The purpose of this study was to examine the educational effect of diversity course graduation requirements on undergraduate students' racial attitudes and support for race-based policy at a predominantly White Midwest university. The present research utilized social structural variables as well as classical and contemporary measures of prejudice and racism to analyze undergraduate students' racial attitudes toward African Americans and support for race-based policy. An 85-question survey on the Internet was administered to undergraduate students at two different time periods during the 2000--01 academic year employing two different data collection techniques. Undergraduate students were contacted in their classes and asked to complete an 85-question survey on the Internet that consisted of primarily five-point Likert items with responses ranging from "strongly disagree" to "strongly agree." Factor analyses were used to ascertain the empirical independence of the classical and contemporary measures of prejudice and racism. Similar to some previous studies, even though there were strong correlations between some of the classical and contemporary measures of prejudice and racism, the factor analysis yielded old-fashioned and contemporary measures of prejudice and racism as separate measures that can be empirically differentiated. There was no significant course effect on undergraduate students' level of racial prejudice and racism in the first study, and the number of interracial friendships and political conservatism did not have an impact on this relationship. However, in the second study that used a pre/posttest design, students who had already fulfilled their diversity course requirement and were taking additional race-based courses, were less prejudiced than those students who just started their first race-based course. Egalitarianism and affective prejudice are most consequential in predicting levels of opposition for race-based policy designed to reduce racial inequality. Anti-Black affect (negative stereotypes) and individualism significantly explain symbolic racism. Controlling for racial prejudice, political attitudes, and socio-demographic variables revealed that even though undergraduate students adhere to the basic American values of equal opportunity, they are less likely to support race-based policy.