FAQ
It is currently Sun Apr 30, 2017 1:20 am


Author Message
marnixR
Post  Post subject: why did the polls get it so wrong ?  |  Posted: Fri May 08, 2015 2:30 pm
User avatar

Joined: Thu Aug 04, 2011 8:35 pm
Posts: 4758
Location: Cardiff, Wales

Offline
this is not a thread about politics, but about the astounding matter of all pre-election polls being vastly at odds with the exit poll, which turned out to be far closer to the truth than all other polls combined

How did the pollsters get the UK general election so wrong?

although the New Scientist article touches on one major difference which is the difference between the intention to vote and actually having voted, i also must question the topic of the representativeness of the sample

after all, all statistics stand or fall with how good or bad the sample represents the population it aims to cover - if your sample isn't representative you're going to get a skewed start-off point for any conclusions you want to draw from it
imo contacting people over the phone or over the internet has the fault of not being able to tell for sure if you have sampled your audience in the same proportions as the actual voting population

_________________
"Reality is that which, when you stop believing in it, doesn't go away." (Philip K. Dick)
"Someone is WRONG on the internet" (xkcd)


Top
marnixR
Post  Post subject: Re: why did the polls get it so wrong ?  |  Posted: Fri May 08, 2015 7:46 pm
User avatar

Joined: Thu Aug 04, 2011 8:35 pm
Posts: 4758
Location: Cardiff, Wales

Offline
oh, one thing i picked in which the current election polls were different from previous ones, is that this time round there was far more reliance on internet polling

no doubt the search will be on how using this method leads to the bias that must have crept into the sampling
i can already see one possible source of bias, that is that the younger generation are far more connected at all times to the web, whereas some of the generation before me don't go on-line at all - the former are more likely to support left-leaning parties whereas the older ones are more likely to vote for right-leaning policies

there's also mention of a telephone poll which wasn't published because it was so out of line with previous polls

Election 2015: Inquiry into opinion poll failures

Quote:
Meanwhile, market research agency Survation said it had "chickened out" of publishing a telephone poll on Wednesday evening, which showed the Tories on 37% and Labour on 31%.
Its chief executive Damian Lyons said he would "always regret" the decision, but the results seemed "so out of line" with previous polls.

_________________
"Reality is that which, when you stop believing in it, doesn't go away." (Philip K. Dick)
"Someone is WRONG on the internet" (xkcd)


Top
iNow
Post  Post subject: Re: why did the polls get it so wrong ?  |  Posted: Fri May 08, 2015 10:10 pm
User avatar
Original Member
Original Member

Joined: Thu Aug 04, 2011 11:40 pm
Posts: 5500
Location: Austin, Texas

Offline
What do we know about their source data and assumptions? That's where US pollsters frequently go wrong and where Nate Silver has consistently been right.

A cool article actually just hit my feed on this topic:

http://mindhacks.com/2015/05/08/samplin ... us-friend/
Quote:
As the UK election results roll in, one of the big shocks is the discrepancy between the pre-election polls and the results. All the pollsters agreed that it would be incredibly close, and they were all wrong. What gives?

Some essential psych 101 concepts come in useful here...
<snip>
The exit polls predicted the real result with surprising accuracy not because they minimised sampling error, but because they avoided the sample bias. By asking the people who actually turned up to vote how they actually voted, their sample lacked the bias of the pre-election polls.

_________________
iNow

"[Time] is one of those concepts that is profoundly resistant to a simple definition." ~C. Sagan


Top
Olinguito
Post  Post subject:   |  Posted: Sat May 09, 2015 8:27 am
User avatar

Joined: Mon Aug 19, 2013 3:56 pm
Posts: 132

Offline
They got it even more wrong at the 1992 UK general election. Everyone thought Labour would win, given that the Conservatives were in a bit of a crisis, with the introduction of poll tax and Margaret Thatcher resigning two years before. In the end, John Major still won by a majority of +101, and Neil Kinnock resigned.

_________________
Blog


Top
marnixR
Post  Post subject: Re: why did the polls get it so wrong ?  |  Posted: Sat May 09, 2015 8:05 pm
User avatar

Joined: Thu Aug 04, 2011 8:35 pm
Posts: 4758
Location: Cardiff, Wales

Offline
i seem to remember that in that case the exit pools got it totally wrong

_________________
"Reality is that which, when you stop believing in it, doesn't go away." (Philip K. Dick)
"Someone is WRONG on the internet" (xkcd)


Top
kojax
Post  Post subject: Re: why did the polls get it so wrong ?  |  Posted: Fri May 22, 2015 1:33 pm
Original Member
Original Member

Joined: Mon Aug 15, 2011 11:43 am
Posts: 582

Offline
marnixR wrote:

although the New Scientist article touches on one major difference which is the difference between the intention to vote and actually having voted, i also must question the topic of the representativeness of the sample



Especially since many people you sample probably aren't even going to vote in the first place. They'll forget what day it is on, or be too busy or something.

If you could find a statistical way to focus the sample on people who are very likely to actually vote, I imagine that that might help. But then we'd probably be neglecting some other attributes, or failing to factor in the people who have low statistical likelihood of voting but who vote anyway.

Perhaps sample them too but give them a lower weight?

So like maybe only one in 10 tree huggers are actually going to wake up on the morning of the vote and go to the poll, so we assign them a factor of 0.1 when we poll them? But the conservative small business owner with coveralls on ready to work might have a likelihood of 95%, so we give him a factor of 0.95 (And probably the differences aren't quite that extreme, but I'm just throwing silly numbers out there for example sake.)


And, we can determine our factoring by analysing the demographic makeup of the people exiting the polls in the previous election, and comparing that with the overall population.


Top
marnixR
Post  Post subject: Re: why did the polls get it so wrong ?  |  Posted: Fri Jul 17, 2015 7:39 pm
User avatar

Joined: Thu Aug 04, 2011 8:35 pm
Posts: 4758
Location: Cardiff, Wales

Offline
so in the end the answer to the question why the polls had it so wrong had nothing to do with mathematics, but everything with human nature

many people who claimed they were going to vote failed to turn up, and for some reason or another, the "lazy" voters were the ones who intended to vote for labour

Lazy, lying Labour supporters to blame for surprise Tory election win, study finds

_________________
"Reality is that which, when you stop believing in it, doesn't go away." (Philip K. Dick)
"Someone is WRONG on the internet" (xkcd)


Top
Falconer360
Post  Post subject: Re: why did the polls get it so wrong ?  |  Posted: Fri Jul 17, 2015 8:31 pm
User avatar
Original Member
Original Member

Joined: Thu Aug 04, 2011 10:30 pm
Posts: 962
Location: Somewhere in the Great State of Washington

Offline
marnixR wrote:
so in the end the answer to the question why the polls had it so wrong had nothing to do with mathematics, but everything with human nature

many people who claimed they were going to vote failed to turn up, and for some reason or another, the "lazy" voters were the ones who intended to vote for labour

Lazy, lying Labour supporters to blame for surprise Tory election win, study finds


Not all that surprising really. Look at the US re-election rate for our Congress last year. Almost all of the Conservatives that should have been voted out due to having a record low approval rating were re-elected. This is because the majority of the people that disapproved of them were younger and didn't turn out to vote, while the older (more conservative) population voted. Off the top of my head I think somewhere around only 30% ended voting and thus determining the representatives for everyone. Low voter turnout is slowly killing the US, and the UK by the looks of it.

_________________
"For every moment of triumph, for every instance of beauty, many souls must be trampled." Hunter S Thompson
"It is easy to kill someone with a slash of a sword. It is hard to be impossible for others to cut down" - Yagyu Munenori


Top
steve upson
Post  Post subject: Re: why did the polls get it so wrong ?  |  Posted: Tue May 24, 2016 1:45 am
User avatar

Joined: Tue May 24, 2016 12:25 am
Posts: 14

Offline
This is the very first time that I've heard of anyone asking this question on a science forum.

The real answer is pretty disturbing, and is best explained by the discussion here at DailyKos.com:

http://www.dailykos.com/story/2012/10/2 ... for-Romney

Back when I was following the story there were two writers that understood the issues well, and they reported on several political sites and fora. They went by Michael Collins and TruthIsAll.

The discussion at DailyKos.com covers the problem rather well, although there were better articles at the time that covered this issue. Ron Paul supporters were pretty much all aware of the problem and tried to raise it as an issue during the primaries, to no avail.

My personal views are that it has only gotten worse since then.

For a short synopsis, they electronically switched votes from the leading candidate to Romney. "They" repeated this scheme at least 12 times, until all the leading candidates dropped out, one after the other.

It has been shown mathematically that the official certified vote (from the states' election boards) is the only data necessary to prove that this is so. The reason that the vote flipping can be shown by just looking at the published vote count is because they used an algorithm that increased the vote for Romney based on the number of votes in the precinct. Remember, this is based on the number of actual votes, not the size of the precinct. This means that neither precinct-to-precinct voter turnout, nor precinct size, nor geographical location can account for the anomaly in the data.

It was reported that an academic group tried to study the odds of the anomaly occurring organically but it outran their software's ability to calculate the odds. They estimated over one google-to-one odds.

After Bush v. Gore, Voter News Service (a scientific organization) was retired and an outfit called Edison/Mitofsky (a pseudo-science political organization) was used instead. Their main contribution to the way we do things now was to convince the public that weight factors should only be added to the exit polling after the polls close and the vote is in.

Sort of defeats any scientific reasons for doing exit polling in the first place.

Think about it. Their methodology changes the exit polling figures in order to make them match the actual recorded vote. What's the point of doing that? There is no scientific reason. The only purpose that I can think of to do it that way is for the propaganda value.

Now, today, the exit polling is all done by the main stream media's National Election Pool (a propaganda organization):https://en.wikipedia.org/wiki/National_Election_Pool


Thank you for asking. I just had to get that out. It's about time the scientific community started asking questions.


Top
Display posts from previous:  Sort by  
Print view

Who is online
Users browsing this forum: No registered users and 0 guests
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
Jump to:   


Delete all board cookies | The team | All times are UTC


This free forum is proudly hosted by ProphpBB | phpBB software | Report Abuse | Privacy