The headline results from a new full-scale Scottish poll from YouGov have been released by the Sun on Twitter. So far I haven't been able to track down the fieldwork dates, but Calum Findlay mentioned yesterday evening that he'd just taken part in this poll, so it's presumably bang up to date. Let's hope so, because the results are very much at the extreme upper end of what my expectations would have been.
Scottish voting intentions for the May 2015 UK general election :
I've been slightly bemused over the last few days and weeks by the number of London commentators who appear to think that Scottish Labour's problems are being caused by a "leadership vacuum" and that some sort of "honeymoon period" for Jackanory Jim is about to provide a quick fix. In reality, the Scottish media have been shamelessly treating Murphy as the unofficial leader for weeks now, so that's already factored into the polling results. To be fair, that doesn't necessarily mean that what's going on is Murphy's fault. The electorate know that Miliband is the real leader, and above all else it's him that they don't like, rate, trust, or respect.
* * *
SCOT GOES POP POLL OF POLLS
Tonight's Poll of Polls update is based on the full-scale YouGov poll, plus Scottish subsamples from five GB-wide polls - four from YouGov, and one from Populus. That means fourteen-fifteenths of the sample comes from YouGov, which is plainly less than ideal!
Scottish voting intentions for the May 2015 UK general election :
SNP 45.0% (+2.4)
Labour 26.5% (-0.4)
Conservatives 16.7% (+0.6)
Liberal Democrats 4.5% (-1.8)
UKIP 3.5% (-1.1)
Greens 2.9% (+0.5)
(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.)
Scottish voting intentions for the May 2015 UK general election :
SNP 47% (+4)
Labour 27% (n/c)
Conservatives 16% (+1)
Greens 3% (-1)
UKIP 3% (-3)
Liberal Democrats 3% (-1)
There have of course been two post-referendum polls from other firms that were even better than this for the SNP (Ipsos-Mori gave them a 29-point lead and Survation gave them a 22-point lead). But to see a gap of as high as 20 points from YouGov is still a bit startling, because that firm's previous estimate of a 16-point lead was more in line with the average results of their daily subsamples. In fact, the SNP's subsample lead both today and yesterday was exactly 16 points. And perhaps more significantly, the party's raw share of the vote in the subsamples has more often than not been quite a bit lower than 47%. So this poll raises the serious possibility that the daily GB-wide YouGov polls have been understating the SNP's strength, in spite of the party performing outstandingly well in them. To be fair, there's always been a straightforward reason for supposing that might have been going on - in GB-wide polls YouGov use Westminster-oriented weighting by party ID.
The other point that leaps out is just how dreadfully badly the smaller parties are doing. I had assumed that the Liberal Democrats' 4% share in the last YouGov poll was a freakish result, but the chances of them being significantly underestimated by random margin-of-error effects in two consecutive polls is obviously pretty low. The jury is still out on UKIP, though, because they fared a bit better in the last poll.
Should Scotland be an independent country?
Yes 48%
No 45%
It's impossible to know the exact significance of this result until we see the datasets, or find out whether there have been any methodological changes. The last YouGov poll (which was the only other post-referendum poll from the firm to ask the independence question) produced almost identical results to this, putting Yes on 49% and No on 45%. However, it was immediately criticised by John Curtice and one or two others for not weighting by recalled referendum vote, which would have had the effect of keeping No in the lead, albeit only very narrowly. I was slightly cynical about that intervention, because I don't recall Curtice making similar unofficial adjustments to the many pre-referendum polls that failed to weight by country of birth, and which therefore underestimated the Yes vote. However, it's possible that YouGov will have heeded his complaint, in which case tonight's result is even better for Yes, because it suggests there has probably been a further swing in real terms since the last poll.
If they haven't made any methodological change (which is also perfectly possible - they may want to keep the trend figures meaningful) then it means there hasn't been a further swing, but also that the headline numbers remain directly comparable with pre-referendum YouGov polls, which in all but one (legendary) case had No ahead. So either way there is no real doubt that a significant number of voters have been converted to independence since September 18th.
Views on the Smith Commission and its proposals for further devolution :
It doesn't devolve enough powers : 51%
It gets the balance right : 23%
It goes too far : 14%
And there, in a nutshell, is the explanation for why the SNP have either maintained their advantage over Labour or increased it somewhat - it appears that they've comprehensively won the battle of perceptions over Smith. It's all very well for Michael Portillo to sit on a BBC sofa in London, smirking at Alex Salmond like an overgrown schoolboy while tittering : "The Vow has been delivered! You know that! You're just playing games!" But unfortunately for the London establishment and the Daily Record (is there a difference between the two?), it turns out that Scottish voters are not half as stupid as Portillo takes them for. They know what they heard Gordon Brown promise - and they also know that what the Smith Commission has come up with is quite simply not "Home Rule". Nor is it "near federalism". Nor is it the "Devo SUPER Max" promised by Better Together's official representative at the TV debate in the Hydro.
Assuming that YouGov only offered respondents the three options listed above, roughly 58% of people who gave a view said that the Smith proposals are inadequate. It's also fascinating to see how few people thought the proposals go too far, because that must encompass the evidently dwindling group who are opposed to devolution/self-government altogether.
Greens 3% (-1)
UKIP 3% (-3)
Liberal Democrats 3% (-1)
There have of course been two post-referendum polls from other firms that were even better than this for the SNP (Ipsos-Mori gave them a 29-point lead and Survation gave them a 22-point lead). But to see a gap of as high as 20 points from YouGov is still a bit startling, because that firm's previous estimate of a 16-point lead was more in line with the average results of their daily subsamples. In fact, the SNP's subsample lead both today and yesterday was exactly 16 points. And perhaps more significantly, the party's raw share of the vote in the subsamples has more often than not been quite a bit lower than 47%. So this poll raises the serious possibility that the daily GB-wide YouGov polls have been understating the SNP's strength, in spite of the party performing outstandingly well in them. To be fair, there's always been a straightforward reason for supposing that might have been going on - in GB-wide polls YouGov use Westminster-oriented weighting by party ID.
The other point that leaps out is just how dreadfully badly the smaller parties are doing. I had assumed that the Liberal Democrats' 4% share in the last YouGov poll was a freakish result, but the chances of them being significantly underestimated by random margin-of-error effects in two consecutive polls is obviously pretty low. The jury is still out on UKIP, though, because they fared a bit better in the last poll.
Should Scotland be an independent country?
Yes 48%
No 45%
It's impossible to know the exact significance of this result until we see the datasets, or find out whether there have been any methodological changes. The last YouGov poll (which was the only other post-referendum poll from the firm to ask the independence question) produced almost identical results to this, putting Yes on 49% and No on 45%. However, it was immediately criticised by John Curtice and one or two others for not weighting by recalled referendum vote, which would have had the effect of keeping No in the lead, albeit only very narrowly. I was slightly cynical about that intervention, because I don't recall Curtice making similar unofficial adjustments to the many pre-referendum polls that failed to weight by country of birth, and which therefore underestimated the Yes vote. However, it's possible that YouGov will have heeded his complaint, in which case tonight's result is even better for Yes, because it suggests there has probably been a further swing in real terms since the last poll.
If they haven't made any methodological change (which is also perfectly possible - they may want to keep the trend figures meaningful) then it means there hasn't been a further swing, but also that the headline numbers remain directly comparable with pre-referendum YouGov polls, which in all but one (legendary) case had No ahead. So either way there is no real doubt that a significant number of voters have been converted to independence since September 18th.
Views on the Smith Commission and its proposals for further devolution :
It doesn't devolve enough powers : 51%
It gets the balance right : 23%
It goes too far : 14%
And there, in a nutshell, is the explanation for why the SNP have either maintained their advantage over Labour or increased it somewhat - it appears that they've comprehensively won the battle of perceptions over Smith. It's all very well for Michael Portillo to sit on a BBC sofa in London, smirking at Alex Salmond like an overgrown schoolboy while tittering : "The Vow has been delivered! You know that! You're just playing games!" But unfortunately for the London establishment and the Daily Record (is there a difference between the two?), it turns out that Scottish voters are not half as stupid as Portillo takes them for. They know what they heard Gordon Brown promise - and they also know that what the Smith Commission has come up with is quite simply not "Home Rule". Nor is it "near federalism". Nor is it the "Devo SUPER Max" promised by Better Together's official representative at the TV debate in the Hydro.
Assuming that YouGov only offered respondents the three options listed above, roughly 58% of people who gave a view said that the Smith proposals are inadequate. It's also fascinating to see how few people thought the proposals go too far, because that must encompass the evidently dwindling group who are opposed to devolution/self-government altogether.
I've been slightly bemused over the last few days and weeks by the number of London commentators who appear to think that Scottish Labour's problems are being caused by a "leadership vacuum" and that some sort of "honeymoon period" for Jackanory Jim is about to provide a quick fix. In reality, the Scottish media have been shamelessly treating Murphy as the unofficial leader for weeks now, so that's already factored into the polling results. To be fair, that doesn't necessarily mean that what's going on is Murphy's fault. The electorate know that Miliband is the real leader, and above all else it's him that they don't like, rate, trust, or respect.
* * *
SCOT GOES POP POLL OF POLLS
Tonight's Poll of Polls update is based on the full-scale YouGov poll, plus Scottish subsamples from five GB-wide polls - four from YouGov, and one from Populus. That means fourteen-fifteenths of the sample comes from YouGov, which is plainly less than ideal!
Scottish voting intentions for the May 2015 UK general election :
SNP 45.0% (+2.4)
Labour 26.5% (-0.4)
Conservatives 16.7% (+0.6)
Liberal Democrats 4.5% (-1.8)
UKIP 3.5% (-1.1)
Greens 2.9% (+0.5)
(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.)
No comments:
Post a Comment