Abstract
This article leverages the quasi-random assignment of college roommates among freshmen enrolled at a large university in the Northeast between 2012 and 2016 to examine how the racial and ethnic background of first-year roommates shapes subsequent residential choices. Specifically, we investigate the influence of freshman-year roommate composition on the racial and ethnic diversity of students’ sophomore-year dorm rooms and their post-graduation neighborhoods. Our findings indicate that students randomly assigned to a roommate of a different race or ethnicity are more than twice as likely to live in a racially and ethnically diverse dorm room during their sophomore year compared to those who elect to live with a roommate from the same racial and ethnic background as themselves. However, we find limited evidence that these early residential experiences influence levels of diversity in the neighborhoods students elect to live in after graduation. Overall, these results offer mixed support for the contact hypothesis.
Get full access to this article
View all access options for this article.
