Panelbase are well known for their Scottish polling, but today we have what I think is their first GB voting intention poll. Their first voting intention figures are CON 34%, LAB 34%, LDEM 5%, UKIP 15%, GRN 6%.

Apart from the low Lib Dem score the figures are very much in line with averages across all the pollsters (and it’s too early to tell if Panelbase will always tend to show low Lib Dem scores or this is just an unusual result.) The tables should be out later, but in terms of methodology Panelbase are internet based, they weight by recalled 2010 vote, they are filtering likely voters to include only those who rate their chances of voting at 8/10 or more, and they don’t do any reallocation of people who say don’t know.

UPDATE: The full tables for the Panelbase poll are here. They reveal one very unusual difference – this isn’t a GB poll at all, it’s a UK poll. The vast majority of political opinion polls are only of Great Britain and ignore Northern Ireland, given its different party system. Panelbase include it. It shouldn’t actually make much difference apart from bumping up the “Others” a bit to include the DUP, SDLP, Sinn Fein, UUP and so on and knocking down the mainland parties by a fraction.


228 Responses to “Panelbase – CON 34, LAB 34, LD 5, UKIP 15, GRN 6”

1 2 3 5
  1. Thanks Anthony

  2. If they don’t do any reallocation of people who say don’t know, am I right in saying this increases the UKIP and Green score and lowers the Lib Dem score?

    This is on the assumption that the reallocation will be based on the recalled 2010 vote.

  3. Anthony has updated the UKPR average. :-)

  4. Good evening all.

    Panelbase just confirming what the other pollsters are showing…Neck & Neck.

    Should be interesting to see how their cross-breaks turn out!!

  5. By the way the Parliament timeline is interactive. You can double click on a year and get more info.

  6. TheFuture

    Those would be the relative effects. (Doesn’t mean they are wrong or right just that you’d expect those differences compared to a reallocater all other things being equal)

  7. Panelbase have done a UK (not GB, Anthony!) poll.

    Limited value in these crossbreaks – no geographical ones, and I presume NI parties are buried in “others”.

    “The headline voting intention figures for our poll are as follows: LAB 34%, CON 34%, UKIP 15%, LD 5%, GRN 6%, SNP 3%, OTHER 3%”

  8. As a conservative. I am very worried by this historical report from The Guardian that has apparently not had any controls or mitigation put in place to this day. Clearly Labour are the main beneficiaries of this based on where it is prevalent. Is this going to be a problem for the consign GE. Sadly I think yes.

    ‘A senior judge made a scathing attack on the postal voting system yesterday, condemning the government for complacency in the face of
    fraud which would disgrace a “banana republic”.

    Richard Mawrey QC, presiding over a special election court in Birmingham, warned that there were no realistic systems in place to detect or prevent postal voting fraud at the general election. “Until there are, fraud will continue unabated,” he said.

    He found six Labour councillors in Birmingham guilty of carrying out “massive, systematic and organised” postal voting fraud to win two wards’.

  9. Surely we should ignore fishers model?

    he initially projected the tories on over 40% – and them getting a decent majority. That was widely seen as tosh by many people on here – and so it has proved. I cant remember what he had labour getting – but it was pretty low – and this was before the rise of the SNP (which nobody predicted)
    The fact that model gets steadily ‘readjusted’ in the light of the most recent polls is having your cake and eating it. If the polls are right the model as originally set out has demonstrably failed.
    As for the lib dems – again a fair few of us were arguing over a year ago that their vote was not going to start rising again and that projections of them getting into the mid teens were daft.
    The polls over the past three years have shown the tories marooned in the low 30s, with a slow but steady climb over the past 3 months as UKIP has (predictably) slipped back a bit as the election looms – but nothing suggests that the tories are going to suddenly shoot up.
    Similarly – and also predictably – over the same period the labour vote has picked up as green and some UKIP has drifted back to labour.
    All this says that big swings are very unlikely between now and may 7th and the final result will be something like 35/35 with maybe a percentage point either way.

  10. OLDNAT

    I can’t make heads or tails of the Panelbase tables. What’s the SNP on?

  11. Allan Christie

    Panelbase say SNP on 3% of UK (not GB) electorate

  12. Also from the Panelbase blurb –

    In common with many recent polls, our findings indicate that there is a likelihood of another hung parliament. However we’re finding little appetite for another coalition government with almost three quarters of likely voters who express an opinion saying that they would rather see a single party gaining an overall majority.

    Opinions deeply divided on Europe – 52% of likely voters would vote to stay in the EU and 48% to leave.

    Faith in politicians isn’t exactly at sky high levels – fewer than one in ten feel that politicians are less corrupt now than 20 years ago – around half think they’re no different
    and a third that they are actually more corrupt nowadays.

    We asked people in England, Wales and Northern Ireland whether or not they are happy that Scotland voted No in the referendum – excluding undecided, fully 85% said that they are/i>

  13. @ Mr Nameless

    Who ‘won’?

  14. @Rich

    Thank you for revealing your political allegiance, but I don’t think electoral fraud is necessarily confined to any one political party.

    This is a serious issue, and it may not be limited to local elections.

    In an election that is expected to be tight, the ‘temptation’ to rig the result is undoubtedly there. It is entirely possible that we will have such a confused and messy outcome to the GE that a ‘hanging chad’ situation could arise.

    One small point: are we entirely certain of the robustness and scrupulousness of our EROs, and their treatment of postal votes? Is it only ‘front-end’ fraud we are discussing?

    The government of this country could well depend upon a recount, or more probably a fifth or sixth recount. I don’t think it will be decided by tossing a coin, as I seem to recall has happened in the recent past.

  15. @Rich ‘historical report’
    “Richard Mawrey QC, presiding over a special election court in Birmingham” etc
    These events related to Council elections in 2004, so the ‘general election’ referred to was in 2005.
    Things may be better or worse now, but in so far as the Birmingham elections were rerun with the relevant people barred, it could be said that the system worked, though obviously it is much better if fraud is made very difficult.
    The current Tower Hamlets judgement is due in early April

  16. Omni
    Thanks for the links – interesting

  17. We asked people in England, Wales and Northern Ireland whether or not they are happy that Scotland voted No in the referendum – excluding undecided, fully 85% said that they are [happy].

    I’ve inserted the [happy] because the way Panelbase have it written it, taken literally it could mean that 85% are either happy or not, with the other 15% not caring either way.

    But I have confirmed from the tables that 71% are happy, 12% are not happy & 17% are unsure (but Panelbase adjusts, thereby arriving at 85% happy people)

    :-) So we won’t need to have any more conversations about England wanting Scotland to leave the UK, because we now have fairly emphatic polling about it. :-)

  18. “We asked people in England, Wales and Northern Ireland whether or not they are happy that Scotland voted No in the referendum – excluding undecided, fully 85% said that they are.”

    Interesting. It suggests that the media narrative of a fissiparous Kingom may be wide of the mark

  19. REGGIESIDE

    Harsh – but fair.

    The Fisher model is surely not at all credible as a way of forecasting the election. It was so wide of the mark that the next time the model is used for forecast several years ahead, it will have no credibility.

    In short, it has failed the “external validity” test.

  20. And working backwards from 1007 – 923 (EW&NI) it suggests that the Scottish part of the sample is 84 people.

  21. OLDNAT

    Did you ever get to the bottom of those strange vacillations to the Scottish YouGov cross breaks on Tues and Weds? They seemed too weird and wacky to believe.

  22. Do Others include whatever the Official Unionists are now, or are they running as an NI branch of the Tories? That’s the only effect I can see that a UK poll might have – marginally more votes for any GB party that also stands in NI, a category which in 2010 only included the Conservatives IIRC.

  23. Prof Howard

    And today as well (SNP 34%: UKIP 14%)

    No idea. I’ll be interested to see what tomorrow’s Populus shows.

    If there is a genuine shift, then Populus should be picking it up as well.

  24. Important not to cherry pick from a poll, and think that results which suit you must be right, while ignoring others.

    That EU split is much closer than other GB polls have indicated – and I doubt that it’s just down to the effect of adding in those from NI.

  25. PHIL

    “Do Others include whatever the Official Unionists are now, or are they running as an NI branch of the Tories?”

    No: the Tories are standing their own candidates separately in NI this time but it will not make any difference as they have not much support.

  26. 24/84 voting SNP would be 28.6% – but we have no way of knowing what the level of Scottish don’t knows/ won’t votes are for this poll.

    Upweighting (assuming don’t knows & won’t votes at 10%) = ~32%; (assuming dk & wv to be 20%) = ~36%

    FWIW, YouGov has very low DK/WV for Scottish crossbreaks, sometimes as low as 2% WV & 6% DK.

  27. Millie
    “Thank you for revealing your political allegiance, but I don’t think electoral fraud is necessarily confined to any one political party.”

    It will no doubt be used by rogue members of all parties, but recent convictions have predominantly been Labour candidates or their agents. That’s just a fact, not partisan comment.

  28. @Syzygy

    Thanks for the link again.

    That analysis was with an EWMA chart (or exponentially-weighted moving average chart).

    Like CUSUM, it is designed to be a sensitive tool to detect small changes (less than 1.5 sigma).

    The fact that the results mirrored my own give me confidence. Now that such methods are now gracing such a widely respected sites as ForecastUK, these methods that add to psephological understanding might become more used.

    @Hal

    What’s not to love about them? They add a new line of analysis? For picking up subtle trends, they are the right sort of tool. A ‘psephological’ engineer needs more than hammers in their tool box ;-)

  29. Given Panelbase’s background, I am surprised they ran a poll like this one. They already know from Scottish experience that Scotland polling and indeed NI polling will have no relevance to E and W so why bother?

    In fairness to Panelbase, I see no point at this stage to have ‘GB’ polls from anyone. Just Scot polling and a separate E and W is what we all need.

  30. And according to today’s YG, all those ‘missing’ SNP voters have fled to UKIP (which seems jolly unlikely unless they are daft enough to believe that Abu Hamza really is an MSP!!!)

  31. Very interesting voter retention values from 2010 from Panelbase:

    UKIP 86%
    Green 81%
    Labour 79%
    Conservative 73%
    SNP 64%
    LD 23%

    Interstingly they find that 2010 LD vote is split:

    Labour 25%
    LD 23%
    Conservative 18%
    Green 15%
    UKIP 12%
    SNP 6%
    Other 1%

    This translates into:

    Labour +5.8%
    Conservative +4.1%
    LD 5.3% (17.7%)
    Green +3.5%
    UKIP +2.8%
    SNP +1.4%

    This suggests that Labour and Green increase in support UK wde is largely from collapse of LD, and that LD support flowing to Conservative has helped offset UKIP voter exit.

    Interesting that UKIP nearly pick up as much LD flow as Green and I am surprised the LD SNP flow is that low – but maybe that is as a result of the overall size of the Scottish UK vote.

  32. EM has won the Toss and sent DC into bat. It’s a drop-in pitch so it should be good for batting throughout the contest.

  33. RICH.
    There was a very moving speech by a good tory MP about William Hague; better to be an honourable fool than a clever man’.

  34. Pete B

    Am I not correct in thinking that the fraudulent activity was, in fact, in the main, coming from one ethnic minority group who normally adhered to Labour, rather than some Labour party plot?

  35. RAF

    I noted that one of the YG people said they weren’t doing an instant reaction poll to tonight’s non-debate, because Ed is last in to bat, and that would favour him in such a poll.

  36. I saw the speech chrislane. That must have been excruciatingly embarrassing for Hague. I normally like him, but he got this badly wrong.

  37. BristolianHoward
    That was my understanding, but I didn’t mention it in case people started throwing around the ‘r’ word which is never followed by rational discussion.

  38. @OldNat

    I agree with YG that snap polling would be unfair and of very little value as this isn’t a debate (nor that it is that valuable anyway).

  39. ICM are doing a snap poll.

  40. @James

    More fool them. Maybe they will apply some weighting to ofset the bias.

  41. How many will watch the Cameron and Miliband thing live? Maybe a couple of million?

    Unless a pollster has already got a panel lined up and guaranteed to watch, it could take them ages to find enough people who’ve watched the program. Actually, they must have a panel lined up, mustn’t they?

  42. Aren’t snap polls rather rubbish and pointless really?

  43. @Pete B

    You would think so. It would be pretty pointless otherwise. Unless the live TV audience will be polled.

  44. Pete B

    “How many will watch the Cameron and Miliband thing live?”

    More importantly, how many of the viewers will still be alive at the end?

  45. lol!

  46. I noted that one of the YG people said they weren’t doing an instant reaction poll to tonight’s non-debate, because.. […they couldn’t find enough don’t knows who were willing to sit through two hours of Jeremy Paxman & whoever those two chaps he’s interviewing are; now, if it had been Beyoncé & Zayn Malik they’d have been queuing around the block but I’m just stating the obvious now, aren’t I?]

  47. Amber
    Thanks for the interesting figures on voter retention and where the Lib vote has gone.

1 2 3 5