New Angus Reid poll

Political Betting have a new poll by Angus Reid. The topline figures, with changes from their last poll, are CON 40%(nc), LAB 24%(+1), LDEM 20%(+1). Others are at 15%, down 3.

Overall there is no significant change from their previous poll. The Conservative lead is, needless to say, much greater than in most other recent polls, as the level of Labour support is lower. This is not a sign of some sudden Conservative recovery – Angus Reid have been showing the Tory lead up at this level all along – it’s just to do with different methodology.

Specifically Angus Reid assume no false recall when weighting their samples to recalled 2005 vote, which means their samples have fewer people who claim to have voted Labour in 2005 than phone pollsters do. For some reason they also tend to show a higher level of support for others – I’ve no clear explanation for that, it could be their question (they ask who people would “support”, not how they would vote) or it could be something entirely different.

Still to come this year we should have a YouGov poll for the Telegraph, which normally comes out the last Friday of the month (though it’s probably safe to assume it will be a different day this year!), a ComRes poll for the Indy, and Ipsos MORI’s monthly monitor, which was apparently conducted last weekend but has yet to be released into the wild. I expect, as was the case last month, that one of the Sunday papers is going to print it.


124 Responses to “New Angus Reid poll”

1 2 3
  1. Thanks STATTO, your guess is as good as anyone’s.

    @Andy Stidwill

    You only need to look at Anthony’s European Election voting intention page to know that this isn’t true. Only ComRes were considerably out.

  2. Surely the trend for each polster is the important factor and this AR is basically a no change on its’ previous one.
    Seems about right as things appear to have stabilised.
    I suppose if Mori has more than 10% lead for Cons we should concude the gap may have widened a little since the previous one; although, Anthony will disect for us as I recall there was an explanataion for the 6%.
    I hope AR as wrong but won’t trash them as they may be the most accurate?
    Should their next one be 43/21 or 3727 that would be meaningful if mirrored by directional changes from other polsters.

  3. Though I could be be wrong I do get the strong impression that many of our contibutors are quite new
    to politics. I certainly don’t have all the answers but I
    have been around the bouys several times, also contested many elections and for more than one party I’m not
    Greybeard for nothing) Please take it from me Euro Elections are so very different from “the big one” as most voters call it. While it makes little sense after almost 13 years the next GE is still wide open (sorry Mark R) and the polls will increasing show it.

  4. Everyone pretends there not partisan but i think some people underestimate the anger of people who have been crippled by the government

  5. @AW

    You ask for the mystery major difference between this poll and all the others… I already set that out. This is *not* a voter intention poll, it’s a party support poll. And that is a substantial difference in the question asked, not merely a slight difference in wording.

    I suggest that what this poll shows is that Labour voters are substantially less likely than Conservative voters to call themselves party supporters in this election. And that while people won’t actually vote for ‘Others’ on election day, they will say they support them.

  6. @Mark R

    The Liberal Democrats have stated they will not form coalition with any government that would not commit the nation to AV+ voting.

    The Conservatives have pledged *opposition* to AV+ voting. A u-turn on that would see a huge party revolt as swathes of seats suddenly switched from safe to contested or even likely liberal democrat/labour.

    And a Liberal Democrat climb down from that demand would see a huge party revolt for them, as the rank and file, and every single LibDem MP has enacting AV+ as the first priority of government.

  7. Jay Blanc

    “This is *not* a voter intention poll, it’s a party support poll. And that is a substantial difference in the question asked, not merely a slight difference in wording.”

    I had wondered about that. If your hypothesis is correct then that difference should be especially clear in Scotland, where we know from simultaneous polling on Holyrood and Westminster voting intentions, that many SNP supporters will vote (mainly) Lab but also Con for Westminster.

    Comparing the Scottish sub-sample to those in other GB polls, I just don’t see the evidence for that hypothesis.

  8. @Anthony Wells

    Thanks for that Anthony….point taken…

  9. @Old Nat

    The sample for Scotland was 165 people, upweighted by somewhere around 5%. And I can’t see breakouts for unweighted by response, so can’t tell if their weighting might mask the effect in the ten times as large sample-set of England.

  10. Jay Blanc

    Scots and English are not so different! However, because we are quite used to voting different ways depending on which Parliament we vote for, the terms “support” and “vote for” would resonate more strongly here than in England.

    I don’t say your hypothesis is wrong, but you need some evidence to support it – other than your assertion.

    The small Scottish sample might have been expected to give some support to your hypothesis, but it doesn’t.

    You need some evidence if you are going to persuade people.

  11. People may well answer differently in terms of support/vote, since it could produce different answers in terms of tactical voting. What I’m certain they are not doing is giving some sort of “party identification”, since the answers simply don’t add up.

    Party ID is a question that’s regularly asked by other companies – MORI ask it, ComRes ask it, YouGov weight by it, the British Election Studies ask it. Far more people identify with Labour than vote Labour, if people were answering with party ID rather than who they would vote for, you’d see Labour getting a higher, not lower, level of support.

    To know for sure, someone needs to do a split sample comparing results from the two different wordings – as OldNat says, otherwise there is no evidence.

  12. It’s burden of proof to show that the Angus Reid “Party Support” poll is a true voter intention poll, not show they aren’t. The old ‘prove a negative’ thing. You say AR is showing accurate enough voter intent to be considered a voter intention poll, despite the big difference between it’s question asked and the results generated to other voter intention polls. I say you need to provide proof of that before I consider it.

    I’m not convinced by the lack of evidence showing that AR is giving accurate voter intention figures. The benefit of the doubt in suspect cases should go to polls that are in-line with the trend, not outliers.

  13. Anthony

    Re Party ID

    Has anyone done any polling on what that actually means?

    eg Those who say “Labour” – do they mean the traditional Labour Party values that they still identify with? or the New Labour Party that Blair/Brown et al created?

    In other words is “Labour” a code word for a set of values that may or may not have any relationship to the political party that is currently the UK Government?

    If not, it may be that the pollsters are asking a question which is producing false answers in terms of party identification.

  14. Oldnat –

    Yep, MORI do a version where they break it down into sub groups like social democrat, old labour, new labour, thatcherite, old nation Conservative and so on.

  15. Anthony:

    On another page you gave us some insights into responses to invitations answer polling questions. Can you tell me please what is the usual range of rates of refusal to participate?

  16. Anthony

    Thanks.

    Is there a link to any correlation between these subgroups and differential voting patterns?

    (I know I ask a lot of questions – but its your own fault for having so many answers! :-) )

  17. 24% for Labour is not way out unrealistic when you consider that only 26% thought Brown was doing well as PM and Cameron 52% as party leader.

  18. … in the recent Yougov poll.

  19. @Philip JW

    At 24%, it is 6% below the pollsters that have a track record in the UK. Nowhere near margain of error, and never has been.

  20. The only accurate survey would have to ask “which of these candidates in your constituency would you vote for”.Theres not a lot of point finding that people want to vote for a party that is’nt standing in their seat.

  21. Oldnat – well I’m now going to let you down by not having answers to lots of questions!

    Firstly, off the top of my head I’ve no idea. MORI have asked the same question a couple of times, so some may have been in polls where cross breaks with voting intention are available. You’d have to google it.

    Philip JW – remember those numbers are not really comparable. The 26% of people who thought Brown was doing well is 26% of all people, the 24% of people here is 24% of people who gave a voting intention – so that 26% actually represents a significantly larger number of people!

    John B Dick – I’m afraid not. I’ve no personal experience of phone polling, and the figures are not something that the phone pollsters often publish in detail.

  22. Oldnat –

    Here are the past figures for the question when MORI asked them, but sadly none of the detailled tables with cross breaks seem to be on MORI’s website anymore :(

    http://www.ipsos-mori.com/researchpublications/researcharchive/poll.aspx?oItemId=2380&view=wide

  23. D Bliss

    But that’s a question that can only be asked once registration of candidates has closed.

    At the moment, I know that the 4 main (actually probably now 3 main + LD) parties will be standing in my constituency. Whether the FSP, UKIP, Green, Christians, Senior Citizens, BNP etc will put up a candidate here is unknown.

  24. Let’s face it, no-one will know which polls are way out until the actual election. Though it would seem that those polls which closely agree with each other are more likely to be correct than are ‘outliers’, it is not impossible that the outlier is correct.

    Time will tell.

  25. D Bliss –

    That’s how MORI used to ask their question during election campaigns (I assume they will no longer be able to do so now they have fully to telephone polling).

  26. Anthony

    Thanks for that. You didn’t disappoint – so I’ll feel free to ask lots more questions in future! I spent many years as a teacher – it’s quite nice being a pupil again. :-)

  27. Surely if you ask people which party they support many of those who name an “other” may not have an opportunity to actually vote for their choice.15% would soon fall below 10% in an actual election.Many will say they want to vote Green but with no candidate standing will they vote?Who for?

  28. Jay Blanc
    “It’s burden of proof to show that the Angus Reid “Party Support” poll is a true voter intention poll, not show they aren’t.”
    “The benefit of the doubt in suspect cases should go to polls that are in-line with the trend, not outliers”

    You seem to have lost the plot. OLDNAt has made some relevant points which you have dismissed. This is not a court of law and you have not convinced us by your arguments which seem a little arrogent if I may say so. It is too easy to assume that A R is wrong and other polls are right and I for one while recognising there are good reasons why there have been differences in the polling results suspect that the support/vote for question difference has had a minimal impact.

  29. D Bliss . you may not appreciate it but the AR question is “who would you be most likely to support in your constituency”

    There may be people who like to think of themselves as Labour supporters in general but who are thinking about voting for someone else in their particular seat at the next election and Angus Reid may be picking up this difference. If so it may be the Angus Reid poll which is more reliable.

  30. Aren’t we rather crediting the respondents to polls with rather too deep a level of thought here?

    The vast majority of people know basically what you’re asking when you give them a list of parties and ask you to choose. I imagine if you phoned them back two days later and asked them “When you were polled were you asked a) who you would vote for or b) who would you support?” most would not be able to remember accurately if at all.

    I don’t think that people, when polled, immediately start thinking about the dynamics of their own constituency. Most people have a general view about party politics, and place themselves either firmly in one party’s camp, or in a floating position at the interface between two parties.

    There is clearly a differential at work with AR’s methodology, but like Anthony I don’t think its the question.

  31. @Neil A
    I am surprised you credit the population of the UK with any level of thought. Most couldn’t even remember the day it is. And the idea that most have “a general view of party politics” is bizarre.

  32. Apologies, that last post was meant to be Bryan Coombe in vino veritas. Wouldn’t want to be accused of being a glove puppet.

  33. I agree with Anthony and find Jays argument hard to support. I would think the “support” question would tend to produce a higher Labour figure not a lower one. Equally the Conservative figure is in line with other pollsters. No-one yet seems to have explained why the AR polls are different.

  34. @Bryan
    “I am surprised you credit the population of the UK with any level of thought. Most couldn’t even remember the day it is. And the idea that most have “a general view of party politics” is bizarre.”

    To have such contempt fo the population you must be a professional politician!

  35. Summary of the Mori data that Anthony linked me to –

    Political ID, Jun 01, Aug 03, Jan 05, Apr 05, Mar 06, Jun 08
    Old Labour, 12%, 15%, 16%, 12%, 16%, 17%
    New Labour, 29%, 17%, 19%, 23%, 18%, 14%
    One Nation Tory, 11%, 10%, 6%, 7%, 11%, 11%
    Thatcherite Tory, 9%, 12%, 11%, 9%, 11%, 14%
    Liberal, 12%, 15%, 12%, 16%, 12%, 13%
    Social Democratic, 7%, 7%, 6%, 6%, 6%, 6%
    Nationalist, 1%, 1%, 3%, 3%, 4%, 5%
    Other, 5%, 3%, 2%, 2%, 1%, 2%
    None of these, 6%, 13%, 15%, 9%, 13%, 8%
    Don’t know, 8%, 7%, 9%, 11%, 8%, 11%

    From my point of view, the most interesting figure is the Nationalist one. Devolution was supposed to kill off the independence movements, but that’s a huge increase in those identifying themselves as “Nationalists” between 2001 and 2008 – given that Scotland/Wales only amount to c, 10% of the GB population.

  36. @pete B

    nah not a professional polititcian.

    contemptuous though I am.

  37. Well I am a politician but I don’t know about professional.

    I quite like the support question in some ways as it probably gives a better understanding of where we are in terms of the parties current position.

    I would suspect that a lot of people in a marginal Lib/Tory seat might support labour but vote LibDem or in the current climate against an encumbant because of expenses even if they support that party.

    In an odd way that means the Ar might give us a good indication of how the parties stand in terms of votes but be a poor indicator the actual result.

    That might give it poor marks in some people books as we tend to focus on who gets closest to the final result but from a campaigning point of view knowing where you are strong or weak and focusing in making supporters voters is all important.

    I’d be interested in someone doing an encumbency question as I think that “Time for a Clear Out” as much as “Time for a Change” could develope into a theme for this election.

    If Cameron is seen to push hard to get anyone tainted to stand down then he can possibly bring the too together and boost his seats tally, by getting rid of New labour and his own Old Guard.

    Although he’d have to watch being seen as the “New Tories” as the electorates feeling for “New Labour” isn’t great and might rub off the wrong way.

    Anthony,

    Any idea where we can get access to the AR tables…

    Peter.

  38. david in france.

    if you are correct and ar and william hill are wrong,you will make a fortune.good luck.

    sadly the commentry from some on this site is very odd,when you have left commentators wanting a poll taken out they dont want to see,and they are serious!

    sounds like the psbr.

  39. If you take a average of the polls over the last 6 months
    cons lead over lab is:
    July 14%
    Aug 16%
    Sep 14%
    Oct 15%
    Nov 13%
    Dec so far 14%

    So the last 6 months cons been averaging 14% lead
    on or around 40%… Not much change, 2% give and take.. Some polling companies are trying to make media capital in the run up to xmas, espeically left leaning papers ( Observer on sunday another mori poll in labours favour..wait and see??) By trying to keep it in the media they hope to influent the public. Wait until next year and the run up to the election we will be having polls like this every week…Also look back last year again some polls leading up to xmas showed cons lead over labour down to 2 points..

  40. Sadly, all the posts and threads which encapsulate the reliability and accuracy of pollstsers, are, ‘inaccurate’. We can only judge the validity of any one of the pollstsers by the last result they produce prior to the GE, provided that’s relatively close. Anything prior is pure speculation and debating fodder as they can only be measured on how close they were ‘on the day’. If AR etc., are the closest poll on ‘GE day’, they are then promoted to ‘the best’. It is impossible to validate any figures any of them offer mid term as there is only one yardstick and we don’t know what that is until next Spring.

    But having said that, debating the subject is thought provoking and of great interest and, dare I say, great fun.

  41. Old Nat,

    Wouldn’t some of the increase in “nationalist” be for people seeing themselves as English/British Nationalists?

  42. You ask for the mystery major difference between this poll and all the others… I already set that out. This is *not* a voter intention poll, it’s a party support poll. And that is a substantial difference in the question asked, not merely a slight difference in wording.”

    Anthony has convincingly explained that that is not a big difference.

    However, bear in mind that party identification polls tend to show more people identify with Labour than vote for them. If you were correct, ARS should be showing more favourable results to Labour than the rest are.

    WRT a coalition, the problem for the Lib Dems is that their supporters (not activists) are evenly divided as to which of the two big parties they would prefer to see in government. Were there to be a hung Parliament, the sensible course for the Lib Dems would be to allow the biggest party to form a minority government, rather than to choose between sides.

  43. Barry P

    “We can only judge the validity of any one of the pollstsers by the last result they produce prior to the GE”

    Not so, the could be right on the button because of sample error or just as they say in Harry Potter “Sheer Dumb Luck”. We really can never know.

    Polling is a bit like quantum physics, you know its there but its almost impossible to measure without disturbing the measurement so you have to make assumptions that you can never be totally certain of.

    Polls as educated guess work, some more educated than others.

    Peter.

  44. @Roland Haines
    @Jock

    and one or two others who make personal comments such as these

    @DAVID IN FRANCE
    Anthony has given his reasons on this matter at least 4 times to my knowledge. Is there any significance in the fact that the people complaining about this pollster are anti – Tory and much happier with news of Labour recoveries ?

    ———-

    Well, as I am actually a member of a French political party – which is considered Conservative over here – I think you, as others on here, are merely showing your own limited grasp of politics by making statements such as these.

    Facts are facts.

    And having a bias does not change that.

    Angus Reids polling appears skewed.

    Here are the reasons for my saying that:

    Fact 1: Only 5 of the last 23 polls have the Conservatives more than +14.

    Fact 2: 4 of those 5 are Angus Reid. At +17.

    That looks odd. Regardless of one’s political persuasion.

    But by all means, ignore that, and continue to make limited political comments based upon your own bias.

  45. @ Cllr Peter Cairns (SNP)

    “Not so, the could be right on the button because of sample error or just as they say in Harry Potter “Sheer Dumb Luck”. We really can never know”

    To some extent true but, if we are to find pollsters results useful, which I assume everyone here does, to varying degrees, we have to accept that the one nearest the GE outcome is ‘the best’ otherwise, they, and we, are wasting our time considering and asking the public their voting/support intentions.

    At some point we have to say ‘#####’ got it right, or as near right as can be expected. However, to refer to an earlier post of mine:-

    “But, even the opinions and decisions made then cannot be set in stone, things will change again and fluidity will reign again.”

  46. YouGov poll for tomorrow’s People 40-28-18

  47. @ Mike

    “YouGov poll for tomorrow’s People 40-28-18”

    If that is the case, YouGov’s latest offering is less than a point off their mean since the end of September.

  48. David in France. The fact that Angus Reid is showing a higher lead for the Tories (basically a lower vote for Labour) is not good evidence for saying the AR polls are unreliable. It “may” be it is more than the right the others . As well argued by Mike Smithson most of the other polls based on evidence from the Euro/May 09 LA/London Major/by-election results tended to overstate Labour. It may be AR’ s approach which includes who will you support in your “constituency” more accurately reflects potential tactical voting and the unwinding of tactical voting.

    While I am obviously not sure about the above point lets keep an open mind. Also I have to say AR have performed well in getting their results and supporting data out so speedily. Also so far they have showed more consistency than their rivals.

  49. @DAVID IN FRANCE
    You have already given us your reasons for thinking AR are wrong.
    Anthony has I reiterate, explained several times why they, (AR) might be different. Either one accepts his opinion or one has an issue with AR’s different methods. I am quite prepared to follow the opinions of Mike Smithson & Anthony Wells regarding this matter.
    I get fed up with left of centre posters who see a brave new world
    everytime a poll suggests a small Labour increase.
    Therefore I am not about to inhabit a Tory dream world myself. I am prepared to believe that Angus Read are worthy of inclusion.

  50. @ David Greybeard

    Not sure why it matters if some people are new to politics. They are entitled to an opinion and just because they dont share your pro government stance or be seasoned political campaigners like yourself doesnt mean to say that there point of view doesn’t count.

    I happen to believe that this government has not done me or my family any favours, so why should I vote for them. This may not be to your way of thinking but I dont believe I am the only one who thinks this way.

1 2 3