Polling uncertainty
However, pollsters' projections depend upon what assumptions they are making about voter turnout. If turnout is substantially different from what the pollster expects, the actual election outcome will be correspondingly different from what the polls projected. This year, turnout models are especially iffy because of the influence of a new factor (the end of Roe and subsequent wave of abortion bans in red states), for which nobody really knows how much impact it will have. The only hard data point we have is the Kansas referendum, for which polls showed a very close result, whereas the actual result was not close (60% favoring abortion rights), while turnout was very high for an election of that type. Obviously the pollsters misread how the sudden salience of the issue would affect turnout. On the other hand, it was a referendum, and we don't know whether the threat to abortion rights will affect candidate elections in the same way. A referendum is on a single issue, while assessment of candidates turns on all the issues the voters expect an elected official to act on when in office.
Pollsters are also smarting from 2016, where they failed to anticipate how Trump's unique appeal to citizens of a type who often don't reply to polls would drive them to vote. They may be over-compensating by over-estimating Republican turnout. Trump is, after all, not on the ballot this year, so it's unclear whether his loyalists will turn out as much as in 2016 or 2020.
So, basically, this year the accuracy of polling is more uncertain than usual. The election could be close or it could be a landslide for either side. We can't be sure. Be prepared for surprises -- in either direction. Democrats would be wise to fight as if they were behind, and to take nothing for granted.
posted by Infidel753 at 3:33 AM
7 Comments:
polls smolls...we gotta get out and vote.
Astute observations, Infidel. Linked.
I was listening to an experienced pollster on a podcast last week, and they really don’t know how to reach a representative cross-section of the voting public with any particular polling method today. I.e. old fashioned telephone polling typically skews towards low income, older people. Other methods have their biases, and most people really don’t want to answer the door. The polling business is overdue for a fresh model of some sort, new tech/methodology.
And obviously if millions of people are willing to make a guy like Herschel Walker a US senator, a lot of common sense assumptions about the electorate in general have gone out the window.
Well,
We got a kinda surprise PM... Liz Truss was clearly so out of her depth - I mean talk about pissing on the Hiroshima firestorm, Michael Gove is just a spherical git*, Penny Mordaunt was more Penny Dormant so it had to be Rishi - and all this with no vote - basically last one standing. Bringing back BJ would have turned a farce into a tragedy. Oddly enough, if the polls (and stuff) are to be believed... We'd not have been Trussed-up and had Rishi instead of Truss from the start if the Tory leadership was determined by the MPs but this went to the grass-roots membership and they didn't want the Indian but opted for the woman who has spent an entire political career cos-playing Maggie Thatcher (yes, even down to riding around in the turret of a Challenger 2 tank wearing a 1950s style headscarf**) rather than the Asian. I'm fairly neutral (I tend to vote more on local issues). So, sometimes things take very strange turns and polling can't reflect that... because if this had got as far as the votes of the entire membership of the Tory party Rishi Sunak's race and religion (he's a Hindu) would have worked against him. The Tory party have a deep problem. Their natural support is older yet their natural instinct is to cut spending which affects their base of codgers and biddies more than anyone else. Pensions, new hips on the NHS and such...
...but polling is odd because there are almost certainly more folk who hold racist views than would admit to it. Oddly enough I think there was a strange element of something along those lines with Obama in the US. The press here in the UK were so obsessed with a black Prez they didn't even mention the interesting fact you'd got a Hawaiian! Which is very odd considering the whole "birther" nonsense. I mean the guy was born in that state after it became a state and surely that is a matter of public record?
*That is with apologies to Fritz Zwicky.
**A look the late Labour MP Denis Healey memorably described as looking like a "Bargain Basement Boadicea".
Granny: Always true.
Reaganite: I think they're going to need some kind of model where they pay people to participate in polls. There are already online services that do that. It would skew the samples in various ways, but they're already skewed because of all the demographics that avoid participating. The pollsters would just have to weight their demographic categories to account for it.
About Walker, I can kind of understand it. If your choices in an election were (a) a really dumb candidate who had once committed what you consider a terrible crime but now opposes it, and (b) someone who would legalize and normalize that "crime", you'd probably vote for (a). The real problem is still the theocratic control-freak mentality.
Thanks for the link!
NickM: We'll see how long Sunak lasts. Hopefully he'll learn from Truss's mistakes.
I know in the UK it's not unknown for a disgraced leader to make a comeback, but bringing Johnson back so quickly would have been farcical.
It is certainly a matter of public record that Obama was born in the US (Hawaii), but remember the hard right wing in the US is mostly religious fundamentalists, which means they have an entrenched ability to simply dismiss any evidence that goes against what they want to believe. Those who think Obama isn't really American just dismiss his birth certificate as a forgery, etc, the same way they dismiss all the evidence for evolution and suchlike.
That’s a good idea, paying people. It would provide an opportunity to perhaps get better/deeper info on them, or have a longer term relationship where you get a movie not a snapshot of the reviews as events unfold. Or find out what their friends and family think too
Interesting. The polls and constant media reporting of them are really getting old. I Was hoping your post was going to tell me that Democrats were going to win in a landslide. :-)
But, I appreciate your insights and certainly agree that Democrats should take nothing for granted.
Reaganite: That would be a good idea, to keep surveying the same people -- it would give a sense of how opinion is shifting over time. But with any system, there would still be the problem of predicting turnout. Usually the assumption is that the people who voted often in the past are the ones likely to vote in the future -- but that can miss potential game-changing factors like Dobbs.
CAS: Frankly I hope somebody wins in a landslide. If one party loses really badly, that one will then have an incentive to purge out the crazy stuff. That way we'd eventually have at least one sane party.
Post a Comment
<< Home