QUANTITATIVE RESEARCH METHODS WORKSHOP
Abstract/Description: Many empirical analyses of causal effects rely on non-refutable assumptions—where the data alone cannot determine whether the assumptions are true. The literature on robustness and sensitivity has developed in response to this problem. This talk will survey several recent papers by the speaker on this topic. In particular, we will outline a general approach to analyzing robustness of one’s conclusions to deviations from baseline identifying assumptions. We then illustrate this approach by studying two classic settings: selection-on-observables and instrument variables. Finally, we will compare and contrast this approach with several alternatives available in the literature.
Matthew Masten is an Assistant Professor in the Department of Economics at Duke University. He received his Ph.D. from Northwestern University. His research focuses on identification and causal inference. In particular, much of his work focuses on identification using instrumental variables. More recently he is working on robustness and sensitivity analysis.