Something Old, Something New

An older article on The Independent website claimed that 2.6m Leave voters had switched to Remain. A newer article said voters backed a new referendum on EU membership, among other statements about public opinion.

This article examines these claims.

‘2.6 million Leave voters’

In total, it concluded that 2.6 million Leave voters have switched their support to Remain, while 970,000 have moved the other way — a net gain for the pro-EU side of 1.6 million.

Roughly, these figures suggest that 15% of 2016 Leave voters now back Remain, whilst 6% of Remain voters go the other way.

According to the YouGov website, the company conducted three waves of research on similar questions for the People’s Vote campaign prior to 4th September 2018. The Focaldata analysis appears to be based on the first two waves (“two YouGov polls that together surveyed more than 15,000 people”).

We can see immediately see the issue. (Tables: YouGov/People’s Vote survey)

The Independent article is seemingly comparing:

  • 2016 Remain voters who chose Leave in a hypothetical referendum;
  • 2016 Leave voters who selected any other option than Leave (including don’t know).

If non-voters and undecided respondents are removed, the Remain share was 53%.

Among those giving a preference, 8% of Leave voters switched to Remain, and 6% of Remain voters now back Leave. The switching rates are similar. Consequently, the net shift to Remain principally comes from the expressed preferences of those who did not vote in the 2016 EU referendum.

Myths and Epics

In reply to the claim that voters just ‘want to get Brexit done’, Kellner writes:

Recent Opinium surveys asked voters whether they agreed with the statement, “I don’t care how or on what terms Britain leaves the EU as long as we leave as soon as possible.” Just one in three said they did.

This is not an agree-disagree question, but a question which asks which statement is closest to the respondent’s view — with four response options:

About one in three click that option. (Image: What UK Thinks EU/Opinium)

Next, Kellner argues against the claim that ‘voters have not changed their mind’:

Figures in individual polls vary slightly, but an average of recent surveys shows a steady 53–47 per cent preference for Remain. This represents a five-point swing to Remain since 2016.

There has been a glacial shift towards Remain in recent hypothetical referendum questions. Similar rates of Remain and Leave voters have switched to the other side, with some undecided.

These surveys were conducted prior to 10th June 2019. (Source: What UK Thinks EU)

The principal role of 2016 non-voters in recent Remain leads is omitted in the Independent article.

The fourth claim that Kellner responds to is ‘voters don’t want another referendum’. Only two questions about hypothetical scenarios are provided:

The latest YouGov survey finds that, by almost two-to-one, voters say that if Britain is to leave the EU without a deal, the decision should be taken by a referendum than by parliament. If a deal is agreed with Brussels, the margin widens to more than two-to-one.

However, it has been well-noted — by Prof Sir John Curtice (back in August 2018 too) and others — that responses to questions about holding another referendum are sensitive to their wording and conveyed meaning.

There is a wide range of responses to questions about another membership referendum. (Souce: UK in a Changing Europe)

Due to changing political circumstances, the questions used by polling companies have often changed too.

Of the ten questions asked between mid-November 2018 and early January 2019, eight estimate 66% or more 2016 Remain voters back another referendum. Similarly, eight questions estimate more than half of 2016 Leave voters oppose such a vote.

The pattern of responses is indicative. Asking about a ‘People’s Vote’ or a ‘public vote’ gets somewhat more supportive responses from 2016 Leave voters than other variants. Those questions did not specify whether an option to remain in the European Union was on the ballot. Consequently, the implied purpose of such a referendum would be to accept or reject the government’s negotiated agreement with the European Union. Once it is made clear that reversing the 2016 result is possible — as four of those questions specify — that level of support withers.

As Prof Sir John Curtice concludes:

Any claims that the polls show a widespread clamour for a second referendum have to be treated with care.

This blog looks at the use of statistics in Britain and beyond. It is written by RSS Statistical Ambassador and Chartered Statistician @anthonybmasters.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store