Earlier this week NatCen released new polling on what people want from Brexit. The vast majority (90%) of people would like to keep free trade with the European Union. By 70% to 22% people would also like to limit the amount of EU immigration into Britain. Getting these two things together does not, of course, seem particularly likely. Asked if Britain should agree to keep free movement in exchange for keeping free trade, people are much more evenly split – 49% think we should, 51% think we should not (the full report is here).

Personally, I still think the best way of judging public opinion on Brexit is probably not to ask about individual policies, but to test some plausible scenarios – when it comes to it, people will judge the deal as a whole, not as the sum of its parts. YouGov released some updated polling on Brexit today that repeated that experiment, and again found that a Canadian type deal is likely to get the widest support from the public (that is, no freedom of movement and a more limited trade deal). The problem with a Norway type deal – retaining full free-trade with the EU in exchange for keeping freedom of movement and a financial contribution is that most of the public would see it as not respecting the result of the referendum.

I’ve written a much longer piece about the YouGov polling over on the YouGov site here, so I won’t repeat it all. One interesting bit though is looking at the possible outcomes of an early election, fought on the issue of Brexit. Now, I should start with some important caveats – hypothetical election questions are very crude tools. While I’m sure an early election would be dominated by the issue of Brexit, there would be other issues at play too, so a question like this will over emphasise the impact of Brexit policy. Nevertheless, it suggests some interesting patterns. YouGov asked how people would vote if Brexit could not pass a Parliamentary vote and instead an early election happened. In the scenarios the Conservatives and UKIP back Brexit (as they undoubtedly would) and the Lib Dems back a second referendum (as they’ve said they would). YouGov offered three different scenarios for Labour – one, where Labour back Brexit, two where Labour back only a “soft Brexit”, three where Labour also offer a second referendum. In all three cases the Conservatives would win easily – even the closest scenario gives them a twelve point lead. The interesting finding is the Lib Dems – in the two scenarios where they are the only party offering a second referendum their support goes up to 19% or 22% (if Labour also offer a referendum the Lib Dems don’t gain nearly so much). So, while these are hypothetical questions that need to be taken with a pinch of salt, it does suggest that appealing to those voters who really are set against Brexit could be a route back for the Lib Dems, especially if they are the lone “anti-Brexit” party. The full results for the YouGov polling are here.

Meanwhile Ipsos MORI released their monthly political monitor. In terms of voting intention the Conservative lead is halved from last month, but that is likely something of a reversion to the mean after a towering eighteen point lead last month. Topline figures are CON 42%, LAB 33%, LDEM 10%, UKIP 7%, GRN 3%. As ever, wait until you see the change echoed in other polls before concluding that the Conservative lead is waning.

Theresa May still enjoys a positive approval rating – 54% are satisfied with the job she is doing, 30% disatisfied. The new government also have a net positive rating at their handling of the economy so far – 51% think they’ve done a good job, 30% a bad job. Where the public are not convinced is on how the government are handling the biggest issue – only 37% think the government are doing a good job at handling Brexit, 48% think they are doing a bad job. Full details of the MORI poll are here.


Donald Trump has won, so we have another round of stories about polling shortcomings, though thankfully it’s someone else’s country this time round (this is very much a personal take from across an ocean – the Yougov American and British teams are quite separate, so I have no insider angle on the YouGov American polls to offer).

A couple of weeks ago I wrote about whether there was potential for the US polls to suffer the same sort of polling mishap as Britain had experienced in 2015. It now looks as if they have. The US polling industry actually has a very good record of accuracy – they obviously have a lot more contests to poll, a lot more information to hand (and probably a lot more money!), but nevertheless – if you put aside the 2000 exit poll, you have to go back to 1948 to find a complete polling catastrophe in the US. That expectation of accuracy means they’ll probably face a lot of flak in the days ahead.

We in Britain have, shall I say, more recent experience of the art of being wrong, so here’s what insight I can offer. First the Brexit comparison. I fear this will be almost universal over the next few weeks, but when it comes to polling it is questionable:

  • In the case of Brexit, the polling picture was mixed. Put crudely, telephone polls showed a clear lead for Remain, online polls showed a tight race, with leave often ahead. Our media expected Remain to win and wrongly focused only on those polls that agreed with them, leading to a false narrative of a clear Remain lead, rather than a close run thing. Some polls were wrong, but the perception that they were all off is wrong – it was a failure of interpretation.
  • In the case of the USA, the polling picture was not really mixed. With the exception of the outlying USC Dornslife/LA Times poll all the polls tended to show a picture of Clinton leading, backed up by state polls also showing Clinton leads consistent with the national polls. People were quite right to interpret the polls as showing Clinton heading towards victory… it was the polls themselves that were wrong.

How wrong were they? As I write, it looks as if Hillary Clinton will actually get the most votes, but lose in the Electoral College. In that sense, the national polls were not wrong when they showed Clinton ahead, she really was. It’s one of the most fustrating situations to be in as a pollster, those times when statistically you are correct… but your figures have told the wrong narrative, so everyone thinks you are wrong. That doesn’t get the American pollsters off the hook though: the final polls were clustered around a 4 point lead for Clinton, when in reality it looks about 1 point. More importantly, the state polls were often way out, polls had Ohio as a tight race when Trump stomped it by 8 points. All the polls in Wisconsin had Clinton clearly ahead; Trump won. Polls in Minnesota were showing Clinton leads of 5-10 points, it ended up on a knife edge. Clearly something went deeply wrong here.

Putting aside exactly how comparable the Brexit polls and the Trump polls are, there are some potential lessons in terms of polling methodology. I am no expert in US polling, so I’ll leave it to others more knowledgable than I to dig through the entrails of the election polls. However, based on my experiences of recent mishaps in British polling, there are a couple of places I would certainly start looking.

One is turnout modelling – US pollsters often approach turnout in a very different way how British pollsters traditionally did it. We’ve always relied on weighting to the profile of the whole population and asking people if they are likely to vote. US pollsters have access to far more information on which people actually do vote, allowing they to weight their samples to the profile of actual voters in a state. This has helped the normally good record of US pollsters… but carries a potential risk if the type of people who vote changes, if there is an unexpected increase in turnout among demographics who don’t usually vote. This was one of the ways British pollsters did get burnt over Brexit. After getting the 2015 election wrong lots of British companies experimented with a more US-style approach, modelling turnout on the basis of people’s demographics. Those companies then faced problems when there was unexpectedly high turnout from more working-class, less well-educated voters at the referendum. Luckily for US pollsters, the relatively easy availability of data on who voted means they should be able to rule this in or out quite easily.

The second is sampling. The inquiry into our general election polling error in 2015 found that unrepresentative samples were the core of the problem, and I can well imagine that this is a problem that risks affecting pollsters anywhere. Across the world landline penetration is falling, response rates are falling and it seems likely that the dwindling number of people still willing to take part in polls are ever more unrepresentative. In this country our samples seemed to be skewed towards people who were too educated, who paid too much attention to politics, followed the news agenda and the political media too closely. We under-represented those with little interest in politics, and several UK pollsters have since started sampling and weighting by that to try and address the issue. Were the US pollsters to suffer a similar problem one can easily imagine how it could result in polls under-representing Donald Trump’s support. If that does end up being the case, the question will be what US pollsters do to address the issue.


-->

No Bregrets

Almost as soon as the referendum votes were counted people were asking for polling on whether people regretted their decision. There is still a certain audience who seem downright desperate to find polling showing that people do not, after all, want to leave the European Union (and, I suppose, a (slightly larger) audience who want to see polls showing they don’t!). I guess this is the curse of a referendum decision that takes a couple of years to actually implement.

The most straightforward way of measuring Bregret is to ask the referendum question again – how would people vote if the referendum question was asked again now. Several polls have done that:

YouGov/Eurotrack (20th-25th Oct) Remain 44%, Leave 43%
BMG (19th-24th Oct) Remain 45%, Leave 43%
YouGov/Eurotrack (21st-22nd July) Remain 43%, Leave 44%
YouGov/Eurotrack (3rd-4th July) Remain 45%, Leave 45%

All of these suggest a very small movement towards Remain, and given Leave’s lead was only four points that’s enough to flip the result in a couple of cases. However, I’d be a little cautious in reading too much into the results. All of these polls are just straight “how would you vote questions” with no attempt to account for differential turnout, when at the referendum Leave voters were more liable to turnout. If you look at the actual tables for these you’ll find there is very little movement between remain and leave, the shift is down to people who didn’t vote in the referendum claiming that in a referendum tomorrow they would vote in favour of Remain. That’s possible of course (perhaps people who assumed a Remain victory in June and didn’t bother to vote, now realising their vote really would count)… but I’m rather sceptical about people saying they’d vote in an EU referendum who didn’t bother to vote in the one we just had.

The approach alternative is to ask if people think it was the right decision and if they might change their vote.

Just after the referendum there was an poll by Ipsos MORI for Newsnight, which showed 43% of people thought Brexit was the right decision, 44% thought it was the wrong decision. Asked if they would change their mind in a new referendum, 1% of Remain voters said they would definitely or probably change their mind, 5% of Leave voters said they would definitely or probably change their mind. If those people all switched to the other side it would have just edged into a Remain lead.

In contrast YouGov have regularly asked if people think the decision to leave was right or wrong, and have tended to find slightly more people saying it was the right decision. The pattern of opinion is pretty consistent – movement between Remain and Leave is small and tends to cancel out, people who didn’t vote at all tend to split in favour of it being the wrong decision:

YouGov/Times (11th-12th Oct) – Right to Leave 45%, Wrong to Leave 44%
YouGov/Times (13th-14th Sep) – Right to Leave 46%, Wrong to Leave 43%
YouGov/Times (30th-31st Aug) – Right to Leave 47%, Wrong to Leave 44%
YouGov/Times (22nd-23rd Aug) – Right to Leave 45%, Wrong to Leave 43%
YouGov/Times (8th-9th Aug) – Right to Leave 45%, Wrong to Leave 44%
YouGov/Times (1st-2nd Aug) – Right to Leave 46%, Wrong to Leave 42%

Finally the British Election Study asked a question on whether people regretted how they voted or not. Only 1% of people who voted Remain said they regretted their vote, but 6% of people who voted Leave said they regretted their vote. Now, saying you’ve some regrets doesn’t necessarily mean that you wouldn’t, on balance, end up doing the same. For what it’s worth though, if those people who had regrets hadn’t voted the result would still have been leave; if those people had voted the opposite way it would’ve been Remain.

Looking across the board at all this polling, there is a suggestion that public opinion may have moved very slightly towards Remain, and with only a four point lead that’s enough to change the lead in some polls. However, in most cases that apparent movement isn’t people changing their minds, but is down to the opinions of those people who didn’t actually vote last time. That means if there was another referendum right now, if turnout was similar to June the result would probably be similar too.

My expectation is that, given time, we probably will see “Bregret”, simply because Brexit is going to be tested against reality while Remain isn’t. The road ahead has a lot of obstacles and some Leavers’ hopes and expections will be dashed (Remainers’ hopes and expectations of what would have happened if we’d stayed won’t, of course, face the same collision with reality). The lead at the referendum was only 4%, so it really won’t take that many people having second thoughts to flip opinion over. To those who really want to see evidence of Bregret in the polls – have a bit of patience. It will probably come in time, but the data really isn’t there to support it now.


A quick update on two polls released today. The regular ICM poll for the Guardian has topline voting intentions of CON 43%(nc), LAB 27%(+1), LDEM 8%(nc), UKIP 12%(+1), GRN 5%(-1). Changes are since mid October. Fieldwork was conducted over the weekend, and the full tabs are here.

BMG also released a new poll, though this is actually less recent than the ICM one (fieldwork was done between the 19th and 24th of October, so just over a week ago). Topline figures with changes from September are CON 42%(+3), LAB 28%(nc), LDEM 8%(nc), UKIP 12%(-1), GRN 4%(-1). Full details are here.

Both polls show the Conservatives still holding a large, robust lead. Note also that UKIP support is pretty steady in both – the drop in UKIP support that we saw in MORI’s poll does not appear to have been echoed in anyone else’s data.


The Evening Standard have published a new BMG poll of the Richmond Park by-election, suggesting a significantly less exciting race than some people thought (and than the Lib Dems hoped). Topline voting intention figures are:

GOLDSMITH (Ind) 56% (down 2 from the Con share in 2015)
OLNEY (Lib Dem) 29% (up 10 from the LD share in 2015)
LABOUR 10% (-2)
OTHER 5% (-5)

While there is a month to go, this suggests that Goldsmith should hold the seat relatively easily. The idea that, with both main candidates opposing Heathrow expansion, it could become an by-election about Brexit in a pro-EU seat doesn’t really seem to working out at present. 25% of voters say that Brexit will be the most important issue in deciding their vote, but they are mostly voting Lib Dem and Labour already. Goldsmith’s voters say their most important considerations are Goldsmith’s own record and views, followed by Heathrow opposition.

BMG also asked people how they would have voted if the Conservatives had put up an official Conservative candidate against Goldsmith. Topline figures would have been GOLDSMITH 34%, LIB DEM 25%, CONSERVATIVE 20% – so the race would have been far more competitive, but with the Tories trailing in third place. It was an unusual decision not to stand, but the polling suggests it was the right one for the Tories (or at least, neither option would have produced a Tory MP, but the Conservatives presumably prefer Goldsmith winning to a Lib Dem). Full details are here.