Statistical Power in Longitudinal Network Studies

Longitudinal social network studies may easily suffer from a lack of statistical power. This is the case in particular for studies that simultaneously investigate change of network ties and change of nodal attributes. Such selection and influence studies have become increasingly popular due to the introduction of stochastic actor-oriented models (SAOMs). This paper presents a simulation-based procedure to evaluate statistical power of longitudinal social network studies in which SAOMs are to be applied. It describes how researchers can test different possible research designs decisions (e.g., about network delineation and study time) under uncertainty about the prevalence and strength of various social mechanisms. Two detailed case studies illustrate that network size, number of data collection waves, effect sizes, missing data, and participant turnover can have a serious effect on the statistical power of longitudinal social network studies