Averaging polls (as RealClearPolitics does) does not necessarily wash out error: if pollsters make the same systematic errors or herd toward each other, the average is no more reliable than a single poll; averaging only clearly helps when polls use the same methodology, in which case combining them is equivalent to a larger sample, and ideally polls should be weighted by the reciprocal of their variability.

causalpending

Speaker

Doug Rivers

Evidence Quote

“in theory you should average waiting by the reciprocal of the variation variability at the poll so if you have a really reliable poll it should be weighted more than one that's unreliable”

Source

Doug Rivers on polling 7/21/2008— EconTalk
Created: 6/15/2026, 9:28:54 AM

Your notes

Loading notes...