American Political Science Review · 2012 · 198 citations · 33 references
Public HealthBehavioral SciencesCausal InferenceEpidemiologic MethodsExperimental DesignCausal Effect EstimationExternal ValidityNatural ExperimentsTreatment EffectsCausal Effect IdentificationCausal EstimationControlled Experiment DesignExperimental StudiesNatural Experiment EvaluationRandomized ExperimentsTreatment Effects EstimationUnbiased Evaluation
Natural experiments are used to overcome obstacles to causal inference in social sciences, yet even with random assignment, some treatment–control comparisons can be invalid. The authors present a framework that clarifies subtle, often overlooked issues in natural experiments and propose alternative designs to avoid these conceptual difficulties. They illustrate the framework by examining four natural experiments in the literature and suggest alternative research designs for selected examples. The analysis shows that random assignment alone does not yield unbiased causal estimates and that additional, problematic assumptions are required.
Natural experiments help to overcome some of the obstacles researchers face when making causal inferences in the social sciences. However, even when natural interventions are randomly assigned, some of the treatment–control comparisons made available by natural experiments may not be valid. We offer a framework for clarifying the issues involved, which are subtle and often overlooked. We illustrate our framework by examining four different natural experiments used in the literature. In each case, random assignment of the intervention is not sufficient to provide an unbiased estimate of the causal effect. Additional assumptions are required that are problematic. For some examples, we propose alternative research designs that avoid these conceptual difficulties.
33
Statistics and Causal Inference
Paul W. Holland · Journal of the American Statistical Association · 1986 · 5.1K citations
Matching As An Econometric Evaluation Estimator
James J. Heckman, Hidehiko Ichimura, Petra Todd · The Review of Economic Studies · 1998 · 3.8K citations