G. Elliott Morris Profile picture
Data journalist + US politics writer @TheEconomist. My book STRENGTH IN NUMBERS is out via @wwnorton summer 2022. Blog/newsletter: https://t.co/B2HCPcS9LW
Stan Downs (Version 8.2.1) *Per La Resistenza* Profile picture eDo Profile picture Marcelo Martinelli Profile picture Joshua Cypess Profile picture Truth, Lies, and Uncertainty—dlonorse 🇺🇸🇳🇴🌊🎾 Profile picture 7 added to My Authors
8 Oct
I… don’t know. This objection rings hollow to me. In part because educational polarization is a long term trend (we have 70 years of it at this point) with stable growth — but really because our conversations on this are usually conditioned on patterns staying roughly the same.
You can make the argument that, well, correlates of voting will change over the next decade, so the rural white non-college bias of the Senate will get “solved” somehow — but it’s still a huge problem today, and it should be the mean expectation for the future too.
So yeah. I don’t think predictions for 5-10 years from now have too high uncertainty to not be useful. But smart people are conditioning on these things in convos, & I think they’re right. After all I think it’s *un*reasonable to think edu polarization will substantially reverse!
Read 4 tweets
6 Oct
This Pew poll seems to indicate support for Trump among Republicans is a lot lower than either the 2024 trial-heat polling or my general read of the media conventional wisdom suggests. Only 44% say he should run for president again. pewresearch.org/fact-tank/2021… Image
Not really sure how to reconcile this all, to be honest. My hard prior is that, if the GOP had a primary today, Trump would win a resounding victory. But maybe that prior is wrong. When we find conflicting evidence the answer is usually somewhere in between!
Let’s do an exercise:

Poll 1: If the 2024 GOP primary was held today, with all the rules from 2016, and it was Trump v all the top GOP leaders who _might_ end up running, would he win?
Read 8 tweets
25 Sep
For fuck's sake... My aggregate in CA was 60% Keep Newsom if you just allocated undecideds and corrected for whether polls had the right weighting scheme. That's a 2 point error on vote share! Tiny! This is bad for SOME pollsters, not the industry. Jesus gelliottmorris.substack.com/p/polls-of-cal…
Contrary to popular belief, you don't have to dunk on the polls every time an election result is a little bit surprising to you -- or if you looked at the wrong polling averages
Observing error in a 3rd-party polling average (& 1 that ingested data that was basically withdrawn by a polling house and didn't allocate undecideds!) and then projecting that error onto "polls" as an industry-wide error is a huge analytical misstep IMO.
Read 9 tweets
25 Sep
Nothing in this Kagan essay is new or shocking, but I do find the cohesive packaging useful. The Constitution has no checks against proto-fascist factions abusing multiple branches of government for the pursuit of power. Something has to change—and soon. washingtonpost.com/opinions/2021/…
We have seen multiple crises of the confluence of factionalism & US electoral +other institutions over the last year. Life-threatening covid-19 policy & 1/6 are only the most relevant examples. I have to wonder how bad people think it needs to get before we hit the tipping point.
I think the latter paragraph here from this excellent @jbouie article puts the pieces together very well. A faction of leaders holding power across levels and institutions of government can effectively circumvent the checks and balances of our government nytimes.com/2021/09/24/opi…
Read 4 tweets
15 Sep
Aaand there it is folks! Early vote-by-mail results in California's recall election are way ahead of tied-race benchmarks and signal an imminent victory for Governor Gavin Newsom, possibly by high double digits. I'm going to bed early tonight livevoterturnout.com/sandiegoca/Liv…
Listen to the (good) polls, folks!
where is my award for beating dave by one hundred and twenty thousand microseconds
Read 4 tweets
14 Sep
At this point, given the CA recall polling and VBM data, all we're looking for in the early vote tonight is confirmation of projected partisan distributions. Returned ballots are sufficiently Dem that we just need to assess loyalty + turnout. LA county VBM +25 Newsom would do it.
The very fuzzy math here is that early mail ballots statewide (+40) were about 10 points more pro-Biden than the final results (+30) in California in 2020, and LA County (+45) was about 15 pts more Biden than the whole state. So +55 early LA = +30 CA-wide in the end. +25 = tied.
You don't call an election just based on one county, of course. But Los Angeles cast 25% of CA's votes in 2020 so it's a good guide.

Other VBM benchmarks could be:

Recall +11 in Orange County (+19D 2020 VBM = 30 overall :: -11D VBM = 0)
Keep +3 in SD (+33D VBM = 30 :: 3D = 0)
Read 4 tweets
14 Sep
Blog post: Media coverage of California's recall election highlights big issues with popular poll aggregation models


FiveThirtyEight and RealClearPolitics are unable to capture modern outliers or the impacts of non-response & different weighting schemes
Expanding on what I've been saying here and elsewhere
Regardless of what happens in the election, the bottom line is that modern statistical methods can provide better analyses of polling data than the stuff available ten years ago, especially (but not only) because of recent problems in the polling industry. Time to do better.
Read 4 tweets
13 Sep
Here's my final update to this model of California recall polls. I'm calculating an aggregate that adjusts polls based on whether they use partisanship in their weighting schemes, and draws different trends for adjusted v unadjusted data. Newsom +18 +/- 10
The point of this project was to illustrate the different methods we can use to aggregate polling data — esp in how to improve existing popular averages that don't peer under the hood of how pollsters are processing their data, an increasingly important aspect of public polling.
So, note two things:

1. Popular averages magnify unlikely trends in public opinion by being whipsawed by data that is subject to higher standard errors than a decade ago (when the models were first made). Weighting by party flattens trends by decreasing nonresponse
Read 5 tweets
9 Sep
The thing about Rasmussen (and to a lesser extent, some other right-leaning pollsters and aggregators) is that the conservative information ecosystem has provided a top-dollar audience for confirmation bias, and there's not much AAPOR or good political journos can do about it.
538 could start by banning them, though
Read 4 tweets
8 Sep
Two of the enduring patterns of polling over the past 20 years are (a) that pre-election polls tend to underestimate the dominant party in a given state, especially lopsided ones (like California) & (b) that polls underestimate the status quo option on referenda and recalls. So..
Don't worry, I have receipts!

For ballot initiatives/referenda:

And see the graph from @MHeidemanns for potus polling errors Image
@MHeidemanns Also, given that the crosstabs for Newsom v Cox pretty much match perfectly with the 2018 results in the state, I'm inclined to believe the media was hoodwinked by partisan non-response and outliers (esp from lower-quality polls) over the last month.

Read 5 tweets
7 Sep
Maybe Newsom was always ahead (as everyone’s prior prediction should have been all along) and people were just overreacting to outliers and temporary partisan non-response that got magnified by over-aggressive polling averages?
I do think maybe people have been conditioned to rely on basic poll averages (like 538’s) too much at the expense of studying the underlying data and empirical dynamics of historical polling on recall elections (big “yes” bias), both of which call the average into question
Anyway, maybe I will be wrong next week, but the underlying conditions of this contest have been pretty consistent — despite what a couple overhyped self-reported likely voter polls have shown

Read 4 tweets
7 Sep
Not sure people realize this, but the state that's most disadvantaged by the Electoral College is Wyoming.
(DC voters have a smaller chance of flipping the outcome of the election, but, alas, DC is not a state)
Read 5 tweets
25 Aug
We're finally seeing some real movement in self-reported vaccination status. This from the latest Economist/YouGov polling: Image
The increase in vaccinations does not appear to be coming from anti-vaxxers deciding to get shots, but rather from a small-ish number of adults who have been unsure or planning on it for a while finally deciding in favor and getting around to doing so.
You can see that here. Lots of Democrats and some unsure Independents are getting their shots, but Republicans and anti-covid-vaccine Indies are mostly staying the course (for now). Image
Read 4 tweets
25 Aug
Germany's ruling centre-right party has been overtaken in the polls by the Social Democrats, reviving the chance of a left-controlled 'traffic light' or 'Germany coalition with the Greens or classically liberal Free Democrats. 1/2 economist.com/graphic-detail… ImageImage
2/2 There is still a lot of uncertainty, though — in fact, even more than there was last week (a product of changing correlations in the data). At this point in the campaign (a month out) history advises a pretty strong hedge against the polls using the results of the last elec.
CDU/CSU + SPD is viable, but there are some political obstacles and the probability of the parties winning a combined majority is under 50%. Red-Black-Green (sometimes 'Kenya') is not usually entertained federally, though it has worked in some länders.
Read 4 tweets
22 Aug
Hmmm. Might there be some partisan non-response going on in Biden's approval polling?

The aggregates:
Pollsters that adjust for past vote or partisanship:
Read 4 tweets
22 Aug
Not sure of the precise probability, but I think “Rs win Congress 2022, Ds don’t get unified control back again until 2036” is very high, maybe even > 0.5, and Dems are going to regret not blowing up the filibuster, passing redistricting reform & adding states for a looong time
It is obviously a fool’s errand to predict how demographic correlates of voting will change over 15 years, so I’m not trying to. The point is that the probability here is alarmingly high (esp given D pop vote advantage) — and higher than some Dems on Cap Hill seem to acknowledge
Rs win 2028, Ds win back 2036? Move the timetable up 4 years if you think Biden loses in 2024 (which is completely plausible) or if Rs get single-cycled. (The precise year is unimportant, it’s the duration thats key here.)
Read 5 tweets
13 Aug
Yesterday's new Census data matched with election results: America's fastest-growing counties have almost exclusively gotten more Democratic since 2012. Counties shifting towards Republicans have mostly shrunk.

Of counties that grew, 70% shifted Dem
90% of shrinkers moved right
This relationship is not entirely causal, but it does tell us a lot about how people perceive politics — both locally (eg in shrinking or growing places) and nationally (with Republicans serving as the party for "left behind", mostly white places w/ high degrees of status threat.
I imagine the relationship is even stronger with 2008 vote data, in case someone wants to send me a csv of county fips codes and D + R vote shares (I know I have one on my computer somewhere, but CMD + F is not saving me)
Read 4 tweets
4 Aug
#New: 30% of American adults haven’t been vaccinated for covid-19 yet. 18% total are relatively hard anti-vaxxers. Our poll shows around 1/10th could be persuaded to get jabbed by a full FDA authorization, but the vast majority may be impossible to sway.

Via me and @JamesFransham, with @MatterofMatt on chartage
That’s one-tenth of unvaccinated people, BTWS So full authorization might boost overall rates by a few percentage points. (The real number could be higher in the future if the information ecosystem shifts as a result, but I wouldn’t necessarily bet on that.)
Read 4 tweets
1 Aug
I will make a better plot soon, but vax rates in the rural midwest and northeast look to be the highest above trend in all the US. Would be interested in reading about why
My initial hypothesis is that counties with a high concentration of whites but fewer white evangelicals have these higher-than-expected vax rates. But there also seems to be a sizable state-level intercept
Read 4 tweets
30 Jul
I think bias in US electoral institutions is probably a bigger threat to representative government in American than almost anything else. The growing rural bias of the Senate+EC (graphs via @davidshor) nearly ensures Rs will deny Ds fair majoritarian control for at least a decade
Plus, bc of the current bias in the House & 2022 redistricting harms, Ds r very very likely to lose control of Congress. Extrapolating historical trends, current polling indicates they'll be @ 48% in polls on eday. Polarization gets Ds ~1% back. But they need 51.5-52%. (MOE is 6)
When ppl ask me if we are approaching civil war, I tend to say no — Instead, we are rapidly nearing a time when the majority will need 52-53% to win, & a ruling authoritarian minority will enact policies that harm the other 200m+ citizens w/o meaningful electoral accountability
Read 7 tweets
27 Jul
.@BenSasse & others have argued that a decline in religiosity in America has left ppl vulnerable to the false spiritual satisfaction of conspiracy theories & sectarianism, eg in QAnon.

But we find the opposite. The least religious are the least credulous:
And it’s not just QAnon; Our Economist/YouGov data show white evangelical Christians are also disproportion likely to believe other conspiracies — eg about the 2020 election, but also about vaccines and the moon landing. True even after controlling for demographics and politics.
Finally, the relationship between church attendance & conspiratorial thinking is positive even if you omit evangelical Christians, though less so. They are driving the trend—but the most religious Americans regardless are more likely to adopt phony theories than the least devout.
Read 4 tweets