john07
Labour & Co-operative
Posts: 15,818 
|
Post by john07 on Dec 19, 2015 1:09:39 GMT
I see some people are swooning over the Lord Ashcroft poll - "20k SAMPLE! WOW!!" In actual fact, a poll of that size is no more intrinsically reliable than the more usual 1k-2k surveys - indeed if the original methodology is flawed, a bigger sample could make it more "wrong" not less. Also, there were two phone surveys earlier this week - ComRes and MORI - both showing big "Remain" leads. The difference between internet and telephone surveys on this is becoming quite marked now (their results on VI and other stuff is much more similar) and I would be interested in possible explanations? Possibly the largest sample in an opinion poll of all time was the infamous Literary Digest Poll on the 1936 US Presidential Election. The sample size was 2.5 million and it predicted a landslide win for Alf Landon over FDR. Landon was predicted to take 55% of the vote. Landon carried the bellwether state of Maine. As Maine goes, so goes the nation so the adage went. FDR took 46 states with Landon adding only Vermont. The new adage became As Maine goes so goes Vermont. Roosevelt ended up with 61%. It doesn't matter how big a sample is if it is biased. The Literary Review sample was taken from subscribers, car owners and Bell Telephone users. |
|
|
|
Post by Davıd Boothroyd on Dec 19, 2015 9:09:07 GMT
Worth remembering in 1936 Maine still held its election for Presidential electors in September, earlier than the rest of the US which polled in November. So the belief 'as Maine goes, so goes the nation' (not actually very accurate but quite a few believed it) led credibility to the Literary Digest.
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 39,009
|
Post by The Bishop on Dec 23, 2015 10:49:36 GMT
Latest survey from ICM - Remain 45, Leave 38.
|
|
|
|
Post by marksenior on Dec 23, 2015 17:14:36 GMT
Yes v different Bish. Looking at the mori poll I noticed they put both UKIP and Lib Dems on 9%. On the ICM poll it had 258 people intending to vote UKIP and only 118 Lib Dems. ICM also asked people how they had voted at the general election and found the ratio of UKIP:Lib dem voters was 234/146, which was pretty much the ratio it was at the general election. That would make me have a lot more faith in the ICM poll being representative. If mori don't ask how people voted at the general election how do they know they have a good representative sample of actual voters across the spectrum? Mori do ask how people voted at the GE . See their full data computer tables . They do not though weight by this past voting data . What is interesting this month is that both ICM and Mori found far fewer voters admitting they voted UKIP at the GE than actually did so . The ICM weightings will adjust for this somewhat whereas Mori will not . |
|
|
|
Post by greenchristian on Dec 23, 2015 19:26:25 GMT
Yes v different Bish. Looking at the mori poll I noticed they put both UKIP and Lib Dems on 9%. On the ICM poll it had 258 people intending to vote UKIP and only 118 Lib Dems. ICM also asked people how they had voted at the general election and found the ratio of UKIP:Lib dem voters was 234/146, which was pretty much the ratio it was at the general election. That would make me have a lot more faith in the ICM poll being representative. If mori don't ask how people voted at the general election how do they know they have a good representative sample of actual voters across the spectrum? Mori do ask how people voted at the GE . See their full data computer tables . They do not though weight by this past voting data . What is interesting this month is that both ICM and Mori found far fewer voters admitting they voted UKIP at the GE than actually did so . The ICM weightings will adjust for this somewhat whereas Mori will not . Why do we think the figure is low? Is it false recall, a bad sample, or UKIP voters refusing to admit it? |
|
|
|
Post by marksenior on Dec 23, 2015 23:06:38 GMT
Mori do ask how people voted at the GE . See their full data computer tables . They do not though weight by this past voting data . What is interesting this month is that both ICM and Mori found far fewer voters admitting they voted UKIP at the GE than actually did so . The ICM weightings will adjust for this somewhat whereas Mori will not . Why do we think the figure is low? Is it false recall, a bad sample, or UKIP voters refusing to admit it? Good question , a similar phenomenon was often present with Lib Dem voters at previous elections so it could be that the none of the above voters who transferred this time from Lib Dem to UKIP are continuing their false recall or refusal to admit who they voted for . The 2010 GE had a rather different situation with clear evidence that some Labour voters were saying that they had voted Lib Dem but would never do so again because of the coalition . |
|
tim
UKIP
Posts: 602
|
Post by tim on Jan 4, 2016 9:39:55 GMT
|
|
|
|
Post by Andrew_S on Jan 16, 2016 23:55:24 GMT
Survation:
Leave 53%
Remain 47%
|
|
|
|
Post by afleitch on Jan 17, 2016 14:15:16 GMT
Panelbase
Scotland Remain 65%, Leave 35%
Also notes that Yes would win on 52% if there was an independence referendum after the UK voted to leave.
|
|
|
|
Post by Andrew_S on Jan 17, 2016 19:40:29 GMT
Panelbase Scotland Remain 65%, Leave 35% Also notes that Yes would win on 52% if there was an independence referendum after the UK voted to leave. The question is whether people in Scotland are genuinely more pro-EU than England or whether they're just supporting the opposite point of view of whatever point of view they think people in England are supporting.  |
|
tim
UKIP
Posts: 602
|
Post by tim on Jan 17, 2016 21:05:18 GMT
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 17, 2016 22:21:56 GMT
Panelbase Scotland Remain 65%, Leave 35% Also notes that Yes would win on 52% if there was an independence referendum after the UK voted to leave. The question is whether people in Scotland are genuinely more pro-EU than England or whether they're just supporting the opposite point of view of whatever point of view they think people in England are supporting. I think Scots are marginally less attached to the idea of Britain's greatness and therefore marginally more supportive of the EU. But the explanation of the huge gap is that a considerable body of opinion in the SNP favours the idea of constitutional change imposed against the will of the people of Scotland. 'We'll deal with the EU after independence', they say. Really. They do. |
|
Tony Otim
Green
Suffering from Brexistential Despair
Posts: 11,913
Member is Online
|
Post by Tony Otim on Jan 19, 2016 12:23:19 GMT
Well, that, and for a certain typ of Nat, Scotland voting convincingly to remain whilst the Uk votes narrowly to leave is probably the ideal result for reason that have nothing to do with Europe.
|
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 16,044
|
Post by Sibboleth on Jan 19, 2016 18:40:37 GMT
A polling firm that cuts out 'don't knows' on a referendum question should be treated with disdain.
|
|
|
|
Post by greatkingrat on Jan 19, 2016 19:52:49 GMT
A polling firm that cuts out 'don't knows' on a referendum question should be treated with disdain. The raw tables give both sets of figures (Remain 54%, Leave 30%, Don't Know 16%). You would be better off blaming the media organisations who commission the polls as they are the ones who choose which of the findings they want to highlight. |
|
tim
UKIP
Posts: 602
|
Post by tim on Jan 21, 2016 15:52:20 GMT
|
|
|
|
Post by mrhell on Feb 5, 2016 13:11:45 GMT
Yougov
Stay 36%
Leave 45%
Meanwhile an Express online poll has 92% wanting to leave
|
|
sdoerr
Conservative
Posts: 149
|
Post by sdoerr on Mar 14, 2016 16:21:25 GMT
Any more recent?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 16, 2016 11:16:30 GMT
Starting to believe it could actually happen.
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 39,009
|
Post by The Bishop on Apr 16, 2016 13:32:30 GMT
There has actually been little poll movement in recent weeks, despite events.
|
|
