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.

ukipdrop

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 6 8
  1. @Jasper22

    I guess you don’t feel the choke hold of centralisation and increasing dominance of the south east then?

    I agree the NHS needs central funding, but can be run and managed locally.

  2. GO has played a blinder with the NHS devolution, not something you can usually accuse him of. To make Greater Manchester, with all its Labour councils, the first region to take this on is a stroke of genius. I don’t know if this will translate into votes but it will certainly discomfort the Labour leadership and make Andy Burnham look as though he does not trust his own party’s councillors to pull this off. I might also say that it is a darned good idea which should be repeated in other areas when they are ready to take it on. How else are you going to integrate health and social care?

  3. The Manchester NHS thing is actually a major cut in the NHS budget of the region, and it was done in a rather ugly way (towards the staff).

  4. @Jasper22

    If you are refering to the Welsh (the correct word I believe) they are free to the service as they want.

    If people don’t like the way it’s run, they can vote for someone else.

    That’s democracy. It might catch on.

  5. @ Martyn

    Basis linear regression. Fitting y = alpha + beta x, with the null hypothesis that the slope is zero.

  6. @LASZLO

    Figures please, give us figures so we can decide for ourselves.

  7. @ Jasper 22. Wales is not a Principality.

  8. @ RMJ1

    Expected cut: 60% of the low band staff (clinicians taking over). they were promised all kinds of nicey nicey stuff during the consultations. Then informing them Friday afternoon that a contrary decision was taken, so the news would’ve be too polluted.

    The decision, by the way, was made purely on statistical reports.

  9. Would’ve = would not

  10. Good evening all from the Southside.

    OLDNAT
    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!” :-)
    ___________

    I think you will poll quite well OLDNAT ;-)

  11. @RMJ1

    Partial devolution for convenience in England is a terrible idea.

    If recent events have taught us anything it is that piecemiel devolution creates far more problems than it solves.

    As to the politics of it, it seems to me a major gamble by GO to try to divide and rule over Labour. If this fails, it will fail spectacularly. Playing with the NHS is not a wise thing for a Chancellor to do.

  12. The reason that NHS devolution is not necessarily a vote winner is that people hate change, even when it is for the better.

  13. @ Neil A

    I still remember when Gynneth Dunwoody won Exeter and Tony Benn was an MP from Bristol, even though I left the UK in 1970.

    What I observed though is that the Green Party in 2009 in the European election came ahead of Labour and then in 2014 there was only 39,677 votes separating Green and Labour.

    A cross check with local government area results then showed that outside of major urban centres Gren was doing better than Labour and LD.

    Then along come some crossbreak polls, just like before the 2014 European election showing Green ahead of LD, except this time they are also even with UKIP.

    Now does that mean Bristol is dominating the respondent sample? While Labour is in the low 20s, I have also seen Green as higfh as 15.

    So where outside Bristol West might the Green vote be competitive and/or of high enough impact to prevent LD keeping seats or Labour from gaining them.

    I assume Conservative will gain most LD seats in the southwest and given current national polling numbers assumed Labour would not win back the Dorest or Somerset seats.

    If Labour are to get more seats than Conservative they will have to win back some seats from Conservative or take vulnerable ones from LD.

  14. NUMBER CRUNCHER

    “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” :)
    ________

    If you receive a visit from someone in Pyongyang then don’t get too excited because that will be me. I have a proxy IP address and it chucks up Pyongyang as my current location.

  15. @ RJM1

    They hate change when it only benefits others.

    The condition of change: the beficiaries of the status quo cannot maintain their position, the losers of the status quo cannot bear it any longer.

    The Manchester NHS was not political, just a bureaucratic madness. So GO could get the blame unfairly. I saw the report. It’s full of misinterpretation of significance test (honestly).

  16. @AC
    “If you receive a visit from someone in Pyongyang then don’t get too excited because that will be me. I have a proxy IP address and it chucks up Pyongyang as my current location.”

    A Pyongyang proxy is not a wise Korea move.

  17. CATMANUEFF.
    Many thanks for your link.

  18. I will take some convincing that centrally funded and benchmarked but locally delivered, is not the way to go, with the proviso that the devolved area is large enough- and greater Manchester is certainly big enough to do it. I can see no other way of integrating the health and social care budgets unless we simply put it all into the NHS and take it away from the councils. You can’t tell me that is the way to go – well you could but you would be wrong.

    No Laszlo, people just hate change, even when it benefits them.

  19. @ RMJ1

    To run NHS properly you probably have to increase the number of managers (including team leaders) to about 5 times. There is no political party that would take it on.

    A decentralised NHS would need even more managers. As there are no resources for that, but they need it, they will try to run the same level of service with a skeleton lower bound staff to be able to pay the managers. It’s actually in the documents …

  20. @ Allan Christie

    I presume it wouldn’t show up, since DPRK is embargoed by the US…

    @ RAF

    LOL

  21. @ Richard et al

    Has their been any cross-comparison by region or constituency between the rise in UKIP vote and level of unemployment, rates of personal and business bankruptcy?

    And in terms dividing UKIP supporters into blue and red in origin I am not sure occupation always tells the exact tale:

    Employed
    Self-Employed
    Business owner

    UKIP, like the Green Party, has been around for some time so why does support take off in 2014 beyond what it was in 2009.

    As Malcolm Gladwell correctly asks in his book “The Tipping Point”, what causes someone to act.

    Declaring someone a “loonie”, “fruitcake” and/or a “racist” does not help us understand what tips a person over from voting Labour or Conservative to UKIP.

    In the 2010 GE in the Northeast BNP obtaned 4.4% and UKIP 2.7%, Yorkshire and Humber BNP 4.4% and UKIP 2.8%, East Midlands BNP 3.1% and UKIP 3.3%, Eastern UKIP 4.3% and BNP 2.1%, West Midlands UKIP 4% and BNP 2.8% and why is the combined vote in the Northwest and Wales lower at 5.3% and 4% respectively?

    What for example is the correlation between impacts of the “recession” and the rise in the votes for these parties?

  22. @Laszlo

    As a resident of GM I was not a little alarmed at the Manchester NHS announcement and am moreso now after reading your remarks.

    As a devout localist I am not against the idea in theory but this close to an election any major announcement by any government smells fishy to me and dressing up a staff and cost-cutting excercise as devolution, as you’re suggesting, just stinks.

  23. YouGov/Times (London only): CON 34 (+2) LAB 42 (NC) LIB 8 (+1) UKIP 9 (-1) GRN 7 (-1)

  24. @ catmanjeff

    Someone put these figures out last night, so what are the implications for LD in terms of seat retention as they are, on these results, losing 63.8% of their support compared to 2010 GE?

    Conservatives are roughly the same and Labour would be up from 36.6%, so how many seats would change hands?

    Does 9% and and 7% for UKIP and Green give them a shot at any seats.

  25. @ Jasper 22. Er no it is not a Principality. Perhaps you should do some homework. Oh and perhaps you could refer to people from Wales as Welsh.

  26. Moody

    Er, yes it is.

    We have a Prince of Wales, always have.

  27. @ THE STATISTICIANS

    I enjoy reading the posts from our statisticians but recently the maths has gone far beyond my understanding. What do you statisticians expect to achieve from the higher level of maths; more accurate interpretation of trends, or better forecasting, or something else? And are your results likely to be significantly different from, say simple graphical analysis?

  28. @Jasper 22
    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.

    I think you need to have a word then with Attlee government, because NHS Scotland has always been administratively separate from the NHS in England.

    http://en.wikipedia.org/wiki/National_Health_Service_(Scotland)_Act_1947

  29. RAF

    “A Pyongyang proxy is not a wise Korea move”
    ______

    Very good lol :-)

  30. NUMBER CRUNCHER
    @ Allan Christie
    I presume it wouldn’t show up, since DPRK is embargoed by the US…
    __________

    There are ways and means of getting round the embargo but they did intercept my Havana IP ;-)

  31. @ PostageIncluded

    I also believe localism.

    I followed the process (it started some months ago) by having some access I came to the conclusion that I expressed.

    The trouble with services we speak about is that they rely on professional norms rather than performance measurements, bonus, etc. the decentralisation will undermine these norms.

  32. @Unicorn

    Hmm, be careful. If your model is linear all you are doing is testing to see if it changes linearly, which is not quite what you want. Here’s a thought experiment: try fitting a linear model to a sine wave – it’ll ghive you a slope of 0. Here’s another one: fit a linear model to the following dataset: {1,1,1,1,1,1,1,1,1,1….1,1,1,1,999}. The fitted slope will be very close to zero…:-)

    It’s 9pm and I’ve been working since 9:30am. I have to go home otherwise I am going to hurl. Consequently I cannot familiarise myself with the Wald-Woflwitz Runs test nor the Jonckheere-Terpstra trend test and so cannot discuss them sensibly. Apologies for that.

  33. @Lazlo
    Thank you for your post. I take your point that p-values and significance testing can be deprecated: don’t the NHS guidelines these days say that confidence intervals should be quoted instead, or have they changed since the new lot got in? My point was that if one quotes a p-value then one has to at least *mention* what the null hypothesis was.

    If it makes you feel better, I sympathise: you are of course correct. But my experience is that as statistical tools become more and more available, the more and more they will be used: if you have lots of hammers, everything will look like a nail

  34. @Unicorn.

    My reply to you is in automod. I’ll try posting it again with different brackets

  35. @Unicorn

    Hmm, be careful. If your model is linear all you are doing is testing to see if it changes linearly, which is not quite what you want. Here’s a thought experiment: try fitting a linear model to a sine wave – it’ll give you a slope of 0. Here’s another one: fit a linear model to the following dataset: (1,1,1,1,1,1,1,1,1,1….1,1,1,1,999)The fitted slope will be very close to zero…

    It’s 9pm and I’ve been working since 9:30am. I have to go home otherwise I am going to hurl. Consequently I cannot familiarise myself with the Wald-Woflwitz Runs test nor the Jonckheere-Terpstra trend test and so cannot discuss them sensibly. Apologies for that.

  36. @Unicorn

    Arse-biscuits, it’s still not getting through. Anthony, can you deautomod it? I have to go home and eat food

  37. @Jasper22

    Wales and the Principality of Wales do not share the same borders.

    Similarly, Northern Ireland and the Province of Ulster do not share the same borders.

    It’s like confusing Greater London, London and the City of London: they are different things

  38. @Anthony Wells

    Jeez, that was fast: thank you. I have to go: bye, everybody.

  39. Anthony, did someone mess up the weighting data on the latest London Yougov

    https://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/ic295thvbe/TheTimes_BorisJohnson_London_150223_website.pdf

    See last page. It seems odd that weighted = unweighted for every single bucket?

  40. @ Tony

    There’s no better data than the one polling firms produced. The trouble is that they have different methodologies, and you don’t have the means to evaluate them beyond intuition.

    As to graphical representation, I’m generally against it, as past, present and future are on them represented by a continuous one, when it’s probably not such (are all oscillations of UKIP random or systemic?). I tend to agree with Lagrange who in his book at the end of the 18th century announced that there are no graphs in his book (it was a major step forward from Newton).

    However, I like Statgeek’s graphs.

  41. CATMANJEFF
    YouGov/Times (London only): CON 34 (+2) LAB 42 (NC) LIB 8 (+1) UKIP 9 (-1) GRN 7 (-1)
    __________

    76% for the two main parties in London. There used to be a time when this was the norm right across the UK for the two main parties.

  42. Why are my comments awaiting moderation? Especially when those of somebody else are not.

  43. Anyway back to party politics. I sense things just drifting towards Con and moving away from Lab. Tonights YouGov will be interesting. In the past few days the polls have shown more Con leads than in the previous few weeks.

  44. @ Laszlo

    It is certainly sensible to introduce checks against over exuberant reporting of so-called significant results, and establishing practices to avoid what your Nature article refers to as “p-hacking” (essentially cherry-picking and data-mining to squeeze at least something out of the evidence. One method discussed for handling this is “preregistered replication” – a process in which the investigator announces in advance exactly what procedures will be adopted to reexamine an idea and what evidence will be taken as corroborative or otherwise.

    I would submit that – far from exemplifying the risks of over-enthusiastic p-hacking – all of my own contributions to UKPR have pushed in the direction of enhanced caution, and I have even unwittingly been recommending a form of “preregistered replication” here (more below).

    As an example of my bias in favour of caution, I have argued on this post that Anthony doesn’t have solid statistical support for his suggestion that the Ukip VI has dropped since Jan 1. In doing this I am not making unreproducible claims. I am pushing in the opposite direction and suggesting that we wait for more substantial evidence before we accept the conclusion that this is what is happening.

    Similarly in late December @CMJ had an exchange in which I challenged his claim that Labour’s support was rising. Again, my emphasis was on caution.

    Turning to the issue of ‘crossover’ I have been concerned that contributors are all too quick to claim that it is about to happen or that it has happened. In response to this some weeks ago I posted some suggestions about possible ground rules we might consider to decide in advance how we might identify a crossover event. It would be an understatement to say that it was not well received, but if you go back to my suggestion you will see that they meet the criteria for ‘preregistered replication’. I was proposing that we avoid cherry-picking by setting our our evidence-evaluation procedures in advance of any particular forthcoming crossover event, and I set out my own suggestions to achieve this. By applying these procedures to test for crossover, I would suggest I am following exactly the kinds of recommendations set out in the Nature paper.

    In short, I would argue that my general approach is exactly in line with your own recommendation and that of the papers you cite. I aim to be ultra-cautious and avoid making claims that might prove to be fragile or questionable: the exact opposite of the style of thinking your criticise in your comment.

  45. @Mikey

    I suspect all comments on that other matter will be weeded out wholesale.

  46. @Mikey

    It’s not that Labour is falling away. It’s more that the Tories are slowly catching up.

  47. RAF and MIKEY.
    I agree about the Tories catching up, picking up people who were UKIP-inclined and tory-inclined Lib Dems.

  48. AW,

    Sorry for comments that were snipped.

    i should know better.

  49. Twitter reporting a poll due later which has “surprises”..

    Lab in front by two?

  50. Considering a large part of the Tory vote shifted over to UKIP I’m not surprised if we see an increased Tory VI and not impacting the Labour VI.

    It’s a double edged sword for Labour. A decreased UKIP VI and a increased Lib/Dem VI must surely be bad news for Labour!!

1 2 3 4 5 6 8