Thursday, January 23, 2014

Professional buffoon Ian Dunt thrills the nation with another vintage gaffe, as he suggests that a poll showing a surge in support for independence is bad news for Alex Salmond

Ian Dunt, fondly known to millions across Scotland as "the thinking woman's Richard Madeley", has scaled new heights of idiocy by posting an article on bearing the title "Major new Scottish independence poll brings fresh despair for Alex Salmond". To be fair, he is indeed talking about a "major" poll, and it's a "new" poll in the sense that it's only just been released (albeit the fieldwork is several months out of date, meaning that it's already been superseded by at least nine referendum polls that have been conducted more recently). But the slight snag is that the poll shows an increase of 6% in support for independence. Yes, that's right, I said "increase", not "decrease". And this is the news that is supposed to have left Mr Salmond in a state of despair? Presumably Dunt has already got his headline of "Heartbreak for Salmond as Scots only vote for independence by quite a narrow margin" all raring to go for September 19th.

You might be amused to learn that describes itself, without any discernible trace of irony, as "an impartial political website with no political affiliation, which prides itself on standing out as an independent voice in a landscape where the norm is to nail your colours to the mast". I was going to issue the editor with a friendly hint that getting an openly Nat-bashing hack like Dunt with a proven track record of boastful ignorance on Scottish matters to pen a "news" article like this might not be the ideal way of burnishing the website's desired image of studied neutrality. But it turns out that Dunt actually IS the editor. The mind boggles. Who in God's name is the deputy editor of this oasis of impartiality - Nigel Farage?

* * *

To turn to the Scottish Social Attitudes Survey itself, the headline figures on independence (using the traditional multi-option question) are hopelessly tainted by numerous methodological problems, not least the pejorative wording about an independent Scotland being "separate from the UK". All it needed was the addition of the word "entirely" (as in "entirely separate from the UK") and it would have been Alan Cochrane's idea of polling heaven. Another flaw is the suggestion in one of the options that an independent Scotland might be outside the EU, whereas there is no balancing option to take account of the fact that Scotland may be forced out of the EU if it remains part of the UK - and possibly within quite a short timescale. There is also no Devo Max option offered, with respondents effectively forced to choose between independence, Devo-current, Devo-minus, and the outright abolition of the Scottish Parliament. Given that we know from the results of another question that the public would be likely to vote "Yes" in a referendum on Devo Max by a margin of almost 2-1, it's impossible to make sense of the headline results showing a significant plurality for Devo-current - after all, Devo Max is far closer to independence than it is to Devo-current. The contradiction can only really be put down to the misleading and near-obsolete wording of a question that ought to have been abandoned years ago. So all we can meaningfully take from the headline figures is the trend, namely a 6% increase in support for independence, a 6% decrease in support for devolution, and a 2% decrease in support for abolishing the Scottish Parliament.

It's obviously a good thing that the actual referendum question was posed this time around, but we can't really deduce very much from that result either, as there are no baseline figures from previous years to work with. We also know that the referendum question was only asked after the traditional multi-option question and a second tortuously-worded multi-option question about where governmental powers should lie. And while I don't want to be unkind to Professor John Curtice (he isn't Ian Dunt), you don't have to look far for the double standards when he analyses research that he was directly involved in. It's only a few months since he was scathing to the point of absurdity about a perfectly credible Panelbase poll showing a narrow lead for the pro-independence campaign, solely on the grounds that the referendum question was put to respondents third, rather than first. Well, he's openly admitting that his own SSAS poll posed the referendum question at least third, and quite possibly much later in the sequence - so where's his equivalent disclaimer that the results can't be taken seriously? It doesn't seem to be forthcoming so far.

Perhaps the most spectacular finding of the survey is that the anti-independence campaign's cynical attempts to sow doubts about an independent Scotland's future within the sterling zone have catastrophically failed - a full 57% of respondents expect that Scotland will still be using the pound a few years after independence, compared to just 21% who think it will be using the euro, and 16% who think it will be using its own currency.

Encouragingly, the figures on national identity show a modest move towards Scottishness...

More or wholly Scottish : 54% (+1)
More or wholly British : 10% (-1)
Equally Scottish and British : 29% (-1)

That's the good news, but the bad news is that this only very slightly checks a long-term trend away from Scottishness - until 2006, the figure for a wholly or predominantly Scottish identity was routinely well above 60%. It's not entirely clear why this apparent change has happened - as Scottish Skier always points out, national identity isn't normally a changeable thing like voting intention. It may simply be that people who would previously have said they were predominantly Scottish are now telling pollsters something different due to changing fashion, rather than any real change in how they actually feel. That theory is at least partly borne out by several other polls that have found that the "equally Scottish and British" group are by no means a no-go area for the pro-independence campaign, which you would certainly expect them to be if the 'best of both worlds' rhetoric had any real validity. It seems that at least some people are intelligent enough to realise that if it's possible to feel Scottish without needing a sovereign Scottish state, it must by definition be equally possible to feel British without needing a sovereign British state.

Of course, the other possible explanation for the change in the national identity figures is that some elderly "Scottish" people have died off since 2006 and been replaced by "British" young people. But regardless of Jan Eichhorn's wild theorising about the impact of Facebook, that is highly implausible - there is no credible evidence at all that young people are more likely than their elders to feel British, and if anything the reverse is more likely to be true.


  1. You seem to be saying that if you dare to offer a divisive opinion on political matters then you should not claim to be an "independent voice...with no political affiliation" - that's rather like saying a judge should not be allowed to give his opinion on a legal matter unless he admits to having a biased perspective. Quite perverse

  2. "A pro-independence blog by James Kelly - voted one of Scotland's top 10 political websites"

    Hardly claiming to be an 'Independent voice"?

    Is this the best you can come up with?

  3. Apologies I picked you up incorrectly.

  4. War is peace. Freedom is slavery. Ignorance is strength. An increase is a decrease.

  5. I don’t understand why you attribute so much weight to the Merono data. The question is a useful enquiry as to how people might reconcile their duel identies, but only that matter. Ask your self this: who when on foreign holiday answers a question as to their their identity or origin with: ‘I am more Scottish than British’. I don’t mean to attack the Merono question or data. It is useful in coming to an understanding of the complexities surropunding Scottish/British identity. However, it is most certainly not itself an identity question, nor even a surrogate for an identity question (however much the polling community seems to accept it as such).

    Of much greater interest, is that the Scottish Social Attitudes forced choice identity question result is much more closely in line with the identity results from the Census. In regard to that data, I wonder that you say that there is a ‘long-term trend away from Scottishness’. Taking the results from that forced choice identity question from 1999 to 2011 the changes are within or around the margin of error for a poll that size. There is certainly a reduction those claiming a in Scottish identity in 2012-13 data sets (which incidently brings it more in line with the Census result), but this is offset by an increase in those claiming a British identity that is substantially out of line with the Census data.

    Remember, in the game of polling poker the Census is the Royal Flush! It is an enquiry of every household and and evey living sole. Its data and results, therefore, are more authoratitive then even elections. It would, therefore, be nothing short of ridiculous were anyone to question the Census data on the basis that it doesn’t agree with a poll. Further, I think that for reasons of his obvious vested interest we may cast a jaundiced eye on Curtice’s critic of the Census identity question as being leading – it is certainly no more leading that his forced choice identity or Merono questions. So, unless there is reliable data that points to there having been a substantial shift in opinion in the period between the 27th of March 2011 and the period of the Scottish Social Attitudes fieldwork (and I know of no such data) we must ask some serious questions of the Scottish Social Attitudes survey methodology. In particular, we must ask whether their sampling (I believe that it is a stratified, clustered sample) is skewed toward persons with a British identity.

  6. Hamish, you've got James's point completely wrong. James is quite right to say it's misguided to ask someone who manages a survey to comment on the survey as if they were independent of it. There are very simple and straightforward reasons for not doing that. Curtice is of course a noted expert in this field and his analyses of other polls and surveys provides valuable commentary on the subject. But asking him to analyse his own survey critically is never going to yield useful information. At the very least there will always be the suggestion that such an analysis is compromised by the proximity of the work to the person carrying out the assessment.

  7. Hamish : If you're asking me whether I think that Dunt's SNP-bashing and Salmond-bashing antics are inconsistent with his site's grandiose claims of impartiality, then you're damned right I do. Some journalists are perfectly capable of setting aside their own prejudices when writing news pieces - Dunt is demonstrably not. The idea that any journalist writing from an impartial and objective viewpoint could characterise a survey showing a 6-point rise in support for independence as bringing "fresh despair for Alex Salmond" is absurd beyond words (and please note the personalisation of the independence movement in that headline - one of the subtle biases picked up by the UWS report on broadcast news).

    Alasdair : I take an interest in the Moreno data because we know it correlates to a reasonable extent with voting intention. Those who say they are "Scottish not British" or "more Scottish than British" are far more likely to be open to voting Yes, so obviously the more of them there are the better.

  8. Michael: I think you misunderstood me, I was referring to the attack of the website.
    James: I suppose it sounds a bit arrogant for someone to claim true impartiality as I'm not aware of any sensible framework for measuring your "level of impartiality". You could subscribe to the self imposed framework that mainstream news channels work to, where they take impartiality to mean "giving the opinion that offends the fewest number of people". Resultingly they video lots of things but don't actually say anything. Thankfully is not a follower of that approach and looking at a cross section of its content it is easy to see that it does not have any particular party allegiances. I'm not saying I agree with that article btw...

  9. I don't find the reported shift away from a Scottish self-identity all that surprising. People may think of themselves as Scottish or non-Scottish along several different dimensions - ethnic, cultural, political, sporting, etc. - and these dimensions may correlate more or less closely.

    In recent years, particularly since the SNP came to power in 2007, Scottish identity has become much more politicised. It has become harder to think of yourself as Scottish without facing the independence issue. So it would not be surprising if some 'ethnic', 'cultural' or 'sporting' Scots have become more reluctant to call themselves Scottish tout court.

  10. "Thankfully is not a follower of that approach and looking at a cross section of its content it is easy to see that it does not have any particular party allegiances."

    Well, they do claim impartiality, which is a much tougher standard than simply being non-aligned in a party political sense. It's perfectly possible to have no party allegiance and still be violently anti-independence, for example. I'm not a regular reader of the website, so I don't know whether they ever allow pro-independence commentary, but if their coverage of the subject consists solely of southern commentators like Dunt making constant digs about how unlikely or undesirable independence is, then self-evidently they do have a severe problem with political bias. Whether they're sufficiently self-aware to recognise that problem is a different matter, of course.

  11. Sorry, Hamish, I did misunderstand you.

  12. I'm guessing the threat of indepdence will be making those against it feel more British.

  13. I believe the unionists are starting to get worried as they know the gap is progressively closing. Check out the archives 9/2 now 4/1.

    Odds past and now.

  14. Given Dunts recent comments on the attack in Paris this mans name is one letter away in the alphabet from a true description of him. Forgive me if this has been pointed out before. He's just another Guardianista ----hole.