This article discusses the promise and pitfalls of randomized and rigorously
matched experiments as a basis for policy and practice in education.
It concludes that a focus on rigorous experiments evaluating replicable programs and practices is essential to build confidence in educational research among policymakers and educators.