We had a glut of seven point leads at the start of the week – from Ashcroft and Populus on Monday and YouGov on Tuesday. At the end of the week things look like they are back to normal again – we won’t get Ashcroft till Monday, but Populus’s second poll of the week has more typical figures of CON 34%, LAB 36%, LDEM 8%, UKIP 12% (tabs here) and YouGov had a Labour lead of four on Wednesday and three this morning – CON 34%, LAB 37%, LDEM 8%, UKIP 12% (tabs here).

There was also a new Survation Scottish poll for the Daily Record. Like the recent TNS poll, the main movement was actually from don’t knows towards the two campaigns, with the lead once don’t knows were removed remaining the same. Topline figures are YES 41%(+2), NO 46%(+2), without don’t knows that works out at Yes 47%(nc), No 53%(nc) (Tabs here.)


322 Responses to “New YouGov, Populus and Survation polls”

1 2 3 7
  1. This next 9 months are going to be interesting politically; a very close referendum followed by a completely unpredictable general election.

  2. First time I have been first, by the way.

  3. I’m not sure that the referendum will be that close.

  4. @Muddy Waters (FPT)

    I don’t think the jury is still out on this. 83% of Scots voters voted for something other than Conservative in 2010, and yet they got a (mostly) Conservative government.

    In some cases, the default vote for some North of the border is to vote Labour to keep out the Conservatives. In the case of Independence, that dynamic no longer exists, and people aren’t inclined to vote according to a (largely English) two party system.

    I actually believe that the Conservatives will do better in Scotland if severed from Westminster. All the ‘Eton posh boy’ stuff will be redundant, and Labour will have to bang some drum against an SNP that has gained its core aim.

    Lab and SNP might easily get into a “we’re more left-wing than the other lot”, and the Conservatives might pick up some votes here and there as a result. There should be a lot less ‘them and us’. It’s a lot harder to blame people on your doorstep than if 500 miles away.

    (all imho of course)

  5. This has been doing the rounds a bit.

    https://pbs.twimg.com/media/BsQXLzQCAAE-8w8.png

  6. I understand the reason for weighting, but the raw data on Populus and Yougov both show substantially larger Labour leads than the weighted data. The Yougov poll has a 7.9% differential in the unweighted data. This has been a feature of almost all data sets for the last couple of years. Although you might assume that there will be reversion to 2010 voting patterns, this assumption could be wrong, meaning that weighting is understating Labour VI and overstating Conservative VI. MIght the election not be as closely fought as anticipated.

  7. So that’s the way Govt’s go these days….death by a thousand tweets!

  8. I was referring to Mr Nameless’ post of course.

  9. Statgeek. 35% of Scots voted for the coalition parties in 2010, 42% Labour, and 19% SNP. Under FPTP the majority have not voted for the Westminster Government of the day since the 1950s, when it was Conservative, in England and Scotland. Those who perpetuate the myth that the Tories do not have significant support in Scotland need to look at the last test of National opinion in the Euro elections-19% voted Tory. I would expect the Tories to win at least 3 extra seats in Scotland at the GE in 2015, all from the LDs.

  10. RogerH

    You seem to be out of line with the general belief on this site which I have been persuaded by. What are your reasons for thinking it wont be that close?

  11. Yay, Portsmouth!

    It’ll be a shame if IDS goes, one of the few honest politicians out there.

  12. Yes, old IDS will be missed… just as much as he was last time.

  13. Statgeek: I don’t think the jury is still out on this”

    Sorry, crossed wires maybe. My “jury out” remark was simply about which of the polling companies has the right size/weighting for the 2010-Labour-to-2011-SNP sub-group – a question which seems to matter, since, according to the YouGov data at least, their indyref VI apparently differs from those who voted SNP in both 2010 and 2011.

    (As it happens, I pretty much agree with your analysis that the Labour vote in Scotland has reflected a broad anti-Conservative coalition which would be likely to break up post-indy.)

  14. “Yes, old IDS will be missed”

    I’d have thought people would have improved their aim by now.

  15. Steve2

    I’m assuming your post about IDS was being rather ironical. IDS has a rather tenuous relationship with honesty which is rather well evidenced across his parliamentary career.. and prior to that as well.

  16. Reg, there’s quite of few of us on here who have predicted a 60/40 vote since day 1.

    Polls are reacting strongly to mere political comments which suggests that polled opinions are highly emotive.

    Come ref day, once all the partisan and divisive diatribe has disappeared, I believe cooler heads will prevail.

  17. One interesting contrast between the last YouGov poll:

    http://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/1dro426p7j/Times_Scotland_Results_140629_Wednesday_W.pdf

    and this and previous Survation ones (and indeed most other IndyRef polls) is the comparative lack of a gender gap in YouGov. In the latest Survation men are in favour of Yes by 49% to 43% (+6)[1] while Women choose No by 49% to 33% (-16) – a difference of 22 points.

    In contrast YouGov’s figures showed Men 53% to 37% (-16) and Women 54% to 33% (-21) a difference of only 5 points and mostly due to women being a bit more likely to say they are uncertain[2].

    [1] Using the table including Likelihood to Vote weighting. YouGov ask an LTV question, but it’s not clear if they use it to weight their headline IndyRef figures.

    [2] Women are nearly always more likely to say DK or equivalent in polls (look at any YouGov daily). The Survation poll actually shows a much bigger discrepancy in this, as do other polls, but the difference in those who have decided is very noticeable as well.

  18. @Statgeek: “83% of Scots voters voted for something other than Conservative in 2010, and yet they got a (mostly) Conservative government.”

    As did much of England, Wales and Northern Ireland.

  19. @Reg: “You seem to be out of line with the general belief on this site”

    I’m not sure I am. About 3 to 2 in favour of the status quo seems to be the consensus.

    Interesting study here, which was mentioned in a discussion on justthetalk.com earlier today:

    “Judging from Gallup Polls in the United States, the United Kingdom, and Australia, opinion often changes during an election campaign. Come election day itself, however, opinion often reverts back nearer to where it was before the campaign began.”

    For the referendum that’d see support return to around 30-35%.

  20. Sorry, forgot to include the link:

    “Waking Up in the Poll Booth”

    http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=6677236

  21. @ Roger H

    As did much of England, Wales and Northern Ireland.
    ————
    And that’s the point which I make to Yes supporters who trot out the ‘get the government we vote for’ line. They become a bit nippy when it’s pointed out to them that, for the Euros, UKIP did better in Scotland than in London…

  22. @Gazprom
    “The Yougov poll has a 7.9% differential in the unweighted data”
    ——————
    Indeed. I have been puzzled by this for some time myself but have never had any response to my comments.

  23. Remember, when the surprise cabinet shuffle is announced on monday, everyone look surprised.

  24. Amber Star

    [Yes supporters] become a bit nippy when it’s pointed out to them that, for the Euros, UKIP did better in Scotland than in London…

    Possibly because it’s not true. UKIP got 16.87% in London

    ht tp://en.wikipedia.org/wiki/London_(European_Parliament_constituency)#2014

    but only 10.5% in Scotland:

    ht tp://en.wikipedia.org/wiki/Scotland_(European_Parliament_constituency)#2014

    It’s true that some on the Yes side can get irritatingly pious pointing out the perceived moral superiority of the Scottish electorate when the facts don’t support them (as I sometimes have great fun pointing out on Wings) but there are genuine differences, even if they’re not as great as some imagine.

  25. So, the latest Populus and the last 3 YouGov polls all have the Greens on 5%: those who were saying it was just a short term Euro-election effect that had Greens higher than in previous years surely have to re-think now (incidentally, UKIP _have_ fallen 1.5 to 2 points over the last 6 weeks).

  26. RM

    Yes if you only look at vote share, but in terms of seats…

  27. Gazprom,

    I think you make a good point about assumption re. the election, aside from the polling…the consensus view among serious pundits that the tories and labour will have pretty much the same vote share, with possible the tories a point or even two ahead. This translates to a hung parliament and labour with 300+ seats as the largest party.

    I am beginning to think that the swing back, if it occurs, might only be 1 or 1.5%, in which case you’d have the tories on 34 and labour on 35…that’s almost a labour majority….i think labour could surprise on the upside.

  28. @ Roger Mexico

    I said: UKIP “did better” in Scotland i.e. 1 of 6 MEPs than London 1 of 8 MEPs. 1/6th is more than 1/8th.

  29. So the next EU commissioner is going to be IDS?
    Always assuming something like a govt reshuffle is being made privy to the charming Flick.

  30. We have crossover on Stephen Fisher’s GE prediction:

    Forecast Election Day Seats Con : 295 Lab : 296 LD : 31

    He has been predicting Con ahead until now because the model (based on past trends, including past wonky polling data) predicts a swing-back to the government. However as time passes the swing-back effect gets smaller. Since the polls have not shifted substantially it was only a matter of time before crossover. And here we are.

    The model is still predicting a larger Con share of the vote, 35-32-13.

  31. That election result would produce a terrible constitutional crisis, as no stable coalition could be formed with a majority of more than 4 (although technically it would be 14)

  32. @Steve 2

    ‘Cooler heads will prevail’. Yes, but who has the cooler heads?

    The No campaign has done very well in burying uncomfortable truths (for example, confusing the value of all Scottish oil with the North Sea fields instead of including the Clyde and Atlantic basins) and pointing out (quite rightly, of course) the inadequacies of the Yes campaign so far.

    The Clyde basin is ‘off limits’ because of the nuclear weapons subs – yet another reason for getting rid of those criminal things.

    In any case, after the No vote, a lot of rethinking will have to be done – especially if Labour wins the next GE. Scottish Labour will have to find a way of winning back power in Holyrood – and with a Labour government having already set out its economic strategy by the time the Holyrood elections come round, I really don’t see the Lamentables coming anywhere near that……

  33. Amber Star

    I said: UKIP “did better” in Scotland i.e. 1 of 6 MEPs than London 1 of 8 MEPs. 1/6th is more than 1/8th.

    Have you thought of a career as an adviser to Lord Ashcroft? :P

    We all know what ‘did better’ means and it’s percentage of the vote, not of some imaginary delegation.

  34. Well it looks like Ed Miliband and Labour are confounding all the pollsters and commentators who have been expecting a cross-over in VI and the Tories to be ahead in the polls.
    The much heralded new model by Stephen Fisher of Oxford University which he uses to predict the next election based on current opinion polls and the track record of polls in previous electoral cycles allowing for change in opinion in the run up to the election had these figures when it began in October 2013.

    Approximate probabilities of key outcomes

    Pr(Con majority) = 57%

    Pr(Lab majority) = 15%

    Pr(Hung parliament) = 28%

    Pr(Con largest party) = 88%

    Pr(Lab largest party) = 12%

    The new model predictions are:

    Approximate probabilities of key outcomes

    Pr(Con largest party) = 49%

    Pr(Lab largest party) = 51%

    Pr(Con majority) = 23%

    Pr(Lab majority) = 25%

    Pr(Hung parliament) = 52%

    Pr(Hung parliament with Con largest party) = 26%

    Pr(Hung parliament with Lab largest party) = 26%

  35. In May 2014, output in the construction industry was estimated to have fallen by 1.1% compared with April 2014, after increasing by 1.2% in April. Both new work and repair & maintenance fell by 1.1% in May 2014.

    ONS

    Markit/CIPS construction PMI rose to a four month high of 62.6 in June from 60 in May, where anything above 50 indicates growth.

    A bit of a contrast, similiar to the manufacturing data yesterday

    and for a further view, the Bank of England agents report covering the month of May

    http://www.bankofengland.co.uk/publications/Documents/agentssummary/2014/agsum14jun.pdf

    from the report

    Investment intentions had moderated – similar to the British Chambers of Commerce report

    Manufacturing output for the domestic market had continued to grow at a steady pace. Sterlings strength starting to effect exports, similar to the BCC findings

    Construction output had continued to grow strongly – that supports Markit

    Employment intentions had been little changed and still pointed to modest growth in the next six months – The BOE agents reports have never indicated fast growing employment despite what the ONS have been reporting

    Growth in total labour costs per employee had increased slightly, remaining moderate- no big rises in wages in May according to the BOE agents

    Overall i would say these falls in construction and industrial production for May look like a blip, but Markit figures look wildly optimistic.

  36. @RM and AS

    No, we don’t agree on what the term ‘did better’ means. ‘Doing better’ can mean making a stronger impact in terms of votes, or in terms of seats won. Sometimes it means both. What AS failed to do was specify which meaning was being adopted.

    The Tories may well ‘do better’ than Labour at the next GE in terms of votes, but Labour may well ‘do better’ in terms of seats.
    Labour may even do better inboth categories, of course.

    Naturally, we all know that what AS really meant is that Scotland is no different from England.
    Ah, but no, for what AS really meant ‘really’ is that Scotland is not as different from England as some Scots claim.

    But would 45% of English voters choose to leave the UK? I doubt it! OK, so it may only be 40%. Only???

    And this brings me to the serious point: to have such a huge percentage of people quite obviously dissatisfied with the Union must not lead those in the South to sleep easy, thinking that all is well. All is not well.

  37. Is EU Commissioner the modern equivalent of Northern Ireland Secretary, only with better pay?

  38. (And no Special Branch cover.)

  39. @ Roger Mexico

    As N E fuel no… ‘did better’ means percentage of the vote, not of some imaginary delegation.
    ———————–
    I disagree, Roger. ‘Did better’ doesn’t necessarily mean % of vote – because e.g. in 2015 Labour could have more MPs than the Conservatives with a lower % of the vote. That gets mentioned a lot on UKPR.

    And I think that Labour will have “done better” than the Conservatives by being the largest Party regardless of which Party gets the biggest % of vote. ;-)

  40. JOHNB

    @” to have such a huge percentage of people quite obviously dissatisfied with the Union must not lead those in the South to sleep easy, thinking that all is well. All is not well.”

    The idea that large numbers of UK citizens, living outside Scotland, can be persuaded to lie awake at night worrying about whether Scots want independence or not , is bizarre.

    We don’t give a to** do we?

    And if they vote NO , and still keep moaning -well what’s new? It hasn’t caused nocturnal worries to the English before-why should we start now?

  41. ******* IMPORTANT *******

    Okay folks, here’s a real treat for you.

    I’m one of the experts who normally hides out in the “Election Guide” section of this blog but I’ve decided to descend among the hoi polloi of this thread and – JUST THIS ONCE – give you all the benefit of my unsurpassed expertise.

    The following link is to a scholarly thesis I recently posted on the “Labour target seats” thread. It uses historical precedent to analyse whether the current Labour lead is likely to hold until May 2015.

    If afraid it doesn’t look too good for Ed Miliband…

    http://ukpollingreport.co.uk/2015guide/labourtargets/comment-page-9/#comments

  42. Colin
    “And if they vote NO , and still keep moaning -well what’s new? It hasn’t caused nocturnal worries to the English before-why should we start now?”

    Not everyone south of Scotland is English.

  43. JohnB
    “All is not well.”
    Don’t get me wrong, you are my favourite Yea-sayer by a long chalk, however the above has more than a touch of Private Fraser about it.
    All together now,
    “Doomed, we’re all Doomed !”

  44. @ Robin Hood

    The problem is that for this election, past performance simply does not apply.

    There has been no discernible drift in votes from Tory to Labour over the parliament. The swing is from LD to Labour. And then there is the UKIP factor.

    And when you look at the ex-LD voters (of which I am one), they are absolutely adamant they will never vote Tory, and nor are they going to ‘swing back’ to the LD’s.

    The old ‘swing back’ theory is not valid in 2015. This won’t behave like 92, or 87. Just look at the stablity in the polling we’ve had since the end of 2010.

  45. Fischer’s forecast is very interesting: a horribly hung parliament, LD’s unable to supply a proper majority for a coalition, deals being done here there and everywhere. You’d only need someone dying or someone crossing the floor to upset things. The markets would not like the instability either.

    There’s several LD seats up for grabs over the country that are not necessarily vulnerable to Labour, it depends where their vote goes locally. I can see that in Scotland the beneficiaries would not necessarily be Labour. From the 2011 Scottish election, it is plain that the LD’s were offering up votes and seats to anyone who wanted them, as long as the recipients were themselves not horribly unpopular, perhaps also being a party with not dissimliar idealogy.

    As for the Conservative position in Scotland, I think their GE vote share cannot really go much lower, therefore if there are wild swings in other parties’ votes, they might just increase their seat count. FPTP is again shown to be strange.

  46. @ Robin Hood

    ******* IMPORTANT *******

    In your thesis, what value have you assigned to the Party of government being able to decide when it holds the election?

  47. Reshuffle on Monday morning it would seem.

    There have been quite a few speculative articles, most of them contradicting each other. This is generally true of pre-reshuffle rumour. Only the PM, very senior ministers and those civil servants working directly on the reshuffle with them will know about it and even now it is unlikely that the Is have been dotted and the Ts crossed. Most of what is reported is speculative stuff from those very backbench MPs without much of a say.

    Personally I’d be surprised if there were big changes to the very top jobs. One of the CONs current strengths are that people think they have a better team of top ministers than LAB do, and a number have become entrenched in their departments to the extent that they’d be hard to move. The IDS rumour is vaguely plausible – his reforms have been put into law now and he may be happy to relieve himself of the implementation difficulties. However, I’d still be quite surprised.

    There will be more women, for sure, but I doubt Cameron can meet either the ‘1/3 of ministers’ or ‘1 in every department’ promises he made at various times in opposition. This isn’t because of sexism but simply because there aren’t all that many available female Tory MPs, and he isn’t going to promote no men at all. I can’t see Nicky Morgan getting another promotion as some suggest (she is very good but she has had two in the last year already) so if two women are brought in one will probably be Liz Truss and the other either Esther McVey or Anna Soubry (McVey will make headlines but would be a risky appointment IMO – still a very inexperienced politician and probably out of depth running a department). Clegg could also bring Jo Swinson in but it is hard to see who for unless Cameron allows the LDs an extra cabinet position (doubtful).

    Expect Sir George Young to go, along with Andrew Lansley and possibly Owen Paterson. Young’s departure will be voluntary, the other two possibly not. I think the CONs could do with somebody with a bit more gravitas to take over the media side of Grant Shapps’ role – that could be one for Eric Pickles.

    There will be some middling ministerial changes but there was a comprehensive reshuffle of these last October so expect them to kept to a minimum.

    Then there is the issue of the Commissioner. Lansley has been talked about a lot but, given the coverage the Juncker appointment, I think Cameron might have second thoughts. Somebody that won’t be a loss to the government, won’t cause a by-election loss and will please the back-benchers whilst having his head screwed on sufficiently to avoid Brexit is surely a must – Liam Fox? Oliver Letwin?

  48. BRAMLEY

    It was to save having to type “UK citizens, living outside Scotland, ” again.

    But well spotted.

  49. ROBINHOOD

    Great effort-I like it :-) :-)

    Thanks !

1 2 3 7