We Need More Than Just Diversity Training

Several weeks ago, I wrote an article in response to New York Times coverage of race relations at Smith College, which argued for a more methodical and nuanced approach to accountability and investigation of identity-based prejudice and harm. While researching the article, I came across a surprising scholarly consensus — apparently, anti-bias training doesn’t work. 

At least, the forms of training that the college usually institutes don’t work — and can even have the opposite of their intended effect. Some scholars have gone so far as to say that anti-bias training is simply “window dressing” with the intention to “preempt lawsuits” rather than actually solve problems.

A 2006 study of over 700 U.S. companies found that mandatory diversity training was the least effective measure in diversifying the workplace and creating a more inclusive social environment, even leading to a backlash against groups seen as benefiting from the training, like Black women. Another study in 2013 showed that the presence of structures that purport to defend diversity and protect marginalized groups actually lowers dominant groups’ willingness to engage with claims of discrimination because it increases perceptions of a fair system, even when it’s not. And in 2016, researchers demonstrated that even the presence of pro-diversity messaging can make members of dominant groups feel as though they are more likely to be discriminated against unfairly, while still failing to reassure minorities that they will be treated fairly. These conclusions echo nearly a thousand studies on the prevention of prejudice that have been carried out since the early 20th century.

I found this particularly concerning given that educational programs seem to be our college’s go-to disciplinary measure, in addition to probation, when confronted with toxic subcultures on campus. In 2016, when the men’s cross country team was caught exchanging vile misogynistic messages via an email chain to first-years, the college responded with probation followed by a vague “educational process.” More recently, when the men’s lacrosse team was found responsible for a racist team culture that tacitly allowed for team members to feel comfortable directing racial slurs at a Black teammate in 2020, the college again turned to probation and undefined “educational programming” to solve the problem — though notably taking the additional step of firing the coach for allowing such a hostile culture to continue for so long. Even the broad-reaching anti-racism plan — promoted by the college in response to the #IntegrateAmherst op-ed series, the #ReclaimAmherst initiative, the @BlackAmherstSpeaks Instagram page and a national reckoning with American racism — mostly focuses on diversifying the community and educating community members about racism. 

None of this is to say that we should simply abandon the project of advancing diversity and inclusion altogether, only that our means of doing so should be more rigorous than just trying to teach community members about the existence of racism and bias — especially when such a focus is more likely to prompt hostility than change. Bias is a pervasive aspect of human society and behavior, which can be difficult to accept for those who are expected to wield unbiased power, like the police or, more locally, our professors and administrators. But because bias is so hard-coded into the way we perceive the world around us, a simple acknowledgment that it’s there doesn’t really impact the way we behave when it comes to day-to-day decision making. 

Even those who rely on diversity training to make a living acknowledge that it is better suited for short-term and narrow outcomes. Joelle Emerson, founder of diversity and inclusion consulting company Paradigm, wrote that her company has three unique features of its focused diversity training sessions: 1) they aim to strike a balance between acknowledging bias as a human trait (to combat defensiveness) while also pushing for participants to actively challenge bias in their daily lives; 2) they focus on specific situations which could occur in the workplace; and 3) they share specific actions people can take to avoid biased decision-making rather than simply focusing on awareness. These strategies might improve anti-bias training when used as a remedial measure but still don’t provide for long-term solutions.

Other obvious solutions might be focusing on anti-racist bureaucracy or recruiting a more diverse staff (two initiatives also mentioned in the college’s anti-racism plan), but scholars suggest that these quick, one-step solutions can’t solve the problem either. The unfortunate fact of the matter seems to be that bias and discrimination are so deeply rooted in our institutions and culture that working to remove them will take more than one administrative response to a protest. It won’t be something we’ll see in our time on campus and may not even be something we see within our lifetimes. Of course, we shouldn’t give up on fixing the problem, we just need to redirect ourselves toward a comprehensive reshaping of the culture and institutions in play. So what can we do to build for the future?

A recent 2021 policy study provides some building blocks for organizational structures to build long-lasting diversity — structures should foster accountability, intergroup contact, affinity groups, inclusive messaging and processes that circumvent interpersonal bias. These are supported by a synthesis of over 30 years of scholarship on corporate diversity by Dobbin and Kalev. Dobbin and Kalev argue that the best diversity programs engage participants in the process rather than forcing it on them, and take advantage of aspects of human psychology like cognitive dissonance to promote diversity rather than prompt backlash. They propose voluntary engagement, mentoring, intergroup contact and social accountability as the most effective methods for increasing diversity, arguing the best way to do so is to make them a part of an ongoing and inclusive organizational culture rather than implementing them only occasionally as remedial diversity programs. 

After all this, I suppose we’re left with the question of what these measures would look like at Amherst. While I’m far from an expert and am also distant from administrative decision making, my research does provide some ground for a few proposals. First, the college should refocus its remedial educational programs along the lines proposed by Emerson if they do not already align with her philosophy — the last thing the college wants is to inadvertently provoke racist or misogynistic backlash from already hostile athletic environments. But secondly, and more importantly, the college should try to implement long-standing programs that are seen as simply a part of campus life at Amherst rather than separate diversity and inclusion initiatives. 

One example of a useful long-term program is the Enhanced Advising Program offered to Summer Bridge students to navigate the FLI experience. This program engages faculty in mentoring like the corporate mentorships followed by Dobbin and Kalev, which led to changes in thinking on diversity. Opt-in programs allow mentors to feel as though they are part of the solution rather than part of the problem, furthering mentor embrace of pro-diversity or, in this case, pro-FLI ideas. They convince mentors of their mentees’ competence and deservingness, which mentors then project to other members of the same group, ultimately making the campus experience better for everyone, even those uninvolved with the mentors directly.

Another, shorter program worth considering is the LEAP program during First-Year Orientation. LEAP engages different students in unfamiliar environments and projects as equals, exactly the type of intergroup contact that scholarship shows can be successful. However, the program is confined to just a weekend, hardly enough time to build the relationships necessary to challenge preconceived biases, especially in the long-term. Perhaps a long-term program built off of the idea of LEAP would yield more long-term effects, with groupmates meeting again and again over the course of their time at the college and ultimately shifting their perceptions of one another and the social groups that they represent.

The building blocks are already in place at the college, but there is so much more to do before we will have begun to solve the problem. Simply mandating anti-bias training and recruiting higher numbers of marginalized people as planned for in the college’s recently updated anti-racism plan might help, but it won’t be able to make the sort of long-term change that we hope for on its own. Instead, the college needs to pursue the implementation of long-lasting programs, events and activities that become an integral part of campus culture for each and every group on campus, challenging biases persistently and at a deep enough level to effect real change in the ways people think and behave.