The first EU referendum poll conducted since the Brussels deal was finalised on Friday night has already been released, and strangely it's not all that helpful. It's a Survation telephone poll, and as Survation haven't previously conducted any telephone polls during the campaign, it's impossible to reach any firm conclusions about whether Remain or Leave have made any progress as a result of the agreement, or whether the effect has been neutral. We're also not really any closer to solving the mystery of the sudden and unexpected divergence between the telephone findings of Ipsos-Mori and ComRes. The most that can be said is that if the 'house effect' of telephone polling by Survation is roughly the same as that of telephone polling by Ipsos-Mori and ComRes, and if the agreement on Friday has had little impact on public opinion (two very big ifs), then the new poll might lend some weight to my theory that Ipsos-Mori and ComRes diverged mainly because of very extreme sampling variation, and that if they had repeated the exercise immediately afterwards they would probably both have produced results somewhere in between the two extremes.
Should the United Kingdom remain a member of the European Union or leave the European Union? (Survation, telephone)
With Don't Knows excluded, that's a 59-41 lead for Remain, which is roughly similar to the advantage the Better Together campaign enjoyed among the three No-friendly pollsters with a few weeks to go until the independence referendum. The important difference with 2014, of course, is that the Leave campaign already have a small outright lead in the polls that tend to be most favourable for them (ie. online polls).
The oddest feature of the Survation datasets is that Wales, not Scotland, is the region of the UK with the biggest Remain lead. That doesn't tally up with the polling evidence we've already seen that suggests public opinion on the EU in Wales is roughly the same as in England. OK, freakish subsamples can even themselves out within a poll, but the fact that the raw Welsh sample has had to be sharply upweighted from 34 to 63 isn't ideal. It's not impossible that the sampling issues in Wales may have led to the overall Remain lead being slightly overstated.
If the rumours that Boris Johnson will declare for Leave today are true, Survation's figures will soon look out of date anyway, and we'll start looking ahead to the first post-Boris poll. I know that very few people who read this blog rate the guy, but if he actually fronts the campaign and represents it in high-profile TV debates, I think that could make a huge difference and potentially swing the balance. Apart from charisma, he brings a bit of optimism and sunshine to the table - the six Tory cabinet ministers who came out for Leave yesterday look like such a dreary bunch in comparison.
* * *
SCOT GOES POP POLL OF POLLS
Should the United Kingdom remain a member of the European Union or leave the European Union?
50/50 ONLINE/TELEPHONE AVERAGE :
Remain 46.7% (-0.8)
Leave 39.2% (-0.7)
ONLINE AVERAGE :
Remain 41.4% (-0.6)
Leave 41.9% (-0.5)
TELEPHONE AVERAGE :
Remain 52.0% (-1.0)
Leave 36.4% (-0.9)
(The Poll of Polls takes account of all polls that were conducted at least partly within the last month. The online average is based on nine polls - four from ICM, two from YouGov, one from BMG, one from ORB and one from TNS. The telephone average is based on five polls - two from ComRes, two from Ipsos-Mori and one from Survation.)