NEW: "Extraordinary U: The HxA Model of Statement Neutrality"

When should university leaders "weigh in" on controversy? A principled approach.

Read the Brief.
Heterodox Academy
Back to Blog
Diversity scaled
September 16, 2020+Musa al-Gharbi
+Diversity, Equity, and Inclusion (DEI)+Campus Policy

Diversity-Related Training: What Is It Good For?

In wake of George Floyd’s murder and the protests that followed, many colleges and universities have been rolling out new training requirements – often oriented towards reducing biases and encouraging people from high-status groups to ‘check their privilege.’ The explicit goal of these training programs is generally to help create a more positive and welcoming institutional environment for people from historically marginalized and underrepresented groups.

As I have explained elsewhere, there is a long literature on the benefits of diversity on knowledge production. However, many of the approaches to training people how to navigate and utilize diversity were implemented by corporations, non-profits and universities before their effectiveness had been tested rigorously (if at all).

Although the precursor to contemporary diversity training, sensitivity training, actually dates back to the mid 1940s, diversity training became especially important beginning in the mid-80s to early-90s. Why? Starting in the late 70s through early 80s, universities began enrolling significantly higher numbers of women, minorities, and people from middle-class and lower-income backgrounds. Soon thereafter, employers found themselves with a much more heterogenous labor pool. They had to face, often for the first time, some of the challenges that come along with the benefits of diversity — as people with increasingly divergent backgrounds and perspectives were put side by side and tasked with common goals. Diversity training was intended to help mitigate those challenges.

Due to their lack of validation, however, it was difficult to gauge how effective these programs were with respect to their stated goals. Over the decades that followed, it became increasingly obvious that many of the problems that diversity-related training was supposed to help mitigate seemed to be persisting or growing worse, raising the prospect that many widely-used interventions may be ineffective or even harmful. And so, beginning in the mid-90s, social scientists set out to systematically test the actual outcomes of these trainings.

In the years that followed, a robust empirical literature was built up measuring the effectiveness of diversity-related training programs. The picture that has emerged is not very flattering.

Resources on these points:

The limited initial research suggesting diversity-related training programs as efficacious was based on things like surveys before and after the training, or testing knowledge or attitudes about various groups or policies. And to be clear, the training does help people answer survey questions in the way the training said they ‘should.’ And many people who undergo the training say they enjoyed it or found it helpful in post-training questionnaires.

However, when scientists set about to investigate whether the programs actually changed behaviors, i.e. do they reduce expressions of bias, do they reduce discrimination, do they foster greater collaboration across groups, do they help with retaining employees from historically marginalized or underrepresented groups, do they increase productivity or reduce conflicts in the workplace — for all of these behavioral metrics, the metrics that actually matter, not only is the training ineffective, it is often counterproductive.

Training is Generally Ineffective at Its Stated Goals

The stated goals of these training programs vary, from helping to increase hiring and retention of people from historically marginalized and underrepresented groups, to eliminating prejudicial attitudes or behaviors to members of said groups, to reducing conflict and enhancing cooperation and belonging among all employees. Irrespective of the stated goals of the programs, they are overwhelmingly ineffective with respect to those goals. Generally speaking, they do not increase diversity in the workplace, they do not reduce harassment or discrimination, they do not lead to greater intergroup cooperation and cohesion – consequently, they do not increase productivity. More striking: many of those tasked with ensuring compliance with these training programs recognize them as ineffective (e.g. see Rynes & Rosen 1995, p. 258. This essay by The Cut is also great).

Resources on these points:


Training Often Reinforces Biases

Often, when people attempt to do fact-checks, they begin by underscoring the falsehood, and then proceed to try to debunk that falsehood. This can create what psychologists call an ‘illusory truth effect,’ where people end up remembering the falsehood, forgetting the correction – and then attributing their misinformation to the very source that had tried to correct it! A similar effect seems to hold with antibias training. By articulating various stereotypes associated with particular groups, emphasizing the salience of those stereotypes, and then calling for their suppression, they often end up reinforcing them in participants’ minds. Sometimes they even implant new stereotypes (for instance, if participants didn’t previously have particular stereotypes for Vietnamese people, or much knowledge about them overall, but were introduced to common stereotypes about this group through training intended to dispel said stereotypes).

Other times, they can fail to improve negative perceptions about the target group, yet increase negative views about others. For instance, an empirical investigation of ‘white privilege’ training found that it did nothing to make participants more sympathetic to minorities – it just increased resentment towards lower-income whites.

Encouraging people to ignore racial and cultural differences often results in diminished cooperation across racial lines. Meanwhile, multicultural training — emphasizing those differences — often ends up reinforcing race essentialism among participants and can lead to pigeonholing people into stereotypical roles. It is not clear what the best position between these poles is (such that these negative side effects can be avoided), let alone how to consistently strike that balance in training. Mutatis mutandis, similar patterns seem to hold with respect to training on bridging gender differences.

Resources on these points:


Training Can Increase Biased Behavior, Minority Turnover

Many diversity-related training programs describe bias and discrimination as rampant. One unfortunate consequence of depicting these attitudes and behaviors as common is that it makes many feel more comfortable expressing biased attitudes or behaving in discriminatory ways. Insofar as it is depicted as ubiquitous, diversity-related training can actually normalize bias.

Perhaps most pernicious, sometimes by making people conscious of status hierarchies within an organization or society writ large, members of the in-group can become convinced that it is, in fact, in their interests to reinforce said hierarchies — or can come to believe that outgroup members will flourish at their own expense. As a result, they may become more invested in preserving social privilege than they were before training.

For others, the very fact that the company has diversity-related training is proof that it is a non-biased institution. This perception often reduces concerns about bias and discrimination – by oneself or others. As a consequence, people not only become more likely to act in more biased ways, but they also react with increased skepticism and hostility when colleagues claim to have been discriminated against.

Meanwhile, those who are discriminated against become more likely to rationalize mistreatment by others in the institution after undergoing diversity-related training (for the same reason, because they believe the institution must be fair in virtue of its commitment to diversity-related training; indeed, minority employees are often called upon to lead diversity reviews themselves). Consequently, they become less likely to actually report or address wrongdoing.

In other cases, people come to believe as a result of training that reporting incidents of bias or harm will generate an automatic, impersonal and highly-punitive process that will lead to outcomes that victims themselves wouldn’t necessarily desire or endorse (and will have little control over): destroying perpetrators’ lives (rather than merely addressing discreet harms), radically upsetting social ties, and dramatically changing how people interact with both the accuser and the accused. That is, institutional messaging on how aggressively they pursue bias incidents leads many to avoid reporting them out of fear that the process itself will generate more harm than good.

As a result of these unintended effects of training, problems persist unabated and harms go unaddressed — often leading to higher turnover among the very populations that the programs were ostensibly designed to render more comfortable.

Resources on these points:


Training Often Alienates People from High-Status Groups, Reduces Morale

Diversity-related training programs often depict people from historically marginalized and disenfranchised groups as important and worthwhile, celebrating their heritage and culture, while criticizing the dominant culture as fundamentally depraved (racist, sexist, sadistic, etc.). People from minority groups are discussed in overwhelmingly positive terms, while people from majority groups are characterized as typically (and uniquely) ignorant, insensitive or outright malicious with respect to those who are different than them. Members of the majority group are told to listen to, and validate, the perspectives of people from historically marginalized or disadvantaged groups — even as they are instructed to submit their own feelings and perspectives to intense scrutiny.

In short, there is a clear double-standard in many of these programs with respect to how members of dominant groups (typically men, whites and/or heterosexuals) are described as compared to members of minority groups (i.e. women, ethnic/ racial minorities, LGBTQ employees). The result is that many members from the dominant group walk away from the training believing that themselves, their culture, their perspectives and interests are not valued at the institution – certainly not as much as those of minority team members — reducing their morale and productivity.

The training also leads many to believe that they have to ‘walk on eggshells’ when engaging with members of minority populations. By calling attention, not just too clear examples of harm and prejudice, but just as much (or more) to things like implicit attitudes and microaggressions, participants come to view colleagues from historically marginalized and disenfranchised groups as fragile and easily offended. As a result, members of the dominant group become less likely to try to build relationships or collaborate with people from minority populations.

Resources on these points:


Increasing Perceptions and Salience of Bias/ Discrimination Can Have Adverse Impacts on Minority Populations

Diversity related training often emphasizes the prevalence of various forms of bias and discrimination in the workplace and strives to increase sensitivity of participants to subtle and inadvertent expressions of identity-based bias or ignorance. This approach is premised on the assumption that if members of more dominant groups are more readily able to perceive bias and discrimination, they will grow less likely to engage in biased or discriminatory language or behaviors. However, to the extent that members of historically marginalized and disadvantaged groups also become more likely to perceive words, actions, policies and people in their environment as biased or discriminatory, this is likely to have a negative effect on their lives and workplace flourishing.

There is abundant research demonstrating that heightened perceptions of racism, sexism, discrimination, and identity-based violence and inequality have highly adverse effects on the psychological (and even physical) well-being of women and people of color. To summarize the central findings: the more people perceive themselves to be surrounded by others who harbor bias or hostility against them, and the more they view their life prospects as hostage to a system that is fundamentally rigged against them, they become more likely to experience anxiety, depression, psychogenic and psychosomatic health problems, and to act in antisocial ways, often to the detriment of their professional success and overall flourishing. Likewise, holding a cynical view of society and one’s fellow citizens is robustly associated with “less success, less job and life satisfaction, worse health, dramatically less flourishing, more negative emotion, more depression, and increased suicide attempts.”

Put another way, telling women and minorities that everyone around them harbors conscious or unconscious antipathy towards people like themselves (whether they overtly express it or not), or encouraging people from historically marginalized and disadvantaged groups to notice and take offense at an ever-expanding sphere of ‘problematic’ words and behaviors — to interpret things they formerly perceived as benign instead as a form of symbolic violence committed against them — this would likely have pernicious effects on ‘people of color’ and other minoritized populations. Likewise, constantly underlining historical and contemporary oppression and inequalities, pressuring or incentivizing people from underrepresented groups regularly testify about experiences they’ve had with bias and discrimination, encouraging them to interpret unfortunate outcomes in their own lives to prejudice and discrimination — practically speaking, these seem likely to lead to increased anxiety, depression and alienation among people from underrepresented groups rather than making anyone feel more included or empowered or enhancing their well-being and life prospects.

Resources on these points:


Focus On: Implicit Attitudes

Implicit attitudes are one of the most commonly relied-upon constructs in contemporary diversity-related training. However, there are severe problems with these constructs – as hammered home by meta-analysis after meta-analysis: it is not clear precisely what is being measured on implicit attitude tests; implicit attitudes do not effectively predict actual discriminatory behavior; most interventions to attempts to change implicit attitudes are ineffective (effects, when present, tend to be small and fleeting).

Moreover, there is no evidence that changing implicit attitudes has any significant, let alone durable, impact on reducing biased or discriminatory behaviors. In fact, to the extent that discriminatory behaviors are attributed to implicit biases, people often view them to be outside of one’s control — and therefore, view it as unjust to hold people accountable for said behaviors. Implicit bias training can even lead to reduced concern about bias, and adopt a fatalistic approach about it, due to the perception inculcated in training that bias is ubiquitous and not something people from high-status groups can actually rise above.

In short, the construct itself has numerous validity issues. Training to combat implicit bias has no demonstrable benefit — and may be even be counterproductive with respect to changing behaviors.

Resources on these points:

Focus On: Microaggressions

Contemporary diversity-related training often draws significant attention to microaggressions – small, typically inadvertent, faux pas involving people from historically marginalized and disadvantaged groups. The cumulative effects of microaggressions are held to have significant and adverse impacts on the well-being of people from low-status groups. However, although the microaggressions framework goes back to 1974, there is virtually no systematic research detailing if and how microaggressions are harmful, for whom, and under what circumstances (indeed, there is not even robust conceptual clarity in the literature as to what constitutes a microaggression).

A single meta-analysis published in 2019, perhaps the first meta-analysis trying to empirically substantiate the microaggression harm claims, found that some types of microaggressions had a correlation with adjustment outcomes — in particular, ‘micro’ assaults (basically, overt hostile and discriminatory statements, which many have argued should not even be considered ‘micro’ aggressions; they are overt aggression, which is very different from ‘small inadvertent slights’). Across the board, the effects were small, and were primarily related to internalized (rather than physical or externalized) adjustment outcomes. And even this inference is limited because, as the study authors emphasize, “very few studies tested whether microaggression predicted adjustment outcomes above and beyond overt discrimination and individual difference factors, and examined the indirect mechanisms that may link microaggression to adjustment outcomes.”

Given how weak the literature is demonstrating harm caused by microaggressions (let alone to whom, and under what circumstances), it should not be surprising that there is no systematic empirical evidence that training on microaggressions has any significant or long-term effects on behavior, nor that it correlates with any other positive institutional outcomes.

In fact, when presented with canonical microaggressions, black and Hispanic respondents overwhelmingly find them to be inoffensive – and there is ample reason to believe that sensitizing people to perceive and take greater offense at these slights actually would cause harm: the evidence is clear and abundant that increased perceptions of racism have adverse mental and physical consequences for minorities. In short, not only is there no evidence that training on microaggressions is valuable for improving the well-being of people from historically marginalized or disadvantaged groups, there is reason to believe it could actually be counter-productive to that end.

Resources on these points:

Mandatory Training Causes Additional Blowback

Although diversity-related training programs are generally ineffective, and often bring negative side-effects, they tend to work better (or at least, be less harmful) when they are opt-in. Mandatory training causes people to engage with the materials and exercises in the wrong frame of mind: adversarial and resentful. Consequently, mandatory training often leads to more negative feelings and behaviors, both towards the company and minority co-workers. This effect is especially pronounced among the people who need the training most. Yet roughly 80% of diversity-related training programs in the U.S. seem to be mandatory.

If an institution is going to include diversity-related training, it should offer it as a resource for those who want to learn more. To encourage more people to volunteer for the training, its value and purpose should be linked to specific organizational and development goals. Small incentives could be offered for those who take part, rather than the current norm of sanctioning those who do not.

Resources on these points:

Training Comes at the Expense of Other Priorities

We are in a period of educational austerity. Creating, implementing and ensuring compliance with diversity-related training programs is expensive. In a world where these training programs consistently advanced diversity and inclusion goals within an organization, or enhanced intergroup cooperation and overall productivity, then these costs could be justified – even during a time of belt-tightening. However, when the training is typically ineffective or even counterproductive, it seems antithetical to the pedagogical purpose of the university to dump still more money into these programs, even as many departments are seeing hiring freezes or budget cuts, and contingent faculty are being laid off en masse (disproportionately people from historically underrepresented and disadvantaged groups).

Indeed, ineffective diversity-related training programs often crowd out much more substantial efforts that could be undertaken to actually enhance diversity and inclusion within institutions of higher learning. Why do universities instead double-down on training despite its demonstrated ineffectiveness? The short answer is that, even if training is expensive and doesn’t work, it is relatively easy to implement – and it allows universities to show (including, often, in court) that they are doing something to address prejudice, discrimination and inequalities… even if what they’re doing is, in fact, ineffective.

However, universities are institutions that regularly claim to embody and inculcate such values as evidence-based reasoning, respect for facts, commitment to truth, etc. Universities are doing a bad job at modeling those values for students insofar as they force upon them (and upon the faculty who are supposed to be instructing them!) pedagogical materials that are demonstrably ineffective or even counterproductive.

It insults, rather than honors, the memory of George Floyd to offer empty gestures like these in his name. Indeed, as Cyrus Mehri aptly put it, “When you keep choosing the options on the menu that don’t create change, you’re purposely not creating change. It’s part of the intentional discrimination.”


Get HxA In Your Inbox

Related Articles
Abhishek Saha Protecting free speech in universities
Protecting Free Speech in Universities: Insights from the UK
February 14, 2024+Abhishek Saha
+Viewpoint Diversity+Open Inquiry+Campus Policy+Campus Climate
Make a donation
Make a Donation

Your generosity supports our non-partisan efforts to advance the principles of open inquiry, viewpoint diversity, and constructive disagreement to improve higher education and academic research.

This site use cookies.

To better improve your site experience, we collect some data. To see what types of information we collect, read our Cookie Policy.