It’s well known that people with less education were more likely to vote Leave in the Brexit referendum. Which is probably why pro-Remain commentators felt able to question Leave voters’ understanding of the relevant issues. Richard Dawkins referred to an “ignorant and misled public”, while Baroness King claimed that many people “were unaware of the far-reaching consequences of the EU referendum”.
But were Leave voters less intelligent? Yes, according to a new study. Chris Dawson and Paul Baker analysed data from a large UK survey in which respondents were given various tests of cognitive ability, as well as being asked how they voted in the referendum. The authors’ main finding is displayed in the chart below.
The x-axis is divided into ten deciles of cognitive ability. What’s a decile? Well, the first decile comprises the lowest-scoring 10%, the second decile comprises the next lowest-scoring 10%, and so on. And the tenth decile comprises the highest-scoring 10%. Each bar corresponds to the percentage of people in the relevant decile (shown on the y-axis) who said they voted Remain.
As you can see, around 40% of respondents in the lowest decile of cognitive ability said they voted Remain, compared to around 70% of those in the highest decile. So a clear majority of the most intelligent respondents supported Remain rather than Leave.
Dawson and Baker ran a multivariate analysis, and found that cognitive ability remained an important predictor of voting Remain when controlling for age, gender, education, political party and various other factors.
Their findings don’t actually come as a big surprise. At least two studies have already reported an advantage for Remain supporters on numerical reasoning – which is one component of cognitive ability. And another study (by yours truly) reported a positive association between average cognitive ability and intention to vote Remain at the level of local authorities.
What does come as a surprise is the authors’ interpretation. They suggest that “erroneous reporting surrounding the referendum might have complicated personal decision making, especially for those with low cognitive ability”. And say “it is also possible that those with lower levels of cognitive ability… are more receptive to divisive ideas”.
“If those lower in cognitively ability are more vulnerable to misinformation,” the authors write, “then political campaigns based on (mis/dis) information may prevail depending on the ability distribution of the electorate.” The implication being that the Brexit campaign was based, to a larger extent than the Remain campaign, on “misinformation”.
I’m not convinced. The Brexit campaign having (allegedly) engaged in “erroneous reporting” and “misinformation”, and having (allegedly) promoted “divisive ideas”, is not a plausible explanation for the fact that less intelligent people were more likely to vote Leave.
More importantly, the referendum was not primarily a disagreement over facts. It was a disagreement over values. The Leave side valued national sovereignty and control over immigration, while the Remain side valued being part of a large trade bloc, as well as being able to live and work in other European countries. Neither perspective is obviously wrong; each has to be argued on its merits.
Where cognitive ability comes in is that the Leave side’s values appealed more to those who score lower on that trait. Indeed, it has long been known that less intelligent people tend to have more socially conservative attitudes, particularly on issues like immigration. So it’s hardly surprising they were more likely to vote for the side representing socially conservative values.