This morning’s YouGov poll for the Sun has topline figures of CON 35%, LAB 39%, LDEM 9%, UKIP 11%. The four point lead is lower than we’ve seen of late, and comes after a series of five and six point leads this week . In comparison Labour leads were averaging 7 points last month. Tabs are here.

YouGov also asked people if they thought the economy would be doing better or worse if Labour had won the election, and if they thought their own personal finances would be doing better or worse. 21% thought the economy would have been doing better if Labour had won in 2010, 42% that it would be doing worse, 26% thought that it would be much the same. On their own finances, 25% think they would have been better off if Labour had won, 32% worse off, 31% much the same. Tabs are here.

Meanwhile Populus’s twice weekly poll has a similar Labour lead, with topline figures of CON 33%, LAB 38%, LDEM 13%, UKIP 9%. Tabs here.

UPDATE: Ipsos MORI’s monthly poll for the Standard is also out, and it also has a four point Labour lead. Topline voting intentions are CON 33%(+1), LAB 37%(-1), LDEM 9%(+1), UKIP 10%(+2)


196 Responses to “Latest YouGov, Populus & MORI figures”

1 2 3 4
  1. It looks like a small shift has taken place, resulting in a narrower Labour lead.

    I don’t really buy the improving economy argument . I don’t think anyone has seen a noticeable improvement from one week, one month or two months ago.

    I suspect it’s down to:

    1. The autumn statement became lost after the death of Nelson Mandela. Potentially days of talking about increased pension ages just didn’t happen.

    2, I think all Prime Ministers benefit from looking and acting stately and mingling with world leaders (I think the selfie issue is meaningless to most folk – thankfully).

    3. I think the days and weeks of tactical announcements by Labour about the cost of living has been both pushed off the headlines by Mandela and just fizzled out naturally.

    The issue all oppositions face is that they can do nothing, but make announcements. It’s hard to keep ambushing Governments on a day to day basis, when Governments have the benefit of controlling set piece events that make a difference (budgets for example) and increase their profile when international events occur.

    To win an election an opposition needs the electorate to be really, really dissatisfied with the Government andsee the opposition as having a clear plan of what they would stand for.

    I think Ed has done well on the day to day street fighting. I think spelling out clearly what his Government would do has been some what less effective.

  2. Labour are going to struggle as long as they keep Miliband in charge.

    He just isn’t going to win over floating voters.

    (None of which is to say they can’t win. So long as they hang onto most of those LibDem votes they probably will win. Or at least be the largest party. But I cant see them getting a comfortable majortity).

  3. I wonder whether Balls is more of a problem than Miliband. Lots of people of my acquaintance really dislike him and his style – and the headlines about the autumn statement thing were all about him having a red face.

    I think Miliband will get more respected the more he’s heard (which is why DC will try to avoid TV debates)

  4. @ Andy,

    The average Labour lead is about half of what it was this time last year.

    Well, of course it is. They can’t expect the Chancellor of the Exchequer to use his budget as a party political broadcast for the Opposition every year.

    @ Catmanjeff,

    Yeah, I agree with all of that. That said, it would be helpful if they could keep the Government on the defensive a bit more, and it’s not like it isn’t giving them enough targets. So I think a charge of ineffective opposition can be leveled at them.

    On the plus side for Team Red, every time Ed Miliband does show up in the news they get a brief boost in VI, which may bode well for the campaign. I suppose we won’t really know how he fairs in a head-to-head comparison with Cameron until the debates. That Which Must Not Be Named is… uneven, for both of them.

  5. The pups were listening to the news recently and were most offended that ole Kim Rong-Un’s uncle was described as:

    “worse than a dog”

    They were less fussed about his execution.

    I told ’em the announcer had translated it wrong and they meant “better” than a dog.

    They weren’t impressed with that either.

    What have NK got against dogs anyway?

  6. Don’t they eat them?

  7. mr mint

    Shush.

  8. Spearmint

    Then your last seems to support my point that there is no such thing as bad publicity.

  9. There are more than enough facts to refute the “blame Lab for the economic mess” story. Any comparison of GDP data from 97 – 08 vs 10 – 13 will show that the economy was better under Lab. GDP when Lab left office was better than it has been for the first three years of this government. It can be shown that it was a worldwide economic crisis (at least in countries with a large financial industry) caused by the bankers. It can also be shown that the country which has exited the crisis best is the US which did not attempt mass austerity cf. Europe and UK. And then of course there is reference to “we are all in this together” which must be a major area of attack from Lab.

    Any government which has failed to raise GDP to the level it was in 2010 can’t claim to be best with the economy.

    I suspect Lab is just keeping its powder dry until late 2014. It can then bring out all the facts which must surely defeat any argument from Con unless the economy has further improved beyond all expectations. The only drawback I can see with this approach is the means by which Lab can ensure the public is familiar with the facts. Without the backing of most of the press and and the fact that TV news is frequently based on the newspaper headlines dissemination of the facts could be a difficulty. I suspect it will depend on a lot of door knocking and leafleting from Lab to overcome these drawbacks. Whether or not there are TV election debates could also be very key to the success of Lab.

  10. howard

    ” there is no such thing as bad publicity”

    Kim Rongun’s uncle wouldn’t agree

    FOX NEWS

    It is a “verifiable fact that both Santa and Jesus were both white. “That’s just the way it is kids.”

    I find in strange meself that not only have they never been seen in a room together, they’ve never been seen at all.

    Actually my daughter Jools did see Santa when she was four [ Jools, not Santa] and that IS a verifiable fact ‘cos Jools was there.

  11. PC
    Well, I’d never heard of him until he got bumped off. Fame at last!

  12. Oh I see what you mean, posthumously.

  13. ”But the opinion polls say that people don’t agree with your view and do blame Labour at least in part…”

    Indeed some think that way, TOH, it’s the Tory message and many among their target audience buy it! But I don’t think there were many of that opinion before 2010, when Osborne and the LD’s made it their message. It wasn’t what people thought when the crash happened.

    Sadly (?) you and I aren’t allowed to argue the facts of the case here, but this isn’t about facts anyway. It’s about propaganda.

  14. Yet another example of no such thing as bad publicity.

    Convicted criminal gets executed, and here Fox News are banging on about him two thousand years later. Even if they do seem confused about his race.

  15. mr minto

    You won’t be going to heaven.

  16. I think everyone is taking the “If Labour won the last election” thing far too seriously. These people had just been asked asked who they would vote for if the election was tomorrow. If they don’t say “Labour” to that, they’re hardly going to say that they should have won last time. And vice versa. This seems particularly true for Conservatives 88% of whom said worse.

    The real interest lies in the difference between the two questions. While 42% say the economy would be doing worse under Labour, only 32% said the same for the finances of themselves and their family. So part of what Labour has to do is to convince people to vote in their own self-interest – not usually a difficult task, especially as austerity rhetoric may be wearing thin after five years.

  17. People don’t just vote on the economy – it might not even be the main issue. That could be the NHS, tuition fees, ID cards, education, law & order or whatever else is flavour of the month.

  18. I don’t like Ed Balls, but I don’t know of any statistical evidence that says that he’s a liability to Labour. Insofar as he’s a shadow minister who knows his brief and pushes the right buttons for the faithful, he seems like an asset to me.

  19. From time to time IpsosMORI asks: “Which political party do you think has the best team of leaders to deal with the country’s problems?”

    Conservatives fell off the cliff by this measure during the Major years, recovering under Cameron.

    Labour only really hit dangerously low levels in 2008… it remains to be seen whether Miliband can forge a cohesive team in time for the next election.

  20. @spearmint – ” …they do seem confused about his race”

    Any classification of different “races” will inevitably end in confusion. The concept of a species rests on much firmer ground.

  21. rogerh

    People don’t just vote on the economy – it might not even be the main issue.

    True, but it’s what people say is the main issue. And not just for the country but for them and their family. YouGov’s two parallel Issues(2) trackers:

    http://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/jul6j6xizi/YG-Archives-Pol-Trackers-Issues(2)-041213.pdf

    ask these two very same questions and people consistently give “The Economy” as their most ‘important’ issue in both categories[1].

    The query here is what people actually mean by “the Economy”. It’s usually discussed in the media as being about marginal changes in a variety of meaningless (to most people) economic statistics[2]. At the same time the state of the economy does have an actual impact on their lives – in wages, jobs, shops shutting or opening on their High Street and so on.

    Given that Labour clearly has a greater advantage in the second definition of the economy, maybe what they need to do is convince people that they are part of the economy too.

    [1] Obviously not everyone, but always over 50% when given a choice of topics with (up to) three to pick. As usual the two categories affect each other as you’d expect. For example over 90% of UKIP voters not only give ‘Immigration and Asylum’ as one of their three ‘facing the country’ (compared to only 42% who pick the Economy) but over 40% also say it is one of those issues ‘facing you and your family’ compared to only 17% of the population at large. This is despite the fact that UKIP support doesn’t usually correlate with areas of high immigration – if anything the opposite.

    Presumably if, say faced with a long waiting time for a hospital appointment, a UKIP supporter will be more likely to blame service pressure from immigrants rather than government cuts. (The fact the NHS would have collapsed without immigration – 30% of doctors qualified abroad for instance – presumably flies over their heads, despite the evidence in the origin of the people treating them).

    [2] Of course, as often discussed on UKPR, some these or their changes may actually be meaningless. But that doesn’t stop the media treating them with great reverence and attention (depending on how good they are for their particular corporate viewpoint).

  22. I think it’s rather tiresome to say you do or don’t like a particular politician, you may dislike their polices but keeping individual dislikes out of the debate might raise the standard beyond that of the School playground.

    Certainly Parliament would benefit considerably from the removal of the sight of Parliamentarians [snip] baying like [] donkeys every time the representative of the opposing party gets on His or Her feet and reacting negatively to whatever they say even when they secretly agree.

    The Government has actually come up with a couple of half way decent and important ideas in the last week, increasing the age for the fostering allowance to 21 and additional support for Alzheimers research which no doubt will have cross party support but this good work will be for nought if the public’s perception of parliament is that it’s a bunch of hoorays who appear to have had a few too many bottles of Bollinger.

  23. Just out of interest, the Conservative leader of Tendring District Council resigned yesterday: http://www.clactonandfrintongazette.co.uk/news/10876683.Tory_council_leader_sensationally_quits/?ref=var_0

    We now have two pending by-elections in Conservative held wards. Bothnormally fairly safe seats but with UKIP standing for the first time (probably), the collapse nationally and locally of the LibDem votes, who knows? Could be two seats to keep an eye on.

  24. The student loan system seems to me to be a way of racking up a massive debt, much of which will be written off.

    http://www.dailymail.co.uk/news/article-2523527/How-foreign-students-fleecing-Britain.html

    Surely there must be a better way of financing higher education. Rather than have these rules regarding no payback of the loan if earnings are below £21k or whatever amount it is, would it not be better to have minimum payments always made. Then the student gets use to having to make payment and the SLC may have more contact with the student. For a student from the EU getting a loan, I would suggest UK government obtains some form of guarantee of payment from the relevant countries government, if the student defaults.

  25. BP
    “Insofar as he’s a shadow minister who knows his brief and pushes the right buttons for the faithful, he seems like an asset to me.”
    I’d agree on the ‘knows his brief’ part, but the ‘pushes the right buttons’ depends on which sect of the faithful you ask.

    The Blairite wing seem to hate Balls, because his solutions are broadly Keynesian so he doesn’t necessarily accept the need for huge spending cuts (despite him sticking with the narrative of ‘tough choices’) and his insistence on government subsidies for capital/infrastructure spending.
    This doesn’t chime well with the folks who wish that the economic policy was far more Osborne-esque than Brownite.

    StatGeek,
    Have a quick question, since I haven’t been tracking YouGov’s polling for a long time and I assume you have.

    What percentage of the average change in Labour lead is down to changes in Labour VI and how much is down to changes in Conservative VI?
    From my, admittedly unscientific (since I haven’t consulted the data), the Labour VI is fairly static and changes to the Lead are largely the result in changes to Con VI.
    But if you wouldn’t mind taking a quick look, I’d be grateful.

  26. @R Huckle

    (Chooses his words carefully)….

    There is a broad acceptance within Whitehall that the student loan system now in place is in need of fundamental revision, but it can’t happen yet because the political consequences, especially for the Lib Dems, are too severe.

    Whoever gets in in 2015 will need, urgently, to sort it out. At the very least the repayment threshold needs to come down to below £21k, and I would expect the interest rate to have to go up as well if the system remains. In fact, the easiest way to sort it out would be to lower fees and increase Treasury support – this would probably cost the taxpayer *less* in the medium to long term (but more in the short term) and would not risk the financial stability of the whole HE system.

    The system is already being studied around the world – other nations are fascinated by it. I will say no more about the nature of their fascination other than there is no danger of any other system choosing to copy us.

  27. Lord Gwyn Roberts has died. If you look at the constituency of Conwy on Wikipedia (was Conway and now abolished) it has an extraordinary history of marginal, three way, and safe seat (for more than one party!!!) since 1950, without its boundaries being altered until done away with for the 2010 election (and gifting Plaid a seat it seems to me!).

  28. It’s time for the By-Election Report!

    Aylesbury TC, Elmhurst
    December 12, 2013

    LD 584 (51.5)
    UKIP 324 (28.5)
    Lab 120 (10.6)
    Con 77 (6.8)
    Green 30 (2.6)
    Majority 260
    Turnout 22.7%
    LD hold.

    Either this is one of those odd areas where the LDs have done tremendously well locally, or I think I spy a “Stop UKIP” panic vote from Lab and Con. Or the local Labour and Tory candidates were only paper candidates. Possibly a little of all three – but we have no changes from last election to confirm.

    Richmondshire DC, Hornby Castle
    December 12, 2013

    Con 127 (46.2; -53.8)
    Ind 98 (35.6; +35.6)
    UKIP 50 (18.2; +18.2)
    Majority 29
    Turnout 19.23%
    Con hold.
    Percentage change is since May 2011.

    CON COLLAPSE TO 45% SWING. Haha, not really. Last time the seat was unopposed, so we don’t know how it would have gone. There wasn’t a Lib Dem or Labour candidate, so presumably those voters went for the Independent – a rare case of UKIP splitting the opposition vote!

    Hambleton DC, Topcliffe
    December 12, 2013

    Con elected unopposed. (100.0; +25.2)
    [LD (0.0; -25.2)]
    Con hold.
    Percentage change is since May 2011.

    CON SURGE – Oh alright. This seems to be a bit of a problem in local government – maybe the opposition don’t have a hope if they can’t find anyone to stand, but the public should have a right to vote for them if they choose.

    South Buckinghamshire DC, Iver Village and Richings Park
    December 12, 2013

    Con 422 (46.9; -6.0)
    UKIP 377 (41.9; +41.9)
    LD Peter Chapman 101 (11.2; -35.9)
    Majority 45
    Turnout 20.4%
    Con gain from LD.
    Percentage change is since May 2011.

    A big LD>UKIP swing here, which seems, er, unusual. Given last time it was a straight Con-LD fight, presumably the LDs were just a receptacle for the ABT vote, and have now been supplanted by UKIP, rather than this generally reflecting a massive ideological shift among the electorate.

    Warwickshire CC, Bedworth West
    December 12, 2013

    Lab 904 (62.6; +4.5)
    Con 353 (24.4; -2.7)
    UKIP 142 (9.8; +9.8)
    TUSAC 46 (3.2; -3.4)
    [Green (0.0; -8.2)]
    Majority 551
    Turnout 16.94%
    Lab hold.
    Percentage change is since May 2013.

    Unclear where votes are going here. The most obviously matching numbers point to a Green>UKIP shift, which is unlikely for obvious reasons. Presumably UKIP have gained some protests votes from the Socialist Workers Par- sorry, TUSC – and that little bit from the Cons.

    Presumably there were also some anti-immigration Lab voters who’ve gone over to the purples, but one imagines that would be absorbed by Labour taking most of the Green votes.

    So to conclude:
    Con Hold 2, Gain 1 from LD.
    Lab Hold 1.
    LD Hold 1.

  29. @CMJ

    “To win an election an opposition needs the electorate to be really, really dissatisfied with the Government andsee the opposition as having a clear plan of what they would stand for.”

    Or in other words, regardless of government or opposition, the party with the best policies ‘should’ win the election. Not so, I’m afraid, with spin doctors and televised debates. It’s all presentation, and no substance.

  30. @Mrnameless

    Funny, isn’t it that there are still “odd” areas where the Lib Dems are doing well? Nice one.

    Iver, I presume is about HS2 by the way. Supporting the long term interests of the country by investing in transport infrastructure for the whole nation is presumably locally very unpopular.

  31. Roger Mexico and Peter Bell

    If one looks at the tables about ‘worse off, better off’ it will be seen that hardly any (6%) eeyores of *Labour* voters think they would be worse of under Labour. It’s just their ‘better off’ figures that are lower than Con ‘worse off’ under Lab.

    It also noticeable that the figures generally don’t change much between ‘personal economy’ and ‘economy’ when the question is put the way it was.

    One other thing from that table, LD 2010 voters noticeably replied with better opinions for Labour than present LD voters, so that proves the latent goodwill among those who defected (I would say returned).

    So Labour strategists know that this supposed forelock tugging that the toffs manage things better does not carry very far with *Labour* voters and that is really all they need to concentrate on.

    I use asterisks for emphasis as I can’t find out how to use italics when replying on this site.

  32. RC,

    I meant rare ‘cos they have quite widely spread support, not that LD councillors are necessarily incompetent!

    Also, why do LDs run as “Liberal Democrat Focus Team”? It makes them sound like the least useful superhero team ever.

  33. Ah, done some research into this – Parties have several “registered” names they can use on ballot papers. Here are the Lib Dems’:

    Democratiaid Rhyddfrydol Cymru – Welsh Liberal Democrats
    Focus Team
    Liberal Democrat
    Liberal Democrat Focus Team
    Liberal Democrats – for a fairer Britain
    Liberal Democrats – for a greener Britain
    Nick Clegg’s Liberal Democrats
    Scottish Liberal Democrat Focus Team
    Scottish Liberal Democrats
    Welsh Liberal Democrats
    Welsh Liberal Democrats – Democratiaid Rhyddfrydol Cymru

    I can’t see many candidates standing for “Nick Clegg’s Liberal Democrats”.

  34. “@ Chris Riley

    @R Huckle

    (Chooses his words carefully)….

    There is a broad acceptance within Whitehall that the student loan system now in place is in need of fundamental revision”

    Not sure what you mean by choose his words carefully, but I suspect that you have some inside knowledge about the current student loan scheme.

    I said at the time the coalition changed the student loan scheme that it would not work, because it was just setting up a larger loan book, which was going to be much more expensive to government in the long run.

    As you say, it would be better for government to reduce the annual tuition fees to say £4500 and for them to fund HE more from general taxation. The debt loan book is then smaller as a consequence and the potential for write off is lower. Students should have to pay back a minimum amount if they are in work, so they get use to making repayments.

    To be fair to the coalition, I think this new student loan/tuition fee system was implemented following an independent review.

  35. @ That Old Bloke

    “Labour are going to struggle as long as they keep Miliband in charge.

    He just isn’t going to win over floating voters.”

    Without getting into an Ed is carp/Ed is great thread and accepting his presentational issues, I think you cannot knock the fact that he has got Lab into a position where they are still more likely than not to win the next election. This was from a base of 29% which was the worst since Michael Foot where it took another 3 elections for Lab to get elected.

    Of course a lot of it is nothing to do with him as coalition produced the LD collapse from which most of his increased vote share has come. But even though this is like shooting fish in a barrel you still have to be able to shoot those fish and he has done this pretty well. He has positioned himself to mop up those LD votes and maybe has done enough to get some of the left wing deserters back in the fold without scaring the natives. I don’t think it is realistic to expect a massive swing back so soon after the 2008 crash so he has made the best of what he has.

    I doubt whether his brother could have done the same. Former defectors (like me) would have been a bit nervous that it was more of the same whereas with Ed there might not be Obama style first election hope but he has done enough to make you think things could be different.

  36. @Fringe

    “What percentage of the average [change] in Labour lead is down to changes in Labour VI and how much is down to changes in Conservative VI?”

    I had to read that about 6 times. Must be the ‘Saturday morning’ lack of focals or lack of braincells. What is ‘the average [change] in Labour lead’ (i.e. from what points in time do we measure that) ?

    Do we measure the change on a day to day basis, and take an average of the changes?

    Answer? How long is a piece of string, when you have three sticks by which to measure it? It sounds very in-depth.

    UK only – MAD, three month change: Con -0.2%, Lab +1.0% (bear in mind that that Lab were lower about then, and they are still above that low point after the conference), but the changes are so minimal, it’s polldrums in my opinion.

    However, the variance of the Con most recent 30 polls is 1.56, while Lab’s variance is 0.87, if that helps. That says to me (and I might have interpreted variance wrongly) that the Con variance is almost double (64%) that of Lab variance (36%).

  37. Colin Davis

    ” I did, I sold my whole portfolio in 2006, and re-eatblished it mainly in 2010 with splendid results for my personal finances”. Your view is a minority view and as you say its about propgander and public perception.

  38. Mr Nameless
    I admire your scholarship in digging up those results but i am afraid I see no significance about any of them. If a local election gets a turnout above 40% then it becomes mildly interesting.

  39. When we look at the MAD lead, we get far smaller variances, as we’ve already taken out the outliers on a given day for the past 30 polls. MAD tends to lack changes for days, then it can change by a sizeable margin suddenly due to median shift.

    The variance for the Con MAD for the past thirty days (i.e. we have taken a note of the UK Con [MAD] value every day, and compared the variance of all 30 values) was 0.044, while Lab’s was 0.013, while the [MAD] lead variance was 0.093.

    Note that the previous post was looking at the variance of the pure poll values and the percentage of each to the sum of the two. However, when I just went and calculated the [lead] for thirty polls, the variance of the 30 values was 2.9 (as opposed to the 2.43 sum of the previous post’s values).

    All calcs using Excel VAR function. I considered VARP, but the differences in answers are similar, so I doubt it matters too much:

    http://office.microsoft.com/en-gb/access-help/var-varp-functions-HA001231525.aspx?CTT=1

    It’s all too easy to get caught up in this, and miss the point. :))

  40. Mr Nameless

    Thanks for the council election rsults. Although i understand the point Howard makes I do take an interest in local bye-elections and do find it interesting that the last two weeks have seen conservative gains at a time when they remain behind in the polls.

  41. Moving away from LAB-CON-UKIP debates, can anyone explain why it is that in Scotland YouGov and Populus are consistently returning very different results for the two main parties north of the border? Friday’s results are typical: YouGov give Labour 44% and the SNP 21% (lower than normal, but not dramatically so) whilst Populus give Labour 34% and the SNP 27%. Such large variations could have a big impact on Labour’s ability to form a Government in the next Westminster Parliament if the race turns out to be tight.

  42. Interesting, and disappointing turnouts for those council election results (perhaps 20% or so is normal). That suggests that if you vote, your vote has five times the power it should have (but so has every vote, so it’s all relative).

  43. At John B. If I might quote Anthony from:

    http://ukpollingreport.co.uk/blog/archives/5717

    2) Remember polls have a margin of error
    3) Beware cross breaks and small sample sizes

    “A poll of 1000 people in Great Britain as a whole might have fewer than 100 people aged under 25 or living in Scotland. A crossbreak made up of only 100 people has a margin of error of plus or minus ten percent. Crossbreaks of under 100 people should be given extreme caution, under 50 they should be ignored.”

    In this case, the general sample size of Scotland is weighted to be 8.7% of the total poll sample (while there were 184 samples, out of a total sample of 1766, 8.7% is 153.6, or the weighted value in the recent poll of 154).

    The margin of error for either is 7.5% to 8%, which can mean the SNP are actually on 29%, while Lab are on 36% (or SNP on 13%, with Lab on 52%). Hence why I use the MAD system for the cross-breaks. It’s not perfect, but it does weed out the worst outliers from a sample. The MAD values for Scotland to the most recent poll are:

    Lab 39.4%
    SNP 26.1%

    The most recent ten polls are (most recent to the right):

    Lab 30 32 42 40 43 44 38 36 33 44 (Average – 38.2)
    SNP 30 32 25 21 29 25 31 26 24 21 (Average – 26.4)

    The MAD calc includes 16 values for Lab (14 omitted due to heavy variation from the sample of 30), and 15 samples for the SNP:

    Lab – 36 37 37 37 38 38 38 40 40 41 41 41 42 42 43 43
    SNP – 24 24 24 25 25 26 26 26 26 26 27 27 27 29 29

    If you’re not familiar with ‘MAD’, http://www.statgeek.co.uk/polling/median-absolute-deviation-mad/

  44. At John B. If I might quote Anthony from:

    http://ukpollingreport.co.uk/blog/archives/5717

    2) Remember polls have a margin of error
    3) Beware cross breaks and small sample sizes

    “A poll of 1000 people in Great Britain as a whole might have fewer than 100 people aged under 25 or living in Scotland. A crossbreak made up of only 100 people has a margin of error of plus or minus ten percent. Crossbreaks of under 100 people should be given extreme caution, under 50 they should be ignored.”

    In this case, the general sample size of Scotland is weighted to be 8.7% of the total poll sample (while there were 184 samples, out of a total sample of 1766, 8.7% is 153.6, or the weighted value in the recent poll of 154).

    The margin of error for either is 7.5% to 8%, which can mean the SNP are actually on 29%, while Lab are on 36% (or SNP on 13%, with Lab on 52%). Hence why I use the MAD system for the cross-breaks. It’s not perfect, but it does weed out the worst outliers from a sample. The MAD values for Scotland to the most recent poll are:

    Lab 39.4%
    SNP 26.1%

    The most recent ten polls are (most recent to the right):

    Lab 30 32 42 40 43 44 38 36 33 44 (Average – 38.2)
    SNP 30 32 25 21 29 25 31 26 24 21 (Average – 26.4)

    The MAD calc includes 16 values for Lab (14 omitted due to heavy variation from the sample of 30), and 15 samples for the SNP:

    Lab – 36 37 37 37 38 38 38 40 40 41 41 41 42 42 43 43
    SNP – 24 24 24 25 25 26 26 26 26 26 27 27 27 29 29

    If you’re not familiar with ‘MAD’, h ttp://www.statgeek.co.uk/polling/median-absolute-deviation-mad/

    (Reposted, due to two links.)

  45. At John B. If I might quote Anthony from:

    http://ukpollingreport.co.uk/blog/archives/5717

    2) Remember polls have a margin of error
    3) Beware cross breaks and small sample sizes

    “A poll of 1000 people in Great Britain as a whole might have fewer than 100 people aged under 25 or living in Scotland. A crossbreak made up of only 100 people has a margin of error of plus or minus ten percent. Crossbreaks of under 100 people should be given extreme caution, under 50 they should be ignored.”

    In this case, the general sample size of Scotland is weighted to be 8.7% of the total poll sample (while there were 184 samples, out of a total sample of 1766, 8.7% is 153.6, or the weighted value in the recent poll of 154).

    The margin of error for either is 7.5% to 8%, which can mean the SNP are actually on 29%, while Lab are on 36% (or SNP on 13%, with Lab on 52%). Hence why I use the MAD system for the cross-breaks. It’s not perfect, but it does weed out the worst outliers from a sample. The MAD values for Scotland to the most recent poll are:

    Lab 39.4%
    SNP 26.1%

    The most recent ten polls are (most recent to the right):

    Lab 30 32 42 40 43 44 38 36 33 44 (Average – 38.2)
    SNP 30 32 25 21 29 25 31 26 24 21 (Average – 26.4)

    The MAD calc includes 16 values for Lab (14 omitted due to heavy variation from the sample of 30), and 15 samples for the SNP:

    Lab – 36 37 37 37 38 38 38 40 40 41 41 41 42 42 43 43
    SNP – 24 24 24 25 25 26 26 26 26 26 27 27 27 29 29

    If you’re not famil1ar with ‘MAD’, h ttp://www.statgeek.co.uk/polling/median-absolute-deviation-mad/

    (Reposted, due to two links, and possible problems with word f a m i l i a r.)

  46. @R Huckle

    Browne recommended that fees go up but not the mechanism by which that took place or would be funded.

    The scheme itself appears to have been dreamed up at least in part at certain right-wing think-tanks, and elements are traceable to them from before the 2010 election. The think-tank particularly responsible is full of very clever young people who know absolutely nothing and who will all become SPADS and, probably, MPs.

  47. “YouGov also asked people if they thought the economy would be doing better or worse if Labour had won the election, 21% thought the economy would have been doing better if Labour had won in 2010, 42% that it would be doing worse”
    ________

    I’m not at all surprised at these findings and with the economy on the up it must be a huge blow to Labour.

    3 polls and all 3 showing Labour with a 4% lead. We all know the UKIP VI will collapse come the election and many will return to the Tories when they realise what the alternative might be.

    Economy on the up spells disaster for Labour.

  48. I just wonder if the small increase in Cons and reduction in Labour is realisation amongst some voters that some of Labours big policy announcements have real problems.

    Take the energy fix, since this was announced, energy firms have been knocked about and lost value, taking billions off pension fund values, therefore costing the tax payer billions. At the same time , it has added uncertainty to the industry, damaged green investment and taken around £10bn of windfarms investment off the table through cancelled projects linked to the uncertainty in the industry. Is this the business environment we want? Am not singing the praises of capitalism, but we do need a balance if we want growth, tax revenues, and money for social/welfare projects too.

  49. That Old Bloke

    Labour are going to struggle as long as they keep Miliband in charge.

    He just isn’t going to win over floating voters.

    (None of which is to say they can’t win. So long as they hang onto most of those LibDem votes they probably will win. Or at least be the largest party. But I cant see them getting a comfortable majortity).
    _______

    The Lib/Dems will return to the Libs, creature of habits and UKIP’ers will return to the Tories.

    If the economy is still showing growth in 2015 then Labour can cancel Hicks removals for Ed.

  50. Allan Chrstie

    “3 polls and all 3 showing Labour with a 4% lead. We all know the UKIP VI will collapse come the election and many will return to the Tories when they realise what the alternative might be.

    Economy on the up spells disaster for Labour.”

    I just hope you are a Conservative party strategist.

1 2 3 4