Introduction to our ‘EURef2 Poll of Polls’

Posted on 28 September 2018 by John Curtice

One of the most widely cited features of this site before the EU referendum was a ‘poll of polls’. This was simply an average of the level of support for Remain and Leave as recorded in the six most recently conducted polls of referendum vote intentions. Its purpose was to smooth out some of the ups and downs in support for one side or the other that can arise simply as a result of the chance variation to which all polls are subject. (Polls are, after all, attempting to estimate the distribution of attitudes across the population as a whole on the basis of the views expressed by maybe no more than a thousand people or so.) That chance variation means that the level of support for one side or the other recorded by an individual poll may well go up or down by two or three points, even though amongst the population as a whole the balance of opinion has not shifted at all. However, if we average the results of a number of polls the effects of this chance variation may well be reduced, and thus we might secure a more robust measure of the balance of opinion.

However, once the referendum in June 2016 was over there was no obvious use for a poll of polls on the site. No further referendum was in prospect. Few polls asked people how they would vote if presented once again with the choice between Remain and Leave. The focus instead was on the much more nuanced question of what kind of Brexit voters wanted and how well they thought the process was being handled, a focus that this site has shared during the last two years.

But now the position has changed somewhat. Polls of how people would vote in another referendum have become rather more frequent. Speculation about whether voters might be asked to revisit the issue has become more common, not least as a result of the high-profile campaigning in favour of a second referendum by the anti-Brexit People’s Vote campaign and Labour’s decision to leave open the possibility that it might support some kind of ballot , though the idea is strongly opposed by pro-Brexit organisations such as Change Britain. But even if a second referendum does not take place, it might be thought important to ask whether or not, as the Brexit process comes to a conclusion, there is still a majority in favour of leaving the EU. After all, the answer to that question might be thought central to any evaluation of the success or otherwise of the EU referendum as a way of deciding what Britain’s relationship with the EU should be.

As a result of these developments, we have now instigated a new poll of polls. This summarises the results of the most recent polls that have asked people how they would vote in another referendum in which the choice was between remaining in the EU and leaving. Some of these polls have asked people how they would vote in response to exactly the same question as appeared on the ballot paper in the 2016 EU referendum. Others have used a slightly different approach, while still asking respondents whether they would vote to remain or to leave. Our poll of polls is based on both types of question (including recent readings of YouGov’s long-running Eurotrack series). However in both cases we have excluded from our calculation those who say they do not know how they would vote or indeed declare that they would not do so. Thus, our total of Remain and Leave support will always add up to 100%.

As well as showing the current position in the poll of polls, we have also calculated what our poll of polls would have been saying if we had been publishing it on a regular basis since the beginning of this year (before which polls of EURef2 vote intentions were largely very infrequent). This gives us a picture of whether support for Brexit has been going up or down during recent months.

Of course, our poll of polls figures should not be taken uncritically. Even when we average across as many as six polls, our estimate will still be subject to some random fluctuation. So not too much should be made of a one-point shift up or down in our estimate, or at least not until that new estimate is confirmed after several more polls have been released.

At the same time, polls are not necessarily conducted perfectly. A poll of polls cannot insulate us against the risk of collective failure by the polling industry, as has happened at more than one recent general election. Meanwhile, less dramatically but no less importantly, the polling companies may systematically disagree with each other about the level of support enjoyed by Leave and Remain. In that event, changes in which companies’ polls contribute to the poll of polls may induce a degree of artificial instability. We thus give details of which companies conducted the polls that contribute to our latest poll of polls, while, where necessary, we will warn in our Commentary section if there is a risk that a change in whose polls are included may have induced a change in the estimates.

Finally, bear in mind that while any set of six polls may have all been conducted at much the same time, equally it is possible that quite a few weeks may have elapsed between when the first and the last were conducted. Even though polls of EURef2 vote intentions have become more frequent, at present it is still the case that it can take a month for half a dozen new readings to appear. You will thus see that we also always provide details of the dates between which the interviewing (fieldwork) for the polls included in the latest poll of polls took place.

John Curtice

By John Curtice

John Curtice is Senior Research Fellow at NatCen, Professor of Politics at Strathclyde University, and Chief Commentator on the What UK Thinks: EU website.

There are 21 comments

Your email address will not be published. Required fields are marked *