Following the Populus/Times Eastleigh poll that showed the Liberal Democrats five points ahead, there is a new Survation poll of Eastleigh in the Mail on Sunday tomorrow that shows the opposite picture, with the Conservative party four points ahead. Their topline voting intentions for Eastleigh are CON 33%(nc), LAB 13%(nc), LDEM 29%(-7), UKIP 21%(+5) – changes are from the previous Survation poll of Eastleigh a fortnight ago.

Both polls were conducted on the telephone and while I haven’t seen the Survation tables both companies tend to use a similar methodology in terms of weighting and reallocating don’t knows to the parties they supported at the last election. I understand they were carried out at about the same time, so it shouldn’t be a “Rennard effect”. The two polls show UKIP with the same level of support, and no significant difference in Labour support – the only difference is the figures for the Conservatives and Liberal Democrats.

I’ll have a proper look when the Survation tables appear, but at first glance the most likely explanation for the difference between the two polls is just the normal variation within the margin of error… suggesting that the race really could be neck-and-neck.

I don’t know if there are any polls of Eastleigh to come – I’ve heard rumours of a poll in the field over this weekend, but we shall see.

UPDATE: The tables for the Survation poll are up here. There are some minor differences in approach (Survation reallocated don’t knows at a lower rate, and didn’t weight up people who didn’t vote in 2010 to as a high a proportion of the sample as Populus did), but none that would explain the difference. Not that the difference really needs a fancy explanation – once you take into account the high level of don’t knows the difference between the two polls can easily be explained by normal margin of error.

389 Responses to “Survation/Mail on Sunday show Tories 4 points ahead in Eastleigh”

1 2 3 4 8
  1. I’m surprised that YouGov didn’t also ask –
    ‘Should Sir Nicholson lose his knighthood?’

  2. “Should stick to deficit reduction (even if growth is slow) – 31 (-5)
    Should focus on growth, even if that makes the deficit worse – 41 (+4)”


    See this is the problem. It demonstrates the intrinsic bias in such thinking.

    The questions are framed so as to suggest growth is liable to be at the expense of the deficit. At best, there’s no suggestion that growth might help the deficit.

    But a big part of the reason for many in focusing on growth, is to help reduce the deficit.

    Because growth increases income via taxes, and reduces outgoings by reducing welfare costs.

    The coalition’s deficit reduction strategy of cuts hasn’t actually done much to reduce the deficit for this reason. Because in the wake of cuts announcements, the tax take has fallen and benefits costs have risen.

    Because the cuts make people and business less willing to spend and invest, and so depress demand for products, so companies get less profits to tax and people were laid off or had reduced wages requiring top up benefits.

    Growth, in increasing our income and reducing knock-on costs, is a deficit reduction strategy.

  3. Double figure Lab lead again. Any perceived connection between this and the performance of the Coalition must be purely imaginary and it should be level pegging in any case. If ‘the good news keeps coming’ it soon will be.

  4. Carfrew
    How would you expect YouGov to phrase the question in a way that seems balanced?

  5. Ozwald,
    I feel confused by your point. Sarcastic or?

  6. @TF
    I am a fan of EM if that helps to clarify a tad.

  7. John Pilgrim

    @”Perhaps more relevantly to the Eastleigh polling and actual voting, what prescription do you see in Moody’s focus on fiscal consolidation?”

    No idea-sorry.

  8. Tinged


    He won’t though.

  9. @tinged

    Suitably phrased to avoid agree/disagee issues etc.something along the lines of. .. which is the better deficit reduction strategy: cuts/austerity? Or growth?

  10. Colin
    With both Labour and Tories attached to Nicholson, I doubt he will too – but who knows what sort of distraction either party will need if things go bad.

  11. Carfrew
    That still implies mutually exclusive categories – that you either have cuts or growth.
    Perhaps it would seem “better” to ask Austerity vs Stimulus spending vs Tax cuts (or combinations thereof).

    “Growth, in increasing our income and reducing knock-on costs, is a deficit reduction strategy”

    It is also the means for investment in future growth, for the wider and fairer distribution of income, and for the creation of and greater access to services and social benefits, all contributing to a longer term deficit reduction strategy.

  13. Tinged

    Fair enough… You could make it a balance question. Not an either/or. ..

    Of course another issue is that Tories argued that the cuts are also a growth strategy. Though that rather stretches credulity given the reality. ..

  14. @john pilgrim

    Yep. The coalition call their strategy a “deficit reduction” strategy but in practice there isn’t much evidence of that. It is rather more a rebalancing strategy, for which there us ample evidence. Rebalancing in terms of public vs private, most notably, but also in terms of taxation/benefits etc.

    How many polling questions do we get on all that?

    If you take housing, roads, railways, health, education, security, social services and benefits to be – which of course they are – part of the economy,, as of course are industrial investment, technology and skills development,.then cuts are reductions in the economy, and reductions in the resources and systems by which to reduce deficits. Policy driven cuts are either, as you argue specifically directed to rebalancing between private and public sector or rebalancing between production and social costs, or they are made without any proven knowledge of their effect on the economy and thus without any knowledge of their likely effect in deficit reduction.

  16. @john pilgrim

    Yes, there are numerous ways in which public spending supports the private sector and the wider economy.

    Which is why it makes little sense to assume more cuts will naturally reduce the deficit. You may make savings in one way, but increase costs and reduce income in another.

    This is not a controversial point: various Tories and LDs have acknowledged this in one way or another, whether Clegg saying they shouldn’t have cut infrastructure so much, or IDS saying capital spending has a bigger multiplier than welfare (another example of rebalancing).

    Perhaps it would be fair to have another question on that. Is it good to shift spending from welfare to infrastructure, if it is better for the economy?

    Regarding the cuts/growth question, might be better to frame it as what’s the best way to reduce the deficit: more cuts vs more investment/stimulus. This bypasses any presumption re: growth.

  17. @Phil Haines – “Lord Ashcroft’s funding for polling has been spared the axe… ”

    In the run up to the 2005 GE Ashcroft donated £26,500 to the Bedford constituency party as part of his targeted approach to marginal seats:


    Bedford is now a Conservative seat, but in 2005 it remained Labour, with the Tory candidate gaining a mere 877 votes. It was LD who gained in that year at Labour’s expense, and any significant row-back from that in 2015 would turn the seat red once again. Who better placed than Lord Ascroft to determine whether that is likely?

    From 2005-2010 Ashcroft reduced his donations to individual constiuency parties but spent many millions on research/polling/focus grouping combined with a switch to funding Conservative Central Office and direct marketing methods.

    Some estimates put Ashcroft’s contributions to the Tory party at around £10 million. It will be interesting to see whether he really has pulled the plug for good and has no office suite in Central Office come 2015 – also whether his polling provides a counterpoint to the private work Mark Textor will be carrying out.

    It was Boris Johnson who recommended giving Lynton Crosby a free hand to run the 2015 campaign: “kill the fatted calf, break the piggy bank, go for Lynton… ”

    Maybe the calculation has been made that new voter profiling software makes it possible to dispense with expertise built up over the years – and that the money can be raised elsewhere.

    “Perhaps it would be fair to have another question on that. Is it good to shift spending from welfare to infrastructure, if it is better for the economy?”

    “Regarding the cuts/growth question, might be better to frame it as what’s the best way to reduce the deficit: more cuts vs more investment/stimulus. This bypasses any presumption re: growth”

    Both these seem to me to be understandable and meaningful in terms of effects of economic policy on people;s lives. I hope Anthony wills think about a yougov poll on economic policy which includes these qestions.

    Adam Afriyie presents his ideas for the budget, calling for a bigger focus on growth (largely through tax cuts).

  20. @carfrew It seems to me that you raise a very interesting methodological question. There are two rather different issues a) how do the public frame the debate over austerity b) given this context on which side of the debate do they come down.

    As you point out, this particular way of asking the question assumes that growth is likely to be gained at the cost of increasing the deficit. As John Pilgrim points out, a realistic framing of the debate would be much more complex than this.

    At the moment I guess that we don’t know how the public sees this debate or even if they have thought it through at all.
    If Oaborne has had his way, people will distinguish between ‘real growth’ that can only occur after we stop frittering our money on state spending, and phoney growth that is the fools gold obtained by further borrowing. Within this context the questions are perfectly fair. If, however, people have a more complex view of the debate the questions are of the ‘have you stopped beating your wife’ variety and assume a context that the respondent may not accept.

    I am not sure what the answer to this problem should be. Perhaps we should stop expecting simple answers to complex questions and ask instead a number which get at the pubic’s perceptions of the different elements in a debate and the stand they take on each. It is likely that the answers will in fact be correlated and that any overall score which is calculated on the basis of them will be correlated with VI.

  21. Economics is fun, ain’t it?

    What cuts in Government spending can be made without reducing activity in the economy?

    (answer, probably none. So we need to look at what we can cut that will make LEAST adverse impact)

    What taxes can we raise that won’t reduce activity in the UK economy?

    (answer: see first question…my answer would be tax rises for the wealthy and cuts for the poor and middle class who will spend it all, spending cuts should be aimed at reducing wasteful private contracts which add no value to the taxpayer but just make rich people richer)

  22. @CARFEW

    The the responents in not aggreeing with either proposition are actually saying: ‘we should have increased spending in order to get growth’ this could be inferred as it is basic keyensian economics that most people know.

  23. Gropegate.

    What did Clegg know & when did he know it?

    Could be fun.

  24. On 5 March, the Health Select Committee will be taking evidence from Sir David Nicholson on the Francis Report.

  25. Colin
    Let’s hope that they give him a good grilling.

    Interview with Lord Ashcroft

    First answer about the importance of polling –
    “I think the single most important contribution polling can make is to be a reality check.”

  27. Tinged

    Yes-I think they will.

  28. @tingedfringe

    Andrew Gimson ends his companion piece (New Statesman) by speculating about his continued influence through a controlling interest in ConHome/TotalPolitics etc, and any future leadership challenge:

    “Ashcroft the pollster is now a convincing enough figure to supply the ammunition that could destroy a prime minister.”

  29. One of the amusing things I always find about the Sunday Times polls is how they reflect the passing indignations of those who run the Press and how they want validation for their views – even though the only people who actually care are a few hundred wealthy types with homes in London and the Cotswolds.

    So we have questions about the RSPCA prosecuting people for hunting (how dare they stop people like us breaking the law!) or whether Hilary Mantel insulted Kate[1][2] or what age IVF should be offered at.

    There are also regular obsessions that crop up regularly such as the evil of wind farms because they ruin the view from one’s weekend retreat[3].

    There’s the way in which questions are asked. For example there’s a question today about the motive for Ed Miliband’s apologising for the abolition of the 10% tax rate. But not about whether people think if should be brought back (and at what level). This isn’t just attempted character assassination, it’s because these people see such actions from the village gossip viewpoint of Westminster rather than looking at what the effects of policy changes would actually be in the real world.

    None of this would matter, except that instead of being minor characters in a novel of social comedy, the UK’s malfunctioning democracy means that these people actually have real power over the lives of other people.

    [1] This gives rise to one of the longest questions I have ever seen on YouGov. As so often, you imagine the meetings, with the ST people wanting to quote very selectively and YouGov saying that they’ve got to put things in context to get a valid answer. Of course the only two relevant things to ask are: (a) why should you be expected to make a judgement on a tiny snippet of an hour-long speech? and (b) either way do you really care?

    [2] Incidentally the YouGov question is wrong because it says “Last week, Hilary Mantel, the booker prize winning author gave a controversial lecture…”. It was actually on 4th February, but no one cared till last week because the Daily Mail had a really slow news day and they had to print something. The lecture’s rather good as it happens and you can listen to it (introduced by Neil MacGregor) here:

    [3] Obviously fracking is completely different because it would only ruin the view in areas where poor people and Northerners live. And Rupert has commercial interests in it

  30. @Ozwald/Carfew

    You make interesting and telling points about the intrinsic nonsense and contradictions buried inside some of these opinion polls and the foolishness of abiding by their every finding as if some unarguable political tablet of stone was being handed down to us from on high. Anthony has already given wise counsel on the tendentious nature of some of the questioning and its capability to skew responses.

    Somebody, I can’t quite remember quite who amongst the multiplicity of posts that ping around UKPR, was telling us the other day that there was a direct link between Government disapproval and the VI ratings of the parties, using this allegedly inviolable “rule” to claim that a Labour lead was an outlier because Government disapproval had remained static. This is nonsense though, isn’t it, and suggests to me the establishment of a “rule” to try and invalidate something you don’t like to be the case, irrespective of whether there is any truth in it. I’ve seen no connection between Government disapproval levels and support for the coalition parties and it seems to me, if you look at the polls over the last 18 months or so that the fairly consistent 10% Labour lead has co-existed with disapproval ratings that have ranged between wide extremes (-19 to -40). Once negative to the tune of -20 or so, it doesn’t appear to matter if it’s -30 or -40 as far as I can tell. Let’s look at the last 6 YouGov polls: –

    Lab Lead 11 App-33
    Lab Lead 14 App -34
    Lab Lead 10 App -34
    Lab Lead 15 App -37
    Lab Lead 9 App -35
    Lab Lead 11 App -32

    So, the lowest Labour lead of the six polls had the second worst Government disapproval rating and, I recall not long ago a Labour lead as low as 8% with a -35 disapproval rating. When the Labour lead has occasionally dipped to very low single figures (5 or 6) then the disapproval has twitched to the -27 or -28 levels, but these polls have proved to be outliers and atypical. Sampling error induced changes, I would think. Certainly over the last ten or so polls there has been no statistically significant correlation between disapproval ratings and Labour leads and party VI ratings.

    I do get amused by the tendency of a few posters on here to make solemn pronouncements, usually quickly disproved, and outlandish assertions based on little more than wishful thinking. Polling for them becomes an endless search for evidence to support their prejudices.

    Anyway, I’m an Opinium man when it comes to Ed’s ratings. Oi, hang on, YouGov still say he’s a duffer. Wait a minute, Ipsos/Mori reckon he’s improving. Naah, Populus reckon he’s second only to Gary Glitter in public esteem………………………………………..zzzzzzzzzzz

  31. @Tinged F
    “Adam Afriyie presents his ideas for the budget, calling for a bigger focus on growth (largely through tax cuts).”

    So a Conservative MP agrees with Balls that a stimulus package is needed now to kick start growth, which means some short term borrowing whether it’s to support spending or tax cuts. The only difference between them is in the form of the stimulus.

  32. @Crossbat11

    Actually, there has been a very strong correlation between government approval and Tory VI throughout this parliament, but this breaks down once they hit their floor of 31-33%. On the basis of limited recent evidence, it looks like this still holds true – the recent temporary bump in Con VI corresponded to somewhat improved approval.

    I would agree though that trying to identify outliers on the basis of this correlation is problematic. There is just too much data noise – the VI range is approx 5 for a given approvaI, but for a given VI the approval range is around 10.

    There’s a similar correlation (and degree of data noise) between Lab lead and gov app, except that here the approval range for a given Labour lead is around 20 points.

  33. @Martyn
    “So basically…we just don’t know”

    Yes indeed, that was the gist of my point, and I agree that a huge pool of undecideds only adds to that. Just to add further to the uncertainty, bear in mind that turnout in by elections is always vastly lower than the polling suggests, which hardly adds confidence when interpreting by-election polling. My gut feeling is that UKIP are now the most likely of the three to come through by a nose, but that “prediction” based mainly on past overperformance on by-election polls has the status of an informed guess from someone whose earlier pronouncements (on Labour’s chances) seem now to be well wide of the mark.

    But for anyone who shares the view that it’s a three way toss up, the current best odds must seem way out of line: LD 2/5, Con 10/3, UKIP 10/1. That implies that the LDs have an 8 times better chance of winning than UKIP.

  34. A UKIP win would be a likely ‘By-election’ win a la Respect etc. So given that the rating downgrade and sex scandal won’t have improved LD or Cons enthusiasm. My best guess would be UKIP (caveat I am not very good at guessing).

    How much damage would Labour coming in 4th do to EM or to Labour VI?

  35. @Couper2802

    None. No one expects good performance from Labour in this seat, it’s not one they seriously contest.

  36. @Couper2802

    “How much damage would Labour coming in 4th do to EM or to Labour VI?”

    None at all. This by-election is all about the coalition, Labour aren’t expected to do well, and not doing well will have no effect. And if UKIP continue to show well in the polls I think we may see a late switch of a proportion of Labour-leaning tactical voters from LD to UKIP. That would be a crushing blow for Con/LD, if in a straight head-to-head neither of them can win.

  37. For Labour, if they can’t win the seat themselves, then a UKIP win would be a wonderful consolation prize!


    Skynews seem to think it will be a devastating blow to EM’s ‘One Nation’ strategy. But then again they do have a DM journalist on. .

  39. Labour in 4th is obviously bad for them, regardless of what Labour loyalists will say. Last week they were wondering at a possible Labour win. That or the Labour voters stopping the Conservatives.

    If UKIP push the Lib Dems or the Conservatives into 4th, what do the Labour supporters say about that?

    There’s your answer.

  40. @couper2802 – “How much damage would Labour coming in 4th do to EM or to Labour VI?”

    I’d say not at all, but who knows? The consensus seems to be that Ed is a slight drag on Labour VI (it would otherwise be 50% according to Lord Ashcroft) – the acceptable “price to be paid for a Labour government”:


    Meanwhile Mike Smithson speculates on the consequence of an Eastleigh upset being that Ed may be the last man standing out of the three party leaders by 2015.

    Now that would have ramifications, my feeling is that for all his faults Cameron would have 5 years expirience as PM behind him, whereas Ed looks young for his age and untried. Fresh leaders in other parties could add or detract to Ed’s appeal – again we don’t know the effect.

    As a postscript Smithson mentions a possible sting in the tail for Nigel Farage if Diane James is the one who makes it to Westminster.

  41. @Phil Haines: So a Conservative MP agrees with Balls that a stimulus package is needed now to kick start growth, which means some short term borrowing whether it’s to support spending or tax cuts. The only difference between them is in the form of the stimulus.

    A stimulus will be far more acceptable if there was some indication of an end to the global slump.

    To start borrowing billions, and it would have to be tens of billions a year to have any stimulus effect, could turn out to be disastrous if the money is spent before a major recovery begins, leaving us back where we started but now with a considerably bigger debt and corresponding servicing costs.

  42. Updated charts folks. Slight Labour rise overall, with a lead of 10.0% on the MAD data.

    Government approval for the month is down (mostly in RoS and Scotland), and the Con/Lab leadership rating are up a little.

    “No one expects good performance from Labour in this seat, it’s not one they seriously contest.”

    Contests or expects to be in with a shout?
    My experience is that throughout the country, regardless of chances of winning, Labour contests every seat seriously – for a number of reasons: 1. The seat may have had in the past or may have in the future, as demonstrated at Eastleigh, a strong support and a Labour member; this is particularly relevant in constituencies like those of the South West, with Lab/Lib/Con marginals; 2. The party base and its activities, including local, Parliamentary and EU elections, are there to get Labour policies across to the electorate and to get them adopted when possible; 3. The party represents in almost all constituencies either a majority or a significant minority of voters and is founded on a principle of making sure that all with the right to representation are in fact represented and can vote for a serious candidate and a serious manifesto and compaign; 4. The party membership and especially its active workers have at least the passion and commitment for politics that contributors to this blog exhibit, and many dedicate their lives to policy debate and campaigning; 5. Candidates make a huge sacrifice of time and often of earnings, to stand, and often do so with a view, often unrealistic but real for all that, to a career in politics; they may have varying capacities and talents but no one in my experience should doubt the sincerity and often quixotic and total commitment, including that of their families, with which they campaign or the strength of support and contribution of time that they have from their branches and constituency parties.

  44. @Steve2

    You’ve nothing to fear but fear itself. But I wasn’t trying to persuade you, just pointing out that support for Osborne’s position is waning amongst even Conservative MPs if not yourself.

  45. @Pilgrim

    “Does not Contest” does not mean “Does not Campaign”. Just that here, Labour know they will not be getting this seat, they are not putting up someone they think should have a seat, they are not spending more than the minimum to be visible.

    And basically, yes, Labour’s expectations here are really really low. They were down at 9.6% in 2010, 13% would be a significant improvement on their voter numbers in this seat even if they do place fourth. (And you can be sure they will point this percentage gain out, and then point out that the coalition parties lost voters using vote count.)


    “Updated charts folks”

    Great blog, superlatively informative and concise, and the very essence of non-partisanship

    ” they are not putting up someone they think should have a seat, they are not spending more than the minimum to be visible”

    Do you have information that I don’t? Their candidate is highly electable, and in other circumstances would improve their vote beyond the increase at Eastleigh.

  48. Here’s a bit more detail on the earlier Populus Eastleigh poll, to explain why I keep banging on about the need to interpret these by election polls without the pollsters’ assumptions about “don’t knows/refusals” being included.

    In the figures giving VI as actually disclosed, 66% of UKIP’s support came from people who had voted Con. Lab or LD in 2010 and had decided to switch.

    The impact of reallocating the “don’t knows” adds the following to the total for each party: Con 23 extra, Lab 6, LD 29 and Others just 1.

    So the Populus published headline VI figures are based on the premise that just 1%-2% of undecideds will decide to vote UKIP. That’s why their VI share fell from 25% in the unadjusted figures to 21% when undecideds were added back.

    Yet of those who have decided to vote UKIP, 66% previously voted for a different party in 2010. Having already persuaded so many to switch, the idea that they’re going to pick up effectively none of the undecided vote is ludicrous in that context.

  49. Questions
    1.What’s the score with the other two credit rating companies have they downgraded us to AA1.

    2.Whats the difference between America’s rating of AA+ and ours of AA1.

    3. Or chile’s AA-

    There’s obviously a difference between AAA, AA1,AA+,AA-, and AA
    It makes you wonder what all the fuss is about if a company can produce so many different ratings for tiny veriations in economic proformance all, of course based on opinion and predictions. And the bigger question why GO thought it was that important to keep on about it.

    I see Belize is on SD whatever that is any idea’s.

1 2 3 4 8