Time for a post-mortem of the European election polling. I’m not a fan of the sort of horse-race approach to these things – just because they are the final poll of the race, they still have normal margins of error, so if one pollster is a fraction more accurate than another it is often just the luck of the draw. Realistically the best a pollster can ever hope to do is get all the results within the margin of error. Better than that is luck. However away from the public “who won” stuff, comparing poll predictions to actual election results is an absolutely critical tool for pollsters – it’s our chance to compare our figures with reality, to improve and finesse our methods.

The final polls from each company are here. Note that I haven’t included Populus – they did conduct one European poll, but it was a fortnight before the election when just a week is a long time in politics! While I’ve included it in the comparison, one should allow some leeway for ICM for the same reason; their poll’s fieldwork finished a week before the actual election.

CON LAB LD UKIP GRN Average Error
ACTUAL RESULT 23.9 25.4 6.9 27.5 7.9
YouGov 22 26 9 27 10 1.4
(-1.9) (+0.6) (+2.1) (-0.5) (+2.1)
ICM 26 29 7 25 6 2.0
(+2.1) (+3.6) (+0.1) (-2.5) (-1.9)
Opinium 21 25 6 32 6 2.1
(-2.9) (-0.4) (-0.9) (+4.5) (-1.9)
TNS 21 28 7 31 6 2.2
(-2.9) (+2.4) (+0.1) (+3.5) (-1.9)
ComRes 20 27 7 33 6 2.6
(-3.9) (+1.4) (+0.1) (+5.5) (-1.9)
Survation 23 27 9 32 4 2.6
(-0.9) (+1.4) (+2.1) (+4.5) (-3.9)

The most obvious current difference between Westminster polls is the reported levels of UKIP – there is a big gulf between the levels of UKIP support report recorded by companies like ICM, MORI, YouGov and ComRes’s phone polls and polls from newer companies like Opinium, Survation and ComRes’s online polls. We don’t know what the reasons for this are – there are a couple of things like prompting and re-allocating don’t knows that we can account for, but mostly the difference is not easily explained. It may be something to do with interviewer effect, or the representativeness of different companies samples. We can’t tell.

The European elections were obviously an opportunity to check figures against reality. I half expected the polls to all converge together in the run up to the election, as they have a tendency to do before general elections, but in reality we got the same sort of contrast as we do in Westminster polls. Higher figures for UKIP amongst newer online companies, lower figures from YouGov, lowest from ICM… and when the votes were counted the YouGov figure was the closest.

Of course, European elections aren’t general elections. On the issue of prompting, for example, every company prompted for UKIP in their European polling, whereas only Survation do it for general elections. There were no telephone polls for the European election, so it can tell us nothing of them. European elections are low turnout elections, so some of the errors may have been down to too strict turnout filters (ComRes used a very strict turnout filter for Euros and would probably have been better if they’d used the method they use for general election polling. There was the issue of the Independence from Europe spoiler party on the ballot paper and so on. At a purely personal level though, getting UKIP right at the next election is the biggest challenge currently facing pollsters, so I’m relieved that in the first real proper national test we got it right. Phew!


440 Responses to “European polling post-mortem”

1 7 8 9
  1. FPTP would have done exacts what it claims to. Selected the first preference candidate for a local constituency based on a plurality of votes. The crisis would be in what people want the electoral system to do.

    Report comment

  2. I think that ‘other’ voters going back the main parties may occur less and less.

    It feels that over the years people have wandered off mid-term, then come back for the GE, but are the rate of return is decreasing every time, like an increasingly worn piece of elastic.

    Eventually worn elastic just snaps…….

    I will do some digging….

    Report comment

  3. @OldNat

    “Ashcroft poll – England only”

    Is your interest in this something to do with predicting the political colour of the Government that would be negotiating the detailed terms of Scotland’s secession should Scotland decide to take that course?

    Report comment

  4. Green Party are on 7% in the Ashcroft poll, like the UKIP 17% probably a record high.

    As ever we need to be careful about these telephone polls. These figures are based on a sample of 543 people so MoE will be around 4 points.

    Most Green votes seem to come from those who voted Lib Dem in 2010, few from Labour (there’s actual more ex-Cons but were talking single figures here). Most UKIP are still ex-Cons but there are almost as many Lab and L/D together.

    As we saw in many of the Locals what may be happening is that UKIP and the Green are to some extent intercepting votes that would otherwise go to Labour as well as attracting some directly.

    Report comment

  5. @origimbo

    The trouble is that I think in 2015, the winning party will be at around 35-36%.

    On 65% turnout that’s 22% of the population.

    Given that UKIP could well get about 10-15% of the vote, and probably no more than a couple of seats, FPTP would seriously look grossly unrepresentative.

    Report comment

  6. In 2010 UKIP fielded 558 candidates but won 919,471 votes, which was 3.1% votes cast. The recent Lord Ashcroft poll showed that UKIP would hold onto about 51%, with 21% and 11% going back to Tories and Labour respectively at the general election.

    Let us say that UKIP score about 10% at the general election, which will be something like 3 million votes and the Lib Dems may obtain about the same. The question is how much of the UKIP vote will actually go back to the Tories and Labour in the key marginals. If the Tories do get back 21% of the possible votes from UKIP, then they stand a chance of holding onto seats.

    Just trying to get my head around the numbers, based on 30 million people voting in May 2015. How much will UKIP affect the outcome in key marginals ? If UKIP maintain much more than 51% of their current predicted vote, they are going to hurt the Tories more than Labour. But if more UKIP votes do actually go back to the Tories, than to Labour, then Labour could be more affected.

    No doubt some of the guys who are good with statistics will come up with a range of likely general election outcomes, based on the UKIP vote level.

    Report comment

  7. @Catmanjeff

    “I think that ‘other’ voters going back the main parties may occur less and less…. I will do some digging….”

    Lord Ashcroft has already done the digging as far as the Euro elections are concerned.

    Those voting at the Euros for any party other than the main 4, 16% of the total vote, say they would be likely to vote as follows at the GE.

    Con 4%
    Lab 21%
    LD 5%
    Plaid 2%
    SNP 15%
    UKIP 2%
    BNP 2%
    Green 25%
    Any other 3%
    Would not vote 1%
    DK 19%

    Report comment

  8. This tweet seems to be from a genuine account.

    ” Ed Miliband @Ed_Miliband · 42m

    I have changed Labour’s position on immigration. But I won’t make false promises, or cut ourselves off from the rest of the world.”

    Whatever does he mean?

    Report comment

  9. From conversations with my Green colleagues who campaigned hard for the locals and Euros, I think that assuming a Green voter will automatically go to Labour for the GE is assuming a bit much.

    The Greens are beginning move from a ‘luxury vote of protest’, to a vote of specific choice.

    Report comment

  10. Phil Haines

    Since rUK would not just consist of England, that seems a remarkably peculiar observation on your part.

    Report comment

  11. Phil Haines

    Since rUK would not just consist of England, that seems a strange observation on your part.

    Report comment

  12. @CATMANJEFF

    True, but then, as has been pointed out for the Euro elections, that 35% which doesn’t vote doesn’t get represented under FPTP, d’Hondt or STV, and I’m yet to meet an electoral system which does*.

    How would we expect the parties to line up on new voting systems in 2015 anyway?

    * Unless we apply a variation of the system Boris Johnson suggested ought to be implemented for strike votes, and say that an MP can’t be elected unless 50% of the electorate vote for him or her.

    Report comment

  13. @R Huckle

    According to Lord Ashcroft there is a 10% lead for the Conservatives (21% – 11%) over Labour in their share of the returning UKIP Euro voters in the GE.

    However, there is a 17% lead for Labour (21% – 4%) in their share of the returning “Other” Euro voters at the GE.

    Allowing for the relative size of each group, they just about cancel each other out.

    Report comment

  14. @Phil

    That’s interesting data. So of the ‘others’ at the Euros, 25% will vote Green, and 21% Labour at the GE. The Conservatives at 4% looks worrying for them. It looks like they need their UKIP voters back.

    Report comment

  15. How would we expect the parties to line up on new voting systems in 2015 anyway?

    They won’t

    Turkeys don’t vote for Christmas.

    Report comment

  16. @ Oldnat

    If you’re right then EM is the first to blink!

    Report comment

  17. @OldNat

    I certainly was a bit puzzled as to why you singled that cross break out in the first place. I’m still none the wiser after your response.

    Report comment

  18. @origimbo

    I my point about turnout is how can a party with barely a fifth of the electorate voting have a majority?

    Back to ‘elected dictatorships’.

    Report comment

  19. The turkeys won’t, but presumably the chickens might? Or to drop the analogy, UKIP, the Lib Dems , the SNP and the Greens.

    From discussions with fellow electoral system geeks in the past, I understood most Lib Dems to favour 3/4 member STV (which of course happens to benefit a third party).

    Report comment

  20. Phil Haines

    Why should seeing the importance of the distribution of VI in the largest political unit of the UK be a surprise to you?

    Report comment

  21. Origimbo

    ” I understood most Lib Dems to favour 3/4 member STV (which of course happens to benefit a third party).”

    It does, but as the LDs discovered after forcing Lab to introduce STV for Scottish local elections – it’s bloody awful when you become the 4th or 5th party!

    Report comment

  22. @origimbo

    Without the support of the Conservatives or Labour, electoral reform is a pipe dream (them being the said turkeys).

    I quite like AMS, and Scotland had demonstrated it works in the UK too. Without AMS the Conservatives would get nothing in Scotland, and the Lib Dems will in the same boat shortly.

    Report comment

  23. “@ bantams

    @ Oldnat

    If you’re right then EM is the first to blink!”

    What EM means is that he has already changed Labours immigration policy. This was announced some time ago and his tweet is not about some new policy change today.

    As I understand it, Labour would make it harder for non EU immigrants to come to the UK, with an enhanced points system for skilled worked. Labour would also look to agree changes with the EU about EU citizens who wanted to come to the UK to work. e.g they would need to be able to support themselves and not be able to claim any benefits in the UK for a period.

    Report comment

  24. @OldNat
    (“” Ed Miliband @Ed_Miliband · 42m
    I have changed Labour’s position on immigration. But I won’t make false promises, or cut ourselves off from the rest of the world.”
    Whatever does he mean?)

    That’s a quote from Miliband’s speech in Thurrock today (http://press.labour.org.uk/post/86997808779/britain-needs-real-change-not-false-promises-ed)

    Report comment

  25. @ R Huckle

    Doesn’t matter if it’s old policy, his tweet intimated it’s a change of policy so he blinked first. He needs to be made of sterner stuff if he stands any chance in the GE.

    Report comment

  26. @OldNat

    You haven’t convinced me. I think I’ll stick with my original supposition.

    Report comment

  27. “Doesn’t matter if it’s old policy, his tweet intimated it’s a change of policy so he blinked first.”

    Lol, ignore the facts given to you by R Huckle because you prefer to think EM means something else !

    Dear God.

    Report comment

  28. @R Huckle, @Liz H, @Bantams

    To be fair to Miliband, he needed to make that announcement today because next to no-one had yet noticed Labour’s new stance and policies on immigration.
    (@Bantams: his tweet does intimate that it’s a change of policy – what it doesn’t intimate is that the change of policy happened today)

    The same applies more generally to the stream of popular policy positions that Miliband has announced over the last year – next to no-one yet knows of them.

    Report comment

  29. @PHIL HAINES: “Those voting at the Euros for any party other than the main 4, 16% of the total vote, say they would be likely to vote as follows at the GE.”

    But they’re not a representative sample of a General Election electorate. For every seven Euro voters there are about another five Westminster voters who didn’t vote. Nor are the two groups likely to be in any way comparable – one is the sort who vote in Euros and one is the sort who don’t. It’s completely futile to try to guess the views of the second group from the actions of the first.

    Report comment

  30. So best guess after lent votes return is that Lab are a little ahead of the Tories 36-34 or similar.

    Report comment

  31. noo fred

    Report comment

  32. There’s been some discussion on here about the higher turnout in London where there were also local elections.

    Not sure if this is cause and effect, or whether the same would apply (if it ever does) in a general election, but I have looked up what is happening in 2015 in England.

    According to the gwydir site http://www.gwydir.demon.co.uk/uklocalgov/elec2015.htm

    there are elections in 36 Metropolitan Boroughs (one third of councillors in each council)
    32 Unitary Authorities (all councillors in each council)
    17 Unitary Authorities (one third of councillors)
    131 second-tier districts (all councillors in each)
    63 second-tier districts (one third of councillors)

    As far as I can see there are no London local elections in 2015. The London Assembly elections are in 2016, as are the Scottish Parliament elections and elections for the National Assembly for Wales (source: Electoral Commission).

    Report comment

  33. I don’t imagine local elections will increase turnout for a General Election in the way they might an EU one.

    Report comment

  34. 30% VI at 65% turnout translates to 19.5% of electorate.

    Something has to give. This situation can’t last, as any party forming a government with 30% of the vote, and potentially less than 20% of the electorate has no mandate to govern.

    Report comment

  35. @PHIL HAINES
    “To be fair to Miliband, he needed to make that announcement today because next to no-one had yet noticed Labour’s new stance and policies on immigration.”

    I agree with you. Miliband has mentioned the following points before which all feeds into a new stance on immigration:
    having better controls when people arrive and leave here, a living wage, ensuring people in public services speak English and people need to earn their entitlements.

    Report comment

  36. Blair got 21.6% in 2005. Is the difference significant? If you’ve got the MPs you’ve got the mandate.

    Report comment

  37. Phil Haines

    “I think I’ll stick with my original supposition.”

    I’m quite sure you will, and doubtless always have.

    Comforting things – self-created certainties.

    Report comment

  38. @OldNat

    It’s up to you to give me a good reason not to.

    As for “I’m quite sure you will, and doubtless always have”, that’s your comfortable self-creating certainty.

    Report comment

  39. Results for Bury lancs show Lab still comfortable in Bury South but UKIP intervention would hand Bury North from Con to Labour.

    In 2010 GE Lib dems in Bury North achieved 17%, in last weeks locals for the 8 wards of Bury N they polled just 2%

    Lib Dem vote held up better in Bury South but that was because two popular, hard working well known incumbents achieved decent results ( although one of them still lost his seat).

    In Local Election Lib Dems scored 8.3% but their vote vanished for the Euro election – down to 4.1 % a loss of 51% of Lib Dem voters from local to euro elections. Did half of Lib Dems in local election switch to UKip. Probably not, it was most likely Tories voting tactically in the two ward elections mentioned above to keep Labour out.

    In the ward elections Lab scored 43.6% but in the euro poll it fell to 33.4% ( a loss of 25% of their vote to UKip ?)
    Cons scored 29.5% in wards and 23.3 in euro poll ( loss of 21% to Ukip).

    But Ukip intervention still delivered two traditionally Tory wards to Labour, one of them (Pilkington Park in Whitefield by only 19 votes).

    Labour switchers from local election to euros might not all have gone to Ukip. The total ward vote was 52,421 but the euro vote was 51,353. That is 1000 people who voted in the locals didn’t bother in the Euro ( presumably these euro abstainers might have been mostly Labour – unwilling to grapple with a list of 11 parties and 88 names on the NW England Sheet.

    In addition there were six other parties apart from the main 5 on the euro sheet including BMP (872) English Democrats ( 685) and the Pirate Party (252). Would Labour voters have been more inclined to go to them ?

    Ukip only contested 9 wards and got 11.2 % at an average of 654 per ward contested. If it had contested all 17 wards at that average its notional total vote might have been double the 11.2%.

    The Green vote held steady between the Local and the euro election at 6.2 and 5.9 respectively.

    The overall impression suggests that Con and Lab affiliations held up in the local election but sizable chunks switched to Ukip for the Euro election. This in turn suggest that when it mattered ( the locals) party affiliations ( except Liberals )held up. This again might suggest that the Ukip vote in Bury was an ephemeral protest vote with Ukip being seen as Second Choice preference in a “harmless” Euro vote.

    Report comment

  40. Hi

    A question from Australia.

    Can anyone suggest a site that has seat by seat information, ideally in a map? The BBC video coverage seemed to include a “floor walk” graphic showing this, but doesn’t seem to have any online equivalent.

    Report comment

1 7 8 9