How do global sources of information such as mass media outlets, state propaganda, NGOs, and national party leadership affect aggregate behavior? Prior work on this question has insufficiently considered the complex interaction between social network and mass media influences on individual behavior. By explicitly modeling this interaction, I show that social network structure conditions media's impact. Empirical studies of media effects that fail to consider this risk bias. Further, social network interactions can amplify media bias, leading to large swings in aggregate behavior made more severe when individuals can select into media matching their preferences. Countervailing media outlets and social elites with unified preferences can mitigate the effect of bias; however, media outlets promulgating antistatus quo bias have an advantage. Theoretical results such as these generate numerous testable hypotheses; I provide guidelines for deriving and testing hypotheses from the model and discuss several such hypotheses.