Design Experiments: Engaging Policy Makers in the Search for Evidence about What Works

This article presents an argument for the greater use of design experiments, which can assist policy making because they provide both robust and timely evidence. We discuss their origins in education research, set out the methodology and propose some adaptations to the techniques used in these education studies to foster their application to a range of policy fields and problem areas. Design experiments need to meet two challenges. Can they provide valid evidence? Can they provide evidence that will be used by policy makers? Our argument shows how design experiments are robust when set against the classical canons of scientific study. We further claim that the design experiment approach offers a more viable means to developing evidence-based policy making than other forms of evaluation because of the timeliness of the insights that it provides.