Testing intergroup contact theory through a natural experiment of randomized college roommate assignments in the United States.

Many colleges and universities seek to leverage the promise of intergroup contact theory by adopting housing policies that randomly assign first-year students to roommates, with the goal of increasing intergroup contact. Yet, it is unclear whether random roommate assignment policies increase cross-race contact, whether this (potentially involuntary, but sanctioned by authorities) contact improves racial attitudes or behaviors, or how these effects may differ for racial majority and minority students. The present studies used a natural experiment of random roommate assignment to directly test roommate relationship, attitudinal, and behavioral changes based on roommate race. Across three samples drawn from two student cohorts, the random assignment policy increased the likelihood that students had a cross-race roommate because without the policy, students tended to self-segregate by race. Moreover, selecting (Study 1) or being randomly assigned (Study 2) a cross-race roommate was associated with having more racial outgroup friends and demonstrating more positive verbal and nonverbal behavior during a novel cross-race interaction (Study 3). There were no roommate group (same vs. cross-race roommates) differences in relationship quality, and the results were largely independent of participant race. These findings suggest randomized roommate assignment is a promising avenue for universities to promote cross-race contact amid persistent racial segregation on college campuses with limited negative consequences for relationship quality. (PsycInfo Database Record (c) 2024 APA, all rights reserved).