Matching Methods for Observational and Experimental Causal Inference (Facultad Latinoamericana de Ciencias Sociales)

We show how to use matching methods for causal inference to ameliorate model dependence in observational data – where small, indefensible changes in model specification have large impacts on our conclusions – and to vastly improve the efficiency of randomized experiments. We introduce methods that are simpler, more powerful, and easier to understand than existing approaches. We also show that propensity score matching, an enormously popular approach, accomplishes the opposite of its intended goal – increasing imbalance, inefficiency, model dependence, and bias – and should be replaced with other matching methods in applications. See http://bit.ly/causeI for papers and easy-to-use software to implement all methods discussed.