Testing for two-stage experiments in the presence of interference

Seminar | February 15 | 4-5 p.m. | 1011 Evans Hall

 Guillaume Basse, Harvard University

 Department of Statistics

Many important causal questions concern interactions between units, also known as interference. Examples include interactions between individuals in households, students in schools, and firms in markets. Standard analyses that ignore interference can often break down in this setting: estimators can be badly biased, while classical randomization tests can be invalid. In this talk, I present recent results on testing for two-stage experiments, which are powerful designs for assessing interference. In these designs, whole clusters (e.g., households, schools, or graph partitions) are assigned to treatment or control; then units within each treated cluster are randomly assigned to treatment or control. I demonstrate how to construct powerful tests for non-sharp null hypotheses and use these results to analyze a two-stage randomized trial evaluating an intervention to reduce student absenteeism in the School District of Philadelphia. I discuss some extensions to more general forms of interference, as well as some current challenges.