Yesterday the British Polling Council had an event talking about how the polls had changed since 2015. This included collecting up data from all the companies on what they’ve done to correct the error and what they are now doing differently – all those summaries are collected here.

In looking at what’s changed it’s probably best to start with what actually went wrong and what problem the pollsters are trying to solve. As all readers will know, the polls in 2015 wrongly overstated Labour support and understated the Conservatives. The BPC/MRS inquiry under Pat Sturgis concluded this was down to unrepresentative samples.

Specially, it looked as if polls had too many younger people who were too engaged and too interested in politics. The effect of this was that while in reality there was a big difference between the high turnout among old people and the low turnout among young people, among the sort of people who took part in polls this gap was too small. In short, the sort of young people who took part in polls went out and voted Labour; the sort of young people who weren’t interested and stayed at home didn’t take part in polls either.

So, what have polling companies done to correct the problems? There is a summary for each individual company here.

There have been a wide variety of changes (including YouGov interlocking past vote & region, ICM changing how they reallocate don’t knows, ICM and ComRes now both doing only online polls during the campaign). However, the core changes seem to boil down to two approaches: some companies have focused on improving the sample itself, trying to include more people who aren’t interested in politics, who are less well educated and don’t usually vote. Other companies have focused on correcting the problems caused by less than representative samples, changing their turnout model so it is based more on demographics, and forcing it to more accurately reflect turnout patterns in the real world. Some companies have done a bit of both.

Changes to make samples less politically engaged…

  • ICM and YouGov have both added a weight by respondents level of interest or attention to politics, based upon the British Election Study probability survey. YouGov have also added weights by level of educational qualification.
  • Ipsos MORI haven’t added political interest weights directly, but have added education weights and newspaper readership weights, which correlate with political interest.
  • Kantar have added education weighting, and also weight down turnout to the level they project it to be as a way of reducing the overall level of political engagement in their sample.

Changes to base turnout on demographics…

  • ComRes have changed their turnout model, so it is based more on respondents’ demographics rather than how likely they claim they are to vote. The effect of this is essentially to downweight people who are younger and more working class on the assumption that the pattern of turnout that we’ve seen at past elections remains pretty steady. ICM have a method that seems very similar in its aim (I’m not sure of the technicalities) – weighting the data so that the pattern of turnout by age & social grade is the same as in 2015.
  • Kantar (TNS) have a turnout model that is partially based on respondents age (so again, assuming that younger people are less likely to vote) and partially on their self-reported likelihood.
  • ORB weight their data by education and age so that it matches not the electorate as a whole, but the profile of people who the 2015 British Election Study who actually voted (they also use the usual self-reported likelihood to vote weighting on top of this).
  • Opinium, MORI and YouGov still base their turnout models on people’s answers rather than their demographics, but they have all made changes. YouGov and MORI now weight down people who didn’t vote in the past, Opinium downweight people who say they will vote for a party but disapprove of its leader.
  • Panelbase and Survation haven’t currently made any radical changes since 2015, but Panelbase say they are considering using BES data to estimate likelihood to vote in their final poll (which sounds to me as if they are considering something along the lines of what ICM are doing with their turnout model)

In terms of actual outcomes, the pollsters who have adopted demographic turnout-models (ComRes, ICM and Kantar) tend to show larger Conservative leads than companies who have tried to address the problem only through sampling and weighting changes. We cannot really tell which is more likely to be right until June 8th. In short, for companies who have concentrated only on making samples more representative, the risk is that it hasn’t worked well enough, and that there are still too many of the sort of young engaged voters who are attracted to Jeremy Corbyn in their samples. For companies who have instead concentrated on demographic-based turnout models, the risk is that the pattern of turnout in 2017 differs from that in 2015, and that Jeremy Corbyn’s Labour really does manage to get more young people to come out to vote than Ed Miliband did. We will see what happens and, I expect, the industry will learn from whatever is seen to work this time round.

199 Responses to “How the polls have changed since 2015”

1 2 3 4
  1. CB11

    That’s a likely as anything I would agree.

  2. Back from garden – too hot.

    Re campaign posters my favourite window near me (Cambridge) has a Vote LDem in the top half of a very nice sash window and a vote Labour in the bottom half.

  3. RICH
    “Even moderate mosques now often have leaders who say the real enemy of Islam is liberalism and even liberal Islam.”
    This appears to be mainly related to two aspects of attitudes towards women: that of dress and that of involvement in public leisure and entertainment.
    It also, I think, relates to the association of violence with conflict between sects within Islam and between them and other cohabiting religious groups, certainly going backin recent history to pre-Indian-Pakistan Partition and continuing, particularly in Pakistan ever since, notably, for example, in Sunni-Shia riots in Karachi between groups who are themselves rural-urban migrants.
    The extreme practice of exclusion of women, which in its own setting is rational and peacable, is that of Pathan in the Swat Valley and similarly in large areas of Afghanistan – explained by an eminent Pathan with whom I worked as attributable to their way of life and the Muslim religion itself as belonging to the desert.
    Others are more informed than I on this, but it may be reasonable to understand their extreme codes of dress and seclusion, there and in much also of the Middle East,, as derived from conditions where protection of women and a separate women’s culture within walls or behind the burkha relates to the mediaeval conditions still seen in desert areas of these countries, and which would express itself also in conservative attitudes of senior generations towards exposure to different cultures in migration to cities.

  4. You get the impression that the Tory campaign has finally started in the last day or so. 3 weeks or so too late however, assuming it does become like a normal Tory campaign, I’d expect them to gain some ground in the polls. They have far far more recognisable figures than Labour and some very easy ways of attacking Corbyn/Labour

  5. I expect a slender firming up of the Tory lead after the collapse of the last couple weeks.

    My poll prediction for today (assuming we have one!):

    Con: 44%
    Lab: 37%
    LD: 9%
    UKIP: 5%

    If Labour *still* continues to advance today, however, then I’m forced to believe all bets are off on the idea of a Tory majority.

  6. The ‘dementia tax’ will return nearer to polling day, if Labour have their way.

    Keep it as fresh in minds as possible. Not hard to do – just announce what their cap would be, and then say to the Tories….over to you.

    While we’re sharing anecdotes…

    I knew of 5 people, after the election was called, who had ALWAYS voted Labour – including my own mother – who said they probably wouldn’t vote Labour this time.

    Partly through dislike of Corbyn and partly through admiration of Theresa May.

    Well… a few weeks later they have all had a complete reversal of position. All absolutely staunchly now voting Labour and see TM as weak and brittle.

    This must be an experience which has been replicated in millions of people across the country, since Labour have moved from 25% toward mid/late thirties.

    I still predict a moderate Tory win. But I wouldn’t rule out a Lab minority government forming with SNP support, in a scenario of a hung parliament with the Tories the largest party.

    “The fact that they aren’t completely out of sight is down to the awfulness of their campaign ”
    How can you be sure that it is not down to the awfulness of the Coalition and Cameron’s governments,, or to a real resurgence of demand for an end to austerity and a demand for public services at the level of other major European economies, or to a demand of people below 45 for an economic system which provides them with an adequate income and a house?

  8. ADAM

    In what way has it re-started? All I see are them responding to Corbyn with some personal attacks which don’t reflect what he actually said.

    Did you see the car crash of Fallon on Channel 4 news which is going viral? It’s incredible (and not in a good way).

    (he was provided a quote abour foreign intervention in Iraq increasing the threat of terrorism, not told it was from Boris Johnson, assumed it was from Corbyn, and launched an attack… when he was told it was a actually from Johnson…priceless)

  9. Having been a veteran on here of the 2010 and 2015 general elections I am enjoying the predictable (campaign period) resurgence of the anecdotal “I know of X number of people who would never vote for Y (until now)” or “…who originally planned to vote for Z but will now for for Y” or some derivation thereof etc etc



  10. Joseph

    It could be an (anecdotal) indicator that Labour have peaked too soon. Memories are short, and most people revert to long-held voting loyalties in a general election don’t they?

    Anyway, I wouldn’t be entirely surprised if polls next week start showing a slow drift back to Conservative. Though in the end it’s likely they will fall short of their expectations when they called the election, and incredibly unlikely they will exceed them.

  11. @CROSSBAT11

    I wonder if this election might be a re-run of 1992 in terms of vote share for the two main partie

    I keep on going in a loop wondering between 1983, 1987 and 1992. Recent polls are making me think ’92, but given all that’s happened in the last two weeks who knows?

  12. DRMIBBLES, it’s on the previous page (Fallon).

  13. Rob sheffield, Well yes, but last time was much much less exciting and I genuinely feel the VI is more febrile as the polls vacillate.

  14. On the topic of anecdotes, I was at my son’s cricket club bbq last night., and must have chatted to about 40 people. Not once did the topics of the election, terrorism, social care, May or Corbyn come up (even after 4 pints of stella).

  15. Rob

    Nice that you’re happy

  16. @ Rob Sheffield

    Ah, but in prior campaigns people used to use “on the doorstep” a lot more. Today’s anecdotes are so much more sophisticated – full marks for creativity ;-)

  17. Well, I try to assume good faith, as I strain every sinew.

  18. @Sssimon – bear in mind 37 would constitute an advance for Labour with all pollsters except YouGov

  19. Hi SSSimon

    Haven’t various political memoirs shown that the use of “…that is not what I am hearing on the doorstep..” was mainly a device to deflect from whatever bad news was being presented and usually not entirely true?

  20. @ Wes

    Good point. That prediction applies only to a YG poll.

  21. On the topic of anecdotes a colleague of mine was mocking Corbyn yesterday on his foreign policy/terrorism speech even though I recalled this chap on several occasions pretty much making that point – the anti Corbyn sentiment is so strong with some that it makes them change their opinion on something rather than agree with him

  22. @ Woody

    Certainly – it’s been the ultimate ‘get out clause’ for politicians trying to evade questioning on whatever piece of bad news was on their plate that day :-)

  23. Whatever ones political persuasions, it is fairly clear that LAB are playing their hand well.

    National security was to be the CON trump attack card, but LAB stole a march on it and according to polling, most people agree with Corbyn’s position. He has not only neutralised their strongest attack line, he may even have benefited from it.

    LAB are unquestionably controlling the agenda, and briefing the media effectively (for example, by following Corbyn’s speech with handing out quotes to journalists of Tory Ministers saying exactly what Corbyn said). I bet CON are wishing their campaign was so well planned and thought through.

    CON cancelled their campaign launch event yesterday and don’t seem to have anything I’m aware of planned – unless I missed something? Seems like they are not quite sure what to do.

    I expect a predictable return to strong and stable and brexit soon – but is anyone listening?

  24. DrMibbles

    There’s definitely been a massive ramping up of the “Corbyn is a traitor” line in the last 24 hours and I think that will morph into Brexit fairly soon. In fact, I think Con will abandon any pretence to making a case for their own positives and will absolutely hammer away at the perceived weaknesses in Labour and particularly Corbyn.

    This is Lynton Crosby territory now. Though, as demonstrated by Fallon, it relies on the Tories not making a complete horlicks of it. And also on it not turning into a Zac Goldsmith-style disaster.

  25. Terrorism definitely has something to do with our foreign policies according to some vile Islamists scum.

  26. My partner received another you tube ad today from the conservatives. Yesterdays was very anti JC stressing his previous associations.

    Today it was ‘ strong and stable’ again, stressing JC would mean chaos for Britain during Brexit.

  27. I thought this was going to be the most boring election ever. True, I could see that May is a wooden public performer, with some muddled views, and therefore vulnerable to someone of the calibre of say Robin Cook, but everything I’ve seen in the last two years made me think that Corbyn’s team, however right-on their views, were unprecedentedly incompetent. How wrong I was! And very happy to admit it! Were they lying low all this time, deceiving us? Anyway, it’s all got exciting and I sense a few others here suddenly getting interested.
    Not that I think for a moment May won’t win. But I don’t think she’ll ever fully recover either, whether in the polls or in the Tory party. It’s even a while since I saw one of those paeans to her competence and trustworthiness that used to crop up regularly on this site.


    I know, you couldn’t make it up…oh wait

  29. Anyway, enough of this frivolous stuff, has anyone noticed that Corbyn has changed his hairstyle? It looks a lot like an old man version of the George Osborne style caeser-cut.

    Clearly preparing for power.

  30. @Rich

    You would find it interesting an useful to read “Empire of Sand” by Walter Reid to provide you with some historical context regarding the UK and the Middle East.

  31. Tuition fees.

    The massive political shift within the Labour Party from 2015 to today, apart from anything else, has been from Milliband’s obsession with portraying fiscal rectitude and economic competence to Corbyn’s targeting of social mobility first.

    This has had a huge disruptive impact on the status quo even when compared to the recent local elections.

    But polling data is not passively received and the recent YouGov poll, more than any I can remember, is feeding back into Tory strategy. Not particularly successfully. The Tories are looking like fumbling amateurs.

    With the tuition fee revolution, Corbyn will in any case poll more highly than Milliband. What other explanation is there for the never retained losses suffered by the Lib Dems after 2010?

  32. Rich Hireton,

    This is a trivial representation of some of the turmoil.

  33. @Rob @MarkW

    I remember 2015, 2010 and 2005. No idea what my handle was back in 05 but my recollection was that 2010 was the worst for partisanship amongst those three. 2017 is in a whole new league though. Everyone needs to calm down. AW has an impossible job.

    @Laszlo – From an earlier thread on the Con Manifesto

    – I agree with your opinion on it. Shockingly I agree with most things you’ve said over the last couple of weeks. You’ve either mellowed or become more considered or I am now more radical. It’s confusing – please return to baseline!

    @S Thomas – The so-called Mao thought on the impact of the French Revolution of 1789., “too early to say”

    – Unfortunately, this is a myth even though the quote is so wonderful if one compares it to the belief in the eon-term thinking of the Chinese.

    The reality is it was almost certainly Zhou (1st Premier) who was talking in 1971 about the French student uprising of 1968.

  34. In economic news, EY has published its annual survey of inward investment in the UK. It shows a solid short term performance but overall EY conclude that every positive indicator is offset by a negative medium to long term development:

    Scotland continues to perform well attracting record numbers of inward investment projects and EY concludes that “Scotland has firmly established itself over the last five years as the UK’s second most attractive destination behind London.” Scotland has also emerged as the UK leader in attracting software R&D projects.

    The EY report shows Scotland’s attractiveness fell 1%. EY conclude that “although not alarming in itself, the research suggests that the EU Referendum vote and its aftermath may be having an influence on global perceptions of Scotland and the UK’s long-term attractiveness.”

  35. The oddest thing about the Con manifesto was how far removed it was in reality from the expectation that May had encouraged.

    I had expected some genuinely new thinking which would move the public’s view of the Tories well to the centre and maybe even a bit to the left.

    In the event there was nothing in it, that I am aware of, that came remotely close to fulfilling that, and certainly absolutely nothing that one could say defined May against the image of her own, self-styled, “nasty” party.

    I really don’t understand why she said what she did and then cam up with a “boring as she goes” strategy.

  36. any poll expected tonight?

  37. @markw and @rich

    That’s a good graphic but doesn’t capture the UK/USA activity in undermining governments ( including the promotion and supporting of coups) in both Iraq and Iran when they didn’t like those governments’ economic policies even when the borders remained stable!

  38. @Steven Wheeler

    >@Paul Croft
    >In fairness he did make pretty clear he wouldn’t negotiate with ISIS. >Perhaps he should have explained his reasons more clearly though.
    >From text of the interview:-
    >AN:Would you talk to them [terrorist leaders/ISIS]?
    >JC: No, I wouldn’t.

    As so often, that is interestingly different from his previous position, which was to seek a political solution including ISIL.

    Marr interview Jan 2016:

    “AM: My question is should we be talking to Isis?
    JC: There has to be some routes through somewhere, because
    remember a lot of the commanders in Isil, particularly in Iraq but
    to some extent in Syria, are actually former officers in the Iraqi
    army, because we made many catastrophic mistakes, one of
    which was to destroy the whole Iraqi state structure after 2003.
    AM: Absolutely. So we could have a dialogue with these people?
    JC: No, I’m – dialogue is perhaps the wrong word to use. I think
    there has to be some understanding of where their strong points
    are, where their weak points are and how we can challenge their

    I think Mr Corbyn needs to explain how you understand the strong and weak points of a gang of genocidal, slave-trading murderers.

    OTOH he may come around to admit that military intervention has been quite effective in degrading and shrinking DAESH / iSIS.

  39. baldbloke,
    “I wouldnt pay any credence to booing from an R4 AQ audience”

    Well you should. Dimbleby as chairman in an election period, so hyper sensitive to bias, sounded incredulous that Davis refused to accept the views quoted of a government security officer. Day before, differnet program, when they tried to find people to comment on Corbyn’s statements, they were again no doubt sensitive to impartiality and said they were hard pressed to find any members of the public who disagreed.

    Even if it is only one segment who feel this, it is a constituency who agree with him. Only takes 1/3 to win. Alan Johnson, observing that in his time as home secretary, libya was a helpfull country when it came to stopping terrorists. Its a drip drip, which people will see. Little things in the media. Dont look good signs, to me.

  40. There are two ways in which western intervention in the Middle East may have led to terrorism:

    1. Provoking it.

    2. Providing it with the opportunity.

    The latter is fairly undeniable. No Iraq invasion, no ISIS – at least until as Adam died.

    But I think Corbyn is talking more of the former. The West has created the grievances. At least historically that has been his line since forever.

  41. @phil

    DAvies was booed because people felt he was (deliberately) missrepresenting corbyns argument (as is much of the tory/media attack on him).
    One of his main points was that western intervention in the middle east has created the chaos which has allowed the like of ISIS to prosper. And this a point of view held – and argued for – by everyone from the former head of mi5 to boris johnson.

    However – what we should be discussing on here is how it will play in the polls – will the tory attack on corbyn hit home with voters – or will they see (as the any questions audience seemed to) as a straw man.

  42. Hi Reggieside

    A question: What proportion of the 40 million + people who have a vote do you think

    a listen to/watch AQ/QT or similar and

    b of those, are swayed by what they hear or see?

    It seems much more likely the tory attacks will hit home because most people will get their news from News soundbites and if that is all the news has that is what they will show.

    Whether they believe them or not is a moot point and it seems likely to me that it will only entrench current views not change them, but that is all the Tories need to do, they are leading in the polls.

    National security was to be the CON trump attack card, but LAB stole a march on it and according to polling, most people agree with Corbyn’s position

    I usually agree with most of your posts but I think you’re wrong in that being the main factor. I suspect May thought it would be the Brexit election and that all they needed to do was to repeat strong and stable ad infinitum in order to harvest all the non-Lab leave voters.

    As it is, Macmillan’s “events dear boy, events” has given Lab a real opportunity to propose popular policies for which the Cons seem not to have prepared.

    If the previously non-voters who turned out to vote for leaving the EU like those policies then Corbyn could have a majority. If they do not the best he can hope for is a hung parliament.

  44. @woody

    oh agree – whatever happens on Any Questions is very small beer. I was talking about the wider tory attack of saying that corbyn is somehow arguing for “appeasement” with ISIS and suggesting their is some justification for jihdai terrorr attacks.

  45. GB,

    Since we seem to be interpreting election posters in a sort of biblical dream way, does that mean the Lib Dems are on top in Cambridge?

  46. I eagerly await tonight’s polls but my hunch is this is still the Brexit election despite the news not really talking about it. Domestic policies have had an influence but ultimately I think it’s still in people’s minds, and unless Labour/Corbyn has a change of heart and takes a strong stance on Brexit, we won’t see anything other than a Tory majority.

  47. First ever post…

    In relation to people quoting ISIS sources (including Andrew Neil yesterday) as saying interventionism was not a terrorist recruiter, do you really think they would be honest on this? If interventionism did help turn some people to terrorism then surely ISIS would have motive to downplay the role of interventions.

  48. Election poster update from Teesside:

    Darlington (LAB held, CON target): no posters evident from any party. 1 LAB flyer through the post, many friends have had letters from TM (but not me)
    South Stockton (CON held, poss LAB target): 1 CON poster noticed so far.

  49. Barbazenzero,

    I agree that the u-turns have made Theresa very vulnerable on the “Strong and Stable” front.

    Nevertheless, the movement in the polls still seems to be Labour collecting votes from Don’t knows (mainly), Lib Dems, Greens, and even UKIP, rather than the Tory vote share slipping appreciably (other than in Wales, and that from an improbable looking high in one poll).

    The Lib Dems will I think be hoping that if trust in Theresa May erodes and Farron manages to pass the Andrew Neil and other tests (when the refugee policy among other things is likely to come up), that they can get some Tory Remainers into their camp who will never vote Labour

  50. Tweet from Mike Smithson.
    6pm – Comres online for IoS/Sunday Mirror. Last time Con led by 18%.

1 2 3 4