2009): "Design of Randomized Experiments to Measure Social Interaction E¤ects

We consider the use of randomized experiments to measure social interaction effects. Ran-domization at two levels—across groups and within groups—can resolve an omitted variables problem for a linear-in-means model of endogenous social interactions. We examine how the randomization should be carried out to estimate the coefficients of interest most precisely, and calculate the optimal treatment probabilities under different criteria.