Polling Errors

2016 should have told you what you needed about polls. Don't trust them or the media on anything. As a conservative guy Fox News is a joke to me these days, and they've been a disaster tonight. CNN had better coverage.

And quit with the false dichotomies. Just because someone doesn't agree with the blm BS doesn't = they don't like black people.

 

Simply people lie to pollsters

in 2016 is was a fun thing to do . . . .
now . . just a matter of course

Rocket River

 

What does. . .. "doesn't agree with BLM" - What do you not agree with it . . .. . . .

Rocket River

 

People will have to reboot and figure out the "outlier" that happened in Florida. Elsewhere, this paper posted today might serve hints.

Also think "shy Trumpers" would be a lot more a factor this time around than in 2016. It's now a defined behavior instead of a reflexive habit to instinctively avoid social discomfort.

www.thecrimson.com /article/2020/11/2/2016-election-polls-kuriwaki-isakov/
Harvard Researchers Warn 2016 Polling Mistakes Serve as a 'Cautionary Tale' in 2020 | News | The Harvard Crimson
4-5 minutes

Harvard Government Ph.D. candidate Shiro Kuriwaki and Michael G. Isakov ’22 cautioned against “overconfidence” in polling data in a paper published Tuesday on their analysis of pollsters’ incorrect predictions 2016 Democratic presidential nominee Hillary R. Clinton would win the previous election.

Kuriwaki and Isakov’s paper, published in the Harvard Data Science Review last week, identifies mistakes in 2016 polling and posits ways to correct these mistakes.

The paper explains that sampling biases were partly to blame for polling mistakes in the 2016 presidential election.

Shortly after the election, some analysts attributed errors in polling to the so-called “shy Trump” effect, in which voters would purportedly lie to pollsters as they were embarrassed to express support for Trump. Kuriwaki and Isakov, however, wrote that analysis of the 2016 polls reveals that mistakes were more likely due to a “sampling problem,” rather than the “shy Trump” effect.

“There was very little evidence of that, partly because there were similar errors for other Republican candidates,” Kuriwaki said in an interview. “So if people were just lying about Trump, you won't see that pattern.”

Kuriwaki and Isakov wrote that polls are most reliable when the sampling pool is representative of the general population, but noted that polls often fall victim to selection bias and disproportionately sample certain demographics.

Pollsters improperly corrected these biases in 2016, which caused the polls’ distorted results, according to the authors.

Poll aggregators also contributed to polling error in 2016, according to Kuriwaki and Isakov.

“When aggregators say there's a 93 percent chance of Hillary Clinton winning, some people think like, ‘Oh, she'll get 93 percent of the votes,’ which is not true,” Kuriwaki said. “But if you say the model tells you that Hillary Clinton will get 53 percent of the popular vote, which is one of the predictions, then people might have the right level of uncertainty.”

“2016 headlines made people more overconfident about the actual results,” he added.

They wrote their research describes a “cautionary tale” they urged pollsters to be responsible in the way they present results.

“It is the polls’ responsibility to not overhype the results and interpret them properly,” Kuriwaki said. [UNLIKELY]

Kuriwaki and Isakov also developed models for 12 states in the 2020 election with the assumption that pollsters make the same mistakes as four years ago.

Some pollsters, however, have already updated their methods, such as by weighting the responses from certain demographics which tend to be underrepresented.

In median battleground states in the 2020 election, Kuriwaki and Isakov found that Trump could receive 0.8 percentage points more than polls conducted in late October predicted, though their model has double the margin of error compared to previous polls. [USELESS]

Another factor Kuriwaki and Isakov considered was voter turnout in median battleground states. Higher voter turnout in this year’s election could result in even less confidence in polling predictions because the polls’ sample pool would become less representative of the general electorate, according to the Kuriwaki’s and Isakov’s paper.

The researchers said that, while polling may not be perfect, it “keeps everyone honest.”

“I feel like having more information is always, always good,” Isakov said. “There's a lot of other countries where we do not really have polling and that has been problematic in various ways.”

“Polling makes the pulse of the country public,” Kuriwaki said. “It shares information that puts people on similar grounds.”
​

 

2016 was within the margin of error come Election Day. It was a 3% national average, Hillary ended up with 2% national lead

This year, Biden had a 7% national lead, which will end up only being around 2-3% at best, 4-6% off is not within the expected margin of error. Something is flawed in the method for modern elections.

 

The problem is that polls don't address turnout.

I think Biden's poor ground game is a big reason for the discrepancies.

 

Margin of error +-4%.....LOL

 

Likely voter polls should be factoring in turnout. Biden really did **** up by mostly taking a knee after halftime.

I was prepared for Trump to outperform polls by about 3%. The reality is I believe a large portion of trump voters hate the media so much they lie to pollsters or decline to respond making it difficult to know.

 

I think it's hard for the polls to pickup the fact that Trump's ground game was registering a ton more Republicans than Dems which is reverse of normal elections.

But yes, something is definitely off with the polls

 

Pollsters are more interested in their desired outcome than actual data.

Polls should be getting more accurate with modernization, not less.

 

Yes. Even with the COVID-19 the rules didn't completely change. The candidates still could've campaigned more in person and they could've sent out volunteers in person and registered and gotten people to the polls.

 

You believe pollsters are lying or tampering with their data?

The turnout theory seems the most plausible. Seems unlikely to me that Democrat pollsters would lie about their data and sabotage their own outcome. I do think Republicans seem to be shy about admitting they're Trump voters but I don't know why they would do that in an anonymous poll.

 

Here’s the issue with polls.

who are you asking?

do you want to tell people you are voting for Trump when the left is constantly questioning his voters character?

 

Lying or manipulating? Absolutely not.

A question can be asked in a manner to get desired results.

When was the last time you have been polled? People in your circle?

The idea we can go off polls is antiquated. We need polling models. Data needs to be extrapolated. A pollster can't ask 5000 people a small set of questions and use statistics to fill in the other 99.9%.

 

Isn’t that what pollsters do? Use polling models to make adjustments to the data in order to more accurately predict the outcome? They look at systematic deviations in polling vs results in past elections and use that to improve their model, is my understanding.

I suspect many Trump supporters have been conditioned to be quiet about their support in public, and that mentality had an impact on the accuracy of the polling results. Even if it’s anonymous, if you are worried that there is a deep state apparatus spying on you or a human on the other end judging you, that will have an impact.

 

This is going to be controversial.. As much as things like the amount of lawn signs, car and boat parades were derided they actually did show that there was a lot of support for Trump and Republicans that weren't being captured in polls. In the future such things like this should be factored in assessing where an election is going.

 

I think when they do the post-mortem analysis, the thing that will come out is that Trump's ground game (and Biden's lack of one) skewed the polls as I am not sure all those new registrations come up.

For example, most likely voter calculations measure voting history, interest in the campaign, and whether a voter knows where their polling place is. A new registration therefore is viewed as less likely to show up to the polls. And if it turns out that Trump way out registered what was expected and Biden underperformed, yet Trumps new registrations voted in high numbers, that could explain why the polls were off

 

Florida is the prime example. There are more registered Ds than Rs in Florida. The Rs kick the Ds ass in GOTV. It is that simple.

 

They’re not errors. Trump voters lie, same as 2016. That’s why polls never mattered.