newsfilter
← Interviews
80,000 Hours2h 2m

#30 - Dr Eva Vivalt on how little social science findings generalize from one study to another

Dr Eva Vivalt, Keiran Harris, Rob Wiblin

Interview

Y Combinator launched a $100 million, three-to-five-year randomized controlled trial providing $1,000 monthly to approximately 1,000 low-income young adults to analyze shifts in time use, health, and well-being while deliberately withholding results until the study's conclusion to prevent narrative distortion. This initiative is situated within a broader academic discourse highlighting the limited generalizability of development interventions and the prevalence of optimism bias among policymakers, who often neglect variance and negative results in favor of optimistic point estimates. Experts argue that despite these methodological challenges and academic incentives prioritizing novelty over synthesis, aggregated expert priors and robust RCTs remain essential for refining evidence-based policy and overcoming the "naturalistic heuristic" in emerging fields like clean meat adoption.