June 25, 2016

Are there any insights to be gleaned from Brexit vote?

By Paul Laughlin

BrexitSo, it’s happened. For good or ill, the UK electorate has voted for ‘Brexit‘; that is to leave the European Union (EU).

There will be plenty of coverage over coming weeks & months. Much will cover political or economic implications & the many options ahead.

But the focus of this blog is customer insight. So, what can the Brexit vote reveal for our specialism? Are there lessons to be learned for those using behavioural economics, research or data visualisation to understand & explain customers actions?

Where was Status Quo bias?

Given our interest in the application of Behavioural Economics, to interpreting both behavioural analytics & research results, one bias didn’t play as much of a role as expected. A number of commentators had followed past British voting behaviour & suggested that ‘status quo’ bias would play a major role. That means the unconscious bias, whereby people will choose to remain with the status quo, when they find decisions complex or too demanding. Often this is followed by post-rationalising more sophisticated reasons for their ‘choice’.

Anyway, that is not what happened & here is an interesting post, which The Economist published about ‘the morning after’:

Brexit and the markets: a seismic shock

INVESTORS are waking to a deeply unpleasant surprise. Despite the closeness of the opinion polls, most people seemed to think that the “status quo bias” would cause Britons to vote for Remain, especially as it was perceived to be in their economic self-interest. A remain vote was virtually priced in.

Research & forecasting public opinion

But beyond our focus on behavioural biases, we are also interested in best practice research. For that reason, elections are normally a great opportunity to test the robustness of polling methods & forecasts made. Of course, referendums like this are a much rarer event & so harder for pollsters to predict accurately. Undecided voters & ‘shy brexiters’ can come from a more diverse demographic distribution than traditional voting heartlands. Nevertheless, it must be disappointing for a number of polling agencies, that most appear to have called this one wrong too. In this thorough post, expert statistical commentator Andrew Gelman provides a useful analysis of why YouGov got it wrong:

Statistical Modeling, Causal Inference, and Social Science

Commenter numeric writes: Since you were shilling for yougov the other day you might want to talk about their big miss on Brexit (off by 6% from their eve-of-election poll-remain up 2 on their last poll and leave up by 4 as of this posting).

Brexit data visualizations can be so unhelpful

As well as great resources available online (like Andrew’s very consistent stats blog), events like this also highlight the opportunity for poor quality data, analysis or research. The case that struck me most this time was a truly awful data visualisation. Not only is it rather distressing, on a day when almost half of the UK population is feeling raw about the result. It also completely fails to convey any useful numbers, nor insights as a result. Check out this poor example of data viz, as a salutary warning (if you’re tempted to rush out quick infographics):

Big Crash in the world as the BRexit from the European Union

Pond just flat down on to the knees, hitting the low of decade. As the long awaited result of the Referendum Pole came up with the exit of Britain from the European Union. Fig seems to be very close …

That’s all, I can manage today, folks

That’s plenty for now. To be honest, the result has left me a bit shell-shocked today & I’m finding it hard to concentrate on my normal blogging output.

Anyway, I hope that content was useful & topical for you. Whether you are delighted or gutted today, I wish you a peaceful weekend & an insightful summer.