Is the EU referendum outcome "on a knife edge"?

"EU referendum outcome on a knife edge, new poll reveals"

The Independent, 15 December 2015

"UK exit from European Union on a knife edge, as poll shows British public are now 50/50 over leaving"

The Telegraph, 14 December 2015

You may have seen these or similar headlines this week claiming that the EU referendum outcome is on a "knife edge".

While two polls have shown a narrow gap between people wanting to 'remain' and people wanting to 'leave', others have shown significant gaps between the two.

Polling experts have pointed to differences between online and telephone surveys which may be causing the contradictory results. One, Professor John Curtice, has commented that:

"Alas, we cannot be sure which approach is right and which is wrong (if either); the divergence simply means that we are left with very considerable uncertainty about just how close the referendum race really is."

ICM poll found 42% opting to 'remain' and 41% opting to 'leave'

The poll those "knife edge" headlines were reporting on found 42% in favour of staying in the EU, and 41% in favour of leaving.

According to YouGov's Anthony Wells, it's the closest gap pollster ICM has found of late. He said the company's polls typically show 'remain' as having a lead of around six points.

It's not the only poll this week to show such a close gap. Another pollster, Survation, found a similar gap, although it was the other way around. It found 40% support for 'remain' and 42% for 'leave'.

Since then, two more polls have shown a rather larger gap. ComRes and Ipsos MORI found strong leads for 'remain' (21 and 26 points respectively).

The methods of the poll can affect the results

It matters whether the poll was conducted by telephone or online.

The telephone polls found a large lead for 'remain', while the online polls showed a more even picture.

This is consistent with the findings of an experiment by ComRes.

It asked the same question—the one that will be asked in the referendum—on an online and telephone survey. Both were weighted in the same way, and were run within days of each other. In the online poll it was neck and neck, with 41% opting to remain and the same proportion opting to leave. The results of the telephone poll were very different: 56% opted to remain and 35% opted to leave.

ComRes says this may be down to differences in the types of voters in the referendum. For general election polling, the vast majority of the electorate vote the same way every time, according to ComRes. A referendum is different, because it said there are many more "low information" voters—those who haven't made their mind up yet—meaning that any polling will be particularly sensitive to differences in the political engagement of the people being surveyed.

People choosing to sign up to an online panel are 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", ComRes says.

Being in a panel and being regularly asked about their beliefs can make people more engaged and aware of the issues.

It concludes:

"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."

Some polls ask different questions

As Independent columnist John Rentoul has said today, part of it is also down to question wording. He points to analysis by Ipsos MORI finding that when people were asked if they support "giving 16 and 17 year olds the right to vote" in the referendum, more were in favour, but when asked if they supported "reducing the voting age from 18 to 16" for the referendum, more were against it.

Most of these polls asked the question that will be in the referendum: "Should the United Kingdom remain a member of the European Union or leave the European Union?", with the options "Remain a member of the European Union", "Leave the European Union". Pollsters also offer the option "don't know".

Ipsos MORI tried out two different questions, one asking the referendum question and the other asking whether people wanted to "stay in" or "get out". Those responding to the stay in/get out question were less likely to pick to stay than those asking the remain/leave question.

In a separate poll published today, Lord Ashcroft asked people to place themselves on a scale between zero and 100, zero meaning they would definitely vote for the UK to remain in the EU, 100 meaning they would definitely vote to leave.  38% placed themselves between zero and 49 and 47% placed themselves between 51 and 100, meaning more wanted to leave.

Some evidence that the gap has tightened since summer

Professor Curtice says that on the basis of the Ipsos MORI poll, Survation poll and another one by YouGov, there is some evidence that there has been a smaller gap during the autumn than during the summer.

But as with all polls, that's still uncertain.