As promised there are two new polls tonight. YouGov for the Sunday Times have topline figures of CON 40%(+1), LAB 27%(-2), LDEM 18%(-1). The second poll is by ICM in the Sunday Telegraph; their topline figures are CON 40%(-2), LAB 29%(nc), LDEM 19%(nc).

So there are no major shifts in either poll, ICM continues the trend of a narrowing in the Conservative lead, while the move in YouGov is in the other direction (in fact, YouGov looks more like a reversion to the mean than anything, prior to their last poll they had shown Labour at 27-28% for four polls in a row).

While it is not statistically significant at all (a move from 39% to 40% from one poll to another really doesn’t mean anything), politically it’s significant that after several polls with the Tories sub-40% and in hung Parliament territory, that both polls have the Conservatives up to 40% again and with a lead that would translate into a majority. It doesn’t necessarily mean much, but it will work against a hung parliament narrative establishing itself.

Some interesting stuff in the other questions in both polls. YouGov asked some questions about whether the Conservatives were seen as the party of the rich, and about potential tax hikes and cuts. On the issue of the Conservative party image 52% agreed with the statement that the Conservatives are still the party of the rich, with 31% disagreeing. It was largely a partisan response though, 90% of Labour supporters thought so, only 14% of Conservative supporters.

On taxation, YouGov continues to find the public opting for public spending cuts over tax hikes (by 52% to 30%). If there are to be tax rises though, putting extra taxes on the very rich remains as popular as ever. Asked whether taxes should be spread evenly across the population, or concentrated on rich people, 66% go for the latter. YouGov also asked about the Conservative proposals to recognise marriage in the tax system, and found the public pretty evenly divided: 48% of repondents supported the idea, 43% disagreed.

ICM meanwhile asked about people’s belief in man made climate change. They found 52% of people thought that climate change was happening and that humans were largely responsible, 39% of people thought that it had not yet been proven that it was man-made, while 7% did not believe the world was warming at all.

This isn’t vastly different from the Populus poll on climate change taken in November, suggesting no vast change in opinion. However, the questions were probably asked in a different way and giving different options, so we cannot be sure. We really need a truly comparable question in order to see whether the recent leak of emails has shifted public opinion on climate change: a question on climate change asked exactly the same way as one before the leak.

151 Responses to “New ICM and YouGov polls”

1 2 3 4
  1. There’s an ICM and a YouGov poll out today.
    New polls and averages up at

  2. No significant difference in my trend model yet from these yet. Still predicting a slim Conservative majority.

  3. I think it’s still the case that if Labour are doing better in Scotland than elsewhere as some recent polls have shown, it probably means the Tory majority would be slightly higher than most of the calculators are indicating.

  4. The Tories are still grounded and can’t rise above the 40 mark.My opinion is that they are now stuck on 40 as their best effort and that the polls will begin to narrow over the next few months.A hung parliament now seems more likely as these last two polls only confirm what we knew.

  5. My trend model is also showing that the Liberal Democrats are making slow gains. Perhaps to push them back up to 22-23, which would put all those assumptions about “the marginals” into the rubbish bin as any ‘edge’ they had dissolves into Libdem defectors.

  6. Andrew

    Can you please check the LDEM figures from the latest ICM in your Tables. You appear to have them at 13 rather than 19. Does this affect your projected results?

  7. I’d wondered why Andrew wasn’t commenting on the apparent slight strengthening of the LibDem figure.

    Interesting how the Telegraph reports this. Instead of reporting that the Conservatives have still slipped down in comparison with their past ICM polling, they lead with “Tories hitting the key measure of 40 per cent”. And only a couple of paragraphs down mention that Labour have gained four points while the conservatives have lost two points over their last ICM poll. That’s worthy of a ‘Gold Star Standard of Reporting’.

  8. LIN REES “The Tories are still grounded and can’t rise above the 40 mark”, This appears to be a rather silly analysis. What evidence do you base your forecast on apart from 3 polls showing the Tories just below 40% followed by 2 showing them at 40%? I have no idea whether or not the Tories will benefit from the problems Labour may face from the Iraq enquiry, the new problem associated with the Hutton enquiry & next weeks public sector borrowing estimates but it is naïve to assume the Tories “cannot” rise above 40%.

  9. Since 1987, most polls have overestimated Lab support and underestimated Con support also the Con vote was hit hard with tactical voting in the 2005, 2001, 1997 and even 1992 general elections, therefore if the polls point to a 20 seat majority for Con it could be more like 50 or 60. Also most pollsters believe that the Cons need to be 10 points on front of Lab to win a slim majority, it would possibly be a slim Tory majority with an eight point lead as it was in 1992 if what I stated above is correct.

  10. @Mike: See my article here –

    “Going from bellow 39 back up to 40” is a misleading reading of these results. The ICM poll, rightly or wrongly, has been giving higher figures than the Ipsos-MORI. Comparing a ICM poll to an earlier Ipsos-MORI poll will make it appear that the Conservatives have made a gain. While if you compare an Ipsos-MORI poll taken after an ICM poll, it looks like the Conservatives have lost out.

    This is why you either compare only within on polling outfit, or compare across a moving average trend.

  11. LIN REES your comments look even more unreliable when one examines the equivalent 2008 polls:

    Con 38
    Lab 33
    LD 19

    Con 41
    Lab 35
    LD 15

  12. The country is not mad – there will not be a hung parliament.

    As I said in my last post, we’re seeing polls with a CLead >12% – the WMA is 39:27:18. But the retrospectives show that the last YouGov, which looked out by 2.2, was only out by 0.8.

    153 days to the next election and over the last 153 the average CLead has been 14.5%. There is a weak (R2=0.47 downtrend of this lead which if continued would indeed leave the CLead around 8% at the time of the election, but I’m very sceptical. I think we will see a reversion to the mean. Brown’s stupid lie about Spain being in the G20 can’t have helped and the “Class war” tactic is doomed IMHO.

  13. @Mike

    Your post rather backs up Lin’s point. In what can only be described as one of the worst years for a governing party in a generation, and the main opposition sees virtually no change in its polling numbers over the year.

  14. Anthony,

    Can you confirm whether the polling averages include the 2.6% of the electoral vote that comes from Northern Ireland? As far as I’m aware they don’t, but I’ve been looking into seat projection methods lately and I wanted to check I’m using the correct percentages.


  15. quite a few people point to the fact that the Tories are averaging 20% in Scotland as opposed to 40% across mainland UK as saying the Tories must be doing far better in England.

    However as Scotland only makes up under 10% of the mainland vote ( most polls don’t include NI’s 2% of the Uk population), then if the tories are getting 40out of a hundred and 2 of that is in scotland then they are getting 38 out of 90 elsewhere.

    That works out at 42% which is better but not vastly so.

    The battle will be one in the key english marginals, but the fact that Scoland only acounts for 10% of the vote and is a four rather than three party system means that you really can’t use the comparison between UK and Scottish figures to show the Tories doing significantly better than they are.

    Having said that they are still on targets for a 50+ majority, but that will be overwhelmingly from English seats.

  16. Is this really a trend model Jay Blanc, or wishful LD thinking?
    There is no evidence atall that the LDs are matching the levels of support in the last Parliament or at the 2005 election.

  17. Lin Rees

    The Tories are still grounded and can’t rise above the 40 mark.My opinion is that they are now stuck on 40 as their best effort and that the polls will begin to narrow over the next few months.A hung parliament now seems more likely as these last two polls only confirm what we knew.

    As Mike and others have said before him, there is no evidence that Labour gains on tories in the closing months before an election.

    If the tories have a more effective front man, which I think they do and if the tory electoral machine is stronger than 5 years ago, which I think it is, then surely they should be able to reach 42% by May.

    If the Lib dems can further strengthen their vote, then labour may be lucky to reach 25%.

  18. We still have others at 15, +7 on the “norm”, seemingly keeping Labour in the doldrums of the 20s and the tories in their own doldrums, the 30s.

    SNP progress contributes 2 points to others, maybe, but 4 or 5 points may must be disillusionment with the main UK parties in england.

    Do people think some of these voters will return to the mainstream? If so, where?

    I notice that 4 points on the tories moving average gives them a majority of 110 whilst 4 points on Labour gives us a hung parliament. If the split is equal, it favours the tories.

  19. As the Tories even got more votes than Labour in the last election in English constituencies, I reckon a 40% overall vote should be enough to give the Tories a decent majority.

  20. Mark M –

    No, all polls are Great Britain only, excluding Northern Ireland (so for comparisons with actual election results, make sure you are using just GB figures).

  21. While labour will be happier than a month ago, they will still be very worried that after a ‘good’ month their own vote is not really moving much. This could yet happen, but while they have been successful in softening the Tory vote they will not be competitive unless they firm up their own support by a good few points.

    The poll data on how the Tories are seen is significant. The ‘class’ war could easily backfire on Labour, but this finding shows there is something to work on if they can get it right. Its interesting that while Cameron lables these attacks as silly and vinditive, the Telegraph reported yesterday that 14 out of 15 shadow cabinet members who went to comprehensive/grammar schools named their school on the CCHQ website, while only 3 from 17 who went private named the schools. This included one who said she went to her ‘local school’ but declined to mention it just happened to be Cheltenham Ladies College. Clearly the Tories have some nervousness about this issue themselves, but Labour will need to be careful to craft an attack thet doesn’t turn people off.

  22. Its totally unscientific but the way i’ve always looked at who will win a UK FPTP election is to sum the squares of the party percentages and each parties share of that is the share of seats they will get.

    On this basis the Tories will get 367…. a majority of about 90…..


  23. I have noticed over the last 4-5 elections that the actual results tend to be slightly better for the Tories than polls show and lightly worse for Labour than the polls show. Is this a fair coment?

  24. Eric – yes, it is.

    The caveat is that the polls have been improving over that time, from 1992 when they were grossly skewed towards Labour, to 2005 when they were very close to the actual result.

    The pollsters have further refined their methods since then, so there is no guarantee that any Labour bias will remain in this election’s polls.

  25. @ Anthony in his opening remarks – heaven forbid that anything alters the prevailing pro-conservative narrative – let alone the published figures from ICM which clearly show Conservative support going from 45 to 42 to 40 in consecutive polls. But that is a triumph according to the narrative ! I regret to say that this will only get worse as we enter what will be a very bitter election period. We need objectivity not wishful thinking and distortion.

  26. To me, this forthcoming election looks and feels like the 1979 election. The similarities are manifold. Callaghan, who had taken over from Wilson mid-term, had dodged an opportunity to go to the country the previous autumn, there had been the debacle of the IMF bailout, the winter of discontent, plus sundry other failures of labour in government – not least the fact that Callaghan lost a vote of confidence on the Scotland Act which forced an election not at a time of his choosing. There was a new tory leader whom Callaghan thought was untried, had a plummy voice, and (shades of toffs!) was a woman. Voters, he thought, would turn against her. Incidentally, there was also a new Liberal leader too. Although Callaghan was more of a gentleman than Brown is, the campaign was rough and was fought on many of the same grounds as this coming one will be. Mrs T won an overall majority of 43, but the swing was not uniform. The further north, the smaller the swing. In Scotland, they swung against her. In 1974, the voting % were Con 36%, Lab 40%. In 1979, they were Con 44%, Lab 37%. I would expect Cameron to get a majority of at least 50.

  27. So out of every ten people

    4 support the Conservatives
    3 support Labour
    2 support the Lib Dems
    and 1 supports some other party

    Not much of a grounds for anything other than a National Coalition really.

    Shame party politics prevent such an idea being put into practice.

  28. Gotta say, I agree with ALAN W in as much as this should be about realistic analysis and not wishful thinking. That said, I can’t see the excitement of the labour group everytime a poll comes out showing a drop in torie support. It’s like saying ‘YIPPEE we are not going to lose by as much as we thought we were’.
    My feel from the media today is a much more positive one for the Conservatives who seem to be offering a fightback. As said on this and previous blogs I think the whole ‘tory toff’ thing is a mistake and will backfire on labour.

    When you are in a hole you don’t care about the background of the person pulling you out.

  29. @Dont tell em Pike

    Such a good start: “Gotta say, I agree with ALAN W in as much as this should be about realistic analysis and not wishful thinking.”

    Shame you then go on to wave your own particular flag.


    “a National Coalition ”

    What on earth would be the point of that?

  31. The Tory party have always been the party of wealth. No change
    whatever has occurred in that department.
    What has changed is the perception of the non chattering class’s
    regarding all politicians as self serving swine. This backfires on the party of “fairness and compassion for OUR PEOPLE” more than it does on the Tories, even though the pigish greed displayed by both was 50/50.

    The days of Winston Leonard Spencer Churchill vs Ernie Bevan,
    patrician born leader vs working class hero, are long gone.
    If the Conservatives can do better than this lot, we dont care where we went to school.

  32. @David in France, you could say the same thing about even 97….I’d hold me nose and vote for either party if they promised a swtich to first alternative, or some sort of borda….or rather, I wouldn’t… I live in a safe seat and realistically don’t get a vote.

  33. @David in france

    Not so much as flag waving but more of an observation.

    There seems to be a almost hysterical reaction (from both sides) everytime a conservative drop in voting intention is mentioned. I feel many other factors will influence this election not least of all the number of party activists willing to canvass door to door etc.

  34. @ALAN W
    Please excuse my stupidity ( I am a Tory) but where do you get a drop from 45 to 42 to 40 in consecutive polls for the Tory Party?

  35. According to Anthony’s list the ICM tally for Conservatives is

    44 42 42 40
    27 25 29 29

    Lib dems
    18 21 19 19

    Not a 45 in sight. All within margin of error and with 42 being a plausible average. However ….

    Why should the lib dems drop 3 points (indeed after going up 4 points)? Why does nobody comment on the wild fluctuations in the lib dem vote. 3% change on a 21% share is a lot.

    The concentration is on the tory share but if we are getting these wild fluctuations in lib dems – well is there something funny with the polls? Seems to me we see labour down, libdem up and vice versa.

    Just how confident can we be in internet polling and telephone polling with such small samples?

    Thanks to the UEA I am quite suspicious of statistical manipulation.


    “Thanks to the UEA I am quite suspicious of statistical manipulation”

    That puts you in Ed Milliband’s category of – “flat Earthers trying to undermine the science.”.

  37. King Harold,

    Notably, Ernest Bevin never became Labour leader. Instead, the Labour leader at the time was the son of a lawyer who was also a trained lawyer. So it was more upper class vs. professional class.

    David Cameron does break the mold of “From Humble Beginnings, just off the M1” that has been a key criterion for Tory leaders from Heath onwards. That’s why class politics feels so archaic: the Tories were loathed to push it vs. Tony Blair (Michael Howard did it only once, I think) but most importantly there was no narrative like that to be used against Heath, Thatcher, Major and the like. The closest was the way people made fun of them (Heath in particular) for their sergeant-accents.

    Will it do well with the British public? I suppose Labour have nothing to lose at this point and it’s as good a distraction from other news as any. The important thing will be to avoid looking desperate, which will require good media management of the sort that was used over the Lisbon treaty. For this reason, it’s probably best if it’s not done by an incompetant like Brown, who fudged things up at PMQs. It would be best if one of the many working-class Labour ministers used it.

  38. It comes to something when the party of government, once lead by a Public School boy who espoused education & aspiration; now mocks Public School boys in HoC , ruins education & decries aspiration.

    We will see how much of a vote winner this turns out to be.

  39. Trevor – why don’t people comment on the Lib Dems? The brutal answer is that people who aren’t Lib Dems don’t care much, most people look at the polls in the light of whether the Conservatives or Labour are more likely to win.

    Of course, in terms of whether either party will get a majority, how well the Lib Dems do is actually of huge importance. How much the headline polls can tell is about that though is a different question. The Lib Dem swing is less uniform (though not that much less uniform, they are not immune, and if they are down by 3% nationally, it has to come from somewhere), but a lot will be down to personal votes, tactical voting and so on.

    How confident can we be? Roughly speaking, with a sample size of 1000 we can be 95% confident that the figure is within 3% of the “real” figure.

  40. @Bill Patrick
    Yes Bill I am aware Major Attlee led the Labour party at the time.
    However I so admire Ernie Bevan (and so did Churchill) that he always seems the sort of self made man who has a right to represent the working people of the 30s, 40s and 50s.
    How very different to the home life of our own dear Prime Minister.

  41. Several people have made comparisons with the coming electoral battle and what happened in the 1979 and 1992 elections.Whilst there are some parallels, there is also one significant difference which now makes the task for the Tories more demanding than on either of the earlier occasions – the fact that in 1997 and 2001 the LibDems achieved a major breakthrough largely at Tory expense. The LibDems now hold something like 35 seats which were comfortably Tory in both 1992 and 1979. Whilst the Tories are likely to regain some of them , I would expect the LibDems to retain most – so that the Tories end up being deorived of at least 25 seats they relied on for a majority in 92 and 79.. Without those 25 seats the Tories would certainly faced a Hung Parliament in 1992 – and probably in 1979 too!

  42. KH, Do you mean Ernie Bevin or Aneurin Bevan?!

  43. Graham – I was in North Devon a few weeks ago which has been LibDem for years.

    Places like Illfracombe were a dump a shocking dump, with the evidence of it being a benefits dumping ground clear for all to see. I would not like to be either a lib dem or labour defending that record.

    So I think LibDem seats are as vulnerable as any. After all what has voting LibDem given us – 13 years of Blair and Brown and a deficit and national debt to make your (and your grandchildrens) eyes water

  44. GRAHAM

  45. @ALAN W
    Your remarks regarding the latest polls are really more silly than they at first appear. The Tories have perhaps lost a point or two,
    since the late summer, then again it may well be normal variance. There is no ongoing trend showing Labour catching up in these figures.
    Also your comment last night regarding the huge media bias in support of the Tories really does display utter desperation.

    Before your desperation leads you to mock the Tory inability to exceed 40%, look at where they came from in the public perception. Like or lump it, Labour are now facing the same situation. You would do better working on a redefined future than hopeing against hope that “something will turn up”.

  46. Anthony,

    Does the ‘within 3% of the “real” figure’ apply to all the parties within a poll?

    i.e. if the Conservatives are at 40%, then the 95% confidence is that they are within 37-43% and if Greens are on 3%, then the confidence interval is 0-6%?

    Or, does this apply to only the more well supported parties?

  47. If I am in Ed Millibands bad books then I must be doing something right.

    Blair was a public schoolboy,
    Foot was a public schoolboy,
    Callaghan sent his daughter to public school
    Gaitskell was a public schoolboy
    Atlee was a public schoolboy (I think his son and certainly his grandson went to public school – Stowe , as did George Monbiot, so AGW was born on the playing fields of Stowe)
    People like Dianne Abbot send their children to public school.

    But with 20% of people in Scotland being technically illiterate – does Brown care?

    Given a 5% swing either way and with 1 high number for ‘others’ I think we should expect these swings. Difficult to place too much credibility on a single poll.

  48. “with 1 high number for ‘others’”

    sorry – ‘with A high number for others’

  49. Mike R – nope, it depends on the size of the figure, there are lower margins on lower figures. Technically the 3% margin only applies in full to a figure of 50%.

    I wouldn’t worry too much about it though – technically the figures also only apply to a genuine random sample, as opposed to the quota samples and quasi-random samples the pollsters actually use. It’s best to treat the margin of error as just a rule of thumb.

  50. Oh shock horror I forgot
    Balls went to a public school.

1 2 3 4