This morning’s YouGov poll for the Sun has topline figures of CON 35%, LAB 39%, LDEM 8%, UKIP 11%. The four point Labour lead follows a lead of seven points yesterday and three points on Monday. Normal caveats apply about taking a single poll out of context – a four point lead is clearly towards the lower end of YouGov’s current range – but the underlying average does appear to be falling. Full tabs are here.

While I’m here, I’m seen comments on twitter getting excited/concerned about today’s poll showing the Conservatives ahead amongst under 25s. This is something I wrote about last month. The brief version is that age cross breaks are only small so have large margins of error, especially for under 25s which tends to be the smallest age group with the largest proportion of don’t knows or wouldn’t votes. This means figures for under 25s are extremely volatile, and will swing about wildly from day to day. Taking just one unusual looking one is extremely misleading! Looking at the trend in YouGov’s recent polls it is very clear that, on average, Labour still enjoy a solid lead amongst under 25s.


225 Responses to “YouGov/Sun – CON 35, LAB 39, LD 8, UKIP 11”

1 2 3 4 5
  1. Latest YouGov / The Sun results 25th July – Con 32%, Lab 38%, LD 11%, UKIP 11%; APP -27

    Oh someone’s beaten me to it.

    Labour seem surgically attached to those high 30′ ,the polish seems to be coming off the UKIP

    Real wage levels have now fallen back to that they were under the Last Conservative Government and the UK average incomes continue to fall compared to other industrialised countries it is a very odd kind of recovery when for the majority it is accompanied by falling not rising standards of living.

  2. Another odd Age Cross break with the only age group where Tories in the lead the under 24’s.

    I think YouGov seriously needs to look at their survey process for this as I don’t believe it is credible .

  3. @Jim Jam –

    “With 3 x 39% and 2 x 38% in this weeks YG safe to say Labour between these 2 levels”

    around 38.6

    “but the Tory score harder to judge with 3 x 32% and 2 x 35%?”

    around 33.4…

  4. AC

    “I’m sure he is a really nice person”

    ………………………………………………………………………………..

    My pups certainly think so.

    I’m up early today and am looking forward to your political wit and wisdom.

  5. wes

    “around 38.6”

    Can you be more specific – perhaps to three decimal places?

  6. I think 6 points is about average for Yougov at the moment.

    I would still say Labour are favourites to at least win the most number of seats (and, hence, form the next government) in 2015, but the Tory’s chances have been boosted in the last couple of weeks when compared to before.

  7. The Yougov age cross breaks are totally unreliable and volatile. I would be extremely surprised if the under 25s have suddenly become Tories!

  8. ‘What it seems Labour should do is keep trying to drive the wedge between those supporters over things like Europe’

    xxxxxxxxxx

    err no. They should try some positive politics and come up with some policies.

  9. Wes – crude averages do not tell us if we just happened to have had 2 in a week at the higher end of moe for the Tories as they have only had one other score above 33% this month (34% on July 3rd)).

    It could be that there has been a further modest improvement in Tory VI in the last week but 2 polls imo is not enough evidence. Whether favourable to Labour or not I have a 3 YG polls required in a row rule of thumb before believing a shift has occurred.

    I may be accused of bias but I would place the Tories on YG at 33% rather than 33.6% which is still an encouraging improvement in recent weeks.

  10. 38.6% v 33.2% over five polls gives a lead of 5.4%, but until we see some more 35s from the Conservatives, we have to assume top end MoE.

    Mind you that’s less than +/- 2% either way.

  11. It’s suddenly hit me that Labour aren’t going to win the next election.

  12. Bill C
    “Labour has peaked and Scotland is off”

    -Labour at 37% anti yes vote parties on 76% of the share.

    So precisely where is Scotland going to?

  13. @Ambi

    “I would be extremely surprised if the under 25s have suddenly become Tories!”

    I don’t tend to look at age spreads very often, so am not asking this from a partisan or knowledgeable standpoint; more a ‘devils advocate’ position.

    I’ve seen that comment or similar ones from time to time on here, and at what point do folk accept that a typical age trend has changed (be it blue rinse folk going predominantly Labour, or students predominantly Conservative?

    Or is it one of those things that folk believe will not happen (perhaps with reasonable grounds)?

  14. @Steve

    “So precisely where is Scotland going to?”

    The Seychelles…late September in fact. We’ll all come back in April though, in time for elections.

  15. @Jim Jam,

    As ever, a sensible and balanced post.

  16. thanks Statgeek I did not spot Wes’ arithmetic error and compounded it with my error saying 33.6 rather than 33.4.

    I think we are on the same page in that we need more evidence before thinking 35% or even a higher average due to the 2 x 35% applies’ whilst Lab with 5 within 1 point is clearer.

  17. @StatGeek,

    “I’ve seen that comment or similar ones from time to time on here, and at what point do folk accept that a typical age trend has changed (be it blue rinse folk going predominantly Labour, or students predominantly Conservative?”

    Good question. I don’t think there is a specific rule that should be followed, but the trend should be sustained over many polls. Also, I would want to see the same trend across more pollsters than just Yougov, especially as its cross-breaks are not meant to be taken as totally reliable.

  18. Stat Geek
    So that’s why there were an extra 5 million people having dinner at the Domaine Du Soleil Restaurant Probably why we had to wait for a seat.

  19. Nice to see Welby is trying to ensure the Church does something positive in fighting exploitation of the poor and desperate.

  20. @Paul – sorry. Around 38.600

  21. So we wait to see whether two periods of growth, summery weather and royal/sporty shenanigans affect the polls.

    The economy is/was always the key. But is reported growth going to help without improved living standards? Or will standing still be enough for most people? What will happen come winter if heating bills and petrol and food keep increasing, and wages don’t?

    Watch this space…

  22. “It’s suddenly hit me that Labour aren’t going to win the next election.”

    Even very large polling leads during a parliament are no guarantee of an election victory.

    As yet, Labour are nowhere near giving the impression of being a potential government-in-waiting.

    The same could have been said about the Conservatives in 2010, and they didn’t get a majority.

    The difference was that Cameron to some extent out-performed his party. Essentially Cameron applied for the post of PM and he got it.
    In effect Labour left the position vacant… by viewing Brown as not an asset, not to be promoted, absent from any campaign literature.

  23. No error, 33.2 *is* around 33.4…. I was just going for a number a bit shy of 33 and a half to be honest. The point I was really trying to make is that those Con scores are no more volatile than the Labour ones really, they’re all within margin of error.

  24. Statgeek

    It’s the “suddenly” bit that’s suspicious. If you add today’s 38-29-16-9 to the list of recent cross-breaks for the under-25s I put up at 4:04 yesterday on this thread, those last two look very odd. It’s probably just coincidence, but I suspect that YouGov wil be looking to see if there have been any changes in methodology.[1]

    Of course it doesn’t mean that the headline figure is wrong – it could be balanced by the slightly unusual Labour lead in the over-60s for example. But you do get a slightly Lady Bracknell feeling that “two looks like carelessness” even though it’s probably chance.

    It would also be interesting to see if polls tend to be more variable in the summer months, possibly because of the lower prominence of politics or people being away on holiday.

    [1] The other possibility is infiltration, particularly of people claiming to be from hard to reach groups such as male, under-25, DE, Sun-readers. But the YouGov panel should be big enough to cope and I think they have safeguards to prevent anyone being asked too often.

  25. fair enough Wes.

  26. The economy is/was always the key. But is reported growth going to help without improved living standards? Or will standing still be enough for most people? What will happen come winter if heating bills and petrol and food keep increasing, and wages don’t?
    Watch this space…

    Standing still would be a distinct improvement for most,

    -Real incomes (except for the top 1%) have undeniably fallen and will continue to do so by policy in relation to public sector workers and by the increase in the proportion taken by the wealthiest in the private sector.

    We will be back to real income levels last seen in the 1980’s by 2015 and the UK average income is going down by international comparisons.

    Tax revenues are static or falling and wide scale legal corporate tax avoidance remains the norm nice example with Formula 1 Yesterday how £375 million UK based profit generated less than £1 million in Corporation tax .

    I think Balls was right to shift the emphasis onto income and fairness from growth because despite the fact that growth and all other Government targets have been undershot substantially the media are prepared to portray this as some sort of success.

    It will be interesting to see if this message gets across in VI

  27. On a slightly more serious note, the Scottish cross breaks (caveats etc.) are interesting.The five polls prior to this week show Con on:

    18, 17, 14, 16, 16 (Average of 16.2%, which is below the MAD for the same period of 17.4%, and the 2010 results of 16.4%).

    The unweighted sample average for the five is 159, while the average for the 30 polls preceding is 150 (the samples ought to be 162 and 159 respectively if going on the 8.7% population).

    Four polls this week: 18, 21, 21, 25 (Average of 21.3)

    Sample average of 151 (ought to be 161, based on the UK samples).

    The SNP numbers:

    Previous five polls: 27.2
    Preceding 30 polls: 24.6
    This week’s four: 23%

    So what? Nominal seats from last week to this week (seat change from 2010):

    Lab 43 (+2)
    SNP 8 (+2)
    Lib 6 (-5)
    Con 2 (+1)

    Seats on the four polls this week:

    Lab 43 (+2)
    SNP 7 (+1)
    Lib 5 (-6)
    Con 4 (+3)

    Lab and Con both on the increase in Scotland, based on those numbers, which can only be bad for the SNP (it’s only four polls folks). Of course, the SNP tends to do better in recent elections, than the long-term polling suggests.

  28. @Billy Bob

    Nick Clegg was the spanner in the Conservative works in 2010. I’m pretty sure out of the 23% of Lib VI in 2010, at least 2-3% of those would have been Con voters, but for NC, giving the Conservatives 39-40%.

    Had NC been ineffective in the debates, Cameron would have come out looking good over Brown, and there would have been a Con OM.

  29. I can’t see Labour’s vote collapsing unless the LD vote recovers. And since (I suggest) we are talking about an anti-Tory vote really, that’s only likely to happen where Lab aren’t in contention.

  30. Statgeek – I reckon you’re bang on the money there.

  31. In other News Prince Harry realises babies are small

    That A Level in Art that His teachers “Helped” Him with at Eton probably came in handy with this stunning revelation.

  32. @Colin

    “An interesting view of the outcome in 2015:-”

    Old news on previous thread

  33. Statgeek

    Clegg was ineffectual in the debates he only managed to increase the lib dem vote by 0.5% which is pretty shocking given the circumstances, if he hadn’t managed to hook up with Cameron lots of lib dems would have been looking at his performance disapproving, they should never have got rid of Charlie, I bet he would have brought home the bacon

  34. Statgeek – re LDs taking 2-3% con vote in 2010 I tend to agree with you hence believing cons going above their 2010 %age at the next GE is quite possible.
    The impact on the GB v DC debate impression, though, is more speculative and perhaps less certain than you suggest. There is one view that GB was up against 2 smooth PR operators and looked uncomfortable; and, that had it been more GB v DC (with NC peripheral) the substance v style contrast may have helped GB.

    Also probably without Cleggmania Lab would have added a point or 2 as well taking them over 30. Probabaly still would have been net better for Cons but may be not enough to carry to OM; close though and imo certainly enough to have allowed at least a minority cons Government.

  35. @Steve – Cool figures pal, haven’t a clue where you got them from, not even the Bitter Together mob are claiming stats. like that. The latest official polling shows it is now neck and neck e.g. Prof. John Curtice’s (Strathclyde University) “What Scotland Thinks” website,, private polling confirms this and evidence from the hundreds of fairs, shows and Highland Games all over Scotland suggests a big swing to the YES vote. The word on the streets of Scotland is “YES”.

  36. @statgeek

    This page shows how poorly Labour performed in picking up ‘don’t knows’ during the campaign, relative to Con and LD… also the small LD base at the start of the campaign:

    h
    ttp://www.ipsos-mori.com/researchpublications/researcharchive/2410/When-Did-You-Decide-How-To-Vote.aspx?view=wide

  37. Today in 1945 the results of the general election were declared – Labour won with 393 seats and 48% of the vote.

    Those were the days. Sigh!

  38. @Laszlo
    Assumptions!
    Earlier I deliberately and a little disingenuously made the assumption that nothing had changed during the month.
    However, you take the view that what we are seeking to do is measure any change which might/does occur, then we must consider how to make a measurement with sufficient precision to detect a change of a given size.
    Let us assume that a real change of 1% in the L-C VI difference is significant. Taking eight rather than sixteen YG results provides a precision of around +/-0.6%
    If we then take the first eight, the middle eight and the last eight results in July, and treat each eight as successive measurements of the early, mid and late July differences, the results are:
    early L-C 7.5+/-0.6; midmonth 7.0+/-0.6, late 6.0+/-0.6
    Those results were obtained using the ‘difference figures’ in the last column of the table, and correlate with the complete month average of 7% obtained from the sixteen sets of Con and Lab VI shares I first used.
    There is then some evidence for the gap closing, but it is largely due to a couple of recent 35% figures for Con support.

  39. @statgeek – no offence, but your figures for Scotland are completely useless. Latest Scottish opinion polls show the SNP is way ahead of Labour for both Scottish and UK parliamentary elections. Tories and the Lib are ‘also rans’.

  40. Bill c

    Link please?

  41. Bill c

    Auto predict tells me your last name is Clinton, would that be correct?

  42. @NickP,

    It’s more than possible IMO that the Libs could recover slightly (and, hence, take some of the Labour vote)…but it’s also more than possible that Labour could lose votes to the Tories (or UKIP, others etc.) as the GE nears. Not saying it will/is likely to happen, but it’s more than possible. Anyone can lose votes to anyone – it’s not just the Libs that Labour have to worry about.

    Of course, by the same token, the Tories should have similar concerns…their vote could fall further….and the threat isn’t limited to just a rise in UKIP.

    @Billy Bob,

    “Even very large polling leads during a parliament are no guarantee of an election victory.

    As yet, Labour are nowhere near giving the impression of being a potential government-in-waiting.

    The same could have been said about the Conservatives in 2010, and they didn’t get a majority.”

    Agree with everything you have said. I’ve never believed this ‘Labour should be leading by X%’ rubbish. Every parliament is different.

    I agree that Labour have yet to convince as a potential government post-2015, and I would add that the same can be said of the Tories. Both are yet to convince the general public, which is why I still believe the next GE is wide open.

  43. statgeek

    Nick Clegg was the spanner in the Conservative works in 2010. I’m pretty sure out of the 23% of Lib VI in 2010, at least 2-3% of those would have been Con voters, but for NC, giving the Conservatives 39-40%.

    Had NC been ineffective in the debates, Cameron would have come out looking good over Brown, and there would have been a Con OM.

    I’m not not sure that your second part follows on from the first. It is quite clear that there was 2-3% of those voting Lib Dem were tactical Tories. Since they started publishing the figures, YouGov’s analysis has consistently shown about 10% of 2010 Lib Dems now choosing the Conservatives. This began soon after the election and hasn’t been affected with events (for example the rise of UKIP), so I can’t see any other plausible reason.

    But the main reason why such people would vote tactically would be that they were in strong Labour areas and voted Lib Dem to keep Labour out. Now some of these may have got it wrong and happened to be in a seat that the Tories could have won, and some may have got it right and helped a Lib Dem over Labour. But in the vast of cases I suspect it didn’t make a difference – a vote that would have been ‘wasted’ on a Conservative candidate was ‘wasted’ on a Lib Dem one. So the percentages might have a bit different but the seats would have stayed the same.

  44. 2 months ago Anthony collated all the polling on the Scottish referendum & there is no suggestion at all that the yes vote would come out on top.

    http://ukpollingreport.co.uk/blog/archives/category/scotland

  45. Bill C
    I just took the latest YouGov from yesterday and looked at the vote shares for parties which shows 76% support for those opposed to independence.

    I am sure the actual result will be a tad closer as people will not vote on party lines however, not all SNP members want independence either so this is a two way street

    I don’t know where you get the idea than SNP is ahead of Labour for the UK GE election the latest figures (Yesterdays) are 28% SNP 38% Labour

  46. @Bill C
    Care to provide a link as to where those “latest scottish opinion polls” are?

    The latest purely Scottish one I am aware of is the Ipsos Mori one in May – but then that was for Holyrood, not Westminster. The results of that (compared to 2011 result in brackets)

    SNP 39 (-6.4)
    Lab 36 (+4.3)
    Con 16 (+2.1)
    LD 8 (+0.1)

    Or are you referring to all these private polls the SNP claim to show them as being universally popular, and independence a sure fire thing, but they refuse to publish any details of?

  47. Bill

    The latest Ipsos Mori poll on Scottish independence has the following results.

    The Times finds that support for Scotland remaining part of the UK has increased to its highest level since August 2011.

    Among those certain to vote, 59% would vote ‘No’, up four points since February, while 31% would vote ‘Yes’, down three points. One in ten Scots are undecided, down a point.

    With the exception of a bounce following the SNP’s election victory in 2011, support for independence continues to hover between 30% and 35%, which is in line with the historical average recorded by Ipsos MORI over the last three and a half decades.

    Maybe you are on the wrong street?

  48. I was following the Kingston by-election a bit being where I used to live. One thing I saw (looking on the Lib Dem website for the result) was they had a ‘virtual phonebank” as follows:

    F125BF-9603 for no data
    2B45BB-9892 for Soft Cons
    3B45BC-6491 for D&P

    It made my mind wonder to all sorts of theories of high tech canvassing techniques. People dialing into the virtuall phonebank and results of phone canvassing recorded automatically on a central computer type thing.

    In my day (20 years ago now) we just got an electoral register and went knocking on doors. I wonder if any of the still active members of any of the political parties could shed some light onto modern canvassing methods?

  49. Come on, Anthony, you’ve had two dedicated threads this week on the YouGov polls showing only 4% and 3% Labour leads but none on the two showing 7 and 8% leads. Anybody would think you were a Sun Tweeter!!

    The variability could be explained by the vagaries of sampling and MOE and it’s therefore difficult to draw any firm conclusions about whether we’re seeing a Tory recovery of any lasting significance. 35% says yes, 32% says no and although we haven’t had a non YouGov for a while, most of the other pollsters, bar ICM, still have the Tories still languishing in the 20s.

    What’s beyond doubt, barring our old friend ICM again, is the stickability of the Labour vote in the high 30s. They’ve undoubtedly had a few points shaved off their VI over the last two months, and 40s are a rarity now rather than the norm, but they appear to be riding out a turbulent period pretty well. The Tories have raised their game, buoyed by some favourable news stories and sharpened up in terms of messaging by Crosby, whilst Labour have encountered some significant turbulence. Good days for the coalition.

    Call me a “half glass empty” Labour man if you like but, as cricket captains are often heard to say, I think Miliband and his party can take “some positives” from these polls. Taken in the round, and bearing in mind recent political events and where we are in the electoral cycle, I don’t think these polls are at all bad for Labour. In fact, I think they could be seen as quite heartening, pointing as they do towards a relative resilience in the Labour vote.

    I now expect the “glass half empty” Labourites on here (for there are many and they know who they are(lol!)) to tell me I’m a deluded fool!

    I’m now off for four weeks Down Under in Kevin Rudd-land, visiting some old friends and relatives and striking out into places I haven’t visited before (Adelaide, Whitsundays and Darwin). I’m particularly looking forward to viewing the Aussies Ashes humiliation through their eyes, although it may mean getting up at strange hours to tune in! I was there in 2006/07 when they stuffed us 5-0 and I’m looking forward to seeking out an Aussie I bumped into in a pub in Port Fairy at that time who kept tapping me on the shoulder, God knows how many times, to tell me: “Hey Pom we won the bloody Ashes mate!”

    Happy holidays to one and all and I look forward to returning in time for the September Conference shenanigans.

1 2 3 4 5