Why the "Political Polls are rigged" mantra is wrong and dangerous

[brooklynite]

It is an article of faith among many here that "the media" (not just Fox News and the like) are biased against Democrats. It is also an article of faith that said media is releasing Biased or fake polls to shift voter opinions away from Democratic candidates.

The problem with these assumptions is that "the media" doesn't do polling. They hire commercial market research firms to do it for him. While there are certainly polling firms that go into the business with a political bias, most do this as part of a broader business and have other, non-political clients (Coca Cola, General Motors, Delta Airlines etc.) To get that business (which is a much more substantial part of their business than high-profile political polls), they have to be seen as accurate and objective in their data collection.

There's also another category of polls: Universities (Quinnipiac, Monmouth, Siena, Emerson, Marist). These are from academic market research programs that using poll preparation and data analysis as teaching tools. As with corporate pollsters, there's little value in carrying out polls that are intentionally biased.

The reason discounting polls out of hand is dangerous is that developing good political strategy depends on objectively analyzing available data. Almost every poll says the national political choice (yes, not applicable State by State, but there's no data suggest the races in WI, MI, PA etc are significantly different) is a matter of single digits. Can that change over a year? Of course. But to discount that data now is to say that you believe the race ISN'T close and that either Biden or Trump is comfortably ahead. Either assumption will skew your political behavior to the detriment of the camapaign.

[Native]

1. Thank you for reminding everyone about this! It seems almost every poll posted here is captioned as fake or biased or

discounted because "landlines."

[everyonematters]

2. According to 538, the polls for 2022 were historically accurate. They have figured it out.

[brooklynite]

3. What everyone forgets about 2022...

...is that polling DIDN'T predict a "red wave". It predicted multiple close races, which is what we got. Analysts (not unreasonably) assumed those races would break for the Republicans).

[JohnSJ]

4. I have never said the legitimate polls are "rigged". What I have said is I believe that a lot of thing

have changed that the polls do not account for.

1. Many people do not answer calls from numbers they do not recognize because of all the spam calls.
2. National polls and favorability polls are not how presidents are elected and may give misleading impressions.
3. Some Democrats may "prefer" a different candidate in a polled question over Biden, but that does not translate that they won't vote for him in the general election. Republicans for the most part will stand in unison in their view of trump, and will vote for him in the general election.

I also think the mindset that the past will predict the probable action may not be correct. For example, it is assumed that young people won't turnout in large numbers to vote, because that has been the case in past election. I don't think those assumptions can be relied upon anymore because of the acceleration of changes that have occurred that will affect people. Such as abortion, climate change, voting rights, civil rights, etc.

State polls on the other hand, I believe more accurately predict the outcome in a state.

[brooklynite]

8. This is also flawed thinking

What you're saying is: "I can't think of a a way to accurately collect polling data, so it can't be done".

This is part of their job. If their clients didn't find the data accurate, they wouldn't hire the firm (nb: that includes candidates up to and including the Biden campaign).

Polliing firms have moved to cell phones, texts and online collections. There are variable levels of accuracy, but that's why people pay them for the service.

[JohnSJ]

13. National surveys of the 2020 presidential contest were the least accurate in 40 years,

state polls were the worst in at least two decades, according to the new, comprehensive report from the American Association for Public Opinion Research.

.....


Those “prime suspects” will hardly be comforting to pollsters and those who depend on them, from political campaigns to the news media. The most likely — if far from certain — culprit for off-kilter polling results is that key groups of people don’t answer polls in the first place.

Decreasing response rates have been a major source of concern for pollsters for more than a decade. But the politicization of polling during the Trump era — including the feedback loop from the former president, who has falsely decried poll results he doesn’t like as “fake” or deliberately aimed at suppressing enthusiasm for answering polls among GOP voters — appears to be skewing the results, with some segment of Republicans refusing to participate in surveys.

https://www.politico.com/news/2021/07/18/pollsters-2020-polls-all-wrong-500050

As for your assertion if the data wasn't accurate, they would not hire them, I disagree. They use them because it has been the traditional method of trying to access public sentiment, and the dynamics are changing faster than they are able to account for those changes.

Whether it is cell phones, landlines, online collections, a good number of people are not participating, and giving out that information.

https://thehill.com/opinion/campaign/3710954-2022-polling-the-bad-the-ugly-and-the-really-ugly/

[PortTack]

23. Indeed they were

[Tomconroy]

32. They were inaccurate because they underestimated

trumps support.

[inthewind21]

11. What does

People not answering their cell phones have to do with anything? I see this repeated over and over. You think NO ONE in the entire country ever answers?

Pretty sure we are ALL aware a poll isn't how a president is elected. It's called a poll or survey for a reason. They ask a question and I answer.

It seems to me that you have never actually been polled and are assuming quite a bit.

[JohnSJ]

14. No I am not. You ask why it matters, well, it matter to the pollsters, they are concerned about it.

National surveys of the 2020 presidential contest were the least accurate in 40 years,
state polls were the worst in at least two decades, according to the new, comprehensive report from the American Association for Public Opinion Research.

.....


Those “prime suspects” will hardly be comforting to pollsters and those who depend on them, from political campaigns to the news media. The most likely — if far from certain — culprit for off-kilter polling results is that key groups of people don’t answer polls in the first place.

Decreasing response rates have been a major source of concern for pollsters for more than a decade. But the politicization of polling during the Trump era — including the feedback loop from the former president, who has falsely decried poll results he doesn’t like as “fake” or deliberately aimed at suppressing enthusiasm for answering polls among GOP voters — appears to be skewing the results, with some segment of Republicans refusing to participate in surveys.

https://www.politico.com/news/2021/07/18/pollsters-2020-polls-all-wrong-500050

https://thehill.com/opinion/campaign/3710954-2022-polling-the-bad-the-ugly-and-the-really-ugly/

I don't answer calls from numbers I don't recognize, and I don't do online polls when surveyed.

and while I may be the exception, I wonder how many exceptions are out there that do the same thing?

[inthewind21]

15. OK

you are 1 in millions. So you don't do polls. Why all the concern over something you state you aren't even involved in?

[JohnSJ]

18. Because while polls try to gauge public opinion, they also can

influence public opinion

[PortTack]

24. And this is what makes even a slightly inaccurate poll dangerous

[JohnSJ]

27. I agree

[leftstreet]

5. It's hard to figure out WHO they're polling

Polls were deemed more accurate when it was assumed polling was done by landline, and people were more willing to pick up an unidentified caller.

That's not the real world now. If pollsters could explain how they're gathering their data it could lend to their legitimacy.

[GuppyGal]

6. I think a lot of these polls are conducted online nowadays ... but it would be helpful for the poll to do what you

said.

[inthewind21]

7. You're

assuming they are all done by phone. Some are, some aren't. I have got polls via text, email AND land line. Yes, they ARE contacting cell phones.

[Sympthsical]

17. I get so many text polls. It's obnoxious.

Funny story. The other day I got a text. It was a man telling me about what he'd been up to lately and included a pic.

I had never met this person before in my life. Not a bad looking man. I'd wondered if, perhaps, he was someone I'd talked with on a dating app once and had simply fallen out of contact with. It happens sometimes. Go on a date or two, lose contact, then they reach out up to five years later, and I'm sitting there like, "I do not remember this person at all."

Anyway. So, I'm like, "Why is this handsome fellow texting me out of the blue?" I'm all settled down now.

Then I scrolled up and saw previous missed messages.

No, I will not be sending you a campaign donation.

And stop sending me pics! Weirdos.

[Silent3]

9. Thank you!

I have to argue with people here about this all of the time.

A related canard is "national polls are useless" when talking about presidential races.

No, they aren't!

While it's true that the electoral vote, not the popular vote, determines the winner, you still don't need state-by-state information to tell if the race is close or not. Whenever someone wins by electoral vote and not popular vote, it's a close race.

It's also true that conditions often can change a lot from what polls show many months before an election, but that doesn't make early polls "worthless" (a word that so many love to use about polling). Those polls still tell you how much work needs to be done between the time of the poll and election day.

[Jarqui]

10. out of the 212 general election polls before the 2020 election

Trump lost something like 211 of them.

https://www.realclearpolitics.com/epolls/2020/president/us/general_election_trump_vs_biden-6247.html#polls

Then Trump says "they stole it"

As part of the theft did they rig 211 polls from a variety of pollsters?

I realize they're just polls but always wondered why the media didn't confront him on that ... because Trump often loves to cite polls.

[former9thward]

25. And we are told by posters the "2020 polls were the most inaccurate in 40 years"

It seems that would be something Trump would say.

[Iwasthere]

28. He was shocked they lost because he was told it was taken care of

He thought it was rigged because he couldn't have lost because HE felt he had it rigged.

[Aepps22]

16. Polls are just Data Points

Polls are data points that are useful to identify weaknesses or strengths but they aren't a guarantee of actual results. I think we all would do well to remove the emotional connection to polling (we're ahead vs we're behind) and focus on what the data says are the weaknesses of our candidates and do our part to help mitigate those weaknesses.

Voter turnout will always win the day. To the extent polling impacts turnout, we as individuals need to do our part to help get people registered and to the polls.

[Happy Hoosier]

19. I don;t think the polls are "rigged," But they seem to be under-estimting Dem turnout right now.

Polls are only as good as their population models.

[Elessar Zappa]

20. Basically agree but

polls haven’t been as accurate lately as they were in the past. But yes, we have to treat this as a close race and act accordingly. Donate, volunteer, and get out the vote.

[Fix The Stupid]

22. I like how some conspiracies here are immediatley rejected...

But other conspiracies..such as the ENTIRE Media in the US (even North America) is colluding to hurt the democrats.

You just gotta laugh sometimes.

"MY" Conspiracy theory? Totally legit...

"YOUR" Conspiracy theory? Met with ridicule and hand waving...

Again, comedy gold.

[totodeinhere]

26. It seems that some people here don't want to believe any pull that doesn't show

a Democrat winning. Yet we continue to see Republicans winning elections.

[keithbvadu2]

29. There are many polls. Pick the one you like and swear by it. Swear at the others.

(from Statistics 101 class) A telephone poll taken in the 20s or 30s indicated that a certain candidate would win.
Didn’t happen. The other guy won.
Turns out that it was not a sample of the general population.
Only the relatively well-to-do could afford phones so they did not sample the ‘common folk’.

[doc03]

30. I would rest easier if Biden was at 60. But many thought Hillary

had this "Blue Wall" and "She got this" in 2016. Hopefully these bad polls will motivate people
to get out and vote. I heard just the other day that Nate Silver had Obama's chances at being re-elected at
17% at this time.

[DestinyIsles]

31. Polls aren't a great predictor on what's going to happen in an election from over a year from now.

Remember Republicans were supposed to win the House the by 10 points with getting 25 + seats near the end of 2021 as polls were predicting. They wound up winning only 5 seats and winning popular vote by 3 points. No one is saying discount polls but they should be suspect this early out. I would be saying this even if Biden was leading by 10 points over Trump. So give me a break on the poll lecturing.