RobDownSouth
No Kings
- Joined
- Apr 13, 2002
- Posts
- 77,656
Who Called the Election Right and Who Got It Wrong?
Interesting hot take from Newsweek magazine on who got it right and who got it wrong polling the past election.- Many pollsters got national and swing state votes polled pretty well, within margin-of-error.
- Many final predictions by pollsters with phenomenal track records got the pick wrong this time.
- Nate Silver made a fascinating prediction that at the time seemed a basic "cover your ass" ploy: He predicted EITHER candidate would sweep ALL of the swing states or conversely win NONE of the swing states. He was correct.
The overall most accurate pollster by far was the British firm J.L. Partners, retained by the U.K. newspaper The Daily Mail. They were the only firm that recognized, and accounted for, a somewhat unique anomaly detected early on:
Polls try their best to get an accurate representation of the population as a whole: party registration, age, gender, race, education, location, etc. In order to get the best possible representation, they attempt to get a polled voter to complete a "long form" questionairre by phone.
J.L. Partners noticed a clear pattern early on that no other polling firm noticed: Kamala Harris voters gladly provided this demographic information, but the majority of Trump supporters, whether due to ADHD attention spans, embarrassment and/or a "fear and loathing of the lamestream media" (or a combination of any or all of those three) routinely abandoned a call after stating their preference for Trump. J.L. Partners had to use other, much more expensive ways to attempt to determine the demographics of a Trump voter.....but it paid off very nicely as their accuracy was much better than other polling firms.
As rightwing distrust of polls and media grows with each passing election cycle, I suspect election polling will become much more difficult.
