Nate Cohn

@Nate_Cohn ↗

Deleted No

Hibernated Yes

Last Checked Feb. 2, 2021

Created

Tue Jul 14 14:06:13 +0000 2020

Likes

Retweets

Source

Twitter Web App

View Raw Data

JSON Data

View on Twitter

Likely Available

Nate Cohn @Nate_Cohn

It's totally possible to still produce compelling topline results, even if you have too many highly engaged activists in your polls. Your average error might be still be low, since maybe you have a decent partisan balance even though you have too many activists, etc. — PolitiTweet.org

Posted July 14, 2020 Hibernated

Preceded By

Nate Cohn @Nate_Cohn

If you show those things, I take it as a sign that you have some normal humans in your poll. Here on political twitter, there are no undecideds and there's no ticket splitting because you're mainly a bunch of ideological consistent partisans. But that's not most of the electorate — PolitiTweet.org

Posted July 14, 2020 Hibernated

Followed By

Nate Cohn @Nate_Cohn

I'm pointing this out because of the Civiqs poll in MT this AM. I think the Civiqs folks are doing about as well as anyone to meet the challenge of modeling out bias in online polling. I think it's good stuff. But nonexistent undecideds/ticket splitting is a warning to me — PolitiTweet.org

Posted July 14, 2020 Hibernated