There's good reason for that too -- party ID is very fluid. Age/sex/race is not.
-
-
--or than the voter base? Probably why we were so wrong in campaign 2016.
1 reply 0 retweets 0 likes -
Replying to @SharylAttkisson @RayGuy3
1/?? The polls weren’t wrong in 2016. Many forecasts were off because they misinterpreted the polls.
1 reply 0 retweets 0 likes -
2/ The nat’l polling average missed the final vote by 2 points in 2016. That’s a perfectly normal election year.http://fivethirtyeight.com/features/why-fivethirtyeight-gave-trump-a-better-chance-than-almost-anyone-else/ …
2 replies 0 retweets 0 likes -
3/ State polls were a bigger problem, but that’s more a function of missing data. Not enough polls lead to bad forecasts.
2 replies 0 retweets 0 likes -
4/ As for your hypothetical about 95% Ds answering a question, there’s no way for that to happen as long as you’re getting a representative
5 replies 0 retweets 0 likes -
5/ sample on all the fixed demographic groups.
2 replies 0 retweets 0 likes -
6/ It is a fair question to ask whether there were “shy Trumpers” in 2016, but the polls don’t back that up. https://fivethirtyeight.com/features/shy-voters-probably-arent-why-the-polls-missed-trump/ …
1 reply 0 retweets 0 likes -
7/ The problem with weighting by party ID is that it leads to “unskewing” polls, which never works out well.https://www.theguardian.com/commentisfree/2012/sep/27/unskewing-polls-party-id-bunk …
1 reply 0 retweets 0 likes -
8/ That’s because party ID is very, very fluid: http://elections.huffingtonpost.com/pollster/party-identification …pic.twitter.com/SZXCTgDHlp
1 reply 0 retweets 0 likes
Less so when you are looking at "registered" though--
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.