understanding.” The response scale ranges from 1 (not at all important) to 5 (crucially important). (Cronbach’s a =.61. M=2.41, SD=.68).


The predictions were tested in three steps. First, a one-way multivariate analysis of variance (MANOVA) was conducted to determine if the IGR had a significant impact across the fourth-year outcomes. Results revealed significant differences between the participant and control groups on the multiple dependent measures of democratic sentiments and civic activities (Wilk’s l =.755, F (14,82) = 1.896, p = .039). Then t-tests were conducted to assess mean differences on these measures between participants and control students at the end of the 4th year. Finally, for those measures where we also had entrance scores, regressions were run using the pre-test entrance measure and a dummy variable of participation/non-participation as predictors. These regressions control for the possible role of self-selection into the IGR program.


Senior Year Differences between Participants and Control Students

Nearly all of the predicted relationships between program participation and democratic sentiments as well as civic participation during college were supported by the senior comparisons of participants and control students (see Table 1).

The participants as seniors, compared to the matched control students, more frequently expressed democratic sentiments. They showed significantly greater motivation to take the perspective of others. They less often evaluated the University’s emphasis on diversity as producing divisiveness between groups, and in fact showed greater mutuality in their involvements with their own groups and with other groups. During the college years they had thought more about their own group memberships but they had also enjoyed learning about the experiences and perspectives of other groups more than the control students. They also reported having learned more about other racial/ethnic groups and their contributions to American society. They expressed a greater sense of commonality in values about work and family with groups other than their own. In all of these ways, the IGR had fostered an appreciation of both group differences and commonalities. Finally, the participants normalized the role of conflict in social life to a greater extent than had the control students. They had significantly more positive views of conflict, as well as significantly less negative views.

Specifically on civic engagement, Table 1 further indicates that the participants were more interested in politics and also had participated more frequently in campus political activities. However, they had not taken part more frequently in community service activities during college. With respect to the importance they placed on post-college civic activities, the participants were more committed to helping their group or community and helping to promote racial/ethnic understanding, although this proved to be the result of self-selection rather than an effect of the program (see below).

Controls for Possible Self-selection

It is possible that students who participated in IGR might have entered college with stronger democratic sentiments, and if so, the effects of IGR that we have discerned might result from these predispositions and not from the program itself. Our matching procedure controlled several sources of possible self-selection (gender, race/ethnicity, in/out state pre-college residence, and college residence hall). In addition, for eight of the senior questions (representing three concepts — perspective taking, perception of commonality in values, and commitment to post-college civic participation), it was possible to control for identical measures taken at the time students entered the University of Michigan four years earlier. Even after controlling for first year scores as covariates in analysis of variance, the participants as seniors had significantly higher scores than did the matched controls on the measure of perspective taking (F (1,114) = 4.34, p =.04). Similarly, the program is also associated with an increase in their sense of commonality in work and family values with groups other than their own after controlling for how much commonality the students had felt toward these groups when they entered college. Participants, as compared to the matched controls, judged themselves more similar in values to non-membership groups (F (1,96) = 6.82, p = .01). This analysis showed, however, that students who participated in the IGR program were already more disposed than the control students when they entered college toward post-college civic participation. Once their initial motivation to help their group or community and to promote racial/ethnic understanding was controlled, participation in the program had no effect, neither increasing nor decreasing these post-college civic commitments.

Finally, for four other senior measures (representing mutuality of own and other groups), we were able to use a related, though not identical, baseline measure to control for possible self-selection. At time of entrance students were asked how important various possible college experiences were to them personally. A high importance placed on two of these, “Learning about cultures different from my own,” and “Getting to know people from backgrounds different from my own,” might have predisposed students to take part in the IGR and might account for the apparent effect of the program on their involvement with their own and other groups as seniors. However, this proved not to be the case. After controlling for an index of these two first-year measures, the participants as seniors scored significantly higher than the controls on enjoying learning about the experiences and perspectives of other groups (F (1,117) = 11.7, p =.001), thinking about memberships in various groups (F (1,117) = 11.2, p =.001), and learning a great deal about other racial and ethnic groups and their contributions to American society (F (1,114) = 9.9, p=.002). Neither the predisposition measure nor program participation was significantly related to learning a great deal about the contributions of one’s own group(s).

Study 2: The Michigan Student Study

The IGR was designed explicitly as a quasi-experimental field study of diversity and democracy. Other educational activities have also been created to help students make educational use of Michigan’s ethnic and racial diversity. These activities share certain features of the IGR, although they are not part of a coherent undergraduate program. We were interested in whether or not these other educational activities have similar effects to the IGR in fostering democratic sentiments among undergraduates.

One activity, participation in intergroup dialogue, is closely aligned with and actually grew out of the IGR Program. On the Michigan campus, intergroup dialogues are offered within courses beyond those that are offered in the IGR, and also within various campus organizations. They always meet over time, although the time varies from one month to ten weeks depending on the particular course or campus organization. A second activity is participation in campus-wide educational events about the cultures, histories, and politics of various groups in American society. These events expose students to knowledge about race and ethnicity in settings that draw highly diverse audiences. A third is exposure to knowledge about race and ethnicity in formal classrooms. All undergraduates in the College of Literature, Sciences, and the Arts are required to take at least one course before graduating that covers theories and research on race and ethnicity in American society. The Race and Ethnicity requirement (or sometimes called the diversity requirement) reflects the University's strong commitment to use its racial and ethnic diversity in an explicitly educational manner.

A value of examining the impact of these activities, although they were not part of a unified program, is that the Michigan Student Study (MSS) includes a large enough number of students to analyze data from four racial/ethnic groups (African American, Asian American, Latino, and White) separately. Thus, the MSS allows us to see if diversity activities have similar outcomes in all groups. The number of students in the IGR study was too small to allow analyses of separate groups.


As indicated earlier, both the participants and the control students in the IGR were part of the larger, comprehensive study of the class that entered the University of Michigan in 1990 (see Gurin, G, 1992). This made it possible to explore the extent to which the effects of the IGR apply to other diversity experiences that a broader longitudinal sample of students had during the four years of college. The Michigan Student Study is a longitudinal study that followed students from first year through the senior year. The data analyzed here comes from students who were measured both at entrance and at the end of the senior year: European Americans (n= 1129), African Americans (n=187), Latino(a)s (n=88), and Asian Americans (n=266).


Experience with Diversity. The survey instrument that was given to the students as seniors included reports of their experiences with diversity. Students were asked two questions about classes, namely how much exposure they had in classes to information and activities devoted to understanding other racial/ethnic groups and interracial/ethnic relationships, and if they had taken a course that had an important impact on their views of racial/ethnic diversity and multiculturalism. A third measure assessed the number of five annually-held multicultural events (Hispanic Heritage Month, Native American Month, Pow-Wow, Asian American Awareness Week, and Martin Luther King, Jr. Symposium) the student had attended during the four years of college. A fourth asked if the student had participated in an intergroup dialogue.

These indicators of experience with diversity in classrooms, multicultural events, and dialogues seem to capture fairly well the important features of the IGR Program. Accordingly, we formed a summary measure—by standardizing individual items and then averaging across them—of curricular and co-curricular diversity experiences for the students in the Michigan Student Study.

Democratic Sentiments. With only two exceptions, the same measures of democratic sentiments already described for the IGR study were available in the Michigan Student Study dataset as well. The two that were not available in the broader MSS sample are attitudes toward conflict and interest in politics.


The relationship of this diversity experience measure to democratic sentiments and civic activities was analyzed separately for White, African American, Asian American, and Latino(a) students, using multiple regression. In the regression equation, initial position on outcome measures was controlled when available. Gender and in/out state pre-college residence were also controlled to make the analysis parallel to the analysis of the IGR program. [Post-test score = &bgr;1_(Diversity Experiences) + &bgr;2(Pre-test score) + b&bgr;3(Gender) + &bgr;4(In/out-state)_+ &bgr;5(Constant)]. Multicollinearity was not a problem as none of the intercorrelations among these predictors exceeded a correlation of .20.


Table 2 shows the relationships between having had these diversity experiences and measures of democracy sentiments and citizen participation for each of the four groups of students. Several conclusions can be drawn from this analysis. First, the broader campus study clearly supports what we learned about the impact of IGR Program. Across the four groups, there is evidence of a fairly consistent effect of having been exposed to knowledge about racial/ethnic groups and to interaction with students from varied backgrounds in classrooms, events, and intergroup dialogues.

For White students, the index of these diversity experiences was significantly related to perspective taking and also to a sense of commonality in values with African Americans and Latinos, even after adjusting for entrance measures of these same sentiments. It was also significantly related to having learned about both other groups’ and own group’s contributions to American society, and to actual participation in the activities of both their own groups and of other cultural groups. This effect held even after adjusting for motivation to learn about other backgrounds and cultures that the students expressed when they entered college. Furthermore, White students who had experienced diversity in classrooms, events and intergroup dialogues more often than other students who had not experienced such diversity contended that difference is not inevitably divisive but instead can be congenial to democracy. They had been more engaged in citizenship during college through community service campus political activities. They were not more active in student government, however.

For the three groups of color, Table 2 shows that these experiences were also influential in citizenship preparation. Nearly all of the predicted relationships were statistically reliable. One exception is the lack of relationship between this diversity index and perspective taking for the three groups of color. However, as noted in Table 2, there was a significant relationship for African American students between perspective taking and participation in dialogue groups, as well as participation in multicultural events. Thus, the aspect of the diversity experience index that most directly asks students to consider the perspectives of members of other groups did show the expected relationship between diversity and perspective taking for African American students (although not for Asian American or Latino students).

Finally, the broader study illuminates some subtlety in the impact of diversity experiences on perceived commonalities with other groups. It shows that diversity experiences increased the sense of commonality that White students perceived with both African American and Latino students, whereas diversity experiences were not significantly related to the expression of commonality with White students by the three groups of color. Because the sample size in the IGR study was too small to distinguish the sense of commonality different groups of students felt with particular other groups, the differential impact of experience with diversity on White students and students of color could not be discerned. It is important to note, however, that these results from the broader study do not show a statistically significant negative relationship between diversity experiences of groups of color on their sense of commonality with White students. The broader Michigan Student study also shows consistent relationships for all groups between diversity experiences and involvement in their own groups, while the IGR study showed mixed results about this relationship.


The results from both studies demonstrate important consistency in the effects of diversity experience across two situations (a multicultural educational program, and in the Michigan campus at large); across two longitudinal assessments; and across four groups of students (White, African American, Asian American, and Latino(a) students.) It might be asked if the similarity in results across studies is produced by including IGR participants and control students in both sets of analyses. We analyzed the broader campus data with and without the participants and control students of the IGR study, and found no differences in the results of the two analyses. That would be expected, of course, since the IGR students constituted a small proportion of the broader study.

A notable exception to the picture of consistency across groups is revealed in the analysis of perceptions of commonality in the Michigan Student Study where the results differ for White students and students of color. This difference raises the question of why diversity experience does not foster among students of color a stronger sense of commonality with White students. One possible reason is that experience with White students is less novel for students of color than experience with African American, Latino(a), and Asian American students is for White students (Gurin, Peng, Lopez, & Nagda, 1999). This may account for the different impact that diversity experience has. It is also important to stress that the lack of relationship between the diversity experiences of students of color and perceived commonality with White students does not support the contention among conservative critics of multiculturalism that it fosters division among groups. Students of color with the greatest experience with diverse peers show greater, not less, interest in learning about groups other than their own and they perceive less, not more, division among different racial and ethnic groups.

We use the term effects in these conclusions because in many instances it was possible to control for self-selection by using identical measures of the outcomes that were collected when the students first entered college, and in the case of the IGR study by matching participants and non-participants on relevant demographic characteristics. In all instances except one (importance placed on post-college civic activities in the IGR study) where it was possible to control for the student’s initial position on the outcome measures, the difference between the participants and controls, and the relationship between amount of diversity experience and outcomes in the MSS study was statistically reliable. Thus, we feel assured that these differences did not reflect merely a tendency of certain kinds of students to participate in the IGR program or in other diversity experiences on the Michigan campus that were measured in the MSS. A limitation of the studies is that we did not have college entrance measures for all of the outcomes. Moreover, controlling for self-selection by using pre-measures of outcomes as covariates is not the definitive test of causality that random assignment provides. In addition, future research should more closely study the experiences students of color (and specific ethnic groups) have with diversity in education, and when or how this may result in different types of outcomes related to democratic sentiments and participation (e.g., Gaines, this issue; Gurin, Peng, Lopez, & Nagda, 1999; Tatum, this issue).

In conclusion, these studies provide an examination of the potential impact, and promise, of diversity experiences, through curricular and co-curricular activities taking place in higher education today, for democratic citizenship. These studies support the claim by Guarasci and Cornwell (1997) in Democratic Education in an Age of Difference that democratic citizenship is “strengthened when undergraduates understand and experience social connections with those outside of their often parochial ‘autobiographies,’ and when they experience the way their lives are necessarily shaped by others” (Preface, p. xiii). The discrepancy that racial and ethnic diversity on college campuses offers students for personal development and preparation for citizenship in an increasingly multicultural society depends on actual experience that students have with diverse peers. Just as positive educational benefits of racial and ethnic desegregation depended on real integration of children from different backgrounds, higher education institutions have to make use of racial/ethnic diversity by creating educational programs that bring diverse students together in meaningful, civil discourse to learn from each other. In arguing in Regents of the University of California v. Bakke case that the use of race as one of many factors to achieve racial/ethnic diversity was constitutional, Supreme Court Justice Powell (1978) appears to have understood the critical importance of actual experience with diversity. He uses an article written by President Bowen of Princeton University that a great deal of learning 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” (Powell, 1978, p. 412).

For diverse students to learn from each other and become culturally competent citizens and leaders of a diverse democracy, institutions of higher education have to go beyond simply increasing enrollment of student of different racial and ethnic backgrounds. These institutions must also attend to both the quality of campus racial climate and actual interactions among diverse students. As Gurin (1999, p. 41) conveyed in her testimony in support of the University of Michigan in the two legal challenges to its admission policies at the undergraduate level and in its School of Law, the onus is on higher education institutions:

to make college campuses authentic public places, where students from different backgrounds can take part in conversations and share experiences that help them develop an understanding of the perspectives of other people. Formal classroom activities and interactions with diverse peers in the informal college environment must prompt students to think in pluralistic and complex ways, and to encourage them to become committed to life-long civic action. Otherwise, many students will retreat from the opportunities offered by a diverse campus to find settings within their institutions that are familiar and that replicate their home environments.


Allport, G. (1954). The nature of prejudice. Cambridge, MA: Addison-Wesley.

Barber, B.R. (1989). Public talk and civic action: Education for participation in a strong democracy. Social Education, October, 355-356, 370.

Bowen, W. (1977). Admissions and the relevance of race. Princeton Alumni Weekly, 7, 9.

Clark, K. B., & Clark, M. P. (1947). Racial identification and preference in Negro children. In T. N. Newcomb & E. L. Hartley (Eds.), Readings in social psychology (pp. 169-178). NY: Holt.

Coser, R. (1975). The complexity of roles as a seedbed of individual autonomy. In L.A. Coser (Ed.), Groups in contact: The psychology of desegregation. NY: Academic Press, Inc.

Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology 44, 113-126.

Deutscher, M., & Chein, I. (1948). The psychological effects of enforced segregation: A survey of social science opinion. Journal of Psychology, 26, 259-287.

Gaines, Jr., S. O. Color-line as fault line: Teaching interethnic relations in California in the 21st century. Journal of Social Issues, this issue.

Guarasci, R., Cornwell, G.H., and associates. (1997). Democratic education in an age of difference. San Francisco: Jossey-Bass.

Gurin, G. (1992). The Michigan Study: Expectations and experiences of first-year students with diversity. Ann Arbor: University of Michigan, Office of Academic Multicultural Initiatives.

Gurin, P. (1999). Expert testimony in Gratz, et al. v. Bollinger, et al. & Grutter, et al. v. Bollinger, et al., in The Compelling Need for Diversity in Higher Education. Retrieved from University of Michigan website:

Gurin, P., Dey, E. L., Hurtado, S., Gurin, G. (2002). Diversity and Higher Education: Theory and Impact on Educational Outcomes. Harvard Educational Review, 71, 3, 332-366.

Gurin, P., Peng, T., Lopez, G., & Nagda, B. A. (1999). Context, identity, and intergroup relations. In D. Prentice & D. Miller (Eds.), Cultural divides: Understanding and overcoming group conflict (pp. 133-172). NY: Russell Sage.

Orfield, G. (2001, July). Schools more separate: Consequences of a decade of resegregation. Retrieved from Harvard University, The Civil Rights Project Web site:

Parker, W.C. (1996) Advanced ideas about democracy: Toward a pluralistic conception of citizenship education. Teachers College Record, 98, 104-125

Pettigrew, T.F. (1998) Prejudice and discrimination on the college campus. In J.L. Eberhardt & S.T. Fiske (Eds.), Confronting racism: The problem and the response (oo 263-279). Thousand Oaks: CA: Sage

Piaget, J. (1971). The theory of stages in cognitive development. In D.R. Green, M.P. Ford, & G.B. Flamer (Eds.), Measurement and Piaget (pp 1-11). NY: McGraw-Hill.

Piaget, J. (1975/1985). The equilibrium of cognitive structures: The central problem of intellectual development. Chicago: University of Chicago Press.

Pitkin, H. F., & Shumer, S. M. (1982). On participation, Democracy, 2, 43-54.

Powell, L. (1978) Regents of the University of California v. Bakke. 439 I/S/ 312. 98 S. Ct. 2760

Ruble, D. (1994). A phase model of transitions: Cognitive and motivational consequences. Advances in Experimental Social Psychology, 26, 163-214.

Saxonhouse, A. (1992). Fear of diversity: The birth of political science in ancient Greek thought. Chicago: University of Chicago Press.

Tatum, B. D. Family life and school experiences: Factors in the racial identity development of Black youth in White communities. Journal of Social Issues, this issue.

Wood, T.E. & Sherman, M.J. (2001). Is campus racial diversity correlated with educational benefits? Retrieved from California Association of Scholars Web site:

Zirkel, S., & Cantor, N. 50 Years after Brown v. Board of Education: The promise and challenge of multicultural education. Journal of Social Issues, this issue.

Zúñiga, X., Nagda, B. A., & Sevig, T. D. (2002). Intergroup dialogues: An educational model for cultivating engagement across differences. Equity and Excellence in Education, 35(1), 7-17.


Table 1

Democratic Sentiments and Civic Activities in the Fourth Year of College

Participant M Control M
Democratic Sentiments
Perspective-taking (range: 1-5) 3.91 (.66) 3.71* (.73)
Non-divisiveness of difference (1-5) 2.71 (.57) 2.51* (.68)
Perceived commonality with other groups (1-4) 2.79 (.85) 2.44*** (.57)
Positive evaluations of conflict (1-4) 3.34 (.36) 3.09*** (.43)
Negative evaluations of conflict (1-4) 1.85 (.43) 1.99* (.54)
Mutuality in learning about own and other groups (1-4):
• Enjoyed learning about experiences of other groups 3.56 (.50) 3.20*** (.62)
• Thought more about my memberships in different groups 3.31 (.53) 2.91*** (.75)
• Learned about other groups and contributions to society 3.05 (.65) 2.68*** (.75)
• Gained knowledge of my group’s contributions to society 2.58 (.84) 2.38 (.87)
Civic Activities during College
Interest in politics (1-7) 5.29 (1.12) 4.94*** (1.23)
Participation in campus politics (1-4) 1.34 (.72) 1.14* (.39)
Participation in community service (1-4) 2.49 (1.16) 2.25 (1.19)
Civic Activities Anticipated Post-College
Helping my group or community (1-5) 4.06 (.90) 3.78**(.90)
Helping to promote racial/ethnic understanding (1-5) 3.43 (1.01) 3.16* (1.10)
Influencing political structure (1-5) 2.79 (.99) 2.70 (87)
Note. Standard deviations presented in parentheses. Higher scores indicate higher attribute.


Table 2

Regression analysis of the effects of diversity experiences on democratic sentiments and civic activities of college seniors (Michigan Student Study)

  B SE B Beta
Perspective Taking
Whites (R2=.280) .135 .031 .132****
African Americans (R2=.186) .109 .102 .1081
Asian Americans (R2=.273) .049 .065 .049
Latino(a)s (R2=.143) .007 .139 .007
Sense of commonality: White students with groups of color
With African Americans (R2=.093) .296 .038 .288****
With Asian Americans (R2=.051) .091 .072 .042
With Latino(a)s (R2=.079) .176 .071 .094**
Sense of commonality: Students of color with White students
African Americans (R2=.018) .183 .221 .095
Asian Americans (R2=.059) .005 .128 .003
Latino(a)s (R2=.125) -.313 .314 -.147
Mutuality: Participate in own group activities
Whites (R2=.010) .068 .020 .100***
African Americans (R2=.076) .356 .091 .276****
Asian Americans (R2=.077) .402 .086 .278****
Latino(a)s (R2=.146) .509 .135 .382***
Mutuality: Participate in other group’s activities
Whites (R2=.061) .171 .020 .247****
African Americans (R2=.097) .384 .088 .312****
Asian Americans (R2=.053) .251 .067 .230***
Latino(a)s (R2=.130) .445 .126 .361***
Mutuality: Learned about own group’s contributions
Whites (R2=.044) -.239 .039 -.209****
African Americans (R2=.054) -.288 .090 -.233***
Asian Americans (R2=.098) -.321 .065 -.312****
Latino(a)s (R2=.143) -.471 .131 -.378***
Mutuality: Learned about other group’s contributions
Whites (R2=.117) .553 .045 .341****
African Americans (R2=.098) .513 .115 .314****
Asian Americans (R2=.125) .514 .086 .348****
Latino(a)s (R2=.133) .540 .151 .364***
Whites (R2=.050) .273 .035 .233****
African Americans (R2=.043) .179 .063 .208**
Asian Americans (R2=.031) .192 .066 .176**
Latino(a)s (R2=..032) .193 .098 .180*
Political Participation: Student Government
Whites (R2=.003) .048 .024 .050
African Americans (R2=.013) .098 .064 .113
Asian Americans (R2=.007) .053 .038 .86
Latino(a)s (R2=.018) .068 .055 .133
Political Participation: Campus political activities
Whites (R2=.043) .175 .025 .206****
African Americans (R2=.034) .148 .059 .184**
Asian Americans (R2=.125) .295 .048 .354****
Latino(a)s (R2=.042) .203 .100 .205*
Political Participation: Community Service
Whites (R2=.033) .386 .062 .183****
African Americans (R2=.029) .254 .110 .169*
Asian Americans (R2=.094) .455 .088 .307****
Latino(a)s (R2=.020) .209 .157 .143
* p < .05,** p < .01,*** p < .001
Note. The analyses of perspective taking and sense of commonality with members of other groups included pre-measures of these outcomes taken when the students entered college. The analysis of mutuality included the same pre-measure of motivation to learn about people from different backgrounds and cultures used in the IGR study. These pre-measures were used as controls in the analyses of these outcomes, thus these analyses provide a reasonable assessment of effects.
1 The effects of dialogue and multicultural events, without classroom exposure, are statistically reliable for African American students. The relationship between perspective taking and an index with just those two diversity experiences has a B of .233, SE B of .101, and a Beta of .223*.

Biographical Information for Authors

PATRICIA GURIN, Ph.D. is the Nancy Cantor Distinguished Professor, Emerita, of Psychology and Women’s Studies at the University of Michigan. She was the expert witness for the University on the educational value of diversity in its defense of its admission policies at the undergraduate level and in its Law School.

BIREN (RATNESH) A. NAGDA, Ph.D., MSW, MA is Associate Professor of Social Work and Director of the Intergroup Dialogue, Education and Action (IDEA) Training and Resource Institute at the University of Washington. His research and teaching interests focus on cultural diversity and social justice, intergroup dialogue, and empowerment-oriented social work practice and education.

GRETCHEN E. LOPEZ, Ph.D. previously held position of Assistant Professor of Psychology and Africana and Latin American Studies at Colgate University and is now Research Director of the Syracuse University Violence Prevention Project — a federally funded intervention and research program examining the effectiveness of school-wide prosocial behavior programs for elementary school students. Her research interests include intergroup relations, multicultural education, social psychology of gender, and youth development.