Does enrollment diversity improve university education?

Debate over the value of admissions policies designed to increase racial diversity at American colleges and universities has relied on surveys of students, and sometimes faculty and administrators, which are designed to measure educational environments and intergroup relations. This article evaluates the role of survey research in supporting the diversity model-the argument that increased racial diversity in college enrollment both enriches the educational experience for students of all racial and ethnic backgrounds and also improves relations between students of different races. We found that much of the supporting data suffers from methodological defects, which range from poor item formulation to interpretive problems linked to selective recall and social desirability response set. We utilized a more indirect approach that asked members of the university community non-controversial questions about their perceptions and experiences, and then correlated their responses with an independent empirical measure of diversity. Data were obtained from a 1999-2000 survey of a structured random sample of over 4,000 US students, faculty, and administrators, and from the National Center for Education Statistics. When student, faculty, and administrators; evaluations of the educational and racial atmosphere were correlated with the percentage of minority students enrolled at a college or university, the predicted positive associations of educational benefits and inter-racial understanding failed to appear. Thus, the findings failed to support the argument that enrollment diversity improves the education and racial milieu at American colleges and universities. Our study also raises questions about survey instruments and designs that affect inferences about respondents' beliefs and behavior.