Nate Cohn

@Nate_Cohn ↗

Deleted No

Hibernated Yes

Last Checked Aug. 28, 2020

Created

Wed Sep 11 17:21:31 +0000 2019

Likes

Retweets

Source

Twitter Web App

View Raw Data

JSON Data

View on Twitter

Likely Available

Nate Cohn @Nate_Cohn

Yes, this is the first time there's been such a large, illusory shift. But the risk would increase as the number of live polls declines and response rates decline, and we know both those things are true. Summer response rates might have been especially bad. IDK — PolitiTweet.org

Posted Sept. 11, 2019 Hibernated

Preceded By

Nate Cohn @Nate_Cohn

I can imagine two major theories. The pro-online case would be that the live polls were, wrongly, swung by differential nonresponse during the summer. Perhaps it was exacerbated by the small number of live polls, which meant maybe a few relative outliers moved things too — PolitiTweet.org

Posted Sept. 11, 2019 Hibernated

Followed By

Nate Cohn @Nate_Cohn

The pro-phone case, I think, would look at the declining variance of online polls and wonder whether there might industry-wide panel conditioning, where so most online-poll takers have taken these polls some number of times and it's affecting their responses, even IRL — PolitiTweet.org

Posted Sept. 11, 2019 Hibernated