There is a new YouGov poll in tomorrow’s Sun. Topline figures, with changes from YouGov’s previous poll are CON 42%(nc), LDEN 28%(+2), LDEM 17%(-1). YouGov’s last poll was done just one day before this, with their fieldwork periods overlapping, so it’s almost certain that the differences between the two polls are just random sample error.

The Sun, however, has got very excited about the poll, saying “it is the most cruicial opinion poll for five years” as “every August poll in the summer before a spring general election has predicted the real result accurately to within one percent since 1996.”

This is, to put it kindly, not entirely accurate. To start with, YouGov’s poll in August 2004 were not within 1% of the 2005 election result, it showed Labour and the Conservatives equal on 34%. Populus too showed Labour and the Conservatives equal in their August 2004 poll. Two August 2000 polls from MORI showed Labour on 51%, evidently not within 1 point of the 42% they got at the 2001 general election.

The claim stems from a couple of weeks back when Nick Sparrow wrote to Mike Smithson in response to him querying whether August polls are a bit dodgy, pointing out that in fact ICM’s August polls prior to elections have a record of being very close to the actual result.

If you look at ICM’s past polls their August polls before elections do indeed come very close to the subsequent election result. However, it is only ICM’s polls where it works, not other companies. Neither is it the case that ICM’s polls don’t change between August and the next year’s election, in fact they then diverge away from it again.

It’s just possible that this means something, perhaps August polls show underlying support at a time when the news agenda is very quiet. However, if a lack of political news did allow polls in August to see through to underlying levels of support and predict the next election, we would expect to see the same sort of thing with other pollsters, and we don’t. With only three data points (1996, 2000 and 2004), it’s just as likely that it is purely a happy co-incidence. Voting intention polls tell you how people would vote in a hypothetical general election tomorrow, they can’t show you how people would vote in a year’s time.


43 Responses to “Do August polls predict the next election?”

  1. An AVERAGE polling for August may well be a FAIRLY good predictor. But more importantly is understanding the story line and foreseeing how it will come to its conclusion. Admittedly, this is not an exact science but quite fun.

    I’m still looking forward to the polls carried out in September and post the fuel duty rise.

  2. Labour’s average is obviously being held down at 26% by their 24% showing in the last ComRes poll to be published on August 23rd. They do seem to be slowly moving upwards towards that 30% mark which really is the bare minimum for them to be on when the 6 months to election day period begins.

  3. Andy that’s a bit finessed. The 24% was hardly an outlier, and even if you exclude it the recent average barely rises. All their results from the past month are within MOE from each other. There is a very small upward trend overall but it could be a complete phantom.

  4. I have been keeping track of the polls to run my monthly forecasts.

    It is indeed true to say that Labour’s polling has improved. At the end of July it was 25.0% (average of 6 polls). In August it had risen to 25.8%.

    However in the same period the Tory rating had increased from 39% to 42% so the gap has increased by 2 points.

    Roll on the September and October polls.

  5. WMA remains unchanged. From the low point at the end of May (37:22:19) the 2 main parties have each got back some of their natural supporters.

    I still think we have a fight between:
    a) Recovering Economic Optimism – boosting Labour
    b) Labour’s collapse of morale and political force.

    It surprises me that (b) isn’t much stronger. But now that Brown is back in the saddle we’ll see what happens.

  6. @NBeale – almost by definition b) will slow (or stop) at some point as you get down to the more hard core residual Labour support. As it stands, I can’t really see Labour going any lower than they have been, and on balance they are likely to rise to a degree. One thing to bear in mind is that although economic optimism has risen sharply, it’s still negative, so more people are still nervous about the future than optimistic. If this goes into the positive that could be a trigger for something of a Labour upturn, but as many have said, it’s doubtful Brown will really capitalise to any great degree.

  7. Presumably, by next May the plans of both main parties will be clear. The diffference between the two might well be be the starkest for many elections.

    For that reason, I can’t see the polls in August presaging the exact result next May. If there is a similarity, it will be a bigger co-incidence than in previous ICM polls.

  8. I agree with JOHN TT-and would throw in the Party Conferences as potentialy significant factors in future polling outcomes.

    We also have the very very interesting Sky initiative on TV debates-prompting much muttering from BBC & ITV.& silence as yet from No.10.
    Assuming they go ahead, they will have an impact as yet unknown.

  9. Following on from that Colin if one of the leaders does a no show they will be accused of cowardice by just about every paper on the other hand if they show they could be soundly beaten question is do yougo down in battle or not face the battle at all which would have the worst effect on the polls?

  10. @NBEALE

    Why should recovering economic optimism help Labour? It didn’t help Major in 1997 and, indeed, the last poll published by the Observer showed a negative correlation between Labour support and economic optimism, as was seen in 1997.

  11. ‘COLIN
    I agree with JOHN TT-and would throw in the Party Conferences as potentialy significant factors in future polling outcomes’

    I disagree; party conferences meana party dominates for one week, then the next one rolls along. Overall I’d argue Josephine Public switches off.

  12. And re economic confidence–it’s worth remembering that John Howard (the Tory Australian PM) was thrown out at the height of the economic boom–he could not resist a ‘time for a change’ momentum. The point I am making is that voters do not just vote on economic issues.

  13. Colin’s comment about live TV debates is an interesting one.

    I can’t see Brown coming out the winner here – he is ardantly against the idea, presumably because he is ill at ease compared to the other leaders in front of the camera and knows he will get some very difficult questions.

    However with both Clegg and Cameron having already said yes, Meddlesome Mandelson publicly volunteering Brown for the debate and the TV channels saying that if anyone refuses to turn up they will make that obvious with an empty seat Brown has no choice – oh dear Gordon!

  14. @JACK:-

    “I disagree; party conferences meana party dominates for one week, then the next one rolls along. Overall I’d argue Josephine Public switches off.”

    mmmm-you sure about that Jack?

    Cast your mind back to the Party Conferences of 2007-look at Anthony’s post 2005 Polling Graph.

  15. Richard,
    I agree with you and most others that Economic recovery in itself will not help Labour and that time for a change is still the strongest message around.
    As Alec has stated elsewhere, however, there is a chance (only a chance) that Cameron will be seen to have been over pessimistic and unnecessarily severe in his prognosis for the Economy.
    Should the recovery be stronger and quicker than many predicted ; and therefore, the numbers better (less bad) than anticipated GB will say look how I steered us through the Recession, the Toires got it wrong would have been worse under them etc.
    May be worth a few points – enough to produce hung parliament.
    In the short term in a bizarre way Camerons hard line over Magrahi demonstrtaes his opportinism and his lack of understanding of being in office, real politick etc. He is still sound-biting with the priviledge of opposition, does anyone think he would have done any different. I think this will play as a subliminal negative.
    Blair had faults but he did come across as a PM in waiting whilst Cammo doesn’t (but then again neither did Thatch).
    Hence a 78/79 feel to this GE with similar majority but with more LD’s.

  16. On the TV debate, these things often don’t go the way people expect. As Jim Jam hints at above, a good many people see Cameron as too smug and cocky, and rather than a statesman in waiting, he’s still in the school debating society. I know many people who ridicule his PMQ performances and are waiting for him to blame Brown for the weather. A TV debate may not be the most appropriate way for Cameron to show his assetts at their best, and as the clear favourite, with very high expectations, he is the only one with anything to risk in a debate.

  17. It would be interesting to see some poll data from 1978-79 to see where the parties were ahead of the 79 election.

    Last night I watched some footage from the 79 election night on You Tube and the BBC exit poll showed a range of seats for each party (rather than a prediction of a majority). Their prediction showed a range of 309-339 seats for the Tories and 269-298 for Labour. The result was actually 339/269 so right at the edge of the prediction.

    Apparently the polls almost closed to neck and neck in the election run up.

  18. Excellent analysis about the validity of August polls, Anthony. Thanks.

    I agree with Colin about the importance of the Sky initiative about TV debate. It could have a large effect switching votes during an election, with consequent implications for interpreting pre-election polls.

    Whilst in principle I approve of the idea of TV debates, there are a number of practical concerns:-
    (1) There are implications for democracy of leaders holding debates on a commercial TV station where even if the actual programme is free vieweres have to pay a substantial subscription for the satellite dish. From a psephological point of view, we would have to consider how many people, from which demographic groups, had access to see the debate in particular constituencies. I think any Prime Minister should hesitate before accepting an invitation to debate before a TV audience selected by cost, albeit a Murdoch invitation may be hard to refuse.
    (2) Scotland (and to a lesser extent Wales) is a problem in relation to TV debates. On the one hand, the level of SNP support is such that you could hardly exclude an SNP representative (who? Alex Salmond or their Westminster leader); but if you gave the SNP reasonable time English viewers would be turned off. And you can hardly have separate debates for Scotland in a UK election. In any case, you don’t know at what moment in which debate the crucial point will occur.
    (3) There is also a problem with minor parties UK wide, given that Other support is currently running at about 10%. The Greens and UKIP would justifyably expect fair treatment. But what do you then do about the BNP, who also have MEPs now but with whom the other leaders would (most of us think rightly) refuse to debate?
    One reason Other support is growing is already that the three major parties often seem to have similar policies conflicting with popular opinion (UKIP would instance Europe, of course). If the debate was confined, by Sky invitation, to Brown, Cameron and Clegg they could look even more than ever like three Westminster insiders debating in a bubble from which they can keep out the main issues.
    (4) These days there are effectively two different polling days for two different groups of electors, the postal and the personal voters. The timing of any debate would have to be carefully agreed not only to avoid an immediately pre-election debate but also to avoid one immediately on top of the issue of postal votes. Voters should reflect on their choices, not just bung in a ballot immediately after a TV programme, which would tend to increase the potential swing depending on the leaders’ performances, Indeed, if there was a TV debate it would further strengthen my fervently held view, posted several times on this site, that the use of postal votes should be minimised.

    The UK is not the US, and Murdoch would be unwise to assume that political institutions can be transferred easily. In addition, Murdoch is not Berlusconi of Italy and as I hope he doesn’t want to be I hope he will take heed of some of the problems arising with political coverage in the Italian media.

    It would be important who chaired the debate. Given current problems in attracting younger voters to be intersted in politics, the party leaders might want to think twice before accepting a veteran political commentator as the question master, however able and distinguished. The person leading the debate could have an impact on voting behaviour not just in terms of politcal balance (Sky would obviously make every effort to be fair on that) but also in terms of which age groups were encouraged to the polls.

    Given my comments above, particular issues about TV debates that psephologists might want to consider include the implications for regional behaviour in voting behaviour and possible effects on the size of the Other vote, which might vary for different minor parties.

  19. @ Alec & Jim Jam

    You echo the Labour party’s propaganda line of “economic optimism and recovery” However, whenever we see independent sources of data (such as the OECD report out today) we find a much more sober (and in my opinion realistic) appraisal that the UK will be the last of the major economies out of recession.

    I suggest that one of the reasons why the public do not respond in the way that you appear to hope is that Brown’s strong-arming of the ONS, and misuse of statistics, is now such public knowledge that no-one believes a single number quoted by the current incumbent of No 10.

    If the OECD predictions have any validity, the public are far more likely to punish Labour for “wasting our money and not achieving anything” than to reward then for non-existent “good mamagement”.

  20. @ Frederick Stansfield

    Nicely argued piece, with a number of valid points. However, given the choice between “No debate, because a perfect system cannot be devised” and “A flawed debate, but one where the contenders are forced to answer/evade questions in front of a mass audience”, I would go for the second.

    So, from that viewpoint, I say thank you, Sky. The threat, and embarassment, of “the empty chair” should force the contenders into the studio.

  21. Superb post Frederick.
    I assumed that even though a UK Election, Scotland and Wales would have their own debate with the local leaders representing, e.g. Ian Gray, Rhodry for Labour.
    The subsciption TV thing I had not considered and I guess it has to be free to air.
    Although if on Sky News a digibox gets you that and this is nearly evrey one now. They could repeat on BBC the day after I suppose as well.

  22. Thank you Frederick.
    You have made me think-not sure about it now-unless it is on BBC & ITV as well.

  23. Cynosarges,
    My line is consistent that I expect a Conservative Majority and that Cameron is an effective communicator, appealing to swing voters.
    I say above ‘only a chance’ whn talking about Economic recovery but it is one worth comtemplating.
    I do find myself trying to compensate for the overwhelming pro-Tory line (much of it blinkered) trotted out on this site.
    Like Alec I was never convinced by Blair and see many of the same weaknesses in Cameron.
    It may be that in office he will be better than he appears now, I hope so.
    He may be more right wing but Hague is far more impressive.

  24. Regarding the possibility of a boost for Labour with economic recovery, isn’t it just as likely that economic recovery will make it feel less risky to change government and so favour the Tories? The “steady hand on the tiller”/”no time for novices” argument loses potency as things improve.

  25. I find the fuel duty rise somewhat comforting. I mean….one can be absolutely certain that they’re doing it because they think it’s what’s best, and not playing politics…

  26. @Cynosarges – I have said on another post that any upturn could well help the Tories, and I have made it lear that I have grave doubts whether Brown has enough credibility left to capitalise on any improvement.

    In terms of the TV debate, it’s always been assumed that it would have to be available through a terrestial TV network. The minor parties are an enormous issue. If the BNP fronts up 650 candidates, why shouldn’t they get a leader into the debate? As Frederick says, we’re not the US, where 2 parties conspire to keep out new entries and the vote is for the individual candidate. I’m against such a debate on principle anyway – I’ve had enough of politicians looking for killer debating blows and smart one liners. Stage managed politics that benefits the established political groupings and doesn’t help us find new ideas.

  27. Anyone know the confidence interval based on the sample size?

  28. Alec,

    Nothing would please me more than the BNP being put on the spot. At the very least, I think every party which already has representation in Westminister should be up there, so George Galloway gets put under the spotlight as well.

  29. Re the debate – or not–

    It’s not a question about the amount of candidates; I am sure the Loony Party could perhaps front more than the British Loony National Party. It’s a question of who could actually get enough votes to be credible as PM (or assistant PM in a coalition Parliament.).

    Having watched too many of these in Australia I think the way I’d kill it is have a minimum of 3 tv debates for the PMs and have say another 3 for the Deputy Leaders and so we are all then bored. Actually I’ll take spin doctor fees for this or can I go and bet that if we have a tV debate it’ll be more than one on the basis that nobody watched

  30. Oh for heaven’s sake – now having read this lot backwards – this debate if it happens is about possible PMs (or deputy PMs in a coalition Govt.). Simple. Much as I agree with devolved Parliaments. It’s got nothing to do with the rights of Leaders of minority parties (such as PC SNP, UkIp or other Loony Parties etc). Get real. It would not be a debate about a platform for parties but solely one about who could lead the Country ( or- as said – deputy).

    I recognise this is a tory dominated site but do try and understand the basis for a modern issue…

  31. The Sky proposal is interesting and the points made are all very valid indeed. However I did read that Brown was planning to make his own proposal at the Party Conference.

    Clearly that is an unknown quantity but with Mandleson operating as the “puppet master” one could envisage him portraying a threesome as two against one and therefore clearly grossly unfair!

    IF he were to propose two separate debates with firstly Clegg and then Cameron he might have more justification in hoping to win the first and thus possibly minimise the risk of losing some city centre seats to the Lib Dems. With that under his belt it might give him more confidence in tackling Cameron. Whether he could win that I don’t know but this scenario might be a better bet than a threesome, and, of course would be on BBC/ITV.

    Incidentally the argument against including the minor and/or nationalist parties is surely that only these three are contesting across the UK (OK excluding NI)

  32. Why can’t we be good Europeans and model the debate on It’s a Knock Out’?. Cammo, Brown and Clegg, in giant puppet suits (just imagine what Brown’s would look like) running up the slippery slope tied to an elasticated rope, while UKIP, Nats and the BNP (or the Belgians, take your pick) try to knock them off with wet sponges. Each side gets a Joker to play, (Redwood/Balls/Opik) with Paxman taking the role of Stuart Hall laughing hysterically.

    Could it be any worse? Well we could be in Japan, where the new First Lady elect has said she’s been taken to Venus by aliens. Even I wouldn’t claim we’ll get that if Cameron wins.

  33. To me the advantage of a debate is not in how the “candidates” do against each other, its in the opportunity to get them to answer a straight question. Unlike on Newsnight or elsewhere, if you prevaricate in a debate then one of the other candidates can make you look a fool by ably answering the question. So its in each of their interests to actually answer to the best of their ability. I don’t think there’s any other forum where this really happens.

  34. I think JACK & NEIL have it.

    The debates will be about testing the ability of candidates for PM of our country, to answer straight questions , under the direction of a skilled Chairman.

    This sort of clos-eup can also expose character flaws which we need to know about. We expect prevarication of course-but as in Paxman vs Howard, that too can be devastatingly destroyed.

    The format will be important-and it must include question of one candidate by another through the chair.

  35. NEW YOUGOV/SNP POLL

    ‘45% of Scots back MacAskill’s decision on Megrahi’

    The YouGov poll of 1556 people found 45% thought Mr MacAskill made the right call to free Abdelbaset Ali Mohmed al Megrahi last month on compassionate grounds. The same percentage said he was wrong.

    It shows a swing in support for the release following a BBC poll last Friday, when only 32% supported it.

    The research, for the SNP on Tuesday and Wednesday, found 49% of white-collar workers were in favour with 43% against. Support among blue-collar workers was 41% while 47% disagreed.

    There was a difference in opinion among age groups. Between the ages of 18 to 34, support levelled out at 39% but among the over-55s, it reached 55%.

    Almost half of those questioned (49%) said Nelson Mandela’s support for Megrahi’s release had enhanced Scotland’s reputation, but a quarter said it had been damaged.

  36. YouGov – Commissioned by the SNP
    Sample 1556 adults, 1st and 2nd September

    1) Do you think the Scottish Justice Secretary Kenny MacAskill was right or wrong to release Abdelbaset Ali Mohmed Al-Megrahi on compassionate grounds?

    Right 45%
    Wrong 45%
    Don’t Know 10%

    2) Regardless of your views on whether it was right or wrong to release Abdelbaset Ali Mohmed Al-Megrahi who do you think this kind of decision should be made by?

    The UK Govt in London 22%
    The Scottish Govt in Edinburgh 60%
    Someone else 11%
    Don’t Know 7%

  37. OK STUART-I think we’ve got the message now!

  38. Beans, I am not up enough on the statistics of individual polls, but I believe the rule of thumb is that you can be confident that the statistics reported from national opinion polls, which usually have a sample size of about 1,000, are accurate to within 3% either way. So, for instance, on the basis of this latest poll the Tories are almost certainly between 39% and 45%.

    These days, you could presumably in theory do a meta-analysis to increase the accuracy of the estimate by combining data from several polls.

    At the moment, we have been getting very consistent reports about the standings of the three major parties from successive polls over a period of months. Even if, like me, you are out in the sticks and don;t have access to specialist statistics software, I think you can be confident for all practical purposes that the figures being reported are accurate to within a pecentage point or two.

  39. Imagine Brown in a live debate he needs notes for pmqs even though he knows the questions beforehand.I think this would be a disaster for labour mabye he could call in sick on the day and let Mandelson deal with it as with all other difficult debates recently.

  40. Jack,

    If the debate is about potential prime ministers, let’s leave the Lib Dems out of it altogether. In terms of leading the next government, they’re just as irrelevant (at this point) as Respect.

    If it’s about potential prime ministers and their coalition deputies, then that becomes more complicated. What if the Tories were just under the cusp of a majority and formed a coalition with Irish parties? What if Labour formed a coalition with Plaid Cymru? And who is to say what is and is not a plausible coalition?

    That’s why I think the best criterion should just be seats in parliament. The Republicans and Democrats have large multi-candidate debates during their primaries, so I don’t think having a lot of candidates is a problem. Of course, if some sort of PR is introduced that creates an Italian-esque situation, a stricter criterion would probably be required.

    I also think it should be about GB, DC, NC etc. in their role as party leaders rather than potential prime ministers, since it is technically the Queen (not the public) who decides who the PM is. On the other hand, a debate between the parliamentary political party leaders prior to a general election makes perfect constitutional sense.

  41. I dont think this will be the next election results, firstly anything could happen between now and then, my prediction for the next GE would be about:

    Con – 42-44
    Lab – 23-25
    LibDem – 20 – 22/23
    Others – 5-10