Peer effects and the freshman 15: evidence from a natural experiment

This study investigates the importance of peer effects in explaining weight gain among freshman college students. We exploit a natural experiment that takes place on most college campuses in the US--randomized roommate assignments. While previous studies suggest that having an obese spouse, friend, or sibling increases one's likelihood of becoming obese, these social interactions are clearly non-random. We collect data from female students living on campus at a private Midwestern university at the beginning and end of their first year of college. Our findings suggest that the amount of weight gained during the freshman year is strongly and negatively correlated to the roommate's initial weight. Further, our analysis of behaviors suggests that female students adopt some of their roommates' weight-loss behaviors which cause them to gain less weight than they otherwise would have. In particular, we find evidence that this effect may be through influences in eating, exercise, and use of weight loss supplements.