ICM’s poll in the Sunday Telegraph is intriguing. Rather than asking voting intention, it asked people to estimate what shares of the vote each party would get in an election tomorrow.

I’m not entirely sure what to make of it – largely because right now, all we have are some figures: Conservatives 31%, Labour 39%, Lib Dems 16%. Taken in isolation they are not particularly interesting, we already knew that people tend to expect Labour to do better than the Conservatives at the next election, and given that ICM tend to show high levels of support for the Lib Dems anyway we don’t know if this is people predicting a higher level of support for the Lib Dems than they currently have… or a sample that would have shown that sort of level of support for them if asked in a conventional method.

What will be a lot more interesting is seeing how it works it over time and how it relates to conventional polling. ICM have been running it in parallel with their regular tracking polls, and have apparently found some consistent patterns of differences and tested it at the last election which it got correct. Martin Boon has a proper academic paper out on the method next month, so I’ll be reserving judgement on what it can tell us until then.

Note that ICM are NOT abandoning their normal method. Their regular Guardian polls will continue in just the same way as usual. This is just complementary.

31 Responses to “ICM and the wisdom of crowds”

  1. The ICM poll doesn’t seem that different from the habit on UKPR of people predicting tonight’s poll!

    I popped in to B&Q today, and found it festooned with Union flags. I knew Rangers FC were in trouble, but selling them off in a hardware store seemed a bit inappropriate.

  2. Doesn’t the fact that 16% of people think the LD’s will win the next election show that this is more based on hope than estimation?

  3. No – it asked people what percentage of the vote they thought the Lib Dems would get in an election tomorrow, not who they thought would win.

  4. This seems a poll which requires a degree of interest in political news , which a normal VI poll does not.

    It takes very little interest in current political news to say how you intend to vote.

    But estimating what shares of the vote each party would get in an election tomorrow would seem to require considerable familiarity with the latest VI polls. What other basis would there be for making this estimation?

    Is the result of this poll just the feedback loop from reading VI polls, mitigated by partisan or other subjective personal political opinions?

    Seems odd to me.

  5. This is fascinating … I presume it wont be weighted in to your UKPR average.

    My own skepticism is on how much people are really clued up to how shares of the vote work and what is realistic in the current climate … but maybe that is because I want to think that this is very complicated and only clever people like me can work it out. Actually maybe all people are pretty savvy.

    What will be interesting is if the figure people estimate for the shares of the vote change a great deal with polling. Will people judge that flim-flam like hackgate and a pasty-tax u-turn will be pretty much forgotten and think of longer-term trends … or will it jump about with the winds of popular change?

  6. Interesting idea. But we will need a lot more data than just the raw figures. That is the same as knowing the mean without knowing anything else about the distribution. It is not very useful unless we also know something about the spread, skewdness, pointedness, and the overall shape of the distribution.

  7. Anthony

    16% for the LDs still sounds like ” hope than estimation”.

  8. @ AW,

    Isn’t there also the danger that this is just reflective … so that people are simply reflecting back what they know to be the case according to the opinion polls.

    So does it tell us anything useful (apart from that people can accurately remember a rough idea of where the polls are)?

  9. @ LDs 16%,

    Even with all their travails, I think this figure will be actually pretty accurate, as they always do better in all elections than in mid-term polls.

  10. I wonder how many parties they ask people to forecast? Presumably it has to be made to add up to 100…

  11. This is interesting. No doubt some of this response will be people repeating what they remember of the latest polls, but many won’t follow the polls, and so presumably are picking up on a general mood.

    When the paper is published it might give us a new angle on how to interpret polls.

  12. Adrian B – quite possibly. An interesting thing is that it’s two stages. ICM ask people to predict what they’d get, then tell them what each party got at the last election, then ask them to predict again. Will be interesting to see what the difference is between those two sets of figures.

    Matthew – dunno. I’ve asked Martin but he hasn’t got back to me yet on it

  13. I’m a thread behind here but, on the basis that obsolete threads tend to die a natural death, I thought I’d post a thought on the Jubilee polls here, if I could. I think they are prime examples of polls that prompt opinions where none really exist and then allow people, maybe with vested interests in doing so, to extrapolate highly misleading conclusions. I suspect there is much more indifference and loosely held opinion towards the monarchy than these polls are measuring. The gung-ho, happy-clappy, flag-waving, “get-out-of-my-cafe-if-you-don’t-stand-up-for-the-national-anthem” types are fairly few in number, probably no more or no less than the “off-with-their-heads” republican zealots. Polar extremes, if you like, with a vast amorphous mass of attitudes and opinions in between. My guess is that most of these in the middle are vaguely sympathetic towards the current Monarch, admiring of the way she has conducted her difficult role, quite happy to watch and participate in a few public events in her honour that allow a knees up and a few bevies in the warm sunshine, but who wouldn’t lose a lot of sleep if they woke up one morning to find that the Royal Family had walked off into the sunset. A lot of these people may well, when prompted for a yes or no response in a poll, plump for keeping the Monarchy but really have no strong opinions either way. It’s probably how I feel too, although I do like to tease arch-Tories that I know with some light-hearted republican posturing. The truth is that I don’t really care and have no strong feelings about the issue. I suspect I’m not alone either.

    As a country, we’re a pretty tolerant and easy going people and, mostly, will go with the flow, but the idea that this is a country where 80% of the population are die-in-the-ditch monarchists is both risible and preposterous.

  14. Crossbat – sat in the YouGov office we thought that too. So after asking if people supported the monarchy or a republic we asked how strongly.

    Of people who said they thought Britain should continue to have a monarchy, 47% said they felt this very strongly, 35% only fairly strongly, 15% not very strongly and 3% not strongly at all.

    Of people who said they thought Britain should have an elected head of state instead, 30% said they felt this very strongly, 40% only fairly strongly, 23% not very strongly and 4% not strongly at all.

  15. On an unrelated note, but something that should be interesting to SoCal atleast – YouGov/Michael Tesler research on the ‘racialisation’ of American politics.

    Effectively those who have positive or negative views of black people will change their ideological positions depending on what position Obama takes – even when controlling for party or self-identification.
    Across many issues – views on gay marriage, healthcare policy, tax policy and even how favourably they view dogs.

    While I’d expect a similar issue over partisanship here (leader x supports policy, policy is awful), it’s interesting to still see such a racial bias in US politics.

  16. The ICM wisd ind
    Con 31%, Lab 39%, LD 16%
    is highly reminiscent of the popular vote projection for the 2012 locals
    Con 31%, Lab 38%, LD 16%.

  17. AW – that last piece of data is quite interesting.

    If one plots figures starting from stongly support for one position through not strongly at all and then to strongly for the opposite position, then one might expect to see an inverse bell curve. The extent to which the bell shifts in one direction or the other, and the flatness of the curve, are both useful guides to how strongly a position is really held and to what extent opinion may be swayed.

    Your figures suggest that not only do those supporting the monarchy outnumber those opposed quite considerably, they are also more solidly supportive of the crown than those who might prefer an elected head of state.

    It may well be that the Jubilee celebrations have made people more strongly supportive of the monarchy at this time, and people may become less committed at a future date….but in my view, events like these have an emotive value on the quality of life index that cannot be measured on economic indices.

    Question for republicans – how does a motorcade with motorcycle outriders compare with a coach & horses with cavalry escort ?

  18. @Crossbat,

    I agree and disagree with you. I do think “flag-waving royal lovers” are a fairly small minority. Apart from anything else its pretty “uncool” to show that much enthusiasm for anything in British culture (I am glad to be terribly uncool in this regard!).

    On the other hand, I do think that a large chunk of the latent monarchism is actually pretty die-hard. There isn’t really that much debate about republicanism in the UK, and I think that generally speaking most people like it that way. I think in the event that there was a serious attack on the monarchy (politically, or God forbid, physically) then you might see the Royalist Lion bear its teeth.

  19. On the wisdom of crowds…

    Firstly, just because 90%+ of the electorate are not very interested in political polling, does not mean they do not understand where the parties stand and how they fare. Indeed, if that were not so, it would justify a Techoncratic approach would would soon see democracy sidelined since the views of the uninformed disinterested public do not matter….

    Now where have we seen that attitude displayed in recent times …?

    Secondly, while many (if not most) people may not change their voting inclination, their propensity to vote will often be driven by where they see the relative strength of parties. That judgement of whether Party A is up or B is down is part of the calculation many people make as to how importnat it is they get out to vote.

    In other words, this methodology is not vastly different from the process employed by the voters themselves.

  20. @Anthony

    “Of people who said they thought Britain should continue to have a monarchy, 47% said they felt this very strongly, 35% only fairly strongly, 15% not very strongly and 3% not strongly at all.”

    Has there been any polling done that compares how important the maintenance, or removal, of the Monarchy is to other issues like crime, employment, inflation, immigration etc? I’m still sceptical, despite the data you’ve shared, about how important an issue it is to people. I mix fairly widely with people of many and varied backgrounds and views, and I get no sense that the continuation of the Monarchy is an issue that stirs real passion or interest.

    I guess if you’re asked a question in a poll and and you are given a menu of responses to choose from, to say that you have a strong opinion on something isn’t a bad default position to fob off a persistent pollster!!

  21. @ Anthony

    It will be interesting to find out how many people felt unable to make a prediction.

  22. Paul H.E, well the coach and horses belong in cinderella,and the motorcade
    Seems very 21st century.

  23. Seems odd inasmuch as you’d expect more people to underestimate the LD position given the history of the past two years. The 16% suggests quite a clued up understanding of their past ability to bounce back which, since I thought I was the only who was clued up, comes as a surprise.

    Mind you I still think it will be less than that.


  24. Now this interesting, I am surprised how low the Conservatives are here, I would have put them at 34% and Labour 38%. Are people really so expectant (keen?) to hand Labour a majority similar to that they received in 2005 (effectively reversing the 2010 election) – that is quite a lot, and quick too. Perhaps people need to see some policies and then maybe the gap would not be so big.

    As for a motorcade with motorcycle outriders, I think there’s a bit less manure to pick up after they’ve been past. Not that manure is bad, it’s just often dropped in the wrong place.

  25. @Neil A

    “I think in the event that there was a serious attack on the monarchy (politically, or God forbid, physically) then you might see the Royalist Lion bear its teeth.”

    I agree with you that there is currently no great enthusiasm for out-and-out republicanism in this country, but I’m not sure that those who would say that they were strong monarchists muster a vast number either. I think, in a typically English way, that there is a sentimental and nostalgic attachment to the Monarchy and it provides a comforting narrative to people who really don’t think too deeply about it as an issue or institution. I’d compare it to people who have no great interest in cricket but who find it’s continuing existence, somewhere in the distant cultural ether, gently reassuring. Would they sit through a Test Match? Probably not, but they quite like the fact that it’s going on because, maybe mistakenly, they think it’s part of what makes us who we are as a nation. In a way, that is the definition of sentimentality to me; a shallow attachment to something more mythical than real.

  26. OLDNAT
    ‘16% for the LDs still sounds like ” hope than estimation”.’

    A rare contributor these days, but one who does from time to time reads the postings; they keep me upto date and informed and you keep me amused. Never forget that you were once a Liberal.

    However, times are changing, we have lost another Councillor to the Greens locally, and his son, a good friend, who just lost in the local elections has also changed allegiance. Given that a few months ago another LIB Dem councillor joined Labour, I am now in a position where my two favourite lib dem activists are Green and Labour respectively.

  27. @Henry

    You might like this one:


  28. CROSSBAT11

    @”I think, in a typically English way, that there is a sentimental and nostalgic attachment to the Monarchy and it provides a comforting narrative to people who really don’t think too deeply about it as an issue or institution.”

    It seems to me that you are trying hard to find some reason why the support for Monarchy in UK is by default-not serious-accidental-anything less than the reality which you fear.

    I think your effort is in vain. I really do think that the popularity of the monarchy in UK is real, and cerebral.

    I suspect that it has to do with a sense of permanence in an ever changing world, and to that extent, the 60th year of a reign is emphasis of that feature , and represents a comfort of sorts to it’s supporters.

    However, the ( understandable) miscalculation which the Windsors made over the mawkish response to Diana’s death; together with the lower level of support for Charles, shows that we have to acknowledge how much of the support for monarchy is vested in the person of the Queen.

    She has certainly managed to keep support for the institution of Monarchy at a level which the wider family of the House of Windsor has consistently done it’s best to destroy.

    Things may change a great deal if and when Charles becomes King.

  29. @Adrian B

    It is my hypothesis that in the past the LibDems have had trouble getting noticed between elections and have thus received a boost from the publicity during an election campaign. It will be genuinely interesting to see whether this boost happens this time as they have had no trouble getting noticed during this parliament!!

  30. Men think in herds, they go mad in herds, they learn the truth one by one. But I can’t recall who said that.

  31. Kieth P “Now this interesting, I am surprised how low the Conservatives are here,”

    I think the coservatives might fare badly on this estimation because so many labour voters live in heavy concentrations. I used to live in Liverpool for a while, and the level of hatred and vitriol aimed at people who aren’t labour can be hefty. I think a good number of them don’t know enough (or indeed any) right-wingers to be able to guage the tories’ hand.

    I’m still surprised by the libs’ standing though… even if it does match my own guess fairly accurately. maybe it reflects that people deep down feel that they are giving them a hard time… we will never know