Pro-independence vote increases to 46.1% in perspicacious Panelbase poll

It turns out that I was wrong in my guesswork last night - when Panelbase said there was "no real change" in their latest independence poll, what they meant was that there was no change at all.  Actually, to be fair, there was a slight change of sorts - one of the many advantages to having a poll commissioned by Wings is that Stuart often publishes the datasets straight away (not exactly something that would happen in the Telegraph), and from that I can see that the unrounded Yes vote has in fact increased, by...well, by 0.5%.

Should Scotland be an independent country?

Yes 46.1% (+0.5)
No 53.9% (-0.5)

Before anyone jumps down my throat - no, of course a 0.5% swing is not remotely statistically significant, and yes, these results are firmly within Panelbase's normal range.  So this poll fails to corroborate the potentially more significant surge we saw in the BMG poll, and thus increases the likelihood that the pro-Yes swing in BMG was an illusion caused by normal sampling variation.  Nevertheless, it's not that long at all since the media were trying to convince anyone who would listen (on the basis of very little evidence) that the Yes vote was undoubtedly on a downward trajectory.  This poll, especially when taken in conjunction with BMG, inconveniently contradicts that narrative as well.

There's a belief in some quarters that recent referendums and elections have proved that supplementary questions sometimes give you a better idea of the state of the race than the headline voting intention question does.  For example, it can be argued that Ed Miliband's poor personal ratings, and Labour's poor ratings on economic competence, were the giveaway clues that the voting intention figures in the 2015 election were leading us astray.  So unionist critics would be wise not to be too dismissive of three more nuanced questions Stuart invited Panelbase to ask, tying the independence question to views on Europe, or to the potential for neverending Tory rule from Westminster.  Those are points that may well be at the forefront of voters' minds by the end of an indyref campaign, even if they aren't at the start.

There's a mixture of good news and bad news on those questions.  When a four-option question on constitutional preferences (indy inside EU, indy outside EU, no indy inside EU, or no indy outside EU) was first asked in July 2015, the two independence options were almost as popular (48.3%) as the two anti-independence alternatives (51.7%).  That gap has now widened to 44.4% for the pro-independence options and 55.6% for the anti-independence options - slightly worse than on the headline independence question.

But as Stuart points out, the four-option question is hopelessly outdated anyway, because the idea of Scotland remaining in the EU as part of the UK is no longer a runner (except in Lib Dem fantasies).  The more realistic three-option question does produce a majority for the two pro-independence options.  After Don't Knows are stripped out, 52.5% of respondents want an independent Scotland either inside or outside the EU, while only 47.5% want Scotland to remain part of a UK that has left the EU.  The snag, of course, is that for this narrow advantage to be pressed home at the next indyref, we'll need to convince all or most of the anti-EU independence supporters that it's still worth voting for independence even if that means remaining within the EU - either that or we'll have to bring across some pro-EU people who haven't seriously considered independence yet.  In reality, it'll probably need to be a blend of the two.

The question that invites people to assume that Labour will never again win a UK general election (not as fanciful an idea as we might have once thought) produces a small boost for Yes - with Don't Knows excluded, it narrows the race to Yes 47.1%, No 52.9%.  So it looks like perpetual Tory rule will not be a decisive argument in itself, but even the smallest of tractions is not to be sniffed at in a close contest like this one.  Stuart also notes that there are seemingly irrational movements within the subsamples for that question, which may cast doubt on whether some of the respondents really grasped what they were being asked.

*  *  *

A Scot Goes Pop reader sent me the following email a few hours ago -

"Just to let you know I have just completed a Populus survey on independence. Lots of questions on the UK Gov rejecting a second ref."

Could be an innocent poll for the mainstream media, but then again Populus have been known to act as a private pollster for the Tories.  Are the UK government seriously trying to work out whether they might just get away with Michael Fallon's "forget it, Jocks" message?  Good luck to them if they are - they'll need it.

*  *  *

If you've enjoyed my writing in recent months and feel a strange inexplicable urge to 'buy me a hot chocolate', bear in mind that my fundraiser from two years ago is still open for additional donations - it can be found HERE.