Two new polls today – the daily YouGov poll for the Sun and the monthly Survation poll for the Daily Mirror.

Survation in the Mirror have topline figures of CON 28%(-3), LAB 34%(+4), LDEM 10%(+3), UKIP 19%(-4), GRN 4%(+1). Lots of sharp changes there since their previous poll, but usual caveats apply – the Tory lead in Survation’s previous poll was rather unusual in itself, today’s large Labour lead also unusual, hence the large changes from one to the other. Note also the drop in UKIP support – Survation consistently show the highest UKIP support, so while 19 is large compared to other pollsters’ figures, its a notable drop from Survation.

Meanwhile the daily YouGov poll for the Sun has topline figures of CON 35%, LAB 33%, LDEM 6%, UKIP 14%, GRN 7%. A two point Conservative lead from YouGov, the first time they’ve shown that for just over a month.

The bigger picture remains the same. The Conservatives probably haven’t moved ahead, or we’d be seeing that across most of the polls, when actually they are averaging out at a tiny Labour lead. Neither is there is big swing to Labour, or we’d be seeing that across most of the polls, when actually they all just seem to be showing normal variation around the margin of error. In terms of the Labour vs Conservative race, 2015 so far has been largely static. The only trend that may be meaningful is the drop in UKIP support.


Now that Survation have published their monthly poll we can compare UKIP’s January and February scores across all the pollsters (I’ve taken an average for those companies who publish more than once a month). There does seem to be a pretty consistent fall in UKIP support, perhaps slightly obscured by the fact that the most frequent pollster, YouGov, shows one of the more modest drops and the second most frequent pollster, Populus, changed their methodology at the start of February in a way that increased UKIP support.

359 Responses to “Latest YouGov and Survation polls”

1 2 3 4 5 8
  1. Survation shifting the onus back to the Lib Dems.

    Does beg the question why the Survation official account tweeted on Friday that the polls would be published “next week”. That message was deleted earlier today, as pointed out above.


    “Support for UKIP – as for Front National and others across the EU like Swedish Democrats – is unquestionably linked to economic despair or depression”

    What a sweeping generalisation! I did not join Ukip in 1996 because of “economic despair or depression”, nor because of immigration. For me the Maastricht Treaty was a step too far in political integration.

    Nor were the many people with whom I worked with in Ukip in the following 18 years motivated by “economic despair and depression”.


    ” by 2020 Carswell will have done at least one of the following:
    Left UKIP acrimoniously
    Assumed the leadership”

    I agree but NF has been very successful at removing opposition or any who constrain his position.

  4. Lord Ashcroft re the Survation / LD polls:

    “Well this is #comfortpolling at its worst. Spinning like a top. Should be totally ignored.”

  5. Presumably, the Lib Dems did the polling to guide their spending/campaigns in target seats. This would mean that they would want them to be accurate but would not want the Tories to know the results.

  6. Carswell is a right-libertarian, so I am not surprised that he is pro-immigration.

    The rigth of the US Republican Party includes the usual paleo-conservatives but also right-libertarians (the Ayn Rand worshippers). There is a parallel here with UKIP.

  7. Rand Paul is the standard bearer for that type of Republican nowadays. Some of them don’t feel comfortable with the predominant Republican views so they join (or stand for) the Libertarian Party instead. Their candidate (Gary Johnson) did a bit better than normal in the 2012 election, winning ~1% of the national vote.

  8. @Unicorn

    I shall look over your points tonight and try to get back to you about midnight. One question I do have to ask you is “if h0 is beta=0, then what is beta?” I assumed that if you were doing linear regression then your model would be something along the lines of “y = a + b1x1 + b2x2 + … + bnxn + e”, where y is your outcome variable, xs are your input variables, the bs are your fitted parameters, a is your fitted intercept, e is the error term. Is that your model (or something similar) and what term is your beta?

  9. Mind you I disagree with Lord Ashcroft, I don’t think these polls should be “ignored”. I’m looking forward to dissecting them with great relish (if they ever appear, which I doubt).


    Absolutely. Carswell is a right libertarian. He is really too independent to fit in with any party and doesn’t like by told what to do by party apparatchiks. That’s the main reason he left the Conservative Party. At the moment he can be at home in UKIP because they don’t really have much internal discipline and no whip. If UKIP grow and win more seats, bith nationally and locally, and their organisation grows as a consequence, my guess would be he’ll leave that too.

    Ironically, of course, he is one of the political class that UKIP voters apparently despise. He is a career politician who somehow manages to give the impression he isn’t. His only difference being his independence, which is anothr reason why he can’t afford to get too tied up with UKIP as an organisation.

  11. Second line..being told, not by told.

  12. OLDNAT

    Are standing for Orkney & Shetland on behalf of the SNP?

  13. #Are” you” standing


    I agree with you that Carswell is “independent”. He is a strong advocate of political reform to make politicians more answerable to the electorate.
    So he will never be a submissive party hack-whichever party he finds himself in. In this respect-which is his central raison d’etre- ,he is cut from the same cloth as Zack Goldsmith & Daniel Hannan.

    But he isn’t a “career politician”:-

    “Carswell worked as Corporate Development Manager for Television Broadcasting in Italy from 1997 until 1999, and for INVESCO, reporting to the Continental Europe regional CEO, from 1999 until 2005 before entering politics.”

    Douglas Carswell.

  15. I’m a bit late for the reminiscences about early political memories (mine include the Korean War, the late 50s China/US standoffs over Formosa/Taiwan and other islands in the Taiwan Straits, coupled with TV news items showing H-Bomb tests with maps of what else would be destroyed by a hit on central London.)
    On a lighter note, some years later I remember clearly a schoolgirl who would now be about Natalie Bennett’s age being asked what was 10% of fifty and getting out her pocket calculator.

  16. With the exception of Scotland, where the SNP are clearly in the lead and have the firmest support, it is far too early to predict the outcome of this election as neither Labour or Conservative have yet managed to take a permanent lead.

    As for the Lib Dems it needs to be remembered that they have lost approximately 66% of their 2010 GE support and it is still not clear that any of that support is coming back. Further they have the softest support and I am starting to detect a an oscillation going back and forth between Lib Dem and Conservative, which will not help the “Coalition” achieve majority.

    The leader of the Labour party continues to remain quite unpopular and outside of London and the North the Party has so far been unable get above the 20s and 30’s for support.

    The big drawback for UKIP is that apart from some constituency pockets of support there is as yet no region in England where they come close being second ahead of the Conservatives, or for that matter Labour.

    As for the Green Party the two places where they appear to be competitive with both UKIP and LD are London and the southwest. where in the latter region two pollsters have had them running neck and neck with UKIP for third spot. That only leaves them with the potential to win one more seat.

    Finally, unless and until, voter probability settles down, as it has done for SNP in Scotland, predicting the outcome of this election could turn on a dime.

    In Engand and Wales only Cameron has negative ratings that are below 50%, except for Leanne Wood and Natalie Bennett, who are not known at all by over half and over a third of the electorate respectively.

  17. @Andy Shadrack

    “The leader of the Labour party continues to remain quite unpopular and outside of London and the North the Party has so far been unable get above the 20s and 30’s for support.”

    Don’t most people in England live in the London and “The North” and, therefore, aren’t there quite a lot of constituencies in these regions?

    If, by “the North” you are excluding the Midlands and Wales, then I’m not sure either where you’re getting your view that Labour are “unable get above the 20s and 30’s for support” anywhere other than in London and “the North”. Regional polling in Wales and the Midlands, rather than statistically invalid regional crossbreaks in national polls, doesn’t support your view at all.

    We’ve also had some recent national polls putting Labour ahead of the Tories in England and Wales.

  18. Allan Christie

    He certainly looks like me. We’re around the same age and both been in l0ts of political parties.

    Maybe I’ve been drafted without my knowledge! :-)

  19. UKIP’s internal problems are entirely predictable and indeed often predicted. The trouble is that have become the refuge for two opposing groups, not just at different ends of the political spectrum but at opposite corners of the political compass.

    So you have the socially liberal but extreme economically right-wing ‘libertarians’ of whom there aren’t very many[1] but who are influential in the leadership. And then you have the socially conservative but economically statist and pro-welfare[2] who make up the majority of the support and membership. Both these groups have fastened onto the old core of UKIP which is a classic petit bourgeois Party in the Poujardist tradition: yearning for the past, isolationist and suspicious of both the status quo and any attempt to change it.

    There are similarities between these groups, but they are more about tabloid-inspired rhetoric and a general feeling of being hard done by than any actual view of what should change. So when it comes to deciding what the Party’s policies should be there are bound to be clashes because the groups have different aims and solutions. The only thing they share is that the aims are all unrealistic and the solutions are either unworkable or equivalent to current policies they dislike the consequences of (or both).

    [1] Because these views are basically unworkable in a society of more than one person, but they appeal to a certain type of male in early adolescence, who demands absolute freedom to do whatever he wants until something goes wrong when he wants to run back to mummy. Admittedly some in UKIP’s leadership do actually appear to be twelve years old (have seen their candidate for Boston?) but Carswell is old enough to know better.

    [2] Naturally they don’t see themselves as such because they consider the pensions, benefits, access to the NHS and so on as their ‘right’. The problem is all the other people who are clearly not entitled to such things. On the grounds of their being other people. You will notice that among the few characteristics they share with the first group is an unwillingness to consider how a society actually operates and an ignorance about the facts of the society they actually do live in.

  20. @ Roger Mexico,

    There’s not really an intellectual difference, I think, just a socioeconomic one.

    The libertarian wing believe they should have everything and everyone else should have nothing, but either are or aspire to be rich enough that they believe this can be best achieved by eliminating state redistribution.

    The Redkip wing believe they should have everything and everyone else should have nothing, but are poor enough to recognise that the only way this can be achieved is if the state redistributes money from the “metropolitan elites”, “banksters”, etc into their pockets.

    The fundamental belief about who should have all the stuff is the same, they just (quite rationally) have different views about how to get there.

  21. @Roger

    I agree very much with your characterisation of the compenents of UKIPism, although your very last sentence I think can also be applied (at least by those who don’t agree) to the more naive and altruistic elements of the left.

  22. @Anthony

    Scratch that, the difference is actually because Survation don’t weight by social grade. Should never overlook the obvious explanation.

    Thanks for confirming. Odd that they include social grade in the tables but apparently uniquely don’t use it to weight by. But good that they show it, so we can see how it is skewing their reported results.

    If nothing else, I think it allows us to answer who is right on the level of UKIP support – not Survation – at least not if C2DE’s continue to be over represented in their results.

  23. Richard – they use income instead.

  24. @Anthony

    Ok, and reading here

    ” However there is a strong correlation between income and social grade as the following chart shows.”

    But then why do they seem to have so many more C2DE’s?

    I guess we can’t answer that…

  25. @Roger Mexico

    Like Neil A, I agree very much with your description of UKIP’s two strands of basic support although, stripped bare, aren’t you really, in admittedly far more erudite terms, describing them as a “bunch of loonies, fruitcakes and closet racists.?”

    This was Cameron’s colourful description of UKIP in 2006, something he now much regrets, apparently. Maybe he realised that he was describing a certain section of his own core vote; a vote that he’d now quite like to recover.

  26. @ Crossbat

    So where outside of Bristol, Plymouth and possibly Exeter do Labour expect to make gains in the southwest?

    Where in the southeast do they expect to make gains outside of communities where they have won back councils since 2010?

    Currently I believe Labour controls Hastings, Southhampton, Gravesham, Reading, Oxford, Slough, Luton, Stevenage and Harlow councils.

    In the Eastern region Cambridge and Norwich.

    The results in 2010 in the southeast were 74 Conservative to 4 LD and 4 Labour.

    In the Southwest 36 Conservative, 15 LD and 4 Labour.

    In Eastern 52 Conservative, 4 LD and 2 Labour.

    Looking at Election Prediction, for example, going down to a 33% probability I can only find two seats in the southeast, Brighton Kempton and Hove, and I suppose you would add in Brighton-Pavillion.

    In the southwest Stroud and in the East Norwich South, Norwich North, Lincoln and Bedford. Remeber I said oiutside Bristol.

    The Midllands as always will be a battleground, and please remember i am looking for places where Labour will make enough gains to overtake the Conseravtive lead in seats in England.

  27. @RM

    “have you seen their candidate for Boston?”

    Is it the one on the right who’s the candidate, or the old geezer with him?

    h ttps://

  28. Anthony

    I guess we may be able to answer it if we had the breakdown of income that they are using, and the weights attached to each income group.

    Are they supposed to publish those – from BPC rules

    “A description of the weighting procedures employed and weighted and unweighted figures for all variables (demographic or otherwise) used to weight the data, whether or not such breakdowns appear in any analysis of sub samples.”


    I agree that you make an excellent analysis.

    I personally dislike right libertarianism even more than paleo-conservatism as when you examine it deeper, the “libertarianism” it espouses is completely phoney (even if you are rich). An ideology that advocates acting in a completely selfish way all the time is hardly libertarian. What if you want to share?

    I also note James’ point about it being a minority interest even in the USA (even though just about every one of them seems to have active role on the internet)!

  30. I think the performance of the Libertarian Party under-estimates the strength of libertarianism in US politics. Like most thinking, it tends to get subsumed within one or both of the two main brands (e.g. the “Tea Party” isn’t an actual party in its own right).

    Rand Paul is a live outside bet to be the Republican nominee next year. Barry Goldwater was certainly of that strand of thinking and had some influence subsequently, far more than you would expect for someone who was thrashed.

  31. Guymonde

    “have you seen their candidate for Boston?”

    Is it the one on the right who’s the candidate, or the old geezer with him?

    LOL. Though given the rate UKIP tends to get through candidates, by now it’s probably the lifebuoy.

  32. Colin

    Carswell stood against Tony Blair in Sedgefield in 2001.

  33. Anthony – should that disctinction (weighting by income rather than job type) really make that much difference? Aside from that I can’t really explain the high UKIP VI other than the panel composition…

  34. Anyway, I’ve put some slightly longer-term charts on UKIP and others. Key points:

    – The vast differences in combined LAB+CON vote share between online and phone polls (on average) has gone

    – Relative to each other, Tories do better in phone polls, Labour online, although this month the difference is exaggerated by outliers in opposite directions

    – For the Lib Dems, the pollsters have converged, largely for technical reasons (DKs making their minds up reduces SoS adjustment)

    – For UKIP, the gap is actually widening – biggest drops are with phone pollsters, who were already lower on average

    – The Greens continue to do best in phone polling, but this is quite a big spread between online pollsters

  35. @Andy S,

    Historically in good years Labour can pick up a seat or two in Cornwall. As you identify they will almost certainly pick up Plymouth Sutton & Devonport (the other Plymouth seat, in which I live, would stay Labour even if Ed M was arrested by Op Yewtree the day before the GE).

    Labour have some strength in the south of Dorset, and there are parts of Somerset that ought to be natural Labour territory (for example Bridgwater).

    Labour may not do all that well in the SW, partly for the same reasons the Tories don’t well in the North. Their message feels quite geographically tailored, so that even their natural supporters don’t feel they are quite on “their side”.

    The big question in the SW is what happens to LD votes. My hunch is that they will hold up slightly better than elsewhere, but there will be quite a complex mix of swings. Some LD seats, particularly Cornish ones, might go Labour or Tory, depending on exactly how things split in 3 way contests.

  36. Good Evening All from a cold beach here.
    Many thanks for your link. As you say the Cons seem to have caught up the Labs; swing back has been happening, almost imperceptibly but ineluctably.

  37. @Chrislane

    No problem. I agree, when you’re talking about moves of a point or two a month, you wouldn’t notice amid daily moves of several points back-and-forth.

    It’s the underlying trend that counts…

    The underlying trend is unmistakeable.

    BTW: how does type become bold on UKPR? (Have tried!)

  39. Burnham says he won’t deliver GO’s Manchester NHS devolution deal if elected.

    I’m guessing that this will please GO. I am fascinated by his well received devolution initiatives with Manchester Councils-via Labour Councillors. They are big deals.

    Will they deliver any votes For Cons?

  40. COLIN:
    Good Evening to you.
    Mr Herbert Morrison, Peter Mandelson’s Grandfather, always wanted the new NHS to be locally-based. ‘Nye’ opposed this.
    Maybe Herbert is being vindicated.

  41. CHRIS

    Thanks-I don’t have the knowledge to comment on the probable outcome of the deal for local people.

    I am interested in the politics of it. These are Labour local politicians who appear to be grasping these new powers ( & the economic deal which went before) with enthusiasm.

    What effects on local VI will this have ?

  42. @Colin

    A national budget being spent and controlled locally by democratic bodies is really good localism, and at face value would be supported many (it would certainly fit Green Party thinking).

    There would need to be some national standards applying so a certain minimum care level is sustained. I think this could be achieved without being too onerously bureaucratic.

    My main concern would be giving inadequate funding, and the central government blaming the people running the service locally for a failure to make the right choices. I’ve seen local government be treated like this before – blame the council, even if in truth they didn’t have enough cash in the first place to fulfill the need.

  43. Today’s YG Midlands/Wales crossbreak

    PC 1% : Lab 36% : Con 32% : LD 4% : UKIP 18% : Grn 8% : Oth 1% : WNV 4% : DK 11%

    Mean of last 20 YG Midlands/Wales crossbreaks

    PC 2% : Lab 36% : Con 33% : LD 5% : UKIP 16% : Grn 6% : Oth : 1% WNV 6% : DK 14%

  44. Polling has shown that people don’t like a ‘post-code lottery’ when it comes to the NHS.

    They like their services to be local but prefer uniformity in treatments, waiting lists, waiting times etc.

  45. WNV and DK for informational purposes only…

  46. A few days ago rather foolishly I tried to compose a post on p values and alike while being rather tired (not emotionally).

    However Martyn’s question to Unicorn a rather surprising news encouraged me to try it again, and hopefully in a simpler language.

    The Journal of Basic and Applied Social Psychology banned the use of null-hypothesis significance testing in all submitted papers (, essentially the treatment of the p value, as it is invalid.

    Why did they do it? For the same epistemological problem that I keep on trying to point out, but cannot do it coherently.

    Essentially, most people think that the significance test tests the null hypothesis (it actually applies for the margin of error too). It doesn’t. The significance test tests the data with a given null hypothesis.

    The 95% confidence level doesn’t mean that the null hypothesis has a 95% probability, but that if we take an infinite number of samples, 95% of the samples would capture the population’s attribute.

    But we don’t have infine samples either, and what we have (let’s say the YouGov almost daily reports), carries varying P values. This is a major problem. Even if you have a super p=0.01 sample, the probability of replicating this p value is actually 50%, not the intuitive 99% (here’s a good article:

    Moreover, the more we want capture the trend, the more variables we introduce (eg more parties). However with this we increase the probability of a good p value (if you have a hundred variables, you have a 95% chance that you find significant variables).

    This is why I suggested the Bayesian analysis. This tests the probability of the null hypothesis by testing a priori data (this is why it is criticised – assigning Y/N – 50% initially) and the test data. To my mind for constituency level analysis the non-Bayasian statistics suits better.

  47. @Amber

    You can have certain minimum standards (all the good stuff you mention), but still allow freedom to deliver services as best befits the area.

    For example, as long as the service time/outcome is met, why not allow one area to deliver it totally different from another? Some urban places allows centralisation, due to the distance. However, this can’t work in the Highlands, so if they can find a different way by mobile services, or sharing local building with other services, why not?

  48. Never knew there there so many Americans (or people in the US) reading this site! Someone in Nelson, Washington and another in Bellaire, Texas, on my site right now via UKPR, according to Google Analytics :)

  49. Amber

    Fully agree. Which is why the NHS should be truly “National ” as in under the auspices of the British Parliament, not devolved to Scotland, Wales and NI.

1 2 3 4 5 8