Confirmation Bias and Vote Leave’s Crimes

Confirmation Bias and Vote Leave’s Crimes

The differing opinions over whether the fact that Vote Leave broke electoral law had any impact on the result of the Brexit Referendum offer one of the clearest examples of confirmation bias in recent history.

In the absence of confirmation bias, opinions should be uncorrelated. The crime wasn’t discovered (or even suspected) until well after the vote, so there’s no way one’s views on the impact could have affected an individual’s votes. And while many other political issues are correlated with voting Leave or Remain, ‘attitudes towards electoral wrongdoing’ is not, notably. I’m very confident that I could sit down with any of my Remain-voting friends to discuss a hypothetical case of electoral fraud in the nation of Elbonia and we’d reach very similar views, without political ideologies coming into it.

Indeed, on the objective case of ‘did they break the law’, almost everyone I know agrees – they did break the law, they deserved to be punished, and potentially they should have faced greater penalties (incidentally, this is your regular reminder that the Remain campaign and the Lib Dems also broke electoral law). But on the more subjective case of whether it had an impact on the result, views cleave almost perfectly along Remain vs Leave lines, with both sides highly confident, and able to give extensive reasons, as to why it did or didn’t change the result.

Now, you might think that more intelligent, or more educated people, might be less susceptible to confirmation bias. Unfortunately it doesn’t work like this. In fact, it may even make it worse.

In experiments done by Perkins, it was found that education made little difference at all and that, while intelligence did correlate with coming up with more arguments surrounding an issue, it only came up with more arguments on the favoured side of the argument, not the opposing side – potentially increasing the weight of ‘evidence’ in support even further! This, incidentally, explains why (from my experience) academics, collectively a group of highly educated and intelligent people[citation needed], appear to be just as, if not more, prone to group think and confirmation bias as any other sector or set of stakeholders I’ve worked with (even on issues as clear cut to anyone outside the sector as ‘is grade inflation happening’).

So what to do? Confirmation bias is with us and we can’t get away from it – as Haidt describes it, it’s the difference between ‘Must I believe this?’ and ‘Can I believe this?’ – and the answers are almost always ‘no’ and ‘yes’, respectively. Engagement with others can help, certainly, as can being aware of the issue – but as we can see on the Vote Leave issue, it’s no guarantee of success, as even those of us who’ve had plentiful discussions with others still cleave along partisan lines. And in groups without thought diversity, engagement with others can simply worsen the issue, creating group-thing that reinforces confirmation bias still further.

There’s no easy solution. On the titular issue, I’ve not changed my mind about whether the illegality impacted the result and nor, good reader, do I imagine that you will either – after all, we’ve each arrived at our view after careful consideration and for very good reasons! But perhaps, on this and many other issues, we could all do with being a little less confident in how certain we are.

This is a post about cognitive bias and its implications for us and for society. This is not a post on which to discuss whether or not Vote Leave breaking the law had an impact on the result of the referendum. Any comments seeking to discuss the latter will be deleted without warning. Exception: if you’re one of the rare people whose belief on that matter does cross the partisan lines, do feel free to share it!

11 thoughts on “Confirmation Bias and Vote Leave’s Crimes

  1. Hi Iain, I’m one of your rarities – I voted Remain but don’t think that Vote Leave’s illegal conduct affected the result. I do however wonder whether the divide is not so much Leave vs Remain as much as ‘second referendum/revoke’ vs ‘implement the result of the 2016 vote’, the latter group including a number of those who voted Remain (up to and including the PM). I suspect that there would be an even stronger correlation measured against that divide. In a way that makes sense – it becomes considerably more difficult to articulate a cogent case for a second referendum (or, a fortiori, revocation) if one can not impeach the result of the first. I’d be interested to see polling data along the various of: Leave, ‘Remain but implement’ and ‘Remain and second referendum/revoke’.
    Kind regards

    1. Excellent, I’m glad we found one! Thank you for sharing.

      I’ve no doubt you’re right about Honour/Revoke being even more closely correlated, but it wouldn’t show confirmation bias as clearly, because in principle (as you say) because the Honour/Revoke view could follow logically from the opinion about whether it made a difference (even if in practice it is often the other way round). Whereas in the case given the two views should genuinely be uncorrelated.

  2. Thanks for another insightful post Iain. I like the point that being more intelligent only really produces more reasons in favour of the favoured viewpoint, but not necessarily a stronger case overall. I think this is one of the underlying currents in the Brexit controversy; the establishment is heavily in favour of remaining, and as it includes a higher proportion of “intelligence” than the leave side (if one takes a measure like tertiary education as a proxy for intelligence, a la Goodhart’s analysis in The Road to Somewhere), the remain case is argued with more and more sophisticated sounding arguments than the case for leave, to the point where many remainers really can’t understand how supporting leave is even a coherent position to take at all. But for all that, and despite a referendum campaign that reflected this split very clearly (e.g. the government flyer setting out many impressive sounding arguments for remaining), the majority voted leave. As you observe in your post, while I don’t think this is necessarily an argument supporting the leave position in itself, it would be grounds for asking remain supporters to at least give the leave side more consideration as a reasonable position to take, and consequently to take some of the heat out of the language used to describe it (e.g. extremist, economic suicide, xenophobic etc).

    1. I agree, the superior attitude taken by (some, but too many) remain supporters has been one of the least admirable parts of the whole business.

      If you enjoyed the part about producing more reasons in favour, you may like (unless you’ve read it already) Jonathan Haidt’s ‘The Righteous Mind’ which looks at these things and how attitudes to morality vary across cultures and across politics/class within cultures. As someone with no background in psychology/sociology I found it very well-written and accessible (review will hopefully be coming at some point!).

      1. Thanks Iain. You’re not the first person to recommend this book to me; it’s on my reading list and getting closer to the top!

  3. Hmm… not sure if I’m an example or not. I’m agnostic as to whether the illegal activity influenced the vote or not – I don’t think is a question that can be reasonably answered. Certainly the Leave campaigns did a better job at persuading people than the Remain campaign, but how much that was legitimate campaigning and how much that was illegal is difficult to quantify.

    But I don’t think you can ever re-run a vote except in case of clear mendacity – such as ballot-stuffing. But the punishment for wilful breaking of electoral rules has to be more than a fine to the institution – there has to be a personal punishment as well. For me, the threat of serious personal punishment (including time in prison) for electoral offences is both morally defensible and necessary to maintain a rigorous democracy.

    The influence of foreign powers on Western democracy (and in particular on elections) is something that worries me greatly (and was/is a strong reason that motivated me to vote Remain).

    1. Completely agree about the serious personal punishment. Given we (rightly) routinely imprison for perjury and perverting the course of justice, there’s an obvious discrepancy here.

      I’m not keen on the influence of foreign powers on democracies either – I also tend to think we should have better disclosure rules for charities receiving money from foreign sources and that DFID/FCO should stop funding lobby groups in other countries, too. I find there tends to be a bit of hypocrisy amongst many who protest this, as they usually welcome the interventions made by Obama, Lagarde and others in favour of Remain. You may not be most though – would you include those interventions in your concerns?

      1. I struggle to draw an equivalency between what Obama did (expressed his viewpoint when invited to do so) and what Russia did/is doing (such as hacking the DNC servers). I’m happy to call foul on both sides when appropriate (for example, I’d agree with you on the govt’s mailshot just before the election campaign opened for example – although I think this alienated more people than it persuaded!, as well as on the Osborne “punishment budget”). But for me, Obama and Lagarde’s contributions were standard campaigning tactics – I didn’t see anything nefarious in what they said or how they said it. And others have weighed into the Brexit debate on all sides (including Le Pen, Varoufakis, Putin and Trump on the leave side) – I don’t have a problem with any of that.

        1. Yes, good point, I hadn’t been thinking of the actually illegal stuff, like hacking. That’s clearly completely unacceptable (though also the sort of thing we should EXPECT hostile, or even just adversarial, foreign governments to be doing and actively use GCHQ and similar to defend against).

  4. I identify with the ‘if I think hard enough I can find a plausible reason to ignore your point’ issue. One offset in my experience is studying the topic – the more data you have the more challenging the mental gymnastics to still believe your bias. Without checking any of the data I can believe anything about Vote Leave’s illegal advertising. If I read the Electoral Commission report on what they did, what the scale was, and how it compared to the size of spend on both sides, and the estimated impact, then my bias has a lot less space to play in.

    1. That’s a good point. I’m certainly not suggesting we go all out ‘we can know nothing’ and your approach, reading the underlying reports and looking at the raw numbers, alongside engaging with those who disagree to understand their arguments, is definitely a very good way to go about it. Certainly much better than joining a group-think self-congratulatory circle. I think though (a) most people don’t do this, including most intelligent/educated people and (b) on matters where it’s particularly hard to plot the causal chain and impacts thereof (such as in this case) the bias can still creep in.

Leave a Reply

Your e-mail address will not be published. Required fields are marked *