Primary Submission Category: Design-Based Causal Inference
Inference for Group Interaction Experiments
Authors: Ye Wang, Cyrus Samii, Jiawei Fu,
Presenting Author: Cyrus Samii*
A common experimental research design is one in which individuals are put into groups and then interact within the groups under different group-level treatment conditions. We present methods for design-based inference for such “group interaction” experiments. A key consideration is that group interaction implies potential interference, which yields dependencies that should be accounted for when making inferential claims. We show that when interference is present, standard cluster robust inference is super-population consistent in accounting for such dependencies for inference on marginalized causal effects that account for interference. When interference is not present but groups are formed through individual random assignment, individual-level heteroskedasticity robust inference is consistent for inference on the usual average treatment effect. We prove the consistency and asymptotic normality of the difference-in-means estimator for group interaction experiments and extend our framework to cases with restricted group compositions and covariate-conditional effects. Finally, we validate our theoretical propositions through simulation exercises and a replication study of such experiments in social science.