The release of two EU referendum polls on Monday night appeared to put the question of Brexit on a knife edge. Some media outlets duly followed suit with headlines including “Britain is on the cusp of leaving the EU right now” and “Voters head for the Brexit”. For some in the market research world, the verdict on the state of the polls has a worrying air of familiarity to it.
During the General Election campaign earlier this year, ComRes put out its own briefing note challenging the prevailing “neck and neck” narrative being given to the election race, and suggesting that the Conservative Party was in fact definitively ahead. We pointed out that while online polling, which is conducted more frequently, was showing a narrow race, the less frequent telephone polling had begun to show a clear trend of the Conservatives being ahead. Our own telephone polling had not shown a Labour lead all year all and during the election campaign had started to show regular 3-4 point Conservative leads. Although of course, the telephone polls from the final week still didn’t predict the scale of Conservative victory.
We are now seeing a similar trend appear in the EU referendum polling. Much of the polling on the EU referendum since May has been done online, and much of it points to a race which is neck and neck. But telephone polling again suggests a different story: this time not showing a modest lead of a few points, but an unequivocal lead – for staying in the EU.
Our telephone poll released this week suggests that 56% of Britons would vote to remain in the EU if the vote took place tomorrow. This is a full 21 points ahead of the 35% who say that they would vote to leave (8% say they don’t know which way they would vote). Now, we would always caution against overemphasising any individual poll, but these figures are remarkably consistent with the last time we ran the question in September (when they were 55% vs 36%) and also in May with the new Conservative Government recently formed (51% vs 33%). Our poll in the final week of the General Election (which had a three point Conservative lead in the headline figures), also showed a 22 point lead for remaining in the EU (56% vs 34%).
ComRes are not alone in this. Ipsos MORI, the only other pollster currently polling the EU referendum by telephone, has also shown consistent double digit leads for Remain. Indeed, their poll released today shows a 17 point lead: 53% to 36%. The largest lead in an online poll has been eight points, with the vast majority showing the race far closer.
We should at this point declare that as a research consultancy we are method neutral – meaning that we use both online and telephone methodologies as well as face to face where appropriate. We match the methodology to the project on a case-by-case basis depending on which we feel is most appropriate for the client’s business and research objectives. All survey methodologies have their different strengths and weaknesses, making each more or less appropriate for certain types of challenges.
During the General Election campaign itself we used only telephone polling for all our voting intention polls. Now in peace-time (outside of official campaign time) we are back to using both methods for our voting intention polls. Indeed we anticipate using online polling for the London mayoral election, as the youthful, highly transient population mostly living in rented housing make it less suited to survey by telephone.
At a national level (where around half of voters are over the age of 55) telephone polling has the better record though: a telephone poll was most accurate at the 2010 General Election, at the 2011 AV referendum, at the Scottish independence referendum and was “least inaccurate” at this year’s General Election.
Furthermore, the British Election Study, which is the one face to face survey on the EU referendum that’s taken place in the same time period, showed a large 17 point lead for remaining in the EU. On this occasion then, it looks as if two methodologies (phone and face-to-face) are showing the advantage currently with the “Remain” side while one methodology is not.
Therefore, in order to understand this apparent difference between the methodologies we undertook our own experiment, asking the referendum question on our online and telephone surveys. The results were telling – both sets of data used the same question, were run within days of each other, had the same demographic weights applied and were past vote weighted in the same way – but despite this, the online polling showed the public split on the issue (as it does in polls conducted by other pollsters online), whereas the telephone polling showed a large lead for “Remain”.
ComRes experiment: EU referendum poll conducted online and by telephone
Q: If a referendum were held tomorrow on the UK's membership of the European Union (EU), how would you vote on the following question? "Should the UK remain a member of the European Union, or leave the European Union?" Base: Telephone (n=1,001) 11-13/12/2015. Online (n=2,049) 9-11/12/2015.
That almost every single online poll has also shown the race neck and neck, suggests this is not just about house effects but an issue of online samples more generally when it comes to this particular subject.
When we conduct Westminster voting intention using both telephone and online methodologies, the results are relatively consistent, as is true for much of the commercial work we do.
The extent of the difference here seems to be instead related to the issue of the European referendum itself. Of course there is no way of telling at this stage which is more accurate, but some of the difference could be explained by the differing levels of engagement between online and telephone respondents.
At a General Election, the vast majority of the electorate vote the same way each time, based on long-term cultural exposure to the parties, being asked to choose between them every four or five years and political leaders gracing our televisions most evenings. However, at a referendum, there is a much more “low information” electorate, making any polling about the issue particularly sensitive to differences in the political engagement of the sample being surveyed.
People choosing to sign up to an online panel are by nature more online savvy. They are more likely to be engaged on social media and exposed to strongly held beliefs that we see in online encounters. Indeed, by being on an online panel, regularly receiving surveys probing your thoughts about particular details of current affairs you are becoming even more engaged and aware of the issues. This engagement gap makes little difference if you are responding occasionally about your opinions of satellite TV or your favourite chocolate bar, but on controversial subjects like the EU, where a minority hold their opinions strongly and the rest are less committed, it can make a crucial difference.
Whatever the reason for the methodological differences, the referendum race is not currently on the knife edge portrayed by some. This is not to say that Britain will definitively vote to remain a part of the European Union – past experience of referendums in the UK suggests the final voting blocs tend to fall into place only in the final few weeks before the vote, and on this occasion we don’t even have a date for polling day. Our polling for OpenEurope released today suggests that if David Cameron fails to secure his renegotiation, opinion shifts fairly dramatically towards a much closer race. But for the time being, we’ll be advising all our clients that if the referendum were held tomorrow – far from being a photo finish, we would be seeing a relatively comfortable victory for the Remain side.