The tables for Ipsos MORI’s poll are now up on their website, so we can dig about and see what’s actually happened.

Regular readers will know that the big difference between MORI and other pollsters is that MORI do not politically weight their sample. All the pollsters including MORI weight their samples by known demographic figures like age, gender, social class and region. All except MORI also use political weighting, normally weighting by how people claim they voted at the last election.

There’s a much longer explanation of why most pollsters do this and MORI do not here, but the very short version is that people aren’t very good at recalling their 2005 vote. ICM, Populus and ComRes take the view that past recall is pretty stable over time and can be estimated well enough to weight by, MORI take the view that it’s too unstable and should not be used for weighting. The result is that MORI’s samples run the risk of varying politically from month to month more than those of other companies (though MORI would claim the opposite – that other companies risk weighting out genuine public volatility). In a case like this, it raises the question of to what extent the shift is down to people changing their minds, and what extent it is just a more Labour sample.

In MORI’s poll last month which showed a 17 point Tory lead, amongst those who voted in 2005 32% said they voted Conservative, 43% Labour and 16% Liberal Democrat. In this month’s poll which shows a 6 point Tory lead the figures of recalled 2005 vote break down as Conservative 29%, Labour 46% and 16% Liberal Democrat – so a 6 point change in the recalled lead from 2005. (For reference, ICM weighted their sample so recalled 2005 vote was Conservative 33%, Labour 38% and Lib Dem 22% – even they don’t weight to the actual figures because of false recall).

So part of the reason for the shift in MORI’s voting intentions since last month is that their sample appears to have had significantly fewer people who voted Tory in 2005 and significantly more people who voted Labour in 2005 (again, from MORI’s point of view at least some of that would be changes in the level of false recall).

However, this does not explain the whole difference. If you look at the rest of the survey’s innards, there is real movement in Labour’s favour too. Likelihood of Labour supporters voting has increased, that of Conservatives deceased – but in both cases the change is too small to be significant. More interestingly there was an increase in the likelihood of people who voted Labour in 2005 to vote Labour now – in October 65% of former Labour voters said they would back the party this time, this month 72% of Labour’s 2005 voters said they would back the party.

In conclusion, while a lot of the massive shift in voting intentions in this poll appeares to be down to sample variation, there’s a modest firming of the Labour vote too. We should still wait for some more polls to see whether that itself is real, and whether other companies also pick up a firming of the Labour vote. I wouldn’t, however, expect any company using political weighting to show quite such a narrowing of the lead.


109 Responses to “More on Ipsos MORI’s 6 point lead”

1 2 3
  1. John B Dick

    “How am I doing in promoting independence amongst English Tories?”

    Not sure if I’m included in that, but nearly everyone I know in England, of whatever party fully supports Scottish independence. Indeed I think we should have a UK-wide referendum on the subject.

    I think Labour would be more anxious to keep Scotland in the Union because of the number of seats they hold.

  2. Pete B

    Has there been any polling on this in England? I haven’t seen any, but it would certainly be interesting.

    Generally, I’m with Peter on nations voting positively FOR their own independence as opposed to AGAINST the other nations in the current UK Union

  3. You are right, Adam
    The announcement by David Cameron that there will not be a referendum on the Lisbon treaty can easily explain the recent decline of the Tory support in that poll.
    He still has a few months to change his mind and get the landslide he will deserve.
    Rudy

  4. I’ve never seem any polling on it either, but I was just being mischievous.

    It would be unfortunate if the rest of the UK opted to cut Scotland adrift when the majority of Scots didn’t actually want to go. (Quite funny though)

  5. Pete B

    It would certainly be a fascinating constitutional issue! :-)

    Technically, I suspect one would have to go back to the pre 1707 partners in the Union when England & Wales were one of the partners that negotiated the 1707 Union and Scotland the other. Under the UN Charter, the population of either country would have the right to vote for their own independence by repealing the Act of Union that their respective Parliaments signed.

    Doubtless the Welsh would have the right to secede from England since they were only subject to the English Crown by right of conquest.

    I don’t know of a case where the conquering power has ever voted to kick out the conquered – so I don’t know how England could rid themselves of the Welsh, if they didn’t want to go.

    How would one deal with NI, where their Union with the combined E&W and Scotland was the result of an Act of the Irish Parliament?

    (I can be mischevious too!)

  6. If Cameron wins then we have to consider the effects a vast popluation shift as people move to Scotland and support total independance from England

    One poll proves nothing though

  7. I think this poll is very good news for the Tories. No one in the Labour Party will be able to get rid of Brown now as they see this as a hope.

    Right wing tories, however will hold their collective nose and vote for Cameron as they would suffer from a Tory/Liberal pact which would be more pro europe.

  8. @WARREN
    Are you seriously saying that despite the appaling debt, the shambles that is Afghanistan, the lies deceit and selfserving, over the last 12 years, a visit to Cockermouth makes it all ok ?
    Have the British people really come to that ?

  9. @SALLYC
    Spot on Sally in every particular.

  10. Nick Clegg has allegedly said he will support the Tories in the event they are the biggest party in a hung parliament so we only need to start worrying if the WMA starts to suggest Labour as the biggest party.

  11. @ JOHN B DICK
    You are doing a grand job of confirming my support for an independent England.

    I have always wished the Scottish Nationalists every success they wish themselves.

  12. What I don’t understand is why MORI didn’t release their result until yesterday – it is well over a week old and doesn’t relate to two polls conducted at the same time.
    Also, it has dominated the news and the CBI conference is this morning?????

  13. @ALEC
    Whilst it is alleged that public confidence in the economy is riseing, the reality is with every day that goes by, the debt becomes less and less manageable. It is all very well Carl and Tracy of Luton not yet feeling the pinch to much and feeling sorry for Mr Brown because the Sun was horrible, but the facts of economic life continue.
    Cameron has every business to make contingency plans, just in case the Tories do scrape home.

  14. Re Scots Independence.

    Do the English care? Is there any polling data?

    That John an Edward are from Dublin, not Scotland, do we see the difference?

    Who was the most foreign? Johnny Giles or Billy Bremner.

    Who’s the most British on a morning? Nicky Campbell or Terry Wogan?

    I don’t know whether it is time, or travel, or the EU, or football or whatever, but Scots independence might well have happened, as far as we’re concerened. When it does happen, we won’t notice, our lives will be unaffected.

    Presently, the only differencebto me between an independent ROI and a “subjugated” Scotland is as follows:

    Bring your Irish Euros into my shop and I will accept them. I do not accept scottish notes at all. Go figure!

  15. Is there a Scottish Breakdown available

    The 15% “others ” is quite high?

  16. OldNat – there are occassionally polls of England or Great Britain as a whole asking about Scottish independence. Here’s one from 2007 (the nationality cross break there is actually which nation people were from, not their nationality!)

    Irere – The Observer publishes on Sunday. Ipsos-MORI do their political monitor over a weekend – therefore by necessity there has to be a week’s gap for MORI’s political monitor to be published in the Observer.

    King Harold – no “alleged” about it. At the very least YouGov, MORI, NOP and TNS all do trackers of economic optimism and they all show it recovering.

    Paul – no Scottish break I’m afraid. If there was, the sample size would be so absurdly small I would have urged you to ignore it anyway! The other breakdown is SNP/PC 3%, Green 3%, UKIP 3%, BNP 2%, “Other others” 4%.

  17. Hi Anthony, the link you provide is dead.

  18. @ Anthony Wells
    I stand corrected.

  19. @King Harold – I don’t necessarily disagree with you, but the question I raised was whether the political presentation of an emergency budget is helpful in the task of gaining votes. It’s slightly unusual for me, in that I may be in the position of criticising Cameron for publically stating a reasonably sensible position to take in financial planning terms that could look bad presentationally. Normally I am highly critical of Cameron for the reverse – for being obsessed with presentation at the expense of honesty (eg, being trailed by a private photographer in a garden of remembrance…)

    To be honest, I don’t know the answer to original question I posed – I was merely interested in what other people might think.

  20. Shopkeeper –
    Whoops! Missed the f off the end of pdf. Should be working now.

  21. @ALEC
    Yes, as you say the most common complaint about Dave is all fluff no substance. This at least is him doing somthing. The message I get is “the economy is a huge mess, it needs planning to tackle the debt now”. I imagine with a firming up of the Labour vote, the Tories will be attacking mounting debt harder. The BBC & ITV have been telling people how bad it all is, but have that many people really felt the pinch yet? Unemployment seems to be the obvious factor which has reduced an individuals or family’s living standards.
    Interestingly a majority support the idea of reducing debt.

  22. Alec – what I noticed was a hint that Cameron recognises that savage cuts + straightkindofguytalk = fewer votes. He’s leaving the bad news rhetoric behind, and his emergency budget is portrayed by him as a budget for growth, rather than simply one for cuts.

    I would have thought that a new Govt would have a new budget pretty quickly anyway, Unless they had committed to sticking to the plans of the old Govt.

  23. @ALEC & JOHN TT
    The issue of telling voters the truth (and starting a new tradition), and not getting into Blair soundbite mode is interesting. Cameron does need to push the point that borrowing at the present level is not only pushing our luck but the luck of our grandchildren. Keyenes lovers will disagree, however take my point we are talking about Tories. If the people would prefer Labours “sweet little lies” because its nicer and easier to live with (for now), Cameron has a real problem. Growth may pick up but will the debt go down, and if not what will be the effect. This has to be Camerons pitch and the truth will hurt.

  24. Cameron clearly doesn’t agree with that. He’s leaving behind the “age of austerity” stuff, and probably listening to the IMF and other sensible observers who are urging caustion when it comes to quick and painful policy errors.

    It was enlightening to hear him talk recently (or did I read him somewhere) of his fear of messing things up.

    The “grandchildren will pay” line only works if the grandchildren haven’t grown up and paid in the past, and of course they have. Long term debt is preferable to short term agony.

    So it’s not people preferring lies that Cameron is worried about, it’s people preferring common sense as approved by the IMF.

  25. @John tt
    Well if thats the case there is no difference between Tory and Labour.

  26. Conservative Home’s “Centre Right” has Stephan Shakespeare from YouGov talking about the details of the poll and the criteria used for arriving at the final figures . One clear conclusion he makes is that MORI’s methodology has volatility bui;t into it.

    The article is worth the read.

  27. Alec,

    If you recall there was much discussion here during the conference season as to what effect telling it straight on the economy would have. In the main, most peole do not fully grasp the inter-connection between tax, spending and debt, but I think we can safely assume that the public prefer palliatives to the bitter truth.

    In that context it is interesting that Cameron has the courage to say things which might not be immediately attractive to the electorate.

    To me, that suggests that he may display a firmer backbone as PM than some of his predecessors (from both sides of the house).

    Whetehr that can translate into votes in the spotlight of an election campaign remains to be seen, but if teh Government’s plans are scrutinised as thoroughly as Cameron’s, then it should prove illuminating and instructive.

  28. Which is why it’ll come down to who looks nicer.

    It’s only after the GE that real differences will emerge. Thatcher didn’t get in in 1979 based on her policies on wholesale privatisation, closing down mines and doubling VAT to make income tax cuts, because those policies didn’t exist before she got in.

    If Cameron sticks to his guns now, he’ll have a small majority and lots and lots of marginals to defend in 2015. The trick should be to say the cuts will be balanced by growth for now, and then blame Brown when the time comes to wipe out large portions of the public sector.

  29. Sorry, I was addressing King harold when I wrote “which is why..”

  30. I think Adam got it spot on. It is the Cameron retreat on Europe amongst my Tory friends which has demotivated them. They are all non-plussed that he didn’t opt for a referendum to endorse the proposed Sovereinty Act, which if endorsed by the public in a referendum would defacto make it binding on any future administration facing a treaty which gave more powers to Europe. Sun Newpaper nastiness to Brown over the Janes letter backfired badly too. It might be a rogue poll, but I suspect it is directionally correct if somewhat too much in Labour’s direction. We shall have to wait and see confirmation from others.

  31. Please can everyone remember that “UKPollingReport is a site for non-partisan discussion of POLLS.”

    There are sites a-plenty for having political discussions. By having these here we are distracted from examining professionally what the polls ac tually represent.

    As Stephan Shakespeare writes “That this particular poll should produce some outlandish figures is even less of a surprise. When pollsters talk about a margin of error of plus or minus 3% it is based on a sample of 1000 people. However, the voting intention figures in this poll weren’t based on 1000 people. Firstly only 799 people gave a voting intention. On top of that MORI excluded all of those who were not 10/10 certain they would vote, leaving only 449 people – the margin of error on that is plus or minus 4.6%.”

    Mr Wells – Could you please read a polite riot act?

  32. Note to all: there is no such thing as a pro-Labour or pro-Tory or pro-Anyone poll. There are polls which show good results for some parties and there are polls that show bad results for some parties. It is not possible, following scientific polling methods, for a poll to be “biased”.

    This is one difference between polls and the anecdotal evidence some people bring up on here. I would think that my generation was entirely made up of small l-liberals (liberal in the sense of liking liberty rather than in the sense of liking a large oppressive socially-interventionist state) if it wasn’t from my exposure to actual hard evidence of their opinions in the form of polls and student democracy.

    John T. T. : I think Cameron will opt to be as explicit as is politically possible come June. The more he puts into his manifesto, the more he castrates the Labour controlled House of Lords.

  33. @John tt
    Well I like the look of your first sentence. We might as well break out the Moet now.

  34. Just read on PH that 50% of ‘insiders’ expect other polls to show the Tory lead reduced to single figures. One would assume that the ‘insiders’ are not naive enough to rush to judgement on the evidence of a single poll.

    @christina speight, this is the same polling company, using the same methodology, which last month produced a Tory lead of 17%.

  35. Bill “as explicit as is politically possible”

    Well don’t they all operate on that basis?

    KH – Nyetimber, surely? :)

    Christina – Are you related to THE ORACLE – a pro among pros?

  36. testing testing

  37. @ Christina – maybe you should relax a little more. After the first 50 responses or so there isn’t much left to say about an individual poll, and AW’s leeway in letting us ramble on harmlessly here in the dark recesses of ageing threads at least keeps us off the streets.

    To the other responses to my query – thanks, it’s interesting. The point that growth is being heavily trailed as a key element of the debt solution today by non other than the CBI is interesting – in general tone it rather backs up Labour’s approach, and there is no question that most economists see more risks in rapid withdrawal of government support as opposed to over shooting on the debt. In truth, there isn’t much between the two parties on this really, which is one reason why I thought Cameron had goofed with the talk of cuts. I also didn’t see Cameron’s conference approach as honest straight talking – they identified a tiny fraction of the cuts they say are needed, and even these didn’t stack up financially. If you are going to claim the honesty mantle you really need to go the distance with it. Unfortunately he has now raised the prospect of deep cuts in many voters minds and potentially given Labour lots of weapons to weild on the doorstep (he’ll cut this,that and the other etc – it doesn’t have to be true to get people worried).

    Please don’t think I am just being critical of Cameron for the sake of it though. I’m interested in what messages will swing the polls, and I feel that Brown’s growth arguement is more palitable and tactically astute, although I’m not necessarily claiming it is either correct or honest. It’s just a much more wobbly target to hit where you end up arguing about how much the economy will grow in the future rather than pointing to real jobs and services that might be cut now. I think a great deal of how this debate will pan out will come down the Labour’s credibility. If there is any sense of recovery that would help in this respect, and make Cameron’s task of attacking a reliance on some nebulous future growth even harder.

  38. @Alec – There’s plenty more to say about the poll if you take the trouble to read the figures. Surely there’s somewhere else to argue political points without hijacking this excellent and vital technical blog.

    @John TT – not related to any other blogger anywhere – I always use my own name except for a time i ADDED Cassandra to the real name when predicting economic catastrophe. . Only relations upwards were my Mum and (I hope ) my Dad.

  39. “By having these here we are distracted from examining professionally what the polls ac tually represent.”

    Well I for one am happy to be distracted from time to time, and grateful that Anthony allows it.

    What the polls actually represent is the problem-and frankly the comments by professionals doesn’t make it any easier. Whilst the minutae of polling methodology may be of intense academic interest to some-I just want to know if I can believe a given poll-yes or no.

    This poll is symptomatic of the problem for a layman like me.

    Is it :-
    A-Flawed because of a methodology which builds in excessive volatility/flawed because of a low sample with the resultant high MoE/ released late by a politically motivated publisher because it conflicted with three other polls which converged with each other but clashed with this one…..ie useless & uninformative.

    or

    B-Accurately highlighting a recent trend which rival polls have failed to pick up.

    You tell me Christina-which should I believe?

    I,ve read the professionals views on it-Stephan Shakespeare & Bob Worcester & they disagree.

    They both run polling companies-rival polling companies , so where does that get me?

    The sooner we have the only opinion poll which matters the better.

  40. If this poll has indeed only considered 449 people how can it be in any way representative?

    There were approx 27 million votes in the 2005 GE. I don’t believe that a poll with only 449 people would be accurate to within 4.6%

    – my guess would be closer to 20-30% although I’m not a statistician!

  41. @Andrew Myers

    Have you actually looked at the IMORI data tables? The poll was of 1006 people. The headline figure was that of those certain to vote, 449. But if you look at the total of those who say they will vote, 799(weighted) 34% would vote Tory, 34% would vote Labour. I think I am right in saying that IMORI are the only polling organization that uses such a strict filter, as certainty to vote, to produce their headline figure.

  42. I also enjoy reading the different opinions here. It’s interesting to know why people think the polls move in one direction or another. The policies of Brown and Cameron will be scrutinised even more as the GE approaches. We can’t divorce them from the polls.

    Anthony may I ask does the Ipsos Mori Poll carry any less weight than the others?

  43. Andrew Myers – well, all margin of error figures are a bit of a polite fiction. If one actually had a genuinely random sample of 449 people who would actually vote out of an electorate of 27 million, then the margin of error would be 4.6%.

    In reality, samples are not random. Most don’t even pretend to be, they are described as “quasi-random” and often include some degree of quota sampling. The majority of people called do not complete the survey, which itself prevents a genuinely random sample. Weighting and stratifying also both affect the actual margin of error. For that reason, treat it as only a rough guide!

    What is worth pointing out is that the size of the population the poll is measuring doesn’t really matter once it gets above 40,000 or so. Assuming the samples were equally represenative a poll of 1000 people out of 100 million would be just as accurate as a poll of 1000 people out of 100,000.

    The metaphor always used is George Gallup’s one about soup. If he wants to know whether a bowl of soup is too salty or not he wouldn’t need to eat the whole lot, providing it was sufficently stirred a single spoonful would suffice…. and it would make no difference whether he was taking that spoonful from a bowl of soup, a cauldron of soup, a bathtub of soup, an ocean of soup, etc, etc.

  44. CLAD – MORI use a far stricter filter on likelihood to vote than any other company. There’s a detailled explanation here.

    AL J – that is entirely a matter for you! In my average of the polls it does, because they do not use political weighting. However, that’s just my opinion and others can and do differ in their opinion of the right way to do things. In terms of reputation, in the Westminster village some pollsters do tend to be taken more seriously than others, but I wouldn’t like to name names (not least because I’m sure it depends who you ask!).

    In my main commentary, I always try to explain the difference between different companies and why they might be showing different figures. People can make their own decisions on which they believe.

  45. I’m with Colin and particularly Al J on this one. Polls are, by definition, historical and backward looking measurements. Christina and others who wish to restrict discussion to a strict technical examination of what is, in effect, out of date data, are I feel missing the point of what polls are really about. The scientific method involves accurate measurement, analysis and ultimately prediction. With polls, this inevitably involves judgements and necessarily disagreements. While we are all interested in what the polls said about yesterday, and how technically accurate or otherwise they might be, I’m much more interested about what they have to say about tomorrow. There is no point trying to restrict debate purely to the poll results (except of course at an election time). Its more about the why than the what.

  46. @ Anthony

    Thanks for the explaination – my question I guess would be in the context of such a small opinion poll sample how could one be sure that the “soup” is sufficiently “stirred”?

  47. If anyone has any Euros or Scots notes lying around that they do not want then please send to our CLP office. Many thanks

  48. Anthony thanks for the reply. I put my trust in your judgement and remember Bob Worcester from when I was a kid.

    I would love to know what the Westminster Village think of the various Pollsters but guess it must be a secret;-)

  49. “So Labour won an election in Glasgow. Is the Pope still a Catholic?”

    They increased their vote from even 1997. And it probably points to the fact that the protest voting in Scotland (and probably in many heartlands is probably over).

    Cameron needs to win lots of “glasgow easts” if they are to even win this election.

    That’s why it matters.

  50. “There’s an interesting article in the Guardian today suggesting that the people who would lose out from a hung parliament the most would be, of all people, the Lib Dems. They would be stuck with the horrors of (a) an alliance with the Tories that could tear the party apart over PR, (b) an alliance with Labour that could destroy the party’s national support, (c) a hung parliament that would let Cameron “do a Salmond” and minority-government Labour and the Liberals to death.

    There’s no wat the LD will be in parliament wit the tories.

    Their principles, manifestos and policies are polls apart.

    They are much more likely to get their way with the Labour party.

    Clegg will tell the media he won’t side with Labour, as they won’t want to be tarred with their brush.

    But I assure you, they will when it counts

1 2 3