Recent polls

After a quiet patch in early August, the past few days have brought a flurry of opinion polls, as editors desperately try to generate some news during the silly season.

There has been much comment across the blogosphere on yesterday’s ICM poll for The Guardian, which gave the Tories a 9-point lead. The first point to make is a pessimistic one from The Daily’s point of view – though this may be a blip, most polls under-estimate the Tory lead. 

ICM over-estimated the Labour lead by 3 points in 2005 and even more in 2001. Furthermore, simple “voting intention” polls currently show lower Tory vote shares than those polls which mention the party leaders – the nominal Cameron-Brown lead has actually been about 9 points since May. This lead may now be even higher.

On the plus side, it is worth rebutting one myth – Labour has fallen to 31% since 1997, in a September 2003 ICM poll, and recovered.

It has been suggested that the poll reflects badly on John Reid’s popularity, but this weekend’s YouGov tracker poll, which got little press coverage, actually shows that his approval ratings have shot up, having initially plunged after he took over at the Home Office.

He and Hilary Benn are now the only politicians more popular than David Cameron. However, Reid’s figures are volatile and likely to dip again, while Benn will have far lower recognition. Tony Blair, incidentally, is now at his lowest ever YouGov approval rating of -37.

Better news for Gordon Brown comes from the weekend’s Ipsos-MORI poll. The Daily has mentioned its findings on immigration elsewhere, but it also included a question on who would make the best PM from Cameron/Brown/Campbell. Brown maintains his consistent lead, currently 7 points, though the Sunday Times neglected to mention that in the story on their poll. It is also worth noting that the Tory lead on immigration has dramatically shrunk since 2005, with a majority of the public reverting to the opinion that no party has the best policy.

Of course, all polling comes with the health warning that it is often misunderstood and sometimes just plain wrong – but given their use as political ammunication, The Daily will endeavour to get behind the headlines and report the facts.


7 Responses to Recent polls

  1. Question says:

    How accurate were ICM at the local elections? They may have improved their techniques since 2005.

  2. That’s difficult to say, because ICM only did national polls and of course the elections did not occur everywhere – in particular, not in Wales and Scotland which is likely to depress the Labour vote.

    But, for what it’s worth, the national vote shares in May were Con 40 LD 27 Lab 26 whereas the final ICM tracker poll was Con 34 Lab 32 LD 24. This was the lowest Tory lead of any of the immediate pre-election polls.

    One other consideration is that this poll was on 23 April, which if memory serves was before the events of “Labour’s Black Wednesday” which hit Labour in all the polls.

    So I would be reluctant to damn ICM on that basis, but at the same time there’s no evidence that they have corrected their Labour bias either.

  3. Al says:

    Actually Labour’s share of the “national vote” in the local elections was higher than the BBC’s numbers; more around 28% or 29% IIRC. Either way it’s kinda irrelevent; local elections are local elections. They are very interesting in many ways, but not for PV stuff (especially when the Valleys or Clydeside aren’t included).

    At the moment we’re unlikely to see any poll showing the Labour vote away from it’s range for the past few months (basically: 33% +/- 3 or 4pts either way) and it’s very likely that any that *does* show Labour out of this range will be an outlier and should be ignored anyway.

    As for underestimation; things have changed enough since 2005 to make it hard to tell whether or not the Tory vote is still being underestimated. In 2001 pretty much all pollsters overestimated Labour (and for reasons that really don’t apply now) and in 2005, if you bear in mind a margin-of-error of 4 or 3 pts, most pollsters did pretty well (o/c they all mysteriously converged towards the close of the campaign, making it hard to tell who’s methodolgy was best. But the pollster closest to the actual result was NOP).
    In 1997 ICM actually underestimated the Labour lead, btw. This may have had something to do with the fact that they were house-pollsters for the Tories at the time.

    Good call on the “lowest since 1066” nonsense, btw.

  4. It is possible that the ICM/Guardian poll was an outlier – we’ll get an idea of that in a month. However, it does seem to me that our position hasn’t recovered from the disastrous week of the local elections – I think I’m right in saying that all polls show a jump in the Tory lead from then on.

    I concur that in 2005 most of the polls were within their MoE – but they did all over-estimate the Labour lead, even NOP which was the most accurate. The problem is more people reading too much in to polls or journos hyping them to increase their news value than the polls being less accurate than the pollsters themselves claim.

    Interesting factoid about ICM in 1997. But I’d still bet that most polls are generous to us at the moment – as much because simple voting intention questions don’t fully account for the Cameron factor as because of methodological issues. Though this poll could just be party affiliation catching up with leadership perception.

    All polling should be treated with caution (and particularly simple voting intention polls this far from an election – as we’ve discussed, we hit 31% in ’03 and recovered to win in ’05) but it can be informative nonetheless.

    BTW, let us know if you’d like a reciprocal link.

  5. Al says:

    Well whatever the polls do or don’t say at the moment is more or less irrelevent (in most respects) as things should change over the next few years… I’m beginning to suspect that their main impact this far out is on the morale of activists.

    But, yes our position hasn’t really recovered from then, while the Tories have (probably) gained some new voters from somewhere (hard to tell whether it’s from undecideds or LibDems. Maybe both).
    And I certainly agree about polls being treated with caution; they give a fairly accurate (to within a certain margin of error, 9 times out of 10. In theory) snap-shot of how things are on a certain day or so, and that’s it. Useful in it’s own way, but often over-used.
    Still, at least Brown leads in most “best P.M” polls.

    And a reciprocal link would be much appreciated 🙂

  6. I think that voting intention can be looked at very much in the context that there is a long way to go, which is why I prefer a wider focus – particularly as journos are obsessed with VI and often don’t report other poll findings.

    I think most polling suggests that the Tories have gained from undecideds and people who previously didn’t vote – quite possibly disillusioned Tories who didn’t vote when they thought their party was completely hopeless; until now they’ve taken some Lib Dem votes but when you took Blair out the picture, so did we.

    Brown’s polling is complicated and somewhat contradictory at the moment. Generally he beats Cameron on competence but loses on personality. But he has a consistent lead on “best PM”.

    Will put up a link to you, and much appreciated if you return the favour! 🙂

  7. Schaumburgsportsregion1956$anchor$basketball Betting,final Four,final Four Betting,final Four Gambling,final Four Sports Book,final Four Sportsbook,march Madness,march Madness Betting,march Madness Gambling,march Madness Sports Book,march Madness Spo…

    Schaumburgsportsregion1956$anchor$basketball Betting,final Four,final Four Betting,final Four Gambling,final Four Sports Book,final Four Sportsbook,march Madness,march Madness Betting,march Madness Gambling,march Madness Sports Book,march Madness Sport…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: