The last three polls, from ICM, Populus and ComRes, have shown slight variation but have all been reporting the same trend: Labour collapsing further and the Liberal Democrats benefitting. The Sunday Times’s YouGov poll however paints a far more static picture. The topline figures, with changes from YouGov’s last poll, are CON 44%(+1), 32%(nc), 14%(-2).

So where ICM, Populus and ComRes have Labour shifts to the Lib Dems, YouGov have pretty much no change (the 2 point drop in the Lib Dems would probably mean nothing anyway, prior to their last poll YouGov had the Lib Dems at 14%-15% for 10 polls in a row).

This obviously begs the question of who is right. Are the Lib Dems up or not? Well, some pollsters are more volatile than others, but ICM and YouGov both produce very consistent figures. We are used to ICM and YouGov showing divergent Lib Dem figures – there are various possible reasons for this but one can normally expect a couple of points difference in their reported level of support. They do tend to show the same sort of trends though – when ICM show the Lib Dems high by ICM’s standards, YouGov show them high by YouGov’s standards (and vice-versa). Besides, eight points is well beyond the normal difference between the two companies.

Normally I would be very suspicious of any change that doesn’t have an obvious explanation – public opinion doesn’t magically move by itself, people react to events. If a party jumps up by a third in the polls when they haven’t really done much or got significantly more publicity than usual, it should ring alarm bells. A six point change in the polls is the sort of thing that normally only happens in either rogue polls, or in response to something big.

For that reason, if it was just the two polls my expectation would be that the ICM poll would turn out to be a freak result, and the YouGov one correct…but in this case the trend in ICM is supported by similar findings from ComRes and Populus. It might not have an obvious explanation, but as I said in a comment on a previous post, perhaps it’s just that Labour’s continuing collapse in support was starting to eat into those people who would never consider voting Tory and see the Lib Dems as their natural alternative. Looking at the details of ICM and ComRes part of what also appears to have happened in both of them is that people who voted Lib Dem in 2005 are much more likely to vote Lib Dem now than a month ago (Lib Dem “voter retention” has gone up from 64% to 72% in ICM, 69% to 84% in ComRes).

Of course, it may just as well turn out that the Lib Dem bounce proves illusionary and the YouGov poll is shown to be right. We won’t really know for sure until we see what happens in the next ICM (should be this week in the Guardian) and YouGov (probably in the Telegraph at the end of the month) polls. We could even, theoretically, find that the difference persists for some methodological reason… but I doubt it very much, if it does we’ll worry about it when it happens!

52 Responses to “A contrasting picture from YouGov”

1 2
  1. I’m glad I wrote last time that we will have to wait for a few more polls to be sure.

    My guess is that the Populus poll is most accurate which shows the Cons 42% Lab 28% and Lib Dems 18%. For this poll lies in the middle of the two extremes from other pollsters regarding Labour and the Lib Dem.

    Hopefully, the picture will become clearer over the coming few weeks.

    Just noticed that the others are around 10% in the Yougov poll.

  2. This rather bears out what I posted on the ComRes thread. ComRes was a bit of a freak result, and I think this YouGov one is too. I agree with Philip that Populus is probably about right
    Con – 42%
    Lab – 28%
    Lib – 18%


    CON 41.8% +8.6% 362 SEATS

    LAB 28.3% -7.9% 215 SEATS

    LD 19.0% -3.6% 42 SEATS

    OTH 10.9% +2.9% 31 SEATS

    a mainly stright swing from labour to conservative and from lib dem to other and the monthly projected majority of 74 to the conservatives should be about right.

  4. I’m in Chicago for the AAAS conference so don’t have access to my full spreadsheet. But it seems pretty clear that the real Labour vote isn’t 32 nor is the real LD vote 14.

    Perhaps people are now so emarassed that they voted Labour last time that they claim they voted LD. This would then cause political weighting to overstate Lab and understate LD.

  5. I am sure that the nature of participating in the You Gov makes a difference. The others rely more on unplanned telephone contact which in the heat of the moment will give a different response to a You Gov panel member who anticpates being asked

    LD support will top 20% at the GE and I am sure will be a close fought thing between Lab

  6. What is consistent is Tory support with a 12th poll in in row now showing 40-45pts. Plus the other 2 main (UK wide) parties 45-50pts combined.
    Pretty much this has been the case since the end of the conference season except a few late in the year when cons support polled a little under 40pts some of which would have been margin of error.
    Accepting that the way these split will make some difference but the key question remains can the cons hold above 40 all the way to a GE.
    If there is a sniff they may not win anti-Tory votes may go where necessary to stop them.
    This is different from 1997.
    Labour is disliked not hated, and whilst the Tories are no longer hated they are not regarded as highly as Blairs new Labour was.

  7. Here we go again !

    A swing to Lib Dems ?-looks like it.

    Lib Dems support level?-who knows?

  8. Just read on AOL homepage an article entiltled ‘HBOS whistleblower:Brown must go’. I think Brown’s fate has been sealed. As long as he is PM the best that Labour can hope for is 28% in the general election.

    There are three people who share the greatest responi
    sibility for the recession in Britain, and they are Bush, Blair and Brown. Finally, this message seems to have a good chance of getting out there. Their failure to lay proper ground rules allowed the overly-indulgent excessive lending and borrowing to take to place.

  9. The interesting part of all this it is only a Poll not an actual election,we will all have to wait until June for a real national voting test, This will give people a chance to vote how they feel without affecting the Westminster Government.
    Labour will do very badly then,the puzzle is how will UKIP do this time? Liberal Democrats were always seen as the party for Europe but as the elections for the Counties take place at the same time this will be a very confusing time for both the electorate and the political pundits.
    The hedgehog in the woodpile is how the BNP will fare
    as they have seen the biggest increase in the share of the vote then of any of the major parties.

  10. @ Jim Garner – “Labour is disliked not hated”

    That depends who you talk to. I know quite a lot of people who currently hate Labour, many of them previously Labour voters (me, for example).

  11. the only poll that matters is the one coming up and the GE next year, if it is next year that is could be later this year yet. the lib dems seam to be holding up well now and this will benifit the conservatives as the lib dems are only taking back votes they lost after the last election so if the lib dems get 23% again and labour poll 25% with the conservatives on 43% then what this means is that the lib dems will have held and the tories will have climed up by 11% meaning a stright swing from labour the conservative a win win for dave & co.

  12. Let’s try again – as i have mentioned previously YouGov POLL predictions are getting less accurate as far as Labour support is concerned – if there were any kind of national election tomorrow Labour would struggle to get between 23 & 25% maximum.

  13. in share of vote but not seats?

  14. Jim Garner – “Labour is disliked not hated”. I who previously voted Labour now hate the party and know others in a similar position.

    I think you should look at the opinion poll results (recent and earlier ) where there is a “forced choice”
    question between either voting for Cameron or Brown. The more recent figures suggest that at the next GE there may well be as much anti Labour as anti Tory voting and there may be substantial un-winding of 2005’s tactical voting to the net benefit of the Tories

  15. Mike – this is based on your alternative measure of public opinion, picking figures you’d like to be true out of your arse isn’t it? Take it to UK Blind Assertions.

    Having seen what readers think about the comments here in the user survey I did a few weeks back, I’m really going to be cracking down on the comments policy at some point in the near future and harshly enforcing the spirit of non-partisanship rule on people who should know better.

  16. Anthony –

    Is it not possible that the divergence in LibDem shares is due to the fact that YouGov don’t apply any likelyhood to vote filter / weighting?

    If it is the case that existing LibDem supporters have become more likely to vote, then this would show up in increased LibDem shares for ICM / ComRes / Populus. However, if there have not been any / many swtichers to the LibDems, then you would not expect YouGov to show any increase in their support (if my understanding of their methdology is right – if it isn’t then obviously my hypothesis is flawed!).

    Of course that doesn’t help explain the sudden increases shown by ComRes and Populus in particular (I think ICM was a part an over-correction as their 16% figure last month seems dubious now), but could this go some way to explain the difference?

  17. Cyberkast – yes, that could theoretically be the reason, and would explain ComRes and Populus just as well, since all three factor in a likelihood to vote while YouGov doesn’t.

    While it is a theoretically possibility, it would surprise me. ICM/ComRes/Populus use more subtle ways of factoring in likelihood to vote than MORI do, so a 6 point difference is probably far too large to be explained by that. Plus, it just seems like implausible behaviour, why should Lib Dem voters massively increase their likelihood to vote all of sudden this month?

    It should be easy to check, anyway, we just need to compare ICM/ComRes/Populus’s poll a month ago wih the ones now and see if there has been a significant increase in likelihood to vote amongst Lib Dem voters.

  18. Anthony,

    Could the nature of the YouGov panel be part of the explanation.

    If we are seeing a desertion from Labour to A N Other, then a fixed panel of people with an above average interest in politics might be less responsive to that kind of trend or even lag behind.

    YouGov may well be the most accurate over time but not the best at picking up a sudden shift in opinion such as a watershed moment for a governments popularity or a late swing in an election campaign.


  19. got to say i wuth the oracle i hate Labour but as much as i would like others to hate them most are just slightly let down by the party give it time the hatred will come and i feel sorry for the guys in the labour party who actually have to the grunt work when it does i remember canvassing in the 90s not a pleasent expereince wouldn’t wish it on anyone.

    the drip drip is turning into a flood Labour are finished now it just needs someone to put them out of their misery

  20. “the Lib Dem bounce”

    I think that is a partisan comment based on short-sighted now-ism, sorry Anthony. If you want to avoid partisan bias, then I think you need to describe events from a consistently balanced perspective to ensure consistency, otherwise such volatility creates for surprises.

    I would argue that this hasn’t been a LibDem bounce, but a return to normal levels of support as the squeeze is gradually relaxed.

    I would also expect a mini-revival in LibDem support north of the border as this narrative takes hold, which would confirm to me that the decisive period for Labour has passed as Obama’s first 100 days come to a close.

    Whether LibDems will now be able to push on above 25% in the lead up to the GE campaign is the next big polling question.

  21. Anthony, haven’t been around the site much for a while but did contribute to the survey and I’m delighted that you are going to moderate comments like those of the ‘oracle’ and ‘onthejob’ which add absolutely nothing to the debate.

    All of us contributing to this board will hold firm opinions otherwise, probably, we wouldn’t be here. Most of us, however, can exercise a degree of objectivity and ensure that our comments have some relationship to the facts as they appear to be!

  22. Anthony,

    You’ve upset Thomas, how can you live with yourself……


  23. DAVID BOWTELL said :- “Anthony, haven’t been around the site much for a while but did contribute to the survey and I’m delighted that you are going to moderate comments like those of the ‘oracle’ and ‘onthejob’ which add absolutely nothing to the debate.

    All of us contributing to this board will hold firm opinions otherwise, probably, we wouldn’t be here. Most of us, however, can exercise a degree of objectivity and ensure that our comments have some relationship to the facts as they appear to be!”

    I AM PLEASED DAVID BOWTELL THAT YOUR OPINION IS THE ONE THAT MATTERS ! Free speech is alive and well in the Bowtell household – NOT !

  24. On the subject of Labour hatred – I have always voted Labour up until now (since 1987 when I was first old enough to vote) and whilst I cannot say that I hate them, I have certainly had enough of them, and will not be voting for them again whilst GB is in charge.

    I no longer think that they can be trusted, and have made unforced error after unforced error after unforced error.

    Haven’t yet decided where my vote will go, but I’ll be watching the other (two main) parties closely over the next 15 months or so.

  25. “whilst I cannot say that I hate them, I have certainly had enough of them,”

    A classic case of “Major Syndrome”.

    Plus ça change……

  26. the point i was making was that Labour are entering the end game they are now unliklety to win the election you will also note that whilst i said i Hate them i also say that most people feel let down by Labour.

    peoples emotions have a very large bearing at the next election if the hatred comes as it did in the 90s then it will not matter what Labour do they will lose and lose big. if this kind of thinking strarts to resonate with the electorate then Labour will facea meltdown and at least three election defeats.

    The lib dem jump in support which is coming direct from Labour is a symtom of this it is the attitude of ” I cannot bring myself to vote conservative but i will never vote Labour again”

    if you want my honest opinion at this stage we have not reached the hatred of the 1990s yet and i would not be surpised to see labour win the next election with a majority on 20 it all depends on what happens in the next 12 months.

    But I satnd by my last comment the drip drip away form labour has stopped and it has turnedinto a flood now the only question is will it be in 12 months or 6 years time either way Labour are finished

  27. ADAM C,

    Move to Scotland and vote for us…..


  28. having looked at my previous comment i can see why you would think it was a partisan

    can i ask a serious question has any party ever come back form a poll showin gthem on 25% to win the next election?

  29. Very interesting point by Nbeale; “Perhaps people are now so embarassed that they voted Labour last time that they claim they voted LD. This would then cause political weighting to overstate Lab and understate LD.”

    Anthony – any thoughts on this? I have heard it said before that in the past polls often underestimated Tory support due to people not wishing to admit it, but then still doing so within the privacy of the polling booth.

    Just for the avoidance of doubt – I am a Tory, out and proud!!

    PS Really pleased you are cracking down on the pointless partisan points – there are other sites that are designed for it!

  30. Peter – I had heard that the SNP are renowned for working for every vote, now I have seen it in practice I am impressed!

  31. NigelJ – an important difference between the weighting of ICM & Populus polls and YouGov ones, which probably doesn’t get as much attention as it should, is the data that is used to weight it.

    YouGov ask people the demographic data they use to weight when they join the panel (updated at various points, obviously). That data is then used to weight them – so a YouGov survey done now, is weighted by getting data collected in 2005 to match the overall picture in 2005 (this is actually a simplication, as people will have joined the panel since then and the target weights adjusted to reflect that, but you get the picture).

    In contrast, phone pollsters collect the data used for weighting at the same time as they do the survey, and weight it to match the overall picture now.

    This means that, were people suddenly becoming embarrassed to admit they voted Labour in 2005, it wouldn’t effect YouGov, since they have their panel’s answers from before they became embarrassed stored away on their servers.

    It could indeed have that effect upon phone polls. ICM and Populus’s methods can adapt and chance to gradual differences in levels of false recall, such as people becoming more embarrassed to admit to voting Labour, but a sudden chance would have that effect. That said, neither ICM or Populus have ever found any evidence of false recall changing swiftly.

    At the end of the day though, if it was happening, we’d see ICM and Populus overestimating Labour and underestimating the LibDems, with YouGov the other way round – so I think we can safely assume that’s not the reason!

  32. OntheJob – they certainly have. The Conservatives were down at 23% in December 1981 before winning the 1983 election, and hit 24% in August 1985 before winning the 1987 election.

    I’m always rather wary of “no one has ever done it before” arguments anyway. There were a couple of rather shamefaced commentators who were arguing before the 1997 election that there couldn’t possibly be a Labour landslide as there had never been a post-war election where there was the massive level of swing that would be required.

  33. I know the aim of this site is to get excited poll by poll. I have been watching local government byelections and those results do not seem to reflect the discussion here, Labour seems to be doing reasonably well in those votes – the tories not that well compared to their commanding opinion poll lead. Does anyone every try to tie together polling with real elections/ Over a month there is a good sample of real votes.

  34. Paul,

    “Over a month there is a good sample of real votes.”

    But it isn’t a representative sample.

    Polls are weighted to create a balance to reflect the country, by elections are random. Just as uniform swing doesn’t tell take in to account local differences by elections are too dependent on them.


  35. Paul – I’m working on a detailed post looking at it. The bottom line however is that they seem to bear very little relation to general election results.

    As a starting point, you need to look at when they were last fought. If they show party X going down, that isn’t necessary a bad thing if the last time they were fought party X was at 50% in the polls, it might still be the equivalent of them being at 40% in the polls.

  36. Anthony, both the 1981 and 1985 example were two years before an election. We’re now just over a year away and Labour is falling again, so perhaps the more pertinent question might be has any party ever hit 25% a year before an election and come back to win?

  37. Paul Smith, local by elections are a VERY dodgy way of assessing a national picture. Most people vote in local by elections on a local issue basis. In local election, people are far more concerned about their rubbish collections than Gordon Brown or David Cameron.

    Last year poster Mark Senior kepts telling us that the opinion polls were over stating the Tories. He based this opinion on local by election results. What happened on May 1st when there was a national local election campaign? Labour slumped to its worst result since 1968.

  38. I had a look at YouGov’s detailed data.

    What caught my attention was the social class figures –

    ABC1 C 48: L 28: LD 15

    C2DE C 38: L 38: LD 14.

    Given recent opinion poll results from other companies showing the Tories ahead with the C2s (and at least level-pegging, or ahead, with the DEs), I just don’t believe YouGov’s figures for the C2 and DE social classes.

    Could the online polling methodology somehow be missing lower-wage (or unemployed) working class people who have switched from Labour to the Tories or Lib Dems?

  39. “But it isn’t a representative sample”

    The people voting in the general election isn’t a representative sample either – tends to be wealthier than the electorate as a whole.

    Although local elections are not representative it is possible to look at shifts in voting from them rather than the absolute number of votes cast – i.e. the swings rather than the results.

  40. “Paul Smith, local by elections are a VERY dodgy way of assessing a national picture. Most people vote in local by elections on a local issue basis. In local election, people are far more concerned about their rubbish collections than Gordon Brown or David Cameron.”

    Where is the evidence for this statement?

  41. Paul,

    But the ones that vote wealthier or not decide the Government so whether they reflect the whole population or not is irrelevant.

    No one really wants to know what the result would be if everyone voted. What they want to know is how those that vote are going too.


  42. Sheesh – there is no point arguing about it. Past local by-election results are freely available on Kevin Edkins site here

    If you think they can be used to reflect how people will vote at general elections, the data is there to back test.

  43. Anthony,

    “Sheesh – there is no point arguing about it.”

    Somebody is grumpy tonight………


  44. The December 1981 figure of Conservatives on 23% was dreadful, but was perhaps the point when the SDP really took off.
    The Crosby by-election was in late November, and although the savage recession actually ended several months earlier, no-one would have believed it at the time. Also, inflation was still a problem and interest rates had to go back up.

    And that 24% figure in August 1985 was a Gallup poll, and they had some outdated methods which were revised at the end of 1996, suddenly reducing 40 point Labour leads to around 18.

    I think there is a difference between hitting a dire figure like 25 or 23 per cent on a few slightly freakish occasions, and settling there for several months.
    The latter happened quite a lot from about 1993 – and the government did not recover.

  45. Gordon is an idiot, Ken Clarke is a fiscal god, Vince Cable would make a better leader for the libs – whew sorry I was feeling repressed with all this non partisanship stuff :-)

    Is anyone aware of local government overspend this year that may stop almost all non essential capital projects next year and specifically in context to this site, does this have a bearing on any ballot systems that were due to be replaced? Sorry but thought someone on here might know.

  46. The question has been raised whether it is possible for Labour to recover from 25%.

    Well firstly, we should remember that at present Labour are averaging 29%. And more importantly as Gin pointed out is there enough time left for a recovery?

    When the polls for the February before the year of the last GE are examined we find Labour to be averaging 35%, and throughout most of the year there is a high degree of consistancy. A slight increase occurs around the end of that year and Labour finally gain 36% of the votes GE.

    If the same pattern is followed then Labour will end up getting 30% of the votes at the next GE. But given the declining state of the economy it is hard to imagine that Labour’s popularity will not decline with it.

    A leadership contest may occur in the summer. But this is high unlikely unless Labour are averaging 26% or less. And while a 10% improvement in 10 months is not impossible, it would be very difficult, certainly an awesome achievement!

  47. I think if they recovered it would be because the economy does turn out to be less severe than forecast – or more specifically, shows some tangible signs of stabilising and shoots about a year from now.
    How receptive people are even to a few shoots when they’ve been through another year of this is a bit of a tall order.

    I suppose making the Tories out to be a do nothing/bankers party remains an option, but do suspect that option is rather running out/is likely to prove insufficient.

    But, looking at it the other way round, the Conservatives have not yet won. And those who avoid the worst consequences of this recession could see more money in their incomes even if wages flatline, although utility bills could rise again.

  48. We also have to factor in events between now and the GE. At the moment the Tories narrative has finally found traction and Brown is clearly a man under fire.

    12 months down the line we may find that Brown has been an instrumental player in this summer’s G20/Obama project to build a new economic system, and that Cameron’s policies are completely isolated and out of step with the new reality being enacted across Europe.

    Thats not to say that Cameron’s tactics may not work – its just that if he finds himself increasingly having to explain how he is right and everyone else is wrong its going to be very hard for him. The one thing most pundits seem to agree on is that the Tory vote isn’t very hard yet – easy to take current polls and say “its a shoo-in” but weren’t polls radically different before Christmas?

    Events dear boy, events….!

  49. I’m noticing that for the past week or so, the BBC has been paying a lot more attention to what the Tories are saying and doing. Tory proposals are being headlined now rather than shunted into Item 5 sidings.

  50. James,

    They’re beginning to realise they can’t stop the Tories from winning I suspect!

    Though I will doubtless be called a sceptic there is a great deal of evidence of the BBC being Labour/Liberal biased.

    Tales of broadcasting house being strewn with champagne bottles after the ’97 election are numerous and Vince Cable is practically treated as their economics correspondant.

    With an incoming Tory administration and tighter purse stings among the general public in a year or two the idea of a reduced TV license or even, god forbid, sharing the pot with the private media companies will be a popular option.

    I suspect a less partisan approach is being pushed by managers now. I doubt it will help them though, the fervour for ‘change’ among grass roots Conservatives will see the power of the BBC curtailed come what may.

1 2