Following on from last night’s YouGov poll showing the Conservatives and Labour neck and neck on 36%, tonight’s again has them very close. The topline figures are Con 34%, Lab 35%, Lib Dems 11%, UKIP 11%. I rather expected that last night’s figures would turn out to be an outlier and we’d be back to a three or four point lead today, but it looks like something might be afoot.

It is, of course, conference season and 11% is at the top end of YouGov’s recent range for the Lib Dems. Perhaps what we’re seeing is a slight Lib Dem conference bounce at Labour’s expense (or perhaps the conference of one of the two governing parties has helped them both!). Either way, polls during conference season go and up down and it rarely pays to get too excited about them. The faltering lead is hardly a good start for Labour’s conference, but next week they’ll probably get their own boost from their own conference publicity.

370 Responses to “YouGov/Sun – CON 34%, LAB 35%, LD 11%, UKIP 11%”

1 2 3 8
  1. First!

  2. Not surprising given YouGov seem to have been weighting their polls strongly towards Sun/Star readers in recent days …

  3. The current publicity on Damien McBride may not help either.

  4. Oi, big mouth, I was first but I was too polite to rub it in everyone’s noses

  5. Weighting is exactly the same as it always is (and the newspaper weighting (a) doesn’t make much difference to the voting intention result, it’s the party ID weighting that makes the real difference and (b) what difference it does make, probably increases Labour support. The reason YouGov keep using newspaper weighting is that without it Labour would be too low (it’s to do with getting the balance of the type of working class Labour voters who’d read the Mirror, vs the sort of Labour voter who’d read the Guardian))

  6. Anthony

    Thanks for the explanation of the reason for newspaper weighting. I hadn’t seen that before.

    Mind you, when you use it in Scottish polls, I’m not sure that YG understands the press here. You simply can’t make these equivalences.

    “Guardian / Independent/ Herald
    FT / Times / Telegraph/ Scotsman”

  7. Anthony, out of interest how far out from the GE would you make a result prediction ?

  8. Oldnat – the important bit is really just tabloid vs broadsheet. Without controlling it you get far too many Guardian, Indy etc, far too few Sun, Mirror, etc.

    Beyond that though things actually fall out naturally enough (we could probably get away with simplifying it a bit. Putting it right down to redtop/bluetop/broadsheet though is too simple, tested it when we last reviewed the newspapers and it wasn’t quite good enough. Probably because the Sun and Mirror need doing separately. I suspect Sun/Star, Mirror, Blue tops, Broadsheets, Other, None *might* do the job… but for now this is where we are. As I’ve often said, the clock is ticking on the usefulness of newspaper readership weighting… but thus far we’ve found nothing that works as well)

  9. Marco – I try not to do it until 24 hours before!

  10. @AW

    Do you make any adjustment for false recall in newspaper readership (or even deliberate misrepresentation)?

    My first inkling of the importance of false recall was a study many years ago, when people outside a newsagents were asked what newspaper they read, and the interviewer then observed what newspaper they were carrying when they emerged. Some people who apparently said that they read the likes of the Guardian/Indy/Times were then observed emerging with the likes of the Mirror/Sun etc.

  11. Hi Anthony,

    Thank you for your reply and your great WebSite.

    …I am sure that you already have a view, even this far out ;-)
    Surely some sort of result prediction could be extrapolated from similar mid-term points in the electoral cycle or have these assessments always proved ineffectual ?

  12. A mini-bounce for the Lib Dems, and a 1 point lead for Labour. I think the LDs will be fairly happy with that.

    PS Did anyone see Harriet Harman on Question Time? Very keen to rubbish the Lib Dem record in government (yawn, yawn, heard it all before) but also extremely careful to keep the door to a post-election deal wide open. Interesting times…

  13. Phil – given the targets are ultimately derived from the National Readership Survey, which is also a survey, no. If there was an issue, it would probably effect both and they’d cancel out!

  14. Clearly a large number of voters have defected from the LDs to Labour since May 2010. But if Labour’s share has only increased from 29.7% to 35% it must mean that Labour has lost a significant number of voters in other directions: otherwise they would be on considerably more than 35% IMO.

    How would others explain this paradox?

  15. Anthony

    Thanks – though unless you’ve actually done such a trial on the Scottish press readership, separately from the GB pattern, I’m not sure that you can say it works here.

    We’ll never know though. :-)

  16. Regarding the last predictions contest, my motivation to initate one came from the fact that a couple of weeks back I was getting a bit cheesed off by those here who were in the habit of jumping on the odd poll to claim that a solid Labour lead of 5%-6% over the previous two months was somehow eroding eroding. Quite clearly it wasn’t and the average of five polls would prove my point, once everyone had put their money where their mouth was.

    Well, things didn’t quite work out that way, did they? And given that result and the even starker result of the two polls since, will you forgive me if I ask that, if anyone does still want another contest, someone else collates all the predictions?

    Mind, there doesn’t seem too much point this week because so far it looks like, of the few who did offer a new prediction, everyone will be miles over.

  17. Maybe I’m misunderstanding sample control, but to use a base which is in an industry which is rapidly changing due to internet, is it a justifiable control measure. For example, I have stopped purchasing my daily, and after daing with free metro type paper now get modt if my news from the web. I find myself reading multiple sources often from papers which I wouln’t have ever bought.
    Just a thought !

  18. apologies for typos

  19. Slightly disappointed that it wasn’t



  20. @Andy JS

    “How would others explain this paradox?”

    My guess (it is just that) is that some Lab voters have gone to DK until they see the Labour conference. They might be ex Lib Dem folk who are wavering. They might be folk who are starting to realise that the economy isn’t as bad as it was (or might have been, or will get a little better than worse, if that all makes sense).

    The 36/36 poll has the DKs at 17, while the previous poll of 33/37 had DKs at 16. No big change there.

    However, the WNV count in the 36/36 has been reduced from previous poll’s nine to six. Perhaps this is part of the answer? Only more polls will tell us.

  21. @Greg

    I don’t buy newspapers ever. I think the last paper I bought regularly was the Times, during the first gulf war, due to its detail on the unfolding drama.

    CNN stole the show though, and since then, I get my diet of daily news from websites (I never use paywall sources though). If they all went paywall, I would just use the BBC / ITV / Sky sites.

  22. It’ll be interesting to see the details of this one. Someone pointed out the large difference between the weighted and unweighted VI totals in the 36-36 poll (Con +11%, Lab -8%). While the whole point of weighting is to get things back to ‘normal’, having to make comparatively large adjustments (it’s usually only 1-2% at most[1]) suggests that there’s something odd with the underlying sample.

    [1] At least for Con and Lab. There is a consistent down-weighting of UKIP which presumably is a reflection of UKIP’s over-representation on internet panels (compare how much lower telephone polls report UKIP VI)

  23. @ Billy Bob (from the previous thread)

    “Carolyn Quinn interviewed “bigoted woman” Gillian Duffy this week for the BBC. She’ll be voting Labour in 2015, but “Ed needs to pull his finger out… no one knows what Labour stands for, not round here any road.”

    Incidentally, Gillian admitted that Gordon thought she swore when she said “flocking here”. In the immediate aftermath of the incident she phoned her daughter to ask what “bigot” meant, her daughter diplomatically told her “someone who holds strong views”.”

    You know what? Gillian Duffy is still way smarter than Joe the Plumber and I presume she has better manners than most Teabaggers.

    Speaking of which, we had another incident this week in Virginia on the campaign trail. One of the local county GOP chairmen in Virginia was leading off a gathering of local teabaggers for the Republican gubernatorial candidate, Ken ‘The Cooch’ Cuccinelli. Well before the Cooch could get up there, this guy led off by making an anti-semetic joke. That race is just funny.

    It’s been some comic relief in a week of pretty awful news. This Navy Yard shooting is very upsetting. Every time this happens, the question is asked “how many more people have to die before we do something?” It’s a rhetorical question but I’m almost wondering if we can count it at some point.

    People who hear voices inside their heads should not be given security clearances and guns for good measure. Nor should we give guns to blind people. I don’t know what to say.

    The one good piece of good political news though this week appears to be that while the Teabaggers in Congress may get their way on a vote in the House but it seems less and less likely that the House is going to try to shut down the federal government and send the United States into default just to try and get rid of Obamacare. A number of Republicans, including some very conservative ones are expressing a distaste for this idea.


  24. @ Chris Lane

    Whenever you’re around, here’s some sports news that made me very happy today (stayed up late last night just to watch my team lose in a pathetic way but today we are champions). :)

    Now I’m just hoping we actually get somewhere in the playoffs.

  25. @ Old Nat

    A few threads ago, you responded to me about Braveheart and I hadn’t gotten around to responding.

    I will say this for Monday’s mass murder. I thought Iain Gray’s cousin, Vince, did an excellent job in a leadership role. He kept the city and national news media well-informed. He remained calm and focused. He didn’t give out facts or unnecessary speculation. He was solemn but composed and dignified. And he kept the news conferences orderly so that his police chief could speak, shutting down questions without being overly rude or overbearing. And that we can all be proud of.

  26. Well, I had a good sleep. Disappointed this morning ? No, we are ahead again.

    I think Labour supporters are holding back to encourage the party to come up with some ‘goodies’ at the conference.

  27. I’d also estimate that the underlying Labour lead is about 3-4%. Two polls which show 0% and 1% aren’t all that unlikely so it’s too early to say that the very strong and steady trend of declining Labour Lead has accelerated.

    But it is possible that people have returned from their holidays and noticed that the economy is picking up, concluded that Ed Miliband really is hopeless, and have adjusted their voting intention accordingly. Things will be clearer in a few weeks.

  28. @SoCalLiberal – ” …better manners than most Teabaggers.”

    I see Ted Nugent has expressed an interest in becoming GOP candidate for President in 2016… I mean we’re lucky to still have him since he promised to be either dead or in prison by now if Obama was reelected. (His comment comparing the Obama administration to “coyotes who needed to be shot”, earned him a little visit from the President’s security team.)

    I didn’t know anything about that, except I just watched Soledad O’Brien (on youtube) taking Allen West to task about his claim that 78-81 Democrats in Congress are members of the Communist Party.

  29. @” it looks like something might be afoot.”

    Looks like it.

    McBride’s book nicely timed too.

  30. Two outliers? It seems unlikely. I wonder if online-polled voters, sat behind their tablets and other PCs, feel more like ‘dabbling’ with their VI than others?

    If one contrast the fieldwork method with phone polling, the phoned voter does not have time to ‘dabble’ but is confronted out of the blue with the question.

    Marco – I try not to do it until 24 hours before!

    -I normally settle for 24 Hours afterwards!

  32. The 60+ group shows a massive improvement for blues considering that Con & Lab were neck and neck in this group not so long ago.

    Con 44
    Lab 22
    LD 10
    UKIP 16

  33. Ozwald

    It seems improbable that the Conservatives are suddenly twice as popular amongst the over 60’s as you point out with just a 4% higher lead for Labour they were level pegging.

    However the over 65’s is prime Tory territory both for votes and membership

  34. Labour remain in the 36-40% range, barring outliers.

  35. I wonder if Nigel Farage’s alleged racist behaviour as a Teenager combined with the alleged Nazi Songs He was supposed to sing will have any impact on the UKIP VI

    Mr Farage vehemently denies the allegations.

    Personally I suspect not Farage appears to be One of those Politicians ,much like Boris Johnson whose personal behaviour is disregarded because they are a cheeky chappy!

    Can any one imagine the gusto with which the Media would attempt to demolish Miliband ,Clegg or even Cameron if similar allegations were made against them?

  36. @Bill Patrick –

    or –

    Labour have moved to the 36-40% range, barring outliers.

  37. There is of course no identified reason for the narrowing, but if the ST YG poll was to repeat it, then we baby boomers returning from unnecessary holidays may well have been an influence in the feel good factor working its magic!

    I wonder, Ozwald (and Steve), if these age cross breaks are not misleading? Was it the same pattern the poll previous?

  38. Morning Everyone,

    Well I honestly didn’t expect another close run.
    A tie on 36% followed by a 1% Labour Lead is quite something.
    Could all be reversed of course by the end of next week and at the end of the Labour Conference.
    But at least the polls are a bit more interesting now and liven things up a bit.
    Of course the big one will be when the Cons overtake Labour even by just 1% it will be interesting to watch the dialogue on here -lol

  39. I don’t think we can start questioning YouGov methodology.

    Maybe it is just being on this site and the regularity of the polls but I’ve always considered them to be the best, mainly because of the panel that clearly identifies past voting intention. Also they seem to be middle of the road in their results as compared to ICM on one end of the spectrum and TNS on the other.

    What is slightly odd is that ICM had a bigger Labour lead earlier this week and also Populus have shown rises in the Lab lead (albeit just two polls).

    If the Lab 35-36% is broken then it’s time for them to worry. This has held so far but I’m very surprised that a year on from expecting a Lab landslide we are now definitely into hung parliament. All will depend on whether the economy continues to grow or whether this is just more green shoots that fizzle out.

  40. @Shevii,

    ICM and the populous polls were really before the big shift in Yougov happened. Will be interesting to see what happens to them next. I expect some movement downwards.

  41. It’s house prices, there has been so much news about rising house prices. All the other economic indicators don’t matter that much but folk have so much invested both financially and emotionally in their houses that it will affect VI strongly

  42. I think AW makes a pertinent point. The Tories have got a boost from the LD conference. I suspect this was due to the universal free school meals policy. However, if the,Coalition is expecting Councils (who have no money) to pay fo it, it may never happen.

  43. @Howard
    Recent polls show wide VI variations in 60+ group and we are cautioned against reading too much into them. Still I can’t resist peeking! I do listen carefully to comments from my circle of friends and colleagues and I detect a slight warming towards LD, no enthusiasm for blue or red, and a noticeable rise in those who ‘UKIP for me next time’.

  44. I am not sure we can trust the opinion polls at the moment. We recently had Lord Ashcrofts polling of marginals showing Labour with a significant lead. Last night Labour won a council seat in Oxford from the Lib Dems, the same seat they came third behind the LD’s and Tories in 2010. Also UKIP won a council seat from the Tories in Canterbury last night.

    I suspect that around the country Labour and UKIP are doing much better than the polls currently show. Why YouGov are not picking this up, I am not sure. It would not surprise me if other polling companies showed different results over the next week or so.

  45. R Huckle,

    The opinion polls are flawed, except Ashcroft’s, and council seat results are good guides to national trends?

  46. Will be interesting to see where we are after the conference season.

    Govt borrowing down today, so more pretty good news.

  47. In this weeks local elections

    The trend of UKIP as the second choice party when one of the main parties is strong continues – 1st in Kent where they seem to be growing in strength and where Farage may be looking for his first mp, second in Hampshire, Dudley East, where one of the main parties has most of the votes, so the minority conservative or Labour voters seem to be tactically voting UKIP.

    Another bad week for the Lib Dems, the comments by the activists in the thread are also interesting as they give a sense of what is happening on the ground.

    Seems the Greens are expecting to pick up a lot of the old Lib Dem student vote. I see the greens are also strong in the 18-24 cross tab today, I think the student vote should be easy pickings for them, but it doesn’t look like that is enough to take whole seats yet. But if the Lib Dems lose their student vote to the greens that could make for some interesting elections in places with a high number of students.

  48. “@ Bill Patrick

    R Huckle,

    The opinion polls are flawed, except Ashcroft’s, and council seat results are good guides to national trends?”

    Not necessarily, but Ashcroft was particularly looking at the marginals, so these are important. Other polls may only pick up opinion from a small number of people living in these marginals.

    In regard to council seat results, these can be important, as these are actual results from people who could be bothered to vote. This is not to say the same result would be found in a general election, because of the lower turnouts for council seats, plus there may be different issues i.e local and not national.

  49. I’ve kept my own poll-of-polls (thanks to Excel!) since 7 May (ie 2 years before the G Election) on a 7-day rolling basis. It’s probably not done quite how a ‘real’ statistician would do it!! I just take off any fieldwork more than 7 days old when I do the next day’s. Most fieldwork is spread over 2+ days and I distribute the results evenly on a day-by-day basis. Gives a sample of 10000+ usually.

    Anyway it does iron-out small variations and outliers don’t have dramatic effects on the results. Over these 4+ months, CONs have increased their share of the polls (29% risiing to 33%) and UKIP have clearly lost some of theirs (16% falling to 11%).

    Labour and the Lib-Dems have comparatively stood still, Labour wavers around the 38% mark, LibDems around the 10% mark. So Conservatives have gained, not at Labour’s expense, but at UKIP’s, by my measurements for what their worth.

    With the UKIP conference starting it’ll be interesting to see if the situation changes, tho’ I do agree that there’s not much point in taking great note of polls til the conference season is over.

  50. I believe we hear from Populus again today and Opinium should be in Sunday’s Observer. It will be interesting to see whether a similar trend is picked up by them.

1 2 3 8