The full tables for MORI’s poll are now available here.

A notable finding there is that economic optimism has significantly dipped since the PBR, down from plus 10 last month to minus 4 now. I think this is the first poll on economic optimism since the PBR, but is entirely in line with the YouGov/Sunday Times poll that showed people becoming much more negative about the country’s current economic situation after the PBR.

Turning to voting intention, there is a shift in propensity to vote, 68% of Tory voters now say they will definitely vote, compared to 61% last time. The proportion of Labour voters who say they will actually vote is down 2 points to 49%.

However, the biggest difference between this month and last is the make up of the sample itself. The past vote breakdown in this poll was CON 22%, LAB 28%, LDEM 9% (the equivalent, if you exclude did not votes and so on, of CON 33%, LAB 43%, LDEM 14%). Compare this to last month’s past vote breakdown of CON 19%, LAB 30%, LDEM 10% (the equivalent of CON 29%, LAB 46%, LDEM 16%). It is the lack of political weighting that has produced such extreme switches in support in MORI’s recent polls – ICM last weekend had a similarly pro-Labour sample, but their weighting brought it back to showing a much smaller swing.

MORI do not weight by past vote because of false recall. The question is not whether false recall exists or not (all the established pollsters accept that), but how variable it is. ICM and Populus believe it is generally steady and doesn’t change wildly from month to month (the implication is that it is largely the result of a social desirability bias making people claim to have voted when they didn’t, and the way people report tactical voting). MORI believe that it does genuinely vary slightly from month to month, and that there is the potential for it to change significantly (the implication there being that it changes with people’s current political preferences).

Politically weighting polls makes them less volatile. However, MORI’s contention would be that the public are genuinely very volatile, and that companies like ICM and Populus risk weighting out genuine volatility. There’s no way of proving with certainty who is correct, though one thing to look at is the results of YouGov, who take a notably different approach. YouGov weight by party ID rather than past vote, but more importantly, they weight using information given by their panellists in the past (either on election day 2005 or when they joined the panel), rather than using data collected as part of that day’s survey. So, while YouGov’s weighting targets could be wrong, they can at least be confident that they are not changing from day to day. Their method would not dampen down genuine volatility, and their figures are one of the less volatile.

For those of you not bored to tears by discussion of methodology there is, incidentally, a one-day conference at the Royal British Academy next month (jointly hosted by the British Polling Council and the National Centre for Research Methods) discussing approaches to polling methodology, featuring Peter Kellner, Andrew Cooper, Martin Boon, Nick Moon and – most relevantly to this, John Curtice discussing whether it is safe to weight by past vote (Simon Atkinson of MORI is also there to put what I expect will be the counter argument!). More info here.

83 Responses to “More on Ipsos MORI’s poll”

1 2
  1. Agreed fully. The numbers they have used haven’t been weighted.

    The weighted figures are 13% lead, which is closer to the consenus.

    It’s OK saying none weighted figures are more accurate. However, you then can’t compare them to other polls which are weighted, and demonstrate this as a change in voting intention.

    It’s always a volatile poll in my opinion. I do think certain polls are done just to create headlines. MORI may be one of them.

    You have to remember that newspapers don’t commision polls for accuracy. They commission them as they want something to make a story out of.

  2. Certain questions you always have to ask about polls.

    Ones done for Sunday newspapers. I’d put a serious argument across that they are always odd. More than likely because they have been commisioned to create headlines.

    The more accurate polls I find that are those that aren’t done for headline making. The ones you get on a Wednesday evening, that don’t get any media attention.

    I’d guess that the papers make concious choices to use certain pollsters at certain times to guarentee themselves something to print.

    An unweighted MORI poll is probably that.

    Secondly, I think that polls influence polls. When you have a series of particularly poor tory polls, it will galvanize support in future polls.

    The problem being, this may not be actual voting intention.Just the “god, Labour are doing better. Better go overly positive for the tories to stop the trend”.

    Definite voting intention has juimped for the tories in this poll. I’d suggest this is a mere knee jerk reaction to weeks of poor polling.

    Finally, I understand why the tories aren’t getting overly excited about any polling. And I really wouldn’t.

    I was watching a political commentator stating they won’t be comfortable of a good majority (50+) unless they go into the polls with a 20% lead.

    I completely agree. You have to take into consideration the political landscape, and the way polls are often used as a protest vote.

    The tories will have to win big in the north, and probably wales to get their majority. They’ll also have to create a total Labour to Tory swing of 7%.

    As an example the biggest swing thus far in any election (euro/council) has been 5%. And that was when the tories were enjoying regular 15% leads.

    Recent council elections, and polling suggests that although the tories lead isn’t dropping, core Labour support is getting more soild

    I would argue that is nothing that either party has done. More that we are just getting into election mode, and people are being more serious on their vote.

    Labour are being very smart in pandering to core vote., On the basis that the tories need a big swing in that core vote to have the success they want.

    In effect the tories can win a heck of a lot of Labour seats, but if they don’t make a big impact in the north, wales, they could still well even lose.

    I do see a lot of 1992 here. Major just won that as Labour couldn’t get the trust of core tory voters, in regards to tax.

    Major played a very dirty campaign, of telling fibs on planned Labour tax rises (income tax has gone down 4% since Labour came to power). To scare the core support into line.

    They should, and will play the samer trick. In effect ignoring their “Blair” seats, that he won from the tories, and just concentrating on strong holds.

    There will be lots of fear mongering about cuts, and public services.

    To harden resistance to the tories in the seats they need to win.

    Personally, I think most polls are top heavy to the tories, whatever they are.

    Just because at least a small part of these polls,will be people angry with Brown, but probably not angry enough to vote Cameron.

  3. It is pretty striking how the certainty to vote filter changes the predicted shares (43/26 at 10, 40/29 at 6-10).

    Do MORI use 10/10 for any good reason? I know if counteracts the otherwise higher estimated Labour share from the lack of political weighting, but surely this can’t be the reason – two wrongs don’t make a right so to speak.

  4. “…and their figures are one of the less unvolatile.”

    Um…my brain hurt when I tried to work that one out.

    Does that mean YouGov are one of the more volatile?

  5. The conference is at the British Academy not the Royal Academy.

    All of this is an excellent reason for using the WMA – the more we understand the detail of the differing methodologies the more we can see an argument for each, yet they will give different errors on different occasions. I still don’t really understand why Angus Reid gives CLeads so much higher than the others though.

  6. It’s funny you mention about voter recall because, in my opinion, the only really accurate weighting variable is past vote. This is because the idea of polls is to predict the election and the only information, the absolutely only direct information, we have regarding the past election is previous vote. For instance, we don’t know how many women over the age of 50 voted Labour – all we know is how many people voted Labour.

    The problem, as you rightly point out, is how do you ensure you get an accurate picture of past vote. False recall is a massive issue on this and really the only way to account wholly for it is the method YouGov use – essentially using historic answers to weight.

  7. Phillipe – If MORI had past vote weighted their figures it would not have shown a 13 point lead. If they’d weighted to ICM’s past vote targets it would have shown an *even larger* Conservative lead. The difference would be down to their very harsh likelihood to vote filter which strongly favours the Conservatives (as Statto has mentioned).

    MORI’s polls are NOT commissioned to make headlines, as they are not commissioned by newspapers. They are a regular monthly series carried out since the mid-1980s. MORI lost their regular contract with the Times around 2001 and since then they have continued doing them without a regular client, sometimes getting individual deals.

    Polls done at the weekends are done using the same methodology as mid-week ones, so should be no more volatile. If there is a difference, it is between regular polls and ad hoc ones. Regular monthly polls occur at roughly the same time regardless. Ad hoc ones are often commissioned in response to events, so will often show sharp movements because of their timing. This does not correlate to weekend/weekday polls though – there are both regular and ad hoc weekend polls, and there are regular and ad hoc weekday polls. This one, incidentally, is very much a regular poll, done in response to a timetable, not events.

    NBeale – whoops. I habitually do that.

    West Midlands – I meant less volatile. It’s because in my draft I wrote “relatively unvolatile”, which I thought in hindsight was actually a rather pompous way of saying less volatile… but only managed to half correct it! :)

    Statto – I believe the basis for MORI’s decision is that the proportion of people who say they are 10/10 certain to vote has at past elections been closest to the actual turnout figure. However, it is demonstratably not the case that people who say they are 10/10 certain do vote and people who say 9/10 are certain not too (in fact the the BES has checked answers against the marked electoral register), so I think it is quite hard to defend.

  8. Yougov’s poll showing Con 40 Lab 28 Lib Dems 18 is virtually identical with the average score of all polls taken over the last two months. The biggest slight difference is that the average polls suggest that the Lib Dems are almost 19.

  9. Thanks Anthony, Bob Worcester’s comments regarding turnout now make more sense to me:

    “a turnout of just 50% would return a Tory majority over all other parties of over 100, a 78% turnout would see a Labour majority of about 25. At about 60%, a 40% Tory, 30% Labour, 20% LibDems, 10% others, would give the Conservatives a majority of around 20 seats over all other parties.”

  10. @NBeale

    I use a Modified Moving Average rather than a Weighted, because with the sole exception of Angus Reid, I think that all the poll takers are within margin of error of each other. And so they’re all within margin of error of the ‘real’ trendline.

    Ipsos-Mori may be volatile, but it’s volatile in a margin of error around the trend line. Angus Reid however appears to have an offset that subtracts from Labour and gives to ‘Others’, that is not reflected in any other polling.

    Personally I think the current Ipsos-Mori poll is above the trend line, and their last one was below the trend line.

  11. Mori poll – England only – unweighted base 860 respondents

    Con 46% : Lab 26% : LD 20% : Green 3% : UKIP 3% : BNP 2%

    Mori poll – Scotland & Wales only – unweighted base 157 respondents

    SNP/Plaid 22% : Con 23% : Lab 29% : LD 16% : Green 4% : UKIP 4% : BNP 1%

    Not that linking Scotland & Wales means much of course! :-)

  12. Oldnat

    Interesting element in your Scot/Wales breakdown – UKIP are a third higher than in England !

    Shows what you can do with stats even when you are not trying !

  13. Paul H-J


    It was that blasted Welshman who couldn’t decide between BNP and UKIP, but opted for UKIP! :-)

  14. @OLDNAT
    Interesting, even if statistically not all that significant, that in England the Conservative support matches Labour and LibDem combined, and that in Scotland and Wales combined the Cons just squeeze in ahead of the Nats. The latter presumably says more about Wales than Scotland.

  15. One of the most worrying indications for Labour at the moment has to be that the breakdowns of nearly every poll recently (which can’t be properly relied upon because of small sample size) have shown that the Labour vote in the Midlands is at almost as low a level as in the south excluding London. And the Midlands of course includes the West Midlands conurbation, as well as Nottingham, Leicester, Derby, etc.

    If figures of that kind prove to be correct Labour wouldn’t stand a chance simply because they would lose so many of their Midlands’ seats. Maybe Gordon Brown is particularly unpopular in the Midlands for some reason. Maybe one of the pollsters might be interested to do some Midlands-specific polling to see if this seems to be the case.

  16. I wander if some of these polls are commissioned to try and encourouge more voting. Mori did a poll for the pro conservative times and it was 6% lead for the Tories. It did one for thr pro Labour guardian and it was 17% maybe they are picking days they think people may be happier or less happy with the governement to encourage either Conservative or Labour voters to vote, or even liberals to vote depending.

  17. Andy Stidwill

    Had a look at the 2005 results for the English Midlands. I hadn’t realised how close the % share was between Con and Lab then.

    What’s been the average swing there in the sub samples – around 10%? Isn’t that around the average English swing since 2005?

    It’s not an area I know at all. Does it have a significant number of marginals, or are the constituencies socially different – ie natural “Labour” or “Tory” seats?

  18. Regarding the sample size [ say, 1000 ], shouldn’t the polls be undertaken in the 200 odd seats on both sides of the most marginal seat. What is the point of asking anyone in the Rhondda or Kensington and Chelsea ?

    After all it is those 200 seats [ actually even less so ] which will decide the result !

  19. I predict Labour to scrape victory. These polls don’t take shy Tories into account. I recall Labour 22 point lead going into the election in 1992. Labour lost giving John Major a comfortable majority. I am the first to sat its time for a change but I cannot trust the Tories.

  20. 2005 election:

    West Midlands:
    Lab – 39%, C – 35%, LD – 19%

    East Midlands:
    Lab – 39%, C – 37%, LD – 18%

    South East:
    C – 45%, LD – 25%, Lab – 24%

    South West:
    C – 39%, LD – 33%, Lab – 23%

    So the Labour vote in the Midlands was roughly 15% higher than in the south excluding London at the last general election.

    Most of the regional breakdowns I’ve seen recently are showing the Labour vote only 2-3% higher in the Midlands than the south. One of the most recent polls I think showed Labour on 20% in the Midlands and 18% in the south.

  21. All very interesting stuff eh?

    All you really need to do is take a look at the English breakdown and compare them with the 2005 figures to see Labour will be slaughtered in May and Wales could prove very interesting too with Labour also in serious trouble.
    January 4th will see the ‘official’ starting gun fired by David Cameron then just watch the Opinion Polls shoot all over the place.

    Excellent site Anthony!

  22. Even to get a turnout of 60% would require a massive increase in a now relatively apathetic base of Labour abstainers. For while a turnout of just 50% would return a Tory majority over all other parties of over 100, a 78% turnout would see a Labour majority of about 25. At about 60%, a 40% Tory, 30% Labour, 20% LibDems, 10% others, would give the Conservatives a majority of around 20 seats over all other parties.

    The above is from Bob Worcester’s blog. Can someone explain to me how a differential turnout giving different results is calculated. After all, an opinion poll only tells us the distribution assuming all parties’ supporters turns out equally. Right ?

    Of course, if 100% of Labour supporters turned out and only 50% of Tories did, Labour would win. But how can we work out that at 78% Labour turnout, Labour will have a majority of 25, even with the same distribution.

  23. MORI filters its results according to likelihood to vote from 1 to 10. So MORI can give you its results for different likelihoods to vote, so-to-speak. Their headline figure only includes results from people who are 10/10 likely to vote, absolutely certain. But they also have results based on other likelihoods to vote as well which is where those other numbers are coming from.

  24. Andy: I see. So starting at 10/10 and going down 9/10 etc. [ and at the same extrapolating what the overall turnout will be ], they can deduce that.

    Interesting. But that leads to the obvious question. What turnout does 10/10 imply ? Surely, anything less tha 60% will be simply unrepresenative. Even in 2005, with many labour voters clearly unhappy with Iraq and staying at home the turnout was circa 60%.

    Does anyone know, what 10/10 would amount to ?

  25. Reigate and Banstead BC, Earlswood and Whitebushes
    Con 391 (37.9;-12.8), Lib Dem 313 (30.4;+6.6), Lab 161 (15.6;-0.7), UKIP 125 (12.1;+2.8), BNP 41 (4.0;+4.0)
    Majority 78. Turnout 16.38%. Con hold. Percentage change is since May 2008.

    Extremely unrepresentative sample. In fact the only local by-election with more than 2 party participation I could find. [ Dec 17 ].

    Presumably, safe Tory ward. But UKIP maintaining 12% of the votes where Europe I am sure was not an issue. It tells me that in Tory heartlands, Europe could snip a few points off the Tories.

  26. Paul B – MORI don’t do polls for the Times (or, for that matter, for the Guardian), both their polls were for the Observer. When pollsters do polls for more than one newspaper they do not use different methodology for different clients, they are carried out in identical manners.

    David C – You recall wrongly, the polls from the 1992 election are all on the site here, and there was definitely no 22 point Labour lead going into the election! More to the point, the 1992 polls were the worst ever performance by the British pollsters and the lessons were eventually learnt leading to major changes. Polling methodology is drasticly different, including adjustments for “shy voters”.

  27. I have done some analysis of the 1992 election at Christmas 1991 and taking the last ten polls of the year, showed Lab and Cons at neck and neck.

    There was certainly no 22 point lead for Labour and a good few of the polls had the Tories showing a lead of 4 to 5 points. So quite what Kinnock was getting so excited about I’m not quite sure,

    There was a solitary double-digit lead in June 1991 (10 points for Labour) but before that the last time Labour enjoyed such a lead was before John Major became PM.

    Other than that the leads were in low single digits for Labour and a few for the Tories. It was a completely different landscape to that which we see today.

  28. Anthony

    Is this Mori poll the last one of the year or are we expecting another poll from another firm (maybe Comres) before the end of the year?

  29. Sorry anthony I don’t know why I thought it was the guardian. I also did not question the methodology. I meant that newspapers may pick a time they think the vote is close or not close to encourage voters to vote. I think the Times may have just quoted someone elses poll to say to Tory voters that if you don’t vote Labour might have a chance. The pollsters aren’t biast but papers do often try to get polls done after certain event such as the budget to sell a point of view to their buyers.

  30. Richard – Assuming the normal cycle of monthly polls, there is ComRes in the Indy and YouGov in the Daily Telegraph still to come.

  31. Anybody heard that the Tory lead in marginals is greater than average. Also a lot of southern suburban seats that Labour won under Blair will revert. Remember Blair had an appeal that went beyond the traditional Labour support. Brown is devoid of this.

  32. Removing AR makes almost no difference to the WMA (instead of 40:27:19 it becomes 41:28:18) but as soon as you make ad-hoc adjustments like that the statistics become intractable analytically.

    All this stuff about adjusting for likelihood to vote is another reason why the real MoE of these polls is nearer +/-5% than 3% which is what the simple theory would predict.

    I remember Bob Worcester telling us in terms at a meeting at the Royal Society that there was no way the conservatives could win the next election. So I’m a bit sceptical about his special pleading. We shall see.

  33. Bob Worcester ended his book on the last election with something very similar to these words: “A third Labour victory? Don’t bet against it.” I was a bit surprised that he put that in myself.

  34. Whoops – “a fourth Labour victory” is what I obviously meant!

  35. @surbiton

    haven’t seen you talk about the collapse in the conservative vote in this thread, Have I missed a post ?

  36. Anthony Wells

    Mori has shown a 11% Labour to Tory Swing in less than a month.

    They are twaddle for the Sunday papers

  37. Paul B

    Polls certainly affect each other. A bad poll for the tories (that makes the news) is often followed by a big surge in support later on.

    The “let’s show Labour” rule. Hardening voting intention, just to prove a point.

    People who may actually vote UKIP, just saying Tory to hurt Labour.

    Works both ways by the way. But I think it happens.

  38. There seems to be much concern about the effect of selecting only those 10/10 certain to vote. Is that at a theoretical General Election tomorrow, or at the real one? The closer we get the more the distinction matters. No one can even be 100% certain that they will vote tomorrow, nevermind whether they will vote next May or June – after all a number may be dead by then! Perhaps the 9/10 band are more honest and are thinking ahead. Can they be isolated?

  39. Should also add that Cameron has most of the media to help him.

    Sky News, Times, Telegraph, Sun, Mail, even Guardian have been on full attack mode since the tory lead started to slip.

    Guardian was even printing a message to readers from David Cameron on their Thursday edition to coincide with Browns Climate Change visit.

    Pretty obvious they’ve given up on Brown as well if you ask me. That was bordering in sabotage by the Guardian. To get Brown off the front page on his big weekend.

    As Blair showed, when you have almost the entire media, it’s a big weapon.

    I’d state these polls could simply just reflect the concerted, across the board media campaign to attack Labour – once the tories started dipping in the polls.

    The boost could well be the fruits of their labour. After they noticed their boys were struggling a bit.

    In truth, it’s next to impossible to win an election without major media backing.

    The media would have been on full attack mode, no matter what the PBR announced.

    Business decide who they want in power. Business pass message on to media. Media tell public who to vote for. Public listen. Leader pays businesses back through tax breaks and favours

  40. Given that Labour do well in London compared to the rest of the Southeast I wonder how the banking crisis and the bankers bonus tax will impact on voters.

    Two possible outcomes;

    The impact on the city and house prices extra means that labour is hurt in both and that its position in London which is vital too its prospects are eroded.


    The banking crisis goes down badly out with London where many commute in to the city but from safe tory seats but the Banking tax goes down well with core Labour voters in London itself and so the difference between London and out with actually widens.

    I’d rule out the third possibility of a general London recovery because Labour have slid back so far in the South that I can’t see the bankers bonus issue turning it around.

    What would be great between now and March would be the Pollsters doing some regional polling? It shouldn’t be too difficult for anyone to do a 2,000 sample in each of their electoral regions.


  41. @ Phillippe

    I feel I ought to answer some of your accusations point-by-point:

    “Ones done for Sunday newspapers. I’d put a serious argument across that they are always odd. More than likely because they have been commisioned to create headlines.”

    They’re not. They’re commissioned monthly, with additionals for special events, e.g. Budget, conferences, PBR. And any journalist worth their salt could make a headline out of any result.

    “As an example the biggest swing thus far in any election (euro/council) has been 5%. And that was when the tories were enjoying regular 15% leads.”

    Euro and council elections are not indicative of Westminster voting. If they had been, the Euros would have predicted a Labour loss in 2001.

    “Labour are being very smart in pandering to core vote., On the basis that the tories need a big swing in that core vote”

    No, they don’t. That is precisely where the votes aren’t needed. You need them in the areas where they lost seats in 1997, i.e. marginals.

    “Major played a very dirty campaign, of telling fibs on planned Labour tax rises…”

    I quote this from the 1992 manifesto: “A new top rate income tax of 50 per cent will apply to individuals with an income of at least £40,000 this year.”


    “Instead of cutting income tax, we will make additional resources… available”

    Whatever Major did or did not fib about, Labour’s planned 1992 income tax increases are not debatable and these quotations are cut-and-paste from the Labour Party’s website.

    “…To scare the core support into line.”

    Again, not really. The core vote is likely to vote a donkey in with the right rosette. That’s their definition. Further, the Tories registered more than 14 million votes in 1992, the biggest vote for a party in UK elections ever, before or since. This is, by definition, not the “core vote”.

  42. Some regional polling would be a good idea in my opinion, especially in the Midlands, London, Yorkshire&Humberside.

  43. Andy Stidwill

    When I first started posting here, suggesting that GB UNS had become meaningless as far as Scotland was concerned, I was met with a certain amount of derision.

    Things seem to have moved on, and English regional differences are now seen as being worthy of analysis.

    If England is no longer seen as a single political dynamic, then that is an interesting development.

  44. It can’t be ignored that relatively few of the 116 seats the Tories need for a majority are in Scotland, Wales, North East, South West, but of course those regions can very much have an effect on the national polling numbers that people are enthusiastically typing into the various swing calculators. For example it’s likely that the Labour vote will fall less than average in the two regions of Scotland and the South West (for completely different reasons – in Scotland because natural support for Labour seems to be solidifying and in the SW because the Labour vote is already very low), and those regions could have the effect of disguising a much bigger drop in the Labour vote in the regions where the mass of marginals are to be found.

    Obviously the PoliticsHome 2009 poll of marginals was extremely useful – it’s only a shame there isn’t another one coming up before the election.

  45. Andy Stidwill

    Damn! I was just having a last surf of the blogs before heading to my pit when I saw your response (if you were, in fact, responding to my post).

    Anthony said to me, previously, that GB wide polls were conducted because that was what the customers asked for (though that doesn’t explain why MORI continue to do that).

    The “British” narrative of polling (and the statistical manipulations of the pollsters to continue that narrative) can be equally explained in at least two ways –

    1. The “establishment” requires that the narrative of Britishness in political expression be maintained, since their power lies in manipulation of the 2 party system in the UK. Thus deviance from a British norm is to be minimised.

    2. The “establishment” is simply lazy, incompetent and ignorant – unconcerned about deviations from the M25 norm.

    Both may, of course, be true!

    What I find interesting is that you are calling for English regional polling on a significant scale. If GB (or England) is the single political identity that the establishment thinks it is (or wants us to think it is), then regional polling is pointless. Instead you would be asking for matched polling of similar demographically described seats in GB regardless of their location.

    However, you have moved beyond establishment thinking. You clearly recognise that location within GB has an effect on voting, regardless of demographic factors. This is dangerous stuff, and you will no doubt find yourself on some MI5 list as a potential threat to the UK state. :-)

    Merry Xmas!

  46. There is no need for an “English” poll. Just the top 200 marginals [ anywhere ] will suffice. After all, these are the seats that will decide the outcome as per FPTP. In fact, polling the rest is a waste of time.

  47. Bryan: My comments related to the Tory local by-elections vote on Dec 17, the Tory vote is still collapsing. The sample consists of only one ward though ! Their share went down 12% in a safe seat. UKIP got 12.2% of the total even though this was a local election.

  48. Sorry – rewrite

    Bryan: My comments related to the Tory local by-elections vote only. As of Dec 17, the Tory vote is still collapsing. The sample on that date consists of only one ward though ! Their share went down 12% in a safe seat. UKIP got 12.2% of the total even though this was a local election. LD went up 6% and BNP came in first time with 4.4%. This was in Reigate, true blue territory.

  49. @ Phillippe – you’re off your rocker if you seriously believe The Guardian supports the Conservatives.

  50. Angus Reid Strategies have a totally different approach, saying that emotional response questions are better predictors of voting than past voting patterns.

1 2