Thursday, February 26, 2015

SNP extend lead to 20% in new Poll of Polls

It hasn't been long since the last update of the Poll of Polls, but I think another one is probably in order, because the full-scale Survation poll has now dropped out of the sample, and there has also been an intriguing flurry of dreadful results for Labour in Scottish subsamples from Britain-wide polls.  And when I say 'dreadful', I mean dreadful even by the standards of recent weeks and months.  That may have happened by pure chance (ie. as a result of normal sampling variation), but it'll be well worth keeping an eye on the remaining YouGov and Populus subsamples this week to see if there's any further sign of public opinion hardening against Labour and the hapless Mr Murphy.

This update is based on eight Scottish subsamples - four from YouGov, one from Survation, one from ComRes, one from Ashcroft and one from Populus.

Scottish voting intentions for the May 2015 UK general election :

SNP 43.8% (+0.5)
Labour 23.9% (-2.7)
Conservatives 17.4% (+0.8)
Liberal Democrats 6.9% (+1.3)
UKIP 4.5% (+1.1)
Greens 3.3% (-0.2)


(The Poll of Polls uses the Scottish subsamples from all GB-wide polls that have been conducted entirely within the last seven days and for which datasets have been provided, and also all full-scale Scottish polls that have been conducted at least partly within the last seven days. Full-scale polls are given ten times the weighting of subsamples.)

22 comments:

  1. The 48% SNP / 22% Lab in the morn's UK Yougov is with the highest level of SNP downweighting post iref. SNP/PC x0.51.

    ReplyDelete
    Replies
    1. Interesting to see that the SNP % is so high despite the downweighting. It seems to suggest the downweighting is correct, in which case the polls are not asking the right people and therefore might not be that reliable. The other thing it could suggest is that the true SNP figure is closer to 60% which does not seem credible.

      Delete
    2. SNP 60% lead?

      Oh please God! Oh please God! Oh. Please. God.

      Delete
    3. Craig : No, it doesn't suggest the downweighting is correct, because the downweighting by party ID doesn't happen in full-scale Scottish polls. The important thing is not where the SNP would be in any individual subsample without the downweighting, but where they would be on an average of several subsamples. The answer is that they'd be higher, but not at an implausible level.

      Delete
  2. Correct me if I'm wrong, but the impression I get is that the SNP vote seems to be incredibly solid and holding at record levels, whereas there still seems to be a little movement between unionist parties. It's as if unionists and SNP-haters just cannot decide whether to vote tactically or opt for one party out of a poor choice. Nothing looks effective for them at the moment as the SNP lead is just too big to overcome. It must be really demoralising for them. My heart bleeds.

    ReplyDelete
    Replies
    1. as a pro-indy snp hater i'm not sure how to vote.

      probably green for principle but a tactical vote would be for labour against the lib dems where i am. and i don't fancy voting labour.

      i'd rather see a labour gov uk wise than a tory gov but i'm an internationalist so i see the well being of workers in rUK as quite important.

      i can see how if a person doesn't give a shit about workers in the rUK and only cares about scot indy then another tory gov isn't the end of the world.

      but i'm probably anti-tory more than i'm pro indy.

      not least cos a rabidly right wing rUK will be an awful awful neighbour and we'll be wee enough that a much larger andf awful neighbour could really fuck things up for people living in scotland.

      in summary: i dunno.

      Delete
    2. oh i'm sorry i reread your post and i see you're viewing things exclusively through the prism of independence and talking about tactical voting from that perspective rather than the usual keep out the tories perspective..

      Delete
    3. Is that anti-red tory or anti-blue tory?

      Delete
    4. People are actually expected to believe that's a real post?

      Delete
  3. Looking at a long term 3 month trend it looks like Labour are losing votes to both the SNP and the Tories. Is this a Jim Murphy effect?

    UKGE trend projections;

    SNP: 44%
    Labour: 26%
    Tories: 17.5%
    Lib Dems: 5.5%
    Greens: 3.5%
    UKIP: 3.5%

    ReplyDelete
  4. New Populus telephone poll in progress. I was polled yesterday.

    ReplyDelete
    Replies
    1. Did they ask a "thinking of the candidates in your constituency" question?

      Could be Lord Ashcroft...

      Delete
    2. Yes, exactly that wording. Checked I was in Kirkcaldy and Cowdenbeath. Asked about my likelihood to vote, who I would vote for, then asked again "thinking of candidates...", then some further questions about UK economy, and then a multiple choice on my preferred government (including Tory/SNP coalition), then finally about Cameron good job/keep him, poor job/prefer Ed, poor job/keep him anyway as better than Ed.

      Delete
    3. Also asked about 2010 vote, but not 2011 or ref.

      Delete
    4. Yup, Ashcroft all the way!

      Delete
    5. Populus are doing a specific East Renfrewshire poll on behalf of Murphy/Lab at present.

      Questions like would you be more likely to vote Labour if you could access alcohol at football matches is one of the questions.
      I think Murphy is tipping his toe in the water before declaring.

      Delete
  5. Why are the SNP downweighted so much

    ReplyDelete
    Replies
    1. Because they weight to party identification figures based on answers people were giving in 2010. The theory is that party identification remains fairly stable over time, even if voting intention changes. But in reality, there has been a sea-change in party identification in Scotland over the last few years.

      Delete
    2. So there is no real way to predict the level of SNP support?
      other than it may be around 48%?
      Trying not to get my hopes up (still haven't really got over September 18th)

      Delete