Natalie Jackson, Discover and Read the Best of Twitter Threads by @nataliemj10

Natalie Jackson

nataliemj10

From the Democratic firms' assessment of 2020 polling error: I find this piece particularly vindicating. I've been saying from the start that likely voter modeling is part of the issue.

As it turns out, I wrote my dissertation (10 years ago, eek!) on how people answer survey questions when they aren't relying on ideological or partisan cues. A brief thread

Here we go with my summary of #AVS2020 findings. How to navigate this beast of a report - there are a few major sections:1. State of the union, which includes

The @guardian article and Holocaust poll running around is reported extremely disingenuously. The headline doesn't hold up if you look at the actual results. People are not certain on #

Message to all the people who will say "it's tightening!": These differences are not statistically significant.https://twitter.com/MorningConsult/status/1299735650783178752 PS- that's without using a design effect, which w

NEW DATA ALERT:The July installment of @PRRIpoll's Trump favorability tracking is here. This time, we've added Biden favorability to the mix. A thread of findings:https://www.prri.org/research/trump-favorability-july-2020/ You want

I just did a couple of reviews for people converting their academic CV to more of a resumé format for applying to nonacademic jobs.A few overarching themes came out that

So the topline is that LGBT rights are pretty popular. Let's dig in a little more. Our sample size makes us one of the few (possibly only, besides Gallup) able

If P(down | rainy days ) = 1, andP(down | Mondays ) = 1, then P(down | rainy days ∩ Mondays) = 1 This is a niche tweet in multiple