Lies, Damn Lies and 'Self-Censorship' Statistics

Are students increasingly reluctant to express their views? Or do we live in a society?

The only evidence of a nationwide Free Speech Crisis is a series of dodgy statistics about “self-censorship.” Last month, in his post announcing the foundation of Anti-Woke U., the former president of St. John’s College gravely intoned that college students were no longer free to speak their minds:

In Heterodox Academy’s 2020 Campus Expression Survey, 62% of sampled college students agreed that the climate on their campus prevented students from saying things they believe.

Whenever we see statistics invoked as evidence of a nationwide trend, we need to ask a series of basic questions before we jump to conclusions.

1. What is this survey measuring?

The Heterodox Academy is not a disinterested think tank that just happens to collect statistics on youthful free expression. It is an active promoter of “viewpoint diversity” as a value and an enthusiastic participant in the Campus Free Speech panic. It’s fine for organizations to have political preferences, but let’s not be naive about the project here.

The number in question comes from the 2020 Campus Expression Survey, the only piece of “academic” work this think tank appears to produce. The survey asks students whether they agree with the statement “The climate on my campus prevents some people from saying things they believe because others might find them offensive.”

This question is meaningless. It’s not asking students whether they, personally, are reluctant to state their beliefs. It’s asking them whether the campus climate causes some people to censor themselves.

If anything, 62% sounds too low. I went to a third-tier regional university where social-justice activities were marginal and edgelord comments in class were commonplace. Even so, is it fair to assume that some students were holding back their opinions to avoid offending people? Of course. We … we live in a society. This question says nothing about which opinions are being held back, nor whether students feel this “campus climate” is a good thing.

The same problem crops up in the rest of the self-censorship “literature.” A cartoonishly biased survey from 2018 found that 60% of students have refrained from expressing their political views, I am not kidding, once or more during their time in college (again: how this is not 100%).

Another pretended to find that 80% of college students were self-censoring but only got to that figure by counting up everyone who didn’t answer that they had literally never had an opinion they didn’t express. Only one in five said they self-censored regularly.

Remember, these surveys are the only evidence that self-censorship is a widespread problem on college campuses. Even with methodologies deliberately designed to produce affirmative results, all they demonstrate is that students have refrained from speaking up a single time.

2. Does this number mean what the researchers say it means?

Even if we accept that most college students think other college students are censoring themselves, we still have to determine whether that’s evidence of a nationwide trend.

It is lazy thinking and shoddy scholarship to measure the prevalence of a phenomenon by asking people whether it is happening to somebody else. Public perceptions on nearly every major political issue, from crime to income inequality to immigration, bear little resemblance to reality.

It’s especially lazy and shoddy when it comes to moral panics. Numerous false beliefs have been propped up by statistics that seemed to indicate a growing threat, but in hindsight only reflected the proliferation of media scare stories. In 1987, at the height of the “stranger danger” panic, 36% of Americans thought children had a 50-50 chance of being abducted before they turned 18. A 2003 survey by the Tort Reform Association — an industry-funded think tank that only exists to perpetuate the “frivolous lawsuits” myth — more than half of Americans agreed with the statement that “many people use the justice system almost like a lottery.”

These results are so absurd that I’m not going to bother debunking them, but notice how they come from polls with the same features as the “self-censorship” survey: They’re not asking respondents to report their own experiences, they’re asking respondents what they think is happening in the rest of the country.

This is a crucial distinction, especially at a time when consumers are shifting their news consumption from local to national outlets. America is an imagined community: We only have direct experience with a tiny sliver of society — our neighborhood, our job, our friends — and rely on the media to inform us of everything happening everywhere else.

Survey respondents are members of the same imagined community as everyone else. College students read The Atlantic and watch YouTube explainers too. When they agree with the statement “some people are reluctant to express their beliefs on my campus,” they’re not only drawing upon their direct experience. They’re also drawing upon their impressions from media reports — the same reports this survey is intended to generate.

3. What does the rest of this survey tell us?

It is abundantly clear that the Campus Expression Survey was designed to produce the panickiest statistics possible. The rest of the poll asks students whether they would be reluctant to express their views on various social issues in class. Here are the results:

Looks bad, right? Nearly half of students feel like they can’t talk about politics in class. A quarter can’t express their true beliefs about race. And the numbers are getting worse. [Cathy voice]: Aack! My views!

But that’s not what the survey asked nor what the answers mean. Check out the wording of the questions:

See it? The survey didn’t ask students whether they felt comfortable discussing race in class. It asked whether they felt comfortable “discussing a controversial issue about race” in class. Same goes for all the other categories: A controversial issue about politics, a controversial issue about sexual orientation, etc.

Controversial issues are, by definition, likely to produce disagreement. Human beings are generally reluctant to bring up things they know will cause dischord, especially in public settings. Is it any surprise that college students don’t want to volunteer their views on “controversial issues” — a term that the researchers let respondents define for themselves — in an auditorium with hundreds of their classmates?

If you keep reading the survey, you find students themselves saying the same thing. Of the respondents who were reluctant to express themselves in class, the majority were worried about being criticized by other students.

This is not evidence of an academic sector crippled by orthodoxy; it is evidence that 19-year-olds want other people to like them. Some of the other reasons for not speaking up (e.g. professors would dock their grade) could be a sign of a worrying trend, but that’s entirely dependent on whether they are actually happening. Students might be afraid of ideological retaliation for the same reason parents were afraid of child kidnappers in the 1980s: Because the media told them to be.

I know I’ve said this before, but it makes no sense to be concerned about “self-censorship” as a problem unto itself. We are a diversifying society in the midst of rapidly advancing technology and changing social norms. In 1960, only 37% of college students were women and 4% were Black. If it was easier to express conservative views back then, that is largely because college students were more monolithic in terms of race, gender and social class — you don’t have to worry about offending Blacks or Muslims if there aren’t any in the room.

It is utterly mundane for broad social shifts to change which views are acceptable in academic and professional settings. Some views (“climate change is a Soros conspiracy”) deserve criticism and it’s good for society that their adherents feel uncomfortable expressing them. Some contexts (physics classes) discourage students from blurting out their politics takes because it’s a waste of everyone’s time. The same organization whipping up a panic about “campus orthodoxy” also found, on its own survey, that 15% of respondents don’t feel comfortable talking about “non-controversial” topics in class.

We finally found it, an actual crisis worth panicking about: One in six American college students doesn’t feel safe saying “baths are nice” on their own campus. Look what the libs have done to this country.

4. How do these findings fit in with other research?

Perhaps the clearest sign of the Heterodox Academy’s bad faith is its hilarious attempt to argue that the “problem” of self-censorship is getting worse. Here’s the first page of findings from the survey’s executive summary:

Wow, students were more reluctant to speak in 2020 than 2019! Say, when did you guys do those surveys?

Ahh yes, so we’re looking at a “campus climate” survey that compares a normal year to one in which campuses effectively did not exist. As usual with these studies, the executive summary blithely notes that its findings are utterly invalid.

This study is oinking with intellectual sloppiness, but there’s a real question here. This survey, and others like it, argue that “self-censorship” is a relatively recent phenomenon, the result of post-millennials entering college and woke libs taking over administrations. The co-founder of the Heterodox Academy dedicated an entire book to this thesis, arguing that campuses were being “transformed” by “new ideas of social justice.”

The problem with this argument is that it’s not clear which period of alleged non-self censorship we should be nostalgic for.

If we’re supposed to believe that today’s college students are more censorious than they used to be, that’s laughable. Public opinion polls show huge percentages of young people cheering the “cancelation” of racist, Muslim, atheist, gay and communist professors as far back as we have polling data. Youth conformity was seen as such a problem in the 1960s that there was a mini-moral panic about it.

Notice too how young people are the least censorious age group in almost every year for which we have polling data. According to these polls, we should be worried about self-censorship in workplaces and nursing homes, not universities.

Nor is there any evidence that conservatives are more likely to self-censor than liberals. Occupations from dentists to accountants to small business owners lean heavily to the right. According to an informal Police Magazine poll, 84% of law enforcement officers supported Donald Trump in 2016. I got an e-mail a few weeks ago from a left-wing dude in finance who says he constantly “self-censors” around the strip-club conservatives that surround him at work. It’s telling that the partisan skew of other professions receives a fraction of the media concern aimed at colleges.

Reluctance to speak about difficult topics is not new, exotic, partisan or confined to academia. As historian Jonathan W. Wilson points out, the organizations whipping up a panic about “self-censorship” often find in their own surveys that college campuses are among the least censorious environments in the United States. Somewhere between one-third and one-half of Americans don’t feel comfortable discussing their political opinions at work. More than two-thirds say they don’t talk politics on social media. Forty percent avoid the topic with their own families.

These are all indicators of the same (actual, not-fake) trend. Dozens of studies have shown that Americans are increasingly walled off from each other according to geography, education, class and race. The cause of this “Big Sort” is disputed — some researchers say Americans are moving closer to their political tribes, others say they’re radicalizing in place — but it’s true that Democrats and Republicans overlap less than they used to.

This is a huge shift with a wide range of downstream effects and I’m not going to pretend they’re all positive. But the Heterodox Academy (and everyone else crocodile-crying about “self-censorship”) wants you to ignore the rest of this trend and worry only about conservatives, only their beliefs on social issues and only their expression on college campuses.

This is what moral panics do, they take broad social trends and amplify the aspects that threaten members of the majority. There is no evidence that white people, straight men or political conservatives are being systematically disadvantaged on college campuses, much less society at large. The only thing these surveys tell us is that conservative students are slightly more likely than liberals to think something and not say it.

That is not censorship, it is adulthood.