The headline results of the two polls published today, one by ICM Research for Scotland on Sunday and one by Survation for the Sunday Post, will doubtless be regarded as encouraging for the Yes side. Some of the details too will do little to dispel the impression that the No campaign is a campaign in trouble. But underneath the bonnet there are also signs that the No vote may have become a little more resilient.
Both polls were conducted earlier last week, immediately after the conclusion of the SNP conference – an event that doubtless the Yes side was hoping would give it a boost.
ICM’s poll suggests that could well have happened. It puts the Yes vote on 39%, No on 42%. That represents a four point fall in the No vote as compared with ICM’s previous poll a month ago, although the Yes vote is unchanged. Once the Don’t Knows are excluded the Yes vote stands at 48%, three points up on the company’s previous poll and two points above its previous highest reading. Indeed, one much criticised poll for the SNP apart, that 48% figure is the highest Yes vote yet to have been recorded by any poll (once Don’t Knows have been excluded) during the course of the referendum campaign.
Survation’s reading is not quite so good for Yes. It estimates the Yes vote at 38% and No at 46%. That represents a one point increase in the Yes vote and a one point drop in the No vote as compared with the company’s previous poll just ten days earlier. At 45% the Yes vote, once Don’t Knows are excluded, is also up one point. However, apart from the fact that such changes are too small to be statistically significant, that 45% figure is exactly in line with Survation’s readings in March and February. In short this poll looks more like further evidence that the Yes side’s progress during the winter is holding up, but that, as quite a few other recent polls have suggested, it has yet to advance further.
Here, by the way, we should note a change ICM have made to their weighting scheme. One of the potential problems with internet polls is that they find it more difficult than face to face or telephone polls to make contact with those who are less likely to vote. Indeed the company noted that in their previous referendum polls fewer people were saying they did not vote in the last Holyrood election than were saying in their British polls (for The Guardian) that they did not vote in the last Westminster election. Yet, of course, in reality fewer people voted in Scotland in 2011 than did across Britain as a whole in 2010.
Bearing in mind that it looks as though many people who did not vote in 2011 will do so in 2014, ICM have thus decided that to weight their samples so that not only are the proportions of people who say they voted Conservative, Labour, etc. in 2011 more or less in line with the actual outcome, but so also is the proportion who state they did not vote (or cannot remember whether they voted). This is not dissimilar to the practice already adopted by TNS BMRB. One of the consequences is to boost somewhat the proportion who say they do not know how they will vote, and what is a four point increase in that proportion as compared with ICM’s previous poll needs to be read in that light. Crucially, however, the increase in the Yes vote registered by the poll (once the undecided are removed) is not an artifact of the methodological change.
Behind the increase in the ICM poll lie signs of continuing progress by the Yes side in persuading people of the economic case for independence. At 37% the proportion that think that independence would be good for Scotland’s economy is now only four points less than the proportion who think it would be bad. Last month that gap stood at five points, in February at nine and last September at no less than 17. Given how close the link is between people’s views of the economic consequences of independence and their willingness to vote for it, this is a trend that must be a source of concern for the No side.
The No side’s campaign has been criticised in recent weeks for being too negative, thereby perhaps creating a risk that its warnings about the alleged adverse consequences of independence are no longer being believed. There is evidence in both of today’s polls that is consistent with that contention.
First ICM report that only 22% think that the Better Together campaign paints a positive picture of Scotland’s future, while 35% believe it presents a negative picture and 26% neither positive nor negative.
Second, Survation report that when voters are asked to mark the two campaigns separately on a score from 0 (overwhelmingly negative) to 10 (overwhelmingly positive), as many as 46% give the Yes campaign a score of seven or more, whereas only 25% mark the No campaign’s card that highly.
Third, and perhaps most importantly, fewer voters trust the utterances of the No campaign than do those of its opponents. Asked to mark each campaign on a score from 0 (completely untrustworthy) to 10 (completely trustworthy) only 30% give the No side a mark of seven or more, while almost as many, 29%, give it a score of three or less. In contrast, 40% give the Yes side a mark of seven or more, though the proportion who give it a score of three or less (30%) is much the same as for the No side.
This contrast is confirmed when voters are asked to compare the trustworthiness of the two sides directly. While 29% say they trust the Yes side more, only 21% say the same of the No side. Mind you, there is also a reminder here of voters’ scepticism about both sides – as many as 39% say that neither side is trustworthy, a feeling that is particularly widespread amongst undecided voters.
However, the rather better news for the No side is that those are inclined to vote to stay in the Union appear to have become a little more determined in their support. According to ICM, as many as 89% of No voters have definitely made up their mind, compared with only 78% of Yes supporters. The former figure is up seven points on the equivalent figure for last month, whereas the latter is only up one. The difference between the two fgures of eleven points compares with one of just three points in February.
Moreover, despite widespread scepticism that more powers will be devolved to the Scottish Parliament in the event of a No vote (only 33% believe it would happen, seemingly another sign of the No side’s credibility problem), those No voters who would like more powers (who comprise some 38% of all No voters) are now less likely to say they would switch sides if by September they were to be convinced that more powers would indeed not be delivered. Only 4% now say they would do so, down from 10% last month, and similar proportions in January and February. Mind you, as many as 16% say they are unsure what they would do in those circumstances, so the No side cannot presume it does not need to worry about what voters think of the prospects for more devolution. After all, despite their differences, both of today’s polls point to a race that continues to look much closer than it did just a few short months ago.