On the margins in the marginals: Ashcroft Poll analysis

Lord Ashcroft’s latest poll should both reassure Labour strategists and give them a fairly clear pointers on where Labour needs to do better to turn a mid-term polling lead into victory.

Below a quick summary, featuring several charts and brief commentary.

One of the good things about Lord Ashcroft’s polling of the key marginals is that it’s done on a scale that allows a reasonable look at the detail.

Being the international ambassador for fun times that I am, I spent some of last night looking at the detailed data for the Conservative-Labour battleground seats, the seats the Tories hold now, but Labour have to win to gain a majority.1

I’ve done some pretty charts in presentation format below, and a brief summary below that.

In the Conservative held marginals, Labour currently has a significant lead of ten points on standard voting intention, and eight points locally.

Clearly, that would be enough to deliver a significant Labour majority as has been reported elsewhere.

Labour voters are strongly disapproving of the government and pessimistic about the future under the Coalition. They also feel Labour is in touch with their values and interests.

However, the internals identify some areas for concern, and this is where Labour will likely need to focus.

In terms of issues, the economy dominates, but the data shows that immigration is also a key issue for potential Labour voters, so the recent focus on that issue is clearly worthwhile.

On the economy, Labour needs to do better with “unsure” and “potential” Labour voters.

Labour also needs to highlight Ed Miliband more effectively. While Labour voters overwhelmingly prefer Miliband to Cameron when forced to choose between the two men, there’s significant doubt among potential voters, and a lack of certainty among Labour voters when offered “None of the above” as an option.

Finally, Labour needs to demonstrate a clearer vision for the future, as “unsure” and “potential” voters aren’t clear what it would do in government.

Labour should also be aware that there is a significant group of “potential” Conservative voters, who are very dissatisfied with the government, but are hostile to both Labour’s alternative and to Ed Miliband as an alternative PM to David Cameron, and so might be tempted to swing behind the Tory party at some point. (My assumption is that these are UKIP voters, who currently represent 11% of all voters, even in these marginals)

When it comes to the difference they’d make in government, Labour does better on jobs and public services, the Conservatives on welfare, deficit, immigration, tax and the economy. However, both Labour and Conservatives are struggling to convince much more than half their own supporters they’d make a big difference on most issues.

This lack of enthusiasm might suggest space for a ‘Fear of worse’ election.

If that’s the case, then Conservative voters and potentials are extremely sceptical of a Labour government. This is more the case than Labour voters and potentials are sceptical of a Tory govt, except on the NHS and Unemployment.

In terms of campaigning, Labour has no advantage in the ground game at the moment, and may even be a little behind in voter communication, both in general and among key voting groups. However, the Tories seem to be using Leaflets and emails, rather than personal contact, which may be a major weakness, as personal contact is more effective in winning votes. Who-ever performs better on voter contact over the next two years may well have a significant local advantage.

  1. though the results in Labour-LD and LD-Tory seats will also be very significant. []

8 Responses to “On the margins in the marginals: Ashcroft Poll analysis”

  1. PooterGeek

    Excellent, sane, crisp analysis. Nice work!

    (I disagree with you about immigration, but I would, wouldn’t I?)

    Reply
  2. Brian Hughes

    Have you come across any convincing, non anecdotal, evidence to support the assertion that “personal contact is more effective in winning votes” ?

    Two recent observations suggest that the reality is more nuanced than the simplistic “contacts=votes” mantra suggests:

    In 2010 Labour made more voter contacts than it had ever made before yet suffered one of its lowest ever popular vote.

    Currently the party is, as you note, not doing much voter contact work yet is riding high in opinion polls.

    Reply
    • hopisen

      I suspect that part of the difficulty in getting solid results from data is that people are strangely reluctant to allow researchers to use their campaigns as research projects with proper control groups!

      However, we do know that face to face contact does boost turnout:

      http://onlinelibrary.wiley.com/doi/10.1111/j.1468-2508.2004.00280.x/abstract;jsessionid=C21A823F71C379EE0231839DA411C67A.d01t01?deniedAccessCustomisedMessage=&userIsAuthenticated=false

      and so do phone calls -especially from volunteers

      https://www3.nd.edu/~dnickers/files/papers/Nickerson.APR2005.pdf

      (that study also refers to US research on Face to face being better than Direct mail)

      Also, I know of a few MPs who selected a list of c1000 voters and contacted personally by phone prior to/during an election campaign. The turn out among this group was significant’y higher than among the non-contacted (who would have got the same leaflets, national DM etc etc)

      Reply
      • Brian Hughes

        I have little doubt that contact from a local Labour MP (or councillor) would boost turnout and, more importantly, boost the number of voters voting Labour. I’m less convinced that repeated contact from enthusiastic but poorly briefed volunteers is quite as efficacious. But I think Contact Creator counts the two equally – it counts quantity rather than quality.

        I only have anecdotal observations to support my theory, alas my pension isn’t sufficient to allow me to fund research programmes. But, at my two local marginals in 2010, Gloucester easily won the most voters contacted award whilst Stroud won the rather more important lowest swing against Labour one. I’ve heard similar stories from other parts of the south of England (OK, I’ll be honest – I heard one similar story but in a court of law I’d’ve risked been upbraided for leading the witness).

        My worry is that party managers focus too much on the one aspect of a campaign that appears to be simplest to measure. This may lead them to neglect other important aspects of a campaign such as leaflets which deliver clear, consistent reasons to vote Labour in a particular constituency*.

        I think high voter contact, before it became such a key target, used to be a symptom of a well run campaign. A well run campaign led to more voters voting Labour than would otherwise have been the case.

        But now, having been misinterpreted as being the cause of voters voting Labour, it has become a simplistic target that has led to campaigns not being so well run overall.

        But, like evolution, hey it’s just a theory.

        * most leaflets I saw in 2009/10 looked more like PPE theses than things which a normal human being, with only a passing interest in politics, might want to read.

        Reply
  3. Alan Ji

    I read Ashcroft’s post before this one. Several things gave me food for thought, but the most distinctive was the LIb Dem seats with Labour 2nd. Ashcroft’s poll puts Labour in a winning position in ALL of them. Given that some of those LibDem MPs have been well dug in for a long time, what do we have to do to make it so?

    Reply
  4. Joy Prince

    Evolution is not ‘just’ a theory, it’s a proven theory.

    Reply
  5. Alex

    Note the difference in the YouGov poll between issues people rate as “important for the country” and ones they rate as “important for me and my family”.

    first observation: for most issues, there’s not much difference, foreign policy is more salient in the first group as you’d expect.

    second observation: one issue is hugely different. immigration. goes from issue no.2 to issue no.6 and from 52% to 16%.

    I think what this measures is that politicians, broadly defined to include opinionators, bang on about it in the national-level media a lot but humans don’t actually experience it as a problem.

    I haven’t checked to see if there is a consistent pattern of this, but if there is it might explain why the politics of anti-immigration has been so ineffective in getting elected when it’s supposed to be so popular.

    Reply

Leave a Reply