The results of a second (and voluminous) Wings Over Scotland poll are gradually appearing on the Wings website during the course of today. For now we will confine our comment to the results of two questions that were published in advance in this morning’s Sunday Herald. A fuller blog will appear tomorrow, by which time the full set of results will be in the public domain.
The poll has been conducted by Panelbase using the much the same method and approach as in their regular polls for The Sunday Times. And this poll’s headline referendum vote intention figure is much the same as in those previous Sunday Times polls.
Amongst who say they are at least 8/10 likely to vote, 37% say they would vote Yes, 45% No, while 17% are undecided. That represents no change in the Yes vote as compared with the last Sunday Times poll (conducted in early September), while the No vote is down two points and the undecided are up by one point.
If we take the undecided out of the calculation, the poll points to a 45% Yes vote, with 55% backing No. All four previous polls of referendum voting intention undertaken by Panelbase for The Sunday Times have emerged with either a 44% or a 45% Yes vote. In short, once again we have a poll that suggests that the balance of forces in this campaign remain remarkably stable.
At the same time, what also remains true is that Panelbase are reporting more optimistic results for the Yes side than any other pollster. But this latest exercise does help to dispel some of the possible reasons why this might be the case.
Some concern has previously been expressed (though not by this site) that perhaps the panel of people who have signed up to undertake Panelbase’s polls has been infiltrated by overzealous ‘cybernats’ who were skewing the company’s results. That prompted Panelbase to protect itself from any such risk by excluding from their Scottish opinion polls anyone who had joined their panel since June this year.
Now in this poll Panelbase have tried to assess whether their panel is disproportionately pro-Yes in composition by conducting half of the poll amongst members of their own panel and half amongst those who are members of another (unnamed) panel. The balance of Yes and No supporters in the two halves was identical (though the non-Panelbase panel respondents did contain a higher proportion (20%) of Don’t Knows than the Panelbase ones (15%)). That would seem to dispel the concern about infiltration.
Meanwhile, we should note too that in this poll the question on referendum voting intention was not preceded by questions about Holyrood vote intention. We amongst others had previously speculated that perhaps having just said they would vote for the (relatively popular) SNP, this practice encouraged some respondents to say they would vote Yes too. In truth this was already coming to seem unlikely – it cannot account for the fact that Panelbase’s polls are now also clearly notable for having more SNP supporters too. And now the company has produced much the same result even when it has not asked Holyrood vote intention first.
With these two issues apparently put to bed, it now seems very likely that what distinguishes Panelbase’s polls from those of other companies is the character of the sample that emerges from the procedures they use to select potential respondents. This aspect of their methodology – and that of their rivals – may well come under greater scrutiny should the pollsters’ results continue to diverge as markedly as they have done so far.
Meanwhile the poll also uncovers some intriguing differences in the motivations (as reported by respondents themselves) of Yes and No voters. Amongst Yes supporters the most common factor influencing their vote (selected by 62% of likely voters) is ‘ensuring Scotland always gets the government it votes for’. In contrast, just 7% of No voters are concerned about this issue. The most common concern of No supporters, cited by 56%, is ‘the future of the economy’, a motivation that only 35% of Yes voters share.
And what of the Don’t Knows whom the Yes side badly need to win over? Just 24% are concerned about Scotland always getting the government for which it votes, while no less than 65% say their eventual decision will be determined by the future of the nation’s economy. The Yes side needs to remember that arguments that make nationalist hearts beat faster may not necessarily be ones that have resonance for those who have yet to be won over.