General article(s) on Polls

[john07]

Apparently the Labour lead in the polls is exaggerated according to Robert Hayward:

www.theguardian.com/politics/article/2024/may/27/labour-polling-lead-conservatives-overstated-tory-election-expert

The Shy Tory syndrome revisited. Apparently some Conservatives and reluctant to admit that they support such a bunch of incompetent losers. That is until they get into the secrecy of the polling booth. And who can blame them?

[nodealbrexiteer]

May 27, 2024 19:05:14 GMT john07 said:

Apparently the Labour lead in the polls is exaggerated according to Robert Hayward:

www.theguardian.com/politics/article/2024/may/27/labour-polling-lead-conservatives-overstated-tory-election-expert

The Shy Tory syndrome revisited. Apparently some Conservatives and reluctant to admit that they support such a bunch of incompetent losers. That is until they get into the secrecy of the polling booth. And who can blame them?

on a serious note doesn't this make you uneasy?

[batman]

He does say it isn't the Shy Tory effect. But I see very limited evidence for what he is saying. There is a limit to how much local elections can be a guide to general ones as there is clearly ticket-splitting. Certainly if the Tories continue to have this dreadful a campaign I can't imagine that the undecided voters will break for them to the extent that he hopes.

[John Chanin]

Rob Hayward is a loyal Conservative and is trying to raise his party's spirits, which is fair enough. Any prediction of what currently undecided voters might do is speculative, although you can reasonably look at past experience as some polling companies, who are producing predictions rather than opinions, are doing. But past experience can only ever be a guide - things may turn out differently this time, and it is not as though there is a hugely consistent trajectory of undecideds in the past either.

[robert1]

'Polling industry expert agrees' might be a slight hint.

The Guardian doesn't publish articles on the basis of what 'he hopes' nor on 'speculation'.

The last public comment I made on accuracy of the polls was in 2016 when I predicted a leave vote. I voted remain. 

[jamesdoyle]

I think it's unfair to robert1 to suggest partisanship might be shaping his views here: he's always been rigorous about his analysis as far as I've seen, and his input is very valuable. If we ignored people with expertise because we disagreed with their political views, we'd miss out on a whole lot of useful stuff - where would I be without Pete Whitehead's knowledge?
Having said that, I would be interested to know if this hypothesis is based on more than just the 2019 voting habits of the current Don't Knows. Robert may have done more analysis  than me, but I didn't see any evidence in the May local results to suggest the Tories might do better than expected in the GE - if anything, it seemed to me to suggest the opposite.
And my hypothesis is that the DKs are more likely to be shy anti-Tories - previous Tory voters who can't even admit to themselves they'll be voting for a different party. It occurs to me that drilling deeper than 2019 into those DK voters might be profitable. If my idea is correct, then I think they would turn out to have consistently voted Con in previous elections, whereas Robert's theory might hold up better if they are recent switchers to Con.

[robert1]

jamesdoyle -Thank you for those generous comments.

The figures I used were subject to scrutiny by both Peston and the Guardian. I will go into more detail in the future. However, in general terms my analysis looked at the movement in the polls over the last two years and the apparently contradictory figures in local government election results.

The Guardian doesn't have a 'house' polling company and was therefore more likely to accept 'criticism' of polls than several other outlets.

BTW if I was aiming to cheer up the Conservatives would I really have my thoughts published in the Guardian? 

[graham]

I always look forward to Robert's informed comments. Re- the 2019 Tory vote - it occurs to me that there were so many traditional Labour voters who voted Tory that year simply 'To Get Brexit Done!'. They had never voted Tory before that election , and it is probably unlikely that many will do so again.I wonder to what extent this group might now appear in the 'Don't Know' column. To the extent that they do, I would have thought they were more likely to Vote Labour or not vote at all - than Vote Tory again.

[nodealbrexiteer]

May 28, 2024 10:39:11 GMT graham said:

I always look forward to Robert's informed comments. Re- the 2019 Tory vote - it occurs to me that there were so many traditional Labour voters who voted Tory that year simply 'To Get Brexit Done!'. They had never voted Tory before that election , and it is probably unlikely that many will do so again.I wonder to what extent this group might now appear in the 'Don't Know' column. To the extent that they do, I would have thought they were more likely to Vote Labour or not vote at all - than Vote Tory again.

I'd never voted Tory at any election before or since so this chimed with me though I wouldn't rule out a Tory vote again in the future

[John Chanin]

I did not mean to suggest that robert1  was anything other than a serious analyst of polling data.  As it happens I agree that opinion polls are likely overstating the Labour lead.  But there are other opinions from equally serious analysts out there, and the extent to which historical experience, or indeed local election results, can give guidance is very limited.  The fact is that it remains speculation.  And Conservatives do need something to cheer them up, given a rather accident prone start to their campaign.  

[Adam in Stroud]

May 28, 2024 10:29:04 GMT robert1 said:

jamesdoyle -Thank you for those generous comments.

The figures I used were subject to scrutiny by both Peston and the Guardian. I will go into more detail in the future. However, in general terms my analysis looked at the movement in the polls over the last two years and the apparently contradictory figures in local government election results.

The Guardian doesn't have a 'house' polling company and was therefore more likely to accept 'criticism' of polls than several other outlets.

BTW if I was aiming to cheer up the Conservatives would I really have my thoughts published in the Guardian? 

I echo jamesdoyle's comments in full and in particular would be interested in your reasoning with regard to local election results. In particular, I wonder how you are factoring in the rather limited availability of Reform candidates in local elections compared to the numbers we expect to see in the General Election (plus the rather greater salience Reform can expect via national media and the largely national policy issues they raise, as opposed to the dogshit-and-potholes aspect of local election campaigns, where I feel their lack of established local teams must hamper them a lot.)

I think one thing we can all agree on is that the huge numbers of 2019 Tory voters now in the Undecided column is a massive headache for the polling industry. There's a big chance that their snapshots of the situation at the date of survey will differ significantly from movement of Undecideds between then and polling day, setting them up for a mass of "pollsters got it wrong" hot-take articles on the weekend after.

On the basis that most elections are decided before the campaign begins, that Undecided pool seems to me where the actual campaign will be won and lost.

Do we have any information on the demographics of that pool?

[John Chanin]

May 28, 2024 12:18:41 GMT Adam in Stroud said:

May 28, 2024 10:29:04 GMT robert1 said:

jamesdoyle -Thank you for those generous comments.

The figures I used were subject to scrutiny by both Peston and the Guardian. I will go into more detail in the future. However, in general terms my analysis looked at the movement in the polls over the last two years and the apparently contradictory figures in local government election results.

The Guardian doesn't have a 'house' polling company and was therefore more likely to accept 'criticism' of polls than several other outlets.

BTW if I was aiming to cheer up the Conservatives would I really have my thoughts published in the Guardian? 

I echo jamesdoyle 's comments in full and in particular would be interested in your reasoning with regard to local election results. In particular, I wonder how you are factoring in the rather limited availability of Reform candidates in local elections compared to the numbers we expect to see in the General Election (plus the rather greater salience Reform can expect via national media and the largely national policy issues they raise, as opposed to the dogshit-and-potholes aspect of local election campaigns, where I feel their lack of established local teams must hamper them a lot.)

I think one thing we can all agree on is that the huge numbers of 2019 Tory voters now in the Undecided column is a massive headache for the polling industry. There's a big chance that their snapshots of the situation at the date of survey will differ significantly from movement of Undecideds between then and polling day, setting them up for a mass of "pollsters got it wrong" hot-take articles on the weekend after.

On the basis that most elections are decided before the campaign begins, that Undecided pool seems to me where the actual campaign will be won and lost.

Do we have any information on the demographics of that pool?

One thing that would be fascinating and would bear on this, is if the undecided former Conservative voters were split between those who voted at the local elections and those who didn't.  It would only require one extra question.  Which polling company is willing to explore this?

[The Bishop]

Surely a lot of the "don't knows" ultimately aren't going to vote, as has been the case for all GEs in my political lifetime.

[Adam in Stroud]

May 28, 2024 12:47:18 GMT The Bishop said:

Surely a lot of the "don't knows" ultimately aren't going to vote, as has been the case for all GEs in my political lifetime.

I think it is the "don't knows" who did vote in 2019 (especially the ones who voted Conservative, who, I understand, substantially outnumber the ones who voted Labour) that robert1 is focussed on, with the assumption that these are people who do normally turn out.

Of course you are right in that sit-on-hands is always an option. Differential turn-out rather than switching may be the story of this election. I think it is a fairly common experience among canvassers of all parties that disillusion with all politics is very widespread this year.

[batman]

Are you saying that ALL polls are exaggerating the Labour lead, or just some of them? If it’s the latter what is your instinct about the correct figure, at least approximately? If you’re saying that even Opinium is exaggerating it I would have to be distinctly sceptical though I wouldn’t dismiss it altogether. Opinium’s last poll showed a top line figure of a 14% Labour lead.

[Sibboleth]

Two things. Firstly, even a fairly significant polling error would result in a terrible defeat for the government all the same: after all, the published polls are absolutely cataclysmic, more so than the media has really internalized. This is actually what happened in 1997. Secondly, I'm not sure if there's that much significance to 'don't knows' being more likely to have voted Conservative than not in 2019 as 45% of people who voted did...

[nodealbrexiteer]

May 28, 2024 13:57:22 GMT batman said:

Are you saying that ALL polls are exaggerating the Labour lead, or just some of them? If it’s the latter what is your instinct about the correct figure, at least approximately? If you’re saying that even Opinium is exaggerating it I would have to be distinctly sceptical though I wouldn’t dismiss it altogether. Opinium’s last poll showed a top line figure of a 14% Labour lead.

I'm not so articulate on the matter to post but see this passage from the link below(albeit over 20 years old):

www.ipsos.com/en-uk/polls-apart

Roger Mortimore has put his finger on what is the difference in philosophy between the approach that ICM takes, attempting to always 'forecast' the general election, year in and year out, while MORI takes the 'snapshot at a point in time' approach. As roger put it to an enquiring email on Saturday:



"The figures are measuring two different things. Between elections, the question on voting intention is effectively a hypothetical one. We try to measure the mood of the nation as a whole regarding the parties. We want to avoid adding any elements that might cause our respondents to be unrepresentative or different from the rest of the public; therefore we don't artificially remind them of the possibility of supporting other parties if such ideas are not in their minds already, because our respondents represent the wider public who do not have such thoughts either.



The function of the final polls in an election, though, is rather different. Our last poll has to be designed to provide as much information as possible to predict the result of the election; here we are interested in finding out what the voters will actually do in the polling booth, and need to confront them with the real choice that they will face so as to judge their reactions. Even in an election-eve poll, we are not really concerned with what the nation as a whole thinks on election-eve but with what those who get to the polling booths will do on Election Day. Therefore a different measuring tool is needed, and the results may be different.



Between the moment where the election is first called and the final poll, it is more of a grey area; we are no longer asking about an entirely hypothetical election, but nor are we predicting the future. We took the view that the most appropriate point to switch methods was once the final lists of candidates were available, at which point the various local factors can be properly brought into the equation. In particular, electors can begin to make their calculations about tactical voting and to realise which other parties apart from the big two are standing in their constituency.



Incidentally, it would be easy to misunderstand the implications of the difference in the figures by thinking in terms of the Labour lead over the Conservatives. Simply because the Labour lead is lower on the new poll figures, this does not imply better news for the Conservatives. In fact the Conservative share of the vote has not been affected; but there is a shift from Labour to the Liberal Democrats and smaller parties. If anything, this implies a worse position for the Tories as the Liberal Democrat increase will be greatest in the seats where the Conservatives and Liberal Democrats stand first and second, where Labour votes would be wasted in any case -- as I wrote at the start of the campaign: www.ipsos-mori.com/election2001/ec0508.shtml

[iain]

I would also note the history of parties where lots of their previous supporters were 'don't knows' at this stage of an electoral cycle:

[greenrichard]

May 28, 2024 10:29:04 GMT robert1 said:

jamesdoyle -Thank you for those generous comments.

The figures I used were subject to scrutiny by both Peston and the Guardian. I will go into more detail in the future. However, in general terms my analysis looked at the movement in the polls over the last two years and the apparently contradictory figures in local government election results.

The Guardian doesn't have a 'house' polling company and was therefore more likely to accept 'criticism' of polls than several other outlets.

BTW if I was aiming to cheer up the Conservatives would I really have my thoughts published in the Guardian? 

Might I suggest the apparent contradiction in local elections results and Westminster Voting Intention could, in part, be explained by the fact they are measures of two different things. So often I speak to voters on the doorstep who say they will definitely vote Green in local elections but tell me openly they will vote Labour or Conservative in the General election. I’ve had party members say the same thing to me. The rise of Greens and the recovery of the Lib Dems at local level is because they target specific wards and divisions. They can achieve 40, 50, 60%+ in places where they struggle to them keep their deposit in the general election. A good illustration of this is in the 2015 General Election in Cannock Chase held on the same day as the locals. The Labour vote tally in locals and GE were almost identical but the Tory vote was substantially higher in the General than in the locals. Lib Dems, Green and UKIP voters seem to split their votes.

More recently in 2023 Reform UK’s presence in the local elections was poor so maybe staunch GE Reform UK voters lent their votes to the Tories.

With regard to methodology. I suppose it comes back down to what are polls. If they are a snapshot then they should just report the data including ‘don’t knows’. If they are a prediction they need to be clear what assumptions they are making.

[Adam in Stroud]

May 28, 2024 14:02:31 GMT Sibboleth said:

Two things. Firstly, even a fairly significant polling error would result in a terrible defeat for the government all the same: after all, the published polls are absolutely cataclysmic, more so than the media has really internalized. This is actually what happened in 1997. Secondly, I'm not sure if there's that much significance to 'don't knows' being more likely to have voted Conservative than not in 2019 as 45% of people who voted did...

The first thing is specifically acknowledged in the Guardian article

Hayward said it was “not shy Tories” all over again, and that Labour still undoubtedly had a large lead over the Conservatives, but he was concerned the polls were “getting it wrong”.

(text bolding by me, not Guardian)

On the Conservative nature of the Don't Knows, it's not just absolute numbers but relative proportions:

There’s three times more people who voted Conservative and are undecided than voted Labour and are undecided,

That's not necessarily a ray of light for the Conservatives - Labour 2019 voters have made up their minds and are probably voting Labour. Lots of 2019  Conservtives are at best unwilling to plump Tory yet. (Plus, plenty of 2019 Conservatives have made up their minds in favour of Labour, Reform, or occasionally LD - the polls are reasonably clear on that).

Another possibility - which is not comforting for the Conservatives - is that some of the Undecideds may be weighing up their options for a tactical vote to remove a Tory.