A New Approach to Diversity Training
On the futility of diversity training and what we can do to fix it.
Reading Time: 4 minutes
The recent spike in hate crimes, xenophobia, and religious intolerance in the U.S. reflects an undeniable trend: we are forgetting how to work together. So it’s unsurprising that corporate America invests over $8 billion every year into seminars teaching employees how to set aside their differences. “Diversity training” has been around for nearly eighty years, but only recently have its inherent shortfalls come to light.
To deliver the ideas of diversity training in a digestible manner, training administrators rely on online courses, group activities, or a combination of both. According to the Harvard Business Review, the ultimate goal of such programs is to “facilitate positive intergroup interaction, reduce prejudice and discrimination, and generally teach individuals who are different from others how to work together effectively.” Following its introduction after World War II, a time during which returning male workers had to adjust to the influx of women in the workforce, diversity training attempted to combat issues stemming from racism, sexism, and other forms of discrimination. In the present day, nearly half of all midsized companies continue to administer diversity training, with all Fortune 500 companies following suit.
Through 30-minute to full-day classes, diversity training aims to reverse decades worth of prejudice, implicit bias, and unconscious perceptions among coworkers. Aside from the impracticality of relying on such an alarmingly short time span to solve problems rooted in the subconscious, the format of diversity training makes it improbable that significant change could take place at all. Questionnaires designed to be passed require little more than common sense to locate the correct answer, which makes diversity training mostly ingenuine and a waste of time. The little effort needed to succeed in such programs is reflects the shortfalls of current diversity training curricula used in the corporate world. As a result, employees show resentment toward a program that preaches surface-level information they already know or can guess themselves. Diversity training is outright ineffective at eroding biases or encouraging a respectful company culture.
Not only does exposure to existing labels and categories heighten employee awareness of potentially dormant yet harmful biases, but it also facilitates an environment of hyper-sensitivity where productivity, honesty, and trust are compromised for fear of offending a colleague. Other research has pointed to another concerning aspect of diversity training; by presenting biases as common and a pressing concern, employees may leave training with an exaggerated perception of the frequency and rampancy of biases amongst their coworkers. Researchers have also found that when people believe everybody else is biased, they feel free to be prejudiced themselves. In one study, a group of managers was told that stereotypes were rare, while another group was told that stereotypes were common. Both groups were then asked to evaluate male and female job candidates. The managers who were told that stereotypes were common were more biased against women. In a similar manner, employees who are told their coworkers are inherently biased against them feel more comfortable making harmful judgements themselves. Why retain a moral high ground when that integrity is not present among your coworkers?
Similar research conducted over the past half-century points to the same conclusion: diversity training just doesn’t work. While a clear, universal alternative has yet to be popularized, certain initiatives taken by various corporations have shown early signs of success. In the previously lawsuit-ridden media company Bedia, diversity training was replaced with communication training for managers and executives, leading to a 10-year-long halt in workplace discrimination lawsuits. What made Bedia so successful was their rejection of a blanket approach in addressing deep-seated issues; they tailored their diversity programs to build upon understanding the impact of one’s actions. Incorporating communication, perspective role-playing, and goal-setting exercises into training curricula is more likely to make meaningful and lasting changes to discrimination in the workplace.
More recently, Brian Welle, Google’s director of people analytics, has initiated a new type of diversity training known as unconscious bias training which is used for avoiding the instillation of guilt and blame among its participants. As part of the workshop, Welle first introduces a presentation on the science of bias, stating, “Yes, we’re all biased. Yes, we’re all trying to fight it, and don’t worry—it isn’t your fault.” With this mindset, Welle believes that all discrimination, conscious or not, can be neutralized through awareness. As part of his workshop, Welle encourages employees to use consistent criteria to measure success, rely on data rather than on gut reactions when evaluating others, and most importantly, call out bias when they see it—even when the culprit is their own boss.
Since the introduction of Welle’s workshop, about 75 percent of the 75,000 employees working at Google have taken the course, with other companies following suit. In the U.S. alone, around 20 percent of all companies now offer unconscious bias training, including almost all big tech firms. By some estimates, 50 percent of all American corporations will offer unconscious bias training in the near future. Though such training has yet to provide evidence of definitive success, the search from companies across the U.S. for diversity training alternatives is a welcome sight. Yet, large corporations will make few changes inside their workplaces if they were to bear the burden of alleviating tensions among different groups on their own. Smaller companies and local businesses must adopt similar programs in place of diversity training and must ultimately address their complacency and “checkmark” attitudes toward a broken system.