An individual's social environment influences many economic and health behaviors. Social network data, consisting of interactions or relationships between individuals, provide a glimpse of this environment and are widely used across the social and behavioral sciences. This talk provides an overview of two challenges that arise when analyzing social network data. First, we discuss the role of survey design in obtaining network data. Collecting network data via surveys is financially and logistically prohibitive in many circumstances, leading to approaches that collect a small, carefully chosen portion of the network. Even when we can observe the complete network, researchers face difficult decisions about the types of relationships to collect and the implications of those decisions on the resulting scientific analysis. A second challenge arises because fundamental features of social structure, such as transitivity (the tendency for a friend of a friend to be a friend), introduce complex dependence structure between individuals. We discuss several statistical approaches to modeling and interpreting this dependence structure.