Controlling for Latent Homophily in Social Networks through Inferring Latent Locations

Social influence cannot be identified from purely observational data on social networks, because such influence is generically confounded with latent homophily, i.e., with a node’s network partners being informative about the node’s attributes and therefore its behavior. We show that if the network grows according to either a community (stochastic block) model, or a continuous latent space model, then latent homophilous attributes can be consistently estimated from the global pattern of social ties. Moreover, these estimates are informative enough that controlling for them allows for unbiased and consistent estimation of social-influence effects in additive models. For community models, we also provide bounds on the finitesample bias. These are the first results on the consistent estimation of social-influence effects in the presence of latent homophily, and we discuss the prospects for generalizing them.