There is a new ICM poll in this morning’s Guardian. The topline figures, with changes from the previous ICM poll in late June, are CON 38%(-3), LAB 34%(+3), LDEM 19%(+3). That’s a bouncing back for the Lib Dems from 16% in the previous poll and the first survey to show the Conservatives below 40 since Populus after the budget.

On other questions, people are pessimistic about the economy, with 51% thinking it is likely to fall back into recession compared to 43% who think it will not. On the defecit reduction plans so far, 38% think they go too far, 39% think they are about right and 16% think they do not go far enough.

YouGov’s overnight voting intentions were CON 42%, LAB 35%, LDEM 15%. Net government approval was just plus 3, the lowest YouGov have shown it so far in their daily government approval figures.

I expect there will be a lot of attention paid to the difference between the Lib Dem figure in ICM and YouGov. The rather unexciting truth is that it’s probably premature to read much into it at all. The previous ICM poll in June was pretty much in line with YouGov’s, with ICM showing 16% and YouGov 17%. All in all, there have been 5 ICM polls since the election, in 2 cases they showed an identical Lib Dem figure to YouGov, in one case they had the Lib Dems lower than YouGov and in two cases higher than YouGov.

The sheer volume of YouGov’s daily polling means we know their Lib Dem score is averaging about 15%, and if you get a 13% or a 17%, it’s likely just noise. In more traditional polling with just one or two polls a month, you can’t really be sure if something is an outlier or not – there may well be no difference here at all (or perhaps a smaller difference than these particular figures imply).

If it persists over time, then I’ll look at it properly, since it’s certainly plausible that there’s a difference. ICM’s “spiral of silence adjustment” consists of reallocating people who say don’t know to the party they voted for at the previous election, so if a lot of former Lib Dem voters are now saying don’t know, this adjustment will help the Liberal Democrats. In Martin Boon and John Curtice’s article on the 2010 election polls they also said they thought they might have been weighting the Lib Dems too highly, and would be looking at it in the future. So I can think of some theoretical reasons why ICM might show a higher Lib Dem score than YouGov, but on the evidence we’ve got at the moment, I’m not certain they consistently are.

We’ve also got a MORI poll due in the next day or two.

UPDATE: Full tables for ICM are here. For what it’s worth, 2010 Lib Dem voters were slightly more likely to say don’t know to voting intention than former Con or Lab voters… but it wasn’t enough to increase their topline figure. In fact, the effect of the topline adjustment was to decrease Labour’s support by 1 point. There do not, as yet, appear to be any changes to ICM’s methodology from pre-election (aside, of course, from the fact that ICM will be weighting to 2010 recalled vote instead of 2005 recalled vote).

69 Responses to “ICM/Guardian – 38/34/19”

1 2
  1. The ‘spiral of silence’ weighting seems to be a logical idea, but is it reallocated differently depending on which party they voted for last time?
    Are ‘don’t knows’ who previously voted Lab’ or Con’ more likely to vote again for the same party than an LD voter who says ‘don’t know’?
    I would guess yes, but is there any evidence to show it? (Any evidence anyway might be completely out of date now with the LDs position being so different).

  2. ICM is possibly a slight rogue poll as it is inconsistent with the daily yougov polls.

  3. Why is YouGov still doing a daily poll? Are they expecting another GE soon?

  4. Julian – good question (in the sense that it allows me to bore on about the minutae of polling methodology – probably not good for anyone else).

    ICM reallocate all don’t knows the same way – 50% of them go to the party the voted for last time.

    Populus do it slightly differently. They reallocate 50% of former Con or Lab voters who say don’t know, but only 30% of former Lib Dem voters, and no former other voters. That’s based on their post-election polling in 2005 which presumably led them to believe that past-vote was a less good predictor of what don’t knows would do when it came to former Lib Dem voters.

    So Populus agree with your view, or at least they did – since it was based in 2005 data and both companies have probably now got nice new 2010 data to consider.

  5. I can imagine there might be a number of Lib Dem voters who would still vote for them but don’t say so as the general impression is now out there that the Lib Dems are performing badly and being blamed for siding with the Tories.

    I suspect that rather than the party shares, the most significant numbers here are ICM’s economic confidence numbers and YouGovs government approval ratings.

    My impression from previous significant shifts in party support is that these are lead indicators that presage voters switching parties, so there is something for both coalition parties to worry about.

    Having said that, I haven’t researched this in depth so I could be wrong, and of course Wayne assures me that the Tories will hit 48% by Christmas so I must be wrong.

  6. This is the sort of poll I’ve been expecting for a while now – Conservatives falling and the Lib Dems rising. As the more unpopular conservative coalition policies start to bite, I suspect that we’ll start seeing a drift of centrist voters from the Tories to the Lib Dems. Of course, that said, it’s just one poll, and is quite probably a rogue. It wouldn’t surprise me however if we started seeing a few more polls like this over the next month or so.

  7. Thanks Anthony. I suspect the LD ‘don’t knows’ is going to be an important factor to consider in the polls for the foreseeable future.
    I would be prepared to bet some of my hard-earned money on the polls being pretty much neck and neck between the Tories and Labour in the near future.
    I wouldn’t bet a farthing on a drift from Tory to LD though.
    Interesting times, as Wayne would say.

  8. ‘and of course Wayne assures me that the Tories will hit 48% by Christmas so I must be wrong.’

    Alec, don’t start him off -he needs no bait!

    Thanks to Amber for your analysis on the previous thread. I have neither the knowledge nor the nous to have produced that and I wondered how you did it if it can be easily explained.

    On ICM we have an awfully long way to go before the polls settle permanently in one direction (IMO). This was brought home to me by the LD newsletters I receive. Today, for instance, I got an email from NC who is ‘proud’ of ending child detention at Yarls Wood (the family unit is to be closed). Forgive me but that seems counter productive in that case. Perhaps the parents are ‘released’ too. Anyone know?

  9. Coalition 57
    Labour 34

    Not much changes.

    Interestingly, in other news:

    Fiscal Doves 38
    Fiscal Hawks 55

    Broadly reflects voting intention. Probably not a coincidence.

    That 3% difference in the two figures may suggest further room for a slip in LibDem support when cuts actually start to bite. On the other hand, if that’s all teh further room for slip then they are not doing badly.

  10. Coalition 57
    Labour 34
    It’s been pointed out before, but it’s not much use to combine the LD and Tory votes if the next election is going to be FPTP.
    Some would argue it’s meaningless in an AV system too.
    The fiscal doves and hawks is more meaningful except this is likely to change too when the cuts start to bite. As the hawks percentage drops, so too could the Tory and LD support.
    All in theory of course. ;)

  11. Stephen
    The local election results do not show up the pattern you produce with your analysis on a ‘coalition’ basis, admittedly with low turnouts, although the latter are the die hard voters who always turn out.

    In other words people do not think in terms of coalition and anti coalition when they enter the polling booth. They still vote for their favourites in the same old way.

    I have opined this before and would like to see polling evidence for this way of thinking, so beloved of contributors to this site.

  12. @Anthony

    I hate to correct you, but re-read the cross-tabs on the ICM poll.

    The headline figures has *NOT* had the spiral of silence adjustments, you’ll note the big bold line saying “THIS TABLE DOES NOT INCLUDE ADJUSTMENT FOR DON’T KNOW/REFUSERS”.

  13. Steve Richards (I think) in todays Independent suggests an interesting idea that Cleggs decision to go with the Tories in such a wholehearted manner is actually rather than a mistake that will lead to the Lib Dems demise, much more of an expression of independence. The act ensures that the centre left will never take the Lib Dems for granted again and means that if we are moving towards an era of coalitions, Labour are going to have to fight much harder to ensure they are able to gather partners.

    Although he doesn’t say so, I guess the implication that Labour could continue to triangulate on issues like justice and civil liberties, safe in the knowledge that the Lib Dems would always back them in a coalition are over.

    This means that while many people are currently highly critical of Clegg’s stance as a sell out, it may well turn out that this one political act forces a much stronger representation of Liberal values across the poltical spectrum as parties key one eye out for potential Lib Dem support. This could end up as a long term political success for the Lib Dems in the same vein that Thatcherism forced an acceptance of the free market on Labour and Blair/Brown’s commitment to public services led the Tories to fully accept the NHS.

  14. @Julian Gilbert

    I’m aware of the fact that the combined Coalition share is not a particularly useful predictor under FPTP. I think it is important to note though as a measure of popular opinion. This is a democracy after all. And although I support FPTP, I prefer it to generally reflect the will of the populace. Further, translating vote % into seats is a rough game at the best of times, so there seems no reason not to measure the actual support for the parties across the ideological divide.

    always generally note the support of the government and main opposition, I don’t think this is any less valid if the government is more than one party. Also, the coalition score has been considerably more stable across polls than the individual party’s score. Further, it underscores my previously stated point, that quite a bit of the despair at the lib dems’ poll collapse missed the point that about half of this collapse has in fact moved to the conservatives and hence in-coalition.

    I think that perhaps the fiscal hawk/dove stats won’t change as much as perhaps people think. Especially if they avoid any more awful cock-ups like over the schools list. We have had so much expectation and hyping of cuts that I think that most of the people who will oppose them on mass (as opposed to being annoyed about cuts to things directly around them), are already opposed. Though, as I said, that % will probably slip a bit further.

    @ woolymindedliberal
    Firstly, I’m pretty sure anthony has covered before how local by-elections are an awful predictor of nationwide political movements. Secondly, local elections are far more local. You vote for your councillor or against him, you don’t think that much about wider commitments. People may think about the coalition nationally, where it is big and obvious, rather than locally, where there is not even a coalition at all and it has almost no relevance.

    The polling evidence for my approach is that we do have a coalition government, it acts as the coalition and is referred to as the coalition and in our confrontational political culture the numbers that really matter (to an extent) are those who support the government and those who oppose it. At the moment that means the coalition and labour.

    Of course this does not mean that the party numbers themselves are not relevant or important. But I think it important to also note the coalition support. That is what the actual government is after all.

  15. @ALEC
    Everything you say is true, if elections were not held under the FPTP system.
    We shouldn’t forget that just because the last election was a hung parliament, doesn’t mean the next one will be. We can actually predict the result of the next election under FPTP. It will be a Tory or Labour majority, not an LD one.
    If NC wanted to ensure independence, he should have demanded PR as a condition of the coalition.
    Unless the LDs are so altruistic they would accept their own demise in order to “force a much stronger representation of Liberal values across the poltical spectrum” Excuse my cynicism, but I doubt that’s the case.
    Not demanding concrete ER was simply a gamble and possibly a huge mistake. IMHO.

  16. @STEPHEN W -“…quite a bit of the despair at the lib dems’ poll collapse missed the point that about half of this collapse has in fact moved to the conservatives and hence in-coalition.”
    Fair enough but it’s comfort for the Tories, not much for the LDs though, is it?
    I said from even before the polls started showing it, that the coalition will be good for the Tories. I’ve also said that the coalition will be very bad for the LDs.
    As Wayne would say, I was right! (so far anyway).

  17. Jay – erm, the pdf does indeed include the figures without the adjustment… but, if you scroll down a page or two more, you get the figures with the adjustment, which are the topline figures published by the Guardian.

    This month the only effect seems to have been to decrease Labour’s support by 1 point, so it didn’t benefit the Lib Dems today.

  18. Stephen W – Rallings & Thrasher’s prediction based upon local government by-elections in 2010 wasn’t that bad this time round actually. It compared well to the opinion polls, but I expect that’s largely to do with the fact that being based on data up to 3 months old it was largely immune to “Cleggmania”, it probably wouldn’t have compared well to opinion polls three months out.

    Anyway, that doesn’t mean that you can look at local election results and get anything meaningful without an awful lot of spadework. Rallings and Thrasher do it by what has become an exceptionally complex calculation. For all by-elections where new parties stand, or parties that did stand before didn’t put up a candidate, they calculate notional shares of the vote based upon that parties performances in the last all out round of elections in that local authority. Changes are then calculated based on the equivalent national share they calculated for the local elections last time it was fought, and finally the Lib Dem score is reduced downwards to reflect the fact they always seem to do much better in local by-elections. Crude calculators based on aggregating up the shares or averaging the changes aren’t of any use.

  19. It will be interesting to see where the MORI figures lie in relation to YouGov and ICM.

  20. I would caution against calling an ICM poll “rogue”. It may however be that it’s at the outside of the MOE. I don’t however give any more credence than Anthony does for the OnePoll figures reported a few days ago.

  21. @ Howard

    Today, for instance, I got an email from NC who is ‘proud’ of ending child detention at Yarls Wood (the family unit is to be closed). Forgive me but that seems counter productive in that case. Perhaps the parents are ‘released’ too. Anyone know?
    They are not. I, too, think it is counter-productive. Either we are a country that provides assylum in a humane & decent manner or we are not.

    IMO, separating children from their parents will be much more traumatic than allowing them to stay together in these circumstances. 8-)

  22. Hi Howard

    My projection of seats lost during boundary changes.

    I used electoral calculus site data for 2010. I considered each region. I divided the electorate by 76,188 (the magic number). Seats with +/5 %, I generally left alone.

    I then reallocated unders (& the occasional overs) to get the regions to their new number of seats. I assumed that the regional seat(s) with the lowest electorate would be the initial loser(s).

    I looked at the seats around the losing seat to see the parties’ vote % & majorities in adjacent seats & made my estimation of how pushing the ‘lost’ votes into new seats would impact the outcome.

    This vote push estimation was based largely on math but where there were obvious non-math factors, I considered them. e.g. In theory, Nick Clegg could have Labour votes pushed into his seat (Hallam is surrounded by red) but I cannot imagine he would lose. Therefore the initial loss to the reds stays with Labour in my concluding seat count.

    There are factors that it is difficult to consider, e.g. would a seat becoming less safe mean parties would ‘get out their vote’. I couldn’t really take that into account because it would be pure guesswork (rather than assumption).

    Nor did I consider the impact of the coalition. I based my estimates purely on the 2010 votes &, as I said, some justifiable assumptions regarding any special circumstances. 8-)

  23. I think if the libdems had got say 15 or 20 extra seats, which some of the polling a few days before the election suggested could happen, they would have been able to bargain for the best PR system possible and whoever offered the best got into no 10. As the LibDem/Labour total wasn’t enough to think about a stable & capable coalition they had to take the only deal on offer – which obviously would only be the “bargain basement” offer of AV.

    I think whenever the next election comes we will find that Clegg has shot his bolt and not got that much for it.
    The TV debates – assuming we have them – won’t boost libdem support that much. We’ll have seen it all before by then.

    LibDem support won’t be as much next time around and I can see then losing approx half their seats. It’ll be an overall majority for someone – probably not that big either, especially with fewer MPs.

  24. @Anthony

    My apologies. Page 7 got filed in with the bumph polling that doesn’t get looked at for me. It’s also a bit annoying, since table 4 is not a normal cross-tab breakdown, so I didn’t notice. And I’m not sure if that has the DK adjustment or not now, and the lack of change for Lib Dems seems to suggest that it isn’t. It’s certainly very close to the non-DK adjusted figures.

  25. Nothing to get excited about here.
    For a bit of fun we could predict when Labour will move ahead of the cons, for 5 polls in a row to avoid outliers as it gets close.
    My money is still on next year after the next budget, so Q2.
    The discussion on here will then be the opposite of what we had in 08 and 09, how big a lead does labour need to be confident of victory. Remember con leads of 15-20 for a long time and a few over 20.
    Being a labour supporter i would want to be over 40 and the cons below 30for a year or more before thinking we had any chance of victory.
    AV, boundary and constituency size chanses will impact our discussion if introduced and maybe tacit collaberation between the coalition partners at the next GE.
    I am assuming the Gov’t will last the full 5 years of course.

  26. I am proposing Attad’s law be added to Godwin’s.

    In any thread about Tory vote % reducing, Attad’s law will be deployed. It will probably be deployed sooner rather than later in the thread.

    A dd
    T ogether
    T ory
    A nd
    D em

    I expect to see it more frequently in the coming months. 8-)

  27. Interesting. Purely on this poll, the Tories would have 293 seats, Labour 289 and the Lib/Dems a mere 40. I wonder what Clegg would have done; given there would be only four seats between the two bigger parties? Prior to the coalition, I would have thought he would join forces with Labour, albeit under a new leader. Now I am not so sure.

  28. @ Anthony

    Daily YouGov polling is great. I hope the Sun & Sunday Times keep this going. I believe that having daily trends will allow us to look back & see which – if any – policies, news coverage & events had a lasting impact on VI.

    Thank you for continuing to cover all the major polling outcomes, even though there is no GE on the horizon. 8-)

  29. @ Valerie

    If it were Brown, he would go Tory. If there is a Labour “true” centralist he might go Labour. Which ever, politics would stay in the “centre”

  30. Alec – Whilst I think he’s as wrong as a wrong thing, I’m sure Clegg DID want a coalition with the Tories for the reasons you quoted.

    Partly because he’s much more right wing than previous lib leaders, but also because he really DESPISED Labour and wanted to show that he and his party would never “tag on” to give them a majority.

  31. @Michael Vaughan

    “It will be interesting to see where the MORI figures lie in relation to YouGov and ICM.


    According to Political betting:

    “Today I’m hoping that we’ll see the latest MORI poll.

    MORI Just out
    CON 40
    LAB 38
    LD 14

    So seriously good news for the reds and bad news for the yellows. ”

    Backs up ICM and YG that show a collapsed Lib Dem vote down to a rump.

    IMHO a rump of orange book Tories and die hards who won’t leave (like the SDP ers who did not desert Labour in 1981)

  32. So the children taken out of Yarls Wood have been separated from their parents. I wonder if they have gone to friends or family, or have been placed with strangers in local authority care? I wonder if any one asked them what they wanted?

  33. @ Valerie

    Given the hypothetical IM seat count, I believe the Dems would have joined a coalition with Labour.

    I’m not even sure that his party would’ve allowed him to hold out for a change of Labour leader.

    At 40 seats, the Dems might have taken the opportunity to have Clegg step aside, if he believed he could not work with Gordon Brown.

    Remember, Clegg only won the Dem leadership with the narrowest of margins & 40 seats would’ve been a very disappointing result.

    IMO, Rather than a Brown decapitation, they would have asked for full PR – presumably with a referendum but with Labour signed up to actively promote a Yes vote. 8-)

  34. @KEITHP -“…they had to take the only deal on offer.”
    A full coalition deal wasn’t the only deal on offer. They could have (should have in my view) gone for Confidence & Supply.
    LOL. ATTAD’s Law has kicked in with impressively immediate effect, hasn’t it? ;)

  35. Combination of latest ICM and MORI monthly polls gives

    Con 39
    Lab 36
    LD 16.5

    Electoral calculus projects

    Con 299 (-8)
    Lab 297 (+39)
    LD 27 (-30)
    OTH 9 (=)
    GRN 1 (=)

  36. Amber – I *heart* you again.

    I imagine anyone combining the Con and Lib VIs does it because they really like seeing Labour 23 points behind rather than 4, but I could be wrong…..

    On a different note, very well done to DC today for getting tough with Merkel/Sarkosy over Turkey’s integration into the EU. They’ve hedged on this for way too long, and given the remarkable economic growth in Turkey, surely Europe should be welcoming them with open arms. That they don’t might start to look like racism….
    On Foreign affairs so far, I like Cameron very much.

  37. Now I *heart* Rob too!!

    I’m liking 40/38/14 I confess.
    In fact, wasn’t YouGov 41/36/14?
    So really all three big guns are very close.

  38. @Amber

    “IMO, separating children from their parents will be much more traumatic than allowing them to stay together in these circumstances.”

    Under Labour we had the sight of children being led away from school by police in front of their friends and classmates to be detained, which was extremely traumatic not just for those detained, but for their classmates as well. Any end to child detention can only lessen the trauma suffered and is to be welcomed. Of course, it would be better if their families weren’t detained either, but an end to child detention is a definite step in the right direction.

  39. @ Sue

    I am rather ambivalent about Turkey joining the EU. ‘Tis okay for Cameron to be in favour. There is no chance of Turkey wanting to join sterling & expect the UK to support its infra-structure expansion.

    From UK business’s point of view, it adds an export market.

    But consider also, many Turks living in Germany & France would much rather be in the UK, where their children would learn in English. Will the UK, whilst under huge public spending restrictions, have the infrastructure to support an influx of people from Turkey & Turks from mainland Europe? 8-)

  40. @Sue

    yes MORI especially good- for that single monthly poll electoral calculus projects Labour as biggest single party on 313- 17 seats in front of the Tories and the LD’s down **41** seats with only 16 !

  41. Germany has historically very close ties to Turkey and Turks who go there will most likely have family connections. I’m not sure that many of them would want to go to the UK.
    DC has surprised me, not for the first time, with the support for Turkey joining the EU.
    Wonder what their rather more xenophobic friends in Europe will think of it?

  42. testing

  43. @ TONYOTIM

    Of course, it would be better if their families weren’t detained either
    That’s what I’m looking for. One parent should be allowed to go with the child(ren). They could have reasonable supervision & maximum support in hostels or special social housing.

    The parent could have an electronic tracker, if necesssary. I cannot imagine being an ‘assylum’ child separated from parents in a strange country, having fled for my life from my home.

    Scotland received many Ugandan Asian’s seeking assylum. Most were not detained in any way. 8-)

  44. Detaining, imprisoning, separating children is wrong. No matter which party is doing it. No matter where they’re from.

  45. @ Amber
    At 40 seats, the Dems might have taken the opportunity to have Clegg step aside, if he believed he could not work with Gordon Brown.
    Remember, Clegg only won the Dem leadership with the narrowest of margins & 40 seats would’ve been a very disappointing result.
    IMO, Rather than a Brown decapitation, they would have asked for full PR – presumably with a referendum but with Labour signed up to actively promote a Yes vote”

    Absolutely. I had forgotten that Clegg’s victory was pretty close cut. I guess it’s the way he behaves and has been treated by the meeja!

    It’s ironic that had the Lib/Dems won fewer seats, they could have been closer to getting what they really wanted. Politics is a funny old game..

  46. Now I think bout it, separation is probably worse for a child than imprisonment.
    I’m sure my son would much rather be in an uncomfortable ‘prison’ with his mum or dad than in a comfortable ‘home’ without us.

  47. @Tinyotim

    I suspect that if you asked children whether they preferred to remain with their parents in a family unit, albeit in a detention centre, or be placed in foster care with strangers, they would choose the former.

  48. @VALERIE
    Snap. :)

  49. @Julian

    You put it much better than me! :-(

  50. And yet this conflicts with how we treat our “own” children. If parents of child are sent to prison, other care arrangements are made for that child, they are not detained with their parents, as that is deemed to be more beneficial and less traumatic to them than locking them up.

    Personally, I would like to see and end to all detention of asylum seekers unless they are guilty of a crime. Its inhumane, but to lock up children is especially abhorrent.

    What was Labour party policy on this at the last election. Oh yes, to increase detention. Enough said!

1 2