Poll: 73% say ‘Brexit dividend’ is a lie

A poll conducted on Monday 18th June 2018 found that 73% of those asked said the claim of a ‘Brexit dividend’ was a lie. 11% of respondents said that there would be a Brexit dividend, with the remaining 16% undecided. The sample size was 1,003, with a margin of error +/-3% (*).


A fake graph to demonstrate confirmation bias

A fake graph to demonstrate confirmation bias

If you’ve read this far, your initial reaction to this ‘poll’ is likely to have been determined by your existing beliefs about Brexit. If you oppose Brexit, you were probably more likely to have seen this as further evidence that your view is right. If you support Brexit you probably haven’t even read this far, but will have dismissed or ignored this article on the basis of the headline itself.

A psychological explanation often offered for this effect is confirmation bias (Darley and Gross, 1983). Confirmation bias is the tendency to seek evidence to confirm your existing beliefs rather than look for evidence that might counter them. Regardless of the actual truth of the information, finding support for your beliefs boosts your confidence in them. Crucially, this makes it less likely that people holding these beliefs will alter them.

Many people on the pro-EU side of the debate are placing a lot of faith in calling for a ‘people’s vote’ on the final EU exit deal. They express confidence (often citing the way that opinion has subsequently changed on the Iraq War pursued by the Blair government) that people won’t be fooled again.

I remain unconvinced that the outcome of any such referendum would be different.

Although opinion pollsters YouGov claim there has been a slight drift towards people thinking that the decision to leave the EU is a bad one, the difference is nothing like as pronounced as the shift over the Iraq War.

There’s also another important difference compared with the Iraq War – Brexit is a current issue. On both sides of the argument, people still have a lot of psychological capital invested in their beliefs. Much of the shift in opinion over the Iraq War seems to have happened afterwards, when it was seen to be both a disaster and with a premise based on a lie.

The challenge for those of us who want no truck with Brexit is to overcome the confirmation bias of the leavers. If I was well enough to attend, I’d be at the march in London on the 23rd June. But no matter how large and well organised it is, it’s unlikely to have much impact in shifting opinion.

What’s needed as well are emotional, media attention-grabbing demonstrations of the benefits of remaining in the EU. The equivalent of the Farage/Rees-Mogg fish throwing incident, if you will.



(*) For the avoidance of doubt, these figures are completely made up. Sorry. (But that doesn’t mean they bear no resemblance to the truth and that the Brexit dividend isn’t a lie, naturally).


Update 18th June – 2200: Sky News has published a genuine poll in┬áthe last few minutes that does indeed indicate that the majority of those asked say the ‘Brexit dividend’ claim is a lie.



Darley, J.M. & Gross, P.H. (1983). A hypothesis-confirming bias in labelling effects. Journal of Personality and Social Psychology, 44, 20-33.

ED209 seen question progress

I think I’ve finally, finally identified three reasonable articles to supplement the children and the legal system chapter from book 4 and the three ‘set’ articles. I also have an outline plan for the essay; now I just have to write it and commit the relevant bits to memory (how simple that sounds … !)

My aim is to have got that done by the end of this coming Sunday, then I can focus my efforts on the rest of the revision I need to do. Sitting on the train this afternoon with my notes, I came to the conclusion that even if I follow my current plan of revising strategically (or what I hope is strategic), there’s an awful lot to cram in.

I’m glad I chose the “legal” option for the seen question – if nothing else, some of the researcher names in the field of deception are familiar from (the pre-2008) DSE212. Aldert Vrij in particular. He seems to have been on a one man crusade since the late 1990s to dominate the published research in this field (325 articles and counting), and my goodness, when he’s not conducting original research he’s busy doing meta studies or writing textbooks. The great thing about Vrij is that his papers are very readable. They’re also full of things that challenge ”common sense’. For example, despite what most people would like to believe of themselves, studies show time and time again that the general public are rotten detectors of lies. People in general perform at worse than chance. That’s right, if you’re trying to work out if someone is lying to you, don’t bother. Toss a coin instead and if it comes down heads they are. Only specific kinds of professionals have been shown to perform at better than chance in detecting lies from demeanour alone.

Findings like these make you wonder about the validity of the whole trial by jury system, especially when you consider the number of miscarriages of justice that eventually get unearthed.

Back to the revision …