Wednesday, August 13, 2014

No lead slumps by 3% among women in new female-only Survation poll

A rather odd new referendum poll from Survation is out, conducted among women only, who of course tend to be somewhat less likely to be Yes voters than men. However, it's still possible to work out the trend from the last Survation poll which was conducted immediately after the leaders' debate, and that trend is favourable for Yes. So if there ever was a significant post-debate boost for No (it may well have been an illusion caused by normal sampling variation), it appears to be fading fast.

Should Scotland be an independent country? (Women only)

Yes 34% (+1)
No 50% (-2)


With Don't Knows excluded, that works out as...

Yes 40% (+1)
No 60% (-1)


Those changes would be consistent with a No lead of about 10% or so among the population as a whole, which was fairly standard for Survation during the spring. With undecideds excluded, that would equate to roughly Yes 44% or 45%, and No 56% or 55%.

This poll was commissioned by the Daily Record, and it appears to have completely replaced their normal monthly referendum poll, which on the face of it rather defeats the purpose of having an ongoing monthly series. The logic may be that there have been two very recent full-scale Survation polls in the Daily Mail, which would render a third one redundant, thus offering an opportunity to do something a little different. But why a female-only poll in particular? I know there are many people in both the Yes and No camps who will welcome a focus on women's views, but given that this is an anti-independence newspaper we're talking about, I find it hard not to be cynical about the motivations for this. I suspect it may be a cunning wheeze to hoodwink the Record's readers into thinking that the No lead is wider than it really is.

Certainly every time we see one of these 'special' polls, it always conveniently seems to be among a No-friendly demographic. I believe this is the second female-only poll, and there was also a Populus poll of over-50s a few months ago. And let's not forget those two bizarre full-scale ComRes telephone polls that were confined to the Borders and Dumfries & Galloway - a particularly No-friendly part of Scotland containing just 5% of the national population. I've got a suggestion for next time - how about a poll of low-income male Glaswegians, aged between 25 and 44?

* * *

As Scottish Skier pointed out a few hours ago, TNS-BMRB have followed in the footsteps of Survation and YouGov by producing a poll that has a slightly lower No lead in the raw unweighted data than it does in the weighted results for publication. It really is extraordinarily unusual for three polls in a row to show this pattern, because Yes are generally weighted up rather than down. It may of course just be pure coincidence, but there could also be something interesting going on beneath the surface which might point to the Yes vote being underestimated in these recent polls. In the case of the TNS poll, part of the explanation seems to be that the No lead is much bigger among people who say they didn't vote in 2011 than it is among the whole sample, which doesn't usually happen. The responses of those non-voters have been significantly upweighted in the overall results, in line with TNS-BMRB's highly questionable weighting procedure. Also, people who recall voting for minor parties in 2011 have broken heavily for Yes this month, and they have been subject to an extreme downweighting, from 33 real respondents to only 6 'virtual' respondents.

I predicted last night that simply stripping out the small number of respondents who say they are certain not to vote in the referendum would be sufficient to increase the Yes vote by an appreciable amount, and so it has proved - it actually trims the No lead by a full 2.6%. Here are the figures for each level of likelihood to vote, with Don't Knows excluded...

Whole sample (equivalent to 100% turnout) :

Yes 41.6% (-2.0)
No 58.4% (+2.0)

Whole sample excluding only definite non-voters (equivalent to 93% turnout) :

Yes 42.9% (-1.4)
No 57.1% (+1.4)

Respondents who say they are certain or very likely to vote (equivalent to 79% turnout) :

Yes 44.6% (-0.4)
No 55.4% (+0.4)

Respondents who say they are certain to vote (equivalent to 71% turnout) :

Yes 45.2% (+0.6)
No 54.8% (-0.6)


The last set of figures represent the narrowest gap of the campaign so far in a TNS poll.

You probably don't need me to point out that the Yes vote gets higher as the assumed turnout gets lower. And I don't want to alarm anyone in the No campaign (much), but a 71% or 79% turnout does sound a hell of a lot more plausible than a 93% or a 100% turnout.

42 comments:

  1. Any numbers on certainty to vote James?

    ReplyDelete
  2. This comment has been removed by the author.

    ReplyDelete
  3. Thanks James - your work is greatly valued,- I share most all your posts on my pages.

    ReplyDelete
  4. BVB1909 : The datasets aren't out yet, but Survation weight their headline numbers by likelihood to vote anyway (unlike TNS).

    By the way, I've had to make a slight correction to this post, because I was looking at the wrong table in the previous Survation datasets. However, it's still a 3% drop in the No lead.

    ReplyDelete
  5. It will be amusing to see just how many out of touch sites stuck in the westminster bubble fall for the Record's laughable spin.

    I doubt that many of them even know what the Record is like to be honest.

    LOL

    ReplyDelete
  6. There was a poll of Big Issue sellers to be fair (who you might expect would be more inclined to vote Yes):

    Yes: 46
    No: 50

    ReplyDelete
  7. And the news is even better in the subset of redheads...Oh dearie me!

    ReplyDelete
  8. I just do not agree with these polls at all ,another weird subset ,chosen to be no friendly (assumed )
    I had a look at the raw data from the last Survation poll and the regional data sets were all over the place ,with Yes leads in places you just would not expect and conversely No leads in strong yes areas
    With the up weighting ,down weighting according to how you voted in the last GE and the same for country of origin do you not think that these polls are just completely wrong ? they are basing the evidence on GE principles which I would argue are non sensical in a Referendum
    Not only that ,there is the question of the PR system that we use here for SG elections skewing the picture even more
    From what I understand they failed to predict the SNP landslide ,I think they are simply wrong in this referendum
    Look at the canvass results ,look at the public meetings ,look at the BT together campaign having to pay people ,look at the tumbleweed blowing through BT meetings
    It simply does not make sense

    ReplyDelete
  9. The Big Issue sellers survey was a straw poll conducted in late 2012. If anyone told me the majority of people who find themselves in this tragic situation support the core aims of the NoScotland campaign for the status quo, ie Trident WMD's, House of Lords, PFI, London Authoritarianism, Austerity, Welfare cuts and preservation of privilege, I'd organise a straight jacket and an ambulance. Poverty and homelessness is a direct result of Westminster's corrupt class based system nurtured by Tory/NewLabour politicians.

    Quite simple really.

    ReplyDelete
  10. The Big Issue sellers survey was a straw poll conducted in late 2012. If anyone told me the majority of people who find themselves in this tragic situation support the core aims of the NoScotland campaign for the status quo, ie Trident WMD's, House of Lords, PFI, London Authoritarianism, Austerity, Welfare cuts and preservation of privilege, I'd organise a straight jacket and an ambulance. Poverty and homelessness is a direct result of Westminster's corrupt class based system nurtured by Tory/NewLabour politicians.

    Quite simple really.

    ReplyDelete
  11. Survation tweeted to say its poll of 1000 women was NOT politically weighted so not comparable.

    Wassat mean? Is this not a bit strange.

    ReplyDelete
  12. It likely means there were "accurately weighted" and "skewed" weighting applied to various groups of respondents. This with be to accommodate the client who may wish to push a particular narrative. Given the majority of the media in Scotland are fundamental imperialists and existential unionists and the pollsters need their custom, they will adapt as far as reasonably practicable to accommodate their clients. A classic examples are YouGov / Murdoch press and ICM / Johnston press polls.

    Better Nation run the Survation poll in conjunction with Trinity Mirror (Record). Their analysis is balanced, and based on the full data sets.

    ReplyDelete
  13. Survation tweeted to say its poll of 1000 women was NOT politically weighted so not comparable.

    Hopefully they at least asked 2011 vote even if not weighted.

    How do you politically weight for one sex anyway? We don't know exactly how each sex voted in 2011 as I understand it? We know it was not massively dissimilar based at least on last minute polling intentions, but exactly?

    It is another caveat to watch. We only see weightings politically for the whole sample. Ideally, the weightings would be done based on sex and vote if we are to believe that women and men will really vote very differently next month (something I personally struggle to believe as both have the same information and arguments presented to them and normally vote quite similarly).

    ReplyDelete
  14. The daily record will have realised that a continuation of their normal monthly polls would start to show significant gains for Yes.That is the sole reason they have changed tack. To give their readers the full picture is a step too far.

    ReplyDelete
  15. surely all these changes are within margin of error? or does that only apply for no increases? :-)

    ReplyDelete
  16. Might just be my limited maths and understanding of polling but does this Survation poll for the record not put in doubt the figures they got for the Daily Mail?

    This one shows Women only to be 34Y/50N/16DK the Daily Mail Survation showed 37Y/50N/DK13

    If the Daily Record is correct on Women then the Daily Mail must be wrong on overall?

    You can't have an increase of 3% just on the male vote without a corresponding fall in No or a more substantial drop in DK's?

    ReplyDelete
  17. Did a straw poll amongst my approx 200 facebook friends, Do you believe the polls? An incredible 29 out of 32 folk who replied said they were a load of rubbish and 3 said don't know. Of these 8 are no voters, all 8 said they were rubbish.

    The people of Scotland believe nothing the media, UK Govt or polling companies are telling us.

    Yes voters tell me the no campaign are lying, no voters are telling me both campaigns are lying. Can't find anyone who thinks the no campaign are telling the truth.

    By now the Scots must be the least trusting of our leaders and media in the entire developed world. Way to go, London/BBC/MSM, proud of yourselves?

    ReplyDelete
  18. Interesting women VI from survation:

    43 SNP
    27 Labour
    15 Con
    8 Lib
    3 Green
    3 UKIP

    Women totally different from men politically, hence the men for Yes and women not.

    Really? Or are the fairer sex just more polite?

    ReplyDelete
  19. Blow for Salmond as 43% of women support SNP!

    With >46% backing pro-indy parties.

    ReplyDelete
  20. The chat is that turnout will be 80 or just above.

    http://www.bbc.co.uk/news/uk-scotland-scotland-politics-28748212

    ReplyDelete
  21. Survation's point about the results not being comparable is a bit of a red herring for the reason Scottish Skier gives above. In any case, political weighting doesn't usually make a huge difference to Survation's results. I'm more concerned about other weightings - yet again, respondents from the south of Scotland have had to be massively upweighted. Survation are really having a problem with that (and with 16-24 year olds), and it's introducing a huge element of randonmness into their results. In this poll they were also way, way short of the number of older respondents they needed, which goes some way towards explaining why - again - the No lead is slightly lower in the raw unweighted data than it is in the weighted results.

    ReplyDelete
  22. Anonymous said: "Given the majority of the media in Scotland are fundamental imperialists and existential unionists and the pollsters need their custom, they will adapt as far as reasonably practicable to accommodate their clients. A classic examples are YouGov / Murdoch press and ICM / Johnston press polls."

    You are utterly wrong about the Murdoch press, Salmond is just about the only politician in the UK Murdoch likes and so the Scottish Sun has pretended a studied neutrality. This is his usual tactic and portends the fact the Scottish Sun will come out for Yes provided (and this is crucial) Murdoch thinks they have a good chance of winning. He always hates backing a loser.

    He will judge this by the polls of course as any sane person would though admittedly much nearer the finishing line than now when they are likely to be approximately right if 2011 is anything to go by.

    Timing is everything here. If we were a week off the referendum right now many on this site would have to concede (barring a massive error of the kind that is so rare it's hardly worth bothering about) No would win. But we are not, we are t-35. At this stage in 2011 the polls were showing a dead heat or tiny lead for the SNP. In nine days time (from now) the SNP would then open a big lead.

    If this happens again Murdoch will of course come out for Yes, probably or a week before polling before the second debate with the intention of inflicting the most massive damage on No he possibly could.

    Welcome at that point to Rupert Murdoch's Scotland.

    ReplyDelete
  23. "If we were a week off the referendum right now many on this site would have to concede (barring a massive error of the kind that is so rare it's hardly worth bothering about) No would win"

    For the avoidance of doubt, that's complete rubbish. No would have to improve their position from where it is now before we could say that.

    ReplyDelete
  24. Correction: should have said: either a week before polling OR before the second debate

    ReplyDelete
  25. James are you seriously saying if it was a week before polling and not a single poll in the main campaign had shown any sort of Yes lead and all the polls were showing the same spread as now you would be confident of winning? Why the heck would you other than blind hope?

    You surely aren't peddling this 'all-the-polls-are-fixed stuff?

    ReplyDelete
  26. "You surely aren't peddling this 'all-the-polls-are-fixed stuff?"

    You wouldn't be trying to put words in my mouth, would you? I'd have thought the meaning of what I said was perfectly clear - unless No improve their position from where it is now, the result will be uncertain.

    ReplyDelete
  27. But now I see I may have phrased the question wrong? I did not mean concede victory, of course not. I just mean accept privately things looked like going the other way and no doubt fighting all the harder to avert it.

    ReplyDelete
  28. "I just mean accept privately things looked like going the other way"

    Nope. To repeat again, No would have to improve their position for that to be the case. You're underestimating the level of uncertainty in the current polling results.

    ReplyDelete
  29. Personally I wouldn't go so far to believe that the polls are fixed (though I have deep suspicions over Yougov). I do however believe that they are way out of kilter with the real situation on tne ground. So yes, whether they shift or not I would still be quietly confident of a YES win. We shall see.

    ReplyDelete
  30. The lastest TNS BMRB poll is unusual in that the swing towards no has happened entirely among women and non voters. I'm not saying they are nessecarily wrong but it does break the usual pattern.

    ReplyDelete
  31. Calum : It's the non-voters part of it that raises a doubt in my mind, because the upweighting of those respondents will have created a lot of phantom No voters.

    ReplyDelete
  32. James: Page 13 of the survation table breaks down the undecided female voters intentions (and the undecided voters are likely to split 50:50 not 40:60) Therefore I believe a more accurate reflection would be 42-58 when dk's are stripped out.

    ReplyDelete
  33. From the same survey

    "Only 1.5% are voting Yes because they want Alex Salmond to be the leader of the country, against 16% who are voting No because they do not. Over half of No voters (51%) say his prominent presence in the Yes campaign will make them more likely to vote No."

    So what single action could tip the balance for yes then? :-)

    Expect the SNP coup d'etat any day now, Sturgeon to be installed before the referendum, not afterwards

    :-)

    ReplyDelete
  34. James as in 2011 I find it hard to equate polls with what i am gettingo n the doorstep.
    Last night for example we had a massive 62% Yes only 22% in No .
    This is happenning throughout our constituency.
    Like 2011 I was being told neck and neck yet we got a landslide.
    Sometihng far wrong with these polls I just do not get it.

    ReplyDelete
  35. The polls are being rigged. If you got kellner drunk he'll admit it to you. Or threaten him with spending more time with his wife. Should do the trick.

    M Carney admitting he's making plans for a Yes vote, having previously said he'd make a currency union work. Wonder why if there's no chance of freedom beating slavery.

    The desperation from the Scotland haters infecting every corner of the media and internet. The last twitches of bigotry from BBBC Glasgow. All for one purpose and they've failed.

    ReplyDelete
  36. James: This poll is quite peculiar in a lot of ways, it's not necessarily wrong but it's something I'm wary of. Apart from the sudden change of opinion among women and non voters, the fact that the certain/very likely to vote numbers are pretty much unchanged despite the widening of the gap in the whole sample is quite odd.

    (Changes from last poll)
    Whole sample Yes: 32% (nc) No: 45% (+4)
    Certain/Likely Yes: 29% (-1) No: 36% (nc)

    ReplyDelete
  37. "James: Page 13 of the survation table breaks down the undecided female voters intentions (and the undecided voters are likely to split 50:50 not 40:60) Therefore I believe a more accurate reflection would be 42-58 when dk's are stripped out."

    Well, the Yes vote would probably be at least 41%. It's not possible to make an exact calculation because the figures for undecideds haven't been weighted for likelihood to vote.

    ReplyDelete
  38. Mick Pork I never denied Cameron and Murdoch were once upon a time in cahoots but you are miles out of date. Everything has changed.

    Murdoch is no longer supporting the main UK parties even if they do or don't want him to The best Cameron can hope for in 2015 is a muted endorsement as the best of a bad job

    But Murdoch likes Salmond and has said so, that is the one thing that has not changed.

    As I say he won't act unless the polls narrow, but then I am quite certain you will get the Sun endorsement.

    If your point is that Cameron and Milliband and Murdoch are somehow acting together just because of the photo of UK party leaders holding the Sun World cup thing you could hardly be more wrong.

    And what about Coulson? I can't see the connection there. I know he faces a perjury charge arising out of the Sheridan case but so what? Coulson even had to sue Murdoch's company to get his legal bills paid. He is no longer connected to him.





    ReplyDelete
  39. "But Murdoch likes Salmond and has said so, that is the one thing that has not changed. "

    How do you know that? I read the same article as you where Murdoch praised AS as the most able politician in the UK. I think that was 4 years ago?

    ReplyDelete
  40. The only thing that has changed is that Coulson and Brooks are no longer the main conduit to Murdoch.

    That means Cameron and Miliband now have to conduct their 'business' with Murdoch at arms length with intermediaries like for example Matthew Freud. Freud is married to Elisabeth Murdoch the second daughter of Murdoch. Freud is also a close friend of both Cameron and Osborne and a member of Cameron's infamous 'Chipping Norton Set'.

    Your other somewhat feeble protestations that Coulson is all in the past, hardly excuses Cameron hiring Coulson to be his own spindoctor when he was specifically warned by many (even by Clegg) not to do so. Coulson was working for Cameron only three years ago before he had to quit and he was at the very top of Cameron's chumocracy. As Cameron's Communications Chief Coulson was directly involved in every aspect of westminster government policy so as to best spin it to the papers, including the Murdoch press where he used to be former Editor. Which I doubt many will be gullible enough to think is a mere coincidence.

    2011 is hardly some long forgotten period. Though I know precisely why the pollsters and No supporters want to forget it after the SNP landslide.

    Nor does little Ed have to go far to get in touch with those closest to Murdoch considering that former Labour PM Tony Blair is about as close as it gets to Murdoch. Just how close you would have to ask Rupert's wife Wendi now that she is getting divorced from the "dirty digger" Murdoch. LOL

    Murdoch has said a lot of things in case you haven't been paying attention. Things like there was no phone hacking at News International.

    The point you appear to be deliberately missing is that Murdoch is quite clearly only trying to use the prospect of Independence to put the screws on Cameron and little Ed.

    As for the nonsense that endorsing the Sun in that shameless manner is of no consequence, how on earth do you think that was coordinated if there is no contact with the Murdoch press from little Ed or Cameron?

    The truth is that Labour and the tories were in as deep as it gets with Murdoch and still fear him.

    If Murdoch backs Independence because the polls narrow (like they eventually did in 2011) you yourself are conceding his primary motivation would be simply to back a winner. Until then the Murdoch press will continue to run scare stories direct from better together while occasionally putting pressure on Cameron and Miliband for his future expansion plans for BSkyB and News International.

    ReplyDelete
  41. We are not in fact in disagreement about much of this but I can't quite see the relevance of your points. Yes there may be conduits but I can assure you Blair is not one of them. When you say Blair is about 'as close to Murdoch as it gets' you may be joking or else you are completely misinformed . It has been reported many times Murdoch has not spoken to Blair since the divorce and has made it clear he has no interests in having any further dealings with him ever. Some conduit!

    Salmond and Murdoch have however spent time together, Murdoch tweeted Cameron should 'let Scotland go' in 2012, and called Salmond the UK's 'clearly most brilliant' politician. He also chortled at Salmond defeating Cameron in the general argument with Cameron. This is not disputed and there is no evidence of any kind Murdoch has changed his mind. If there is some BT bias in his newspapers occasionally, that would be entirely in keeping with keeping his surprise last minute endorsement up his sleeve.


    And of course I don't excuse Coulson or his hiring. I just point out the hiring and Coulson are utterly irrelevant to any discussion of Murdoch and Scotland now.

    And if Murdoch sees his chance and endorses independence and wins, do you really think in your wildest dreams he will not expect something in return?

    I don't for a minute think you are that naïve.

    ReplyDelete
  42. "When you say Blair is about 'as close to Murdoch as it gets' you may be joking"

    You clearly know very little indeed about any of this. The reports around Murdoch's wife and Blair are hardly a secret any more. That you didn't spot it was a joke (the LOL was your clue) may be related to the fact that you have tried to pronounce, supposedly in all seriousness...

    the Scottish Sun will come out for Yes provided (and this is crucial) Murdoch thinks they have a good chance of winning. He always hates backing a loser.


    Then try to claim, with no irony whatsoever...

    "And if Murdoch sees his chance and endorses independence and wins, do you really think in your wildest dreams he will not expect something in return?"

    You contradict yourself at every turn. You have been shown absolute proof that both Cameron and Miliband are more than willing to look like a pair of enormous twats only a couple of months ago by fully endorsing the Sun and Murdoch press with their THIS IS OUR ENGLAND front page PR for the Murdoch press. Yet according to you the burden of proof is somehow bizarrely not on them but on Salmond for a couple of lines Murdoch made years ago.

    At least try to hide your obvious bias.

    Your feigned ignorance over JUST how close to Murdoch's empire Coulson (still in prison) self-evidently made Cameron is also laughable. Coulson was Cameron's right hand man and spindoctor. That's one of Murdoch's key men and ex-Editors at the very heart of westminster and Cameron's government in case you still don't get it. Just because he was caught doesn't exculpate Cameron from his astonishing lack of judgement and obvious closeness to Murdoch. A closeness far greater than Salmond. Unless you really are deluded enough to think a line or two of praise is somehow equivalent to having one of Murdoch's most prominent EDITORS as a spindoctor while shmoozing Brooks to a frankly comical degree.

    The entire point of this is trust and I'm afraid Cameron's actions over Coulson means you would have to be a gigantic fool to trust him with Murdoch now. Nor do the contacts with Freud show anything other than an incompetent twit of a PM who still hasn't learned his lesson.

    ReplyDelete