Abstract
Recent legal and political actions have challenged the use of race-conscious college admissions policies. Earlier research offers mixed evidence about the link between an institution’s racial/ethnic composition (i.e., structural diversity) and the formation of close interracial relationships, so the present study examines this topic directly for several racial/ethnic groups. Within a large longitudinal data set, structural diversity is positively related to having close interracial friendships and a different-race romantic partner for White students, and no main effects of structural diversity are apparent for Black and Hispanic students. Moreover, structural diversity has increasingly positive effects at higher levels of campus diversity.
Keywords
The racial/ethnic composition of American colleges and universities has long been a contentious issue. Affirmative action legislation was passed in the 1960s to address inequalities that stemmed from discrimination, and the use of race-conscious admissions policies was upheld in the Bakke and Grutter U.S. Supreme Court cases as a means of providing educational benefits for all students. The key assumption of the diversity rationale in these court cases is that the increased representation of students of color on college campuses (i.e., structural diversity) will lead to more frequent interactions among students from different racial/ethnic groups (i.e., informal interactional diversity), which then leads to student learning and retention. This indirect effect of structural diversity has received empirical support (e.g., Chang, 1999); however, other studies have shown that the link between structural diversity and interactional diversity occurs only among White students because students of color can easily interact across difference at predominantly White institutions (Saenz, 2010; Saenz, Ngai, & Hurtado, 2007).
Some research in the social sciences suggests that greater structural diversity may actually lead to fewer interracial friendships, particularly for underrepresented students of color. Homophily describes a tendency in which “contact between similar people occurs at a higher rate than among dissimilar people” (McPherson, Smith-Lovin, & Cook, 2001, p. 416), and racial/ethnic homophily is more pronounced than homophily for any other demographic category (McPherson et al., 2001). At structurally diverse institutions, students of color will often have greater opportunities to meet same-race students, which may lead to their having fewer interracial friendships. More in-group friendships among students of color may also lead to White students’ having fewer interracial friendships. Two high school studies are consistent with this logic: The proportion of same-race students in one’s high school was negatively related to the number of interracial friendships for students from all racial/ethnic groups (Quillian & Campbell, 2003), and the proportion of interracial friendships was lowest at moderately diverse schools (Moody, 2001). Organizational characteristics (e.g., ability tracking) also seem to influence the formation of interracial friendships (Moody, 2001), so these K–12 findings may not generalize to higher education.
Some recent research also has examined relevant dynamics among college students. A simulation study suggested that substantially increasing the number of Hispanic college students would result in Hispanic students’ having a smaller proportion of non-Hispanic friends (Mayer & Puller, 2008). Conversely, other researchers have found that the proportion of students of color at an institution is unrelated to the frequency of positive interracial interactions among students of color (Saenz, 2010; Saenz et al., 2007). Using a different methodological approach, Bahns, Pickett, and Crandall (2012) observed that pairs of students socializing at a large, diverse university were generally more similar to one another (in terms of attitudes and behaviors) than were pairs at less diverse colleges. However, this study did not examine student race/ethnicity, and the sampling procedure confounded structural diversity and institutional size; therefore, it is unclear whether institutional racial/ethnic composition was directly associated with interactional similarity.
Present Study
Using a large, four-year longitudinal data set of 3,098 undergraduates at 28 institutions, the present study examined the relationships between structural diversity and seniors’ having close interracial friendships and a romantic partner from a different race/ethnicity. The multilevel analyses modeled the unique effects of institutional size and structural diversity (using an index that accounts for the representation of several racial/ethnic groups) and controlled for several key variables (e.g., interracial friendships in high school). Moreover, the data set contained approximately equal numbers of Black/African American, Hispanic/Latino, Asian/Pacific Islander, and White/Caucasian students, so analyses were conducted separately for each racial/ethnic group. (See the online supplemental document available on the journal’s website for a description of the study’s method.)
Results and Discussion
Among White students, structural diversity is positively associated with having close interracial friendships and having a different-race romantic partner (see Table 1). These effects are independent of institutional size, students’ interracial friendships in high school, and other student demographics and perceptions. Thus, structural diversity is positively related not only to White students’ informal interracial interactions (Chang, 1999; Saenz, 2010) but also to their steady romantic relationships and close friendships, which are likely to persist well after graduation. These findings suggest that a college’s structural diversity may notably shape postcollege behaviors, attitudes, and values, which supports the rationale for considering race/ethnicity in admissions processes.
Results for Hierarchical Generalized Linear Modeling Analyses Predicting Close Interracial Friendships and Having a Different-Race Romantic Partner
Note. For all analyses, the linear effect of structural diversity is similar regardless of whether the curvilinear effect is included in the model. The dependent variables (close interracial friends and different-race romantic partner) are listed in boldface in the left-hand column.
p < .05. **p < .01. ***p < .001.
Among Black and Hispanic students, structural diversity is not significantly related to having close interracial friendships or a different-race romantic partner. Quillian and Campbell (2003) demonstrated that high school students whose racial/ethnic group constituted a small minority at their school often displayed a strong tendency toward homophily, and this same dynamic may be true on college campuses: Although underrepresented students of color may actively seek same-race friendships for social support, this propensity becomes less necessary (and therefore less pronounced) when a sizable number of same-race students attend that institution. The structural diversity measure used in this study conveys the representation of several groups, with higher values signifying that all students are more likely to interact with racial/ethnic difference. In other words, this form of structural diversity is related to—but not synonymous with—the proportion of Hispanic and Black students.
Among Asian students, structural diversity is not significantly associated with having a different-race partner, but it is negatively linked to having close interracial friendships. Institutional size is also inversely related to Asian students’ having close interracial friendships and a different-race partner. It is not clear why Asians’ interracial relationships are often related to the size and composition of the student body, whereas this is less true for Blacks and Hispanics. Asians constitute the largest racial/ethnic minority group at most institutions in this sample, so they likely have a greater ability to choose same-race friends. Future research should explore this issue in more detail.
In sum, structural diversity is positively associated with close interracial relationships among White students, which is consistent with previous literature examining informal interracial interactions (e.g., Saenz, 2010). Furthermore, the lack of significant effects among Black and Hispanic students runs contrary to the logic that increased structural diversity leads to increased homogeneity within underrepresented students’ close relationships.
Finally, significant curvilinear patterns exist for predicting close interracial friendships for White and Asian students, such that the effect of structural diversity is more positive (or less negative) at higher levels of campus diversity. All eight curvilinear coefficients in Table 1 are positive, which is greater than one would expect by random chance (p < .01). Thus, institutional strategies that solely attempt to achieve a minimal critical mass of students of color on campus may not take full advantage of the nonlinear effects observed here.
Footnotes
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
