The Guardian have released this month’s ICM poll, the first telephone poll conducted since the local elections. Topline voting intention figures with changes from a month ago are CON 28%(-4), LAB 34%(-4), LDEM 11%(-4), UKIP 18%(+9!).

The 18% for UKIP is the highest that ICM have shown and, more strikingly, the highest any company have shown. I will advise my usual caution about polls showing extremes and records, more often than not they tend to be outliers. ICM’s methodology tends to produce some of UKIP’s lower figures (ICM reallocate some don’t knows to the party they voted for last time, weight down people who didn’t vote last time and interview by phone… all things that tend to produce lower UKIP scores) so it is particularly surprising to see ICM with a record breaking UKIP score. While the scale of the UKIP increase may well be a bit of a blip though, the broader trend is the same as other companies – they are all showing a surge in support for UKIP.

It does make one ponder about how high we might see UKIP in some other companies’ polls. If YouGov have them as high as 16% or 17%, if ICM have them at 18%, what might we see from companies like Opinium or ComRes who tend to show higher levels of UKIP support?

More unexpectedly the ICM poll also found a jump in support for the BNP, up to 4%, the highest any poll has had then at for years. This is strange. The BNP have certainly not had any great publicity boost, at the local elections they seemed essentially moribund. It may just be an odd sample, or perhaps as Tom Clark suggests it is just a case of confusion amongst respondents, with some people getting the names of the BNP and UKIP mixed up.

ICM also asked about voting intention in an EU referendum, finding voting intention fairly evenly balanced – 40% would vote to stay in (22% definitely, 18% probably), 43% would vote to leave (32% definitely, 11% probably).

UPDATE: ICM tabs are up here. Topline figures without reallocation of don’t knows would have been CON 27%, LAB 35%, LDEM 9%, UKIP 19%, BNP 5%.

That strange boost of support for the BNP is almost wholly amongst women, almost wholly amongst C2s, almost wholly amongst over 65s and almost wholly in Wales. The unweighted number of 2010 BNP voters in the sample was 1, increased to 18 by weighting. What that strongly suggests to me is that there was one little old C2 BNP-voting Welsh lady who got a very high weighting factor, and probably makes up almost all of that 4%! Such things happen sometimes, but it means the BNP blip is probably just a data artifact that can be ignored.

499 Responses to “ICM/Guardian – CON 28, LAB 34, LD 11, UKIP 18”

1 8 9 10
  1. StatGeek (or anybody else) –
    Do you track the ‘Do you support/oppose the coalition agreement?’ question?
    I haven’t looked at it in a while, but in the latest YouGov poll support for the coalition with Con voters is only at +8 (52/44). That seems relatively low for the major coalition partner.

    (Worse if you look at 2010 Cons – with support at -10 (41/51)).

  2. 30:40:10:15
    A statistician’s delight. Beautiful.

  3. Labour up, Coalition down.

    Could be the effect of EU division?

  4. “Why no fans of the lesser teams?”

    There were lots of permutations and what ifs on the last day of the season but sadly my team, Barnet, were relegated from the Football League :-(

    Still good to see the YouGov poll this morning. And the more the Tories tear themselves apart over Europe the more this gap will widen. :-)

  5. Incidentally, I can’t help feeling this is much more important than a sterile debate over Europe:

  6. “Labour up, Coalition down.
    Could be the effect of EU division?”
    My (rounded) weighted average stands at Con 30, Lab 39, UKIP 15, Lib 10
    The last 4 polls have been –
    Con 30 (-),30 (-),31 (+1),30 (-)
    Lab 39 (-),39 (-),38 (-1),40 (+1)
    Lib 10 (-),9 (-1),10 (-),10 (-)
    UKIP 14 (-1),16 (+1),14 (-1),15 (-)

    I suggest, alternatively, that there has been no change over the past few days.

  7. Spearmint,

    I think I remember the opposition putting forward a bill in the Commons in recent times.

    In any case I am sure that if there is a majority in the Commons for a bill then the speaker would be obliged to find time for it.

  8. Blues in the lead in 18-24 age group:-
    Con 37
    Lab 35
    LibDem 15
    UKIP 6

    Usual caveats about crossbreaks I suppose so it will be interesting to see if this holds up in the next few polls.

  9. @Hal

    The suggestion (I think) is that they ask a backbench MP to put it forward as a Provate Members Bill. To do this they need to get a Eurosceptic high enough up the ballot so that it can get any time at all. (It can’t be proposed by a member of the Government apparently if its a PMB)

    That is likely, but even then PMBs as a group only have limited time for debate, so they probably need to hope that there aren’t any specially worthy bills ahead in the queue. I don’t think that the Speaker would have to give time to what is likely to be a minority opinion in the Commons… And how would he judge if it was?

  10. Labour lead of 10 could be movement back to the Labour 10%-11% lead OR within MOE showing Labour lead on 8.5% OR slow increase in Labour lead heading who knows where. But we need a full week of polls to see a trend.

  11. Richard

    If you look at the screenshot, they do seem to have been promoted, I just wasn’t sure if that was a recent thing, or if they had always been in the premier league

    That’s interesting, I always assumed that YouGov only offered the appropriate Nationalist Party if you lived in the right area, but that screenshot seems to comes from Bexley. That said, I know YouGov have done tests before to see if the positioning made a difference (perhaps splitting the sample in half and trying different things on each), so it may be that a question from one of those – Anthony may enlighten us later.

    I’m not sure that it would make much difference, though. In the past the boost for UKIP by putting it on the front page has been to act as a reminder that it existed before (most) people made a choice. They’ve hardly had a shortage of publicity in the last few weeks, so any improvement may be minimal. Indeed it might even reduce the UKIP choice a little as some of the alienated might still click on Other (“They’re all the same”) and choose their least worst of those (BNP, ‘Other’ or even Green).

  12. I’ve always viewed this as the best 13 second summary of Tory backbench MPs take on the EU.

  13. Blues in the lead in 18-24 age group.

    I’ve noticed this recently and I find it very weird. Not least given that the Tories not only wanted to raise tuition fees to £9000, they wanted the limit removed.

    Also, of the 100 or so people in that age group who I know, only two are Tory voters. I’m then forced to conclude that either young Tories are much more likely to vote (barely anyone I know voted in the locals, and the four I did voted Tory, Tory, Labour and Green) or voters are following what their parents do.

  14. @Chrislane1945 – you said previously that you felt DC was doing brilliantly over Europe for the nation, but again I have to differ.

    By launching a confusing period of uncertainty, without any clear objectives transmitted to our EU partners, creating an eruption within his party that is sucking up up time and effort that should be devoted to negotiating reform, he has made a major error.

    I support his overall prognosis and his desire for change, but his tactics are very poor. This doesn’t serve the UK interest well. By kicking off open splits on the EU within his own party, in a completely unnecessary manner, he has degraded his ability to convince other EU leaders that he can deliver a deal, and is therefore undermining our national interest.

  15. Labour lead grows by by just under 43% overnight !!!!!!!!!!!!!

    They are set to win by more than a nose at this rate.

  16. Alec

    Wot you said.

  17. @Mr Nameless

    Most 18-24 year olds will never go to university. Tuition fees may not mean that much to the majority.

  18. I think people criticising Labour’s lead are attempting to compare EM to Blair.

    What we need to readjust to is smaller majorities of the kind we had between 1945 and 1983, before the Alliance came and messed with everything. ’83, ’87, ’97, ’01, ’05 all had pretty large majorities returned and that might cease to be the case. Ed Miliband should be happy with 350+ seats.

  19. Europe yet to settle problems within euro zone: the two other main party leaders against him, including the DPM, his coalition partener: a divided party and cabinet.

    Don’t feel the tactics measure up to the problem at all.

  20. @Chris Riley

    A fair amount do, and for those who don’t the lack of social housing, sluggish economy and youth unemployment rates are still enormous problems.

    Plus there’s the fact that almost all generations have been more left-wing in their youth.

  21. On the subject of ICM and telephone polls, ICM say that they include about 150 mobile numbers in their sample. Whether 15% is enough (at least one Irish pollster uses 50%) and whether there are sampling problems in how the mobile numbers are chosen and the questions asked are other matters.

    It’s also worth repeating a point I think Hal made. These telephone polls tend to start out with smaller samples (especially now the Times seem have stopped using Populus) and have a higher rate of people saying “Don’t Know”, refusing to say their choice or just not voting. So the effective sample size (the number of responses the headline percentage are based on) can be as low as 500 or less. The margin of error is therefore bigger and likelihood of a ‘rogue’ poll more.

    I’m not completely convinced by Anthony’s theory of the whole thing being thrown into confusion by one elderly Welshwoman[1] (the BNP figure for the rest of the Midlands is odd too). I wonder if there is some sort of Reinhart-Rogoff error[2] with residuals being added on the BNP row or similar.

    [1] Of course it’s possible that she was an expat Frenchwoman with slightly imperfect English who thought it was a banking survey.

    [2] The interesting thing about the R-R affair is that the more significant mistakes in their work (such as the weighting given to various events) were not what destroyed its reputation, but the fact that they made a mistake in a spreadsheet. Because, let’s face it, we’ve all been there.

  22. @ Hal,

    The Speaker doesn’t control legislative timetabling. The Government and Christopher Chope, Philip Davies and David Nuttall do.

    The Speaker can make time for debates under Standing Order 24 or by selecting an Urgent Question, but those are just debates; they can’t introduce a bill. The only mechanisms to bring forward actual legislation (and thereby repeal an existing law like the Fixed Term Parliaments Act) are Government Bills and Private Members’ Bills, and Private Members’ Bills are highly vulnerable to being talked out if there is substantial opposition to them from any quarter.

    Practically speaking there’s just no way to get a Private Members’ Bill through first reading if either frontbench objects to it. Miliband and Clegg don’t even have to vote against Cameron’s EU referendum bill, they just have find twenty backbenchers willing to show up on a Friday morning and complain about it. Likewise repealing the Fixed Term Parliament Act. It’s dead on arrival unless Cameron, Clegg, Miliband, AND the Conservative Backbench Filibuster Team all want it to pass.

  23. @MrNameless

    Almost all generations had Tory governments to rebel against. 18-24 year olds had Labour representing The Man all the way through their teenage years and no Tories in sight.

  24. Statgeek inspired me to make myself useful, so here are updated 5-poll YouGov rolling averages. All parties are on comparable y-axes except Ukip, whose VI grew so much I had to give them more space.


    You can see the Thatcher boost, but the Tories have since lost it all to Ukip.

    With occasional blips, Labour seem to have suffered a linear decline since early March. This is also clearly visible in Statgeek’s regional trendlines. It’s not really correlated with major news stories, Tory gains, Lib Dem gains, Ukip gains, or anything else, so I’m a bit of a loss to explain it. Increasing economic optimism? (Cheering news for Ken et al.- if the trend continues they’ll be on 15% or so by the general election!)

    Slight decline for the Lib Dems.

    Ukip hit a plateau after Eastleigh but began growing rapidly after the local elections. So far there’s no sign of a second leveling off.

  25. ALEC

    @”By launching a confusing period of uncertainty,”

    Do you know how long anything takes in the EU ?
    How long do you imagine it will take to agree DC’s agenda for UK?
    THe FO haven’t published the results of the Review of Competencies yet-until this is done the detailed agenda is not decided.

    @” without any clear objectives transmitted to our EU partners, ”

    See Review of Competencies.

    Read DC’s Bloomberg speech ( 23/1/13)

    ….and Merkel’s reaction.

  26. Just for you Alec-its a good read-I feel you will approve, given your well known views on the EU.

  27. @Colin – read a recent Telegraph report into the competencies review. It said they are finding what many of us have said for years – that EU law is actually quite reasonable, but that UK officials copper plate the regs to make them far more complex that required.

    If nothing else comes of this, if action on the EU leads to a better way of translating directives into UK law, then much good will come of it.

  28. Ozwald/mrnameless

    The under-25s are a perpetual problem for YouGov (and the other polling companies). There just aren’t enough of them on the polling panels (particularly male, C2DEs) and those that do reply tend to be more likely to be non-voters (as is representative of their age-group).

    If you look at today’s YouGov, although the weighted sample for 18-24 is 230, only about 105 people actually gave an opinion that fed into the VI for that group. On some days the figures go into italics which indicates that the effective sample is less than 50. So each individual responding is responsible for maybe a point (or even two) of VI.

    The result is that the VI for this group can vary wildly from day to day – UKIP’s figures also jump around a lot, though in general they are lowest for this group. There may also be a problem with the the make up of those members of the panel who are under 25 (I think Anthony once said they had far too many graduates) which may mean that samples are a bit bluer than you might expect. Though of course YouGov will really be interested in whether the sample is representative of those who will actually vote rather than generally.

  29. ALEC

    @”much good will come of it.”

    Hallelujah !

    I have framed this-in a Blue border.

  30. @PaulCroft

    Thanks muchly, Paul. I was leaning towards the open hole thing… apparently it fosters better finger positions and you can do glissando and other groovy things. And groovy is good in my book…

    On the other hand it’s harder, and harder is bad in my book. But it’s now pretty clear from what you said that given open holes is trickier, makes sense to make life a bit easier in turn by going down the offset road. 

    Who knew? I just thought one bought a flute. I recently found out they even do one in B-flat which is more up my street…

  31. Mike Smithson just tweeted “YouGov tell me that there has been no charge in way it polls Ukip. The screen-shot in my Tweet putting it on main prompt was from a test”

    That was the same screen shot I posted earlier.

  32. Spearmint

    Interesting graphs. It does appear that Lab has lost 3 and a bit% to UKIP. There is an apparent co-incidence in UKIP’s rise from Feb onwards and Lab’s fall.

    That ties in with the bit of amateur analysis I did a week or so back on the movement in the Lost LD voters. I reckoned that about 20% of the Lost LDs had moved from Lab to UKIP in the last 3 months. That would equate to 3-4% of the overall electorate.

    My take? A simple and broadbrush one admittedly, but I’m a big picture sort of guy. These are the inveterate awkward squad. They are the people who believe that all politicians are b******s and will support whoever appears to be against those in power. They supported Clegg, the anti-politician in 2010. They supported Labour from 2010 to now because there was no alternative if you wanted to register the fact that you despise the Govt. And since Eastleigh, with UKIP suddenly becoming a credible party to support, they have decided to support UKIP.

    I’m being harsh here, but I don’t see this group as having any core beliefs other than the belief in sticking two fingers up to politics and politicians. Labour now has a big decision to make. Does it go chasing these people (but how? What policies will bring them over to Lab?) or does it hold its nerve and accept that it can still win in 15 without these voters?

  33. Roger – the ICM regional breaks are slightly odd, “the Midlands” break actually includes the Wales break within it… hence it is the same C2 elderly Welsh lady :)

    The giveaway is the weighted and unweighted recalled vote figures. On the unweighted figures only 1 person said they voted BNP in 2010. In the weighted figures the equivalent of 18 people said they voted BNP in 2010. Given weighting only upweights or downweights existing people (it cannot conjure new BNP voters from thin air), the obvious implication is that one former BNP voter had a weight factor of 18.

    Circumstantially the extreme clumpiness of the current BNP support suggests that same one person is making up most (but not quite all) of their support.

  34. “Most 18-24 year olds will never go to university. Tuition fees may not mean that much to the majority.”

    Nearly 50% of 18-24 year olds go to university these days.

  35. @Neil A –

    “It would perhaps be a useful intellectual exercise to set out, given the circumstances and constraints that he is operating under, what would be a sensible policy for Cameron to adopt?……..

    “Retire to dig his own garden?

  36. Amber

    Most of the females I know boast about their age. Mindyou they are either 11 year olds and their younger siblings or over 80.

  37. @ mrnameless

    “What we need to readjust to is smaller majorities of the kind we had between 1945 and 1983, ”

    1945, 1959, 1966?

  38. Spearmint,

    Thanks for that info.

    So in the case that the government loses a vote of confidence, then the majority that voted for the confidence motion can form a new coalition, new PM and all that, with a governing programme to repeal the fixed term parliament legislation and then call an immediate general election.

  39. @Roger M
    “I wonder if there is some sort of Reinhart-Rogoff error[2] with residuals being added on the BNP row or similar.”

    I wondered the same a bit earlier.

    “the obvious implication is that one former BNP voter had a weight factor of 18”

    What we can’t say is whether that apparent weighting was produced through design or error. It can’t be entirely impossible that data was somehow put in the wrong place so that that one BNP voter appeared to have a weight factor of 18. What’s less implausible – a weight factor of 18 or an error?

    If there was a transposition error or the like, more than one party will have been affected. Given that we’re all getting excited about the polling company that regularly produces the lowest scores for UKIP suddenly producing the highest, in the same poll that produces silly numbers for the BNP, I think it’s a reasonable alternative theory. Might ICM be persuaded to say something on the matter?

  40. Norbold – Okay, I may have misphrased that.

    What I mean is that even in those landslides, the party in government never got above 400 seats (Attlee came closest) in the way Thatcher nearly did in ’83 and Blair did in ’97/’01. There was always the possibility of being overthrown in a single term, as happened in the 60s/70s several times.

    Much as I want a Labour majority, it’s precisely that sort of enormous majority which leads to complacency and disunity among parties.

  41. Phil – perfectly possible to get a weight factor of 18 (albeit, both rare and undesirable!)

    Imagine you had to weight up C2 by a factor of 2. Imagine people in Scotland needed to be weighted up by a factor of 1.5. Imagine people in local authority housing needed to be weighted up by a factor of 2, imagine people who didn’t vote also needed to be weighted up by 1.5. Imagine women under 24 needed to be weighted up by 2.

    Someone who then happened to tick every single one of these boxes would end up with a weight factor of 18.

  42. @Alex Harvey

    “Nearly 50% of 18-24 year olds go to university these days.”

    ‘Nearly’ is a very important word in this context, Alex.

    I repeat – most 18-24 year olds in the UK will never go to university, so it’s best to put more significance on issues that affect all of them and not just the middle class ones (even if they’re a bit overrepresented in the sample).

  43. @ Chris

    True, I guess ‘most’ does technically mean 50.1%+, I guess I always thought of it as meaning a largish majority.


    Could you recommend a good site that has some methods and examples of polling calculations showing how the base data is weighted to give final VI values?

  45. This should be fun for Paul and Ken…

    “The Premier League is crossing its fingers that it is not presented with a major headache if Arsenal and Chelsea finish the season level on points, goal difference and goals scored.

    That scenario could happen if Arsenal win 2-1 at Newcastle and Chelsea draw 0-0 at home to Everton this Sunday. That would make the clubs joint third and Arsenal and Chelsea would have to face a play-off for the automatic Champions League qualification spot.”

    The Graun…

  46. Gary – on the left hand sidebar, just above the archive links, there are some links to some old articles I wrote explaining all the methods. They are a bit out of date so I need to do a new version at some point.

    There are no “worked examples” anywhere, althought one can normally get a fair idea from the published tables for each poll – it’s something I keep meaning to do using a free dataset like the BES stuff.

  47. Spearmint,

    Actually I discovered that the procedure to get to a general election in my previous post is probably not necessary.

    The opposition can introduce a bill in the House of Lords (to repeal the fixed term parliament legislation). Google turns up several instances of this but I have yet to find a good description of the procedure. It appears there was an opposition bill in 2012 on constitution reform (also in 1911) and also several other non-constitutional measures over the years.

  48. Slow under-arm bowling with a soft fluffy ball by Harriet H today.

    Re footy and Chelsea/Arsenal, I shall hope for extra time tonight, Chelsea lose on penalties with JT faing over, a a Moscow, for the vital miss.

    Followed by defeat to a very fit Everton on Sunday while Arsenal win 3-0 at St James Park

  49. Mike Smithson [email protected] 1h

    LAB lead down to just 3% in the Ipsos-MORI May poll. It’s CON 31, LAB 34, LD 10, Ukip 13

1 8 9 10