Andrew_S
Top Poster
Posts: 28,241
Member is Online
|
Post by Andrew_S on May 9, 2015 18:28:53 GMT
|
|
|
|
Post by greenchristian on May 9, 2015 18:55:11 GMT
It would be interesting to see an analysis of Ashcroft's constituency polls. Clearly some were more-or-less right, but others were way off.
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 39,015
Member is Online
|
Post by The Bishop on May 9, 2015 20:25:21 GMT
As I was going to say to simonk's post earlier, the (mostly phone) polls that got 5-6% Tory leads had all sorts of odd internals - that is indeed why they were dismissed as rogue, not (just) because some didn't like their results. And phone polls converged in the last few days anyway...... (indeed if there was any last minute movement this time it was to Labour - in contrast to 1992 when a late pro-Tory swing *was* picked up, just not enough) Despite what Survation have now claimed, we also have to be wary of now presuming phone polls are "better". The fact is that since 2010 it was an online pollster (YouGov) who came closet to getting both the 2012 London mayoral election and last year's Euro elections right. I imagine much of the explanation for this week's debacle is, sadly, people not telling the truth (though at least as much as to *whether* they would vote as how) |
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 16,044
|
Post by Sibboleth on May 9, 2015 21:42:44 GMT
I repeat, a large part of it was getting turnout wrong - none predicted it would be only 1% up on 2010 (so any "lies" may have been on whether they would vote, not how) That surely benefited the Tories significantly...... Good old Occam. |
|
|
|
Post by froome on May 10, 2015 6:00:25 GMT
I don't think the national VI polls were that far out and the general error may be attributable to last minute changes of mind amongst voters . The interpretation of the polls into a result forecast was awry. Agreed. And a decent model was pretty difficult to create. True. Though I think Bristol West turned out to be reasonably accurate, so some of them were OK. Perhaps going through them all there's some room to work out lessons for future constituency polls. Agree with all of GC's points. I think the polls were not that far out from the situation one or two days before polling day, but there were a lot of people still wavering or completely undecided. On polling day those who did vote then went heavily for the Conservatives. I think it was a mixture of two factors: a) Opting for the status quo is a common feature for last minute decisions if you are not confident of the alternatives. b) They compared Cameron to Miliband as potential leaders and made their decision. Constituencies like Bristol West were different. The national picture came less into it, here it was a keen campaign between different alternatives to the status quo. |
|
Pimpernal
Forum Regular
A left-wing agenda within a right-wing framework...
Posts: 2,873 
|
Post by Pimpernal on May 10, 2015 6:16:06 GMT
I don't think the polls could have got it right. They have become drivers of opinion and voting rather than reflections of them.
Had the polls consistently shown an 8% Tory lead then the result would have been a lot closer. You wouldn't have had the whipped up 'fear of the SNP' and that drive to keep them out would not have materialised. I think it cost UKIP about 3-4% and probably a few votes for Labour too. We certainly had a lot of UKIP-minded people telling us late on that they would after all vote Tory to stop the Labour/SNP.
Perhaps there is a case, given the fact that they have a false impact on the results, to follow France and ban opinion polls during the campaign itself?
|
|
hedgehog
Non-Aligned
Enter your message here...
Posts: 6,826
|
Post by hedgehog on May 10, 2015 6:56:45 GMT
I dont see how you could stop people carrying out a survey of voting intentions, however the number of national polls did seem crazy.
Just glancing at the last 3 constituency polls, Swindon South, Sheffield Hallam and Thanet South, it struck me how accurate they were.
I dont see where this pollster bashing is coming from, they were opinion polls, last minute switching or not bothering to go and vote cant be blamed on the pollsters.
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 39,015
Member is Online
|
Post by The Bishop on May 10, 2015 9:35:08 GMT
Agreed. And a decent model was pretty difficult to create. True. Though I think Bristol West turned out to be reasonably accurate, so some of them were OK. Perhaps going through them all there's some room to work out lessons for future constituency polls. Agree with all of GC's points. I think the polls were not that far out from the situation one or two days before polling day, but there were a lot of people still wavering or completely undecided. On polling day those who did vote then went heavily for the Conservatives. I think it was a mixture of two factors: a) Opting for the status quo is a common feature for last minute decisions if you are not confident of the alternatives. b) They compared Cameron to Miliband as potential leaders and made their decision. Constituencies like Bristol West were different. The national picture came less into it, here it was a keen campaign between different alternatives to the status quo. Not sure about b) - the Cameron/Miliband thing was already mostly priced in IMO. Much more important was the Great Scots Scare - and many on the ground in the marginals we should have won but didn't agree...... |
|
Richard Allen
Banned
Four time loser in VUKPOTY finals
Posts: 19,052
|
Post by Richard Allen on May 10, 2015 9:35:19 GMT
I dont see where this pollster bashing is coming from, they were opinion polls, last minute switching or not bothering to go and vote cant be blamed on the pollsters. It is coming from people who don't understand opinion polling. |
|
Pimpernal
Forum Regular
A left-wing agenda within a right-wing framework...
Posts: 2,873
|
Post by Pimpernal on May 10, 2015 11:59:33 GMT
I dont see where this pollster bashing is coming from, they were opinion polls, last minute switching or not bothering to go and vote cant be blamed on the pollsters. It is coming from people who don't understand opinion polling. Oh contrare - it's coming from people who do understand opinion polling and realise it is opinion forming rather than opinion reflecting. So much of it was based on assumptions and adjustments that were questionable, especially in the Constituencies. |
|
|
|
Post by swindonlad on May 11, 2015 7:34:33 GMT
I dont see how you could stop people carrying out a survey of voting intentions, however the number of national polls did seem crazy. Just glancing at the last 3 constituency polls, Swindon South, Sheffield Hallam and Thanet South, it struck me how accurate they were. I dont see where this pollster bashing is coming from, they were opinion polls, last minute switching or not bothering to go and vote cant be blamed on the pollsters. Looking at South Swindon (which I know best) the last Ashcroft poll was C 37 L 36 (+1), final figures C46 L 35 (+11), some difference Mind you at 10pm both parties were still thinking it was very close I think it was a last minute decision, the reports coming back from the polling stations (not so much the VIs but the response to the tellers) in the marginal wards were good but it was not being picked up before that |
|
neilm
Non-Aligned
Posts: 25,023
|
Post by neilm on May 13, 2015 11:41:18 GMT
This was the poll that changed my vote. Right up until the day, I'd intended to vote UKIP and then someone reminded me of this. It loomed over me as I wandered to Wavendon Gate Pavilion to vote and I resolved to vote Conservative on the way in. I really thought Labour were in with a chance here. |
|
Andrew_S
Top Poster
Posts: 28,241
Member is Online
|
Post by Andrew_S on May 13, 2015 11:44:20 GMT
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 39,015
Member is Online
|
Post by The Bishop on May 13, 2015 11:46:04 GMT
That's referenced earlier in this thread, isn't it?
|
|
Andrew_S
Top Poster
Posts: 28,241
Member is Online
|
Post by Andrew_S on May 13, 2015 11:46:39 GMT
That's referenced earlier in this thread, isn't it? Apologies if so. |
|
Richard Allen
Banned
Four time loser in VUKPOTY finals
Posts: 19,052
|
Post by Richard Allen on May 13, 2015 12:22:47 GMT
It does raise the interesting question of how many other polling companies might have not published polls because they didn't fit the narrative. |
|
|
|
Post by Robert Waller on May 13, 2015 19:45:15 GMT
Yes, Richard, I remember during the 1992 campaign when we (Harris) published a national poll showing the Conservatives significantly ahead - and received a storm of abuse for what was instantly branded a 'rogue'.
We changed the methodology after that to make sure we were more in line with the received wisdom.
As for the polls in general, I've said these things before, I know:
1. No poll except the exit poll should be treated as a prediction of the eventual result.
2. Far too much attention is paid to the VI figures and not nearly enough to other questions such as best for the economy and best PM.
3. I believe the main problem is not exactly 'shy Tories' or 'late swing' but the fact that communicating (however you do it) with pollsters about voting intention is not the same thing as actually voting. The former is free of charge, the latter is not, they are different acts.
Anyway, it can't be my fault this time!
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on May 13, 2015 20:03:07 GMT
Are we to see the end of the YouGov daily then?
|
|
|
|
Post by manchesterman on May 14, 2015 2:07:03 GMT
Seems that the old "rule" still stands that a party cannot win where it is neither the "most trusted to take care of the economy" nor "has the best-suited PM". Parties can (and often do) win where they lead in one of the 2 but never win where they lead in neither.
|
|
|
|
Post by carlton43 on May 14, 2015 7:53:47 GMT
Yes, Richard, I remember during the 1992 campaign when we (Harris) published a national poll showing the Conservatives significantly ahead - and received a storm of abuse for what was instantly branded a 'rogue'. We changed the methodology after that to make sure we were more in line with the received wisdom. As for the polls in general, I've said these things before, I know: 1. No poll except the exit poll should be treated as a prediction of the eventual result. 2. Far too much attention is paid to the VI figures and not nearly enough to other questions such as best for the economy and best PM. 3. I believe the main problem is not exactly 'shy Tories' or 'late swing' but the fact that communicating (however you do it) with pollsters about voting intention is not the same thing as actually voting. The former is free of charge, the latter is not, they are different acts. Anyway, it can't be my fault this time! That is so important a post. It explains everything . The crucial to points for me being a) snapshot in time v. actuality on the day; and b) abstract musings ahead of event v. concentrated mind on important event on the day. Those are such different animals in nature and consequence. All along my concern was with the graph trend of the poll of polls and that does contain real information. |
|
