Matthew Yglesias

@mattyglesias ↗

Deleted No

Hibernated Yes

Last Checked Sept. 24, 2020

Created

Tue Sep 01 13:33:02 +0000 2020

Likes

118

Retweets

Source

TweetDeck

View Raw Data

JSON Data

View on Twitter

Likely Available

Matthew Yglesias @mattyglesias

National polls which *did* weight on educational attainment were highly accurate in 2016, so pollsters know how to account for “shyness.” You can actually check whether or not a pollster is using a sound methodology rather than speculating randomly. — PolitiTweet.org

Posted Sept. 1, 2020 Hibernated

Preceded By

Matthew Yglesias @mattyglesias

Higher synthesis: The true form of the “shy Trump voter” thesis is that non-college people are less likely to respond to polls *at all* so pollsters who don’t weight for educational attainment end up undersampling Trump-loving working class whites. But it’s fixable! — PolitiTweet.org

Posted Sept. 1, 2020 Hibernated

Followed By

Matthew Yglesias @mattyglesias

But! Accurate polling is harder to do at the state level because the range of demographic factors you’d ideally like to weight for is more nuanced (Mormons in Arizona, churchgoing vs secular non-college whites in Maine, cuban vs Puerto Rican in Florida) and it’s just hard. — PolitiTweet.org

Posted Sept. 1, 2020 Hibernated