There have been three polls over the last week – in the Sunday papers there were polls from ComRes and Opinium, the regular YouGov poll for the Times last week. Voting intention figures were:

Opinium – CON 37%, LAB 25%, LDEM 16%, BREX 13%, GRN 2% (tabs)
ComRes – CON 28%, LAB 27%, LDEM 20%, BREX 13%, GRN 5% (tabs)
YouGov – CON 32%, LAB 23%, LDEM 19%, BREX 14%, GRN 7% (tabs)

There isn’t really a consistent trend to report here – YouGov and ComRes have the Conservatives declining a little from the peak of the Johnson honeymoon, but Opinium show them continuing to increase in support. My view remains that voting intention probably isn’t a particularly useful measure to look at when we know political events are looming that are likely to have a huge impact. Whatever the position is now, it is likely to be transformed by whether or not we end up leaving the European Union next month, on what terms and under what circumstances.

What did receive some comment was the sheer contrast between the reported leads, particularly because the ComRes (1 point Tory lead) and Opinium (12 point Tory lead) were published on the same day.

Mark Pickup, Will Jennings and Rob Ford wrote a good article earlier this month looking at the house effects of different pollsters. As you may expect if you’ve been watching recent polls, ComRes tend to show some of the largest Labour leads, YouGov some of the biggest Tory leads. Compared to the industry average Opinium actually tend to be slightly better for Labour and slightly worse for the Tories, though I suspect that may be changing: “House effects” for pollsters are not set in stone and can change over time, partly because pollsters change methods, partly because the impact of methodological differences change over time.

What that doesn’t tell us why there is a difference. I saw various people pointing at the issue of turnout, and how pollsters model likelihood to vote. I would urge some caution there – in the 2017 election, most of the difference between polls was indeed down to how polling companies predicted likelihood to vote, and this was the biggest cause of polling error. However when those new turnout models backfired and went wrong, polling companies dropped them. There are no longer any companies using demographic based turnout models that have a huge impact on voting intention figures and weight down young people. These days almost everyone has gone back to basing their turnout models primarily on how likely respondents themselves say they are to vote, a filter that typically only has a modest impact. It may be one factor, but it certainly wasn’t the cause of the difference between ComRes and Opinium.

While polling companies don’t have radically different turnout models, it is true to say (as Harry does here) that ComRes tends to imply a higher level of turnout among young people that Opinium. One thing that is contributing to that in the latest poll is that Opinium ask respondents if they are registered to vote, and only include those people who are, reducing the proportion of young people in their final figures. I expect, however, that some of it is also down to the respondents themselves, and how representative they are – in other words, because of the sample and weights ComRes may simply have young people who say they are more likely to vote than the young people Opinium have.

As regular readers will know, one important difference between polling companies at the moment appears to be the treatment of past vote weighting, and how polling companies account for false recall. Every polling company except for Ipsos MORI and NCPolitics use past vote in their weighting scheme. We know how Britain actually voted at the last election (CON 43%, LAB 41%, LDEM 8%), so a properly representative sample should have, among those people who voted, 43% people who voted Tory, 41% people who voted Labour, 8% who voted Lib Dem. If a polling company finds their sample has, for example, too many people who voted Tory at the previous election, they can weight those people down to make it representative. This is simple enough, apart from the fact that people are not necessarily very good at accurately reporting how they voted. Over time their answers diverge from reality – people who didn’t vote claim they did, people forget, people say they voted for the party they wish they’d voted for, and so on. We know this for certain because of panel studies – experiments where pollsters ask people how they voted after an election, record it, then go back and ask the same people a few years later and see if their answers have changed.

Currently it appears that people are becoming less likely to remember (or report) having voted Labour in 2017. There’s an example that YouGov ran recently here. YouGov took a sample of people whose votes they had recorded in 2017 and asked them again how they had voted. In 2017 41% of those people told YouGov’s they’d voted Labour, when re-asked in 2019 only 33% of them said they had voted Labour. This causes a big problem for past vote weighting, how can you weight by it, if people don’t report it accurately? If a fifth of your Labour voters do not accurately report that they voted Labour and the pollster weights the remaining Labour voters up to the “correct” level they would end up with too many past Labour voters, as they’d have 41% past Labour voters who admitted it, plus an unknown amount of past Labour voters who did not.

There are several ways of addressing this issue. One is for polling companies to collect the data on how their panellists voted as soon as possible after the election, while it is fresh in their minds, and then use that contemporaneous data to weight future polls by. This is the approach YouGov and Opinium use. The other approach is to try and estimate the level of false recall and adjust for it – this is what Kantar have done, instead of weighting to the actual vote shares in 2017, they assume a level of false recall and weight to a larger Conservative lead than actually happened. A third approach is to assume there is no false recall and weight to the actual figures – one that I think currently risks overstating Labour support. Finally, there is the approach that Ipsos MORI have always taken – assuming that false recall is such an intractable problem that it cannot be solved, and not weighting by past vote at all.

Dealing with false recall is probably one reason for the present difference between pollsters. Polling companies who are accounting for false recall or using methods that get round the problem are showing bigger Tory leads than those who do not. It is, however, probably not enough to explain all the difference. Neither, should we assume that the variation between pollsters is all down to those differences that are easy to see and compare in the published tables. Much of it is probably also down to the interaction of different weighting variables, or to the very samples themselves. As Pat Sturgis, the chair of the 2015 enquiry into polling error, observed at the weekend there’s also the issue of the quality of the online panels the pollsters use – something that is almost impossible to objectively measure. While we are wondering about the impact of weights and turnout filters, the difference may just be down to some pollsters having better quality, more representative panels than others.


4,413 Responses to “Latest voting intention and the difference between the polls”

1 2 3 89
  1. Anyone here?

  2. Some weak arguments being put to the Supreme Court to justify the long prorogation.

    “”116. The Inner House inferred that the real reason for the advice was to prevent Parliament from sitting and holding the Government to account in the run up to the UK’s exit from the EU on 31 October: per the Lord President at §53, Lord Brodie at §89, Lord Drummond Young at §123. This was an inference which no reasonable court could have drawn on the evidence before it.””

    Also I thought they were joking when they argued that the long length was so as not to clash with an/the SNP conference.

  3. Are there ‘shy Labour” voters? Or do they now deny voting Labour because they have no intention of doing so again?

  4. Doesn’t “false recall” simply imply a regret at having voted in a particular way in the past and, therefore, make it relatively easy to allow for the significance of that fact?

  5. Although the Tories have recovered fairly well, as was somewhat predictable once the exhausted previous PM was replaced, it is a very tenuous support base dependent on Brexit outcomes.

    If a Brexit deal isn’t reached by Halloween, the Tories really will be on the bonfire, and I expect a mass defection to Brexit Party.

  6. I was wondering if AW had left it an unusually long time between thread heading articles in the hope it might persuade at least one or two in here to have the decency to comment on his next one instead of perpetuating their own self-indulgent rants.

    If so, looks like it didn’t work.

  7. Seventh :|]

  8. Davwel

    The positioning in Joanna Cherry’s statement strikes me as interesting.

    It doesn’t just seek to justify the Inner House decision, but concentrates on the justiciability of the Privy Council’s advice to the monarch in this case.

    The Scots constitutional tradition is fully laid out so that the UK Supreme Court would need to either ignore, or find a mechanism to dismiss, Scots legal principles, were they to rule against the Inner House.

    In that sense, it’s a very political statement – but one that drives to the heart of the different (and unreconciled) bases of governance in the UK Union.

  9. “PETERW
    I was wondering if AW had left it an unusually long time between thread heading articles in the hope it might persuade at least one or two in here to have the decency to comment on his next one instead of perpetuating their own self-indulgent rants.

    If so, looks like it didn’t work.”

    Is your own post meant to inspire us all as to how it actually should be done???

  10. Anthony notes “the difference may just be down to some pollsters having better quality, more representative panels than others.”

    I remember someone previously noting that one pollster (I forget which one) having a panel that was dominated by members whose interest was mostly in consumer/brand issues.

    Not that such folk should be denied the right to comment politically!. but it would seem reasonable to assume that, since panels are composed of (filtered) volunteers, if their motivation for joining varied, so might their ability to represent the whole population – even if they are weighted by the relevant demographics.

  11. @Oldnat

    As someone who submitted to a YG panel for a month or so, then faced multiple consumer surveys and zero political ones, I was of the opinion, that to continue as a panel member, one HAS to be a consumer-minded individual (or extremely patient).

    As for false recall, there’s nowt as queer as folk. While I sometimes struggle to remember which party got my vote in 2010 (can’t remember if I voted with heart or tactically), the rest are reasonably easy to remember (back to ’92).

    If however the average punter struggles to remember after 12-24 months, I would dump election votes older than 12 months as a measure of current VI. The current changes in party politics will make it all the more difficult anyway. Half won’t have a clue who they voted for with 4-5 parties equally vying for votes.

    It might be easier to prompt for ‘did you vote in past 1,2,3 elections?’, which would create a reasonable grasp of certainty to vote. I assume most, if not all here are 10s, with exception of active spoilers. Our echo chamber is not representative, sadly.

  12. @Oldnat

    As someone who submitted to a YG panel for a month or so, then faced multiple consumer surveys and zero political ones, I was of the opinion, that to continue as a panel member, one HAS to be a consumer-minded individual (or extremely patient).

    As for false recall, there’s nowt as strange as folk. While I sometimes struggle to remember which party got my vote in 2010 (can’t remember if I voted with heart or tactically), the rest are reasonably easy to remember (back to ’92).

    If however the average punter struggles to remember after 12-24 months, I would dump election votes older than 12 months as a measure of current VI. The current changes in party politics will make it all the more difficult anyway. Half won’t have a clue who they voted for with 4-5 parties equally vying for votes.

    It might be easier to prompt for ‘did you vote in past 1,2,3 elections?’, which would create a reasonable grasp of certainty to vote. I assume most, if not all here are 10s, with exception of active spoilers. Our echo chamber is not representative, sadly.

  13. Statgeek

    “to continue as a panel member, one HAS to be a consumer-minded individual (or extremely patient).”

    Or – prepared to spend a few minutes every so often answering (or not in my case) daft questions like “Which of these brands would you be proud/ashamed to work for?” and racking up a few more points to get to another £50 to spend on the grandweans – Ah! The luxury of retirement!

  14. Thanks for the new thread AW. I particularly agree with this bit of your comments:

    “Whatever the position is now, it is likely to be transformed by whether or not we end up leaving the European Union next month, on what terms and under what circumstances.”

    The political situation is so febrile that I think that only polls after October 31st will have any relevance unless there is a GE first, and I think that could still happen if Parliament is recalled.

  15. ON
    “Or – prepared to spend a few minutes every so often answering (or not in my case) daft questions like “Which of these brands would you be proud/ashamed to work for?” and racking up a few more points to get to another £50 to spend on the grandweans – Ah! The luxury of retirement!”

    Me too! Yet another thing we have in common, as well as being grumpy old gits.
    —————-
    Now here’s a thought – if there were to be a GE before (or even after) Brexit actually happens, why should citizens of the RoI (and hence the EU) be allowed to vote, as they usually are? Their vote may only swing a few seats but surely the principle is important?

  16. @PETE B

    Surely the important principle is that they should be allowed to carry on as before? You can’t start self-selecting the electorate for a specific vote, that’s bonkers.

  17. @Pete B

    Only if Scotland can prevent English-born residents from voting in Indyref 2. They have to be biased.

    :D

  18. @JONESINBANGOR

    “Although the Tories have recovered fairly well, as was somewhat predictable once the exhausted previous PM was replaced, it is a very tenuous support base dependent on Brexit outcomes.

    If a Brexit deal isn’t reached by Halloween, the Tories really will be on the bonfire, and I expect a mass defection to Brexit Party.”

    So long as BJ doesn’t propose the Remain WA, he will keep the Brexit vote. Voters know his hands are tied.

  19. I wonder if the house effects are more prominent now because of the number of adjustments that are applied to the raw figures, it would be more costly, which is probably why it isn’t done. Wouldn’t an attempt to achieve a panel of a thousand, with the correct demographic be a better way? As a guess obtaining info from 1500 people could allow for age, class etc. to be accounted for, wouldn’t there be less need to rely on previous voting etc? Or is MRP the only way forward to accuracy? If so isn’t the difficulty that as a method it doesn’t perform the shop window effect for other businesses that use polling for market research as MRP as a method is unlikely to be (a) cost effective (b) necessary for that type of research.

    These are genuine questions from a lawyer not a statistician (the Roy Meadow problem shows how little lawyers understand statistics). Statisticians can probably tell me why I am wrong and why, statistically, a lawyer is likely to be wrong on (x – y) + (a x 4.33) = 22 occasions. :-)

  20. The TREVS will love this twitter thread from Jon Worth. It contains Brexit Diagram V19

    https://twitter.com/jonworth/status/1172553353097863168

  21. Forgive the incoherence of the first sentence, I initially had the first and second sentence as one long one with a but in between. That looked too long so I made two without considering the sense of the first (too early I the morning)!

  22. Part of a tweet from Tony Connelly gives a part of an explanation of why Johnson did not attend the press conference in Luxembourg. Johnson’s staff wanted the conference moved inside to escape the noise of the hecklers, who in addition to heckling, had sung Ode to Joy at least ten times. That would do it for any sane person.

    “Luxembourg govt source says the request to have the Bettel Johnson news conference inside came at the last minute – it was, said the source, impossible for logistical and security reasons to relocate everything inside because the Lux PM’s office was too small to host all press.”

    https://twitter.com/tconnellyRTE/status/1173616177576366080

  23. The New Statesman has, I think,a good summary of the issues facing the Supreme Court.

    https://www.newstatesman.com/politics/brexit/2019/09/week-supreme-court-must-come-age

  24. At the risk of stating the blindingly obvious, surely the really important question (more than ever), is how the national votes are distributed at constituency level.

    If the ‘Boris bounce’ is winning back Tory voters in safe seats, it may not help the Tories to a majority: if the Lib Dems are winning over Remain voters from Labour in metropolitan areas, that might not be such bad news for the Conservatives; if the Brexit Party’s support is in Labour seats with a strong UKIP vote in 2015 that could also really help the Tories. And it wouldn’t need a huge increase in SNP support to counterbalance whatever happens in England (Jo Swinson could be vulnerable if the SNP can persuade Scots that they are preferable to an ‘English’ unionist party).

    Is there any evidence of regional rather than national trends?

  25. WB61

    Your first sentence made sense to me. The final sentence was beyond my comprehension. Mea culpa.

  26. Interesting post. If Labour voters from 2017 are less willing to recall their vote, then this clearly suggests that Labour has lost popularity. In failing to account for this, I would think this places doubt over ComRes’ results and suggests that the YouGovs are perhaps more accurate?

    @Oldnat – I think the media’s characterization of the Supreme Court’s job today as reconciling two conflicting rulings demonstrates their general lack of understanding of the UK.

    Yes, the two rulings are contradictory, but that doesn’t mean one of them is wrong – just that Scots law may be different to English law, thus both lower judgements were perfectly correct. The job today is to assess how we account for this in a UK wide context.

    In such an event, the judgement would (I imagine) have to rule in line with the Scottish case, for the simple reason that you can’t prorogue the English bit of parliament in isolation, thus meeting the law in both polities. As you said from the outset, before any judgements, the Scots case looked strong, but we’ll see.

    One other observation is that I can’t help thinking the No 10 briefings that they are considering ignoring the Benn act and prorouging parliament again may not help them in court today. It’s a strange briefing to put out there, if you are trying to convince a court that a 5 week prorogation was normal and for purely domestic management purposes, to then be briefing for another one within a week.

    Given our flexible constitution, the Supreme Court is being asked whether any manipulation of this process, for whatever reason, is abusive and illegal. If they side with the government, they are in effect giving a green light to a future government governing without parliament.

    I think I am correct in stating that there needs to be an annual financial approval mechanism, so HMG can continue to raise tax and fund itself, but if the English case is upheld, that would allow a government to recall parliament for a few days a year only. Such a position would be completely intolerable, and if the court thinks through the implications of such a judgement, I would think they must side with Scots law on this.

  27. Reading La K’berg on Podiumgate I was struck by the similarities between approach now and his approach to the referendum in 2016.
    K’berg says that despite questioning about what might change in the backstop to make a deal possible Johnson won’t say and when asked how he intends to get round the legal blocks on No Deal, he won’t say.
    I can see how he might think that his silence puts pressure on the EU and suggests to his supporters that he has some cunning wheezes that he’ll pull out of a bag to resolve the situation. But it could also indicate that he’s fresh out of pragmatic ideas and is reduced to bluffing it.
    In this he relies on the support of Leave supporters who, despite a mounting tide of evidence, continue to believe that the EU have been the cause of all, or maybe the majority, of our woes. It was the same sort of faith that allowed him to busk his way through the referendum campaign relying on bluff and bluster with little hard fact to add depth to what was a garnering of established prejudices.

  28. Jboyd,

    There is evidence of regional trends in the crossbreaks of the polls, and perhaps more securely in the EU election results. Lib Dems are doing better in London and the S than in the North, for example.
    In 2017 the Lib Dems took some target seats in London, the S and Scotland, but lost seats to Labour in Yorkshire and failed to regain Cheadle and Hazel Grove. That trend may continue, perhaps.

  29. @PeteB the right of RoI citizens to vote in elections in the UK pre dates either states’ membership of the EU and is a mutual right

    @Statgeek; the application of removing non-Welsh born voters from elections here (apart from being obviously unjust as they would be subject to taxation without representation) would be much more significant than doing the same in Scotland as there are simply so many more of them. I caused great offence on twitter when pointing out the implications of this different voting so I won’t bother doing again now….

  30. Oh and before BRXT erupts in righteous indignation, I mean ‘prejudice’ in their widest sense. Being upset by having lots of Continental EU citizens in your locality means you are prejudiced against them. Being upset by having to comply with regulations agreed as sensible and voted for by the UK but labelled as EU, is prejudice. Being cross with having to operate within restrictions imposed by commonly agreed policies on agriculture, fisheries etc means you are prejudiced against them etc, etc.

  31. @SAM
    “The New Statesman has, I think,a good summary of the issues facing the Supreme Court.”

    Although a somewhat one-sided view, given that it was written by a left-leaning barrister for a left-leaning publication.

    The key issue is not whether prorogation itself was unlawful; the Miller case is that the 5 week period of prorogation was excessive and disproportionate to any legitimate aim. But before even considering that, the Court has to decide whether the entire matter was justiciable, in other words does this Court, or any Court, have jurisdiction to make a ruling on the matter, or is it entirely a question for Parliament, a political issue in which the Courts should not be involved?

    Proceedings over the next three days will be televised on the UKSC website, for those who wish to follow the arguments.

  32. This business of panel membership is interesting.

    Obviously the panels help with false recall because Yougov can weight according to what people said at the time, but that means they are perhaps emphasising even more people who like to be on panels over those who do not.

    So are people who like filling in consumer surveys more or less likely to be swing voters? Personally, filling in questionnaires and such is one of my pet hates, and i am not keen on quizzes. When I stay in a hotel I ignore the requests for feedback (unless the experience was really bad or good).

    I suspect people working 80 hour weeks as an entrepreneur or high flying (or overloaded) academics are underrepresented, as are activists in politics, charities, churches etc who are very busy. Working parents may also be, although if you can make some small financial gains in vouchers or whatever then the hard pressed JAMS may be there.

    In voting intention the problem now is that a lot of people are saying they would vote differently from 2017. Hence past vote is not a good guide to future vote. Are the panel members more or less likely to be switchers, and does it correlate with demographics too? I have no idea! The only speculation I have is that people switching to Green may be underrepresented, pehaps?

  33. Alec: One other observation is that I can’t help thinking the No 10 briefings that they are considering ignoring the Benn act and prorouging parliament again may not help them in court today. It’s a strange briefing to put out there, if you are trying to convince a court that a 5 week prorogation was normal and for purely domestic management purposes, to then be briefing for another one within a week.

    Given our flexible constitution, the Supreme Court is being asked whether any manipulation of this process, for whatever reason, is abusive and illegal. If they side with the government, they are in effect giving a green light to a future government governing without parliament.

    I think I am correct in stating that there needs to be an annual financial approval mechanism, so HMG can continue to raise tax and fund itself, but if the English case is upheld, that would allow a government to recall parliament for a few days a year only. Such a position would be completely intolerable, and if the court thinks through the implications of such a judgement, I would think they must side with Scots law on this.

    I think you are right here. The SC cannot afford to rule that prorogation is not justiciable, for the reasons you cite and I think that the Scots judgement will give them a good starting point. What they might miss is that if they base too strongly on Scots Law, then in the event of Independence, their ruling could be overturned in rUK and they could paint themselves into a future corner if they rule too strongly against any English and Welsh dimension to justiciability.

    This then leaves them scope to decide whether this particular instance is bad enough to be justiciable.

  34. @pete b – “Now here’s a thought – if there were to be a GE before (or even after) Brexit actually happens, why should citizens of the RoI (and hence the EU) be allowed to vote, as they usually are? Their vote may only swing a few seats but surely the principle is important?”

    That’s absolutely right – the principle is important!

    We must use Brexit to demonstrate just how insular and narrow minded we have become, regardless of what we have allowed before.

    It’s that principle that counts!

  35. Shevii,
    “Also scratching my head on why the BXP vote is holding up so well- what exactly is it they want from Johnson- do they still not trust him to crash out? In the tables this is mostly coming from ex Tory voters”

    Well..has the BJ conservative comeback as born again leavers been less than perfectly convincing?

    Farage/BxP is still maintaining a line that bells and whistles brexit with everything which was promised is possible. The conservative line has diverged quite a bit from that. Could soft leavers be sticking with Farage, because he is still offering the ideal brexit, which con have given up on?

    Edge of Reason,
    “Interesting points. I’d agree that a 70% turnout seems far more instinctively plausible than an 85% turnout, but then it highlights the wider problem we have with polls (as we did in 2017) where they are so far apart that at least some of them have to be actually wrong in their methodology, not just a bit out with their sampling. And once we accept that, it’s an open question as to which ones are likely to be right,”

    There is another problem here, that pollsters do not ask enough respondents who dont vote. Granted, if they always do not vote, that is rather a waste of time if you re trying to get the result of an election, rather than finding out what people really think. But if the sample has already eliminated most peope who never vote, one might reasonably then apply a pretty high turnout figure, because you already eliminated the no-shows?

    Which raises the question, did you eliminate the right groups of non voters? Or perhaps, people are self selecting to answer polls because they are interested in politics. Usually thats ok, provided the parties react in the same way with similar proportions of committed voters. But if we then cut across political parties with something like Brexit, could the samples then be hopelessly out, because they have an innate bias towards the politically committed and are not representing committed leave/remainers.

    Accuracy relies upon the non voters staying non voters, and not interchanging with the group which is voters. Interchange means potential error.

  36. Program about smuggled cigarettes on R4 just now. Could Brexit mean a shortage of cheap smuggled cigarettes? How would that affect VI?

  37. The Supreme Court hearing is not perhaps anything at all to do with distinctions between Scots and English law. The Court of Session found,iirc, that prorogation was not normally justiciable but in the circumstances of this particular case it is because prorogation sought to escape the democratic scrutiny of parliament. This, the Court of Session said, is not a matter of Scots law, but common law.

  38. sam: The Supreme Court hearing is not perhaps anything at all to do with distinctions between Scots and English law. The Court of Session found,iirc, that prorogation was not normally justiciable but in the circumstances of this particular case it is because prorogation sought to escape the democratic scrutiny of parliament. This, the Court of Session said, is not a matter of Scots law, but common law.

    That does not ring true. The normal distinctions are between common law and statute law, I think and between Scots and English law, I am sure.

    So the Court of Session was saying that this is a matter of Scots common law. Common law propagates by precedent. Now I am open to correction on this, it can propagate between jurisdictions on strength of argument but not by unquestioned precedent.

    I think what the Court of Session has done is to frame the decision in common law rather than statute to give the Supreme Court an easier job to adopt their decision, because the Scots statute law will by its nature be less easy to uplift into UK law. [And, referring to my earlier post, using Scots common law rather than Scots statute law will make any SC judgement that prorogation is justiciable much more robust against Scottish Independence]

  39. Thanks for the comments AW.

    I think @”“Whatever the position is now, it is likely to be transformed by whether or not we end up leaving the European Union next month, on what terms and under what circumstances.” is the most sensible one I have read for some time.

    All parties are-to a greater or lesser extent-gaming the Brexit effect.

    Some will lose-perhaps fatally.

  40. @Andrew111
    Thanks: that’s interesting.

    The Lib Dem position is fascinating: they seem to be at risk of alienating some of their traditional support (e.g. the ‘car crash’ interview by the North Devon candidate) whilst picking up votes amongst metropolitan Remainers who voted for Corbyn in 2017.

    I think the election will be decided not so much by how the votes split between Left and Right, but how they are divided between Tories/Brexit Party on the one hand and Labour/LD/SNP/Green/PC on the other.

    But again that’s probably stating the obvious.

  41. JBOYD, ANDREW111

    This was posted in the last thread by BEDKNOBS, it gives a bit of a picture of where the shifts are taking place:

    https://sluggerotoole.com/2019/09/13/the-conservatives-are-failing-to-make-headway-in-the-brexit-supporting-north-and-midlands/

    Interestingly it looks like the Tories are set to pick up seats in London and Wales in particular, due to the relative collapse in Labour support in those areas. The worst place for them in terms of shedding votes is in the SE, which could put people like Steve Baker and Penny Mordaunt in danger. Of course this analysis was carried out based on Comres’ polls which, as AW points out at the top of the thread, have the worst figures for the Tories. If you were to repeat it with Yougov or Opinium then it would probably show the Conservatives winning swathes of the Midlands and North.

    SAM

    To answer your post in the last thread:

    So – that’s extreme?

    No less extreme than the DUP when they were insisting that there be no change at all for East/West trade. Thankfully they’re starting to soften their stance, time for the EU and Ireland to reciprocate. A propoer solution that respects the GFA would be somewhere in the middle.

    There are not alternative arrangements that work at present.

    There are no alternative arrangments that can meet the EU’s arbitrarily high requirements, that’s not the same thing as them being unworkable. Funny as well that you should post a link to a Lords’ report dating back to 2016, when A50 hadn’t even been invoked, as support for your statement. Quite a bit of water under the bridge since then. Still, this paragraph is interesting:

    The preferred approach, we believe, would be for the EU institutions and Member States to invite the UK and Irish Governments to negotiate a draft bilateral agreement, involving and incorporating the views and interests of the Northern Ireland Executive and keeping the EU parties fully informed as this negotiation proceeds. Such an agreement would then need to be agreed by EU partners, as a strand of the final Brexit arrangements.

    Sage advice indeed, it’s a shame the EU and Ireland didn’t heed it.

  42. @Andrew111

    I note that AW’s comments on false recall and the advantage of YouGov’s panel method are substantially the same as in my comment in the last thread (15/9/19 12:23 pm). In the UK, YouGov have about 1.2 million people on their panel, so there is plenty of opportunity for taking account of groups that form a small part of the overall population.

    Undoubtedly there will always be people who are under- or un-represented in a panel. This only matters if their inclusion would significantly change the results. So unless your entrepreneurs and academics would vote differently from their less heavily-loaded colleagues AND there are enough of them to change poll outcomes by more than about 3%, pollsters are justified in ignoring them.

    Indeed, as CMJ has shown, much useful information comes from trend analysis of a single pollster, rather than comparisons between different polling organisations with fieldwork at the same time.

  43. Better late than never, the timetable for the UKSC hearing:

    Robert Peston (@Peston) Tweeted:
    was unlawful. See attached for the details of this legal and constitutional blockbuster https://t.co/agyIopeyAM https://twitter.com/Peston/status/1173639875163279360?s=17

  44. TO, Alec, et al

    Re common law

    It’s important to remember that common law in each of the 3 UK jurisdictions developed separately, and have significant differences.

    https://en.wikipedia.org/wiki/Scots_law

    The obvious course for the UK Supreme Court would be

    to uphold both of the judgments being appealed to them :

    to observe that in the particular circumstances of the use of the prerogative to prorogue it is up to Parliament to resolve the differences between the 3 jurisdictions :

    to note that, now Parliament is no longer prorogued, that Parliament should define in statute the circumstances during which the executive can prevent them sitting.

  45. @peteb

    EU citizens are not able to vote in UK General Elections. Irish citizens are able to vote because of reciprocal arrangements which pre date the EU. Citizens of Commonwealth countries can also vote in UK General Elections if they have leave to enter or remain in the UK ( or do not require leave to do so).

  46. Lord Sumption’s opinion – that the English judgement is the more likely to be backed by the Supreme Court – is being quoted widely by supporters of the government, but I think they should take a little care in their interpretation.

    He has repeatedly said that it is clear that Johnson behaved ‘outrageously’, and, by implication, this makes clear that he accepts, like all the courts have done so, that Johnson l!ed about his reasons for prorogation, both to cabinet, voters and apparently the Queen. Sumption merely takes the view that this is a political, not a legal matter.

    Where the caution comes in is how Sumption describes judicial involvement in cases where there has clearly been a breach of normal standards. Here he repeatedly warns that where governments push the boundaries of acceptable behaviour beyond their breaking point, even while judges may ordinarily rule that such a sphere of action would not normally be justicable, in such cases of obvious abuse they may decide that the law as it stands is defective and needs updating.

    This is where government is pushing them this morning by refusing point blank to rule out a further prorogation. Where there is a matter of legal interpretation, especially within an unwritten constitutional context, judges certainly ought to me mindful of the ultimate implications of any decision they make.

    If they decide that the prorogation was based on unlawful advice and was thus unlawful, all they are saying is that PMs must not l!e to the monarch. Nothing else changes. If they choose the opposite path, they give free hand to all future PMs to do and say whatsoever they like, with absolutely no qualifying requirement or constitutional balance.

    Conservatives should think very, very carefully before wishing for such an outcome. No use complaining in the future if an unelected hard left PM abuses unwritten rules to ram through political measures that don’t have the support of parliament. Just think about what that might bring.

  47. Leftieliberal

    We are not talking here about the CMJ analysis of trends using a single poster (quite robust), but the potential sources of systematic error that lead to the big differences we are seeing between pollsters. Error there must be, because they cannot all be correct (which also means that averaging them is invalid)

    YouGov have a big panel, which means they can easily find enough people who voted for each Party in 2017 and so on. What they cannot do is weight for people who dont like filling in surveys. We have no evidence one way or another as to how significant that might be. Anthony already showed that false recall is not enough to explain all the variations, and the likelihood is that it is a set of factors that happen to pull in the same direction.

  48. GARJ,
    Thanks for reposting that Salmon of Data analysis.

    As you point out using Comres is one problem, as is using cross-breaks which have much larger errors and may not be accurately weighted.

    I think using the EU election as a calibration is worthwhile. Not that it is an accurate indication of GE results, but that the variations in relative position of the Parties should be in the same direction (ie Lib Dems did best out of any region in London in the EU election so I expect them to be best there in any set of cross breaks)

    The EU election results slso show big variations within regions, particulsrly in London where the Lib Dems won every Borough in a N-S stripe through central London, with Labour stripes to each side and BXP in the E and W.

  49. Andrew 111
    “YouGov have a big panel, which means they can easily find enough people who voted for each Party in 2017 and so on. What they cannot do is weight for people who dont like filling in surveys.”

    Yes – I often think this. The vast majority of surveys from YouGov are mind-numbingly boring consumerist rubbish, that I’m very queasy about. The points you get for wasting fifteen minutes are of no real value at all. What kind of person (who isn’t obsessed with politics) can be bothered with them? I hardly ever do myself any more.

    As you say, there’s no evidence about what effect that may be having.

  50. @alec

    Parliamentary approval is needed each year to supply money to meet departmental budgets ( the Estimates). IIRC it is also needed for the Finance Bill for taxation. It will be very 17th century if we descend into an executive v Parliament battle over supply and taxation.

1 2 3 89