There have been several new polls with voting intention figures since the weekend, though all so far have been conducted before the government’s defeat on their Brexit plan.

ComRes/Express (14th-15th) – CON 37%(nc), LAB 39%(nc), LDEM 8%(-1), UKIP 7%(+1)
YouGov/Times (13th-14th)- CON 39%(-2), LAB 34%(-1), LDEM 11%(nc), UKIP 6%(+2)
Kantar (10th-14th) – CON 35%(-3), LAB 38%(nc), LDEM 9%(nc), UKIP 6%(+2)

Looking across the polls as a whole Conservative support appears to be dropping a little, though polls are still ultimately showing Labour and Conservative very close together in terms of voting intention. As ever there are some differences between companies – YouGov are still showing a small but consistent Tory lead, the most recent polls from BMG, Opinium and MORI had a tie (though Opinium and MORI haven’t released any 2019 polls yet), Kantar, ComRes and Suration all showed a small Labour lead in their most last polls.

Several people have asked me about the reasons for the difference between polling companies figures. There isn’t an easy answer – there rarely is. The reality is that all polling companies want to be right and want to be accurate, so if there were easy explanations for the differences and it was easy to know what the right choices were, they would all rapidly come into line!

There are two real elements that are responsible for house effects between pollsters. The first is the things they do to the voting intention data after it is collected and weighted – primarily that is how do they account for turnout (to what extent do they weight down or filter out people who are unlikely to vote), and what to do they with people who say they don’t know how they’ll vote (do they ignore them, or use squeeze questions or inference to try and estimate how they might end up voting). The good thing about these sort of differences is that they are easily quantifiable – you can look up the polling tables, compare the figures with turnout weighting and without, and see exactly the impact they have.

At the time of the 2017 election these adjustments were responsible for a lot of the difference between polling companies. Some polls were using turnout models that really transformed their topline figures. However, those sort of models also largely turned out to be wrong in 2017, so polling companies are now using much lighter touch turnout models, and little in the way of reallocating don’t knows. There are a few unusual cases (for example, I think ComRes still reallocate don’t knows, which helps Labour at present, but most companies do not. BMG no longer do any weighting or filtering by likelihood to vote, an adjustment which for other companies tends to reduce Labour support by a point or two). These small differences are not, by themselves, enough to explain the differences between polls.

The other big differences between polls are their samples and the weights and quotas they use to make them representative. It is far, far more difficult to quantify the impact of these differences (indeed, without access to raw samples it’s pretty much impossible). Under BPC rules polling companies are supposed to be transparent about what they weight their samples by and to what targets, so we can tell what the differences are, but we can’t with any confidence tell what the impact is.

I believe all the polling companies weight by age, gender and region. Every company except for Ipsos MORI also votes by how people voted at the last election. After that polling companies differ – most vote by EU Ref vote, some companies weight by education (YouGov, Kantar, Survation), some by social class (YouGov, ComRes), income (BMG, Survation), working status (Kantar), level of interest in politics (YouGov), newspaper readership (Ipsos MORI) and so on.

Even if polling companies weight by the same variables, there can be differences. For example, while almost everyone weights by how people voted at the last election, there are differences in the proportion of non-voters they weight to. It makes a difference whether targets are interlocked or not. Companies may use different bands for things like age, education or income weighting. On top of all this, there are questions about when the weighting data is collected, for things like past general election vote and past referendum vote there is a well-known phenomenon of “false recall”, where people do not accurately report how they voted in an election a few years back. Hence weighting by past vote data collected at the time of the election when it was fresh in people’s minds can be very different to weighting by past vote data collected now, at the time of the survey when people may be less accurate.

Given there isn’t presently a huge impact from different approaches to turnout or don’t knows, the difference between polling companies is likely to be down some of these factors which are – fairly evidently – extremely difficult to quantify. All you can really conclude is that the difference is probably down to the different sampling and weighting of the different companies, and that, short of a general election, there is no easy way for either observers (nor pollsters themselves!) to be sure what the right answer is. All I would advise is to avoid the temptation of (a) assuming that the polls you want to be true are correct… that’s just wishful thinking, or (b) assuming that the majority are right. There are plenty of instances (ICM in 1997, or Survation and the YouGov MRP model in 2017), when the odd one out turned out to be the one that was right.


1,834 Responses to “Latest voting intention and the mystery of house effects”

1 34 35 36 37
  1. @Trevors – thanks for the 11.03am post. I would echo @Charles’ comment and commend you on a decent, well toned post.

    There are a number of points of contention in your 7 points, some of which are not supported by the AHDB summary document, but it’s not an unreasonable starting point for a discussion of how UK agriculture could survive Brexit.

    It basically backs up everything some of us have been explaining for a while now; namely, that a no deal exit in March would be extremely difficult for this sector, and that the promise of cheaper food from some Brexiters following UFT or similar (not you, I appreciate, but the official leave campaigns did promise this) would be catastrophic for large chunks of the industry.

    We can also now agree that Brexit is going to mean the sector will need substantial government assistance, that the intricate trade arrangements are going to be seriously affected, and that consumers will either have to pay higher prices, pay more taxes, and/or have to switch to products they consider inferior, in order to help UK farmers get by.

    It would have been easier to agree all this a week ago, but I hope we can all be happy that we got there in the end.

  2. The Trevors are on better form than usual with their link at 11.03 on Brexit prospects for the UK Agri–food Trade.

    I noticed “Sheep industry particularly sensitive to a Hard Brexit” “UK prices to decline” and big worry about the needless uncertainties.

    Disaster for sheep farming will have much wider effects countrywide than a similar disaster for sectors like pigs, whose producers are much more localised.

    For the uplands and marginal land too high or poor for cereal cropping, the sheep industry is of key importance and has already declined substantially in the past 25 years.

    A small illustration of general attitudes to sheep farming is in a report in our local newspaper of an utterly gruesome rural murder – pensioner battered on head and thousands of pounds of his money horde stolen. The police went to interview a main suspect known to go into the pensioner`s home two months after the murder, the son of a nearby sheep farmer. They were told by the farmer`s wife to go away “We`re too busy with lambing”, and the police didn`t try again for four weeks till June.

  3. TW

    That is certainly an interesting analysis by AHDB. Extremely detailed.

    I recognise that it is about Trade in Agri-Food products.

    What I don’t understand though is how one can forecast post Brexit effects on UK flowing from its Agri-Food Sector without even mentioning the CAP subsidy & its post Brexit deployment.

    CAP isn’t even in the Glossary.

    According to DEFRA, 2017 Total UK Agricultural Output valued at Market Prices was £ 26.3 bn

    After all costs of production this produced a net “Income from Farming” of £2.5bn………………BEFORE receipt of “Subsidies” of £3.2bn-leaving a “Total Income from Farming ” of £5.7 bn

    ie Subsidy currently represents 56% of Farmers NET Income.

    I don’t see how it is possible to discuss post Brexit Agri-Food Sector economics without discussing post Brexit policies for deployment of that part of UK’s Gross EU BUdget contribution currently returned to UK Farmers as CAP subsidies

  4. Sky News reporting that Nissan will announce on Monday that it will not be producing its X Trail model at Sunderland as previously announced in autumn 2016 although the details of the announcement are unclear.

    https://news.sky.com/story/nissan-casts-further-gloom-on-car-industry-with-x-trail-blow-11625885

  5. @ Steam Driven Andy

    Yes- I see your argument and of course this is a big unknown but I’m not convinced you have the right answer.

    2017 seemed to me less about new voters as much as it was about previous voters returning who had not voted on the basis they were “all the same”. I did that 1997 to 2005 voting Green, my vote wasn’t missed of course (which backs up your argument) but New Labour did preside over large losses on the much easier demographic for them than moderate Tories like @Colin who dipped out very early.

    There was also a section who had perhaps gone Lib Dem in 2010 believing they were to the left of Labour, uninspired in 2015, and also quite a lot coming over from the Greens (600,000 votes there alone).

    I just feel that the standards that held up very well during the Blair years are simply not relevant today. Things change, society changes. Sometimes voters want middle of the road parties if things are going well for them so 1992-1997 was probably mostly about thinking the Tories had gone too far but voters liking the giveaways and general economics but not liking the public services element. 2015 onwards was much more voters wanting radical change- except that is split between left and right solutions and produces weird results.

    So yes there are votes to be had from moderate Tories but less and less of them as more people want radical solutions. We don’t know how many UKIP went to Labour in 2017, despite totally different policies, but I think that is was a big demographic of fed up people and bigger than the moderate Tories. It’s a pool of voters is increasing as people’s lives get worse following 10 years of economic stagnation and worsening services/ safety net.

  6. Colin @ 3.16 pm

    Your comments are good today.

    Yes, we simply have to trust that M. Gove and the government will keep their word on replacing the CAP payments. Using the term “subsidies” shows your bias, since these payments don`t just advantage farmers but the wider public, in giving lower and steadier food prices.

    I also hope you are noticing that myself ,whom you accused of aggressive “anti-Englishness”, is not watching the Murrayfield rugby, but intending tuning in to the Dublin game soon..

  7. Interesting map on national identity based on 2011 census returns. It seems people identifying as British are only in a majority in small parts of London and Northern Ireland:

    https://brilliantmaps.com/national-identity-uk/

  8. @ Hireton

    Thanks, interesting stuff.

    If you look carefully there’s another little blob of purple in the middle of England signifying ‘British’ as the largest identity. I’m pretty sure it’s Leicester.

  9. DAVWEL

    @”Using the term “subsidies” shows your bias,”

    I was quoting-from “Total Income from Farming in the United Kingdom First estimate for 2017″ ( DEFRA 3 May 2018).

    The line item in question is -” Other subsidies on production”

    Even if I had not been quoting I would have been inclined to use the word “subsidy.” Your remark implies that this word has some pejorative aspect.

    Governments Tax National Income(S) and redistribute them. They redistribute them for all sorts of reasons depending on their political outlook & priority. By definition -if a given government decides to subsidise anything , it does so because it thinks it is in the national interest to do so.

    If criticism of “subsidy” is levelled it should be related to the outcome of policy-not the word used to describe it.

  10. @Davwell – “..the sheep industry is of key importance and has already declined substantially in the past 25 years.”

    Has it? There has been a substantial fall in the size of the national flock, but that was largely due to the switch to area rather than headage payments. Shortly after the flock started to drop, prices started to rise. I seem to recall that this was first noticed on wool prices, which were ridiculously low, but then recovered to more reasonable levels. [I was particularly aware of this because it scuppered a project to develop a novel use for wool I was working on at the time].

    There have also been additional pressures, like the rising cost of butchering and the drop in consumer demand, but I’m not sure that we can characterize the last 25 years as a straightforward decline.

    I would also agree with @Colin that the payments are subsidies – not quite sure why farmers are averse to admitting to this. As you know, I’m also less troubled by a shift away from upland grazing, at some level and in some areas at least.

    There are many other things we can do with parts of the uplands that would provide more jobs, better environment and stronger communities, but we’ve been round that one before.

  11. Alec:

    Fair comments.

    But I`m amazed if you and Colin haven`t noticed how the word “subsidy” is used to attack shirkers, people who aren`t fit to be treated generously, and the bitter rows it causes between nations, since there are knock-on implications for the spending of governments if “subsidies” are proved to exist.

    Far better to call them payments to achieve conservation benefits, low food prices, stable certain markets, and not just “subsidies for farmers” the full phrase being normal among our urban majorities.

    As often here, I am defending minorities.

  12. The Trevors,
    “There are significant risks and we have left it too late to mitigate all of them, but we can and should be doing a lot more to help directly.”

    Er, I think in fact these are not risks from Brexit, but are what has been sold as features.

    The farm subsidy might continue long enough to soften an orderly shutdown of the farming industry, but it is hard to see what future it has with such a huge disruption of its business model.

    News from Nissan in Sunderland of the loss of a model, which seems to be the investment they had said would continue after the government gave secret assurances.

    What intrigues me, what with labour sounding very soft brexit these days, is whether disenchanted leavers might go straight across to the libs and greens.

    The Other Howard,
    “I agree it’s madness, I don’t want our economy to continue in slow decline as it will do if we stay close.”

    This always raises in my mind the question of how come germany does so well in the EU but supposedly we do not. What is it the Germans are so much better at than we are? Why do you think leaving the EU will miraculously cure whatever it is that is wrong in the UK, when plainly membership isnt causing Gemany a problem.

    It might even be that membership is boosting the UK economy which would othewise be doing far worse outside.

    Colin,
    “PA says that whilst the UK , after a No Deal exit, could mitigate the economic damage with unilateral responses-RoI could not.”

    However, while the Uk can look forward to a future of companies moving out, ireland has strong reason to believe they could be moving out of the UK into ireland.

    In all seriousness, i think there will be temporary measures to mitigate disastrous infrastructure collapses over the first year post no deal, but as things start to settle, it will be Ireland showing gains. The UK finance industry looks ripe for plucking.

  13. Apparently the Nissan Sunderland decision is a result of the EU/ Japan trade deal. Parts can be exported from Japan to EU with Zero tariffs so they would not want to risk parts being made in the UK with tariffs.

  14. @Hireton

    Very interesting map that. Presumably Edinburgh and Aberdeen are slightly ‘less Scottish’ due to influx of non-Scots, whether business, or educationally related. I’ve never gotten any impression of said areas’ Scots being ‘less Scottish and more British’.

    Having said that, the map is almost 10 years out of date. It’s a pity we can’t get the census data so informatively displayed within a year or two of the census. Given the political events of the past 5 years, I would really like to see a UK map of 2019.

  15. @STEAMDRIVENANDY

    We often get comments like: ‘I would vote for you but I don’t like your leader” when out campaigning. People said it about, Blair, Brown, Miliband and Corbyn.

    I think this is natural CON or LIB voters who see the logic in Lab policies but can’t bring themselves to vote for a different ‘tribe’.

  16. “ROBBIEALIVE

    As I am only reading female posters, not much to detain me, other than Valerie & Rosie/Daisie, who are always excellent value.”

    Daisie wants to know if we can prosecute you for sexism as I [a bloak] ghost-write some of Daisie’s offerings when she is having a well earned kip after taking me for a walk.

    Rose isn’t bothered either was but has requested [again] that I don’t threaten to read her a TW post when she’s been naughty.

  17. “Having said that, the map is almost 10 years out of date. It’s a pity we can’t get the census data so informatively displayed within a year or two of the census. Given the political events of the past 5 years, I would really like to see a UK map of 2019.”

    Well the data was available at the end of 2013 and datashine (http://datashine.org.uk/) was up and running in the summer of 2014. Pretty easy to map this sort of thing yourself though in GIS.

  18. Also new opinium out with a hefty change.

    CON: 41% (+4)
    LAB: 34% (-6)
    LDEM: 8% (+1)
    UKIP: 7% (-)
    GRN: 4% (-)

    Obviously usual single poll, possible outlier caveats apply.

  19. Westminster voting intention:

    CON: 41% (+4)
    LAB: 34% (-6)
    LDEM: 8% (+1)
    UKIP: 7% (-)
    GRN: 4% (-)

    via @OpiniumResearch
    , 30 Jan – 01 Feb
    Chgs. w/ 18 Jan

    No it’s not a wind-up!!

  20. Sorry to do this but…

    From @BritainElects

    Westminster voting intention:

    CON: 41% (+4)
    LAB: 34% (-6)
    LDEM: 8% (+1)
    UKIP: 7% (-)
    GRN: 4% (-)

    via @OpiniumResearch, 30 Jan – 01 Feb
    Chgs. w/ 18 Jan

  21. 5% pts swing -Lab to Con ??

    How does that work ???

  22. Apparently the Nissan Sunderland decision is a result of the EU/ Japan trade deal. Parts can be exported from Japan to EU with Zero tariffs so they would not want to risk parts being made in the UK with tariffs.

    Obviously Operation Fear is in full swing…

    Clearly Nissan can’t see the glorious no-deal future we will face with our British stiff upper lip.

  23. DAVWEL

    @” not just “subsidies for farmers” the full phrase being normal among our urban majorities.”

    So urban residents are on your list too!

    Goodness me-your days must be filled with outrage & indignation.

  24. @wes

    “Sorry to do this but…”

    Sorry to post the same poll 13mins after it had just been posted? Your welcome ;)

    @colin
    “5% pts swing -Lab to Con ??

    How does that work ???”

    Well outlier potential aside, these appear to be excluding DK so it could be a Lab>DK swing that then manifests itself as a lab>tory swing in the headline numbers.

    Not all that different to the 2017 election. Almost all of labour’s ‘gain’ was DKs returning and/or forced back to labour.

  25. Colin,
    “How does that work ???”
    Well most likely an outlier. But if its real, then it can only be because labour has been distinctly brexity of late.

  26. For those interested in the six nations: England defensive game looks formidable. I foresee Biggar picked for Wales on 23 February as the long ball attack seems the way to counteract the English Blitz.

    As for Opinium: I don’t know I have given up everything I once held dear is no longer clear and to steal from Simon and Garfunkel:

    My mind’s distracted and diffused
    My thoughts are many miles away

  27. @WB61
    I’m glad you didn’t go on to the next two lines. I don’t know you well enough.

  28. Colin

    “5% pts swing -Lab to Con ??

    How does that work ???”

    Well, they were a bit out of line last time. The rest is interpretation – in whatever way one wants.

    I don’t think the tables will say much.

    It is more likely to be associated with the perception of “activity” – but then it is only my interpretation.

    I think the Conservatives are ahead – in a way it is similar to 1997 (not the scale) – one narrative, different credibilities.

  29. I would imagine that the war on diesel so poorly handled by Grayling had a lot to do with the demise of the X- trail, a big hulk of a car unlikely to be economic in petrol or electric form This is a similar issue to a lot of the Jaguar Land Rover products.

    @Danny
    Money for old rope farming probably as had its day. Within the EU, there was never going to be the overdue reform to deliver public good from farming!

  30. Opinium siding with YouGov.

  31. No Colin.

    Your comments about me are so descriptive of yourself.

    Instead I am content with life, still turning out the research papers and music articles, enjoying the weekly singing, absorbed by some enjoyable rugby of both codes from Thursday to Saturday this week, children doing important jobs, grandchildren starring with one winning a gold medal in the Emirates today.

    You mistake outrage for efforts to improve life for everyone in the UK.

  32. Jonesinbangor

    I wouldn’t agree that Grayling has mismanaged the diesel issue that was one of those pesky EU directives pushed forward by a increasingly desperate French President .

    However back to cars Nissan is one of or even the biggest Japanese car manufacturer here in the US it produces over a million cars a year over here.
    There have been news items over here of corruption at the heart of Nissan and problems in Europe and Russia with lack of sales of certain types of vehicles in there car range.
    I think we have to wary of the Brexit word being suggested as why Car manufacturers in general are cutting back in production the truth is some car companies are having a tough time due to the sudden downturn in diesel sales and over production.
    Personally I think we can see this trend continue throughout the car industry over the next few years as they switch from diesel to electric .Certainly I think European car manufacturers such as Mercedes and bmw will have problems ,as a large proportion of there prestige cars are diesel and wouldn’t be at all surprised to see them cut back on production throughout there production plants around the world and in Europe.
    Of course blaming everything on brexit makes a good smoke screen for poor management and a excellent “look I told you so” moment for remainers so rather like blaming the unions for everything in the 1970’s brexit is the new catch all when things go wrong.

  33. Turk

    All the arguments in your 8.53 pm are wrong – but yes, the point is correct, the automotive industry has significant problems.

    In addition

    There is no dominant design what the e-car or alternative fuel car would be. It will take 50 billion plus to find it out. It is because the current shorter life cycle (Ford Mustang, one of the most glorious cars has never made any profit).

    If you went to BMW China, you would be surprised.

  34. Re: Opinium.

    Shift to the Tories not entirely unexpected. Yes, it’s just one poll, so we need to see others. However, in these times it appears that people prefer clarity (even if that clarity is paradoxically full of contradictions and arguably undeliverable) over nuance.

    If the poll is repeated it could also be an indication that the Labour vote is being eaten into at both ends – Labour Remainers no longer believing that Labour has any inclination to press for a second referendum; Lexiters being drawn away by the party’s push for an extension to A50.

  35. @RAF

    The movement seems to be Labour to Tories. I suspect its Labour Brexit supporters moving, whilst Remainers are still enthusiastic about Cooper and Co and their attempts to delay Brexit?

  36. JiB

    “The movement seems to be Labour to Tories.”

    Until we see the tables, we will have no idea of whether there has been any “movement” between parties or not.

    The headline figures from all pollsters remove those who currently “Don’t Know”. That practice made a lot of sense in a largely stable political system, when polling was primarily concerned with measuring shifts of opinion among voters who floated between the dominant parties.

    That stable system broke down for GB pre 2010, and the breakdown now extends to England.

  37. @JIB
    “I suspect its Labour Brexit supporters moving, whilst Remainers are still enthusiastic about Cooper and Co and their attempts to delay Brexit?”
    You would suspect that, wouldn’t you?

    I suspect it’s centrist tories who lent their vote to Lab last time on the grounds it was the remain party going home to mummy given that bogeyman JC is as bad on Brexit as mummy, as it turns out.
    I would suspect that, wouldn’t I?

  38. “And always keep a-hold of Nurse
    For fear of finding something worse. “

  39. Government minister saying openly that voters would turn on the Tory party if they allowed a no deal, “and rightly so”.

    Reports that Labour MPs are about to split and form a new party.

    Lots of big stuff swirling around, with difficulties for both parties.
    On the Opinium poll – AW usually cautions us against reading too much into a single poll.

  40. “COLIN
    5% pts swing -Lab to Con ??

    How does that work ???”

    Despair – and loss of what was very little faith to start with.

    [That is Rosie’s view anyway.]

  41. Sunday Times headline –

    “Brexit Plan to Evacuate the Queen”

    I can just see the wee soul with a label round her neck and a cardboard suitcase in her hand, waiting tearfully on a station platform to be evacuated to an unknown destination in a safe location.

  42. WB61

    Suggestion from someone in the Lords that Ministers offering MPs enhanced spending in their constituencies would be in contravention of section 1 of the Bribery Act 2010 and “on conviction on indictment, to imprisonment for a term not exceeding 10 years, or to a fine, or to both.”

    Always a bit cheeky to ask a lawyer for free legal advice, but your view would be interesting.

  43. “Only one poll” seems particularly relevant given the Survation in the same period showed no dramatic change.

    Likewise until we actually know if Nissan are going to invest the same x-trail money somewhere else (and where that is) then trying to say what the decision does or doesn’t show about Brexit seems even more speculative than usual?

  44. DAVWEL

    @” efforts to improve life for everyone in the UK.”

    Wow-you are going to be a busy chap !

  45. DANNY

    @”Well most likely an outlier. ”

    I agree.

  46. More details on the Opinium poll for the Observer

    Labour has lost support from both sides of the Brexit debate. Labour has dropped five points among both remainers and leavers. For the first time in since the election, less than half of remainers (49%) would opt for Labour.

    https://www.theguardian.com/politics/2019/feb/02/labour-slumps-in-polls-as-tories-open-biggest-lead-since-general-election?utm_term=Autofeed&CMP=twt_b-gdnnews&utm_medium=Social&utm_source=Twitter#Echobox=1549140757

  47. Sam:

    I put up a message thanking you for the link on BBC Bias on Brexit earlier today, but it has now disappeared.

    Anyway I was grateful to see the figures on the disproportionate appearances of Leave-voting MPs compared to those voting Remain in 2016 – 3 times as likely to appear as their numbers warrant.

    I noticed also that the article was dated April 2018 – surely the bias has increased since then.

    There was also some interesting info on the rogue Banks, buying the BBCComplaints.com web site, so he could pressure the BBC.

  48. @Oldnat –

    “Sunday Times headline –

    “Brexit Plan to Evacuate the Queen””

    Right royal dose of constipation, apparently.

  49. An introduction to some of the complexities of the yet to be published Withdrawal Agreement Implementation Bill which has to be passed before 29 March if there is a WA to implement:

    https://twitter.com/pauldalyesq/status/1091734003265224708?s=19

  50. Alec

    Her late daughter in law referred to it as colonic irrigation.

    (Though, as you pointed out upthread, Brexit is likely to create a significant problem already).

1 34 35 36 37