Monday, January 27, 2014

SNP enjoy huge 19% lead over Labour in ICM poll

Further details of this weekend's extraordinary ICM poll have been released by the Scotsman. Unfortunately, however, the party political voting intention figures raise more questions than they answer, because the newspaper's report on the poll is a masterclass in ambiguous language. We are told that "the Liberal Democrats have slumped to fifth place in Scotland ahead of this year's European elections", but what does that actually mean? Is it meant to imply that this is a poll of European Parliament voting intentions, or merely that this is a Holyrood poll that just happens to have been conducted in the run-up to European elections? That isn't a point of pedantry, because we know that people vote very differently in different types of elections (they even vote differently on the Holyrood list ballot than they do on the Holyrood constituency ballot). Doubtless the mystery will be solved when ICM post the datasets, or hopefully even before then, but for now all I can do is give the figures without really having a clue how to interpret them.

SNP 43%
Labour 24%
Conservatives 14%
Liberal Democrats 6%

I think the main thing to take away from this is that any SNP supporter attracted by the siren voices from the Green corner suggesting that a tactical vote is the way to prevent UKIP winning the sixth European Parliament seat should forget the whole idea. On these figures, it would be the SNP gaining the sixth seat from the Liberal Democrats, with UKIP and the Greens not really in contention at all. Lallands Peat Worrier has already explained in considerable detail why a tactical vote is likely to backfire (or to have no effect) in this post. But a more fundamental point is that it isn't just the seat breakdown that matters - at the last European elections the SNP and Labour both won two seats, and yet there was no doubt in anyone's mind that Labour had been gubbed, because the SNP had handily beaten them on the popular vote. So misguided tactical voting wouldn't just endanger the SNP's chances of an extra seat - it could also cost them bragging rights on the popular vote, which may prove crucial in terms of momentum going into the referendum. Because frankly I don't expect them to be 19 points ahead of Labour on European polling day.

Which brings me to my other point of concern. Perverse though it might seem, I would actually have been more reassured if the SNP's lead had been more modest in this poll, because it would have made the spectacular referendum voting intention figures from last night look more robust. As it is, ICM seem to be contradicting recent polling evidence from Ipsos-Mori and YouGov, both of which suggested a pro-Yes swing coupled with a simultaneous decline in the SNP vote, in line with what appeared to be happening in Cowdenbeath. So this raises the slight worry that this may yet prove to be a rogue poll, ie. not within the standard 3% margin of error, which is something that pollsters acknowledge happens one time in every twenty - and in reality happens far more often than that due to methodological shortcomings. All we can do is keep our fingers crossed for the second full-scale poll of referendum year, and hope it's just as good as this one.


  1. There should be some internal evidence from the data tables. Since the 5% change in referendum VI apparently occurred among the under 45s, then this change in Euro election VI (assuming that this is what it is) should also be restricted to that age group.

    Otherwise, Curtice's caveat may be of little import - and he certainly didn't mention an age factor in the Scotsman article.

  2. I'd say this is for the European Elections, the Scotsman have just published Scottish figures for an EU referendum.

  3. Callum

    That article also shows that Murdo Fraser is an idiot. I understand why Unionists want to peddle the idea that Scots political views are the same as those in England. However, to have any credibility the numbers would need to back him up, not contradicting him.

  4. oldnat,

    As Stu Campbell put it "These stats are dramatically different, so they are the same".

    As for the referendum poll, it'll have been weighted by past vote so any strange current voting intention figures shouldn't be an issue.

  5. James, I see you are still the cautious optimist.

  6. Tables up:

    Question was: "The Conservatives, Labour, Liberal Democrats, SNP, UKIP, the Scottish Greens and other parties and independent candidates w
    ill fight the European Parliament elections in your
    area. In the European Parliament election in May 2014, which party do you think you will vote for?"

  7. Interesting that in the past two days none of the pundits have said the ICM poll was a rogue. It must be what the internal polling has been indicating.

  8. Spot on Marcia. Darling's behaviour has indicated that their private polling has been trending against them.