Class and the referendum

Most Leave voters are middle class because most voters are middle class

On last night’s Newsnight, Danny Dorling claimed that the vote to leave Europe was largely a middle-class, English vote.

This is true, but misleading. Most leave voters are middle class for the simple reason that most voters are middle class and English. In this way it’s exactly like the claims that Danny made about the North, which I also argued were misleading.

The evidence for Dorling’s claim comes from a post-referendum survey carried out by Lord Ashcroft.

That survey found that 59% of all Leave voters were middle class, defined as belonging to approximated social grades AB or C1.

You might disagree with this definition of middle class. If so, you should gather your own survey data using your preferred class schema. I’ll accept Dorling’s definition.

According to the 2011 census (E&W; SCO), 53% of people aged 16 to 64 are middle class. (The reason these figures apply only to 16 to 64 year olds is that this particular class scheme is based primarily on occupation, and that assigning social grades to those 65+, many of whom have no occupation, is tricky, and was not done in the 2011 census).

According to the same Ashcroft survey used by Dorling, 67% of people who voted in the referendum were middle class (Table 1 in this pdf).

Because middle class voters represent such a large proportion of total voters, it’s possible for most Leave voters to still be middle class, even if the middle class voted overwhelmingly to Remain.

In the referendum we had, 54% of middle class voters voted to Remain. (Compare that to the 64% of working class voters voted to Leave). We can alter that rate, and see whether it affects the claim that most Leave voters were middle class.

Suppose that 67% of middle class voters had voted to Remain. The proportion of middle class Leavers would then be

• 67% (their proportion in the voting population) times
• 100–67% (the proportion of middle class voters who vote Leave)
• = 22%

The proportion of working class Leavers would be

• 33% (their proportion in the voting population) times
• 64% (the proportion of working class voters who vote Leave)
• = 21%

If the claim that “most Leave voters were middle class” is still true when two-thirds of middle class voters vote Remain, then the claim seems to depend almost entirely on the fact that most voters are middle class.

I don’t know why Dorling continues to concentrate on the absolute numbers of people voting Leave and Remain, to the exclusion of information about the rates at which different types of people vote Leave and Remain. It’s not like he hasn’t been told about it. Let’s just hope that this line of analysis isn’t contagious.

P.S. The studio discussion following the clip I linked to above was much more nuanced. See here on iPlayer, from 29:50.