As well as the YouGov and Opinium national polls, there were also two Scottish referendum polls. Regular readers will recall that ICM’s Scottish poll a week ago showed an interesting narrowing in the race, so the question is whether this will be repeated elsewhere.

Survation’s first referendum poll in the Mail on Sunday had topline figures of YES 32%, NO 52% and don’t know 16%, very similar to the recent YouGov and Ipsos-MORI polls. Given it’s Survation’s first Scottish referendum poll we obviously don’t have any changes from last time.

More interesting are the figures from TNS-BMRB which have YES on 29% (up 2 points since December), NO on 41% (unchanged). The change is small in isolation, but looking at the broader trend from TNS there does appear to be a gradual increase in Yes support. In August they has Yes on 25%, October on 25%, November 26%, December 27%, now 29%.


There are three YouGov polls out today, Britain, Scotland and Wales. Starting with the regular daily poll for the Sun, GB voting intentions stand at CON 34%, LAB 39%, LDEM 9%, UKIP 12% – a five point Labour lead, the third in a row from YouGov. Full tabs are here.

There was a second YouGov poll in the Times, this one a Scottish poll on referendum voting intentions. YouGov have the YES vote at 33% (up one point since September), the NO vote at 52% (no change). Excluding won’t votes and don’t knows the figures are YES 39%, NO 61%. This is the first YouGov poll since the independence white paper and clearly shows no significant change in referendum voting intentions. John Curtice has a nice summary of the three post-white paper polls we’ve seen so far on his blog here – a little narrowing in the lastest wave of polls, but “a touch on the tiller, rather than a game changer”. Full tabs for the YouGov poll are here.

Finally there is a new YouGov Welsh poll for ITV Wales and the Wales Governance Centre at Cardiff University, the first of a regular series of Welsh voting intention polls. Welsh voting intentions are:

Westminster – CON 21%, LAB 46%, LDEM 8%, Plaid 12%, UKIP 10%
Welsh Assembly constituency – CON 19%, LAB 43%, LDEM 9%, Plaid 20%, UKIP 7%
Welsh Assembly regional – CON 19%, LAB 40%, LDEM 9%, Plaid 15%, UKIP 10%

No changes from the previous poll as YouGov have changed how they prompt their Welsh assembly polls, as we explored yesterday. More generally the Westminster figures represent a 7.5 point swing from the Conservatives to Labour since the general election (not wildly dissimilar from the GB national picture), while the Welsh Assembly figures suggest an improvement for UKIP, but not a vast change for the other parties – if repeated at a Welsh assembly election Labour would retain the same number of seats they won in 2011, just short of an overall majority. Full Welsh tabs are here.


-->

We finally have a Scottish post-white paper poll from a BPC member: Ipsos MORI have released their regular Scottish public opinion monitor, conducted in the week following the publication of the white paper.

Amongst those certain to vote referendum voting intention is YES 34%(+3), NO 57%(-2). Changes are from the previous MORI poll in September. This represents a slight shift towards the YES campaign in the last three months, though given MORI tend to show one of the largest leads for the No campaign this still leaves them with a 23 point lead.

If the same trend is repeated in other post-white paper polls from companies that tend to show a tigher race we could see some interesting polls. Then again, this is just one poll and the changes are within the margin of error, so we may very well find other polls don’t show the same sort of movement at all. Time will tell.


TNS BMRB released a new poll on the Scottish independence referendum this morning. I expected several polls to appear in the wake of the publication of the white paper, letting us see if it had any effect on referendum voting intentions.

This alas is not one of them, as annoyingly it was carried out almost wholly before the white paper was published. For the record the figures show very little change from the previous TNS poll in October. The YES vote stands at 26%(+1), the NO vote at 42%(-1), 32% are undecided (for some reason TNS tend to show a much higher level of don’t knows compared to other Scottish referendum polls). Full tabs are here.


Tuesday round up

There are a few interesting bits of polling news today. First there is a new chunk of Lord Ashcroft polling, this time on the Conservative’s position in Scotland. Full results are here. I won’t summarise the whole report here, but essentially he segments up the Scottish electorate and as well as that poor sorry rump of Tory support left in Scotland, he also finds a group he calls “reluctant Cameroons” – primary Scots who approve of Cameron, trust the Conservatives on the economy… but don’t vote for them because they don’t see the Conservatives as caring about Scotland and view the party as irrelevant to Scottish politics, or a wasted vote. Therein lies the Conservative problem not just in Scotland, but in much of the urban North too. There are people with some sympathy towards Conservative policies, but they live in places or communities where voting Conservative is simply not done, no one else does it, there’s no point doing it, there’s no longer a recent history of it, what would be the point of it? It’s something people in the South do.

Anyway, I’ll leave you to read Lord Ashcroft’s report for yourselves, but for the record it also contained Westminster voting intention figures for Scotland, concucted earlier this month. CON 18%(+1), LAB 40%(-2), LDEM 6%(-13), SNP 31%(+11), UKIP 2%(+1). Changes are from the 2010 election and reflect a big swing from the Lib Dems to the SNP. If it was repeated as a uniform swing across Scotland the Lib Dems would be reduced to three seats in Scotland, the Tories would gain two seats, Labour would gain two and lose one, the SNP would go up to 11 seats.

Secondly there is the regular YouGov poll for the Sun. Topline figures today are CON 31%, LAB 40%, LDEM 9%, UKIP 12%. While it’s within the normal margin of error of YouGov’s recent polling, the nine point lead is the largest YouGov have show since the start of October. My hunch is that this particular poll is probably a bit of an outlier, but that the issue of energy prices coming back into the news agenda following the energy price rises has boosted Labour’s lead a bit. Full tabs are here.

Finally the Electoral Commission have issued advice on the referendum question contained in the referendum bill currently before the Commons. The Bill currently contains the wording “Do you think that the United Kingdom should be a member of the European Union”. The Electoral Commission have recommended that the “do you think” bit is dropped, so the question is shorter and more formal, and that the wording reflects that the UK is already a member of the EU, as some people thought the question read as if it was whether Britain should join the EU. As such the question would become “Should the United Kingdom remain a member of the European Union?” They’ve also floated the idea that it might be better to move away from a Yes/No question, and instead have a Remain/Leave question, along the lines of “Should the United Kingdom remain a member of the European Union or leave the European Union? Remain a member of the European Union/ Leave the European Union”.

As a pollster you tend to get asked questions about referendum wording. It makes some sense, as writing fair questions is the bread and butter of being a pollster, but in many ways the considerations really aren’t the same. As a pollster I hardly every write questions with just a straight Yes/No as options because there is a fear of affirmation bias, so as a polling question the Electoral Commission’s Remain/Leave is definitely better, giving both sides of the campaign equal prominance. However, it’s NOT a polling question, it’s a referendum question. With a polling question, people are rung up out of the blue (or get an email out of the blue) and get a few seconds to answer the question – those small differences in wording undoubtedly make a difference. In a referendum people have weeks to decide, and will be influenced by the whole campaigns, personalities, arguments, advertisements and so on. What the Yes and No votes mean for the country is something that voters will form their own perceptions of long before they enter a polling station. In that sense, as long as the question is clear and unambiguious, I doubt whether it says yes/no or remain/leave matters much.