Alongside these behavioral insights, we are increasingly using data, well-conducted studies, and evidence-based decision making to evaluate and sharpen the effectiveness of social interventions. This, of course, is nothing new. Some 150 years ago, Florence Nightingale revolutionized medical care in England the same way. We think of Nightingale as a kind lady with a lamp, but she was closer to a data analyst. She wrote: “To understand God’s thoughts we must study statistics.” And she used data to force changes that substantially cut death rates in hospitals and military barracks and led to the formalization of nursing. Today, the social sector remains far from evidence-based. For example, much of the math and writing instruction in American schools is not supported by evidence of what works. Even in medicine, the evidence-based movement is only two decades old. (It was only in the 1960s that the U.S. government began requiring pharmaceutical companies to demonstrate “substantial evidence of effectiveness” for new drugs.) Since the 1970s, a few standout groups like MDRC have pushed for more rigorous testing of social programs. But until recently, if you ran an after-school or Head Start-type program, or a program that claimed to reduce juvenile crime or prevent teen pregnancy, you could keep turning the crank for years without having to furnish proof that you were achieving results. That is still possible, but it’s getting tougher. Private and public funders, as well as groups like M.I.T.’s Poverty Action Lab, the Cochrane and Campbell Collaborations, the Coalition for Evidence-Based Policy, the Washington State Institute for Public Policy, the Institute for Health Metrics and Evaluation, are increasing demands for more, and better, evidence. The upshot is that we’re now in a better position to recognize what works and what doesn’t in a variety of areas — like which methods to reduce child abuse and prevent unwanted teen pregnancies appear most effective, or what studies tell us about how to improve the teaching of math or writing, or which police tactics are most effective at reducing crime. In both the Bush and Obama administrations, we’ve seen early efforts to incorporate evidence in policy making at the national level. People with good intentions have long worked on social problems in the dark; increasingly they are being asked to prove that they are getting somewhere. This is a departure from the past. And like the scientific revolution, if the movement grows, it should foster considerable innovation.