The politics of random assignment: implementing studies and impacting policy

This article presents lessons from social experiments in the US over 30 years testing employment and training, welfare reform and social service programmes and systems. It discusses the challenges in implementing a random assignment study and strategies to overcome them, and also sets out lessons for ensuring that an experiment informs and affects policy. In laying out the ingredients for success, the article argues that creative and flexible research design skills are essential, but that just as important in a complex, real‐world context are operational and political skills, applied both to marketing the experiment in the first place and to helping interpret and promote its findings down the line.