factual

Randomized trials don't guarantee generalization

Randomized trials are often better than observational data but still contain unobservables, and many randomized trial results fail to generalize precisely because we don't understand the underlying causal mechanism; observational similarity for the data you happen to have does not establish that subjects are truly similar at, say, the cellular level.

factualpending

Speaker

Russ Roberts

Evidence Quote

“if you don't understand the causal mechanism that's going on at, say, the cellular level, you don't know that these people really are similar to the other people. They are just similar observationally for what you have data on.”

Source

Pedro Domingos on Machine Learning and the Master Algorithm— EconTalk
Created: 6/13/2026, 12:10:57 AM

Your notes

Loading notes...