This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Find out more here

OPINION18 December 2018

Five Brexit lessons for researchers

Brexit Opinion Public Sector Trends UK

Understanding voters’ views of Brexit has proved a challenge – even for a seasoned political pollster. Deborah Mattinson shares some insights from the past two years.

1. Sometimes traditional methods hit the spot

In early 2017, as the UK hurtled towards triggering Article 50 in preparation for leaving the EU, BritainThinks began the Brexit Diaries project that has been tracking public opinion regularly ever since. Our start point was to recruit 48 Remain voters and 52 Leave voters (see what we did there!) in 10 locations around the UK.

We chose a traditional – even old-fashioned – diary method as we were keen to understand the underlying attitudes of the British public as the Brexit process unfolded. Our thesis was that by inviting participants to send us their thoughts, without framing these by posing specific questions, we would better understand voters’ preoccupations and priorities. And so it proved, with early findings yielding fascinating insights, such as the fact that, while ‘remain’ voters were very worried about the economy and constantly on the look-out for signals that a down-turn was imminent, ‘leave’ voters never mentioned it at all.

2. Our country is divided, complex and hard to understand

We followed this diary exercise with a nationally representative survey of the British public. Our initial research indicated that there were not two, but four Brexit tribes revealing the true extent of our divided nation:


BritainThinks produced a study a few months later that provided more detail on the division that the referendum had showcased. Six out of 10 of us described ourselves as ‘have-nots’, rather than ‘haves’, and increasingly we found people in focus groups in small towns and suburbs around the country talked about their own home towns as ‘left behind communities’.

3. The world can change fast but public opinion is remarkably intransigent

It is perhaps surprising how little these groups have changed in size over time. Basically, we are polarised in our division, with the two ‘extreme’ groups each approximately a third of the nation and the two ‘swing voter’ groups accounting for the remaining third between us. The relative size of these groups has changed little despite campaigners’ attempts to showcase the impact of their efforts, although our latest episode – a nationally representative survey of 2,092 British adults and a qualitative diary project with 26 of our original ‘Brexit Diarists’ – does suggest some change after months of dominating the news.

We learned that the size of the ‘devastated pessimists’ group (the most pro-remain of our EU segments) has increased for the first time for a year; at the same time, the size of the ‘die-hard’ segment (the most pro-leave segment) has decreased.

4. The public engages little with policy detail unless it can easily make a personal link

This consistency in the public views derives at least in part from lack of engagement with events. We often found that people’s reference point for Brexit remained the referendum campaign, even 18 months on. Small wonder then, that little changed. Across all our Brexit segments we found bewilderment about talk of Norway, Canada and Irish ‘backstops’.

People struggled to see the relevance of Brexit to their own lives, and still find it impossible to imagine the future: uncertainty about what Brexit will bring is widespread, with 56% of the population agreeing that they are ‘unsure about what Brexit will mean for me and my family’. Again, there are some interesting differences between the segments with 71% of ‘devastated pessimists’ agreeing with this statement, compared to 30% of ‘die-hards’.

It is for this reason that so many claim to be ‘bored of Brexit’: bored and very much wanting government to ‘move on’ and focus on the things they feel really matter: NHS, education, housing, jobs.

5. Beware confirmation bias when interpreting findings

During the referendum campaign in 2016, I was invited to speak about public opinion at many high-level events all around the UK. The audience was usually senior business leaders, politicians and commentators – classic so-called ‘elites’. I took to asking them who they thought would win. Everyone’s hand would go up for ‘remain’: it was obvious to them that this is what would happen. Everyone they knew – their friends, their colleagues, other senior people – said so.

I would point out that there was no evidence to suggest that they were right (the poll of polls was neck and neck within a margin of error, and throughout the campaign, more polls predicted ‘leave’ than ‘remain’). The same confirmation bias later sparked the ‘polls were wrong’ response.

But the polls had been right – and proved an accurate assessment of the divided nation that we now were, with ‘elite’ stakeholders utterly failing to read the mood of the nation beyond the M25.

Deborah Mattinson is founding partner at BritainThinks

0 Comments