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.


Prorogation polling

Three polling companies – YouGov, Ipsos MORI and Survation – have so far released polling on the government’s decision to prorogue Parliament in mid-September.

YouGov polled on the issue twice – a snap poll on the day of the announcement itself, with the same question repeated overnight. The on-the-day figures were 27% acceptable, 47% unacceptable, 26% don’t know. The follow-up poll had a similar split, but with the number of don’t knows dropping off as people became aware of the story – 31% said it was acceptable, 53% unacceptable, 16% don’t know. Tabs are here)

Ipsos MORI did an unusual online poll (almost alone among pollsters these days, most of their polling is done by phone). They found 30% thought the decision to prorogue Parliament was right, 46% thought it was wrong. Tables are here.

Finally there was a Survation poll for today’s Daily Mail. This found a closer result, with the public fairly evenly split – 39% were supportive, 40% opposed (note this is rounding the totals for support/oppose after they’ve been summed, hence the apparent discrepancy with the tables). Tables are here.

Overall it looks as if the public are opposed to the prorogation decision – though it is unclear to what degree. Whether that really matters or will make any dent in the government’s support is a different matter. Opposition to prorogation is concentrated among Remainers (in YouGov 82% of Remainers think the move is unacceptable, but only 24% of Leavers, in MORI’s poll 74% of Remainers think it was wrong, only 20% of Leavers, in Survation 74% Remainers, 14% leavers). If most of the opposition to the move comes from people who are opposed to the government’s policy anyway (and I expect the more fervent opposition comes from those who were most fervently opposed already) the government are hardly likely to worry too much over losing the crucial “people who hated us anyway” vote.

Both YouGov and Survation included voting intention in their surveys:

YouGov’s topline figures were CON 33%(-1), LAB 22%(nc), LDEM 21%(+4), BREX 12%(-1), GRN 7%(-1)
Survation’s topline figures were CON 31%(+3), LAB 24%(nc), LD 21%(nc), BREX 14%(-2), GRN 3%(nc)

Changes in the YouGov poll are from a poll earlier this week, before the announcement. In Survation changes are from a poll three weeks ago. There is a little movement up and down, but certainly nothing that suggests the announcement has done immediate damage to Conservative support.


-->

A very quick update on voting intention polls over the last few weeks. As usual August is a relatively quiet period – opinion pollsters have holidays too. The fact that we have a new Prime Minister hasn’t made much change to that. In August so far we’ve had five voting intention polls:

BMG/Independent (Dates TBC) – CON 31%, LAB 25%, LDEM 19%, BREX 12%, GRN ?
ComRes/Telegraph (11th Aug) – CON 31%, LAB 27%, LDEM 16%, BREX 16%, GRN 4% (tabs)
Survation (11th Aug) – CON 28%, LAB 24%, LDEM 21%, BREX 15%, GRN 3% (tabs)
Opinium/Observer (9th Aug) – CON 31%, LAB 28%, LDEM 13%, BREX 16%, GRN 5% (tabs)
YouGov/Times (6th Aug) – CON 31%, LAB 22%, LDEM 21%, BREX 14%, GRN 7% (tabs)

Note that the BMG tables aren’t up yet, hence I don’t know the level of support for the Greens or their fieldwork dates. These polls continue to show the boost in Conservative party support following Boris Johnson’s accession filtering though. It is the first “Post-Johnson” poll for BMG and Survation, and they show the Conservatives up by 3 and 5 points respectively. We’re now at a point where the most recent polls from all the regular polling companies show the Conservatives back ahead, though the size of their lead differs given the variation in figures between pollsters.

Normally I would be speculating about how long the government’s honeymoon boost would last. It’s not really the case here given how many political events are going to be crammed into the next few months. Events will likely preempt its natural unwinding: whatever diplomatic negotiations or stand offs occur between the government and the EU (starting with the G7 meeting this week), whatever Parliamentary moves there may be against the government or against No Deal, the party conferences, whatever preparations or announcements there may be on No Deal and, of course, the actual outcome at the end of October. The current levels of party support seem rather irrelevant in the face of that – the Conservatives are probably happy to have a lead at the moment, but there are ten weeks ahead of us that are packed with events that can throw everything up in the air.


So far we have had six opinion polls since Boris Johnson became Prime Minister, from Ipsos MORI, Deltapoll, Opinium, ComRes and two from YouGov (one for the Sunday Times, one for the Times). Voting intentions from them all are below.

YouGov (30 Jul) – CON 32%, LAB 22%, LDEM 19%, BRX 13%, GRN 8% (tabs)
Ipsos MORI (30 Jul) – CON 34%, LAB 24%, LDEM 20%, BRX 9%, GRN 6% (tabs)
Deltapoll (27 Jul) – CON 30%, LAB 25%, LDEM 18%, BRX 14%, GRN 4% (tabs)
YouGov (26 Jul) – CON 31%, LAB 21%, LDEM 20%, BRX 13%, GRN 8% (tabs)
Opinium (26 Jul) – CON 30%, LAB 28%, LDEM 16%, BRX 15%, GRN 5% (tabs)
ComRes (25 Jul) – CON 28%, LAB 27%, LDEM 19%, BRX 16%, GRN 4% (tabs)

The trends across all these polls are very consistent – compared to pre-Johnson polling everyone shows the Conservatives gaining support (up 10 points in Deltapoll, 8 in MORI, 7 with Opinium, 6 or 7 in YouGov, and 3 with ComRes). In each case support for the Brexit party has dropped by a similar amount, while support for the other political parties remains broadly consistent. While in practice things will be a little more complicated (people will have moved in and out of don’t know, likelihood to vote will have gone up and down and so on), you can fairly characterise it as Johnson’s leadership immediately winning back a chunk of support from the Brexit party.

While The Conservatives will no doubt take some cheer from being ahead again in the polls, they should perhaps not take too much. The polls show them back at around 30% – where they were in March – as opposed to figures in the high 30s or low 40s that they recording at the tail end of last year. Boris Johnson has not magicaclly repaired all the damage they have suffered in the last few months – primarily it would seem because they are still losing a significant chunk of their 2017 support to the Brexit party. The fact they are ahead again is as much because of the splitting of the anti-Brexit vote between Labour and the Liberal Democrats. In the early months of this year Liberal Democrat support was around ten percent and Labour were mostly in the thirties; now the Liberal Democrats are typically in the high teens and Labour normally in the twenties.

Secondly, it is very much the norm for a new Prime Minister to receive a boost in the polls. They normally come to power with a flurry of announcements and activity (and that often contrasts with the drift of whatever moribund government they’ve just replaced), their natural supporters once again project all their hopes and dreams upon them, and a fair chunk of the media are normally treating them as the messiah. It happened with John Major, Gordon Brown, Theresa May and now Boris Johnson. Generally speaking those factors don’t last, and neither does the boost – though the temptation is always to think this time is different. Gordon Brown narrowly avoided calling an election during his bump, aborting just before his lead collapsed; Theresa May’s boost in the polls stretched on far, far longer than expected, finally tempting her into an election before rapidly deflating. One probably shouldn’t get too excited about this one either – more important in terms of public support will be what happens in terms of Brexit in September, October and November.

(A couple of quick notes on methodology. You’ll note the usual big gap difference between pollsters in terms of Labour support – with YouGov and Ipsos MORI showing lower Labour support than Opinium and ComRes. My best guess, which I’ve written about elsewhere in more detail, is that this is to do with how and if pollsters weight for past vote. Secondly, I should flag up a methodology change from MORI – previously they hadn’t been including the Brexit party in their question wording for the voting intention question, resulting in lower support. This month MORI included the Brexit party in the prompt for half the sample – presumably in order to see how much difference it made.)


ComRes have a new poll out in the Telegraph. Topline voting intention figures with changes from last month are CON 25%(+2), LAB 28%(+1), LDEM 16%(-1), BREX 19%(-3).

The Telegraph headline their report on hypothetical questions asking how people would vote if Boris Johnson was Tory leader, I’m rather sceptical of the worth of such questions when it’s a hypothetical that actually appears to almost certainly happen in a week or two’s time, but there goes. For what it’s worth, in the hypothetical Boris question the voting intentions are CON 32%(-5), LAB 25%(+3), LDEM 17%(-3), BREX 14%(nc) – a substantial drop in Conservative support from the same hypothetical question a month ago, suggesting perhaps it wasn’t such an effective prediction of Boris’s future impact.

Anyway, my general assumption is that parties normally do get something of a boost from new leaders, if only from the news coverage, enthusiasm of their supporters and whatever the new leader has planned to make an early impact. We shall see for real in the coming weeks.

As ever, other polls are also available – there have been two other voting intention polls this month:

Opinium in the Observer at the weekend had topline figures of CON 23%, LAB 25%, LDEM 15%, BREX 22%, GRN 8% (tabs here)
YouGov in the Times last week had topline figures of CON 24%, LAB 18%, LDEM 20%, BREX 23%, GRN 9% (tabs here)

There remains a significant difference between polling companies, most notably on the level of support recorded for the Labour party. The reason for this is unclear – polling companies these days are not taking radically different approaches towards turnout modelling or reallocating don’t knows, nor in how the questions are asked (though whether the Brexit party or Greens are prompted may be making a difference in some cases). By default that means the differences are more likely to be down to sampling make up – whether by the way respondents are sampled or weighted, companies are interviewing slightly different people. Specially, some companies seem to get Labour voters who are more loyal than others. I suspect some of this may be down to weighting variables (the measures polling companies choose to use, such as whether they control on education or political interest), perhaps some down to when past vote weighting data is collected – whether it is collected in the survey itself, or was collected at the time of the election (or in the case of MORI, whether past vote weighting is avoided entirely).

My advice, as ever, is to avoid the temptation of assuming that the polls that you’d like to be accurate are the ones that are, and that polls with results that you dislike are wrong ones that can be ignored.

However, it is probably worth paying some attention to polling for the European election results in May. At those elections we saw a very similar difference across polling companies, with extremely large spreads in terms of Labour support (final polls varied from 13% to 25%). It did tend to be the same companies showing high and low Labour support, the most obvious explanations did appear to be down to sampling, and when comparing to final results those companies showing low levels of Labour support were substantially more accurate. I am cautious about how much weight to put on these – after all, along with Ipsos MORI who were most accurate, my own company did conspicuously well here, and I wouldn’t want to fall into wishful thinking myself. There are obviously different challenges in polling low and high turnout elections (and other companies have other questions to ask about, for example, Brexit party support), but I would have thought that, in the absence of changes or explanations, it would sensible to be somewhat cautious of polls at the top end of Labour support if those same polls have very recently overstated Labour support in a national election.