Published empirical work hides the 'kitchen' process: researchers who run a single specification usually get nothing significant, then convince themselves they had the wrong specification and iteratively add or drop variables and nonlinear terms until they sculpt a publishable result, so readers cannot judge whether the finding is robust.

causalpending

Speaker

Russ Roberts

Evidence Quote

“you don't just sit down and say I think these are the variables that count... if you do that you don't get anything that's significant which happens tons of times you convince yourself rather easily that you must have had the wrong specification”

Source

Ed Leamer on the State of Econometrics 05/10/2010— EconTalk
Created: 6/17/2026, 9:55:26 AM

Your notes

Loading notes...