Randomized control trials: what are they, why are they promoted as the gold standard for casual identification and what can they (not) tell us?

Over the past decade scholars, research practitioners, and donors have shown an increasing interest in using randomized control trials (RCTs) to evaluate the effectiveness of a wide range of social and infrastructure programs in developing countries. Much as in medicine where randomized control trials are standard practice, the policy community has turned to RCTs in response to increasing calls for evidence-based policy. As a result, quasi-experimental approaches to program evaluation have come under increasing scrutiny and been strongly criticized by advocates of RCTs. In fact, in some policy and scholarly circles the term RCT has become nearly synonymous with “rigorous” and has been hailed as the primary, if not sole, arbiter of “hard” evidence in the policy process.