Diminished access to higher education is arguably one of the more significant predictors of inequality in our society. My research looks to understand who gets to attend selective colleges and universities, in an effort to discover how this process might promote social mobility or reinforce inequality. One of our projects tested the "frog pond" model of college admission decisions. Using data from more than 45,000 applications to three elite universities in the fall of 1997, we found that a given applicant's chances of being accepted are increased if the individual comes from a high school with relatively less talented students.In more recent studies, we find that the athlete admission "advantage" has been growing, while the underrepresented minority advantage has declined. At the same time, we find that preferences for athletes and legacies only mildly displace members of minority groups. The policy impact of such findings is significant. Ongoing attempts to eliminate affirmative action will substantially reduce the share of African Americans and Hispanics among admitted students. The Campus Life in America Student Survey (CLASS), a study of 12,000 freshmen and returning juniors, measures the prevalence and impact of diversity experiences in college. Relying on this data, my research team considers what college and university administrators can do to maximize the educational benefits of diversity.