S4 Ep2: Aggregating evidence

The conversation about evidence-based policy usually asks why good evidence isn't shaping decisions. But we should also be asking, is the evidence base itself actually worthy of shaping policy?In this episode of Ideas in Development, Rafe Meager – Associate Professor at the University of New South Wales and one of the leading meta-analysts in economics – explains why a single paper is not a proof, why we should learn to respect statistical noise, and why a result that holds in Kenya may or may not hold in Ghana.We walk through Rafe’s work aggregating microcredit RCTs, and their research with Noam Angrist and Youth Impact showing that how well a programme is implemented predicts how well it works. Often, the right question for policymakers is not ‘does this work?’ but ‘can my system actually deliver this?’We also discuss the research process in social sciences, some rules of thumb for interpreting a single causal claim, and why the public should be empowered to understand that evidence comes in different qualities.Read the show notes on our Substack: https://ideasindevelopment.substack.com/Check out Rafe’s research: https://sites.google.com/view/rachaelmeager/home