We develop an empirical discrete choice model that explicitly allows for endogenous social interactions. We analyze the issues of multiple equilibria, statistical coherency, and estimation of the model by means of simulation methods. In an empirical application, we analyze a data set containing information on the individual behavior of some 8000 high school teenagers from almost 500 different school classes. We estimate the model for five types of teen discrete choice behavior: smoking, truanting, moped ownership, cell phone ownership, and asking parents' permission for purchases. We find strong social interaction effects for behavior closely related to school (truanting), somewhat weaker social interaction effects for behavior partly related to school (smoking, moped and cell phone ownership) and no social interaction effects for behavior far away from school (asking parents' permission for purchases). Intra-gender interactions are much stronger than cross-gender interactions.