
How to be a data detective (Tim Harford)
Rationally Speaking Podcast
Randomise Trials - Can You Learn Anything From Randomism?
i've had interesting debates with angus deaton am and russ roberts. Both of them are like, no, you learn nothing from randomis control trials. i don't fully understand the reasons for angus's scepticism about randomis control trialsyes, it's possible that. I don't either. Yes, on we knothee's t we so wi,. There's one line of icism that basently says they don't ever measure anything that really matters. This is aunt prichett's idea, like, shaw runs ome randomise trials, but you're only ever going to identify niche variables of niche interest. And then there are other objections
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.