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 3 4 5 6 7 9
  1. ‘My question for Labour is that the past week 2 elections have shown Labour only 1% ahead of the Conservatives, and many objective commentators will accept that polls tend to shift towards the governing party in the final year as the election approaches’

    MIM,
    It’s a bit of a myth that this always happens and is much more likely to occur when the Government is well adrift in the polls – as was the case with Brown pre-2010 and Major pre-1997.. Consider the position in relation to the 1979 election – the Tories only enjoyed a narrow win in the 1978 local elections unlike the landslide they had a year earlier. Despite this Thatcher went on to do a fair bit better in May 1979 than the 1978 locals had implied
    More recently, the PNS from the 1996 locals was Lab 43 Con 29 – a year later the actual election result was Lab43 Con30 so the Tory Govt gained just 1% in the final year. Looking at this set of Euroelection results, if Labour increases its % vote share next May in the same way the Tories managed between June 2009 and May 2010 then Labour will be at circa 35% next year.
    As to economic recovery, London has been the main beneficiary todate – but Labour did not exactly perform badly there on the whole.

  2. @Phil Haines: “Alloowing for all this, I’d guess that the Euro poll would be consistent with a Labour lead of about 2% in a GE context with a slightly higher share of the vote for the main parties than above.”

    I agree with this. And it seems plausible to think that the return rates from the EU poll are already reflected in the national GE polls: if tUKIP voters are self-aware enough to report that they will or may go to another party at the GE, they would presumably report the other party as their GE destination in other polls.

    This, too, seems consistent with the lower UKIP VI which the national polling is reporting. So Ashcroft’s EU poll may be giving us more granular information about the preferences and intentions of UKIP voters, but it’s not necessarily adding anything to what other polls are saying about current headline GE voting intentions.

  3. I still find it frustrating that party leaders beat around the bush on why Ukip is doing so well.

    They still seem scared to talk openly about immigration for fear of being labelled racists.

  4. BLUEBOB.
    I agree with you about this; a pernicious long term effect of Powell’s words in 1968, I think, and Benn’s very silly ‘Belsen’ speech which lost Labour seats in 1970.

  5. John B,
    His last poll was released May 19th and fieldwork was 16th-18th.

    So assuming that it’s similar now (he’s not releasing it today because it’s a bank holiday), we should expect fieldwork of 23rd-25th.
    So before the EU results, obviously, but post-locals.

    Right slap bang in the middle of people having just voted UKIP and the media’s UKIP mania. So it should be interesting to see how much UKIP jumps up.

    Last week had Con 29, Lab 35, Lib 9, UKIP 14.

  6. ‘Internal ICM poll’ as reported in the Guardian indicates Nick Clegg will lose his Sheffield Hallam seat in 2015, with this 53% vote share in 2010 collapsing to 23%.

    Ouch.

    Shame there is no data around the poll itself and sample size.

  7. Is weekly Ashcroft a Populus phone poll? I thought AW confirmed it is?

  8. nickp

    Nobody answered my question…

    Was Thursday’s election Labour’s best ever performance in a Euro poll?
    ______

    Yes but it’s the first time Labour and the Con’s have not won a national election in 100 years.

    Also I think it’s the first time an opposition party has not won the Euro election but I could be wrong on that front.

  9. bluebob

    I still find it frustrating that party leaders beat around the bush on why Ukip is doing so well.

    They still seem scared to talk openly about immigration for fear of being labelled racists
    __________

    Absolutely spot on and it’s also the same reason across other parts of Western Europe.

  10. Next week’s polls will be interesting.

  11. I was just about to post about the internal quote-unquote leaked ICM polling.

    Lab 33, Lib 23 – says that he would be behind the Cons but gives no figure for Con.

    Cambridge would go Lab 41, Lib 28. Redcar is Lab 46, UKIP 20, Lib 16. Wells would go Con 41, Lib 28.

    It also includes ‘Cable as leader polling’ –
    Reduces the lead by 8% in Sheffield, Cambridge by 12, Redcar by 4 and Wells by 7%.

    “Lib Dem supporter” = Lord Oakshott?

  12. @DRMIBBLES

    I think it’s fair to say that there’s a section of the left leaning electorate to whom Clegg is now more toxic than the Tories, and that it isn’t small. However that’s a very large shift and there has just been an election with the word Europe involved.

    So, will pollers see shy Lib Dems in 2015?

  13. With regard to the returning Kippers, it’s worth noting that in Ashcroft’s marginal polls, the Labour lead with Ukip on 3% (in 2011) and Ukip on 15% (last month) is identical.

    It’s possible that Ukip are hurting the Tories disproportionately in the marginals but this is being masked by a fall in the Labour vote overall, so when the Tory defectors return home we’ll see the lead fall. (This is what happened last spring in the national polls). But it’s striking how little difference a fivefold increase in the number of Kippers seems to make to the relative positions of the main parties in Ashcroft’s polls.

    If I were David Cameron I would not want to stake the next election on the idea that Ukip falling back to 3% in 2015 will result in a radically different vote distribution than Ukip on 3% in 2011.

  14. “Also I think it’s the first time an opposition party has not won the Euro election but I could be wrong on that front.”

    It feels at the moment that it is drifting away from Labour.
    In my opinion they should be much further ahead, I am though still impressed on how they have come back from 2010.

    I thought they would not even have a sniff of winning an election till at best 2020.

  15. After yesterday’s late night/early morning I’ll not wait up for Ashcroft. It’ll make interesting reading over breakfast.

  16. Nick Clegg losing his seat might actually solve several of the Lib Dems’ problems for them.

    If I were Tim Farron I’d be tempted to adopt Mr. Nameless’s strategy of wandering around Sheffield Hallam at 3:00 in the morning in a fake beard and a yellow rosette shouting “Vote Lib Dem!” into people’s letterboxes.

  17. But would Lord Clegg of Hallam then resign?

  18. Origimbo

    I don’t see how the champions of democracy and constitutional reform could have an unelected leader. But then, they’ve surprised me before.

    Hey, if Ukip win in Doncaster and the insurgent Chipping Norton Labour Party win in Witney (lol) all the main parties could be decapitated! That would be fun.

  19. Needless to say that poll has cheered us all up immensely. You’ll all be watching out for me at the count, I trust?

  20. If Clegg goes and the Lib/dem VI starts to climb again then where will the votes come from? It ain’t going to come from kippers or the Tories.

    Be careful what you wish for…

  21. NickP
    I think we need to wait a couple of weeks for the last few days to disappear from voters’ memories. They will.

  22. Hello UK Poll

    What about the ‘An independance from Europe Party’? That was clearly a decoy party and cost UKIP a seat in the South West, and another 1.5% points overall? Doesn’t that make their final result a bit higher?

  23. @ Allan,

    Yeah, I said it would solve the Lib Dems’ problems, not Labour’s.

    But it’ll climb anyway as soon as they’re back in opposition. If Labour want to win in 2020 it’s going to take more than being the least toxic party, they’re going to have to prove they’re good at governing.

    (Mind you, the way the Tories are going, the Lib Dems may be getting Heseltine and Ken Clarke pretty soon.)

  24. NickP,

    Check the 1994 Euro election: Labour 44%, Cons 28%, LD 17%.

  25. NickP,

    … and without looking it up, guess who the Labour leader was at the time of the 1994 election.

  26. This graph puts things in perspective

    https://twitter.com/RaviSubbie/status/470991431050727424/photo/1

    What polling do we have on that 66%?
    – How many will vote in 2015?
    – Who will they vote for?

    And what would it take to get more of them out? We know the majority of them are younger and we know the young are facing huge issues – unaffordable housing, large student loans, few jobs. We also know they don’t like UKIP policies. We also know they are the future electorate. Parties with vision will energise those votes and reap the rewards in 2015 and build a base for many years in the future.

    But who is offering a vision for them? No one.

  27. This graph puts things in perspective

    https://twitter.com/RaviSubbie/status/470991431050727424/photo/1

    What polling do we have on that 66%?
    – How many will vote in 2015?
    – Who will they vote for?

    And what would it take to get more of them out? We know the majority of them are younger and we know the young are facing huge issues – unaffordable housing, large student loans, few jobs. We also know they don’t like UKIP policies. We also know they are the future electorate. Parties with vision will energise those votes and reap the rewards in 2015 and build a base for many years in the future.

    But who is offering a vision for them? No one.

  28. Hal, I know but I’ll let someone else guess first!

  29. Posts today seem ragged, maybe everyone is tired from two late Election nights.

    Most people seem to be ignoring the substantial vote that went to the Greens which Labour will hope to borrow/repatriate at GE 15. This will go some way to compensate for any differential in UKIP return rates favouring the Cons. Similarly the exceedingly slender Lab lead in the national totals is being emphasised rather than the much larger lead they have in the key marginals which will actually decide the GE – a lead demonstrated by some very comprehensive data from Ashcroft revealed on Saturday, and substantially corroborated by Labour’s actual performance last week in marginals such as Reading, Cambridge, Ipswich, Swindon, Amber Valley, Hastings, Bury, Various London seats etc etc.

    The other Q in my mind is whether Labour have spent too much effort on the marginals, and now put their underbelly of so-called safe seats at risk, mainly to UKIP. I assume that both main Parties will in fact be hastily working on a Plan B to address this in the light of the Euro results. Whichever Party deals most cleverly with the purple peril may in the end win all !

  30. … and without looking it up, guess who the Labour leader was at the time of the 1994 election.
    ———-
    It was a different voting system.

    And Labour’s leader was Margaret Beckett (John Smith had just died, the saddest thing we’ve ever seen?).

  31. Well done you guys at You Gov….

    though this is not good forLabour the timing of their best results came too late in the results to stitch a narrative thread that was worth a glance…whether this will help or hinder in the long term it is hard to say.

    For what my poor opinion is worth the Conservatives emerge from this with least collateral damage. It is ironic that the last time the Liberal Party got into a political bed with the Conservative party they enjoyed it least and emerged with less than they brought to the marital bed. Maybe Mr Clegg wishes he had studied history’s lessons more closely.

    Meanwhile why not go the whole hog and have an EU referendum on Election Day and shoot UKIP ‘ fox….the chickens voting for self preservation? Maybe not….

  32. John Smith?

  33. @Richard

    Good post.
    I’m afraid the ‘trad’ parties and in particular Labour are still over-dependent on focus groups and excessive psephologising.
    It was rather brought home to me by an exchange with one of the councillors I helped in a modest way to get re-elected.
    Him – We have the wrong brother leading the party
    Me – But the other brother is a Tory
    Him- Yes, but he’d win the election
    I’m not convinced that’s true, and I think the UKIP rise is in part tied up with a reaction to this attitude.
    I’d rather have a single term Attlee government than another 3 terms of Blair.

  34. Provisional estimation for the composition of the new European Parliament (compared to the composition of the outgoing Parliament)
    EPP 214 (-59)
    SD 190 (-6)
    ALDE 64 (-19)
    G/EFA 54 (-3)
    ECR 47 (-10)
    GUE/NGL 45 (+10)
    EFD 38 (+7)
    OTHERS 102 (+69)
    TOTAL 751 (-15)
    My estimation differs slightly from the one at the official site of the European Parliament. I give one extra seat to EPP (the Bulgarian Reformist Bloc), one extra seat to the G/EFA (the German Ecological Democratic Party), one extra seat to the ECR (the Independent Greeks) and 3 extra seats to GUE/NGL (the Italian “Tsipras List”). Of course the final composition may vary because of the probable alignment of some of the 102 “unaffiliated” MEPs to the existing groups or to new ones. Af far as the EFD group is concerned, despite its slight increase in terms of MEP numbers (thanks to the UKIP performance) , its survival is not guaranteed, because, even if the 5 MEPs of the Italian Lega Nord remain in the group, it is still composed of members from 6 countries, whereas 7 are needed (they lost their Greek and Slovak MEPs). On the other hand, even if the LN chose to caucus with the far-rightists of France (FN), Austria (FPOE), Netherlands (PVV), Sweden (SD) and Flanders (VB), this new group would have 40 MEPs, but still from 6 countries. Ergo, the organization of the Euro-phobic (I don’t like the term Euro-skeptical, we never say homo-skeptical or xeno-skeptical!) far-right in the new EP is yet to be defined, given also the fact that nobody among the afore-mentioned wishes to co-operate with the MEPs of Hungarian Jobbik and Greek Golden Dawn.

  35. RogerH

    John Smith was Labour leader between Kinnock and Blair. He tragically died, which is why Blair became leader.

  36. Virgilio,

    Do you have any analysis on the ideologies of the Others?

    It looks like the losses in Spain and Greece to EPP and SD have gone to left wing parties, but in France and the UK, it’s to right wing parties.

  37. @Virgilio

    The SDs are much closer to the EPP than our awful coverage on the BBC suggested last night.

    Will Juncker still be elected President?

  38. “John Smith was Labour leader between Kinnock and Blair. He tragically died, which is why Blair became leader.”

    John Smith was my answer. I wasn’t asking who he was.

  39. The prize goes to Amber.

    It was indeed Margaret Beckett; it was a month after the death of John Smith.

  40. Virgilio
    Thanks very much, as always.

    Good point about ‘euro-sceptic’. It was an euphemism dreamt up to disguise the situation by apologists at the time of Maastricht. If anyone thought the likes of Bill Cash were merely sceptic, they were living in their own political world.

    The same goes for ‘dry’. The opposite of ‘wet’ was ‘manly’ (er, yes I know).

  41. Hi all,

    the figs for the South East in 000s were …

    ukip 751
    cons 725
    lab 342
    grn 211
    LD 187

    can somebody who understands d’hondt run them through again ?
    I keep thinking Labour should have been awarded a second seat with their second total of 171.

  42. NORBOLD

    “All the results by district are on the BBC Vote 2014 web site.”

    Can you give me the actual link for that? I still can’t see them. I can see the Euro results by *region* OK.

  43. Inevitably, Roger scully has the best analysis of the Welsh Euros that I’ve seen

    http://blogs.cardiff.ac.uk/electionsinwales/2014/05/26/the-2014-european-election-in-wales/

    Even though he’s less than complimentary about YG’s Welsh polling.

    “The first, which I’ll review in a future blog post, is that YouGov’s polls in Wales have been systematically over-stating Labour support.”

    That looks like an interesting analysis to come.

  44. @TINGEDFRINGE, RAF
    The wider gap between EPP and SD presented yesterday was due to the fact that the Italian results were not yet factored in. The Italian Democratic Party had an extraordinary all-time high result of 40.8% (+15 from EE 2009, +16 from GE 2013) with 31 seats (+8). Conversely the EPP in Italy had its worse result ever (FI 17 + NCD 4= 21%, vs PLD 35 + UCD 8 = 43 in 2009, and 13 + 3 = 16 seats vs 29 + 6 = 35 in 2009). Furthermore, according to my calculations, because of the great number of PD voters (more than 11 million), the aggregated result of the PES parties in number of votes is similar to (or slightly higher than) the total result of the EPP parties (both around 25%). So, Junker and Schultz begin their course on an equal foot, and there is a strong possibility that a third person will be chosen, with the consensus of ALDE.
    As far as the makeup of the “Others” is concerned, these can be divided into:
    1. The new far-right group headed by Marine LePen, with 35 MEPs (they can become 40 with the LN).
    2. The neo-nazis of Hungary, Greece and Germany (6 MEPs).
    3. The Italian 5 Star Movement of B. Grillo (17)
    4. The German Euro-phobes of AFD that do not appear as xenophobic etc. but only as opponents of the Euro (7).
    5. Various “alternative” movements such as the animal rights defenders (2 MEPs from Germany and Netherlands), the Swedish feminists (1), the Spanish protest movement Podemos (5), the Portuguese Movement for the Earth (2) etc.
    6. Moderate center-left parties moving between PES and ALDE, such as the Spanish UPD and Greek To Potami (The River) (7 MEPs).
    7. Various center, center-right and independent parties and MEPs.

  45. @Virigilio

    Is the Europarliament left or right wing?
    Has this election boosted the left or right?

  46. JOHN B
    “Do we know if people in safe seats (who may well feel their vote achieves very little either way) are more likely to vote UKIP?

    ————–

    I can tell you that of the 10 highest UKIP votes at the 2010 GE, five* were in seats that had produced Conservative majorities of 15 percentage points or more in 2005 (and therefore could arguably have been considered safe in the context of the 2010 election), two** had had more modest Tory majorities in 2005 (but were presumably also considered pretty safe in 2010), two*** were Labour-held West Midlands seats which – as it turned out – would have fallen to the Tories in 2010 if just one-third of UKIP voters had voted Conservative instead, and the other was the Speaker’s seat contested by Farage (Buckingham) and so was a special case.

    * Boston & Skegness (2010 UKIP vote: 9.5%)
    Christchurch (8.5%)
    Spelthorne (8.5%)
    Cambridgeshire North West (8.3%)
    Devon East (8.2%)

    ** Staffordshire Moorlands (8.2%)
    Dudley South (8.2%)

    *** Dudley North (8.5%)
    Walsall North (8.4%)

    Also of interest: Newcastle-under-Lyme (8.1% – half of that would have won the Tories the Labour seat), Wentworth & Dearne (8.1%, ultra-safe Labour); Hull East (8.0%, ditto).

    Of course GE voting patterns have changed since then, so who knows. It’s fascinating how this will pan out over the next year.

  47. Thank you Virgilio – your post was the one that I was waiting for.

  48. NeilVW

    The demographics of the UKIP vote has changed and is still changing, which is one of the reasons their impact is so hard to predict.
    A couple of years ago UKIP supporters were supposed to be mainly right-wing tories from the shires. More recently I have seen surveys that show that UKIP are the most working-class party, and I have seen a theory on this site that UKIP do best in poorer areas bordering big cities. There is also no doubt that a lot of their support comes from people who haven’t voted for years.
    As many of these voters will not be particularly politically-aware, I don’t think that tactical voting will come into play very much. Differential voting for UKIP will be more likely to be caused by local issues and by the effort put in by UKIP activists.

  49. @COUPER2802
    According to my calculations, the total number of MEPs that can be dubbed “left” or “center-left” has increased by approx. 10 MEPs, wheres there are approx. 25 “right” or “center-right” wingers less. (The new EP has 15 MEPs less than the outgoing one). This said, there is still an approx. 57-43% domination of the “right” over the “left”. Yet the “right” is much more fragmented in its makeup because of the increase of far-right parties of all kinds, so we will almost never see decisions taken along this divide. Most of ALDE and EPP MEPs are pro-Europe and pro-Euro and even federalists (with some notable exceptions such as Berlusconi, Orban, Babis etc), whereas ECR, EFD and the new far-right group have the opposite view (but still with divergences between them).

  50. Bury Lad,

    A quarter of the UKIP vote in the SE was 187859 and so they got the last seat.

1 3 4 5 6 7 9