Power and sample size calculations for evaluating spillover effects in networks with non-randomized interventions

Determining the appropriate sample size for desired statistical power is crucial for reliable research outcomes. However, methods for evaluating the power of estimating spillover effects in sociometric network-based studies with non-randomized interventions remain inadequately explored. This study conducted a simulation analysis to assess how design parameters, such as the number of components, number of nodes, node degree, transitivity, and effect size, affect statistical power. Both simulated networks and a real-world network from the Transmission Reduction Intervention Project (TRIP) were utilized. Simulation results suggest that power increases with more nodes or a larger effect size. Conversely, a higher node degree or greater transitivity leads to reduced power. Highly unbalanced networks, where most nodes are in one component, can drastically reduce power. These findings were specific to the inverse probability weighting estimator employed and its required assumptions.