We won’t get any polling reaction to the budget for a couple of days – even if anyone does carry out an instant poll this afternoon for publication tomorrow, I’m not a huge fan of them as the fieldwork ends up being done after the budget speech has been given, but before respondents have actually taken it in or read the media’s analysis of it. Our first real measure will be YouGov’s monthly poll for the Telegraph on Friday – and that’s assuming it’s not going to be held back a day or three to get a better taste of the budget reaction.

In the meantime I’ve now got confirmation that the Marketing Sciences poll carried out for the Sunday Telegraph was indeed done using the same methodology as ICM, so the two polls should be directly comparable. Looking again at them…

MS (Apr 15th-16th) – CON 43%, LAB 26%, LDEM 21%
ICM(Apr 17th-19th) – CON 40%, LAB 30%, LDEM 19%

In theory it’s possible that there was a 3.5% swing to Labour within just two days, even though the media was still unremittingly horrid for them, but more likely the difference is just down to normal sample error. There’s no reason why the real position couldn’t be somewhere between the two polls.

It is a good reminder that polls are subject to sample error, that two polls done using exactly the same method, at pretty much the same time, can still show different results, and we should never get too excited over movements of 1 or 2 points in a single poll, unless it supports a trend that has been observed over a number of different polls.

64 Responses to “More on that Marketing Sciences Poll”

1 2
  1. If they both use the same sampling method, can we not combine the raw data in order to obtain a better picture?

    A very quick calculation suggests that a 1,000 person sample has an error of +/- 3%, a 2,000 person sample has an error of +/- 2%

  2. Does the fact that the one under the ICM name was done over the weekend, while the other was midweek, have any measurable effect?

  3. I am with Anthony on this one, the most likely answer is sample error.

    I can’t remember the exact details but in the run up to the Holyrood elections Yougov had two polls close together that showed different results and they got pilloried for it. I think Peter Kelner did a comment piece on it.

    Being a politician, and therefore surrounded by cynicism, I might be less charitable that Anthony and suggest that using Marketing Science might have let ICM avoid a similar kicking to the one youGov got in the event to to dissimilar results side by side..


  4. Peter,

    You cynic ! But seriously, presumably the choice of Marketing Science ratehr than ICM badge was determined before they had done the fieldwork and arrived at sharply different results.

    If we did as Mark M suggetss then on a simplistic basis we can take the mean as being:
    – Con 41.5, Lab 28, LD 20 –
    In which case this would have been swept into the archive along with all the other “static” polls over the past couple of months.

    Anthony, is Jono’s point worth exploring with ICM ?

  5. Sample error. There is clearly a Conserative lead of around 10 points to 12 points. And at the moment any other poll is probably an outlier.

    How this might change now that Darling has decided to tax the super-rich 50% will be of interest. The next polls may show a more significant move.

  6. I think it is probably accurate to say that since 28th of March Labour have dropped 3%, and that this support has shifted equally towards the Cons and Lib Dem.

    Over the past 6 months the polls have been up and down like a rollercoaster. But in the end Labour will lose all the momentum gained in October and come down to where they were before. But this time the Lib Dems will be about 4 to 5% higher than before.

  7. This is a timely reminder of this characteristic of opinion polls in general. Since following the polls and their analysis and interpretation here and elsewhere, I have come to a better understanding of the whole scene.

    That’s “better”, not necessarily all that brilliant(!)

  8. Great article and reminds people that polls bounce by 1-2% just because of error.
    One item I would like to remind people and discuss is that if the Tories get 42% of the vote, an increase of 9% over 2005 then this means in England they are likely to have increased by 10% or more because I don`t believe the Tories would increase their share of the vote in Scotland or Wales (15% of the UK electorate) by anywhere near 9%.
    Also it is fair to assume that solid Tory areas like Huntingdon and Mid Beds where the sitting Conservative MP already gets 50+% is unlikely to go to 60%. Therefore marginal/swing seats could easily see a 12-15% Conservative increase. This would easily give the Conservatives a solid majority even if 9% nationally is projected to give them a small majority.

    I believe there was a poll last year of marginals in England which showed the Tories doing better compared to 2005 than the national headline results indicated.

  9. It would be interesting to see a graph over, say ,the last two years showing the various methodologies transposed against each other (for each party %age share), thereby seeing whether peaks and troughs were simultaneous.

  10. Mike makes an interesting point. UNS would really be more accurate if applied to England, Scotland and Wales seperately (some irony in that statement!)

  11. Testing testing

  12. Anthony,

    Is there a reason why the latest IPSOS Mori poll is not included in the table? Is it under dispute ?


  13. At last! A sensible budget targeting the culprits of this recession wgo have wrecked our country and society. Capitalism doesn’t work. A Socialist agenda is the only way to get the country working again. This government is finally waking up and ditching all those wasted years trying to mimic the tories. Thirty years of disastrous policies. Still time to squeeze a 60% tax rate in before the next election. This should prove a winner, then going higher after the election should mean that the city crooks who got us into this mess will be forced to get us out of it instead of buying there new Fararis

  14. Wayne – this is a reason. Forgetfulness!

    I can’t access the database that the tables run off from work, so when polls are released during the day I can’t update the tables till I get home that night. I forgot to do it :)

  15. Hmmm Adam C, would that be the same Socialism that worked so well in Eastern Europe, or that of Pol Pot, or even the Peoples Republic of China that have embrassed capitalism in the best interests of the “people”? Or perhaps you would prefer to use Robert Mugabe as a good socialist example. Capitalism may be imperfect, but Socialism is a nasty ideology generally favoured by the envious and the despotic.

    As this is a site for opinion polls, I will say that the opinion polls, and actual polls, tend to confirm that socialism isn’t generally more appealling than other fanatical doctrinaire pipedreams, which is why socialist parties tend to struggle to to get any more of the vote than their equally evil fascist counterparts, which is indeed a blessing. Long live moderate views!

  16. Adam – you put it in a really inappropriate way for this site – just read up a few posts to get a handle on what it’s about.

    Having said that, you’re suggesting that the two main parties are reverting to “type”, and I see that too. The choices are now much clearer than they were a year or two ago.

    To my mind, the more difference there is between Labour and Conservative, the more chance Labour have of recovering. I’d much rather it was about policies than the ability to come across well on the telly.

    Having said that, I don’t think the bookies are completely unreliable, so Labour’s chances, although in my view enhanced by a bit of populist tax hiking for the rich, are still dependant on a massive change in fortune for the economy.

  17. Nifgel – you’re just as bad as Adam, and should probably know better than to join in in the same terms.

  18. I’m completely unsure how the polls will reflect the 50% tax rises. They’re not getting a good press in general, but only affect 1.5% of the population, so who knows.

    Although I don’t want to join the socialism/capitalism debate, it is interesting to see that the highly capitalist world of international finance was so radically wrong, and imposing MBA style contemporary management techniques on public services has failed to deliver effective value for money. On the one hand, governments have to intervene more, and on the other a lot less. How this fits with political dogma I don’t know. One thing I do believe is that just becasue someone earns a shed load of money we shouldn’t assume they have any idea what they are talking about.

  19. Hmmm, not sure that Adam C is joining in the spirit of non-partisan commentary of this site (I’m also not sure that he has realized that the 1970s are over)

    Back to the subject of polling (ahem!)…

    Andrew Neil makes an interesting point in his blog today, that he is now certain (as am I) that the Sun’s pact with New Labour is now over. I think that this will have a big impact on the voting intentions of a lot of people.

    Anthony, do you have any stats on the influence that certain publications have on the voting intentions of the population? I suspect that The Sun is the most influential

  20. John TT – I agree that the parties are reverting to type but I think a lot of people think public spending has been wasted and bloated. Even the government now accepts that by “finding” GBP15 billion in “efficiency savings”. Remember in 2001 and 2005 they critisiced the Conservatives for “cuts” and now they are doing it.
    The 50% tax rate may or may not be popular but it reduces a GBP175bn deficit to GBP170bn (i.e. raise 5 billion). Where do you think the rest will come from??

  21. Mike – I think a better question is where did it go ? (or why did well-paid pro’s believe it was there at some point?) The sheer scale of it means that the public can’t reasonably believe it’s all gone on one-legged dance troupes.

    How to find £170bn without borrowing it? You could ask the entire public sector to take a 100% pay cut, but in the absence of miraces, chipping away over time is probably better than cutting until nothing is left.

    If “efficiency savings” are about closing tax loop-holes rather than firing people, they should find favour in the electorate. I’m not convinced the polls will give a boost to Labour in hte short term, but IF the VAT cut is seen to have worked to boost growth by this time next year, then things might become interesting

  22. @Andrew Kennedy – The Sun has been Tory in all but name for at least 5 years, being incredibly critical of Labour but not having the guts to completely get behind the Tories until they know for certain what side will win – I think this will continue, esp as the polls are rather fluid at the moment, right up to the election and then back the obvious winner – they can say the ‘Sun wot won it’.

    The power of the Sun is no where what it use it be and their publications are way down what it was it was in 1997 and continues to slump.

    There was figures, which i’m trying to relocate, that showed no real difference in voting intentions of Sun readers following their move to Labour.

  23. John TT – I am not advocating no borrowing in a recession but we are borrowing 12% of national income. The highest in the world – why are we borrwing more than EVERY other country if as the Government says the recession is worse in other countries like Germany. Obviuously they ran their finances better during the past few years. We had a structural deficit even before this beagn – we should have been running a surplus after 15 years of growth.

    The VAT cut was a complete waste of 12 billion. Because at the same time as that cut they increase NI which is essentially an income tax. Also by the Government’s opwn measure the VAT cut has not worked because they said at the time of the cut that it should help the economy recover by late 2009. Now they say the recovery will not be until 2010 at the earliest. Therefore the VAT cut was pointless and hence no improvement in Labour’s poll rating.

    We really need polls that show the key swing seats and regions in the UK. Scotland, Wales and the North of England are not where the election will be won or lost. The Tories are targetting predominately the South, London and the Midlands and are probably doing much better than the national polls show.

  24. David in France,

    I agree on sampling error.

    In my view the Con lead is more likely to be over 12% than between 10%-12%.

    So far this year there have been 30 Polls (see chart). Of these:

    – only three polls show Con lead in single figures, while five show a lead of 16% or more.
    – 10 show Labour below 30% (including 3 at 25/26%), while only 3 show them above 32 %. Just over half show Lab at 30-32%
    – None show Con below 40%, 3 with Con at 40%, and 3 with Con at 45% or more.
    – 80% of Polls show Con in range of 41-44%

    Based on the figures for party share (as opposed to lead) it is fair to assume that Lab are in a range of 28-31% and Con are in a range of 41-44%, giving a lead at mid point of 13%, with a min of 10% and a max of 16%.

    Which happens to be in line with the 13% lead and mean figures of 42/29 shown by Anthony .

  25. @Paul

    Good summary, but it’s made me really depressed – clinging onto polling error!

  26. A £12bn spending tax cut is significant, and will boost retailers takings irrespective of NI rate increases, or any other increases in costs of living.

    Darling is depending on the £12bn VAT cut having a visible effect by the time it has fed through (ie by next jan). If it works, his growth forecast looks more reasonable, and his ratings should improve (if not the voting intentions).

  27. Chris Newey:
    ” they can say the ‘Sun wot won it’ …

    …There was figures, ”

    Too much Sun reading, Chris?

  28. John TT – do you really believe a 12bn tax cut on consumption (so a DVD player is 39.40 instead of 39.99!) will make a noticeable difference in a 1 trillion economy?
    The Governmens own measure was that by this summer it was to have had an impact and the forecast is now for the economy to do much worse than when the VAT cut was launched. Do you not think that 12 billion could have been better spent. Why cut that tax but increase income taxes? Don’t they cut cancel each other out and therefore are wasted. You don`t understand economics if you say “it will boost retailers irrespective of NI increases or other increases in cost of living” – if peoples costs are going up (tax or cost of food etc) then they have less to spend and saving 60p on a DVD player (or whatever) is hardly going to make retail sales increase!

    I hope that the Chancellor’s growth forecasts are correct for the sake of the UK but unbiased experts have predicted worse, the chancellor himself was wrong with his November estimates (only 6 months ago) so how can we trust the current predictions. Also remember that this Government for the past 8 years have consistently been worng on the borrowing figures, and always wrong on the optimistic side. Umm I see a pattern there.

    Just answer some facts – why do we have the largest deficit in the world? Why are we doubling the national debt (which has taken 300 years to accumulate) in just 2-4 years? It is a recession but we should never have been running a deficit of 3% a year for the past few years when we have had 15 years of uniterrupted economic growth. Basic economics states you pay back in the good years and borrow in the bad years.
    Labour crucified the Tories when they vorrowed 50billion (8% of GDP) in 1991/92 – so lets be consistent.

  29. As Anthony says the 3.5% “swing” is quite likely to be polling error. As I repeatedly remind people, the theoretical Std of a 1000 person sample may be 3% but the real Std of a 1000 person poll is rather higher. This is because:

    a. The 3% assumes that the 1000 person sample is an unbiased sample of the whole population.

    b. Filtering reduces the effective sample size.

  30. saving 60p on a DVD player (or whatever) is hardly going to make retail sales increase!

    Yes it is, but your mis-conception is shared by many who don’t understand economics (or any principles involving scale)

    You answer your own questions when you say “you borrow in the bad years”. The reasons for the financial collapse (most keenly felt in huge financial centres like The City) were nothing to do with “wasteful” govt spending on extra money for public services.

    The collapse was due to mistakes by bankers. Easily preventable by better over-sight? The polls reflect that the public aren’t too sure about that one (if they were, then I’d say Labour would be in the teens by now)

    There’s no correlation between Conservative borrowing in 1992 and to-days, and I don’t think Cameron will see that avenue as a vote-winner.

  31. @John TT
    “The collapse was due to mistakes by bankers. Easily preventable by better over-sight?”

    Actually, the collapse was entirely fuelled by governments. It was just left for the bankers to start. It was easily preventable, and not just in hindsight. Too much cheap credit was lent to people who had no way to pay it back. That much was obvious from the house price boom (as a non home owner I looked at house prices and said “no way, I’m waiting until they come down before I buy. I’m not taking on 5x my salary”).

    I think the scale of this crisis has caught most people off guard, but to say there were no warning signs until it was too late is false. After all, what has happened following every other house price boom?

    I think what we are seeing now is what happens when the whole world has an asset price boom (USA, Spain, Ireland, UK etc). We all go up together, and we all come down together.

  32. Mike – just to correct, the NI increase has not come into effect yet, so doesn’t cancel out the VAT cut. £12b is a major boost to the economy, and a recent CEBR report suggested it is working very well in terms of keeping spending up.
    Mark – this hasn’t happened after any boom since the 1930s. Its special.

    In terms of economic contraction we’re in a significantly better situation than many other developed countries. In terms of government debt were in more trouble, partly because of an over reliance on the financial sector but mainly due to many previous years of high public spending. Even here though I would urge people to have some perspective – the main reason we have spent so much was that, frankly, by the 1990’s decades of under funding gave us third world public services and a crumbling infrastructure. The roots of this financial mess go back many decades and responsibility lies not just with both main parties but also to the great British public, who have no patience and want everything for nothing.

  33. I am somewhat puzzled by the growth forecasts. Virtually every commentator, including the sainted Vince Cable is claiming that Darling was wildly over-optimistic in his 2010 and 2011 forecasts and of his claim that the economy will start to grow before the end of 2009/10.

    The logical think for a chancellor to do from a political point of view would be to *understate* the growth targets with a view to taking credit for better-than-forecast outturn. So what is Darling up to? Either:

    1. He really does believe in his forecasts – indeed believes them to be an under-estimate (which puts him at odds with the consensus and highly exposed if he is proved wrong before the election).

    2. He is privy to data that suggests that he is right and the consensus view is wrong (and the Treasury have in the past confounded independent forecasters) in which case Labour could announce good news just days before an election.

    3. He is perfectly aware that the numbers don’t add up but Brown would not allow him to take any tough decisions on spending this side of the election, and he is therefore hoping against hope that things will turn up to make his predictions correct.

    My money’s on 3.

  34. PoliticsHome is reporting that they expect 20+ Tory leads in the next few polls. Apparently their polling suggests people are tending to be ‘less likely to vote Labour’ following the budget. Of course, this could just be Conservatives saying ‘less likely’ and wouldn’t make any difference.

    Alec – this is only special because we’ve all busted at the same time so there are no strong economies around to pull everyone else up i.e. if I break my leg and you’re ok, you can carry me til I heal but if we both break our legs we’re in a really bad situation. This is a preview of what will happen frequently if a truly global economy (the ‘New World Order’ continuously referred to) booms then busts.

  35. John TT and Alec – of course the financial collapse has exacerbated the deficit but we were running a deficit of 3% of GDP in good economic times (2002-2007 with GDP growth in the region of 2.5%). Another indication that it isn`t just the financial collapse or global factors that have caused our deficit is the fact that we in the UK have the leargest deficit ( a mind blowing 12% of GDP) even though our economy is contracting less than others. Please answer this point and explain why the UK is more badly hit than any other county remembering that the US and Switzerland have large financial sectors which were just as badly hit.

    The UK public spending was not so under funded – we spent more than other countries (USA for example).

    The Government, of which I assume you support, has admitted there is wasteful spending – the Gershon report of 2005 identified billions of waste and now they have found 15 billion of extra waste. So if 12 billion in VAT tax cut is substantial then 15 billion is waste is even more substantial.

    I am not a partisan hack and will give credit where it is due. I just hope you are the same and can admit that we overspent in the last few years running up debt when we should have had a surplus.
    Spending alone does not guarnatee quality services – it is how you spend the money.

  36. Leslie – I completely agree with you that it is option 3. I say this because all the measures on public spending come into effect from 2011- conveniently 1 year after the next election.

    Just as the Tories in 1996 set up tough spending plans for 1997-1999 knowing they would lose in 1997. Labour is doing the same now.

    Any talk they make of Tory Cuts (used in 2001 and 2005 election campaigns) is now discredit because of their own cuts. Also Labour liked to compalin of unfunded tax cuts, instead we now have unfunded spending commitments.

  37. John TT talking about people “who dont understand economics.” LOL. Brilliant.

    People just dont understand his and Labour’s fairy tale economics where people can have everything for nothing. Apparently the treasury’s books dont understand their fairy tale economics either given that we always end up bankrupt when Labour get their envious grip on things.

  38. My first post.

    I just clicked on the link to the UKPR Polling Average. Personally I like the idea of a Polling Average even though I am aware of some of the arguments against; e.g. You can’t compare apples with pears.

    It’s not clear from the link when the UKPR Polling Average started, how often it is updated and whether there is a graph of its results over time? Are these available?

    I also clicked on the Comments Policy. great ethos. Many Thanks.

  39. in ref to the economy, the govenment of the day had a warning about the upcoming problems when the internet bubble burst and markets went everywhere but up, and it also affected banks as the money that had been lent to the internet firms that went down during the bubble burst did not and had not paid back all of their lones from banks and building sociertys and investers lost alot of money, this should have been the warning sign to better regulate our markets and banks as it was a mini version of what is going on now but the internet bubble burst only had a slight effect and govenments did not respond when they had the chance and the other warning sign was back in 2006 another mini blip in th markets made by some companeys going to the wall and other doing big mergers with other firms which in turn put others out of buissness for good, and lastly house prices yes dreaded house prices they were going up fast and were not helped by people pushing the price of there own houses up for personal gain that in hindsight was not achivable in any great way, so the govenment is now paying for most of thoes mistakes and is also paying for an over inflated public sector that makes jobs out of thin air or non jobs plus non regulation of banks that they had a chance to do some years ago during previous small blips and small down turns like the internet bubble burst.

  40. “of course the financial collapse has exacerbated the deficit but we were running a deficit of 3% of GDP in good economic times”

    Mike – thanks, that’s exactly what I said, but it looks like you didn’t grasp the fact.

  41. “that we always end up bankrupt when Labour get their envious grip on things.”

    You’ve clearly studied all there is to know.

    Blind ignorance won’t get you far, Mike, just pitied.

  42. Yes OK, we are not broke. And Labour have never made us broke. They always run the economy expertly.

    Blindness won’t get you far, John, just pitied.

    Keep your mouth closed. You dont want to get sand in it.

  43. M – How do you come to the conclusion that this site is for swivel-eyed self-convincers to bang on at each other? Your tone is deeply offensive.

    In terms of understanding economics, let me explain so that you might understand about how a 3p reduction in the price of something can make a difference. If you spray an air freshener into a large room, atomised droplets, invisible to the eye will make a difference to the smell of a room.

    It works the same way , and , logically, if a tax reduction costs £12bn, then it must be spent.

    Tax reductions are offset by other tax rise, and so we look at the overall tax take, and in this case, it is clearly lower as a % than it has been.

    There are different types of taxes, which have different effects. An income tax cut (although there has been one to cover the 10p losers) is not as good as a spending tax cut IF you are aiming to boost spending.

    A spending tax cut is likewise not as good as an income tax cut if you want to boost saving.

    I hope you understand a little better now, and get over the ad hominem rot that stinks out this respectable site.

  44. “It works the same way , and , logically, if a tax reduction costs £12bn, then it must be spent.”

    You cant come to the “logical” conclusion that it will be spent at all. It could be saved by consumers/retailers and not lent out by their banks due to it being sucked into repairing the black holes in their balance sheets.

    And I know you will disagree with this since you subscribe to fairy tale economics, but VAT is the lesser evil of the taxes. Having that reduction work and getting people out there blowing money on consumer goods again is not at all what is needed at the moment.

    I have no doubt you will be of the opinion that the stimulation of borrowing and spending is exactly what is needed though. Thankfully the number of people in the UK believing in this silliness is decreasing by the day and should not effect the next election.

  45. “How do you come to the conclusion that this site is for swivel-eyed self-convincers to bang on at each other? Your tone is deeply offensive.”

    Oh come on. Like you have never been in the thick of it in the past. You are no Chris Newey, but dont pretend you are so far above it.

  46. Heh M – Take exception to that; I only give a balancing view to the prodominantly right-wing dominance of views here.

    Just because you don’t agree with me don’t lable me in a way that’s different to what you and others have submitted.

  47. It could be saved by consumers/retailers

    For a start, the retailers and customers cannot be lumped together like that, as they are on opposite ends of the transaction. The money will only reach the retailer once it has been spent. That’s why it’s called a spending tax.

    You refer to “lending out by banks”, whereas the issue is to do with spending money, irrespective of the debt position of the spender. If, say, the average consumer will earn £25k in the year , and spend £12k of it on VATable goods and services, then bank overdrafts don’t come into it.

    “fairy Tale economics” is a pitiful remark. I don’t mind exchanging ideas with people , no matter how silly their tone is, but aAnthony will probably delete your posts if you do it too often.

    VAT and income tax, petrol duty, IHT , NI (all types) have different effects. You prefer higher VAT to higher income tax. However (and there’s legitimate debate about this), a VAT cut cannot leave the money “saved”, it must be spent. The idea behind it was to boost spending at a time when, with confidence very weak, and home purchase interest rates coming down, we risked a catastrophic slump in demand which would have guaranteed a period of deflation and unemployment rates the like of which are being seen in Spain.

    Ken Clarke agreed in the Autumn that a VAT cut was the better route, and he’s not exactly a left-winger is he?

    I cewrtainly believe that borrowing more now in order to pay it back when growth kicks in is the better way forward, and fortunately the Chancellor believes it too.

    Where both sides agree is that it needs to be paid back asap, and that, I think is the difference between the Labour Party of to-day and that of the seventies.

    I owed a lot on my credit card a couple of years ago. If I’d sold my car and paid it off at the time, I’d have put myself out of business.Instead, I borrowed more to see me through a tough period. I now have no credit card debt. Write that large, and you have the other argument from your own. Listening to it might not persuade you that you are wrong, but at least it will show yourself to be a thinker.

  48. Well put John but think your giving him too much credit. His previous posts prove this!

  49. me in disguise?

  50. Not sure people can do that

    [No, they can’t – AW]

1 2