Power to Detect Intervention Effects on Ensembles of Social Networks

• Activity Number: 102
• Type: Invited
• Date/Time: Monday, August 4, 2014 : 8:30 AM to 10:20 AM
• Sponsor: General Methodology
• Author(s): Tracy Sweet*+ and Brian Junker
• Companies: University of Maryland and Carnegie Mellon
• Keywords: social network models ; hierarchical Bayes ; experimentation ; latent space model

Abstract

The hierarchical network model (HNM) is a framework introduced by (Sweet et al., 2013) and (Sweet et al., 2014) for modeling interventions and other covariate effects on ensembles of social networks, such as would be found in randomized controlled trials in community research and mental health studies. In this paper, we develop calculations for the power to detect an intervention effect using the hierarchical latent space model (HLSM), an important subfamily of HNMs. We derive basic convergence results and asymptotic bounds on power, showing that standard error for the treatment effect is inversely proportional to the product of the number of numbers and the square root of the number of networks; a result rather different from the usual effect of cluster size in hierarchical linear models, for example. We explore these results with a simulation study and suggest a tentative approach to power for practical applications.