Orley Ashenfelter, David Card, Using the Longitudinal Structure of Earnings to Estimate the Effect of Training Programs, The Review of Economics and Statistics, Vol. 67, No. 4 (Nov., 1985), pp. 648-660

We study social interactions in the initiation of sex and other risky behaviors by best friend pairs in the Add Health panel. Focusing on friends with minimal experience at the baseline interview, we estimate bivariate ordered-choice models that include both peer effects and unobserved heterogeneity. We find significant peer effects in sexual initiation: the likelihood of initiating intercourse within a year increases by almost 5 percentage points (on an 11% base rate) if one's friend also initiates intercourse. Similar effects are present for smoking, marijuana use, and truancy. We find larger effects for females and important asymmetries in nonreciprocated friendships.