NOVAFRICA Podcast: How Much Should We Trust Observational Estimates? Accumulating Evidence Using Randomized Controlled Trials with Imperfect Compliance

In Episode 4, Season 7 of the NOVAFRICA Sustainable Development Talks, we speak with Professor Jonathan de Quidt from Queen Mary University of London on his paper: “How Much Should We Trust Observational Estimates? Accumulating Evidence Using Randomized Controlled Trials with Imperfect Compliance.”

In this episode, Professor de Quidt discusses the motivation behind the study, the challenge of bias in observational evaluations, and the key assumption of exchangeability. He uses data from 44 development RCTs with imperfect compliance to show how we can adjust confidence intervals to account for uncertainty due to bias in observational estimates.

He also reflects on what these findings mean for policy evaluation and the future of using observational methods in development research.

🎧 Listen to the full episode here