Much has been made of the supposed conflict between Equality, Diversity, and Inclusion (EDI) policies and protection of free speech at universities. Indeed, in a recent publication for the think tank Politeia, Cambridge professor John Marenbon argued that “EDI destroys British universities from within” by creating “an atmosphere of fear and intimidation, leading to creeping conformity.”
Similarly, the investigative report by Dame Nicola Dandridge into the Open University’s handling of the intimidation campaign against Jo Phoenix, found that many academics there believed that the “principles of free speech seemed to be pointing in one direction, with EDI legislation and policies pointing in the other.” While this contradiction may seem obvious to some, there is very little empirical data on the effect of EDI bureaucracies on the culture of free speech at universities.
However, a recent study by Kevin Wallsten, a professor of political science at California State University, Long Beach, has attempted to measure the relationship between the relative number of EDI employees at an institution and the attitudes of its students towards free speech. While his conclusion are less than encouraging for advocates of EDI, a number of caveats are necessary before reviewing the results.
First, the study has not been peer reviewed and was not published in an academic journal, but rather in Professor Wallsten’s Substack newsletter. However, he conducts a meticulous statistical analysis, and is nuanced in all of his conclusions. Second, his sample size is rather small (71 universities) and only involves US institutions.
Finally, the measure of EDI is the ratio between the number of employees tasked with EDI-responsibilities and the total number of students at each university. As such, the measure cannot distinguish between different approaches to EDI, and may not accurately reflect how forcefully the policies are implemented at each school.
Despite these caveats, Wallsten reaches several fascinating conclusions. In brief, he finds that larger EDI bureaucracies are associated with (1) more tolerance of liberal, but not of conservative speakers; (2) no change in how comfortable students feel discussing their views in the classroom; (3) more student discomfort with discussing their views outside the classroom; (4) more acceptance of disruptive behaviour; (5) no improvements in students’ willingness to have honest conversations about difficult topics. It is worth reviewing each of these findings.
Using data from a 2022 FIRE survey of college students’ attitudes towards free speech, Wallsten constructs a measure for how tolerant students are of controversial speakers. Specifically, the survey asked students how likely they were to allow a speaker with a controversial viewpoint on campus, giving them several hypothetical examples of such viewpoints ranging from “Religious liberty is used as an excuse to discriminate against gays and lesbians” to “Transgender people have a mental disorder”.
The result of Wallsten’s analysis is that larger EDI bureaucracies are associated with more tolerance for controversial speakers with liberal views, but no change in tolerance for those with conservative views. It is important to note that the data does not show an increase in intolerance for conservative speakers, but rather a consistent level of tolerance/intolerance regardless of the number of EDI employees. These results remain statistically significant even after controlling for average student ideology on each campus.
With a different self-reported measure from the FIRE survey, Wallsten analysed how comfortable students felt honestly expressing their views in classroom discussions or in submissions of written work. Interestingly, the data showed no difference in this measure based on the size of a university’s EDI bureaucracy. Such feelings are probably more influenced by the individual professor or seminar leader, rather than the overall EDI policies of the school, but Wallsten adds that there is no good data on such factors.
However, the study did find a clear connection between the relative number of EDI employees and students’ level of comfort expressing their views outside the classroom, indicating a stifling effect on the general free speech climate. This relationship remains significant after controlling for potentially confounding variables, such as the average student ideology, perceptions of faculty ideology, and overall student diversity.
The two remaining findings of the study are somewhat more ambiguous. First, Wallsten demonstrates a bivariate correlation between the size of EDI bureaucracies and acceptance of disruptive behaviour, such as “Shouting down a speaker to prevent them from speaking on campus” or “Using violence to stop campus speech.” However, the relationship falls short of statistical significance after controls for average student ideology are added to the analysis.
According to Wallsten, this is most likely because the relative number of EDI personnel is correlated with both average student ideology and acceptance of disruptive behaviour. Indeed, there is a very strong relationship between the latter two measures. Perhaps the initial correlation would remain statistically significant if the data differentiated disruptive behaviour towards conservative and liberal campus speech, but the FIRE survey made no such distinction. As Wallsten explains, this means we can only conclude that EDI “probably does not reduce support for disruption and may slightly increase it.”
Finally, Wallsten considers the FIRE measure for whether students feel comfortable having “open and honest” conversations about divisive issues, such as abortion, racial inequality, transgender issues, or gun control. The results show that larger EDI bureaucracies are associated with students finding it more difficult to discuss these topics. However, the correlation ceases to be statistically significant once the controls are added. Again, this limits the conclusion to saying that although EDI may not directly hamper “open and honest” conversations, it clearly does not encourage them either.
Overall, Wallsten’s study shows multiple correlations between large EDI bureaucracies and less free speech and no correlations with more free speech. However, given the data used, it is not possible to determine the direction of these correlations. It could be that a student body with less tolerance for free speech somehow leads to larger EDI bureaucracies. Moreover, the sample size of the study is quite small and it might be the case that Wallsten unintentionally included highly unrepresentative institutions.
Even with these caveats, however, the results are intriguing, and certainly not encouraging for university EDI regimes. As mentioned above, the results are only from US colleges, but it would be interesting to see a similar analysis for universities in the UK. Likewise, as Wallsten’s study was only focused student attitudes to free speech, a similar survey of academics’ attitudes would surely be interesting as well.