This month’s ICM poll for the Guardian has voting intention figures of CON 31%(+2), LAB 39%(-2), LDEM 15%(+2), UKIP 7%(-2). The topline figures suggest a narrowing of the Labour lead, but this probably a reversion to the mean after what looked like a rogue poll last month. ICM have shown Labour with an eight point lead in four of their last six polls (ICM typically show lower Labour leads and higher Lib Dem scores than some other pollsters for methodological reasons to do with how they treat don’t knows).

The rest of the poll is reported as showing that the Conservatives would do better if they were more anti-European, or were more anti-immigration, or were more on the side of traditional families (whether people thought being more supportive of apple pie would help them was not, alas, polled upon).

I shall only repeat my normal grumbles about polls purporting to show that people would be more or less likely to vote for a party if they did x, y or z. They really don’t, people just use the questions to show their opinion of the issues being asked about regardless of whether or not it would actually shift their vote or increase/decrease their likelihood of voting for a party. Hence what the poll actually shows is most people don’t like immigration or the European Union much and do like families.


232 Responses to “ICM/Guardian – CON 31, LAB 39, LD 15, UKIP 7”

1 3 4 5
  1. There seems to be a definite shift by Cameron to playing the man instead of the ball. I think it might be a bit too early for all the jibes at the two Eds, especially the attempt to create a split between them.

    Crosby, again?

    Report comment

  2. If I understand it correctly, the parent of a soldier killed in action could expect to receive a bedroom tax demand along with their MoD telegram*, because they would no longer qualify for the exemption. That could make for quite some a tabloid story if it happened.

    *Just a bit of WWI flavouring. On a good day I know what century it is.

    Report comment

  3. statgeek

    Look at the last five polls[...]

    Note that the LDs and UKIP are consistent, while Con and Lab are not. Like I say, I think there’s a little fickleness around.

    I think this is just mathematics. The MoE with a sample of around 1,400[1] is 2.4 to 2.6 points if you are trying to estimate a percentage that is 30-40%. However if you are estimating something around 10%, the MoE is only 1.6 points.

    The result is that the figures for Labour and Conservatives will naturally move about a lot more than those for Lib Dem or UKIP without any extra fickleness.

    [1] Rough average of last 5 YouGovs after removing “non-voters”

    Report comment

  4. @Roger

    (My own quote)

    “Like I say, I think there’s a little fickleness around.”

    In fact I’ve changed my mind. It’s MoE.

    Hehe!

    Report comment

  5. Chordata
    The 1% benifit rise will have little impact on polling because most of those folk dont vote. It may however have an impact on crime figures

    Actually as 95% of housing benefits are paid to people in employment your first assumption is almost certainly wrong.

    Regarding crime figures there is a correlation between unemployment levels , poverty and property based crime but more significant than this particular restriction will I suspect be the increasing child poverty which according to the TUC’s analysis will place half of all children below the bread line by 2015.

    60% of low level property crime is committed by the under 18′s and if they are right in their analysis this is almost certainly going to result in serious implications at a time when the policing budget is being cut by 20%

    Wouldn’t do Theresa May’s chances of being the next leader of the opposition much good either,

    Report comment

  6. Tables are now up for the ICM poll:

    http://www.icmresearch.com/wp-content/blogs.dir/1/files/2013/03/OmBPC-Mar13.pdf

    Figures before adjustment for Don’t knows and Refusers are:

    Con 29% (31%)

    Lab 42% (39%)

    L/D 14% (15%)

    UKIP 8% (7%)

    SNP 2%

    Green 2%

    BNP 1%

    Adjusted figures in brackets.

    ICM say “Data excludes those who definitely will not vote (13%), don’t know who they would vote for (27%) or refuse to answer (8%)” and the total weighted response is only 43% with a sample size of 426 from 1,002. It’s worth pointing out just how high this is even for ICM and of course MoE will be high.

    Report comment

  7. @Steve,

    You assume that the reason under 18s commit property crime is because of their financial status.

    I think there is some correlation there, certainly, but I would be extremely dubious about assigning causation.

    It is many a year since urchins stole apples from market stalls to feed their families.

    Report comment

  8. STEVE
    “Chordata
    The 1% benifit rise will have little impact on polling because most of those folk dont vote. It may however have an impact on crime figures”

    I didn’t say that though.

    I was quoting what Richard in Norway said.

    Report comment

  9. Every picture tells a story and looking at the body language of his back benches today signals that he has lost the dressing room.

    Looking grim is one thing, but laughing along with opposition jokes reveals another.

    David Cameron needs to bite the bullet and do his next relaunch in a brewery to survive! ;-)

    The game is up.

    Report comment

  10. @Neil A

    “It is many a year since urchins stole apples from market stalls to feed their families.”

    Hmm. I thought they stole apples because they fancied an apple, or possibly because they were hungry.

    Report comment

  11. A correction to a previous post of mine:

    “The BBC report on the Falklands poll states ”… 1,672 British citizens eligible to vote in a population of about 2,900. ””

    This is incorrect. The franchise for voting was the same as for other elections on the Falklands. This is that all permanent adult residents are entitled to vote, including those not British Citizens.

    http://www.falklands.gov.fk/home/referendum-2013/

    Report comment

  12. Thats right he was quoting me talking about the 1% limit on benefit increases

    Report comment

  13. @ Carfrew

    My word, but this is the thin end of a very thick wedge isn’t it?
    I mean, what’s next? A Cooker Tax?
    ————-
    Indeed; I was alluding to the point that the Government are saying the social housing sector is being brought into line with the private sector when, of course, it isn’t. HB claims in the private sector are not reduced for there being a dining room or study (nor for them being furnished, as you point out) so there is certainly not the equivalence which the Government claim.

    Report comment

  14. STEVE
    “60% of low level property crime is committed by the under 18?s and if they are right in their analysis this is almost certainly going to result in serious implications at a time when the policing budget is being cut by 20%”

    What do you think might be the equivalent anti-social or criminal behaviour among Asian youth with their much higher levels of unemployment?
    Just asking.
    No, on reflecting I know what it might be, and would suggest the need for some very well thought out strategies for inclusiveness and generalised strengthening of vocational and professional training and of support for such a programme from within the Asian community.
    What’s your experience of the effectiveness of any such policy and of community attitudes?

    Report comment

  15. The 95% of housing benefit claimants are in-work is, to use a technical term, complete and unadulterated bollocks. What you are thinking of is a claim that either 93% or 95% of *new* housing benefit claims are from people in work. This is also untrue, as explained by Fullfact here:

    http://fullfact.org/factchecks/90_93_95_percent_housing_benefit_claimants_in_employed_work-27483

    What is true is that that of the overall rise in people on housing benefit between the two dates the Smith Institute and the BSHF chose to come up with those figures, 93% or 95% of the rise was made up of people in work. This is not the same thing at all.

    Anyway that’s by-the-by, that particular stat is normally used to create the impression that of all housing benefit recipients most are in work. Obviously this is not necessarily the case. There isn’t an actual figure, because the DWP don’t collect it because of the way housing benefit works. Essentially you can apply for housing benefit directly (in which case the DWP may have data on whether people are employed or not) or you can get it by virtue of receiving another benefit, like jobseekers, income support or pension credit (called being passported), in which case the DWP does not collect any extra data on whether people are employed.

    What we know is that at least 18% of people receiving housing benefit were not passported and were in employment, so we have a minimum proportion of people getting housing benefit who are in work. This figure has been rising significantly over recent years. At the other end of the scale, 13% of people getting housing benefit got it through virtue of being on JSA, so we know they are out of work. In between we have 69% of people who are getting the pension credit, or income support, or ESA or none of the benefits, none of whom we have any firm data on their employment status.

    In short, all we can be certain about is that between 18% and 87% of people in receipt of housing benefit are in work, probably nearer the lower end, since ESA and Income Support, while available to people working a limited number of hours a week, are broadly classed as out-of-work benefits.

    Anyway, most of this is once again from Fullfact:

    http://fullfact.org/factchecks/one_in_eight_housing_benefit_claimants_unemployed-27479

    For the record, the TUC release today isn’t about poverty either. It projects that the proportion of children living under the minimum income standard calculated by the Joseph Rowntree Trust, will be 54% in 2015, but would have been 49%. Make of that what you will, but being below the minimum income standard is very much NOT the same as being in poverty, they are very different things. The MIS for a couple with one child is £24643. The most recent figures I can find on the Child Poverty Action Group’s website for what they think the poverty line is for a couple with one child is £13354. Anyway, full report is here

    I’ve no real interest in the silly back and forth of statistics here to score political points, but please everyone, before throwing them out there check you are citing them and understanding them correctly.

    Report comment

  16. @Steve,
    You assume that the reason under 18s commit property crime is because of their financial status.
    I think there is some correlation there, certainly, but I would be extremely dubious about assigning causation.
    It is many a year since urchins stole apples from market stalls to feed their families.

    -
    I don’t make an assumption at al however empirical evidence does support this conclusion.

    Also for personal experience I did spend 20 Years as an operational Police Officer in Inner London. At a time when property crime levels were higher than now.

    Can’t say I ever arrested anyone for scrumping apples , not a lot of apple trees in Harlesden and Brixton but I did arrest hundreds of youths for stealing property primarily because they couldn’t afford or couldn’t be arsed to buy it.

    The overwhelming majority of these young offenders were from poor backgrounds.

    Report comment

  17. @ Roger

    Do those ICM stats adjustments you quote include weighting? (Ie- getting more Lab 2010 voters in the sample than there should be so trimming down their vote). If not then I guess ICM is exactly where YouGov is- just different methodologies.

    In my view the background chatter on here about what the polls are doing is meaningless- other than a rise for UKIP (and a small fall for Tories but nowhere near the rise in UKIP) nothing has changed for a year and it’s all MOE or small changes because of a specific event that then revert back after a few days.

    Report comment

  18. AW

    I did not mention Poverty with regard to the TUC Survey the survey does as you indicate relate to the minimum standard deemed necessary for a decent standard of living. Howard Reed calculates this at around £29k pa for a family of four by 2015 and calculates that 54% of children will be in families with incomes below this level.

    Within this figure there will inevitably be those who meet the CPAG’s definition for poverty
    .

    The standards acceptable for living are those drawn up by the respected Joseph Rowntree Foundation.

    Reed also concluded 90% of families will be worse off in 2015 than in 2010

    Report comment

  19. What a great website Anthony.

    Wasn’t aware of it-a wonderful resource.

    Report comment

  20. “silly back and forth of statistics here to score political points, but please everyone, before throwing them out there check you are citing them and understanding them correctly.”

    If only we could have politicians who set an example.

    “QE is a money tree.”" Higher borrowing can easily lead to lower borrowing.” And the opposite, depending on where you stand.

    If only we could be trusted with the proper rationales instead of the Punch and Judy blllcks

    Report comment

  21. You can suggest a fact check too!-great stuff.

    Report comment

  22. Shevii

    As far as I can tell those figures are after all weighting, both for demographics and for turnout. The only adjustment not made is the one for Don’t knows and Refusers. These combined figures are around 30% for those who say they voted Conservative or Lib Dem, 17% for Labour. Of course there may be false recall problems, particularly with (ex-)Lib Dems.

    Report comment

  23. Hal

    A correction to a previous post of mine:

    “The BBC report on the Falklands poll states ”… 1,672 British citizens eligible to vote in a population of about 2,900. ””

    This is incorrect. The franchise for voting was the same as for other elections on the Falklands. This is that all permanent adult residents are entitled to vote, including those not British Citizens.

    I’m not sure that’s quite correct. There’s a leaflet on the website you linked to:

    http://www.falklands.gov.fk/assets/Guidance-Note-leaflet-Oct-2012.pdf

    which suggests that new registrations have to be British citizens and have “Falklands Island Status”. However there is also an inclusion for anyone who was in the 2009 electoral register which suggests the franchise may have been wider in the past and it may be that this included non-Brits and there for anyone who’d been around for longer than five years would have qualified for the referendum.

    Report comment

  24. @Paul Bristol

    I think Cameron’s safe, for now, mainly because I don’t see a viable and electorally plausible alternative to him, certainly not from within the current parliamentary Tory Party. Gove? May? Maybe I’m a bit tribally averse anyway, but surely most Tories can see that both of these politicians have the electoral appeal of horse meat. Although it’s no great achievement at present, Cameron is still polling ahead of his party and, despite his lukewarm personal ratings, he’s still the Tories best electoral asset. Notwithstanding the froth and blather in Westminster, I think most of the realistic wing of the Tory Party recognise this and, accordingly, will cling to nurse this side of the General Election. I base my prediction on the assumption that the Realistic Tendency is still in the majority within the Tory Party. If I’m wrong, and events may disabuse me of my assumption, then all bets off and mayhem and silliness may well ensue.

    I would think that Miliband is dreaming of a Gove or May led Tory Party. Unelectable, the both of them, in my view.

    Report comment

  25. Turk – re
    ”The problem is will Yougov begin to loss some credibility against other pollsters with it’s constant polls results that can very between 8% and 14% leads within a day”

    If YG do lose any credibility it will not be because of daily polls but because some people (including Journos) take too much note of single polls and misread and/or misuse them.

    Unlike the esteemed pollatariat on here who have been priviliged to be educated about such things.

    I think others have said that the LDs seem have consolidated their 2% or so lift since the new year and that UKIP appear to have taken around 2% off both Labour and the Tories since Eastleigh. The rest is moe and methological.

    In terms of effecting the likely GE result other than the LD boost it is the polldrums de dum de dum.

    Report comment

  26. Re: poverty. I am not belittling anyone’s genuine need but I would love to see full separation made clear between “minimum income”, “the poverty line” (both relative measures of poverty and in reality more a measure of the divide between richer and poorer in society) and absolute poverty – the lack of food, warmth, shelter and the like. I would love to see it but I’m not expecting anything any time soon. So I expect there will simply be more misleading conflating of different definitions…

    Report comment

  27. STEVE
    “Reed also concluded 90% of families will be worse off in 2015 than in 2010″

    There are stats and stats: those that ‘triangulate’ with qualitative evidence, as do the reports of the Child Poverty Action Group point to a severe and growing level of poverty which add to Reed’s analysis in saying what the effects of poverty are in children of low income families as a vulnerable group, that is, one which would be harmed by ill-conceived economic policies, and correspondingly helped by good ones:
    “Child poverty blights childhoods. Growing up in poverty means being cold, going hungry, not being able to join in activities with friends. For example, 62 per cent of families in the bottom income quintile would like, but cannot afford, to take their children on holiday for one week a year.5
    Child poverty has long-lasting effects. By 16, children receiving free school meals achieve 1.7 grades lower at GCSE than their wealthier peers.6 Leaving school with fewer qualifications translates into lower earnings over the course of a working life.
    Poverty is also related to more complicated health histories over the course of a lifetime, again influencing earnings as well as the overall quality – and indeed length – of life. Professionals live, on average, eight years longer than unskilled workers.7
    Child poverty imposes costs on broader society – estimated to be at least £25 billion a year.8 Governments forgo prospective revenues as well as commit themselves to providing services in the future if they fail to address child poverty in the here and now.
    Child poverty reduced dramatically between 1998/9-2010/12 when 1.1 million children were lifted out of poverty (BHC).9 This reduction is credited in large part to measures that increased the levels of lone parents working, as well as real and often significant increases in the level of benefits paid to families with children.
    Under current government policies, child poverty is projected to rise from 2012/13 with an expected 300,000 more children living in poverty by 2015/16.10 This upward trend is expected to continue with 4.2 million children projected to be living in poverty by 2020.

    Report comment

  28. Sorry: the references are from the CAP website.

    Report comment

  29. Gawd! CPAG

    Report comment

  30. Roger,

    Ah yes, you are right.

    There was a new constitution on 1 Jan 2009 and after that date the only new registrations to vote allowed are for British Citizens. However anyone on the register at that date is entitled to stay on the register.

    Section 32 of the constitution: http://www.falklands.gov.fk/self-governance/the-constitution/

    Thus a number of non-citizens will have voted in the referendum.
    And presumably there was also a number of disenfranchised residents.
    So it is rather more complicated than the BBC made out.

    Report comment

  31. Apologies for returning to the point I meant to make in my (rather messy) post above: Cutting through the sterile debate on whether poverty is a relative concept or an absolute, the CPAG figures (ww.cpag.org.uk/child-poverty-facts-and-figures) and their causal relationship with welfare policy and public sector investments have the merit of being verifiable. Thus:
    • By 16, children receiving free school meals achieve 1.7 grades lower at GCSE than their wealthier peers.
    • Child poverty reduced dramatically between 1998/9-2010/12 when 1.1 million children were lifted out of poverty (BHC), credited in large part to measures that increased the levels of lone parents working, as well as real and often significant increases in the level of benefits paid to families with children.
    • Child poverty is projected to rise from 2012/13 with an expected 300,000 more children living in poverty by 2015/16.10
    We are not dealing with any academic miserology – these are our children and our future citizens. Their costs and benefits to society depend on decisions which are made in government.

    Report comment

  32. ‘• By 16, children receiving free school meals achieve 1.7 grades lower at GCSE than their wealthier peers.’

    Yes, many academic studies highlight that the biggest predictor of how well a child does at school is home background (measured in many ways- books used in the home, parental educational background, language used…) as the home sets up attitudes and skills which schools need students to have. The vast majority of students who ‘fail’ at school have parents to blame – but politicians cant say that as parents vote.

    Schools are about getting students to jump through qualification hoops and nearly all schools and teachers can do that. So long as the students come into school with the right skills and attitude from the home…

    Report comment

1 3 4 5