I’ve seen various people pick up this article by Julian Glover today, so it probably needs some background. On Wednesday there was a joint conference with the BPC and the National Centre for Research Methods and at the end of the day the chairman put all the speakers on the spot and forced them to make a prediction of the election result (actually he asked them to predict the Conservative lead at the general election, but everyone apart from Nick Moon gave predicted shares).

The title of this article is a line that Bob Worcester likes to roll out at election time when he is asked the question (and in fact, Martin Boon from ICM prefaced his answer on Wednesday with it). Opinion polls ask what would happen in a general election tomorrow – apart from the final eve-of-poll figures when there actually is an election tomorrow, they don’t try to predict what will happen in several months or weeks time. Wednesday’s predictions are just the personal opinions of the particular pollsters who were there.

For the record though, as Julian reports, Nick Moon of NOP predicted an 8 point lead. Simon Atkinson of MORI predicted a tight 4 point lead. Everyone else clustered closer to a 10 point lead, the highest specific lead predicted was Andrew Cooper of Populus who plumped for CON 41% to LAB 29% (though Martin Boon went for a Tory score of “one point more than the next highest”, so I suppose he actually gave the most Tory answer).

Martin’s answer probably gives you the hint that these were not the most serious of predictions! As Peter Snow says, it was just a bit of fun. Before someone asks, I wasn’t speaking so didn’t get put on the spot for a prediction ;)

Anyway – for those anxiously waiting for the next poll, I’m expecting at least one this coming Sunday.


99 Responses to “Polls don’t make predictions, but pollsters sometimes do”

1 2
  1. @ Alec:

    Where’s the fun in that? Poll watching and number crunching can create an infinite amount of speculation from an infinitely small amount of data. It’s what I live for (when it comes to elections), at least as good as the policy analysis.

  2. Quincel

    Exactly! It’s why I love the subtitle of the “SNP Tactical Voting” blog which is

    “Elections – Probably the Best Spectator Sport in the World”

  3. @ Barry P

    I’ve read all your posts since you were pretty rude to/ about Sue Marsh. It seems to me & some other posters that you mis-read her post. An apology would be nice.

    Most of us try to be polite on this site; we’re not here to ‘score points’ off each other. Pointing out errors is fine but there is no need to add your personal (& unsupported) opinion of people’s math skills :-)

  4. Barry P

    You seem to have been rather condescending towards people.

    I would just like to point out as someone with a degree in Mathematics that you’re statistical methods are as flawed as any and where as I do not claim to be an expert in the field of statistical analysis I do have a fair degree of knowledge. (Although I’m sure there are people on here who have more)!!

    Above all it is rather easy to ‘twist’ statistics to support ones arguement (Politicians have been doing this for years.) The essential point being that Statistics are always widley open to intepretation.

  5. I confess I’m somewhat baffled by pollsters predicting that Labour’s share will increase significantly over the next 4 or 5 months. This presumably assumes Labour will campaign better than the Tories in the near future or that the economy will give them a boost. I can’t see anything to indicate that either is likely to happen. If anything, Labour seem more determined to shoot themselves in the foot than the Tories (plus having an unpopular leader) & economic recovery may show in statistics but is unlikely to filter down to the lives of voters in that time.

  6. @ Amber Star

    Indeed, I may have misread Sue’s commnet and if so, of course, apologise.

    However, I make no apology whatsoever for my replies to the posts of members, who it appears, might also have misread her post and, argued consistsently about the Tory vote.

  7. @ Sue Marsh

    If I read your post incorrectly, I must apologise for my inappropriate comments. You were quoting the Tory lead, not Tory share.

  8. I do not apologies for my comments to Sue Marsh or anyone else.
    If tomorrows poll shows C36 L34 LD21, I will not be posting comments which fly in the face of the facts, EG arn’t the Tories doing great. I did not insinuate Ms Marsh is innumerate, because I very much doubt that she is. The comment was the usual effort to shore up Labour morale, which is not a crime but does wear very thin at times, because quite simply Labour are currently not winning this election campaign.

  9. @DEREK PIERSON
    I am assuming that 6 or 7 pollsters participated Derek. I note 2 actually predict bad news for Cameron. In one case just short of an overall majority and in another Labour (just) largest single party. Allowing for just how seriously they all took the matter, the talking down of Tory chances was not particularly great when you look behind the headline.

  10. Roland –

    Simon Atkinson of MORI
    Nick Moon of NOP
    Peter Kellner of YouGov
    Andrew Cooper of Populus
    Martin Boon of ICM
    Prof Paul Whiteley of Essex Uni
    Prof John Curtice of Strathclyde Uni
    Dr Steve Fisher of Oxford Uni (I think – he was a speaker, but I can’t actually remember him being collared for a prediction)

    (Paul Whiteley is part of the British Election Study team which is what he was talking on, but is also BPIX. Steve Fisher was part of the exit poll team for 2005 and will be again in 2010.)

  11. @AW
    Much appreciated gov.

  12. Anthony, did anyone confront Paul Whiteley on why BPIX refuse to join BPC and why they never publish any of their tables or methodologies?

  13. Neil A

    Your list of potential “events” is extensive (though not exhaustive) and each has the potential to change the outcome one way or another but your assumption that one or more will make a difference leaves aside not only the possibility that they will cancel each other out, but that minds are already made up.

    Is it possible that election campaigning has run its course before it has officially started?

  14. GIN – I’ve no idea, I didn’t stalk him around all day ;)

    As I said, he was there talking about the British Election Study, not BPIX.

  15. There is an interesting little poll on Politics Home today.
    About 50% of people think we are still in recession and likely to be for some time to come.
    About a third of people think we are now coming out, further, a similar number give Labour credit for this.
    17% of people think we are up the creek without a paddle, and see no respite from doom and gloom for years.
    The numbers who think things are looking up and give Labour credit for it, are slightly higher than Labour support in the polls.
    On the other hand the vast majority do not believe we have anything to thank Labour for where the economy is concerned.

  16. Roland Haines

    Have to say that I’m dubious about any Politics Home poll since the change of ownership and editorship. The only reference I can find on methodology is that they poll by email.

  17. @JOHN B DICK

    “Is it possible that election campaigning has run its course before it has officially started?”

    It is possible that the underlying partisan assertion of your apparent hesitancy is correct- that minds are made up (you meaning: made up in favour of the Tories).

    But sorry to say it is equally possible that minds will change during the campaign- including during the 10 seconds that one spends in the polling station cubicle (a la 1992)…….

    That ought to give you and some (obvious) others on here pause for thought if not downright nightmares !

  18. @OLD NAT
    Fair enough.

  19. @ROLAND HAINES

    “There is an interesting little poll on Politics Home today.
    About 50% of people think we are still in recession and likely to be for some time to come.
    About a third of people think we are now coming out, further, a similar number give Labour credit for this.
    17% of people think we are up the creek without a paddle, and see no respite from doom and gloom for years.
    The numbers who think things are looking up and give Labour credit for it, are slightly higher than Labour support in the polls.
    On the other hand the vast majority do not believe we have anything to thank Labour for where the economy is concerned.”

    ….and your point/ contribution to the debate is…..???

  20. Rob Sheffield

    Roland was commenting on a poll. This is after all a political polling website.

    His post, was therefore perfectly reasonable.

  21. ROB SHEFFIELD- Some of us have been having nightmares for quite some time.

    Re your post to John B Dick-you are the first to identify
    partisan comment in favour of Tories from JBD.

    A remarkable insight into Mr Dicks opinions ,which had totally escaped me.

  22. John B Dick

    “Is it possible that election campaigning has run its course before it has officially started?”

    You may be right, but on this side of the border we have no way of knowing!

    Considering that in November YouGov had SNP on 24% (Lab 39%) while MORI had SNP on 34% (Lab 32%), that is a huge gulf between the two methodologies.

    While I still suspect that YouGov’s weightings fall down in Scotland due to their lack of understanding of the Scottish Press readership, we could be anywhere between a no seat change election, and an SNP breakthrough!

  23. As one of the contributers to the Politics Home survey for over a year now I do not care who, Ashbourn or Ashcroft, owns it. It samples regularly and measures the change in views of a large panel.

    I also find David of France’s bleating about comments tiresome. We are all adult and can take a bit of ribbing.

  24. @Roland Haines

    You said, “The numbers who think things are looking up and give Labour credit for it, are slightly higher than Labour support in the polls.”

    The specific number there was 36%, which is 7 points above the UKPR Polling Average. I’d not call that “slightly higher”.

    I’d also note that consumer confidence fell quite a bit in December, despite that month actually showing a huge improvement for retailers. It’s hard to translate the PoliticsHome poll into a consumer confidence figure. The December GfK NOP poll had consumer confidence showing the majority thought the economy would get worse in the next twelve months. This one shows only 17% think it’s actually going to get worse, while the rest think it has either recovered, started to recover, or will recover in a while.

    I’ve long held the opinion that improvement in consumer confidence has translated directly into improvement for the government’s polling, and vice versa. So if PoliticsHome’s polling is any way representative of consumer confidence, we should expect a further boost to Labour. But I think that their methodology here is a bit suspect, and the question asked isn’t a good consumer confidence measure, so I’d take it with a huge amount of salt.

  25. Glenn Otto

    But how is the panel constructed? What factors are used to make it demographically representative?

  26. “The specific number there was 36%, which is 7 points above the UKPR Polling Average. I’d not call that “slightly higher””

    It’sa LOT higher – remember that voting intention figures are repercentaged to exclude don’t knows and non-voters while no other questions are. Once you include non-voters and don’t knows Labour are in the low twenties or so, hence there are a large chunk of people optimistic about the economy who are not presently voting Labour.

  27. @ROB SHEFFIELD
    Let me apologise at once for posting a comment on a poll which covers political veiws in the UK. Especially one which does not show the party of government in a very good light.

  28. @JAY BLANC

    “The specific number there was 36%, which is 7 points above the UKPR Polling Average. I’d not call that “slightly higher”.”

    As Anthony says its a lot higher, however, I did not wish to over egg critisism of Labours performance.

  29. @JOHN B DICK
    Just checking you are well John having been mistaken for a Tory?

  30. @ Old Nat

    To be honest I cannot give you an answer. An educated quess is that members of the panel are sorted into “loyalty” and weighted accordingly. I have been faithfully punching in my answers as I feel for seemingly ages, I suspect they tend to measure the changes in the panel answers. Suffice it to say that the questions tend to similar on each accasion and have not changed over the period I have been involved.

  31. 3 local by-elections on Thursday

    1. Erewash BC: C hold [ -22% ] LD [ + 37% ]

    2. Wakefield MBC Lab hold [ + 3.2% ], C [ -7.7% ]

    3. Calne TC . LD gain from C.

  32. @SURBITON
    You are sooo yesterday. I already posted this information for Chris on the immigration thread 2 days ago. Clearly a re run of 1945 in the offing.

  33. LOL! Thanks anyway Anthony. :D

  34. Anthony. Why, as a member, were Angus Reid not present.? A discussion of the difference between their figures for Labour and the rest would have been particularly helpful.

  35. Well if we have a new poll in the Sunday papers that leaves us with about 4 hours to kill.

    Why leave all the fun to the pollsters: Anyone fancy predictions for:

    a) tonights poll
    b) The GE

    Then we could average up all our predictions and make our own UK Polling Pundits Prediction. We have enough staitsticians to do the maths.

  36. a) 40/30/20/10
    b) 40/30/20/10

  37. Collin – they were, but they weren’t a speaker.

    Others – saying “something might affect the polls” is not an excuse to start a thinly veilled political argument, so stop it. There are other places for such things. I’ve pruned both sides. Roland in particular, please stop trying to provoke other posters.

  38. @ BarryP

    Thanks for reading my post. Respect to you for a sincere post.

  39. It’sa LOT higher – remember that voting intention figures are repercentaged to exclude don’t knows and non-voters while no other questions are”

    By the same measure, though, the number who give the government no, or not much credit, also exceeds the combined total of voters for non-Labour parties.

  40. Shopkeeper Man

    a) 39/31/20/10
    b) 38/32/21/9

  41. Shopkeeper Man

    But I retain the right to change my prediction as the election unfolds. Thanks ;-)

  42. a) 41:30:20:10
    b) 41:33:19:7, Tory majority of 30-50

  43. a)40:29:18:10
    b)42:27:18:6

  44. As a stalker of this board I do enjoy the great work Anthony does and also the enlightened and informative posts (most of the time).

    I’ve been interested in the data that political betting and others have been posting for some time. It would appear that pollsters tend to over estimate Labour figures before an election. In addition the personal ratings would indicate a firm Tory win come election day.

    Obviously much can happen between times but I don’t think I have read anything that would give much comfort to those that wish to see Brown as Prime Minister for another 5 years.

    As for Sundays poll:

    42/28/20/10

    Election – I would expect a Tory majority of between 30-50 seats.

  45. My Predictions are as follows:

    a) 41 : 30 : 19 : 10

    b) 42 : 29 : 20 : 9

    And depending on the Iraq inquiry it could be even worse for Labour!!

  46. Tweet from Vincent Moss:

    “New blow for David Cameron and his policies for the family, a Sunday Mirror/ComRes poll reveals tonight. Big drop in Tory lead too.”

  47. Comres

    Tories 38 -4
    Lib dem 19 UC
    Labour 29 UC

  48. ICM News of the World Marginals survey shows 8.5% swing. I am not sure how a marginal was defined. According to this, Tories will have a 38 majority. If repeated against Liberal Democrats, apparently Tory majority of 88 !

  49. Roland Haines

    I’ve had an interest in politics since I was a teenager and am now 70 and this is the first time anybody has made that mistake.

    For the avoidance of doubt, I should say that I have never been a mamber of a political party and have voted for at least five. If Donald Dewar, who was a great influence on me at school, could not persuade me to join a political party at Glasgow University, then nobody can.

    The cleverest man I know, who cites on his CV ten pages of articles in academic journals in languages which are not his first, explained why I can’t join a political party. Parties expect their members to change their opinions when the leaders change their policies, but I want to do it the other way round. So I could join any party that made me the leader first.

    I have never voted for a political party or for a particular candidate. I always place my cross where it will do the most damage to what Tony Blair called “the forces of conservatism” and in that sense it is now hard to choose which is the worst.

    I have never voted Conservative and could only envisage ever doing so in the SP constituency vote for a relic of the past like Lord James Douglas-Hamilton or Mary Scanlon. The last time I voted Labour was in the 2003 SP election when I was minded to vote for the SNP candidate whom I knew personally and whom I respected, but on the day before the poll I was canvassed by an SNP councillor who foolishly imagined I would approve of his “family values” attack on the sitting Labour MSP.

    I voted for the Green list in 2007 because we need representation from minor parties and it was known that they would vote for AS for FM and otherwise support the SNP.

    Before the first SP election I astonished my Australian niece by saying that whether the first Scottish government was led by Donald Dewar, Jim Wallace, Alex Salmond or Annabel Goldie (then only deputy Con leader) I would be content that it would be a pragmatic government and a great advance on anything we had seen previously.

    Since Jim Wallace and other ex-MP’s have gone the LibDems including my MP have never failed to disappoint.

    Indepndence is not my first preference. I would much prefer the second phase of Donald Dewars vision: a UK parliament reformed on the model of his Home Rule parliament. I am as unsympathetic to the Braveheart and the middle-class cultural nationalists as I am to fundamentalists in other parties, but they are well restrained by a leadership hungry for the opportunity to impress as a government.

    “None of the above” has the support of the overwhelming majority, but the SNP is the least worst.

    They have three outstanding ministers.

    Before he became a minister, I was in correspondence about historically Scottish values with Kenny MacAskill arising out of his contribution to the Green Party’s debate on Dawn Raids. I came to the conclusion that he was a person of exceptional integrity, principle and (notably, since) compassion. I have tended to support the SNP since.

    Fiona Hyslop, the Education minister – now demoted – who ended Dawn Raids has my unqualified admiration for the political achievement of changing the policy of a UK ministry without the press being aware of it at some personal cost.

    Having spent my career in the NHS, I am impressed by the Health Secretary. Labour had two excellent ones, but Nicola Sturgeon is, in my opinion, the best health minister in either system since Barbara Castle.

    In a parliament fit for purpose, with ministers like these, I am ready to vote for independence so that my grandchildren grow up in a country which respects education and which promoted child health from conception, rather than one with a failed constitution in which agents of the state in body armour break down doors and take children from their beds to take them to a jail before deporting them to a country where they do not speak the language.

1 2