Why Doesn’t Diversity Training Work? The Challenge for Industry and Academia

Introduction

Corporations and higher education institutions have been offering diversity training for decades. Yet countless studies dating back to the 1930s indicate that anti-bias training does not reduce bias, alter behavior, or improve the workplace. Despite these shortfalls, organizations continue to rely on diversity training due to concerns relating to optics, litigation, and perceived lack of commitment to address structural racism. This study calls attention to the ample evidence that short-term diversity training alone cannot lead to sustained and long-term improvements to attitudes and behaviors, especially towards women and minoritized populations. Instead, diversity and anti-bias curriculum offered over a longer period, such as  through a college course, has resulted in reductions in measured bias.  

This article by Frank Dobbin and Alexandra Kalev identifies reasons for the failure of diversity programs and remedies that address problematic features of training. The authors also present evidence that training is most effective as part of broader, systemic efforts to address bias and discrimination. Further, the study finds that even if employers design a diversity course that reduces bias, the intervention may not lower workplace discrimination. This is because intercepting unconscious bias does not decrease discrimination, since discrimination is a result of learned behavior or organizational practices. Organizations must view training as part of larger diversity initiatives that address both implicit biases and structural discrimination. 

Frank Dobbin and Alexandra Kalev are professors of sociology at Harvard University and Tel Aviv University, respectively. Dobbin’s field of study includes organizations, inequality, economic behavior, and public policy. Kalev’s work focuses on gender, racial and ethnic diversity in the workplace, and organizational restructuring. Dobbin and Kalev collaborated to develop an evidence-based approach to diversity management that has been published in The New York Times, CNN, and National Public Radio, among other platforms. 

Methods and Findings

Dobbin and Kalev conducted a literature review to understand why diversity and anti-bias training do not lead to sustained improvements in bias and discrimination. The authors find that short-term interventions, such as diversity training, do not change attitudes and behaviors that have been learned over a lifetime of media exposure and real-world experience. Instead, anti-bias training can be counterproductive by making stereotypes more prominent and thereby reinforcing them. Further, training and anti-discrimination programs often results in organizations and individuals developing unrealistic confidence and becoming complacent about their own biases. This can result in employees not taking responsibility for discrimination and becoming blinded to hard evidence of discrimination. Research also finds that training often makes white individuals feel excluded and therefore reduces their support for diversity. For example, white people may exhibit hostility and resistance during training regardless of whether the trainer uses a message of multiculturalism or color-blindness. In addition, diversity training is less effective when it requires mandatory participation and emphasizes legal curriculum, as this can make participants feel a loss of power. For example, white participants who participated in diversity training in one experiment resented the loss of autonomy and external pressure to control their prejudice, thereby compounding white participants’ bias. 

The authors offered key lessons to organizations. First, organizations must acknowledge that when people do something good (e.g., attend diversity training) they are likely to feel permitted to do something bad afterward (e.g., discriminate in hiring). For example, when experimenters described subjects’ employers as non-discriminatory, subjects did not censor their own biases.  By recognizing this tendency,  organizations can prevent managers from becoming complacent and believing that the organization has addressed discrimination through training. Second, organizations should frame diversity training as being inclusive of the majority culture; this can help to ensure the majority responds more positively and is more willing  to support multiculturalism. Third, organizations should ensure that training is seen to be voluntarily chosen, not externally imposed. By mandating and imposing anti-bias training on employees, organizations tell employees that they need to change, which can lead to weakening their commitment to diversity.

Conclusions

The authors recommend that organizations view training as part of wider diversity programs, since diversity training has been found to be ineffective in isolation. Similar to health and safety training, diversity should be integrated into other workplace training. Additionally, organizations should develop long-term diversity initiatives that not only address implicit biases but also target structural discrimination. Dobbin and Kalev found that discrimination is a result of learned behavior or organizational practices, and therefore must be seen as distinct  from unconscious bias. While workplace programs may address unconscious bias, organizations will not lessen discrimination without making structural, systemic changes. 

Many corporations and higher education institutions are faced with the same challenge of increasing diversity in executive and faculty ranks. Research from social science can help organizations identify successful strategies from ineffective initiatives, and therefore allocate resources more effectively. For example, traditional human resources policies that aim to reduce discrimination and promote diversity can be  counterproductive when not incorporated into a suite of long-term organizational reforms. Formal hiring and promotion criteria, such as those in job tests and performance rating systems, may both limit managerial discrimination and  hinder diversity in leadership, depending on how the measures are structured. In addition, although diversity training can improve the effects of certain diversity programs, employers should complement training with the right programs. Initiatives such as special college recruitment programs, formal mentorship, diversity task forces, and management training programs have been found to be successful. These programs offer an opportunity for executive staff to connect with individuals from different identity groups (i.e., racial, ethnic, gender), empathize with minoritized populations, and become more adept at supporting systemic changes to enhance diversity.