Experimental designs in public policy evaluation: uses and abuses

The present essay discusses the strengths and, above all, the limits of using experimental and quasi-experimental methods in evaluating public programs. It begins with a brief presentation of classical experimental design, its requirements and related concepts such as internal, external, and counterfactual validity. Next, it addresses the modalities of quasi-experimental evaluation design, which relax the requirements of the classical experiment. Two critical sections point out the ethical, political, and operational limitations of experimental and quasi-experimental designs in program evaluation and, subsequently, the political-institutional motivations for the resilience of this approach in spite of well-known and recurrent robustness problems