There's not a huge amount more I can say about the new YouGov poll, because unlike John Curtice I haven't had the advantage of seeing the datasets - and given that this is a Bank Holiday weekend, it's not completely impossible they might not be published until Tuesday. However, Curtice's own article does reveal a few interesting titbits. Firstly, the independence question was asked again -
Should Scotland be an independent country?
Yes 47% (-1)
No 53% (+1)
This is more likely to be margin of error 'noise' than a genuine easing down of support for independence. It looks to me as if the Yes vote with YouGov has been hovering around the 48% mark since the big methodological change a few months ago. Similarly with Survation, it's been hovering around the 49% mark - sometimes a little higher, sometimes a little lower, but it was bang on 49% in the most recent poll a few days ago. Bear in mind that both firms now weight by recalled referendum vote, so we can be fairly sure that the Yes share has genuinely increased from the 45% recorded on September 18th. This also means that poll results now are not directly comparable with pre-referendum poll results, so the people (and unfortunately they do exist) who claim that "YouGov's figures are much the same as before referendum day, and look what happened then" are barking up the wrong tree.
The percentage of respondents who want another independence referendum within ten years has slipped slightly from 40% last month to 36% now. That figure is obviously much lower than the 59% who said the same thing in the Survation poll, but that's largely because of different methodology. Survation don't seem to offer a Don't Know option on that question - so every respondent is forced to give an opinion. More importantly, they also offer a far more realistic range of options than YouGov, with the two extremes being that a referendum should never take place, and that there should be another referendum within two years (there's also an option of within five years). In YouGov's case, the option of a referendum within ten years IS one of the two extremes, and seems to be presented to respondents last. Other than Don't Know, the six options are "never", "not for at least 50 years", "not for at least 25 years", "not for at least 15 years", "not for at least 10 years" or "within 10 years". The unspoken message being sent to respondents is that you're on the fringes if you select the last option - and the fact that 36% still did so speaks volumes. Even more significantly, a combined total of 49% selected one of the last two options, which clearly implies there should be a referendum within 15 years (ie. less than the fabled 'generation') -
Less than fifteen years : 49% (-2)
More than fifteen years : 44% (n/c)
By the way, a mere 16% say there should never be another referendum.
Encouragingly, there is now an absolute majority of respondents who think that, regardless of their own preference, there will be another referendum within ten years. Clearly, Jim Murphy's sterling efforts in talking up the prospect of that happening is helping to normalise the idea. Thanks, Jim!
There probably WILL be another referendum within ten years : 54% (+4)
There probably WON'T be another referendum within ten years : 33% (-4)
Curtice also implies that roughly half of the Lib Dems' 7% support is made up of tactical voters. This may be a cause for concern for the SNP, because in most cases that will presumably be anti-SNP tactical voting. The only likely exception is in Berwickshire, Roxburgh & Selkirk, where some voters may still (wrongly) assume that the Lib Dems are the best hope of keeping the Tories out. It's noted by Curtice that many people voting tactically this time did the same thing in 2010, so in a sense the phenomenon is already 'factored in' to the baseline numbers - but that may not be much comfort to the SNP in Lib Dem-held seats. There won't have been many Tories voting Lib Dem to keep the SNP out five years ago - of if there were, they were behaving totally irrationally.
I still haven't been able to track down the fieldwork dates for the poll, although it seems to have concluded on Friday, which presumably means that only a small proportion of interviews took place after Miliband revealed on Thursday night that he might be prepared to help Cameron stay in office.