Nate Cohn

@Nate_Cohn ↗

Deleted No

Hibernated Yes

Last Checked Oct. 1, 2020

Created

Thu Jan 16 13:39:30 +0000 2020

Likes

Retweets

Source

Twitter Web App

View Raw Data

JSON Data

View on Twitter

Likely Available

Nate Cohn @Nate_Cohn

Many pollsters have a tougher second screen, particularly late in race. They often give 0 weight to anyone who doesn't say 'very likely' or 'almost certain' to caucus. In our poll, this would have cost Sanders (our Oct. result in par.) W 21 (22) Bu 19 (18) B 16 (17) S 16 (19) — PolitiTweet.org

Posted Jan. 16, 2020 Hibernated

Preceded By

Nate Cohn @Nate_Cohn

In Oct poll, we had two LV ?s. First, we asked whether they'd probably caucus for Ds, R, or wouldn't caucus. If D caucus, asked how likely to caucus, and down-weighted less likely voters. The vote after 1st screen, then 2nd was W 22 --> 22 S 20 --> 19 Bi 18 --> 17 Bu 16 --> 18 — PolitiTweet.org

Posted Jan. 16, 2020 Hibernated

Followed By

Nate Cohn @Nate_Cohn

This latter result, with the tight second screen, is closest methodologically to, say, Selzer's screen. And indeed, she had Sanders/Biden at 16/15 two weeks later. The result after the 1st screen (no second screen), is more like Emerson, which had Biden/Sanders in 1st kn Dec.. — PolitiTweet.org

Posted Jan. 16, 2020 Hibernated