factual

Judgment-based scoring introduces noise

Scoring intellectual discourse involves many judgment calls—whether someone really gave a caveat or steelmanned an opponent—and where there is judgment there is noise, as Kahneman and Sunstein describe, so different judges or the same judge on different days would score differently, requiring noise-reduction effort or AI/machine learning to do it at scale.

factualpending

Speaker

Arnold Kling

Evidence Quote

“this is something raised by a couple of... fantasy intellectuals, namely Daniel Kahneman and Cass Sunstein is the problem of noise. It's a judgment call whether this person really gave a caveat or didn't”

Source

Arnold Kling on Reforming Government and Expertise— EconTalk
Created: 6/13/2026, 12:29:57 AM

Your notes

Loading notes...