As we had cause to remark only yesterday, the polls have been remarkably stable in the last couple of months. Since just before the end of March most have reported little or no change in the balance of Yes and No support. True, ICM suggested in April that Yes support had reached a record high, only for it to report a month later that it had fallen back heavily again. Otherwise no poll has indicated anything greater than a one point change in the balance of Yes and No support.
True, last week Ipsos MORI reported that there had been as much as a four point increase in Yes support (once Don’t Knows were excluded). But its previous poll had been conducted as long ago as February and had put Yes support at a markedly lower level than at any other poll conducted this year. That simply seemed to be a case of Ipsos MORI coming into line with the remainder of the polling pack.
Today, however, we have a poll that suggests the Yes side might now have made some further progress in reducing the narrow lead. In a poll for the Daily Record, Survation report that Yes are on 39%, up two on its previous poll last month, while No are on 44%, down three. Once the 17% who say they Don’t Know how they will vote (a figure that is unchanged on last month) are excluded, that equates to a 47% Yes vote, up three – and represents the highest Yes vote to be recorded by Survation in any of its polls to date.
At this point some caution is in order. There is still a fair chance that even a three point swing could be the product of the random variation to which all polls are subject. The result will certainly need to be confirmed by other polls before we can be sure that the No lead really has narrowed further. In the meantime today’s result does see the No lead narrow in our poll of polls from 16 points to 14, that is back to where it was in the middle of May.
Even so, to day’s poll cannot but produce furrowed brows at the Better Together headquarters in Glasgow’s Blythswood Square. This is the first poll to have been conducted since the UK government advised us that the Union was worth £1400 a year to every man, women and child in Scotland, and went on to illustrate the value of this bonus with the help of bricks from Lego (that subsequently had to be removed). The poll was also conducted immediately after remarks from Barack Obama that were seized upon by the No side as evidence of American hostility to Scottish independence.
These manoeuvres by the No side all came in for criticism in respect of their accuracy, wisdom and/or effectiveness. Even if we wish to reserve our position on whether there has been a significant swing to Yes, this poll hardly suggests that the No campaign’s recent tactics have proven particularly effective.
Meanwhile, the poll suggests that the Yes side’s attempts to woo women voters may finally be reaping at least some reward. Most of the swing since Survation’s last poll has occurred amongst women rather than amongst men. At 10 points the difference between the level of support amongst men (51% after Don’t Knows are excluded) and that amongst women (41%) is lower in this poll than in any previous Survation poll. This at least is another possible trend to look out for in future polls.
Doubtless the Yes side will also leap on another finding in this poll – that the prospect of David Cameron continuing to be Prime Minister could see the No lead evaporate entirely. Asked how they would vote ‘if they were certain David Cameron was going to remain Prime Minister after the Westminster General Election in 2015’, 44% said they would vote Yes while only 38% stated they would vote No. So the Yes vote is five points higher than the standard referendum voting intention figure in this poll, while the No vote is six points lower – representing a net swing of five and a half points.
This is a rather bigger net swing than that obtained by a number of other such exercises in previous months. Survation themselves found in January that the prospect of a Conservative victory was worth a three point swing to the Yes side, while ICM recorded potential swings of one point and four and a half points in April and May respectively.
One possible explanation for the somewhat higher swing this time around lies in the question wording. Whereas in this poll respondents were asked what they would do if they were ‘certain’ Cameron would remain Prime Minister, in its February poll Survation asked what they would do if the Conservatives were ‘expected to win’ in 2015. Meanwhile, ICM asked what people would do if they ‘thought the Conservatives were going to win’. These alternative formulations arguably pose something a little less than the ‘certainty’ posed in today’s poll, and thus perhaps were less likely to stimulate people into admitting that their referendum vote might be swayed.
However, whether in three months time many voters will feel ‘certain’ about what will happen another eight months thereafter remains to be seen. After all political prediction is a hazardous business!