Tuesday, December 23, 2014

Support for independence reaches 50% for the first time ever in a Survation poll

I'll keep this relatively brief, because after the tragedy in Glasgow only ten hours ago, I know none of us are really in the mood for politics at the moment. However, this is a landmark finding from Survation, and many thanks to Calum Findlay for pointing it out.

Imagine there were another referendum on Scottish independence held today. How would you vote if the question were ‘Should Scotland be an independent country?'

Yes 50% (+3)
No 50% (-3)


The percentage changes are from the last Survation poll, which was (and remains) the only post-referendum poll from any firm to have No in the lead. YouGov have had Yes slightly ahead in both independence polls they have conducted since September, and there was also a narrow Yes lead in the sole Panelbase poll to ask the question. In some ways, though, tonight's result is the most impressive so far, because there's no obvious "alibi" to cover for the fact that Yes have drawn level. When the Panelbase poll was published, John Curtice was mildly critical of the question wording that had been used, and in the case of the YouGov polls he pointed out that Yes wouldn't have quite made it into the lead if there had been weighting by recalled referendum vote. With Survation there is no such get-out clause, because the question wording is scrupulously neutral, and weighting by recalled referendum vote has been applied. If that weighting hadn't been used, ie. if Survation had simply stuck with the same methodology they used in their pre-referendum polls, then they would be showing Yes in a clear lead tonight. That tells you all you need to know about the swing in favour of independence that has occurred since September (something that all pollsters agree upon), because at no point during the referendum campaign did Survation ever have Yes higher than 48%.

Respondents were also asked when, if ever, a second independence referendum should be held.  12.7% think it should take place immediately, a further 21.7% think it should happen before 2019, and 18.4% want it between 2019 and 2024.  14.5% think it should be at some point after 2024, while a hard-core of 26.3% (almost exclusively No voters) don't think a second referendum should ever take place.  Probably the best way of making sense of those numbers is to group together those who want a referendum within ten years, and those who don't (with undecideds excluded).

There should be another referendum within the next ten years : 56.4%
There shouldn't be another referendum within the next ten years : 43.6%

It's tempting to point out that the voters have once again given short shrift to the outrageously anti-democratic narrative of "it's over for a lifetime" that David Cameron tried to get off the ground on the morning of September 19th, but in truth I think even the majority of unionist politicians have long since given up on that one.  Incidentally, even with undecideds included, there is still an absolute majority (52.7%) in favour of a new referendum within a decade.

*  *  *

There have been two new GB-wide polls published over the last twenty-four hours - one from Populus, and yet another one from YouGov that I wasn't really expecting.  The Scottish subsample from the Populus poll (conducted between the 19th and 21st) has the SNP ahead of Labour by 37% to 23%, while the bang-up-to-date YouGov subsample (conducted on the 21st and 22nd) has the SNP in the lead by 43% to 27%.  Both of those results may look very similar to what we've become accustomed to, but in fact Populus are showing a significantly bigger gap than their recent average.  It's also striking that the SNP have maintained such a huge advantage in a YouGov poll that sees Labour at a Britain-wide level move up to an unusually high 36% - it seems to be in London and the North of England that Miliband is making some progress (although of course that may just be a one-off statistical blip).

And unless any newspaper is crazy enough to publish a poll on Christmas Eve, that really should be that!

17 comments:

  1. It is easy to choose the radical option when you know it won't happen, but it is encouraging to see Survation, YouGov and Panelbase all showing roughly 5% swings towards indepedence. Dead in the water it definitely isn't.

    ReplyDelete
    Replies
    1. True, although similarly, it's not uncommon for someone who's swithering between two options to end up plumping for the "default" or so called "safe" option, only to bitterly regret their lack of courage afterwards, and think "never again, if I get a second chance I'll be taking it."

      Well, that's the hope anyway!

      Delete
    2. I think quite a few NO voters thought on the morning of the 18th:


      "I'm going to give the buggers one more chance, and if (nay when) they blow it, that's it - next time I'm Voting YES!"

      They wanted to be reasonable -one last time.

      Delete
    3. Much as they did for John Major in 1992. Only for Blair to give us another 13 years of Thatcherism.

      Delete
  2. "A quarter of all Scots say NEVER AGAIN to referendum" - all newspapers tomorrow.

    ReplyDelete
    Replies
    1. As a lovely Christmas present for all who read it!! :)

      Delete
  3. Aye three quarters say there will be and so do I,hell need one quicker than 10 years I might not last that long.

    ReplyDelete
  4. The fact that Labour are doing so badly in Scotland must mean they are doing much better in England and Wales than UNS would predict. If the election result is close to the polls this week and the SNP's vote holds up, Labour could scrape into a Labour/Lib Dem collation without winning in Scotland.

    Scotland getting a Lab/Lib Dem Government that it didn't vote for is a pretty interesting idea. It would totally demolish the argument that it's our duty to vote Labour anyway.

    ReplyDelete
  5. 'And unless any newspaper is crazy enough to publish a poll on Christmas Eve, that really should be that!'

    Don't know if it was for a newspaper but I just did a Panelbase poll asking past Indyref vote, future Indyref vote, and past Holyrood and WM vote - presumably for weighting, also country of birth, etc.

    Also qs about who I trust to act in Scotland's interest - Nicola Sturgeon, John Swinney, Patrick Harvie, Willie Rennie (I know!), Jim Murphy (lulz), etc.

    ReplyDelete
    Replies
    1. That could be a private poll for the SNP, as they used Panelbase during the referendum campaign.

      Delete
  6. Survation say they have another poll out tonight (groans).

    ReplyDelete
    Replies
    1. Just seen their Tweet. Sounds like it's a UK one.

      Labour's loyal servant Daily Record frantically panicing and tyring to get a swing to Labour before Christmas? And put that as momentum for Murphy.

      Why another poll so soon! Unless it's a UK wide poll with a small Scottish sub-sample.

      Panelbase could be interesting, but if folk aredoing them now then I doubt we'll see it released until after Christmas.

      Delete
    2. Survation have done some random constituency polls before, so it could be one of those. e.g. Unite union recently paid for them to survey two safe Tory seats (held by David Cameron and Jeremy Hunt) for no apparent reason.

      Delete
    3. It was just a standard GB voting intention poll for the Mirror. Very small sub-sample size (79, weighted up to 84). FWIW (not much) it had Lab 49, SNP 27, Tories 10 and UKIP 9. Survation's GB polls tend to have very high UKIP shares.

      Delete
  7. I've just taken part in a Panelbase poll.
    The first question was who did you vote for in the 2010 UK elections.
    Would more SNP MP's better protect Scotlands interests.
    Should Scotland get DevoMax? DevoMax explained as all powers except foreign affairs and defence.
    A question giving all the various outcomes of the GE and asking which was best.
    Questions asked if you trust all the Scottish and English leaders to look after Scottish interests.
    It had the flavour of a SNP poll?

    ReplyDelete
  8. It must be a different one from HapLeg, as I wasn't asked any referendum questions.

    ReplyDelete
    Replies
    1. All the questions sound the same. I only joined Panelbase after the referendum so maybe in my excitement I thought I'd answered a question about Indyref intention when it was only a recall weighting question. I didn't keep a note of the questions as you seem to have done so I will defer to you.

      Delete