Educational experiments.

Summary Among the methods used to evaluate educational policies, randomized experiments, also known as controlled experiments, are currently of particular interest. In the field of education, randomized experiments can be used to evaluate the effects of a pedagogical innovation, a modification of school rhythms, the introduction of support courses, etc., even before such interventions are generalized or not. Advocates of experiments often point to the statistical biases inherent in evaluations conducted with non-experimental observational data, and in particular to selection biases, which controlled experiments can reduce.