Why and how can we experiment in the educational field ?

This presentation introduces the experimental approach to education and shows how it can support rigorous evaluation. While this approach is often used on a small scale in highly controlled environments, to establish precise learning processes for example, it is more rarely mobilized on a large scale in so-called " ecological " contexts, to measure the potential effects of a school system in its generalization. However, provided that reference situations are observed (and not just experimental schools or classes) and that precise and varied measuring instruments are used, including surveys, tests and administrative data, it is possible to measure the impact of an action and to understand its driving forces in a given social and institutional context. Using examples drawn from French and international research literature, we will show what such experimental evaluations can contribute to public decision-making.