Time for a round up of Sunday’s polls, with new stuff in the Sunday papers from YouGov, Survation, Opinium and ICM.

YouGov’s weekly poll for the Sunday Times has topline figures of CON 33%, LAB 33%, LDEM 7%, UKIP 16%, GRN 6%. The poll also asked a series of questions about how people would vote with different Labour leaders. In a control question asking how people would vote if the leaders remained Cameron, Miliband and Clegg the answers were CON 33%, LAB 31% (so the effect of reminding people of the current party leaders still seems to produce a slight positive Cameron effect or negative Miliband effect). If Yvette Cooper were Labour leader the position would be the same, a two point Conservative lead. If Ed Balls was the leader it would be worse, a three point Conservative lead. In contrast with Alan Johnson as leader Labour would be two points ahead (CON 31%, LAB 33%.

I’ll give my usual caveats about questions like this – people are answering them when on very little information, they don’t know what policies or priorities those alternative leaders would set, how the media would react to them and so on. In the same poll, YouGov found that only 42% of people think they could recognise Yvette Cooper from a photo… if you don’t even know what Yvette Cooper looks like, I’m guessing you don’t have a thorough understanding of what she would prioritise as Labour leader. It’s a response based on a very crude impression of those potential leaders based on what tends to be the very limited public awareness of opposition politicians. Nevertheless, those crude first impressions count, so it’s a good sign for Alan Johnson.

Survation also had a new poll with topline figures of CON 29%(+2), LAB 34%(+3), LDEM 6%(-3), UKIP 23%(-1), and they too asked a series of hypothetical voting intention questions with different Labour leaders. In the Survation poll they displayed a biography and played a video clip of each potential leader and asked people questions about them before the questions. This allowed them to include people with extremely low public awareness like Chuka Umunna, though does of course rely upon the choice of biogs and video clips (given bias is often in the eye of the beholder, choosing clips that even those who don’t like the eventual results think are fair is incredibly tricky). The control question with Ed Miliband had a Labour lead of 4 points. In the Survation poll Yvette Cooper did worse than Miliband (neck and neck with the Tories), Andy Burnham just the same (4 point lead), Alan Johnson and Chuka Umunna did best – both extending Labour’s lead to 8 points. A voting intention question asked after video clips of Labour leaders is obviously skewed towards Labour, but it’s the relative performance between the different leaders that counts, and again it’s good for Alan Johnson, and now also for Chuka Umunna.

Meanwhile the fortnightly Opinium poll for the Observer has topline figures of CON 29%(-4), LAB 32%(-1), LDEM 9%(+3), UKIP 19%(+1).

Finally there was an ICM poll in the Sunday Telegraph. As usual with ICM/Sunday Telegraph polls, this asked the public to predict vote shares rather than ask people how they would vote themselves. The average response now has the Conservatives getting slightly more votes than Labour.


281 Responses to “Sunday polls and alternative Labour leaders”

1 2 3 6
  1. All part of the depressing media campaign, I suppose. [snip]

  2. The choice of clip used could be an uncontrolled skew.

    I’m very dubious of this methodology.

  3. Two new polls put Labour ahead by 5 points and 3 points. With Ed as leader it seems Labour looks likely to be the largest party in the GE. Why would they want to change leader?

  4. “YouGov’s weekly poll for the Sunday Times has topline figures of CON 33%, LAB 33%, LDEM 7%, UKIP 16%, GRN 6%… In a control question asking how people would vote if the leaders remained Cameron, Miliband and Clegg the answers were CON 33%, LAB 31%”

    Who do the 2% think is leader now then?

  5. Rather a large variation there then. I found the headlines in my newsagents this morning, all very predictable. The red tops concentrated on Shelby Perkibody, sleeping with 12 soccer players.
    The Mail is intent on seeing the blood of a nice Jewish boy from Hampstead, all over the floor. The Telegraph was far more circumspect. Clearly believing that Mr Miliband is the Tories best friend. The prospect of a working class, trade union educated, highly personable postman, leading Labour, does not make Tories happy bunnies.

  6. Charts updated folks. Some nice stuff going on today, and the thirty-poll charts’ last data point is a full thirty.

    http://www.statgeek.co.uk/polling/periodic-averages/thirty-poll-averages/

    Some factoids in no specific order:

    – Miliband’s leadership ratings at an all-time low in Scotland at minus 72.
    – Conservatives seven points ahead in the Midlands.
    – UKIP on 21% in RoS.
    – Conservatives’ Scottish VI slips for 4th month in a row.
    – UKIP up approximately 4% in all parts of the UK (except Scotland – down a fraction) over the last three months.
    – Conservatives down 1-2% everywhere, except London, where they’re up a fraction
    – Greens up everywhere and down in Scotland (what’s that all about?)

    And finally – Miliband and Clegg set to slog it out for least popular leader, while Cameron’s M&W rating hits a positive value.

  7. @ Norbold,

    I suspect they weren’t thinking about it at all. Putting the leaders’ names into the question concentrates the mind on that aspect of the party identity.

    Of course, the leaders’ names won’t be on the ballot papers either unless the Tories pull an SNP and change their party name to “Cameron for Prime Minister”, so I don’t know how accurate it is as a VI test- presumably if that wording gave better responses YouGov would use it all the time.

  8. @DAVID BUTLER
    My question is, are you the David Butler who was the “Labour Man on the Telly” .

    My answer to your question is, because a large proportion of people, think he is (politely), not up to it.

  9. I have to say, I’m impressed with how few of the YouGov panel fell for the Andrew Farmer trick!

  10. Spearmint – “presumably if that wording gave better responses YouGov would use it all the time.”

    Exactly, and we don’t.

  11. The pollsters should include an imaginary candidate as a control. It would reveal what a nonsense it all is.

  12. @Roland

    But not enough to stop Labour winning the GE.

    The premise of the Survation report and tables is that under EM Labour would only win a majority of 40 seats when they could be doing so much better. Better than a 40 seat majority?

    [I expect (without putting thoughts in the mouth of some hypothetical “Labour party MP who thinks Ed should go”), that it would be because they are looking at the broad trend across all polls, not picking out an outlier. You know better than that – AW]

  13. No, Rolandgatinois, I am not the psephologist David Butler, though when I was an undergraduate I received an invitation to lecture on British elections at a Californian university. Sadly I lacked the nerve to accept.

  14. Labour List have launched a “Put up or shut up” petition available for all to sign here:

    http://labourlist.org/2014/11/anonymous-labour-mps-seriously-you-need-to-put-up-or-shut-up/

  15. The Ed Miliband story is ridiculous. I can understand [NewsUK] running with it – but the left leaning press should know better. There is no mechanism in the Labour Party rules that would allow Miliband to be deposed prior to the GE. & A guy that wrestled the leadership from his brother must have very strong convictions. He is currently on course to be PM why on earth would he give that up.
    The Cons are the ones with the real issues 29/27 in the polls, a poor by-election showing, and the EU bill fiasco. The media and the BBC have lost their grip on reality.

  16. @Statgeek
    The Greens in Scotland are down probably because pro-Yes parties & voters are planning on voting SNP (lending their vote) in May
    Report comment

  17. @Phil Haines

    Perhaps I’m biased, but I see that article as “Come out and challenge Ed, or shut up or the Tories will win.”

    There’s still the knotty subject of Ed actually having leadership skills, and Labour having election-winning policies (and I would apply that to any or all parties as the election approaches).

    If these ’20 Shadow Ministers’ are real, then Ed really is in deep trouble:

    http://en.wikipedia.org/wiki/Official_Opposition_Shadow_Cabinet_%28United_Kingdom%29

    “Not all Opposition frontbenchers are members of the Shadow Cabinet, which is composed of the most senior Opposition Members (usually around 20).”

    Smoke and mirrors, methinks.

  18. Interesting YouGov this morning. Top line figures show Cons and Lab neck and neck but when asked how people would vote in 2015 if the same leaders were in place then the Cons are 33% and Lab on 31%. So really the Tories have a small lead in the latest YouGov. This is backed up the ICM Wisdom Index which also puts the Tories ahead. Two other polls put Labour firmly in the lead.
    The papers still full of EM crisis talk, probably much of this is rubbish, Certainly EM is unpopular with YouGov showing that even among Labour voters he has minority support.
    On the EU extra payment question the only polling I have found shows 42% think it was “a result” for Britain with 33 against. So probably no damage for the Cons on this issue, maybe even a slight positive.

  19. Wow. 27/29 figures are very dismal looking for Tories. That’s an even lower percentage than their ’97 wipe out and one of the lowest if not the lowest figure for an incumbent government as far as I can remember.

    Couple more of these polls and I expect more defections to UKIP and possibly UKIP overtaking the Tories in the polls.

  20. Sorry for the duplicate comments. There was a glitch.

  21. @Spearmint – Some proportion of online respondents answering “knowledge” questions Google the answers. I remember in particular a question I ran once, asking if people knew which brand owned .com, only to have about 10% of the respondents name some obscure American media agency. When I Googled “who owns .com” the first result was a Wikipedia page, and Google’s text said “-channel.com is a website owned by …”. So when I see knowledge questions for obscure things, like Vernon Coaker, I automatically knock the numbers down a bit.

  22. @Spearmint – Some proportion of online respondents answering “knowledge” questions Google the answers. I remember in particular a question I ran once, asking if people knew which brand owned (name).com, only to have about 10% of the respondents name some obscure American media agency. When I Googled “who owns (name).com” the first result was a Wikipedia page, and Google’s text said “(name)-channel.com is a website owned by (American media agency)”. So when I see knowledge questions for obscure things, like Vernon Coaker, I automatically knock the numbers down a bit.

    (That’ll teach me to use angular brackets…)

  23. @TOH

    Do you really have any faith in the ‘wisdom’ index?

    Surely that is no more than 1000-2000 people guessing how everyone else will vote?

    I’m sure everyone has an opinion, but I’m also sure that this opinion is worth bobbins, plus a minus a smidgeon.

  24. @MacTavish – Given that the “incumbent government” is a Conservative and Liberal Democrat coalition, the relevant comparison figure is the Conservative + Liberal Democrat total. Otherwise, one could as easily describe the Lib Dems’ lows of 5% as “the lowest figure for an incumbent government”.

  25. @CatManJeff – The literature suggesting that wisdom indices are accurate predictors of a binary result is quite plausible. Here is an interesting MRS paper on the subject:

    http://www.mrs.org.uk/pdf/IJMR_54_%284%29_Boon.pdf

  26. @Chris Green

    Thanks.

    I’ll read that later :-)

  27. @RAF
    Far be it from me to expand on AW’s comment, but yes exactly.
    Please, lets be realistic, even if your sympathies are left of centre.
    To quote the Iron Duke on this armistice Sunday, “its a damn nice thing, a damn close run thing”.

  28. Catmanjeff (may I call you CMJ?)

    I’m sure Anthony has warned us many times that people don’t always successfully predict their own behaviour let alone other people’s.

  29. CATMANJEFF

    Why not, it is actually asking what people think will happen in 2015 not what they think now. That is more interesting to me as are the scientific groups attempting to forecast the result in 2015.

    It seems to be in tune with YouGov at the moment which I find encouraging. Are you suggesting the opinions expressed in the YouGov polls is also ” worth bobbins, plus a minus a smidgeon” as you put it?

  30. @Bramley

    CMJ is fine :)

  31. @Bramley – And yet, wisdom indices demonstrate good accuracy. One hypothesis is that the effective base size is much higher because one is accessing not only the 1,000 respondents, but also, indirectly, the 1,000 respondents’ friends and family.

  32. CHRIS GREEN

    Interesting and not surprised.

  33. Chris Green

    But gathering of public estimates of things like immigration statistics, level of JSA payments etc show the public regularly give answers that are way off the scale.

  34. @TOH

    I don’t doubt YG polls in the slightest.

    I’m interested in the methodologies involved and the Wisdom Index seems superficially to be getting ordinary folk to guess what is an ‘unknown’.

    This is why polls asking people about what percentage of welfare is spent on x or y for instance, are interesting. Most people don’t have a clue about the real value, but that doesn’t stop them making a guess (often very badly wrong).

    In market language, people aren’t answering on partial knowledge, they are mostly answering on zero knowledge.

  35. The Morning Star hails Ed Miliband as the saviour of The Labour Party, while the TG is running 105 seconds of Ed Miliband gaffes (find the links yourself).

    I have a feeling it’s more a case of the media are trying to drive the outcome of the election, long before it happens. It just shows that the media are financially interested in election outcomes, as much as any business, and have no faith in the electorate to make the decent decision.

    On another note, I spotted one Tweet today:

    “Bit strange to see Labour supporters complaining about media bias now, after staying silent throughout the indyref.”

    Leaving aside my own opinion on said bias or lack of it, is it just a case that supporters of all political parties or causes only moan when their side is being attacked or when their opponents are not being attacked?

    Ergo, are all politically active people hypocrites? Probably :))

  36. I guess that if people absorb snippets on polls from newspapers and the like, the wisdom index may be them basing a VI percentage on that.

    Therefore, it is value based on weighted polls anyway.

  37. CATMANJEFF

    As I say I am much more interested in what people predict will happen in 2015, than I am in how they would vote if there was an election today. You really should read the paper Chris Green referred to.

  38. That Observer story about “20 shadow ministers” might at face value be harder to dismiss than an anti-Labour smear story from the Daily Mail.

    However, it turns out that the Observer journalist who wrote it was previously employed writing smear stories for the Daily Mail.

    http://tompride.wordpress.com/2014/11/09/why-is-the-observer-employing-a-mail-journalist-to-smear-ed-miliband/

  39. @Statgeek

    As a Green, I don’t what media attention is, positive or negative!

  40. Correction

    @Statgeek

    As a Green, I don’t know what media attention is, positive or negative!

  41. Statgeek,

    “Leaving aside my own opinion on said bias or lack of it, is it just a case that supporters of all political parties or causes only moan when their side is being attacked or when their opponents are not being attacked?”

    Pretty much yes.

    “Ergo, are all politically active people hypocrites?”

    No, but all those political active people who strongly take sides (supporting a party, a campaign etc.) are likely to suffer from confirmation biases and other framing effects. I wouldn’t call that hypocrisy, so much as just humanity.

  42. I note that ElectionForecast.co.uk has now joined Elections Etc (Fisher) in forecasting that the Cons will have the most seats in 2015. For the last couple of weeks they had been showing Labour with the most seats but that has changed with today’s forecast.

  43. @CMJ

    Well some Greens tend to get media coverage during [insert anti- non-renewable energy protest].

    @Bill

    “I wouldn’t call that hypocrisy, so much as just humanity.”

    Ah ha! Now we’re into the realms of “Is it more human to ignore or recognise one’s built-in political bias?”

  44. @CHRIS GREEN

    “…The literature suggesting that wisdom indices are accurate predictors of a binary result is quite plausible…”

    Serious question: how do you measure the error of a wisdom index?

    If 50% think Dog Party will win the election, and 30% think Cat Party will win the election, and 20% think Mouse Party will win the election, and the Cat Party actually wins..then what is the error expressed as a number? I know how to measure the error[1] of a conventional poll but I don’t know how to measure the error of a wisdom index.

    [1]: one way is the square root of the sum of the squares of the differences between each predicted number and the matching actual number. There are other ways.

  45. Catmanjeff

    What would be interesting would be to see how the expected vote share was distributed. I suspect it won’t be a nice normal distribution curve with a standard error of 3%. It’s possible it’s be a two humped beast with partisans of both sides adding a few percent to their side and taking off the other. I’d expect a whole bunch of crazy outliers as well.

    Wisdom indices when people are regularly told about the measured answer are going to be more accurate than those where the answer is not commonplace.

    For political polling it may give reasonably accurate predictions as most people have some level of understanding of the positions right now, try it with asking people “what do you think the share price of these 5 stocks will be in April 2015” and see how it works out.

    The reason that these wisdom measures predict close to the polls is people are using the polls as a basis for their estimate to a greater or lesser level.

    If you created a trial and isolated a group of people and fed them fake polling data at the same level as the real news for a month, suggesting party x was on 40% and on for a landslide. At the end of the trial, ask for a wisdom index from them you’d see a great influence in their views.

  46. Back in the day, the Tory membership liked IDS, but the Westminster party did not. He failed.

    Is this the same sort of deal? The trade unions liked EM, but the party preferred DM. EM won the leadership but, I suggest, will not win the war. Especially now, as his perceived weakness will damage him.
    It will be very interesting to watch Cameron’s attitude towards EM. Because about the only person he wants to hang on to more than Ed,
    is Samantha.

  47. Toh

    The reason that electionforecast has drastically changed it’s prediction is due to changing how it handled the Scottish polls, before they ignored them as they had no way of incorporating that data and now they’ve shoehorned that data into their model.

    It’s hard to have confidence in their model when they keep altering their methodology so much due to errors, it’s not a mature model yet but appreciate what they are trying to do. They really needed to iron out the bugs before going public with predictions.

    Also they use the 2010 election data to both create the model and as a check to make sure that their model predicts the 2010 election. That is a pretty big nono when producing data based models, crosschecking data should be isolated from data used to form the model.

  48. Alan

    I agree with most of that, but they are at least trying to take account of what clearly is a “moving feast”. They are asking for comments so I for one am pleased they have gone public with their predictions.

  49. I think this BBC article summarizes the situation well

    http://www.bbc.com/news/uk-politics-29967737

    it mentions this mornings papers, but not too much, records the Labour rebuttals and moves the story on to what ED M is going to do next.

    Some Lab supporters were complaining about the initial BBC coverage of this story, but this article seems fine

    “Labour backbencher Simon Danczuk told BBC One’s Sunday Politics Mr Miliband would lead the party into the election, but added: “The numbers show us he’s not popular. That’s the reality.”

    Simon Danczuk was mentioned in the Telegraph as one of the plotters, but he has obviously recanted (or clarified)

    Ian Austin was the other one, I wonder where he has gone, perhaps the Lab hierarchy have had a word with him.

  50. Sloppy journalism in the Observer editorial today. Twice it refers to ‘barely seven months to the election’, Can these people not count?

1 2 3 6