Bayesian longitudinal social network models: An implementation in R using STRAND

Across the social and biological sciences, researchers study social networks that are often longitudinal in nature, featuring the same set of individuals interacting with one another over long periods of time. Such network data may be collected via panel studies in human communities, or via long-form behavioral observation programs in animal communities. Important research questions often hinge on how dyadic ties or flows in the network at one time-step are associated with dyadic ties or flows at subsequent time-steps. Similar questions focus on the relationship between nodal characteristics across time-steps: for example, is an individual with a high out-degree at one time-step more likely to have a high in-degree in the next? These questions are effectively addressed by longitudinal extensions of generative social network models like the Social Relations Model (SRM). Here, we present a novel longitudinal extension of the SRM, provide an implementation of the model in the STRAND R package, and provide tutorials teaching end-users how to fit the model to their own data. We provide worked examples of data analysis, parameter visualization, and results interpretation, using data-sets from both social science and animal behavior. The software package allows end-users to deploy complex longitudinal network analysis models using only simple, base-R model syntax.