EU Referendum Summary

EU Ref Blog Summary

EU Referendum – Summary


In the aftermath of the EU referendum much ink has been spilled in an attempt to explain the outcome and why people voted as they did.

In the month leading up to the referendum we conducted a study, in collaboration with the Guardian, that explored the subconscious ‘feelings’ that drove voters’ decision making. By feelings, we mean the information that has been embedded into the subconscious over time that subsequently informs our decision-making.

So what did the data show and how does it relate to the actual result?

Based on the research we conducted in February 2016, before the campaign got into full swing we, like other research companies were surprised by the final outcome suggesting a small majority in favour of the Remain campaign, and I will suggest some of the reasons for this error in a moment. But let me start by looking at what we got right.

  • Among those most in favour of Brexit (for further details see Blog 2) 74% prioritised immigration above all other issues (See blog 3 for further details)
  • At the other end of the scale, our data clearly suggested that feelings about the EU were most positive in Scotland (Remain 61%), London (Remain 95%), and Northern Ireland (Remain 99%)  (See blog 2 for more detail)
  • Our data also clearly showed a significant increase in ‘Leave’ voters according to age with half (49%) of 18 – 24 year olds feeling strongly we should remain in the EU compared with around a quarter (24%) of those aged between 50 and 60. (See blog 2 for more details)

Last but not least, our findings indicated that the feelings of those who were still undecided at the time of testing (Feb 16) were aligned closer to those who were in either of the two ‘out’ groups (Weak
out and Strong out).

We found that a latent mistrust of the EU lurked in many people’s subconscious, but at the time of testing was of insufficient power to override conscious doubts about the future outside of the EU. This might have tipped the balance towards voting leave at the last minute. When decision making becomes overly complex people tend to rely  on their ‘gut’ feelings.


Basing our prediction on the data collected in February, we expected that Britain would remain in the EU. Clearly, implicit attitudes change during campaigning and more so than we predicted.

As explained in Blog 1, we can only ever accurately estimate outcomes over which we have no control. It is possible to make increasingly precise weather forecasts, since their publication exerts absolutely no effect on the outcome. The same cannot, of course, be said about public opinion polls whose widespread dissemination can and does significantly influence voter intentions.

A second possible reason is that people who wanted out were more strongly motivated to actually go out and vote. Data shows that only around a third of younger voters actually did so on the day and, of course, no one knows how non-voters would have voted.

Since the methods we employed have never before been used to capture people’s implicit attitudes and motivation towards complex and abstract concepts like EU membership, we probably need even finer tuned tools to do so.

All in all, however, we believe the detailed results are a powerful demonstration of both the accuracy and the potential of this radically new approach to market research.

Written by Insa-Annett Tiaden & Dr David Lewis-Hodgson