Speaker: Aislinn Bohren is an Assistant Professor of Economics at University of Pennsylvania. More »
Abstract: In recent years, empirical researchers have become increasingly interested in studying settings with interference between units, in which an individual's outcome depends on the behavior and outcomes of others in her group. This paper formalizes the optimal design and analysis of two-stage randomized controlled trials to measure causal estimands in the presence of interference. Our main contributions are to map the potential outcomes framework for causal inference to a regression model with clustered errors, calculate the power of different two-stage designs and derive analytical insights for the optimal design of such experiments. We show that the power to detect average treatment effects declines precisely with the ability to identify novel treatment and spillover effects. We provide software for optimal design.
Last Updated: Oct 14, 2016