by Eric Kaufmann
Tuesday, 16
November 2021
Factcheck
10:45

Yes, there is an educational divide over ‘wokeness’

There is a political element too
by Eric Kaufmann
Credit: Getty

People who oppose wokeness don’t speak from personal experience of an elite educated culture of repression, but do so for ideological reasons, claims Olga Khazan in a new piece for the Atlantic. Citing new polling data on cancel culture, she finds little difference between the opinions of those with university degrees, who one would expect to have experienced speech policing more, and those without. Instead, variation in opinion on political correctness and cancel culture largely revolves around whether one is a Republican or a Democrat.

There is truth to Khazan’s observation, but it misses the mark because the survey only asks people about their attitudes to cancel culture and identity politics rather than their personal experience of speech repression. It therefore lacks the instruments that would paint an accurate portrait of the reality of cancel culture and self-censorship.

The Atlantic poll found, in common with other surveys, that most Americans agree the country is too politically correct. But around 55% of both college and non-college educated respondents agreed. The share of respondents who say cancel culture is a problem is only slightly higher (51%) among those with degrees compared to those without (45%). From views on defunding the police, to using gender-neutral pronouns to endorsing the use of ‘Latinx’, only a few points separate college graduates from those with a high school diploma.

In many ways this is not surprising. Indeed, scholars of public opinion repeatedly find that people are generally more concerned about society than their own situation because they form their opinions from the media and peers, and are influenced by the ideological lens through which they view the world. In Britain, for instance, Bobby Duffy’s work showed that 70% of people thought immigration was a problem in the country, but only 20% said it was in their local area. Views on immigration and other emotive issues are shaped far more by perceptions of what is going on nationally than personal experience. Hence it is unsurprising that the Atlantic survey finds that 70% those who voted for Trump said ‘cancel culture is a big problem in society,’ compared with 31% of Biden voters.

So does that mean cancel culture is just a Right-wing moral panic? Not quite. Cato and YouGov’s 2020 National Survey of 2,000 adults — twice the sample of the Atlantic survey — showed that 32% of Americans said they personally worried about missing out on career opportunities or losing their job if their political opinions became known. Overall, 23% of those without a degree worried compared to 34% of those with one, a statistically significant difference. 62% of Americans also said the ‘political climate these days prevents me from saying what I believe.’ On this question there was no significant difference between those with a college education and those without.

Emily Ekins, who authored the study, was kind enough to share the raw data with me. The numbers show that exposure to higher education really matters for how fearful conservatives are about expressing themselves. There are also important demographic differences. Young people, men and those who live in more ethnically diverse ZIP codes — all indicators of potential exposure to more sensitive speech environments — are significantly more fearful of their views becoming known.

Let’s focus on education level. In figure 1, controlling for age, gender, race and the share of minorities in ZIP code, we can see that among those who voted for Trump, those with higher levels of education are more worried about their career if their views became known. Among Trump voters with at least a Masters degree, 6 in 10 are worried for their careers.

Source: Cato Summer 2020 Survey

And while there isn’t a difference based on education in answers to the question about being free to say what you believe, there is when you break the results down by 2016 vote. As figure 2 shows, the higher a Trump voter’s level of education, the more they feel they have to self-censor. The reverse is true for Democratic and Independent voters, who feel less need to self-censor the more educated they are.

Source: Cato Summer 2020 Survey

Views on cancel culture, defunding the police and using gender-neutral pronouns are generally determined by ideology and partisanship. But when it comes to feeling the chill of cancel culture and self-censorship, characteristics which increase a person’s exposure to a more politically-correct environment — such as being young, male, or inhabiting a highly educated or diverse milieu — really matter.

Join the discussion


  • No one should ever have to fear that he might lose his job or damage his career for expressing unfashionable views. If only 1% of the respondents to this survey had experienced anxiety on that score, something has gone seriously wrong. That’s why the stats are largely beside the point.

  • Fair play. Let’s debate. First, I just don’t like this guy and the way he writes. You can agree, disagree, up to you. He’s an academic, writes like an academic, and it’s not my cup of tea.
    Now are these statistics useful? What can be gained from them? Another attempt to divide America. I personally hate these pseudo-surveys, even polls. A sociology professor, quite unethically, sends out a thousand fake resumes to potential employers from Usama bin Laden at 1 Jihadi way and when Usama is not called for an interview, it’s “proof” of racism. Can we return to reality for a moment–a concept most academics are unfamiliar with? Another example, my elderly mother was meeting with her doctor who looked at his laptop, not my mother, and, based on the data on his laptop, ordered an active stress test. My mother responded Why? The doctor responded: OK, you’re right. No test. Explanation: a stress test was called for when someone presents with these numbers–the stress test could indicate a heart transplant is needed. But not for an 80 something lady. Just because you have data does not mean it’s useful!
    I simply disagree with you that half of America doesn’t hate the other half. I haven’t conducted a survey–don’t need to–but this is what I believe. Agree, disagree, fair play. Don’t give up.
    Yesterday a lady, Katherine Eyre if memory serves, celebrated Austria locking down the un-vaxxed. Many, including me, strongly disagreed. But she hung in there, kept posting, kept defending her position. Good for her! That’s what UnHerd is supposed to be for!

  • This is an astute observation, and a reason why it is important that we don’t all become comfortable in our bubbles. We, who just know it is a giraffe, need to know why others think it is a rhinoceros. Of course, it may come as a suprise to when we find out, after investigations, that it is in fact an aardvark.

  • To get involved in the discussion and stay up to date, become a registered user.

    It's simple, quick and free.

    Sign me up