The British Election Study have released their data from the election campaign waves today – one large wave straight after the election was called, a wave of daily rolling polls from throughout the campaign itself and a third large wave conducted straight after the campaign. All three of these datasets were collected online by YouGov (the face-to-face element of the BES is still to come). If you’re au fait with stats software like SPSS, Stata or R the raw data is available for download on the British Election Study site here.

There’s already some analysis of the data by the BES team here (a longer version of the article you may have seen on BBC this morning), focusing on how people changed their votes between 2015 and 2017, and between the beginning and end of the election campaign.

The article breaks down 2015 vote by Remainers and Leavers. Looking at how 2015 voters who backed Leave ended up voting in 2017, the Conservatives kept the vast majority of their 2015 leave voters and picked up over half of the 2015 UKIP vote (as well as a chunk of Labour Leavers). The collapse of UKIP wasn’t all to the Conservatives’ favour though, 18% of UKIP Leavers ended up moving to Labour.

Turning to the Remain vote, Labour were the clear victor: around a third of 2015 Tories who voted remain drifted away from the party, either to Labour or to the Lib Dems, but Labour also picked up a chunk of the 2015 Lib Dem vote and most of the 2015 Green vote. Of course, while this is easy to view through the prism of Brexit, that doesn’t necessarily mean Brexit was the main driver (to give an obvious example, yes – a large proportion of Green Remain voters moved to Labour… but a large proportion of the 2015 Green vote had already moved to Labour before the referendum, presumably as a result of the direction Jeremy Corbyn had taken the party).

More interesting is the movement during the campaign itself. 19% of people changed how they would vote between the start and the end of the campaign. This is not in itself unusual – in 2015 the figure was 17%, and according to the BES team it was higher in 2010 and 2005. The difference in 2017 is that this movement was overwhelmingly in favour of the Labour party, whereas at previous elections the churn largely cancelled itself out. Hence during the campaign we can see significant numbers of Tory votes, Lib Dem voters and, most of all, don’t knows moving towards Labour, but very little movement away from Labour.

In terms of explanations for the movement – while the voters Labour attracted during the campaign were those you’d expect to be the most receptive (that is, tending to be opposed to a hard-Brexit and left-leaning), the most obvious movement was on leadership ratings, that sharp collapse in Theresa May’s ratings and the steady increase in Jeremy Corbyn’s, and those people who moved to Labour during the campaign were also those who displayed the biggest increase in their perceptions of Jeremy Corbyn.

Ed and Chris’s full article is here.


372 Responses to “British Election Study release their campaign data”

1 2 3 4 5 8
  1. CARFREW

    I’ve found the tables of the 2017-06-05 ICM poll here. As I expected, no question of police numbers was asked. The media [or perhaps Reuters, who syndicated the story] obviously assumed causality.

    My guess would be that they were right to do so, but it’s a pity that the grauniad, who commissioned the poll, didn’t ask for that question to be added.

  2. @Carfrew

    At the start of the campaign the Lib Dems were polling at 10-11% – quite a few former Con/Lab Remain voters were looking to vote for them. That suggests the 2015 effect isn’t responsible, at least for some electors, but Corbyn’s decent campaign (and Farron’s poor one) put paid to any chance of a revival.

    The Tory vote is pretty hard – polling shows many were very unhappy with the campaign but voted Conservative anyway, partly due to traditional voting patterns, partly due to Brexit and partly due to a fear of Corbyn.

    Labour didn’t win many votes from the Tories compared to the start of the campaign – they mostly won them from Greens, UKIP and most of all from undecideds.

    Greens swinging to Labour was pretty inevitable from the moment Corbyn was elected.

    UKIP mostly went Tory as expected, but some of them were working class former Labour voters who felt betrayed by years of New Labour which they felt had done little for them. Many of them were raised to never vote Tory, and the collapse of UKIP along with Corbyn’s Euroscepticism and disconnect with the Blair years proved enough to send them back to their old party.

    The undecideds were the key, and absent any evidence to the contrary I don’t see why the Labour manifesto would have been more important than the Tory clown-car. Farron made sure the Lib Dems weren’t a viable option, UKIP and the Greens were dead, making it a two way choice for most undecideds, and looking at the campaign as a whole, it makes sense that most broke for Labour.

    If I was truly undecided going into that campaign, the combination of the Tory’s awful manifesto, May’s inability to connect with the electorate, Corbyn breaking through the Daily Mail inspired preconceptions, and the positive Labour manifesto would all have influenced my vote in that direction. People aren’t that stupid – they make up their minds due to a combination of factors.

    As it happens I knew how I would vote long before the campaign, but I’m not telling you how I did!

  3. Alec

    This…?..is the word of the Lord?

  4. Me

    “The arguments/debate over “reasons” seem a bit like trying to figure out whether one footy team won a match because they payed well or the other team played badly.

    In reality it’s all part of the same thing” [etc etc]

    A clear analogy [I would have hoped…] for the fact that there can be a variety of reasons for results, either in footy or politics because, by their very nature, they involve multiple factors.

    carfrew”@Paul Croft

    “In reality it’s all part of the same thing ”
    ——-
    People have been at pains to point out that it probably isn’t.”

    What that last line is supposed to prove I have no idea. If it is that “people” are of the collective view that there were actually NOT multiple factors at play, than I can only say that people have been in pain for nothing.

    Anyway, my last word on subject – except to say that I did actually go out and vote Labour, rather than just approve of their policies.

  5. New voter registrations also played a part in the result of the GE.
    Figures elude me but iirc lots of new voters registered befor the referendum and the GE.

    This may not have been a huge effect but it may have helped labour.

  6. BTW although I am not party to Labour Party election strategy but the idea that putting money into “safe” seats was a Blairite plot looks particularly disingenuous.

    At the start of the campaign many, many Labour seats looked vulnerable. Seats like Dewsbury, where I live, were universally expected to fall to the Tories like confetti and people added up the Tory and UKIP votes and concluded that seats like Huddersfield could be marginal. No-one expected Labour to be gaining Colne Valley, let alone Canterbury..

    Labour put a lot of resources into Dewsbury. Three excellent leaflets promoting the sitting MP delivered by Royal Mail before the expenses started – well over £50k. In contrast the Tory campaign was slow to start and never promoted their candidate, only Theresa May, but Labour can hardly have expected that degree of incompetence on the ground…

    As the polls changed it of course became apparent that Labour were going to do better, but still most people on here (politically aware people, mostly, bruised by previous wishful thinking over selected polls such as on referendum day) predicted an increased Tory majority. The fact that the Labour performance in the end matched more or less the most favourable polls for Labour was hard to anticipate, and I am sure Labour strategists and sitting MPs favoured a safety first approach to Labour held seats irrespective of whether they were Blairites. I will credit the Party with realising that Huddersfield was never under threat, BTW, and NOT putting in much effort there, but I still know several Lib Dem members in Huddersfield who voted Labour, not because of any love for Corbyn, but because they were genuinely worried Barry Sheerman might be replaced by a Tory..

  7. It’s the manifesto, stupid

  8. Hi Anthony Wells,

    My comments seem to be stuck in moderation on the constituency threads.

    Please could you help.

    Thank you.

  9. @Paul Croft

    “What that last line is supposed to prove I have no idea. If it is that “people” are of the collective view that there were actually NOT multiple factors at play, than I can only say that people have been in pain for nothing.

    Anyway, my last word on subject – except to say that I did actually go out and vote Labour, rather than just approve of their policies.”

    ———

    There may be multiple factors at play, but the point you are dodging is that it can be possible to screen lots of potential factors out, and settle on the few.

    To use a footy analogy, one factor affecting the outcome can be a player being sent off. But if it turns out you’re already three nil up when your goalie gets sent off and there’s only thirty seconds left on the clock, prolly didn’t affect whether you win or not.

    Then when you’ve boiled it down, there may be ways to figure out what was most influential from what remains.

    A footy example might be how England lost in Mexico against ze Germans in 1970. Some say it was the goalie, but if Sir Alf hadn’t taken off Charlton early, allowing the Germans to start steaming through, the goalie issue might not have become such an issue.

    Furthermore, yes, sometimes may not be able to definitively decide. But in the process you have at least determined it’s not possible to decide in this instance, have determined key factors, gotten better at analysing, maybe figured out some ancillary stuff etc

    (To use a music analogy, just to try and play a piece that’s beyond someone’s current abilities, may still benefit them.)

  10. @ANDREW111

    “BTW although I am not party to Labour Party election strategy but the idea that putting money into “safe” seats was a Blairite plot looks particularly disingenuous”

    ——–

    Well to suggest it’s portrayed as a plot when someone didn’t actually suggest that, they just noted that it might have happened, for whatever reason, would be especially disingenuous.

  11. carfrew

    others

    Little time – out for lunch

    Is this of use/interest? Page 15

    http://www.centreonconstitutionalchange.ac.uk/sites/default/files/news/One-year-on.pdf

  12. @ Andrew 111
    “How about a “10 posts in 24 hours” limit AW??”

    Haha

    Used to sit on this god-awful committee, went on for hours. We introduced the each-person-has-5 matches rule. Each intervention cost match & when the matches were gone . . .
    It saved a lot of time.

  13. @Barny

    Yes on reflection am prepared to accept maybe the LD campaign played a bit of a part, but clearly it’s a bit of a sideshow in terms of overall Lsbour gains.

    We also have to allow for summat that has become apparent reading Alec’s post. The question of what one means by a “good campaign”.

    I mean, there’s “good campaign” in terms of the mechanics of it. E.g. stealing a march by leaking a manifesto.

    Then there’s the policies. This seems to get lumped into “good campaign”, which carries the danger of minimising the contribution of the policies, or indeed Corbyn’s personability.

    So if Labour did pull votes from LDs, how much was this due to Farron laying an egg, to campaign tactics, and to the attractiveness of the policies?

    A potentially key point here in determining this is the one made earlier by Kester, who notes how Labour took votes from parties across the board.

  14. Robbiealive

    It probably improved the quality of the interventions too.

  15. There was a plot to unseat JC, this failed but had the unfortunate effect of destroying the confidence that many on the labour right may have had.

    They undermined their own confidence in a spiral of loathing and a priori regret.

    This had the effect that the NEC ran a defensive campaign.

  16. “Used to sit on this god-awful committee, went on for hours. We introduced the each-person-has-5 matches rule. Each intervention cost match & when the matches were gone . . .
    It saved a lot of time.”

    ———-

    Oh God, closet liberals and Blairites would love this. Especially given a situation where they already dominate a board and can share the load so the few who counter have to do proportionately more posts.

    Then they pepper with lots of false charges and straw men too, so that you have to make even more responses and then they complain about your number of posts!!

    Still, it’s all part of the challenge…

  17. @ Alan
    “It probably improved the quality of the interventions too.”

    It focussed things, that’s for sure!
    We called it the Oliver Cromwell rule, tho I have since been unable to find out if he really imposed such a rule when the Godly assembled.

  18. @CARFREW
    @THE OTHER HOWARD

    I find it amusing that to english speakers can have different interpretations of the student debt needs to be dealt with

    To my mind the interpretation is pretty much partisan:

    http://www.huffingtonpost.co.uk/entry/jeremy-corbyn-student-debt-general-election_uk_5975a97de4b09e5f6cd0619c

    he did say that student debt needed to be delt with as have a number of commentators after all it stand now at £100B and will be at £330 by 2024 as I understand it. Now does that mean he said he would wipe it out, clearly no one would say that is what he said so the problem I see is that there is a level of projection going on

    What I find interesting is that TOH would be happy to say that there is a projection exists with respect to this (because I am sorry it is projection pure and simple) but if as Lord Ashcroft said :

    “whatever was printed on the ballot paper the question large numbers of voters heard and the answer they gave had nothing much to do with the European Union … ultimately, the question many saw was: ‘Are you happy with the way things are and the way they seem to be going?’ And their answer was: ‘Well, since you ask … no'”

    We need to be careful of what we are taking as projection it is clear that some people according to the polls believe what he meant was that he would abolish the debt( one poll puts this at 17%) but the majority correctly identified that he made no such promise and to turn that iit needed to be dealt with into a promise to wipe it out is definitely good politicking but would be considered a fail in english comprehension.

  19. The match analogy is false as we are not bound by time or place on a forum such as this.

  20. @Pass The Rock

    “he did say that student debt needed to be delt with as have a number of commentators after all it stand now at £100B and will be at £330 by 2024 as I understand it. Now does that mean he said he would wipe it out, clearly no one would say that is what he said so the problem I see is that there is a level of projection going on”

    ———-

    As already pointed out numerous times, in the transcript it’s clear he went a lot further and gave examples that showed explicitly that he didn’t mean wiping out all the debt.

    This is another reason one has to make more posts, to keep pointing out what’s already been said to peeps who don’t read all the posts, even during the same discussion.

  21. @Carfrew

    Shrug. I don’t believe any one factor was the tipping point, you clearly do.

    I try to take a holistic view of these things.

  22. @Barny

    No, that’s to majorly mischaracterise my argument, to say I think there’s just one factor responsible. Indeed initially came in to counter that idea, specifically over Brexit.

    I’ve also made clear there’s evidence that both Corbyn himself, and his policies played a part, and ceded the LD thing somewhat, then there’s tactical stuff I mentioned like leaking the manifesto early…

  23. @ Carfrew The 5 match rule.

    “Oh God, closet liberals and Blairites would love this. Especially given a situation where they already dominate a board and can share the load so the few who counter have to do proportionately more posts.”

    You see “Blairite” plots everywhere — weird: & this Stalinist habit of lazy labelling. . .
    he rule was adopted for ONE committee & it was designed to stop two garullous people, nicknamed East & West Prussia, from going on and on and on. — after one meeting lasted from 2,00-7.30.
    This committe did not have coffee breaks: clearly a Blairite plot to undermine Marxist-coffee-producing states in central America.

  24. @Carfrew

    I apologize, I meant only that I understood you felt the Manifesto was the tipping point, not that it was the only responsible factor.

    My only point of contention is that I believe the poor Tory campaign was just as influential.

  25. @Robbiealive

    No I don’t see Blairite “plots everywhere”. I just noted a specific issue with dominating a board and peppering with distractions, as you are still doing. Whatever floats your boat, but don’t be blaming me for it!

  26. @Barny

    That’s ok. I don’t have a cast-iron position on the matter, to me it’s something that evolves as peeps keep bringing things to light. Hopefully in the end one can settle on summat, albeit tentatively…

    I don’t think the Tory campaign had zero effect, I just don’t think it’s necessarily that big a deal, for reasons already given by me and others.

  27. @CARFREW

    I do find it strange, where this whole idea came from and more straneg that english speakers are having difficulty in comprehension. Whatever ones view of Corbyn it is hilarious to see this level of projection over something so clear.

    What actually scares me is that if we have to debate data and facts then we have a real problem having a level playing field to discuss opinions.

    Now I accept some people would interpret what Corbyn said as wiping out the debt despite what he said but then if someone said that people voted leave on projections of what the leave campaign said, would be met by howls of derision

    Still I think the strategy was to muddy the waters of what a policy that the Tories have extended such now your average student will be close to £50k in debt. I fear that the problem of student debt, the loans, interest rate and the fact that fully one third will not be paid back seems have been lost in the politicking of it all

  28. barny

    “Shrug. I don’t believe any one factor was the tipping point,”

    To add to that point, and observe just how silly it is to try and list the “reasons for voting” in some sort of definitive, prioritised list, my feeling is this:

    Most of the small minority of people who were either non-voters or uncommitted would probably find it difficult to say [or actually, even know] what was the most important factor in their decision, never mind “expert”, armchair analysts.

    Further, I am not actually convinced that there really is a “tipping point” for these decisions: rather a build up of a feeling to go one way or another.

    What is pretty certain, given that virtually everyone does agree that the Tory campaign was dreadful and TM rather worse, is that they are holding on to an incredibly solid chunk of current voters and that these are not likely to move elsewhere – as things stand.

  29. @Andrew111
    “the idea that putting money into “safe” seats was a Blairite plot looks particularly disingenuous. At the start of the campaign many, many Labour seats looked vulnerable. Seats like Dewsbury . . ”

    The bookies had the Tories heavilly odds-on to win dozens of Lab seats at the beginning of the campaign, including Dewsbury, Halifax etc, usually as you say where the UKIP 2015 vote was > than the Lab majority. Dewsbury was a marginal by any standard of course. Lab funding must have followed the pre-campaign polls I guess.

    I think we went round the houses on tuition fees & it was concluded that School Funding was as important, particularly for 25-45 females who disproportionately shifted to Lab.

  30. @Pass The Rock

    Read the full transcript! Don’t just go by the selective quoting of “deal with it”. He clearly gives unambiguous examples.

  31. @ Paul Croft
    “To add to that point, and observe just how silly it is to try and list the “reasons for voting” in some sort of definitive, prioritised list”

    Yeh. Causal explanations come in all shapes & sizes, the “prioritised list of factors to be taken into account” the most common. Often it’s the last thing that happened that is prioritised: the window broke because the vandal smashed it with a stone. Or the penultimate thing. The security gurad took a sickie & hence the vandal wasn’t stopped. Then the infinite regressions: the builders contracted to put in toughened glass, which would have withstood the stone, but defrauded.

    I’ve run out of matches.

  32. @Paul Croft

    To add to that point, and observe just how silly it is to try and list the “reasons for voting” in some sort of definitive, prioritised list, my feeling is this:
    Most of the small minority of people who were either non-voters or uncommitted would probably find it difficult to say [or actually, even know] what was the most important factor in their decision, never mind “expert”, armchair analysts.
    Further, I am not actually convinced that there really is a “tipping point” for these decisions: rather a build up of a feeling to go one way or another.
    What is pretty certain, given that virtually everyone does agree that the Tory campaign was dreadful and TM rather worse, is that they are holding on to an incredibly solid chunk of current voters and that these are not likely to move elsewhere – as things stand.”

    ———

    On a polling site, people are apt to want to explore what goes on with polling. Bit pointless to keep complaining about it. Would you go on a footy site and complain if they discuss what influences footy results? Cos they do an awful lot of that!

    It’s true some people may not always know the reasons for a decision, but that doesn’t mean you can’t ever tease out what influences them nonetheless. And also rule some things out.

    There clearly can be tipping points, check Black Wenesday for one of the more obvious examples.

    Even if it’s a collection of things, it’s still useful to work out the collection.

    Once again, no doubt, it’s tricky, but that doesn’t make it all impossible or worthless.

  33. “Yeh. Causal explanations come in all shapes & sizes, the “prioritised list of factors to be taken into account” the most common. Often it’s the last thing that happened that is prioritised: the window broke because the vandal smashed it with a stone. Or the penultimate thing. The security gurad took a sickie & hence the vandal wasn’t stopped. Then the infinite regressions: the builders contracted to put in toughened glass, which would have withstood the stone, but defrauded.”

    ———

    Many, many things are constantly improved by trying to attend to all the factors, and work out which are the most salient. The fact that some may struggle to handle such systems does not make the exercise inevitably worthless.

    Polling itself tries to do this, hence the new Yougov model that has thus far proven quite successful. The BES survey is indeed useful ammo in the process of improving polling. So you can’t really expect people not to discuss it or its implications.

  34. These were the main points of the Conservative Manifesto, how many will be delivered

    Key policies
    Real terms increases in NHS spending reaching £8bn extra per year by 2022/23
    Scrapping the triple-lock on the state pension after 2020, replacing it with a “double lock”, rising with earnings or inflation
    Means test winter fuel payments, taking away £300 from wealthier pensioners
    Raising cost of care threshold from £23,000 to £100,000 – but include value of home in calculation of assets for home care as well as residential care
    Scrap free school lunches for infants in England, but offer free breakfasts across the primary years
    Pump an extra £4bn into schools by 2022
    Net migration cut to below 100,000
    Increase the amount levied on firms employing non-EU migrant workers

    It’s interesting that some people were fixated on a promise Corbyn never made, but even if he had was not in Government to implement it, but seem to turn a blind eye to actual promises (many of which since the election have been binned) the Conservatives made in black and white in their Manifesto, who are of course in Government.

  35. O happened to play around with the data. Statistically Brexit is far the most important, but it is partially a derivative of the questions.

    There were obviously a number of factors, but the ad-box inclusion or rejection of dominant influences because “there were other factors” is just agnosticism (in the philosophical rather than the religious sense of the worr) and aims at closing the discussion.

    As everything is related to everything, the resultant (the election outcome) actually has an infinite number of factors. However, stating this is equivalent of saying of not knowing (as we cannot know the infinite). Also, defining the resultant doesn’t require all the factors, it likely requires very few factors.

    While the election outcome is homogeneous, the voters aren’t. It is clear from the BES data that different factors affected different groups, but as the structuring was done from the perspective of the outcome (a kind of (not real) reversal of cause and effect), which have the prominence to Brexit. However, it is very likely that Brexit had an important role in YouGov’s model too, but the socio-economic factors were also likely be more pronounced than in the BES data.

    The hunting for the factor or factors, when the narrative is provided, however, cannot ignore these structural differences of voters, so the relative importance of the factors could vary significantly. From the data it’s not possible to disaggregate it (without the begging the question fallacy).

    So, I think, that the various dominant factors (Brexit, election promises, JC vs TM, economic conditions, discontent vs being contented) in their overall negative and positive effects on particular groups resulted in the outcome. This is very different from saying that more people were affected – it accounts for socual, economic, geographic and accidental circumstances, that is, assumes some objective basis, coloured by individual, spectrum-like differences.

    There was also a factor – negative choice instead of a positive one. While it is a common sense, it is not really shown in the data (apart from Brexit – another reason for the emphasis on it).

    Then there are individuals whose choice is not based on relative importance of particular factors, but certain factors create a refusal of the particular party, which may mean no vote or switch. The data cannot show this as the whole thing is calibrated for defining the relative importance of the factors, while for these people the presence of one factor nullifies the importance of the other factors.

    So, the BES data shows by design the importance of Brexit, and it is probably true. The churn indicated by the data suggests the importance of the existence of a distinctly anti-Tory platform (attributing this to JC, manifesto, etc is meaningful, and probably right, but using the data is problrmatic. While I think JC’s performance was probably the key factor in his rise in approval, but the rise of Labour could have contributed – just as in the fall of TM’s approval could have been partially caused by the drop in Conservative support. If this is true, the importance of the manifestos is bigger than suggested by the data. The same applies for the higher turnout – the referendum, the campaign performance of the leaders and policy promises would not necessarily be independent of each other.

    So, I can’t support my own view with the data – the dominant factor was the gradually refined and better broadcast development of an anti-government (and hence anti-Tory) platform – which was still not enough to get the majority.

    Considering the churn in the data, I don’t see from where the needed vote would come from, unless the differential turnout helps this platform. And, of course, events.

  36. Carfrew

    It’s a shame that the details of the Yougov regression model are hidden from sight. In the same way that some transparency is demanded by the BPC. As it stands it is no more than a black box which produced a good result. It makes it difficult to argue strongly in favour of the model due to this lack of clarity (only to dismiss the obviously false criticisms of the model from people who were obviously clueless about how linear models work).

    As you say, proving causation is difficult due to the sheer number of factors which confound the relationships, introducing new factors might totally change the relative importance of these factors.

    Models pick out patterns in the data, they don’t necessarily explain the data.

  37. LASZLO

    ‘Considering the churn in the data, I don’t see from where the needed vote would come from, unless the differential turnout helps this platform. And, of course, events.”

    The needed vote would not come from one factor but from progressive spread and adoption of the recognition that it would be good and feasible idea to get rid of this government.

  38. @ Andrew111

    ‘although I am not party to Labour Party election strategy but the idea that putting money into “safe” seats was a Blairite plot looks particularly disingenuous”’

    Try telling that to the MPs like Margaret Greenwood or Kelvin Hopkins (Corbyn supporters) who received no funding whilst their neighbouring MPs, Alison McGovern and Gavin Shuker (prominent Progress members) had so much election material that it couldn’t all be distributed.

    Try telling that to my colleague who saw the face of her Blairite PPC fall when the exit poll was announced.

    Try telling that to the successful candidates in marginal seats like Brighton Kemptown or Sheffield Hallam who were left unfunded but ended up with majorities of 9k and 2k respectively (and unseating Nick Clegg).

    Try telling that to the thousands of LP members who have watched two years of dirty tricks, negatives briefings and party grandees trying to dislodge the labour leader with 62% of their votes.

    Of course, you may be right that there was no plot but to say that that the idea is ‘disingenuous’ is disingenuous of you.

  39. @Sam

    “Is this of use/interest? Page 15”

    ————-

    Well it would help if you quoted the specific bit, bit that page opens with this:

    “Two claims were made during the referendum
    campaign concerning the implications of Brexit for
    the UK’s regions, cities and nations.”

    I.e. it goes on to talk about the impact during the referendum campaign, not during the 2017 GE.

  40. @ John Pilgrim

    Thank you so much for your response on the last thread… I did reply yesterday but (inevitably) AW started this new thread. I am aware that your sharing of your knowledge on the thread (and also from others) is a great example of what I believe is called ‘generalised reciprocity’.

  41. @Alan

    I accept your qualifications regarding the Yougov model. Not that this refutes my point about the potential benefits of analysing such systems of course. It’s more that you’re saying that to some extent keeping the model to themselves frustrates the abilities of others to do this.

  42. neilj

    I blame jezza for all the tory promises which will not now be implemented. There we all were thinking that he would campaign like he did in the referendum and even possibly go on holiday during it and then he goes and plays a blinder. Who would have credited it? and now in consequence we have a tory government but without any of the giveaways!
    when they come to write the Ladybird book of Corbyn’s achievements this will surely feature in the index.

  43. @Sam

    Reading some more, that bit of the document does look at things like whether economic predictions made during the campaign came to pass, normal stuff about impact of Brexit, which would get priced in before the GE, but haven’t found anything specifically about the impact of Brexit during the campaign to explain Corbyn’s rise etc.

    Have you got anything in the document on that?

  44. If I were May I’d have called the snap election for May 4th and made the campaign slogan ‘May the Fourth Be With You’ as opposed to ‘Strong and Stable’.

  45. Alienated labour

    I miss the perspective of AL on this survey. it is a real shame that he is on holiday at Workers Collective Camp 3 in the Republic of Uzbekistan.I am sure the dam will benefit from their exertions. I will await the trip advisor report with interest.

  46. PAUL CROFT

    “What is pretty certain, given that virtually everyone does agree that the Tory campaign was dreadful and TM rather worse, is that they are holding on to an incredibly solid chunk of current voters and that these are not likely to move elsewhere – as things stand.”

    Except that you could almost make the reverse argument about Corbyn. He’s still out on the wing of British politics and it’s still true that UK elections are usually won from the centre. He still doesn’t command his party. He still appears Mr Marmite. He still has a weak front bench because half the plp won’t serve on it.

    All the reasons we might have said two months ago (indeed people did here) that it was incredible he was even holding at Milliband levels, and that it was hard to see anything better – as things stand.

    Things didn’t stand.

  47. I genuinely believe a May 4 election would’ve resulted in a Tory landslide – having a longer campaign through May and into June allowed Corbyn to make inroads and win voters.

  48. Carfrew

    I wasn’t intending to refute but to qualify, overall I am in agreement with you.

    Yes, I agree very much that opening up the model should lead to discussions about omitted variables etc. which could lead to improved models.

    If you want to model something, then exploring what your model does and doesn’t contain, identifying areas of concern where you might be measuring the wrong thing, or making decisions about which interaction terms to include is all part of the process.

    Particularly with regression models, variable selection is a challenge and can often be informed by expert guidance. The good thing is any suggestion is testable and so any changes to the model can be justified by the data.

    A model is only as good as the collected data, it won’t be able to tell you anything about the impact of data which you don’t collect. That’s where discussing the model can be of a huge benefit. To do that, the model needs to be more than a black box.

  49. @ Paul Croft

    I pretty much agree with you about tipping points and ‘a sort of a feeling’ (neurologically speaking our brains process ‘stuff’ and then let’s a version of that stuff become conscious).

    However, I think you’re forgetting the former Ukip voters when you write:

    ‘What is pretty certain, given that virtually everyone does agree that the Tory campaign was dreadful and TM rather worse, is that they are holding on to an incredibly solid chunk of current voters and that these are not likely to move elsewhere – as things stand.’

    I accept that Ukip seem pretty much finished atm but isn’t it
    likely that disillusion with the Tory transition policy might affect former Ukip support negatively for the blues? Is it also completely impossible to imagine the return of Farage and a rejuvenated party?

1 2 3 4 5 8