Here's my initial answer to the question of why Trump is losing so much ground with non-college whites:

The upshot is that "the president still harps on racial issues, *but voters are less racist and swayed less by sexism than in 2016*"

Short thread:

economist.com/graphic-detail…
I first took 318k interviews from UCLA + Democracy Fund's "Nationscape" poll to measure average levels of racism and sexism over time. Those scales avg Qs like whether voters think slavery still harms black Americans & whether there's racism against whites into a single measure.
I then compared how racism & sexism correlated to support for Trump in 2016 with how they correlate in 2020. It turns out that racism is MORE predictive of support for Trump now — IE that voters are more polarized by racial attitudes — even though RACE itself is less predictive.
This growing disconnect between how *racial attitudes* predict support for Trump or Bide and how self-described *race* do presented to me what at first to be a paradox. If non-col whites are most racist, what explains the shift?

The poll has answers:
According to Nationscape, average levels of racial resentment in the electorate have fallen over the last year. (Pew finds the same). That's both bc 2016 voters have gotten less racist since Trump's election & bc new voters are less racist. True overall & for non-college whites.
This means that there are fewer voters with high degrees of racism for Trump to court before November.
Sexism also plays a big role. In 2016, male discomfort with a female nominee could have cost Clinton the WH. But since Biden is a man, attitudes about gender are less salient v 2016; voters who score highly on the sexism scale are less likely to vote for Trump this time around.
The final piece: since non-college whites are by far the most racist and sexist voters in the electorate, at least by this measure, this means that the shifting roles of each in predicting support for Trump most impacts their votes, rather than for BIPOC voters (esp women).
Here's the chart, and don't forget to read + share the piece!



economist.com/graphic-detail…

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with G. Elliott Morris

G. Elliott Morris Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @gelliottmorris

24 Sep
It would be nice if the most famous person in political data journalism would actually ask people for the data they're referring to before amplifying totally misleading takes Image
as far as i can tell, the "polling" ryan is referring to is far from conclusive (i can't even tell if it exists), and nate's take is pretty transparently too clever by half. pretty standard punditry here
in fact, i think we have enough yougov data to actually model this. turnout ~ faith in the process + demo and party controls. will report back.
Read 9 tweets
23 Sep
In all seriousness, this tweet from @SeanTrende exemplifies a very important difference in how various poll aggregators view actual polling microdata — one that has the potential to reshape how news consumers view political polls writ large (for the better, IMO). Short thread:
Some polling aggregators (eg RCP) take the raw polling toplines data and assume that enough pollsters will follow best practices in weighting, sampling, etc to give you the best average prediction possible. This serves us well sometimes but is pretty naive when you drill down.
Others (eg 538) think that there are consistent differences between pollster modes, populations and polling firms that allow you to makes good predictions of which methods are best. They weight polls by accuracy and do other math to debias data to squeeze out all the extra juice.
Read 7 tweets
23 Sep
10% chance of a trifecta? color me skeptical, very skeptical Image
The one huge thing I learned from modeling this year is that it is a mistake to model uncertainty as uniform over time. Polarization makes elections much easier to predict (on balance) today versus, say, 1970. I think that's making Nate's models too uncertain across the board.
If people in the media are to take one thing away from the theory behind our models, it's that how we think about certainty in a time when voter behavior is constrained by very predictable forces can lead to better predictions, especially earlier in the year, in modern elections.
Read 5 tweets
22 Sep
I think there is still a fairly sizable chance of a systematic error in the polls, and how big it is depends on which forecasts/averages you look at. (This is an important thread, please read it carefully.) 1/8
2/8 While more battleground-state pollsters are weighting by education than at this point in 2016, about half still aren't. And there are other issues with pollsters who have clearly politically biased samples because of other errors (like weighting to the 2016 exit polls).
3/8 We adjust for some of this bias by including a term in our model that balances systematic difference between correctly and poorly-weighted polls. In 2016, it shaved about 2pts off Clinton's margin in the Midwest, so it IS helpful. BUT the model was still surprised on Nov 8.
Read 11 tweets
22 Sep
Trump can afford to lose 36 electoral votes from his 2016 total and still hold on to the White House. But Biden currently leads in polls of all 6 of the closest Trump states from 2016:

MI (16 EVs)
PA (20)
WI (10)
FL (29)
AZ (11)
NC (15)

A short thread:
projects.economist.com/us-2020-foreca…
On the one hand, Biden is clearly in a good position. If the polls look like they do today on election day, our model will give him close to a 92% chance of winning the election. We're still a long way away, but that's roughly where things are headed.
But on the other hand (and especially because of Trump's relative electoral college advantage), it actually wouldn't take much to nudge the race closer to 60/40 or even 50/50. A streak of good polls in FL or PA would get him much of the way to a tied race.
Read 7 tweets
21 Sep
Some more numbers on the Senate's rural bias:

If all Senate seats were up at the same time and we assume D pres states go D down-ballot, Dems would have to win a national landslide of ~19 points to control a supermajority. Reps would just need to win by just 2(!) for 67 seats.
If you order the states by their partisan leans, the 67th seat for Democrats falls somewhere between Mississippi (R+19) and Missouri (also R+19), whereas the 67th seat for Republicans is between Nevada (D+1) and VA (D+2).
The thing about the stupid "We're a republic, not a democracy!" comments is that republics are supposed to have indirect **representation** for voters — and under no reasonable national parameterization can you call what the Senate is doing today "representation."
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!