Massive learning gaps persist in most developing countries, undermining rapid gains in school attendance. While a pedagogy based on rote learning of facts is often cited as a factor for limited school effectiveness, evidence of success of introducing new pedagogies to teachers is scarce. Indeed, most in-service teacher interventions evaluated in the literature have been ineffective. In this paper, we report on the randomized evaluation in Uganda of an intervention in which teachers were trained in a “learning how to learn” approach. The curriculum, called Preparation for Social Action, trained teachers to teach students to learn like scientists: posing sharp questions, framing specific hypotheses, using evidence and data gathered from everyday life whenever possible. We find evidence that the approach was adopted in class, with dramatic effects on learning: The intervention raised the pass rate in the national exam that determines progression from elementary to secondary school from 51% to 75%, which places the program in the top five percentile of all rigorously evaluated education interventions in terms of learning-adjusted years of schooling per USD.
(co-authored with Nava Ashraf and Abhijit Banerjee)