The US presidential election remains on a knife-edge 45 days before voters go to the polls, despite Kamala Harris enjoying one of her most encouraging spells of opinion polling since becoming the Democrats’ nominee nearly two months ago.
During yet another momentous week that began with a suspected second assassination attempt against Donald Trump, the latest Guardian 10-day polling averages survey shows Harris increasing her lead to 2.6 points, 48.5% to 45.9%.
While still within error margins, that is an improvement of the 0.9% edge Harris held last week and a significant shift from the statistical dead heat of a fortnight ago before the candidates held their only scheduled televised debate in Philadelphia on 10 September.
Jimmy Kimmel shows polls on (I think) his Thursday monologue that showed Harris ahead, Trump ahead, and both of them tied.
Polls clearly don’t mean shit anymore.
I mean, they are significant, and they’re one of our better ways to get information about what’s going on. But there are complexities, and you can’t ignore them.
First, obviously, there’s the issue that any polling thing has, which is the possibility for error. If you call people on phones, some people won’t respond, and those people may not be an even cross-section of society.
Second, the Presidential election isn’t a popular vote. What matters is what’s happening in the swing states. So presenting the output of a national popular poll – which for some reason, a lot of media organizations do, I can only imagine because it’s simpler to understand – isn’t a very good metric for who will win the Presidential election. There’s correlation, sure, but the difference can be quite significant in very close elections like this.
Third, voter turnout matters, not just who a voter prefers.
To compensate for various issues like this, you develop a model that takes input – for which polls are one, but only one, source of inputs – and then train it on past elections. You can model things like increased or reduced response rates to polls among various demographics.
That can be useful, but it’s still not an oracle that can see into the future. Models always deal with a simplified view of the world, aren’t omniscient. We won’t know for absolutely certain how well a model performs in a particular election until that election. Maybe it turns out that there’s a ton of hail on that day, and that there’s a Twitter trend to not go out in hail, or something equally off-the-wall.
If you want to know for sure, the answer is wait for the election.
If you want to get our best guess, the answer is “we can make some predictions, using polling data, statistical analysis, and other historical information, though it’ll inevitably have some level of uncertainty”. Right now, I think that it’s a pretty decent consensus that the race is pretty close, but that Harris has been generally improving her position. No legitimate person modeling the election is going to simply outright call the election at this point, though. They’ll just give estimates as to likelihoods of outcomes.
Yeah, they’re entirely meaningless this election cycle. These poll leads have been sitting in margin of error zones the whole damn time, which means you cannot make a useful inference from the data in the poll.
It’s so close you can slightly tweak the conditions and who you ask and WHAT you ask them to push the poll data to support anything you want.
Good for propaganda, bad for anything else.
Also, who answers an unknown number from their phone anymore other than the elderly and the lonely? That’s not exactly a good cross-section of voters.
There was an interesting thing from the NYT where they talked about their phone poll response rate being 2%.
I can’t imagine that they had worse or better response rates annoying people with their shit than anyone else would, so it’s probably a good wager that all the polls see 2% response rates and thus are based off a shockingly small profile of people.
A small dataset can still be statistically significant, depending on how and where the data is obtained and structured.
That being said, everything involving Trump has had bad polling since he first started running. Poor polling isn’t particularly new when he’s involved. Some of it is spot on, others wildly inaccurate.
I’m surprised marketing companies don’t release their stats since they have a better idea of who everyone is going to vote for than anyone else. But then they’d have to admit that they know.
They’d also have to not have a horse race, which is what is in their best interest.
That doesn’t mean it’s meaningless. It means it’s a very close race.
Polls are simply educated guesses. Nobody knows for sure until the election is certified and POTUS is sworn in.
But how do you even make an educated guess anymore? So few people answer their phone now that there’s no way you’ll get any idea.
They use other survey methods. Go read any methodology and it will describe how they collected their data.
But that would require learning stuff instead of spouting off whatever negative shit feels good
They just keep going down the list until they have enough respondents to use for the data.
I don’t think there is a list of phone numbers out there that can give you anything close to an educated guess for people under 30. They don’t even like to talk on the phone. They’d rather be texted.
Polling companies have a standard number of people they need that will cover their demographic requirements (ie: age, income, location). They keep going down their list of phone numbers until those requirements are fulfilled.
But again, how can they accurately get that in terms of opinion when so few people under 30 are willing to answer their phones? It doesn’t matter what your demographics are when you’re still restricted to the small number of people under 30 that are willing to answer their phones, making them outliers to begin with.
Their lists are millions of names long. It takes them a lot longer than it did 30 years ago, but they slog away at it until they get enough.
It’s faster. They use more methods than just calling, but the calls can be automated and the data entry digitized. (I know there are restrictions on robodialing, but I’m sure surveys have an exemption or approval process.)
I would suggest that the fact that the polls are all over the map supports my point.