Monday, August 11, 2014

Boost for Yes as the gap narrows by 2% among definite voters in new poll from No-friendly firm YouGov

As I pointed out in last night's post, YouGov are unusual in that they don't filter or weight their headline results by likelihood to vote.  However, the datasets for the new poll have been released, so we now know what the figures would have been if YouGov used the same method as Ipsos-Mori, and headlined the results for respondents who say they are absolutely certain to vote in the referendum.  The logic for doing so is that people typically overestimate their own likelihood to vote, and actual turnout figures tend to correlate quite closely with the number of people who tell pollsters they are absolutely certain to vote.  The percentage changes given below are from the last YouGov poll which was conducted in late June.

Should Scotland be an independent country?  (Definite voters only)

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

As always, it has to be borne in mind that YouGov are now by far the most No-friendly of the six BPC pollsters, almost certainly due to the artificial and highly secretive "Kellner Correction" which is used to suppress the reported Yes vote, so the size of the No lead has to be seen in that light.  

Among voters who say they have at least an 8 out of 10 chance of turning out to vote, the No lead is completely unchanged since the last YouGov poll at 61-39.  So there's really very little comfort here for the "No have been given a boost by the leaders' debate" narrative.  Admittedly there's no absolute proof that there hasn't been a post-debate bounce for No, because only half of the sample for this poll was interviewed after the debate.  But if by any chance there was a substantial bounce, that must mean that the pre-debate half of the sample was extremely favourable for Yes, possibly with the Yes vote approaching an all-time high for YouGov, because there's no other way that the overall figures would average out as showing a small swing to Yes among definite voters.  That in itself would be troubling for the No campaign, because if any post-debate bounce proves to be superficial and transitory, you'd expect the state of play to return to roughly where it was prior to Tuesday evening.

It may also be worth making the point that, due to YouGov's clear sympathies with the arguments of the No campaign, they'd have been likely to point it out if there were any significant differences between the two halves of the sample.

There's one uncanny similarity between the YouGov and Survation polls of recent days, which is that the No lead is actually slightly lower on the raw unweighted data than it is in the weighted figures.  That's reasonably unusual - generally the weighting lifts up the Yes vote, because groups that favour Yes (such as lower-income people) are under-represented in sampling and have to be upweighted.  This should perhaps set a few alarm bells ringing, because it might mean there is something strange about both polls that is suppressing the Yes vote.  In the case of Survation it was fairly obvious what was going on - there were implausibly huge swings to No among the small samples of young people and residents of the South of Scotland electoral region, who had been upweighted massively in the overall results.  But I'm struggling to spot such an obvious explanation in the YouGov datasets.

We've got used to the highly unusual disparity between different polling firms in this campaign, but we're also going to have to start facing up to the fact that this looks likely to prove to be a factor in the next Holyrood campaign as well - unless of course the firms that prove to be the most inaccurate in September subsequently put their house in order.  YouGov are currently showing a modest Labour lead for Holyrood (although that lead has actually narrowed since late June, which again flatly contradicts the "post-debate blow for Salmond" claim).  That contrasts with the other traditionally No-friendly pollster Ipsos-Mori, who are continuing to show a decent SNP lead.  The more Yes-friendly pollsters tend to show much bigger SNP leads.  So although it would be an over-simplification to say it's "YouGov versus the field" in terms of Holyrood polling, there's certainly an element of truth in that, and the explanation is most likely to be the Kellner Correction.  The fact that this disparity isn't just happening in referendum polling must surely place a big question mark over the credibility of YouGov's approach.

The most interesting of the supplementary questions in the poll asks whether there should be another independence referendum in the future if there is a No vote this year, and what the timescale for that should be.  A full 53% of respondents say there should be another referendum at some point in the next 30 years.  I must say that really surprises me - although I personally think people's tolerance for another referendum would be quite high in the long-run (after all we had a second devolution referendum just 15 years after the first one), I wouldn't expect them to realise that right now.  Most startlingly, 31% of No voters say there should be a second referendum.  What's going on here?  It's hard not to conclude that at least some of these people are soft No voters who are not at all sure they are doing the right thing, and would like the safety-net of thinking they might have a second bite of the cherry one day.  If so, there's every reason to think that many of them will be open to persuasion between now and September.

* * *

REQUIRED SWINGS

Swing required for 1 out of 6 pollsters to show Yes in the lead or level : 3.5%

Swing required for 2 out of 6 pollsters to show Yes in the lead or level : 4.5%

Swing required for 3 out of 6 pollsters to show Yes in the lead or level : 5.5%

Swing required for 4 out of 6 pollsters to show Yes in the lead or level : 6.5%

* * *

SCOT GOES POP POLL OF POLLS

This update of the Poll of Polls takes account of both the Survation and YouGov polls, and therefore the slight increase in the No lead is almost entirely caused by Survation - the YouGov figures make an absolutely negligible difference.

MEAN AVERAGE (excluding Don't Knows) :

Yes 42.8% (-0.7)
No 57.2% (+0.7)

MEAN AVERAGE (not excluding Don't Knows) :

Yes 36.5% (-0.5)
No 48.8% (+0.8)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 42.8% (-0.6)
No 57.2% (+0.6)

(The Poll of Polls is based on a rolling average of the most recent poll from each of the pollsters that have been active in the referendum campaign since September 2013, and that adhere to British Polling Council rules. At present, there are six - YouGov, TNS-BMRB, Survation, Panelbase, Ipsos-Mori and ICM. Whenever a new poll is published, it replaces the last poll from the same company in the sample. Changes in the Poll of Polls are generally glacial in nature due to the fact that only a small portion of the sample is updated each time.)

*  *  *

Those of you who read the comments section of this blog can't really have failed to notice that there has been an infestation of anonymous No-supporting trolls recently - some of them abusive, but none of them remotely interested in constructive debate.  A number of people have contacted me to suggest that I ban the trolls or mass-delete their posts.  I'm not going to do that.  One thing that makes the Yes campaign different (with a very few unfortunate exceptions like James Mackenzie) is that we believe in open debate, and we don't go around censoring our opponents in the way that Labour Hame, Vote No Borders or the Better Together Facebook page do.  I fully appreciate how irritating it is, though.  Heaven only knows where all these people have suddenly sprung from - they might be from the dark hordes at Political Betting, they might be "risk assessors" and risk assessor groupies who have followed me over from Twitter, or something more organised than that might be going on.

To answer a specific question that a couple of people have asked me, there's no need for anyone to post anonymously if they don't want to - simply select the "Name/URL" option when you comment.  If you don't have a website or profile of some description, just leave the URL section blank.

81 comments:

  1. 5 August:

    REQUIRED SWINGS

    Swing required for 1 out of 6 pollsters to show Yes in the lead or level : 3.0%

    Swing required for 2 out of 6 pollsters to show Yes in the lead or level : 3.5%

    Swing required for 3 out of 6 pollsters to show Yes in the lead or level : 4.5%

    Swing required for 4 out of 6 pollsters to show Yes in the lead or level : 5.5%

    ReplyDelete
  2. Kerching! At least 7p for you from McDougall Central for that hilarious choice of name.

    ReplyDelete
  3. You have a strangely calming influence on me James.

    The trolls will still be 'amonymous'.

    News from William Hill to cheer us up.
    http://www.ross-shirejournal.co.uk/Independence-Debate/Punters-backing-Scottish-independence-11082014.htm

    ReplyDelete
  4. I read that earlier, David. I wonder what odds the bookies would be giving minus the huge bets from No Thanks / London?

    ReplyDelete
  5. "Those of you who read the comments section of this blog can't really have failed to notice that there has been an infestation of anonymous No-supporting trolls recently"

    I've always found the "cybernat" thing hilarious from these people.

    When the worst, most prolific, most aggressive online presence has always been from the Tory Party.

    I do find the way the debate over Scotland (and therefore the UK's) future has been conducted pretty depressing all round.

    Patronising fearmongering from No, and I'm afraid not enough attention to nitty gritty practicalities from Yes.

    Hugh.

    ReplyDelete
  6. What are the 'nitty gritty practicalities'?

    Just curious.

    You might find some here:

    http://wingsoverscotland.com/weebluebook/

    ReplyDelete
  7. James, why not at least disable the ability to post as "Anonymous"? I can't think of anyone else on the Blogger platform (and there are a lot) who allows this.

    It would at least give us a chance of figuring out who was who when there are several anonymous posters in the conversation at once. Sometimes they're even Yessers, which is confusing.

    ReplyDelete
  8. Oh heck, sorry, software playing up. Can you delete the two duplicates?

    ReplyDelete
  9. Shouldn't this be classed as margin of error, as you cautioned with regards to the Survation poll?

    ReplyDelete
  10. James, our comedy BritNat troll is only highlighting that the swings required for a Yes Win are easily achievable. Of course that's only IF the polls are accurate and properly reflecting the situation on the ground. They certainly weren't only three years ago. While the absurd "Kellner Correction" and so many other methodological changes hardly point to the pollsters themselves being remotely confident they are accurately reflecting the real level of the vote.

    "I do find the way the debate over Scotland (and therefore the UK's) future has been conducted pretty depressing all round."

    Hugh, there is a a yawning chasm, light-years wide, between what the unionist media and westminster bubble sites and pundits are portraying as the debate, and what is actually happening on the ground in scotland.

    There have been thousands of town hall style meetings up and down scotland with a quite remarkable level of engagement. We've seen music concerts, artistic endeavours, plays, books, festivals and every type of community driven event embracing the Independence debate and getting extraoridnarily encouraging numbers of ordinary scots engaged in politics and the referendum itself.

    This is of course well away from the prying eyes and smears of a unionist dominated media who actively demonise Yes supporters on their front pages. With 37 national or daily papers in scotland, ALL of who oppose independence, is it any wonder that this grass roots movement is being totally ignored by the Westminster bubble?

    Meanwhile as another Iraq War escalates a perfect time then for some jawdropping 'comedy' from one of westminster's 'finest' MPs.

    Foreign Office Minister Mark Simmonds has just resigned over a matter of principle.

    So what is this noble principle that has driven a westmisnter politician away from a lucrative and plum foreign office job? The grotesque bombings in Gaza? Military Intervention in yet another Iraq War escalating daily with still no weapons of mass destruction found from the previous one?

    Nope.

    Westminster MP and Foreign Office Minister Mark Simmonds has resigned over the principle that he doesn't think MP's are getting enough expenses.

    He feels the £170,000 or so he got last year in expenses is simply not sufficient for him.

    I shit you not.

    Even Alistair Darling manages to get by on his massive expenses claims. Admittedly the fact that Darling is getting paid thousands by NHS privatising companies just to 'top him up' obviously helps.

    To be fair though westminster politicians are still waiting for their their paltry 11% pay rise to come though. This at a time when so many ordinary people can only laugh at such a pitifully small number.


    "Better Together" Indeed.

    ReplyDelete
  11. Do you believe that Labour is on 37% for Holyrood Constituence and 35% for the list vote. That just isn't credible considering Labour got 26% in the Euros and have hardly made friends during the campaign. The YouGov sample seem is far too pro-Labour

    ReplyDelete
  12. Another interesting examination of the latest poll. We near-innumerate types owe sites like this a great deal.
    Just one question: I think it was mentioned in the previous article on the Survation poll that these polls are over-representative on the opinions of non-Scots. Is that correct? [It might have been Mick Pork who raised this point - if not I apologise]. However I still don't know how a nonrepresentative high level on non-Scots can result in an accurate poll. Also, I can't remember other analysts mentioning this discrepancy.
    Sorry if this has all be covered before.

    ReplyDelete
  13. James,

    I am sorry you are suffering from an infestation of idiots, but I am afraid you do rather encourage them with posts like this. You know well enough that a margin of error movement in a subset that the pollster itself does not use for its headline numbers is irrelevant, and yet by devoting a main blogpost to such irrelevance you reinforce the impression that you either you don't know how to analyse polls (which obviously isn't true) or you don't really care about objective reporting of them. I can appreciate your desire to rally the troops, but reporting this poll as a boost for "yes" is simply absurd.

    Even on the point you have chosen - arising from the fact 90% of "yes" supporters claim they are certain to vote, versus 88% of "no" voters, your point is somewhat undermined by the fact that 5% of "no" voters say they are 90% certain to vote, compared to 3% of "yes" voters - i.e. exactly the same proportion of "yes" and "no" voters are at least 9/10 likely to vote. The distinction between a 9 and a 10 on that scale is so slight as to be almost meaningless. And that's before one takes into account the effect of rounding.

    In recent weeks you have variously celebrated or sought solace in numerous margin of error movements, sometimes even at a decimal place level. Your willingness to see hope in even the most grim poll, while admirable, is leading you into ever greater contortions and, regrettably, distortions. It provides a momentary boost for your comrades but serves them ill in the long run.

    A case in point is your increasingly strident attacks on YouGov. Only a few posts ago you referred to the Kellner correction as being "notorious". It's not. It is in fact quite the opposite of "notorious". It's a little known, carefully reasoned decision to apply an additional weighting filter to try to ensure each sample is as representative as possible of the population it seeks to sample.

    In this post you openly contemplate that there could be something supressing the "yes" vote, but have to concede you can't see what it might be. In fact remarkably little weighting was required on this poll, other than on social grade, and even that is unlikely to have had a very significant impact because "no" was 18pts ahead even amongs C2DEs.

    It is a pity, because you clearly are capable of understanding the polling and some of your analysis is both correct and much more detailed than on other sites with bigger reputations. For example, you are probably correct that the recent Survation poll had a sampling error amongst the lowest age group. I would expect the next Survation poll to revert to the mean (subject to events, naturally).

    Current polling is somewhere between poor and very poor for "yes"; poor in the sense that the momentum "yes" enjoyed last Autumn and early in 2014 has stalled, poor in the sense that time is running out and the breakthrough remains elusive, even amongst the most "yes" friendly pollsters, poor in the sense that the "no" majority seems quite firm. Of course, polls can be wrong but it is unlikely all of them can be THAT wrong.

    ReplyDelete
  14. Part 2

    Having started well, the "yes" campaign is also performing poorly. Salmond lost the first debate; not comprehensively, and perhaps only because he was expected to win comfortably (but that matters). The currency issue has been completely mishandled; it should have been easy to diffuse but Salmond chose instead to pour petrol on it, meeting genuine questions with belligerence. A change of tack, a more inclusive and mature approach, is desperately required. The "no" campaign is being let down by its army of online idiots (no, they are not being paid, they volunteer to be that stupid). The "yes" campaign is being let down by its leadership. It's also being a little let down on here, as much by your loyal commentariat as by you, as grievances are nurtured, suspicions harboured, uncomfortable facts glossed over and negatives portrayed as positives.

    All that said, good luck with this site. It is often very good, and a must read site in the referendum debate.

    ReplyDelete
  15. The most important thing for me is that I have been actively involved 'on-line' since the campaign was announced and always 'Favourite' a tweet in which people confirm that after some consideration they or perhaps their loved ones/friends have decided to vote Yes.

    I see this happening every day and anyone who tweets will also confirm.

    It was reported last week that a former No campaigner/Labour councillor was now campaigning for Yes, and just earlier today a newspaper article reported that two former No activists, had now joined the Yes campaign.

    ...and yet still the polling companies show a very slow move to Yes, with some like Ipsos or Yougove laughingly showing swings to No or even more laughingly no moves to Yes.

    If people who have openly campaigned for No are switching, just how many ordinary people are doing the same?

    We also saw the figures for the amount of Labour voters who now say they are backing Yes steadily rise, I think the last figures were that 28% of Labour voters now say they intend voting Yes! nearly a third! yet still the polls say there's not much movement for Yes?

    A complete joke and something that although expert insights and analysis help join the dots, it's clear to anyone who has followed the debate on twitter and on-line blogs that the yes vote is growing day by day, but for whatever reason the polling companies are not reflecting this reality.

    I Personally believe this is by design, but whatever you believe you can be assured that Yes is growing day by day.

    ReplyDelete
  16. Flockers. You overdid the condescension. Nice try though.

    ReplyDelete
  17. Final thought - I think you're wrong to read anything into the fact that 31% of "no" voters would like a second referendum. For one thing 87% of "no" voters either said they did want another vote (66%) or that they wanted another vote 20 or 30 years down the line (21%). Only 4% wanted another vote within the next ten years. So having been given a range of options that included some nice sensible midpoints, the vast majority of no voters wanted the matter settled for a generation.

    Moreover, "yes" voters were more likely to say there should not be a second referendum than "no" voters to say there should be one within the next ten years, and 34% of "yes" voters were happy to see the issue kicked into the long grass for the next ten years or more.

    The "no" voters are no softer than the "yes" voters on this emasure - all this shows is that there are a decent number of pragmatists out there who recognise that circumstances might change, their veiws might change and it is wise never to say never.

    ReplyDelete
  18. One thing that is very very clear, is the concerted effort to discredit James by the new posters.

    This is an openly Yes supporting blog, and it would come as no surprise to any fair minded person that the author of the blog would explore polling results, to see what good news The Yes campaign could glean from the numbers, yet a whole bunch of people some following the well worn path of 'I'm a Concerned yes voter...and'

    or 'I'm a very reasonable No voter and usually like this site...but'

    And the go on to insult James and people who back Yes on here, and insinuate James's evaluations are dishonest or misleading.

    We also have the claims that some commenters are 'conspiracy theorists'

    We also get the downright abusive, who make no effort to engage in any debate.

    So my question to all fair minded people is simple:

    Why do you find it so strange that a Yes blogger like James looks for positives in polling stats, for the Yes supporters/campaigners, who frequent Scot goes Pop?

    ReplyDelete
  19. "Why do you find it so strange that a Yes blogger like James looks for positives in polling stats, for the Yes supporters/campaigners, who frequent Scot goes Pop?"

    Because dismissing a 2% increase in the No lead as margin of error and claiming a 2% increase in the Yes lead as a boost for Yes is a bit disingenuous. Despite being an 8% boost for No, the previous Survation poll was dismissed as largely an illusion of margin of error:

    "To turn to the headline voting intention results, there are two specific reasons to be sceptical about the supposed swing to No (over and above the possibility that it's largely an illusion caused by the standard 3% margin of error)"

    ReplyDelete
  20. So I just went on the Wings twitter account.

    Third tweet in;

    Revelo1630

    "I used to be a firmly on the no side, now I can't imagine what I was thinking. Don't swallow media spin do your own research"

    Kinda proves my point doesn't it?

    .

    ReplyDelete
  21. I can assure you James has never said that a 2% increase for any side is margin of error, what he has repeatedly said about all polls is that there is a 3% margin of error in the tolerances and therefore any move within these can simply be a MOE.

    If you are needing to lie to discredit the man, what does it say for the character of the people on your side of the debate?

    Please stop lying!

    ReplyDelete

  22. That represents a swing of 7.5% Lab to SNP in the constituency vote and 12% Lab to SNP in the list vote since 2011 HR election. That doesn't seem credible considering the Scottish Parliament is popular and the FM and DFM have positive approvals

    ReplyDelete
  23. O/T Daily record online referendum poll 11/08/2014, currently 65% Yes, 32% No, 3% undecided. I know what poll I believe in and its not Ipsos/Mori or Yougov!!

    ReplyDelete
  24. The Labour and SNP VI in the latest YouGov show a swing of 7.5% Lab to SNP in the constituency vote and 12% Lab to SNP in the list vote since 2011 HR election. That doesn't seem credible considering the Scottish Parliament is popular and the FM and DFM have positive approvals

    ReplyDelete
  25. Strange how all the polls lately are mostly You Gov and Ipso Mori. Almost as if the trend is manipulated to reverse

    ReplyDelete
  26. Haartime: of the last 10 polls, there has only been one YouGov, and one MORI.

    ReplyDelete
  27. When will YouGov finally lift the very telling secrecy that surrounds their absurd "Kellner Correction"?

    When will the No campaign actually begin to debate ALL the subjects that scots want answered when it comes to who the scottish public trust with scotland's future? (as opposed to the BritNats demented fixation on one issue they wrongly think they can scare and bully scots into submission with)

    Comically patronising and inept declarations of just how badly the Yes campaign is doing while No desperately reheats the currency variation "too poor, too wee, and too stupid" for the umpteenth time, seem more than a trifle misplaced.

    The 2011 campaign was subject to the same unionist spin, bad polling and media onslaught. A landslide win for the SNP was the reward for a complacent scottish labour and westmisnter establishment.

    If they truly believe they are doing so well then why the endless spin and reliance on a rabid and distrusted press?

    This is why.

    The fundamentals of the entire scottish Independence Referendum.

    http://survation.com/as-the-gap-closes-between-the-yes-and-no-campaigns-who-do-scots-trust-with-scotlands-future/

    Trust.

    ReplyDelete
  28. Sorry about my comments the swing is SNP to Lab 7.5 and 12 since 2011. Suggests a very proLabour panel

    ReplyDelete
  29. I saw a comment UK poll report that an ICM poll is imminent.

    What the MSM in general are not picking up are debates held all over Scotland such as the one in Irvine tonight. I assume that this is/was a pro Labour area and its these voters the pollsters are underestimating IMO.

    ReplyDelete
  30. There was a UK-wide ICM poll released a few hours ago showing a big boost for Labour. Not sure about an independence question poll though.

    ReplyDelete
  31. Yougov have already explained what you describe as the Kellner correction (actually just a more refined and sophisticated model that takes into account how people said they voted at the time), on its website, in a blog by Peter Kellner. Despite your desperate hope to the contrary, there is no scandalous silence or even unwarranted opacity. It's all pretty open.

    The 'no' campaign has been fairly poor throughout, but they have a few key messages that look to be effective enough and now they have a theme, of Salmond's arrogance preventing him from acknowledging the need to address legitimate concerns. I expect the next debate will focus in part on Salmond's position on EU membership, which is similarly built on assertion rather than fact and carries major consequences if he is wrong. The currency issue really matters. I am very surprised at how illprepared Salmond has been on the topic.

    The only "too poor, too wee, too stupid" sloganising is coming from you. That's not the view of the 'no' campaign. But there are very significant questions about the costs and consequences of independence that have to be answered. So far, Salmond has not been up to the task. For your sake that needs to change.

    ReplyDelete
  32. internet polls ffs stuart - vote early, delete cookies, vote often :-)

    james : what about the kelly correction, where you switch your poll of polls methodology from showing a lead to showing a "swing needed", thereby halving the numbers quoted into more agreeable single figure territory? :-)

    ReplyDelete
  33. Oh dear, yet another amusing Kellner groupie jumps on to the site praising their 'polling god' to the skies talking absolute shit.

    Why are YouGov so obsessively secretive about the most important detail of their referendum polling?

    http://scotgoespop.blogspot.co.uk/2014/06/why-are-yougov-so-obsessively-secretive.html

    "This is all guesswork, of course, because for some reason YouGov only ever reveal the voting intentions of SNP voters as a whole, after weighting has been applied - there's never a breakdown for the two distinct groups they use for weighting. It's very hard to understand why they would keep that information a secret, unless it shows such an improbable disparity between the two groups that eyebrows would be raised about the wisdom of the methodology."

    This would also be the comical Kellner who got it so very wrong on the night of the EU elections to the mirth of many.

    SNP win the European elections - and the big losers are blundering Peter Kellner, and the BBC results programme

    And yet just a couple of hours ago, YouGov's Peter Kellner smugly informed the nation that Labour were heading for victory in Scotland. Viewers of the BBC results show will have been forgiven for taking that wildly implausible claim seriously, because it was shorn of all context - it was the first time any Scottish results had been mentioned (other than a very brief comment from Kellner himself about Aberdeen). Now, of course, the trends of early results can sometimes be unrepresentative, and if that was the case he could be forgiven for leading people astray. But that categorically isn't what happened. He immediately explained the percentage changes that had led him to conclude that Labour were winning, and they made no logical sense whatever - they in fact suggested that the SNP were heading for victory by about 2% or 3%. It seems he hadn't even bothered to check the baseline figures from 2009 before coming into the studio. The BBC's political editor Nick Robinson then had an immediate opportunity to correct an obviously flawed piece of arithmetic, but clearly he didn't know the baseline figures either, and instead eagerly seized on the latest concocted "blow for Alex Salmond" narrative."

    Not forgetting what Kellner himself said over three years ago just to prove he isn't biased in the slightest.

    YouGov's Great Wall of Secrecy, Bluster and Evasion

    In spite of the technical detail Kellner goes into in support of his attack on fellow pollsters, my view is that this is a monumental bluff on his part. For more than three years, he has been telling anyone who will listen that it is literally impossible for Yes to win the referendum.

    http://newsnetscotland.com/index.php/scottish-opinion/9406-yougovs-great-wall-of-secrecy-bluster-and-evasion

    Sounds like a very trustworth and open minded chap, don't you think?

    LOL


    There's also no question at all about YouGov continual methodological changes mere weeks away from the actual referendum indicating that they are very far from confident in their own "sophisticated" analysis even this late in the day.

    ReplyDelete
  34. "I expect the next debate will focus in part on Salmond's position on EU membership"

    Do you indeed? I expect most of the scottish public are well aware than an IN/OUT EU referendum is on the cards for westminster. They'll also know that the scottish public would get dragged out of the EU whether they like it or not by rUK with UKIP rampant and winning EU elections.

    I also expect Iraq to feature. It's somewhat implausible it won't what with UK special forces boots on the ground and ever increasing bombing and military intervention.

    What do you think Alistair Darling's response should be when he's asked why he supported the Invasion of Iraq and scaremongered about Weapons of Mass Destruction? Hmmm? Furious blinking for a couple of minutes or incoherent stuttering and finger pointing for five minutes or so?

    If that doesn't make scots trust westmisnter politicians like Darling then what will? Being paid thousands by by NHS privatising companies? His expenses claims and house flipping?

    LOL

    ReplyDelete
  35. "Margin of error stuff" yes if it's within 3% it might be margin of error stuff.

    Again if you were not trying so desperately hard to discredit James you would concede that he was remarking on the fact that any respectable commenter would not headline with 'Blow for Salmond' when the figures fell within what they know to be MOE.

    Oh just had another tweet, with a tweeters friend saying they have managed to convert their previously staunch No voting wife to a definite Yes.

    It just keeps happening again and again, but the polls aren't picking this up!

    Oh just in, 79% of all bets in William Hill Bookmakers in Scotland are backing Yes, with 100% in Dundee, backing Yes!

    100% in Dundee? I'm so proud of my Yes supporting city!

    ReplyDelete
  36. "Again if you were not trying so desperately hard to discredit James you would concede that he was remarking on the fact that any respectable commenter would not headline with 'Blow for Salmond' when the figures fell within what they know to be MOE."

    So why is a 2% increase in Yes a "Boost for Yes", whereas a 2% increase for No is "margin of error stuff"?

    ReplyDelete
  37. I am lead to believe that the Kellner Corrected correction, splits the SNP vote between those who have voted SNP before the last election and those who voted Labour but switched at the last election.

    He then mark's up the switchers in his polls as more likely to be No voters who are simply voting SNP because they do a good job in the Scottish Parliament.

    Fair enough, you might say, but there's a fundamental flaw in this and it will hugely down play the Yes vote.

    If he does this to the SNP voters but doesn't do this to the Labour Lib dem and Tory voters, he misses out the large % of those voters who are Tory, Labour (LFI) or Lib (LIBSFY)

    Labour have 28% yes voters! big % for LD's reasonable % Tory, but the Yes doesn't get these voters marked up in the Polls!

    It doesn't work both ways because a SNP voter isn't given an opportunity to vote Labour/Lib/Tory, only No,

    But all of the three unionist part voters can indeed vote Yes, and many have said they will.

    So why not say that 25% SNP voters will vote No, but 40% Unionist Party voters will vote Yes, so when you look at the party preferences you mark up the Yes vote By 15%?

    Might seem very simplistic and probably a silly idea, but at least I'm not being secretive about how I came to this figure, and it would give a polling result that would seem more like the feeling on the ground than YouGov/ Mori give us.





    ReplyDelete
  38. "So why is a 2% increase in Yes a "Boost for Yes", whereas a 2% increase for No is "margin of error stuff"?

    Because the point James was making, was related to the fact that the 2% was not a Massive blow to Salmond!

    Of course a 2% increase in your vote is welcome from both sides, but when the polling company reports his findings with a headline screaming that it's a massive blow, then his credibility and neutrality must surely be brought into question.

    Most general public don't understand the MOE but a polling company fully understands it, so anything under 3% should be treated with some caution until it has been repeated in a following poll (the trend) so screaming that a 2% increase for No is a blow is simply propaganda for the BT campaign, something any credible polling company would be careful to avoid.

    Then we see the data fields and it gets even more strange, as YouGov say they have introduced two new methodological changes.

    A poll can only really be compared to another poll that uses the same calculations, and saying that the changes are Yes friendly doesn't cut the mustard after you have been screaming that your poll is a Blow for Salmond!

    Alex Salmond is not the leader of the Yes campaign, but the BT campaign have worked very hard to put it into peoples heads that he is, the YouGov polling headlines supported this BT campaigning strategy, with that headline and therefore they have lost any claim to impartiality or balance.

    ReplyDelete
  39. Patrick mentioned something I've brought up in previous threads. The disconnect between what the polls (and the bookies' odds) are saying and what is blatantly happening right in front of our eyes is grotesque.

    A couple of years ago I thought to myself, better get used to the idea of seeing window posters saying "No", because it's going to happen. Er, right, where? Meanwhile places like Bruntsfield and Morningside seem to be wallpapered with Yes posters. (And bear in mind people have to source these actively. The window-poster fairies aren't shoving them through their letterboxes.)

    By any metric you choose other than those I mentioned above, Yes is rampant. Window posters, car stickers, social media metrics, sheer numbers of activists, canvassing returns, attendance at meetings, and yes, online polls. (Anyone can clear cookies and vote early and often. No don't seem to be able to arrange this though.) It's beyond anything I ever experienced. Not just political activism either, but artistic - the Yestival, the Fringe, Scotland Yet, various concerts and performances all over the country. And look at the amount of money that has been raised for various Yes endeavours by crowdfunding. Half a million? Compared to tuppence ha'penny for No.

    More importantly, the Yes narrative isn't one of a static level of support talking to itself, but one of constant conversions coming over. Very very few people ever seen to declare a move to No, while people are declaring for Yes every day. Even previous No activists and Labour politicians are moving to Yes. And Yet the polls seem static or even creeping towards No.

    One or the other has to be wrong. The pollsters seem to be saying to us, "Who are you going to believe? Me or your lying eyes?" Hence the way the polls are being scrutinised.

    I really, really don't get it. On one hand, I'm not a fan of conspiracy theories. On the other hand, spy stories aren't entirely based on fantasy, and this is the British State fighting for its very life. Would it hesitate to throw a few million quid at the bookies to skew the odds, if it thought that would help? And if it would do that, is it inconceivable that some way might be found to keep the polls telling the same story?

    It's not as if this is a small point. The polls (and to a lesser extent the odds) are what allow the media - and in particular the BBC and STV - to maintain an anti-Yes narrative with all the "blow for Salmond" and so on. If Yes was ahead, the TV coverage which is so damaging would have to be different.

    So, I wonders. Yessss, I wondersssss.

    ReplyDelete
  40. So Scotcen are saying that their attitudes survey is showing that the number of Scots who say they are Scottish not British is 23%.

    http://www.scotcen.org.uk/news-media/press-releases/2014/august/referendum-campaign-divides-public/

    What was it in the census? 62%? People change their national identities just like that? Really? This strikes me as insane. Would we believe it if they told us 40% of Americans now feel Canadian?

    I lived in the States for several years, never did I feel any less Scottish. My grandparents lived abroad for about 2 decades, they still felt Scottish. How can nearly 40% of Scots suddenly stop feeling Scottish? It is utterly implausable.

    I'm now seriously worried that we are getting set up for some kind of fraud no vote.

    ReplyDelete
  41. Anonymous, If the papers say it is a "blow for Salmond" and then James shows that the papers are talking rubbish then that is a "boost for yes".

    Personally I love it when James points out the polling flaws. It gives me a wee boost. It reminds me to pay less attention to the polls.

    ReplyDelete
  42. Rolfe, I agree with everything you said, it is my own experience.

    Another reason for keeping up the appearance of a no lead would be to discourage Labour politicians from declaring for yes as they face being ostracised or deselected if it is a no vote.

    ReplyDelete
  43. Without the massive single bets from the anonymous english punters the bookies odds would be very different. The overwhelming majority of real people placing real bets are on Yes to win.

    If No was such a certainty then the bookies would be flooded with people taking advantage of the free money on offer. The question is why not and the answer is obvious.

    ReplyDelete
  44. I don't remember the 1979 campaign very well. It was the end of a long hard winter, and I had my head down, stuck into my PhD research. I wasn't in Scotland in 1997, I was freaking out in isolation in Sussex, so I can't say what the temperature was then.

    People who were awake and paying attention at these times though are saying that this is something else. The visible level of support for Yes far outstrips anything seen at the time of either of the two referendums on devolution.

    There are differences of course. In the two earlier votes, Labour and the LibDems were supporting Yes, at least to some extent. And the media weren't going with the wall-to-wall "you're doomed" narrative. And independence is a bigger deal all round.

    On the other hand, there was no internet in 1979 and it wasn't much of an influence in 1997 because it wasn't really interactive. Now, look at the online advantage Yes has, that it didn't have before.

    Nevertheless, we WON those two previous referendums. On a far far more subdued public show of support, and without any online activity.

    No wonder people are picking the polls apart and wondering where the spindizzy is.

    ReplyDelete
  45. "If No was such a certainty then the bookies would be flooded with people taking advantage of the free money on offer. The question is why not and the answer is obvious."

    Good point, never thought of that.

    ReplyDelete
  46. A lot of people in England do seem to think it's free money. Not in Scotland though!

    ReplyDelete
  47. Was a variant of the..."highly secretive "Kellner Correction"... used when YouGov made a cock up for the 2011 election? SEE PIC:

    https://twitter.com/JamesCo77979225/status/498801807821660161/photo/1



    ReplyDelete
  48. "Most general public don't understand the MOE but a polling company fully understands it, so anything under 3% should be treated with some caution until it has been repeated in a following poll (the trend) so screaming that a 2% increase for No is a blow is simply propaganda for the BT campaign, something any credible polling company would be careful to avoid."

    The same could be said about suggesting a 2% increase for Yes is a "boost for yes", that it is simply propaganda for the Yes side.

    James can't criticise people for suggesting a 2% poll change is good or bad for a particular side, when he does exactly the same in a post a few months later.

    ReplyDelete
  49. You know, I don't think some people here fully appreciate the gentle humour in many of James's headlines, where he mocks the mainstream media headlines by doing the same thing himself.

    ReplyDelete
  50. I think there has been more than one. I remember some discussion of this apparent anomaly some time ago. On the other hand, newspaper gets its facts wrong, don't hold the front page!

    ReplyDelete
  51. William Hill bets: Are they falsely trying to get us to think the support for yes is higher in Scotland than it actually is?

    Or are did they lie originally that the £200k bet came from Glasgow?

    Or maybe they just accidentally told us that the original bet was London?

    Could someone make a mistake like that?

    ReplyDelete
  52. Rolfe, I've done googling and can't find any other mention of a bet of £200,000.

    Also if there was a £200k bet in Glasgow then surely the Scottish yes percentage would be less than 79% and No would have to be much higher than 21%?

    I have a feeling the original bet was London not Glasgow. If it existed at all.

    ReplyDelete
  53. Sorry Rolfe, I'm being tin hat here, I know.

    OK I found another £200k in London just the other day.

    But doesn't explain why they failed to mention the Glasgow one.

    William Hill produced a local breakdown, it says bets in Glasgow were 89% yes. Surely if there had been a £200k bet in Glasgow the yes percent would be less than 89%? They would have to have taken, like £1.9m bets on yes and no other bets on no to reach that percentage.

    I don't buy it, I'm convinced they're hyping up the yes bets in Scotland for some reason.

    ReplyDelete
  54. If they are pretending that a £200,000 bet didn't happen they must be doing it for a reason? Or maybe they just forgot?

    ReplyDelete
  55. Flockers - I haven't got the time or energy to deal with all your points, but I'll just pick out one which I found particularly breathtaking, about the Kellner Correction...

    "Despite your desperate hope to the contrary, there is no scandalous silence or even unwarranted opacity. It's all pretty open."

    So in that case you can tell me how much it's increasing the No lead by. Oh wait, you can't, can you? Why not?

    Exactly.

    ReplyDelete
  56. Rolfe : As far as I can see, if I disabled anonymous commenting it would automatically disable "Name/URL" commenting as well. So in order to comment people would need either a Google account or an OpenID, which is an awful lot of hassle.

    ReplyDelete
  57. Anon at 7.39 : This poll is the first YouGov poll to be weighted by country of birth, so Scottish-born people are no longer under-represented in the weighted results. Survation haven't introduced country of birth weighting yet.

    ReplyDelete
  58. JK - if you genuinely found my comment breathtaking you must be asthmatic. Let me rephrase it. Peter Kellner, the president of Yougov, has explained in a blogpost on a freely available website that Yougov takes care to ensure that their sample reflects historic voting behaviour, specifically in the context of the SNP surge in 2011. Peter volunteered this information and in a long and thoughtful piece of analysis, which you are free to disagree with if you wish, he explained why another pollster taking a different approach might be oversampling a group of 2011 SNP voters that is not representative of the whole. All very open.

    I will address your question directly because I have seen your twitter wars before and don't want to get into a tedious "answer the question" quagmire. No, I cannot tell you, because Yougov have not published, what impact their sample refinement has on the overall figures. But that, I submit, is not unwarranted opacity. Yougov do publish voting intentions broken down by 2011 Holyrood vote, so you can see for yourself how much less "yes" inclined the 2011 SNP voters in Yougov's sample are than in Survation's (and indeed other pollsters'). The answer is that Yougov's sample of 2011 Holyrood SNP voters are notably less "yes" inclined than some other pollsters, but not to a degree that would fundamentally alter the overall polling. And you probably could quite easily calculate what impact applying a different "yes/no" split to that group would have. So it's hardly fair to describe Yougov's approach as being "unwarranted opacity" (my words, which you found breathtaking).

    The source of your complaint seems to rest solely on the fact Yougov do not publish further data about how an unrepresentative sample of 2011 SNP voters respond to the independence question. You can legitimately ask for that info, but Yougov have no obligation to share it, nor can I see an obvious reason why they would. Yougov believe - and again I stress they volunteered this information in a public blogpost specifically to help people understand the volatility in polling - that refining the 2011 SNP voters to reflect past voting behaviour is necessary to produce a representative survey. Why publish the results on a different basis, that you believe to be wrong, just to satisfy the whims of random members of the public?

    Look how many tin hatters you're now attracting - your comment threads routinely contain accusations of pollsters deliberately misrepresenting the true picture, being part of the unionist campaign, or even Tory stooges. These people are feeding of the conspiratorial tone you are setting.

    In this case it simply isn't warranted.

    ReplyDelete
  59. In terms of criticising polls...

    I recall discussing the problems with 2010 weighting on here due to inherent flaws. Subsequently, Survation stopped using that leaving only Yougov (Labour 2010 + SNP 2011 group requires 2010 weighting).

    There was much discussion on the sense of country of birth weighting. Subsequently, Panelbase and Yougov adopted this.

    So, two major criticisms discussed on this site are addressed by pollsters yet making such criticisms is wrong?

    Yougov have been at odds with all other polls since 2007. They just, by chance, started to match again due to the large gap that emerged 12/13 and are now becoming an outlier as all the others close.

    Wondering why they show this pattern at odds with all the others is perfectly legitimate.

    ReplyDelete
  60. I'm Anon 7:39. Thanks for your reply. All this weighting or non-weighting just clouds the water even further. Polls cannot be trusted. If I'm asked in the future I will make it my business to lie to any poll organisation.

    Flockers. Once again you undermine your comment by being blatantly condescending. It only wastes our time and your own.

    ReplyDelete
  61. SSAS coming out.

    41% (+10) support for Holyrood to make all decisions for Scotland, i.e. full indy.

    https://pbs.twimg.com/media/Bu06GREIgAA97md.jpg

    Indy (41%) + devo max (29%) at 70% combined I believe.

    Support for Westminster control at various levels down 10%.

    ReplyDelete
  62. The 28% for Status quo (22%) or end of Holyrood (6%) will be those saying they are definitely voting No and won't change their mind / totally against indy (e.g. TNS and ICM).

    ReplyDelete
  63. Amazing that I am now a tin hat wearing mentalcase for pointing out that the opinion pollsters are all biased and do the job they are paid to do.

    Explain why there used to be polls showing majority support for independence before the referendum campaign but not now.

    Explain the massive movements from yes to no that have never been witnessed by actual living breathing humans. Nobody goes from independent thinking to wanting slavery. Not now and not ever.

    Explain how kellner being a rabid british nationalist married to a labour peeress and unelectable euro-commissioner doesn't mean he has any bias at all. The death to Scotland campaign admit that they will do or say anything to get a no vote, no matter the consequences

    In short you are either a paid shill or a moron. Feel free to kill yourself.

    ReplyDelete
  64. As far as I can see, the polling companies have no real incentive to predict correctly.

    Their main political poll customers are the MSM and the Westminster parties, who are vigorously pushing the BT agenda.

    If they get the referendum numbers wrong they can brush it off just like they did in 2011. They will still be in business.Westminster and rUK parties will still use them.










    ReplyDelete
  65. Anonymous at 11.51, I didn't use the phrase "mentalcase", but given that in one post you managed to accuse all pollsters of being biased and delivering results to order, describe the Scottish people as being enslaved, describe the Scottish-led "no" campaign as the "death to Scotland campaign" (slogan: we've been trying for 300 years but we really mean it this time?), declare that nobody has ever moved from "yes" to "no" and invite me to kill myself for no other reason than daring to suggest that a pollster has been quite open about its methodology, you're certainly not helping yourself.

    bjsalba - doubtless the pollsters would survive calling the referendum wrong but that doesn't mean they have no incentive to call it correctly. They also want to sell their services to commercial customers for whom accuracy (rather than narrative) is very important - their reputation is therefore vital to them. Also, if the pollsters were trying to fix the result you wouldn't expect them to be producing polls that would encourage such complacency as we are now seeing from parts of the no campaign. It just doesn't stack up.

    ReplyDelete
  66. RIC mass canvass results from 6th August:

    42% Yes, 28% No, 30% DK
    Headline 60% Yes, 40% No
    5100 people canvassed

    Again these are from heavily working class areas of Scotland which I believe the polls are not reaching.

    ReplyDelete
  67. SayNoToSexistNoTrollsAugust 12, 2014 at 12:45 PM

    Does anyone have an opinion on the Scottish attitudes survey saying that only 23% of Scots now identify themselves as Scottish not British, down from 62% in the census?

    Scottish_Skier?

    ReplyDelete
  68. I've just realised your post is mince. Sorry for responding sensibly.

    See forced choice ID. Next month we need to pick between Scottish and British.

    The SSAS gave this to that forced choice:

    65% Scottish
    23% British

    ReplyDelete
  69. Flockers : Can you honestly, truly not see the contradiction between claiming that YouGov have been "pretty open" about the Kellner Correction, and then launching into a long and astonishingly condescending explanation of why they have every right to be secretive about it, and why we should just be grateful for the scraps of information they so generously choose to give us?

    By the way, that explanation was uncannily similar to the excuse Laurence Janta-Lipinski tried on me when I asked him for more information. Have we ever seen the two of you in the same room?

    ReplyDelete
  70. The Radical Independence Mass Canvass on 6th August produced the following results:- 587 canvassers 42 areas 5089 respondents Yes 42% No 28% DK 30% A previous Mass Canvass on 22nd June produced the following results:- 978 canvassers 46 areas 8317 respondents Yes 40% No 30% DK 30% so overall that's 88 areas 13406 respondents Yes 41% No 29% DK 30%

    ReplyDelete
  71. SayNoToSexistNoTrollsAugust 12, 2014 at 2:21 PM

    Scottish Skier said "I've just realised your post is mince"

    The 23% Scottish not British came from the BBC:

    "When presented with a range of options ranging from "Scottish, not British" through to "British, not Scottish", the proportion who say they were "Scottish, not British" was 23%, the lowest it has been at any time since 1999."

    http://www.bbc.co.uk/news/uk-scotland-scotland-politics-28740329

    ReplyDelete
  72. Can anyone help with this? I just saw a FB link to a Herald story with the encouraging-sounding opening line "The No campaign's lead has reached a record low among people certain to vote in the independence referendum, a new poll suggests." However, the link was dead, so I Googled the phrase to see if I could locate the article. I found a Daily Record story, posted two hours ago, with the exact same opening line - but again, the link was dead!

    Can anyone shed any light on what's happened to these stories, and what poll they're referring to?

    ReplyDelete
  73. Keaton : I suppose in theory it might be an accidental early release of an embargoed poll. If it's TNS, that would imply a No lead among definite voters of less than 9% - but because TNS are so slow, the fieldwork would have been done before the debate.

    Alternatively, it may have just been an old article referring to an old poll.

    ReplyDelete
  74. I've seen the link to the poll too and saved a screen cap.

    38% Yes, up 6% on ~a month.

    No up too (46%), but it's getting hellish volatile in TNS, with 41% last time, 46% prior to that, 42%...

    Yes in contrast is apparently on up and steady.

    Volatility suggests weakness - people saying No, then DK, then No...

    Anyway, those headline figures would yield:

    45.2(+1.7)% Yes
    54.8(-1.7)% No

    Ex DKs and would be something of a milestone, putting Yes in TNS - not the most friendly pollster for Yes - over 45%.

    All a bit mysterious.

    ReplyDelete
  75. SS : If it is TNS, my guess is that the 38/46 split is for definite voters only.

    I think there was a Survation poll a few months ago that had headline numbers of 38/46, so that's the only other explanation I can think of.

    ReplyDelete
  76. James,

    For the avoidance of (very tedious) doubt, I am not Laurence Janta-Lipinski nor have I ever met him or indeed (to the best of my knowledge) any employee, director or owner of any pollster.

    Of course I cannot see "the contradiction between claiming that YouGov have been "pretty open" about the Kellner Correction, and then ...[explaining].. why why they have every right to be secretive about it and we should just be grateful for the scraps of information they so generously choose to give us." There is no contradiction, because that is not what I said.

    I didn't say they have "a right to be secretive", nor did I say you should be "grateful for scraps". There is a world of grey between absolute transparency and absolute secrecy. All I have done is point out that your accusations against yougov are unfair, using evidence to back up my position that they have been fairly open (including the fact they voluntarily and publicly disclosed the methodological differences and that you can see from their own published data how their cohort of 2011 Holyrood SNP voters behaves compared to the equivalent cohort polled by others). The fact they they have chosen not to share even more information does not automatically make them secretive, or their behavious suspicious. They've just drawn a line at a reasonable place - and I have suggested a good reason why they might choose to do so. They might choose to draw the line elsewhere in the future; that's a commercial decision for them.

    Incidentally it's a little disappointing you took issue with my post (particularly in such hyperbolic terms) while allowing the anonymous post suggesting I kill myself stand unchecked. It's your site, of course.

    ReplyDelete
  77. "including the fact they voluntarily and publicly disclosed the methodological differences and that you can see from their own published data how their cohort of 2011 Holyrood SNP voters behaves compared to the equivalent cohort polled by others"

    No, you can't. That is simply untrue.

    ReplyDelete
  78. "Incidentally it's a little disappointing you took issue with my post (particularly in such hyperbolic terms) while allowing the anonymous post suggesting I kill myself stand unchecked. It's your site, of course."

    You're damn right it is.

    ReplyDelete
  79. What I screen grabbed from google:

    Latest poll: 38% Yes, 46% No, 16% Undecided | Herald ...

    www.hearldscotland.com/...

    3 hours ago - In all the TNS pre-referendum polling, one consistent theme is that the...

    ReplyDelete
  80. On it's way to you by e-mail.

    ReplyDelete