Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

One pollster’s explanation for why the polls got it wrong

vox.com

Post reply on HN
Print view View original

81–90 of 565 posts

Re: One pollster’s explanation for why the polls got it wrong

#81

Post by wombatmobile »

jxramos wrote: post #5

I'm curious where polls actually work, are there any retrospectives polling agencies conduct and post online for scrutiny and constructive feedback? It'll probably help to enumerate all the premises that polling begins with and question whether those premises are in fact the case.

> I'm curious where polls actually work

If your definition of "work" is that a poll accurately predicts the outcome of the election, take a step back. Polls don't do that.

I'm not just referring to inaccuracies that are introduced by non-representedness of the sample, although that's clearly a factor as well.

Polls invite respondents to indicate how they would vote "if an election were held today". But the election isn't being held today, and the respondents know it. Polls are an invitation for respondents to send a message to politicians before the poll, to influence the politicians' offerings. Respondents know that, and use the opportunity that way.

So what are polls? Polls are their own little universe, with their own characteristics, dynamics, and effects. What they aren't is a crystal ball.

Re: One pollster’s explanation for why the polls got it wrong

#83

Post by dragonwriter »

bryanlarsen wrote: post #74
dragonwriter wrote: post #65

Earlier quoted context omitted.

> People 2020 said that in 2016, yet the average income of Trump voters was higher than Clinton voters. One person with $100 million income has a higher average income than 999 people with a $100K annual income. Averages aren't particularly useful here.

There aren't many people with $100 million income. Most billionaires don't have income that high -- there is a much higher wealth disparity than there is income disparity. So while I'd rather the media reported the median income than the median income, it's the number we have and hopefully it's a useful approximation.

> There aren't many people with $100 million income

Sure, but there's also a couple with over an order of magnitude more.

> So while I'd rather the media reported the median income than the median income, it's the number we have and hopefully it's a useful approximation.

It's not a useful approximation of anything, and we have a lot more detailed information that “average income of voters of one side vs. the other" on the relationship between income and voting / party preference. And it tells a more complicated story (e.g., in 2016, Clinton won significantly among those with income of $150K+ and under $30K, was nearly even among those with $100-150K and $50-75K, lost significantly among those with $74-100K and $30-50K.)

https://www.pewresearch.org/fact-tank/2020/10/30/a-voter-dat...

Re: One pollster’s explanation for why the polls got it wrong

#84

Post by vorpalhex »

dragonwriter wrote: post #50
TechBro8615 wrote: post #27

Earlier quoted context omitted.

It’s pretty obvious from election maps and donor data that the democrats are now the party of coastal/urban elites, despite what they may advertise.

> It’s pretty obvious from election maps and donor data that the democrats are now the party of coastal/urban elites Urban populations, sure. Not particularly coastal aside from the fact that there's more urban population on the coasts, as county-level maps show pretty clearly. “Elites” are more dubious. Sure, Democrats and Republicans (counting leaners) are closer to parity for voters with less than a college degree…

> OTOH, Blacks, who are overwhelming not elites, overwhelming are (and even more overwhelming lean) Democratic, and the same is true predominantly (though less overwhelmingly) of Hispanics.

I think that's actually the rub. Blacks and Hispanics are voting less and less for Democrats, and that is true more for this last election than 2016 [1]

Despite the very visible rhetoric of racial equality and other movements, the Dems are becoming the party of rich white people.

[1] - https://news.yahoo.com/exit-poll-black-men-drifted-204522278...

Re: One pollster’s explanation for why the polls got it wrong

#85

Post by webmaven »

tshaddox wrote: post #73
dragonwriter wrote: post #50

Earlier quoted context omitted.

> It’s pretty obvious from election maps and donor data that the democrats are now the party of coastal/urban elites Urban populations, sure. Not particularly coastal aside from the fact that there's more urban population on the coasts, as county-level maps show pretty clearly. “Elites” are more dubious. Sure, Democrats and Republicans (counting leaners) are closer to parity for voters with less than a college degree…

What are we calling "urban" here? More than 80% of Americans live in urban areas. Are we using a circular definition of "urban" that means "big cities with a large Democrat majority"?

> What are we calling "urban" here? More than 80% of Americans live in urban areas. Are we using a circular definition of "urban" that means "big cities with a large Democrat majority"?

Urban mostly means "denser than suburban tracts of single-family homes".

Re: One pollster’s explanation for why the polls got it wrong

#86

Post by bra-ket »

gnusty_gnurc wrote: post #15

It seems obvious to me that there was massive bias in media towards Biden, and they all happily accepted the narrative that Biden would win in a landslide and dismissed anyone saying otherwise as know-nothings. IMO that's tantamount to election interference, if there's near unanimous media coverage for the entire election season of inaccurate polling that assumes a landslide Biden victory. I didn't vote for Biden or…

absolutely, Biden has a chance only thanks to constant brainwashing by the media, 'orange man bad evil racist' broadcasted 24/7 for 4 years to millions of people.

If you repeat lies many times people start to believe in them even on subconscious level. Especially if you manipulate people emotions and subjective opinions.

Try asking anyone supporting Biden why do they so firmly believe Trump is 'racist'.

Their response will eventually come down to regurgitated media spin, which you can trace to news coverage by particular group of particularly biased sources.

MSM won the elections.

Re: One pollster’s explanation for why the polls got it wrong

#87

Post by gnusty_gnurc »

bryanlarsen wrote: post #63
gnusty_gnurc wrote: post #43

Earlier quoted context omitted.

> If the media should be giving equal status and weight to all the parties I'm not saying they need to give them equal status, necessarily - I'm saying that the employees of these institutions exist in a bubble mostly cause of class&geographic reasons. The point is that all of these positions "Biden will naturally win in a landslide per polling" are tacitly accepted and challenges are dismissed as hearsay. There need…

I didn't see any of the elite media saying "Biden will naturally win in a landslide per polling". They were saying that in 2016. In 2020 much more common was "the polls are showing a landslide, but I don't trust the polls completely any more". If you are saying that the polls were deliberately biased, I disagree strongly. It's a really freaking hard job if only 1-3% of the people you try to poll don't hang up the pho…

> but I don't trust the polls completely any more

That certainly was not the case. There was a bit of shell-shocked fear about 4 more years of Trump, but they most certainly did not expect a nail-biter election where Republicans outperformed in most areas, whether it's the presidential election in states(and pretty darn close to 10% off in some places), the Senate, the House or state legislatures.

Re: One pollster’s explanation for why the polls got it wrong

#88

Post by jerf »

koolba wrote: post #17
0x53 wrote: post #3

One of the biggest issues I think is that the polling is still done by just randomly calling people. I think I agree that there is a large skew introduced by the simple fact of only reaching people willing to answer a call from an unknown number.

Even the method of contact would skew toward people who have a means for the method of contact or location. Cell phones might include less seniors. Questioning people walking into Whole Foods is going to give you a different set than people walking into Cracker Barrel. If the source set of names and numbers is truly representative, I think random calling could work if you make sure to include "did not response" as a…

I suspect "did not respond" is already more than 15% and increasing constantly. I could easily see it already being 50%+. Even a layman would know to be suspicious of "12% Biden, 11% Trump, 77% No Response".

I can't even tell you if a pollster called me... I'm getting about half-a-dozen calls a day from rando numbers that want to refinance my nonexistent student loans or talk to me about my home's nonexistent warrantee. I went to permanent do-not-disturb months before the election was even in the news. If you don't leave me a voice message I don't even know you exist; if you do leave me a voice message you've got about 3 seconds to convince me you're not a spammer before it's trashed.

Re: One pollster’s explanation for why the polls got it wrong

#89

Post by orangecat »

tshaddox wrote: post #71
TechBro8615 wrote: post #27

Earlier quoted context omitted.

It’s pretty obvious from election maps and donor data that the democrats are now the party of coastal/urban elites, despite what they may advertise.

What does "coastal/urban elites" mean? Just anyone who lives in a coastal or urban area? 80% of the American population lives in urban areas, and that's only going to increase when you include whatever your definition of "coastal" is. Where does the "elite" factor in?

Where does the "elite" factor in?

As a rough approximation, almost everyone in your social circles having a college degree.

Re: One pollster’s explanation for why the polls got it wrong

#90

Post by macintux »

_AzMoo wrote: post #67
rootusrootus wrote: post #18

Maybe it doesn't really matter much. I mean, sure, we need some amount of polling because it informs decisions such as who gets to participate in televised debates, etc. But other than that, what does the general public get out of polling? It's a spectator sport. In the end, all that matters is the votes, and people should really be encouraged to remember this. I'd think 2016 and 2020 would be a healthy reminder of t…

Exit polls were historically used as a method of detecting election fraud, because they've always been very accurate. It's only the last few elections where we've seen such a large disparity in polling and actual results.

Is exit polling no longer working? All of the discussion I’ve seen seems to focus on polling in advance of the election.
Post reply on HN