Most of the research on intergroup anxiety has examined the impact of people's own anxiety on their own outcomes. In contrast, we show that in intergroup interactions, one's partner's anxiety is just as important as one's own anxiety (if not more important). Using a diary study among college roommates, we show that partners' anxiety predicts respondents' anxiety across time on a daily basis, as well as respondents' interest in living together again the next year. We discuss the importance of taking a relational approach to understanding intergroup interactions.