S3E27: Propensity Scores -- I Meant To Do That!
Quantitude
00:00
The Biggest Critique With Causal Inference Methods
The biggest problem with propensity scores and causal inference methods that rely on cobert adjustment in general, is that you really have to collect every single variable. Confounding is just what we call this situation where there are variables that cause the treatment and the outcome. If you can collect all these variables and adjust for them in a specific way, then you can claim that your effect estimate is causal.
Transcript
Play full episode
