Survation Change Methodology – But Still Find the Battle for Second is Tight

The latest poll of voting intentions for Thursday’s election, conducted by Survation for the Daily Record, is of particular interest for two reasons. First, in contrast to Survation’s previous polls it was conducted by phone rather than via the internet. Second, the wording of the question used to ascertain voting intentions on the list vote is markedly different from the one that the company has used previously.

Despite these methodological differences, today’s poll shares one feature in common with Survation’s previous online poll – Labour and the Conservatives are reckoned to be neck and neck. This time, Labour are estimated to be two points ahead in the constituency vote (last time the figure was one point) but  a point behind on the crucial list vote (exactly as in Survation’s previous poll) – figures that imply that the Conservatives are more likely to emerge with the larger number of seats. The fact that Survation have replicated the finding that the battle for second place is very close even though this time it has used a very different approach in its polling lends considerable weight to the conclusion that the outcome of that battle is far from certain – despite Panelbase’s suggestion at the weekend that Labour might have pulled ahead somewhat.

What, however, Survation have not replicated from its previous poll is its unusually high (11%) figure for the Greens on the list vote. Indeed, today’s figure, 7%, is lower than that recorded by the company for the Greens in any of its polls in the last two years. This drop is almost undoubtedly a reflection of the second methodological difference between this poll and Survation’s previous polls.

We have pointed out before in this blog that one possible explanation for the relatively high level of support for the Greens in the company’s polls was the wording that it used to ascertain how people might vote on the list. This read:

‘Your second vote will be a party list vote to elect representatives from your region of Scotland by a form of proportional representation. If the election were tomorrow, which party would you be most likely to vote for with your second, regional list vote?’

The repeated use of the word ‘second’ seemed to carry with it the risk that some respondents might think they were being asked for their second preference rather than how they would vote on a distinct and separate ballot. Moreover, the proportion of people who said they would vote differently on the list vote than on the constituency vote has tended to be higher in Survation’s polls than in those of other companies.

In today’s poll, however, Survation asked:

‘Your other vote will be a party list vote to elect representatives from your region of Scotland by a form of proportional representation. If the election were tomorrow, which party would you be most likely to vote for with your regional list vote?’

The word ‘second’ no longer makes an appearance. Meanwhile, we now find that considerably fewer respondents now say that they will vote differently on the two ballots. In today’s poll, 15% of those who say they will vote for one of Conservative, Labour, Liberal Democrat or the SNP on the constituency vote say that they will vote for a different party on the list vote. In Survation’s previous online poll the equivalent figure was as much as 27%.

The principal beneficiaries of the reported willingness of respondents in Survation’s previous polls to vote differently on the two ballots were the Greens. So the fact that fewer respondents in today’s poll say that they will vote differently on the two ballots almost inevitably means that it has reported a lower estimated level of support for the Greens.

At 7% Survation’s estimate of the level of support for the Greens is, however, above the 5-6% level that a party needs to secure before it is likely to pick up a list vote in a region. So even if this lower reading for the party is correct, it would still be expected to win a list seat in most of the eight regions into which Scotland is divided for the purpose of allocating list seats. But with the Liberal Democrats reckoned to be on 8%, today’s figures imply that the battle for fourth place at Holyrood may be much tighter than some polls have previously suggested.

There is one other feature of the results of this poll that should be noted. At 49% support for the SNP on the constituency vote is four points lower than in Survation’s previous poll (whereas support for both the Conservatives and Labour is higher). Indeed, it is more than a year since Survation have put the nationalists below 50% on the constituency vote.

It follows upon a similar slip in SNP support in Panelbase’s poll that was released on Sunday. In that case the drop in support for the nationalists appeared to be most marked amongst those who did not vote Yes in the referendum. There is some sign of that pattern in this poll too. While, at 86%, support for the SNP is two points lower amongst those who said they voted Yes in September 2014 than in Survation’s previous poll, it is, at 16%, five points lower amongst those who voted No. It is certainly not obvious that it has been in the SNP’s interests for the referendum debate at the tail end of the election campaign to have focused so much on the possibility of a second independence referendum.  That, after all, is the one thing that most No voters do not want.

Topics: Elections, parties & leaders

About the author

John Curtice is Professor of Politics at Strathclyde University, Senior Research Fellow at ScotCen, and Chief Commentator on the What Scotland Thinks website.