ABSTRACT The paper addresses the empirical significance of the social context in economic decisions. Decisions of individuals who share spatial and social milieus are likely to be interdependent, and econometric identification of social effects poses intricate data and methodological problems, including dealing with self-selection in spatial and social groups. It uses a simple empirical framework to introduce social interactions effects at different levels of aggregation, and examines estimation problems in linear models, the impact of self-selection and of nonlinearities. It also examines neighborhood effects in job matching and proposes a research agenda that offers new techniques and data sources.