"As the gold standard for programme evaluation, experimentation is gaining increasing attention from education researchers and policy-makers. In this article I discuss how experimentation can be used to shape education policy moving forward. I first discuss how well-designed experiments can both build upon and inform a general framework for the education production function. Experiments within this framework can be particularly powerful when they draw on a wide range of disciplines including child development, psychology, and behavioural economics. Insights from these areas can help identify underlying mechanisms of the education production function and inform the design of interventions in ways that increase (cost-) effectiveness. Additionally, I argue that there should be a rich array of experiments in education, ranging from lab-like basic research to policy-level efficacy trials. Finally, I discuss the policy-maker's role in integrating experimentation into policy decision-making, including how to address concerns commonly raised with experimentation in education."

Support: Digital