It ain't a shy Trump thing if Susan Collins is beating her polls by a ton... That is, the problem doesn't seem to be Trump specific...
-
Show this thread
-
But to judge the extent of the polling errors we really got to wait until vote counts are more complete. I'll leave you with a scary thought: I'm not sure any individual poll average miss in a swing state was outside the 95% confidence interval for state polling.
63 replies 43 retweets 745 likesShow this thread -
This Tweet is unavailable.
-
Replying to @zeynep @ForecasterEnten
Huh? You can have confidence intervals even without the central limit theorem applying. Certainly for even the smallest samples. You just need to use the correct formulation rather than assume normality.
1 reply 0 retweets 2 likes -
Replying to @arikmotskin @ForecasterEnten
Yes, but for polling data. That's not what we're doing here. (I meant it only for what the 95 percent confidence intervals are for these polls.)
2 replies 0 retweets 0 likes -
Replying to @zeynep @ForecasterEnten
I think it's fair to say that if there is a systematic bias, then traditional ways of computing CIs are meaningless or misleading. But that's separate from the central limit theorem
2 replies 0 retweets 1 like -
Haven't looked. I'd assume/hope competent ones are adjusting CIs after doing the LV reweighting. However since there still exists systematic bias even after that (whether big or small) there isn't much they can do other than maybe add some fudge factors to account for the unknown
1 reply 0 retweets 2 likes
Also, I think there must have been a lot of herding going on. I think some x part of the systematic bias isn't sampling, but pollsters looking at each other to adjust for the sampling issues, and then arriving at similar numbers. I don't know how on earth to think of a CI then.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.