So much for opinion polls!
Discussion
Whilst the exit polls turned out to be surprisingly accurate, can we all now agree that opinion polls in the run up to an election are completely meaningless?
I am sick to death of having predictions based on these polls shoved down my neck on a daily basis for weeks before the election as though they were a serious prediction of the outcome.
From memory they were calling for around 33% Lib Dem and late 20%s for Labour right up to the last minute. What a load of s
t.
I am sick to death of having predictions based on these polls shoved down my neck on a daily basis for weeks before the election as though they were a serious prediction of the outcome.
From memory they were calling for around 33% Lib Dem and late 20%s for Labour right up to the last minute. What a load of s
t.Remember that the percentage of the vote is not related in any simple way to the number of seats. The percentage figures were pretty accurate.
| Party | Seats | Gain | Loss | Net | Votes | % | +/-% |
|---|---|---|---|---|---|---|---|
| Conservative | 291 | 95 | 3 | +92 | 10,213,492 | 36.0 | +3.9 |
| Labour | 251 | 2 | 89 | -87 | 8,307,487 | 29.3 | -6.3 |
| Liberal Democrat | 52 | 8 | 13 | -5 | 6,481,602 | 22.9 | +1.0 |
Edited by Goochie on Friday 7th May 12:48
Goochie said:
Remember that the percentage of the vote is not related in any simple way to the number of seats. The percentage figures were pretty accurate.
Isn't the share of the vote something like this so far:Con 37
Lab 29
LbD 23
So with many polls putting Lib and Lab neck and neck at 32 or 33 the polls were way out.
turbobloke said:
Goochie said:
Remember that the percentage of the vote is not related in any simple way to the number of seats. The percentage figures were pretty accurate.
Isn't the share of the vote something like this so far:Con 37
Lab 29
LbD 23
So with many polls putting Lib and Lab neck and neck at 32 or 33 the polls were way out.
Interesting article, still doesn't come to the right conclusion though:
http://news.bbc.co.uk/1/hi/uk_politics/election_20...
http://news.bbc.co.uk/1/hi/uk_politics/election_20...
elster said:
The opinion polls were roughly correct before the debates.
Apart from a bit of a switch the Brown and Clegg.
Is that a joke?! Are you saying that the opinion polls were not totally incorrect at one point in time, several weeks before the election, and so they're ok? Apart from a bit of a switch the Brown and Clegg.
Even a stopped clock is right twice a day!
youngsyr said:
elster said:
The opinion polls were roughly correct before the debates.
Apart from a bit of a switch the Brown and Clegg.
Is that a joke?! Are you saying that the opinion polls were not totally incorrect at one point in time, several weeks before the election, and so they're ok? Apart from a bit of a switch the Brown and Clegg.
Even a stopped clock is right twice a day!
They got UKIP share of the vote almost spot on.
They correctly predicted what happened in Scotland, though not the scale.
They got the Tories and Labour share wrong by about 3% each way. A significant difference . What was key was how national voting intentions were translated to marginal seats. Also how Conservatives were doing better than nationally in many areas.
There have been some suggestions that polling companies were seeking to lead public opinion rather than record it. Whether that is the case is another matter.
They correctly predicted what happened in Scotland, though not the scale.
They got the Tories and Labour share wrong by about 3% each way. A significant difference . What was key was how national voting intentions were translated to marginal seats. Also how Conservatives were doing better than nationally in many areas.
There have been some suggestions that polling companies were seeking to lead public opinion rather than record it. Whether that is the case is another matter.
Like it or not (and I don't) the pollsters are onto a winner with a media that craves storylines or generic info they can spin/disseminate/blather on and on and on about...
It's like rating agencies service to financial media (they have been majorly wrong in the past but continue to be 'story fodder' whenever it suits).
As many have said, the only poll that matters is the GE - everything else is immaterial.
It's like rating agencies service to financial media (they have been majorly wrong in the past but continue to be 'story fodder' whenever it suits).
As many have said, the only poll that matters is the GE - everything else is immaterial.
Exit polls were virtually on the money, so much so that Paddy Ashdown needs to eat a hat soon.
It was the previous polls that were out. Consistently, regardless of who conducted them.
If you engineer a gearbox and it turns out that your assumptions were wrong and it is noisy and vibrates, do you:
1. Agree that it is not possible to engineer a decent gearbox, pack up and go home, or
2. Challenge your assumptions, figure out what you did wrong and have another go?
I'd say this gives us a VERY good opportunity to understand what went wrong with the polls leading up to the election.
This is worth a read, if you are interested.....
http://www.newscientist.com/article/dn27484-how-di...
And for those that think that predicting the outcome of an election should be much easier than designing a gearbox.....
http://www2.warwick.ac.uk/fac/sci/statistics/staff...
It was the previous polls that were out. Consistently, regardless of who conducted them.
If you engineer a gearbox and it turns out that your assumptions were wrong and it is noisy and vibrates, do you:
1. Agree that it is not possible to engineer a decent gearbox, pack up and go home, or
2. Challenge your assumptions, figure out what you did wrong and have another go?
I'd say this gives us a VERY good opportunity to understand what went wrong with the polls leading up to the election.
This is worth a read, if you are interested.....
http://www.newscientist.com/article/dn27484-how-di...
And for those that think that predicting the outcome of an election should be much easier than designing a gearbox.....
http://www2.warwick.ac.uk/fac/sci/statistics/staff...
Edited by NelsonP on Friday 8th May 13:22
Edited by NelsonP on Friday 8th May 14:29
So there's going to be an independent enquiry....
http://www.britishpollingcouncil.org/general-elect...
http://www.britishpollingcouncil.org/general-elect...
The polls fail to factor in the two essentials of Englishness. We love to moan and we don't like change.
So loads of people genuinely say "I'm not voting this lot back in" because of factor 1, but when they come to actually vote, think "better with the devil you know" due to factor 2.
History shows we only vote PMs out after something really bad happens.
Heath - 3 day week
Callaghan - winter of discontent
Major - sleaze
Brown - financial crisis
This lot hadn't done anything bad enough. We don't vote out incumbent governments for mediocrity. In fact we half expect it.
I predicted tories would get 326, and my 17 y/o son said 336. Unfortunately, I never put it in writing beforehand! But lots of family witnesses.
So loads of people genuinely say "I'm not voting this lot back in" because of factor 1, but when they come to actually vote, think "better with the devil you know" due to factor 2.
History shows we only vote PMs out after something really bad happens.
Heath - 3 day week
Callaghan - winter of discontent
Major - sleaze
Brown - financial crisis
This lot hadn't done anything bad enough. We don't vote out incumbent governments for mediocrity. In fact we half expect it.
I predicted tories would get 326, and my 17 y/o son said 336. Unfortunately, I never put it in writing beforehand! But lots of family witnesses.
article said:
A new report traces the roots of the pollsters’ failure to predict the Conservative majority in last year’s election to their lack of adequate contact with Tory supporters.
Earlier speculation about what went wrong has focused on poorly designed questionnaires, a late swing to the Conservatives, a failure of “lazy Labour” supporters to turn out, or reticence on the part of “shy Tories” to reveal their leanings.
But, ahead of Prof Patrick Sturgis’s presentation of the first findings of his postmortem for the British Polling Council next week, analysis by the leading psephologist Prof John Curtice blows these theories out of the water and suggests the problem was pollsters’ failure to reach the right people.
Earlier speculation about what went wrong has focused on poorly designed questionnaires, a late swing to the Conservatives, a failure of “lazy Labour” supporters to turn out, or reticence on the part of “shy Tories” to reveal their leanings.
But, ahead of Prof Patrick Sturgis’s presentation of the first findings of his postmortem for the British Polling Council next week, analysis by the leading psephologist Prof John Curtice blows these theories out of the water and suggests the problem was pollsters’ failure to reach the right people.
article said:
Work by pollsters themselves, including ICM and YouGov, is increasingly also pointing to the awkward explanation that they had the “wrong people” in their samples. But Curtice’s study sheds valuable new light on how they got the mix wrong.
He uncovers a link – independent of social class and age – between the ease with which voters can be got hold of, and their political leanings. Among those respondents whom the BSA researchers succeeded in talking to on their first visit, Labour was six points ahead. But among those who required between three and six knocks at the door, the Tories enjoyed an 11-point lead.
http://www.theguardian.com/politics/2016/jan/14/election-polling-errors-blamed-failure-speak-tory-votersHe uncovers a link – independent of social class and age – between the ease with which voters can be got hold of, and their political leanings. Among those respondents whom the BSA researchers succeeded in talking to on their first visit, Labour was six points ahead. But among those who required between three and six knocks at the door, the Tories enjoyed an 11-point lead.
article said:
A new report... uncovers a link – independent of social class and age – between the ease with which voters can be got hold of, and their political leanings.
The ones that can be got hold of are at home watching daytime TV while the ones that can't are out earning.Explains why the opinion polls for the Scottish Indyref showed a greater support for Yes than the actual result too...!
Gassing Station | News, Politics & Economics | Top of Page | What's New | My Stuff