John Ioannidis on Statistical Significance, Economics, and Replication
EconTalk
00:00
Is an Observational Study Having Any Real Chance of Getting It Right?
"I am highly skeptical of these observational studies," he says. "There may be many situations where they could give an answer that is fairly reliable" But some designs are just not to be used, and should be abandoned for certain types of questions,. He adds: "The noise is many orders of magnitude more than the signal"
Transcript
Play full episode
