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 5 6 7 8
  1. How the Tory campaign is going from 10 MP’s

    “Panic would be far too strong a word. But there is a puzzled unease about why the polls aren’t moving – and candidates in the marginals are much more aware of polling in their individual seats.”

    ““I’m a subscriber to the 1992 theory: in other words, that the electorate will come to a decision very late on. It won’t take a definitive view before Easter or the Budget”

    ” I note that CCHQ has picked up the anxiety about seats outside the 40/40 that some of us have been going on about…and is now referring to the 50/50″

    “There’s that moment in the long-distance race when the bell sounds, and the man who’s going to win gets his head up and ‘kicks’ – in other words, goes for it. I don’t think that moment’s come yet, or that it will until after the budget.”

    ““.. not enough people are feeling the benefit of recovery, too many others have defected to UKIP and will stick with them – and the “party of the rich” problem remains. But there’s a very limited amount we can do about this in the short-term.”

    “But it’s true that there’s still this privilege problem, so it’s important that there are some measures in the budget that help to tackle this”

    “It’s not at all bad on the doorstep. The Labour vote is very soft, some UKIP support is coming back to us ”

    ““I’d say the campaign is not bad – we could do with being a few points up in the polls by now…bearing in mind how unpopular Miliband is, we really ought to be further ahead.”

    So pretty much the same as it is going on UKpollingreport. Nothing moving.. we are all waiting for the ‘big event’ – is that the budget, is that the debates…or are we just going to continue ‘as is’….

  2. You might have a long wait for the debates-cant see them happening .

  3. Thank you Jonathan for reminding us that there are regions other than Scotland that matter in the GE ! A Martian visiting this site might think otherwise …

  4. @Alan “Exactly the same applies to the question, “is the UKIP VI declining?””
    The question being asked is not that one, but
    ‘Is the UKIP VI shown by the polls less than it was?’
    “If there was no variation, there would be no need for statistics and statisticians” (Snee, Int. Statistical Review 1999, p. 257).”
    What that really means is that there is no need for statistical analysis if random errors are small compared to the size of the measurements, or the accuracy needed. That is why carpet fitters use steel tapes, but polling companies have to use statistics. Graphs are a means of presenting their results in quick easily visualized form.
    What people reading the results of polls do is take their inaccurate results and make unjustified deductions from them.


    Thank you for your long and interesting response at 10.54pm yesterday. I look forward to more statistical analysis in the weeks up to the GE.

    Thank you also to other contributors who have entered into the discussion. I accept that interpretation of graphs is “in the eye of the beholder”.

  6. @Barney Crockett

    The problem with passing all the NHS funding over to English local government is that if that became a general trend it would make such pooled NHS funding far easier to cut without central government being accountable for that. English local government has suffered just about the highest % funding cut of any set of services over the past five years. It’s easy for central government, because the decisions on precisely where those cuts fall are ostensibly taken locally, and it becomes easier for central government the more its popularity wanes and such that authorities are not run locally by the Conservatives. e.g. “It’s all those Labour authorities (and the Greens in Brighton) that are responsible for those 15 minute care visits, not us, don’t you know?”.

    There is a history of services being passed over to local government when central government finds it hard to control costs, only to be cut subsequently. We had “Care in the Community” in the 1980s and “Supporting People” (supported housing services for the vulnerable) in the last decade. Both services are now on their knees as their initially ring-fenced funding has been subsequently transferred to the general pool and then cut drastically as part of the pool, without central government ever having to admit that it was cutting those services in particular.

    Having spent my early working years up to 2006 in English local government, I am not normally one to decry localist solutions. But in this case, the funding risks to the NHS are enormous. So rather than give the NHS funding to local authorities, I would go the other way and give what remains of local authority care funding to the NHS, with the caveat that the local health and community care boards that would run those integrated services should then be run by fully elected members and be subject to equivalent standards of openness and accountability as those for local authorities now. It would be a lot harder to cut funding to local NHS and care boards than to cut funding to local authorities.


    @”The problem with passing all the NHS funding over to English local government”

    I don’t think that is what is proposed in Manchester.
    CCGs & the local hospitals will be involved as they are now-but Local Government will now have input too.

    As Simon Jenkins writes this morning :-

    “The catalyst has been the gulf that has opened in care for the elderly, with GPs and hospitals under siege from NHS Direct and an ageing population, and local care services trying to pick up the pieces with ever-shrinking budgets. Past concessions to GPs and pharmaceutical companies are draining even a “ringfenced” NHS of money, while council care has been cut by a third. The result was predictable: blocked beds and jammed wards. GPs flush with cash hurling patients at hospitals, which cannot cope. Emergency services take the strain. Cost and chaos ensue.”

    The two key figures who persuaded GO that this could work, and remove barriers between local councils , and between NHS care & LA care are council leader Richard Leese and his chief executive, Howard Bernstein.

    The total budget involved is three times the £2bn of spending over housing, transport, planning and skills training which GO devolved to the same city region.

    I read that Balls has said he will offer this to Leeds/Sheffield-whilst Burnham says he will scrap it.

  8. A simple Q&A from the Indy:-

    As Barney says upthread, this is really significant, both politically & economically.

    More details on Friday at the launch apparently.

  9. 07052015: Well Ed Miliband wants the debates to happen, and if there is a feeling that the Conservatives need to make something happen (as conservativehome reports) then David Cameron should want them to happen to.

  10. @Colin

    I stand by my point and remain more than a tad sceptical as to Mr Osborne’s real fiscal motives here, assuming that this is not just a short term pre-election gimmick.

    Of course, local authorities will jump at being given responsibility for running anything, regardless of the consequences. And presumably Mr Leese has visions of becoming an elected mayor with sole responsibility for it all, which was what Osborne was pushing for Greater Manchester. It’s exactly the same basic motivation that led Nick Clegg et al to be wooed by the ministerial trappings of red boxes and chaffeurs even though the LDs were then marginalised from real power, sacrificing the alternative of having real political clout through distancing themselves and putting pressure on the government from outside the tent.

    I think Burnham is right to be sceptical.

  11. Number Cruncher

    I though your blog post on the differences between phone and online polling very interesting. There’s one little ambiguity:

    […] phone polls are an attempt to approximate a random sample, but are limited by landline coverage

    Actually some of the phone pollsters try to include interviews by mobile. Ashcroft’s report say that “Half of the interviews were conducted by landline and half by mobile phone”. ICM get 15% of their sample by mobile. I’ve a feeling that MORI do as well (10%?), though I can’t find any detail (their methodology reporting is rather scanty) and I can’t find any details about ComRes either as their methodology section went missing in their recent website redesign. The site is less chaotic than when I last looked so hopefully it will reappear sometime.

    The position of UKIP with regard to online and phone polls is an interesting one. It was very noticeable during the rise of UKIP that online polls reported the rise first and the phone ones took a long time to catch. This was particularly strange in the difference between the two ComRes polls where there was often a five or six point gap between phone and online despite similar weighting etc otherwise in 2013.

    At the time I thought there were actually two problems. The online polls were too high due to panels having too many of the sort of people who switched to UKIP even after allowing for other factors[1] (the opinionated basically). But the phone polls were also too low with there being a ‘shy UKIP’ factor, perhaps particularly among women.

    This gap has diminished over time but never entirely vanished (as Anthony’s examination of house effects shows). Presumably voters got less ‘shy’ as voting for UKIP was seen as more normal and felt easier about telling phone interviewers they were supporting them. Some online pollsters (notably Opinium recently) may have got better at correcting for panel bias. Survation remains the outlier, but Richard’s examination of their SEG weighting may give us the reason for this. The oft-discussed issue of whether UKIP was prompted for or not doesn’t seem to have made much difference.

    However there’s nothing to say that a ‘shy UKIP’ factor may not emerge again. The Party has been getting a fair bit of bad publicity with the TV programmes (fact and fiction) and this could influence less declarative voters to say they are less sure they will vote for them. In Anthony’s summary above the phone polls tend to be showing the biggest drop (if you accept that much of Opinion’s fall is methodological).

    When you also consider that the smallest UKIP changes were with pollsters averaging two or more polls, while the once a month ones were less stable, it all suggests that any fall in UKIP’s VI has been quite small if anything at all.

    [1] The exception was Populus, but their ‘raw’ figures show exactly the same effect and only their ridiculous weighting reduced their headline figures well below the other online pollsters (this has now mainly been corrected).

  12. Today’s YG Midlands/Wales crossbreak

    PC 1% : Lab 41% : Con 29% : LD 4% : UKIP 17% : Grn 7% : Oth 1% :

    Mean of last 20 YG Midlands/Wales crossbreaks

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

    2010 General Election Actual Result (of Midlands/Wales combined)

    PC 2.6% : Lab 31.6% : Con 37.0% : LD 20.5% : UKIP 3.4% : Grn 0.5% : Oth 4.4% :

  13. Well Ed Miliband wants the debates to happen, and if there is a feeling that the Conservatives need to make something happen (as conservativehome reports) then David Cameron should want them to happen to.

    The debates represent a huge gamble for the incumbent. Ed’s image among the public (if polls mean anything) is that he is not very convincing, when compared to the Statesmanlike Mr Cameron. The debates have a factor built in that this theme will continue.

    What if Ed doesn’t have a ‘Nat Ben’ performance? As long as he is just steady or mediocre, that would exceed the perception of him at large. Unless Dave wipes the floor with him question after question, the Prime Minister’s perceived superiority halo might well slip a peg or too.

    I have no view on what would happen, or what I would want to happen. It’s just obvious to me the debates are bigger risk the Prime Minister than the would be Prime Minister.

  14. The FT mentions a 75 billion hole in the NHS budget that has to be filled in with efficiency savings. As it’s rather difficult to persuade virtual, bacterial and other illnesses to cure quicker, it means cuts.

    The document I saw was a statistical analysis (modelling) how the same service could be delivered with fewer people.

    As it is so sensitive politically, no failure is permitted in the Manchester experiment, so no money will be spared if necessary.

    Every single major drives in NHS in the last ten years has been about cost cutting. Then a later study shows that actually it cost more (especially if the cost of management consultants is included (although the going rate in the NHS is lower than in the Treasury).

  15. Hi all, polls tight but overall a small Labour lead still…. I still wish people would not refer to 1992 as a comparison, in 1992 the tories had a majority of over 100, as things stand now no party has a majority , there could be a comparison with 1979 but there was an appetite for the tories then after 5 years of a more or less hung parliament….cant see that at the moment, I still compare the present situation with February 1974…..


    Ipsos Mori do include mobile-only respondents in the VI polls. I know, because I have been one of them.

  17. Stardaz
    I see what you’re getting at, but there are more parties this time. In Feb 1974 Con+Lab got 94% of the seats. In 2010 it was 87%. Even if they get as many between them this time, it means that there will be an extra 7% (about 43) MPs from different parties than there were then.

  18. PHIL

    I think that’s a pretty cynical assessment of what the Manchester politicians are trying to do.

    But if you believe that everything must be managed from Whitehall, the idea of one , common, local approach to the totality of health / social care is not going to appeal, I suppose.

  19. Returning health and social care to a local authority – well innovations abound – once we do the same for Water, Electricity and Gas we will have come full circle – oh and local taxation too of course!

    Revolutions it appears often end up taking you back to where you were at the beginning…..

  20. @Colin

    “But if you believe that everything must be managed from Whitehall….”

    That’s just about the polar opposite of what I said. I’ll repeat it:

    ” I would go the other way and give what remains of local authority care funding to the NHS, with the caveat that the local health and community care boards that would run those integrated services should then be run by fully elected members and be subject to equivalent standards of openness and accountability as those for local authorities now. It would be a lot harder to cut funding to local NHS and care boards than to cut funding to local authorities.”

    In other words, devolve control to a locally elected board for NHS and care services.

  21. Hmm, I wonder whether the news today about immigration to UK will lead to a resurgence in UKIP VI with attendant decline in Con (and Lab) VI?

  22. As ever, this is NOT a place for policy debate. Knock it on the head please.

  23. “Hi all, polls tight but overall a small Labour lead still…. I still wish people would not refer to 1992 as a comparison, in 1992 the tories had a majority of over 100, as things stand now no party has a majority , there could be a comparison with 1979 but there was an appetite for the tories then after 5 years of a more or less hung parliament….cant see that at the moment, I still compare the present situation with February 1974…..”

    my sentiments exactly. the tories still lost 40 odd seats in 92…If they lose 40 seats in ’15 they are done for.

  24. Dave

    Firstly, I’d consider the terms declining and “less than it was” to be fairly interchangeable. Yes there is the pedantic subtley about “Is it STILL declining” so perhaps “Has the UKIP VI been declining over the last month?” would satisfy anyone who insists on making that distinction.

    Secondly, and more importantly, I’d say my question was the more accurate, what interests people is what is happening in the country, not what is happening in a weighted sample of 1000 people.

    Polling is about inferring the former based upon the latter. Polling aims to answer the question “what is the VI on a national basis?” not “what would the polling result be if I took a poll now?”

    Being able to infer properties of a population from a small sample is fundamental to making measurements of a population so large measuring the entire population in unfeasible.

    Polling is essentially the art of trying to correct for the fact that taking an unbiased sample is impossible. After that, normal statistical inferences are made about the population.

  25. @ Peter Crawford

    I agree. Perhaps the 2015 GE will resemble 1992.

    Perhaps there will be a late undetected surge of support for the Conservative party.


    Perhaps there will be a 2% swing to Labour and The Conservatives will lose 40 seats and Labour gain 40.

    Actually, whatever happens will probably resemble 1992 in some ways and be radically different in others. Personally, I think that the differences will outnumber the similarities.

  26. I have just been looking at what the polls were saying 10 weeks prior to the last three general elections

    End of Feb 2010 Con +6 Outcome Con +7.3

    End of Feb 2005 Lab +6 -7 Outcome Lab +3

    End of March 2001 Lab +15 Outcome Lab +9

    In each case, therefore, the final 10 weeks saw a swing to the Opposition.

  27. I suggest that the reason that the polls have not moved much for Labour and the Conservatives is that both parties have lost nearly all their floating support to third parties — they are both at their minimum floor of support (well, I think Labour is there, and the Conservatives are a couple of points above it) and, barring a complete catastrophe, cannot lose any more. The bad news for Labour is that they no longer can hope to benefit just from Conservative failure. The bad news for the Conservatives is that even at its lowest ebb, Labour’s support is still stronger than their own.
    As things stand, the two parties together seem likely to take about 66% of the vote in the election. The only way for this to change is for either of them to make a strong positive case for their own party. More likely they’ll be reduced to badgering third-party voters about supposedly wasted votes.

  28. @Jonathan

    I’m always a little sceptical of regional crossbreaks in national polls, based as they are on statistically invalid sample sizes, but I think the general gist of the data you’re sharing about the Midlands and Wales does rebut the claim that Labour are only electorally competitive in London and the North, albeit those are rather populous parts of the country.

    There’s no doubt that Labour polls poorly in the less populated and rural, non-urban, parts of the South and South West, but the result of the 2015 election isn’t going to be decided in those regions where there are relatively few constituencies and where even fewer are marginal. Tory heartlands in the main and they voted Tory in the 97 and 01 landslides too.

    I think what’s starting to emerge from both the national and regional polling is that if Labour were maintaining their 2010 performance in Scotland then they’d almost be home and hosed by now.


    Only just saw your post as I was penning my own. I’m seeing the current electoral and political situation in much the same way as you.

  29. Personally I don’t give much credence to the various attempts to use statistical jiggery pokery to project an election result from today’s(or yesterday’s or tomorrow’s) VI figures. The electors will however be (as always) swayed by a number of known factors and a few unknown ones that have yet to raise their heads. I think we can assume today’s even Steven numbers represent the likely present position.

    The immigration figures are likely to move voters rightwards but from where and to where is not clear. The economy is likely to gradually gain friends for the Conservatives as people start to feel better off but most other issues are already factored in to people’s VI. The budget may have an effect and is the one big card the government can play. The debates, if they happen, are, barring the possible meltdown of one of the leaders, only likely to entrench present positions as they will not be such a novelty.

    I don’t believe that the other stories that everyone gets so excited about, I.e. NHS devolution, cash for access, enquiries into everything, etc. Will have the slightest effect . All rather boring really. It does, however, suggest to me that there will be a slight upwards movement in the Conservative VI numbers and a probable, as per usual, further rise when it comes to the actual vote. Where that leaves us in terms of actual seats is another matter entirely.

  30. As we approach the actual election, I feel sure the smaller parties will get squeezed and we shall see a return to the Big Two.

    Who gets what from these returnees is the 64 dollar question, my hunch is the Tories will benefit most.

  31. CB11

    Geographic crossbreaks are always an unreliable guide to regional variations in party support – though, even without internal weightings, they will still pick up particular party strengths/weaknesses in them.

    The Scottish situation is different only because the activity of a particular party exactly matches the geography of the Scottish cross break.

    Where a pollster (YouGov and Populus) weight by Party ID, then that enhances the likelihood of the SNP’s VI being appropriately measured by aggregating samples.

    A similar situation would apply in Wales, Cornwall and Yorkshire, if

    1. their unique parties were weighted for
    2. they were displayed in precisely those geographic crossbreaks.
    3. party strength is more than a few %
    4. crossbreak sample size is large enough to be aggregated in a sufficiently short period to even out moe

    Still nothing like as good as a proper poll, of course.

  32. @ Graham

    All perfectly correct, but if you look at the three elections before that, 97, 92 and 87, the reverse is true. Each produced a swing to the government of between 2 and 4% in the final 10 weeks. You pays your money….

    I find David’s analysis persuasive. Both parties are at their rump or core (or close to it). The election will likely be decided by how many UKIP supporters return to the Conservatives or how many Green supporters will vote Labour in marginal constituencies.

    The final factor, which I haven’t seen discussed much here, is the incumbency factor for Labour MPs in Scotland. I don’t think that Ashcroft’s Scottish constituency polls detected much but that might change as the great day draws near.

  33. @ Unicorn

    We have an eight hour time difference, so the sun is just coming up here.

    In response to your observation a 1point distortion for the Conservatives or Labour in these polls is a makes an actual 3% difference in the value assigned to them.

    A one point difference for Lib Dem or Green right now can mean they are tied or one is ahead of the other or vice a versa, but the actual value difference on reporting 7 instead of 8 is around 14.3%.

    For UKIP such a drop in polling value diminishes their support by up to 10%. For the SNP with these Scottish crosbreaks a 1 point distortion means a change by about 1:5 and for Plaid Cymru aound 1:2.

    So why show cross breaks at a sub-UK wide level at all versus doing a proper poll on the southwest, Scotland, Wales, etc.

    Yesterday’s You Gov crossbreak for London had UKIP on 13, but a London YouGov poll for the Times, taken back to back, had UKIP on 9. At some point these pollsters lose all credibility with me.

    In 2001, using a volunteer, I organized polling in four constituencies in British Columbia by simply taking the phone book a having phone 1 in ten names within the polling area.

    Not only did we get the order in which candidates placed for the various party right we were pretty damn accurate on percentage of the vote too.

    These samples pollsters are using are far too small to be accurate, so I admire Ashcroft for doing his constituency polling.

    And one last thing when Party A fears losing all it’s seats and writes up the questions to get the outcome it wants, and then pays polling company B to conduct the fieldwork, but not pay them to either write the questions or do the statistical analysis on the results of the fieldwork, that is not appropriate and accurate polling – that’s yet another distorted view of reality that should be publicly challeneged.

    I rest my case and would, if I was an MP, propose a bill banning polling during the writ period. Imagine an election campaign in which we had to focus on the campaign itself and what was beng said by leaders and candidates, instaead of trying to second guess what Grandma Huggins down the street might vote or not vote.

  34. Little Red Rock

    It is persuasive, although there is the alternative that it’s not the con/lab waverers that have jumped to the greens but people from what was previously the “core”. Same goes for UKIP.

    There could still be a decent size of population who could conceivabley vote for either party.

    What once used to be considered the floor from either party might well be different now. (Even if a large portion is coming from non voters, the dilution effect will still lower what was considered the floor by a few %.)

  35. @Little Red Rock

    Not in 1987. In early April that year the polls were putting the Tories 12 – 13% ahead compared with the actual outcome of a lead of 11.8%
    !997 – like 1979 – did see a swing to the Government , but that probably reflected the fact that they were trailing the Opposition by such a wide margin. Under such circumstances some rallying was hardly surprising.
    I do agree that 1992 represents the best precedent for Tory optimism – though it is a bit lonely and occurred in the context of a fairly recent change of PM.

  36. @Sam

    “The subsidy is unreal. It is the additional amount that would be paid (by both the renewable sector and the O&G sector) if VAT was levied at 20%.”

    A tax expenditure is a subsidy. The candidate was articulating a perfectly reasonable policy analysis position.

  37. Any thought form people on what 300,000 net migration might do for UKIP? Probably the headline the tories least wanted.

    Oh and more than two years ago NHS Highland & Highland Council signed up for joint services and budgets.

    There are legal complexities which mean each entity is still responsible for it’s own spend, but effectively the two swopped/transferred staff so that the Highland Council through Children’s Services (combined Education and Social Work) is responsible for Children and NHS Highland took on Care and support and provides all services for Adults and the elderly.

    Lots of teething problems and culture clashes particularly on the NHS side, but it seems to be working and leading to improvements.


  38. Little Red Rock

    The final factor, which I haven’t seen discussed much here, is the incumbency factor for Labour MPs in Scotland. I don’t think that Ashcroft’s Scottish constituency polls detected much but that might change as the great day draws near.

    Well the Ashcroft Lab-SNP seats actually showed Labour polling exactly the same in total (37%) for the SVI and CVI questions:

    Obviously there was some variation but nothing more than a point or two in either direction in any constituency.

    On the other hand the rating of the SNP went up by an average of two points between the two questions. Which is odd because the difference is meant to reflect on the relative strengths of the candidates and SNP candidates had only just be selected if at all at this point. It may be that the in total sitting Labour MPs had a slight negative incumbency effect which was balanced by a tactical switch from some Conservatives.

    The other factor could be the presence in some of the areas of sitting SNP MSPs, though the constituencies do all match up. It’s also worth pointing out that the biggest increase was in Glasgow East which had an SNP MP 2008-10.

  39. @ Alan,

    That is a very good point. Just because 30-35% is the old core it doesn’t necessarily follow that the 32-33% currently supporting Lab and Con are those same people.

    However, when I look again at AW’s churn analysis for 2012, 2013 & 2014 I see almost no net churn between Labour and Conservative, which makes me think that those in either of those camps are unlikely to cross to the other (they may go elsewhere of course).

    So I am prepared to accept that Lab and Con may not be at their lowest possible levels but I don’t expect to see much Lab to Con or Con to Lab traffic in the next 10 weeks. I still think its about whether either party can make it to OM territory by adding 3 or 4% to their VI from “others”.

  40. @ Little Red Rock

    On what basis would those currently considering voting for “other” go to Labour or Conservative? These voters attachment to Labour and Conservative is at best marginal.

    Particularily in the North, Midlands and Eastern regions a whole swath of voters first supported BNP and then UKIP.
    They have been joined by a whole new swath of voters supporting UKIP in the soouthwest and southeast.

    One way to determine where a voter might actually go is to ask them who they have voted for over, say, the last six elections they have particpated in.

    If they respond BNP or UKIP every time then it is far more likely that they may go back to BNP than Labour or Conservative.

    That is why asking who their second choice is, is so important because then you know who they are choosing between.

    In Scotland SNP have become a “mainstream party”, so you have to ask the question are there other parties in England that are becoming “permanent” and “mainstream”.

    UKIP and Green come to mind. It is ludicrous to suggest that a first time voter between 18 -21 will “swing” to Conservative if they actually have never voted.

    And I hear youth get quite angry when the suggestion is made that their vote for a “minor” party is a wasted vote, because from their perspective the two mainstream parties “are fossilized and out of touch with their reality”.

    So a better way to track Green Party support might be to look for constituencies with higher than normal 18-39 populations and for UKIP higher than normal 55+ populations.

  41. Little Red Rock

    I agree that there has been little movement (or at least the amount of movement to the other parties has dwarfed the lab con switches)

    There is an element of strategy here, if the major parties try to claw back their votes exclusively from others we’ll see much smaller swings than if one party successfully appeals to the waverers.

    Bear in mind the doubling up effect still makes a interparty swing worthwhile, even a net 1% could have a decent effect on the next election. The one upcoming feature which could have an effect on this scale will be the head to head debates, I don’t see much chance of both leaders ignoring each other and appealing to UKIP or Greens respectively.

    The other element that might have an effect is the polls themselves. If it looks like the other side will get in there might well be a last minute change of heart on a large scale, particularly in the marginals.

    The other other element is are the pollsters right? It’s not unreasonable the they might all be over predicting (or under) the scale of support of people. The defence “well the day before they election the electorate told us the were all voting OMRLP, it’s not up to us to capture the swing in the booth” might be played out but this scenario would lead to a different outcome than predicted by the polls.

    All these considerations will have to be taken into account when devising a strategy, “who to target with which message?” I tend to agree that a lot of focus will be on people prepared to say they are voting for other parties. We’ll see how that pays off, a substantial difference in the rate at which those votes come home (or not go further away) could be enough on it’s own.

  42. @ Andy Shadrack

    “On what basis would those currently considering voting for “other” go to Labour or Conservative?”

    Lots of possible reasons:

    1) To influence the outcome of the General Election (the wasted vote argument)
    2) Because their stated preference for their chosen “other” has until now been posturing
    3) Because their chosen “other” looks less appealing when the spotlight of the campaign falls upon them
    4) When they appreciate the tactical position in their own constituency
    5) When they see the candidates on offer in their own constituency
    6) Because their minds are changed in the course of the campaign.

    “a whole swath of voters first supported BNP and then UKIP.”

    Firstly, I think that your definition of swathe must be different to mine. BNP supporters are mercifully rare. They make up a tiny proportion of UKIP support.

    UKIP got less than 2% in the 2010 General Election. Most UKIP supporters previously voted for Lab, Con or Lib , and in many cases did so for decades. It is not fanciful to think that some may return to one of those parties.

    The same is true of SNP voters although my view is they are more likely to stay with their new party..

    Green voters are slightly different. Many are young and will not have voted for anyone before.

    Finally, the polls. Polls often ask how certain voters are to stick with their current choice. In Lord Ashcroft’s most recent poll only 57% said that they were absolutely certain to vote they way that they had indicated.(the figure for UKIP was 56%, the figure for the Lib Dems was 35%)

    You write as though you think it extraordinary that some people currently expressing a VI for UKIP, SNP or Greens will, in the event, actually vote Labour or Conservative. It really isn’t extraordinary at all..

    Good early evening from a nice beach here in sunny Bournemouth.
    I agree that UKIP, Green, SNP and LD VI could well slip from their possibly artificially high levels, and help the Cons and Labs to improve their own poll figures.

  44. I think those who will vote SNP and Green have a differnt mindset and will not vote for Labour or Tories unless they are using a tactical vote. Neither mainstream party appeals to the key issues of these electorates…

  45. “GRAHAM
    I have just been looking at what the polls were saying 10 weeks prior to the last three general elections
    End of Feb 2010 Con +6 Outcome Con +7.3
    End of Feb 2005 Lab +6 -7 Outcome Lab +3
    End of March 2001 Lab +15 Outcome Lab +9
    In each case, therefore, the final 10 weeks saw a swing to the Opposition.”
    Or maybe its a swing to the Tories, who knows.

    I do think we are in a new paradigm, UKIP,SNP and Green surge, whilst the big 2 are in long term decline and both hated in certain areas. IE I see greens and ukip taking over as the opposition in the south,

  46. We can’t believe in swing back when times are changing in British politics.

  47. Jihadi John, will not help the pro-European cause. Nor will the figures on immigration. LD’s to dip further? I’m not sure why people vote LD apart from tactically. Democracy at its best.

  48. @”Jihadi John, will not help the pro-European cause.”

    I find it difficult to comprehend , how, or why anyone could compose that sentence.

  49. John

    I don’t think people will link a London terrorist to Europe either, seems like a non sequitur.

    If you had argued that the libs might suffer for their softer stance on terrorism via giving security services fewer tools, that is an argument that might be proven.

    I’m not at all sure how being a pro European party has anything to do with a Kuwiti immigrant raised in London. Perhaps if you used more words to support your claim we might understand the link.

    To then challenge anyone who disagrees with your statement as “pro-muslim” (as opposed to what? a racist?) and anti british values, my mind is well and truly boggled.

  50. @ Little Red Rock

    With all due respect I must respectfully, from a statistical point of view, refute your claim about BNP.

    In the 2009 European election BNP attained 6.3% support and English Democrats 2.6% UK wide, more combined than LD obtained in 2014 at 6.9%. So there was a greater statistical probability that a voter would vote BNP and/or English Democrat in 2009 than LD in 2014. You may remember 2 BNP MEPs were elected, more than the lone LD MEP in 2014.

    So if you track the rise in UKIP vote from 2009 to 2014 in the European election there appears to be a high statistical probability that someone who voted for BNP or English Democrat in 2009 likely voted UKIP in 2014. Further while some voters switched from Conservative and Labour to UKIP, the statistical probability exists, on the numbers voting BNP and English Democrat in 2014 versus 2009, that more voters switched from BNP and ED, than individually from Conservative or Labour in certain regions of the country.

    In fact in London it appears that a number of Labour voters who had sat out 2009, came back and voted Labour in 2014. So I am, to some extent, dubious about the actual volume of Labour voters moving back and forth between Labour and UKIP, but did notice some oscillation between the number of voters supporting Labour at the European election in 2014 and in the London borough elections, but compared to the swing between Conservative and UKIP it is quite minimal.

    As for the extensive list of why someone might switch the question needs to be asked as to what is the statistical probability of a voter switching. This is after all a site dedicated to discussing polling in which statistical probability and not political speculation should be the main forum of discussion.

    My perspective, from a statistical probability point of view, is that if someone has previously voted BNP, ED or UKIP, then they are probably not likely to switch to Labour or Conservative. Likewise someone who switched from LD over LD joining a coalition with the Conservatives is not likely to switch to Conservative or back to LD, but might contemplate moving from Green to Labour or vice versa.

    Likewise some LD supporter who feels LD is finished in his or her constituency might switch to Conservative to prevent Labour winning the seat. But if I have been voting Green at local government elections and in the European election I am probably not inclined to switch back to either Labour or LD if I think the Green Party has more support than in 2010.

    Asking voters how they voted in immediately previous elections other than 2010 will actually help track the changes in voting patterns and thus you come up with a better statistical probability of how someone actually voted and might vote in the future. It is, after all, about the numbers and not what you and I might want to happen

1 5 6 7 8