The Impact of Diversity Training Programs in the Workplace and Alternative Bias Reduction Mechanisms
The Impact of Diversity Training Programs in the Workplace and Alternative Bias Reduction Mechanisms
- Alexandra KalevAlexandra KalevSociology and Anthropology, Tel Aviv University
Summary
Corporations often start their diversity journey by providing their managers or all workers with diversity training. These trainings were first offered as race-relations sessions in the 1960s and are among the most popular tools for diversity managers. Diversity training programs have changed their content during the decades, but they usually include live or online explanations about unlawful discrimination and bias, often supplemented by discussions of cultural differences and business needs for diversity. Despite their popularity and often high costs, a large body of research conducted over decades shows that most diversity training programs do not lead to long-term improvements in participants’ bias, attitudes, behavior, or workforce diversity. Some studies also show that training has negative effects on bias and diversity.
Factors that impede the success of diversity training or make them backfire include the hardwiring of cognitive biases and people’s complex reactions to direct attempts to change their biases, as well as the broader systemic biases rooted in everyday organizational routines. These suggest that common diversity training simply may not be the right tool for reducing bias and generating the changes needed for increasing workplace diversity. Some studies suggest that trainings’ effects could be improved by carefully designing them. This includes: avoiding training features that increase participants’ alienation, such as mandatory attendance, quizzes, and legalistic content; designing long-term training such that meaningful learning can be achieved; calibrating training to specific organizational challenges rather than using off-the-shelf content; and including ongoing collaborative contact with members of underrepresented groups and integrating training as part of broader diversity and accountability efforts. More research is needed to determine whether these types of training indeed produce sustained improvements in bias and diversity.
Alternative bias reduction mechanisms can be found in popular management models that increase collaboration between workers, such as cross-functional teams and cross-training. Such collaborative teams and training improve corporate performance and, as a byproduct, also reduce bias. Cognitive biases are affected by the work contexts in which individuals operate. Highly segregated workplaces, where White people and men meet women and people of color (or other underrepresented groups) primarily in marginalized jobs, deepen group boundaries and strengthen stereotypes. When organizations create cross-functional collaborations using self-directed teams and cross-training, workers from different groups have more opportunities to collaborate and, as studies show, biases and group boundaries are reduced, and leadership diversity increases.
Keywords
Subjects
- Human Resource Management
- Social Issues
Introduction: Diversity Training and Other Bias Reduction Mechanisms
The management of workforce equality and diversity has been on the agenda of U.S. corporations almost continuously since the 1960s, but the integration of women and people of color into good jobs has not followed suit. Trends in underrepresented groups’ entry into management have been underwhelming. The seven demographic groups for which there exist continuous annual corporate data collected by the U.S. Equal Employment Opportunity Commission (Black, Latinx, and Asian American men and women and White women) witnessed increases in their shares in management during the 1970s; but this progress slowed beginning in the 1980s, and for most groups it has virtually stopped (Dobbin & Kalev, 2022; Stainback & Tomaskovic-Devey, 2012).
This slow progress in diversifying corporate leadership took place alongside adoption by U.S. corporations of anti-bias and diversity programs. These programs have focused mainly on individuals’ biases. The most popular approach to debiasing individuals focuses on diversity training for educating workers and managers about inequality, their personal biases, and the law. Topics such as cultural differences and business needs are sometimes added. Some trainings require participants to take a quiz at the end; some take the form of role playing. Training is done mostly in sessions that last 1–4 hours and take place in classrooms, in conference rooms, or online, isolated from everyday work routines. Such diversity training has been almost the only diversity measure to spread and grow in popularity since 1964, when the Civil Rights Act was passed. It is also the only program for which robust scientific evidence shows little to no effects on management diversity (Devine & Ash, 2022).
Another approach to bias reduction entails replacing work routines that strengthen stereotypes and deepen group boundaries with routines that provide greater opportunities for collaborative contact across groups. Examples of such work routines are cross-functional work teams, which bring together workers from different parts of the organization to work on a project, or cross-training programs, where workers train in different skills from across the organization. These are not part of the diversity programs toolkit, but they lead to increases in diversity. Because work tends to be segregated along gender and racial lines, such teams and training tend to bring workers from different demographic groups together to collaborate. Research shows that these work structures reduce bias and improve the odds that workers from underrepresented groups will enter management. Taken together, then, the diversity efforts in most U.S. corporations since the 1960s have focused more on ineffective in-class training than on effective de-biasing of work routines.
This article discusses the impact of diversity training programs in the workplace and that of bias reduction mechanisms focused on collaborative work routines. It begins with a brief overview of research on cognitive biases in decision-making and interactions at work, and the malleability of these biases. It then reviews the history of diversity training and evidence of its effects on cognitive bias and diversity outcomes. The review is followed by a reflection on why diversity training does not boost diversity and whether and how corporations can make diversity training more effective. The third part of the article discusses the connection between segregated work routines and bias and shows that ameliorating task segregation and expanding opportunities for visibility and collaboration across groups reduces bias and improves workers’ career outcomes.
Cognitive Bias at Work
The categorization of people into in-group and out-group and the exaggeration of differences between groups are normal cognitive functions that help humans handle complex environments. In nature, these processes are necessary for survival—for example, by quickly distinguishing between a friend and a foe—and they operate on an automatic level, regardless of actors’ intent or awareness (Fiske, 1998; Greenwald & Lai, 2020; Krieger, 1995).
Although categorization is a natural process, the content of our categories is shaped by society. Categories and the stereotypes they entail feed off available representations in our social environments and lead to attribution errors and evaluation biases. Thus for example, movies and other popular cultural artifacts that portray men as heroes and women in supportive roles, or heterosexual lives as the norm and LGBTQIA+ lives as a rare otherness, strengthen our perceptions about what normal sexual orientation is (Ward & Grower, 2020; Welch, 2007). Similarly, racial and ethnic segregation in housing and schooling reaffirms group boundaries and stereotypes. At the workplace, managerial ranks that are filled mostly with White men confirm our categories and evaluations of who fits leadership jobs and who does not (Krieger, 1995; Sturm, 2001). Categories and stereotypes about gender, race, ethnicity, sexual orientation, ablism, and more are thus pervasive and deeply rooted.
A rich body of research based on laboratory and field studies has shown that such biases affect decision-making at every step of the labor process. At the recruitment stage, women’s scores on skill tests are judged more harshly than those of men (Rivera, 2016), and Whites and men are more likely to receive job offers and higher pay offers than their nonwhite and female counterparts (Correll et al., 2007; Pager et al., 2009). While at work, women and people of color receive lower evaluation scores for the same job performance as men and Whites (Castilla, 2012; Elvira & Town, 2001; Tsui & O’Reilly, 1989) and are less likely to be offered job training (Knoke & Ishio, 1998; Osterman, 2022). During layoffs, women and people of color are more likely to lose their jobs and less likely to be hired in bad economic times (Dencker, 2008; Kalev, 2014; Kim et al., 2021). Similar findings are shown for older workers as well (Canduela et al., 2012; Roscigno et al., 2007; Truxillo et al., 2015).
Cognitive bias also affects everyday interactions at work. Sociologists have long understood social interactions as a process in which the actors involved mutually define each other and the roles they play in the situation (Goffman, 1967; Ridgeway, 1997), often drawing on salient categories such as gender, ethnicity, and race. These interpersonal definitions and expectations affect participants’ behavior in ways that can perpetuate stereotypes. For example, both sides of the interaction tend to expect higher-status group members to outperform lower-status group members, an expectation that acts as a self-fulfilling prophecy (Fiske, 1998; Reskin, 2000). Thus, studies show that when whites and Blacks interact, whites’ racial stereotypes create hostility toward their partners—and this hostility has negative effects on the partners’ performance (Chen & Bargh, 1997). And in mixed-sex interactions, men have more opportunities to perform than women, and their performance is evaluated more positively (Reskin, 2000; Ridgeway, 1997). Cognitive biases, then, affect both decision-making and everyday interactions at work (Green & Kalev, 2008).
But our cognitive categories and biases are malleable. Studies point to two factors that may affect cognitive processes: individuals’ bias awareness and the structural conditions in which decisions and interactions take place (Blair, 2002; Fiske, 1998; Reskin, 2000). Research suggests that making people aware of their own biases and of the unequal outcomes of their actions may increase their motivation to change and lead to suppression of stereotypes and less biased behavior (Brooks & Purdue-Vaughns, 2007; Dasgupta, 2004; Devine, 1989; Monteith & Mark, 2005). Awareness-raising is often the first step in developing a diversity training curriculum (Roberson et al., 2013).
Beyond awareness, the structural context in which work is done affects bias. Hierarchy and separation strengthen group boundaries, while low power differentials and interdependence reduce actors’ reliance on cognitive shortcuts and weaken group boundaries. Consequently, work structures that allow for collaboration between workers from different groups can reduce biases (Pettigrew & Tropp, 2006; Ridgeway & Smith-Lovin, 1999). Barbara Reskin thus termed organizational structures as the proximate causes of discrimination—they can reinforce or constrain cognitive biases (Reskin, 2000, p. 320).
Taken together, research on the malleability of cognitive biases implies two major ways to reduce cognitive bias at work: changing individuals and changing work structures. The first strategy has been the most common since the 1960s, with no convincing evidence for its success. The second strategy is pursued by management models that are not part of the diversity repertoire, such as self-directed teams and cross-training programs, and have been shown to reduce bias and increase workforce diversity in good jobs.
Diversity Training: Decades of Attempts to Reduce Bias
Antidiscrimination regulations in the United States do not require training, but many large U.S. employers established race relations workshops soon after John F. Kennedy signed Executive Order 10925 in 1961. The Order required federal contractors to “take affirmative action to ensure that applicants are employed, and that employees are treated without regard to their race, creed, color or national origin” (U.S. Department of Labor, n.d.), and it gave federal contracting agencies the authority to discipline non-complying contractors. Executives of large firms wanted to signal that they were listening and taking action. They also wanted to inform their managers about the new requirements (Dobbin, 2009).
Early trainings blended speeches by key figures in the civil rights movement, such as Martin Luther King Jr., James Baldwin, and Malcolm X, with training materials on societal racial inequality and prejudice and on legal requirements. These were gradually replaced through the early 1970s by race relations sensitivity training, modeled after management sensitivity training that had become popular at that time as a means for increasing workers’ productivity. Race relations sensitivity training followed the key psychological insight that people are not always aware of their deep-seated prejudices that lead to biased decision-making and the belief that bias awareness will lead to behavioral changes (Dobbin, 2009).
The change in the political regime in the early 1980s and President Reagan’s emphasis on small government brought about a rebranding of antidiscrimination compliance as diversity management. Declines in antidiscrimination enforcement budgets freed employers from the feeling that Washington was looking over their shoulders. To survive the changing ecology, affirmative action officers and equal employment opportunity specialists needed to reinvent themselves as diversity managers, and race relations training quickly became known as diversity training (Kelly & Dobbin, 1998). Despite the new branding, the content of training did not change much; legal and bias awareness have remained the centerpieces of diversity training.
Diversity training traveled quick and wide. In 1976 the Bureau of National Affairs survey found that over 60% of the large companies at its Personnel Policies Forum offered training (Dobbin, 2009). By the turn of the century, national samples of small and large companies showed that between 40% and 50% of midsized and large employers had training (Kalev et al., 2006). Diversity training came to stand as a proof of diversity efforts. It is often recommended as a remedy by judges or agreed upon in consent decrees (Schlanger & Kim, 2014) and has become the go-to solution when a company gets into hot water. An oft-cited example is Starbucks’ closing of more than 8,000 stores across the United States for an afternoon on May 29, 2018, for diversity training, after an employee called the police on Black men waiting at a table. Delta Air Lines similarly offered unconscious bias training for all of its 23,000 flight attendants after the credentials of a Black physician were questioned when she attempted to provide emergency medical care to a fellow passenger (Crespo, 2018, cited in Devine & Ash, 2022). The beauty retailer Sephora shut down all its locations for an hour-long sensitivity workshop on June 5, 2019, after R&B star SZA reported that a Sephora employee had called security to make sure he wasn’t stealing (Bromwich, 2019). Diversity training has become so popular that it featured in sitcoms (e.g., The Office, Superstore) and in two presidential campaigns, Hillary Clinton’s in 2016 (Hahn & Gawronski, 2019) and Donald Trump’s in 2020 (Kalev & Dobbin, 2020).
Despite variation in detail, most diversity training curricula focus on educating managers about stereotypes and biases and about the legal implications of discriminatory behavior. Some training programs also discuss cultural differences and their business implications. The vast majority of the programs last between 1 and 4 hours. Rarely do training programs span more than one session (Ameri et al., 2021; Fitzgerald et al., 2019; Roberson et al., 2013). Most training is mandatory, further attesting to executives’ deep belief in its importance at least for signaling a good-faith effort to reduce discrimination. But research on the effects of training on biases, behaviors, and actual changes in workforce diversity does not support viewing them as such.
Diversity Training’s Effects
Several reviews of research on the effects of diversity training over the past decades have reached a similar conclusion: there is no evidence that diversity training alone has long-term positive effects on individuals’ biases, attitudes, and behavior or on organizational recruitment and promotion outcomes. There is mixed evidence about positive effects, especially on diversity knowledge and skill, especially in the short term, as well as findings on negative effects on attitudes, bias, and leadership diversity (Bezrukova et al., 2016; Devine & Ash, 2022; Fitzgerald et al., 2019; Kalev et al., 2006; Naff & Kellough, 2003; Paluck & Green, 2009; Roberson et al., 2013; Williams, 1947).
Most research on diversity training has focused on knowledge about diversity, skills for dealing with diverse populations, and bias. These are important outcomes that shape everyday routines in the workplaces as well as decision-making. Studies suggest that in the short term, diversity training can have positive effects on the first two outcomes. Kulik and Roberson (2008) found that 27 of 31 studies that used either pre-test/post-test assessments or a control group design showed some significant short-term improvements in knowledge of, or attitudes toward, diversity. A meta-analysis of 65 studies concludes that training typically has measurable positive effects on both diversity knowledge and skills (Kalinoski et al., 2013). Findings on the effects of training on unconscious bias are more mixed. In their review of 985 studies, Paluck and Green (2009) found inconsistent evidence and concluded that researchers need better-designed studies. Reviewing 47 studies of implicit bias reduction published between 2005 and 2015, FitzGerald et al. (2019) found at least partial positive short-term effects in the majority of studies, especially those where training involved exposure to counter-stereotypical exemplars. The authors observed, however, that most studies rely on small samples, and many lack clear methodological details. They concluded with advice to use caution when interpreting these positive effects and called for more robust research and replication. In a study that examined eight implicit bias interventions and one sham, all nine reduced bias—suggesting that subjects were gaming the implicit bias test. All effects dissipated within a few days (Lai et al., 2016). Findings on positive changes in explicit bias and in behavior following training are meager as well. A meta-analysis of 492 interventions conducted by Forscher et al. (2019) found weak immediate effects on unconscious bias and little evidence of reductions in explicit bias or changes in behavior. A test of 17 interventions to reduce whites’ bias toward Blacks found that 8 reduced unconscious bias, but none reduced explicit bias or any bias toward Hispanics or Asians (Lai et al., 2014). Looking at short-term results, then, knowledge and skill seem to be more malleable to training than implicit and especially explicit bias.
Studies on long-term outcomes are rare. One study has shown that training in several strategies for reducing bias led to a reduction in implicit bias that lasted for at least 8 weeks (Devine et al., 2012). A meta-analysis of 260 studies found positive long-term effects of training on cognitive learning, and smaller positive short-term effects on attitudes and behavioral learning (Bezrukova et al., 2016). They also find that training had better effects when it was complemented by other diversity initiatives, targeted to both awareness and skills development, and conducted over a significant period.
Does training lead to improved career outcomes for workers from underrepresented groups? The few studies that look at this question are not encouraging. When Rynes and Rosen (1995) asked 765 human resources experts whether their diversity training program increased hiring and promotion outcomes, fewer than one in three said yes. A study of 72 federal agencies found no effect of diversity training on promotion ratios for women or nonwhites and on dismissal and quit ratios for Blacks (Kellough & Naff, 2004; Naff & Kellough, 2003). Research that looked at nomination of women for star projects or awards found no direct training effects (Chang et al., 2019). A study that examined whether diversity training leads to changes in the share of women and people of color in management jobs found small, mixed effects. Mandatory training and training sessions that discuss the legal aspects of diversity have no positive effects and show some negative ones (Dobbin & Kalev, 2022). Training that avoids legal discussions has some positive effects on the share of men of color in management, but not of women (Dobbin & Kalev, 2022).
Most diversity training programs consist of a single session, although longer interventions may be more effective. In one study, a group of self-selected non-Black students showed a significant decrease in their implicit and explicit prejudice and stereotype scores after a 14-week-long prejudice and conflict seminar, with class discussions and readings about diversity. Students from the control group, taking another seminar, did not show such change (Rudman et al., 2001). Corporate diversity training, however, rarely lasts more than a day.
Taken together, meta-analyses of research on diversity training shows that this management tool has drawn a lot of research attention, albeit with much room for methodological improvement. In a nutshell, studies point to mixed short-term effects of training on bias and diversity knowledge and skill, and to meager long-term effects on bias and organizational diversity.
Why Does Diversity Training Fail?
Several factors impede the success of diversity training. Some of these factors can be alleviated by well-designed diversity training, such as adjusting the content of training to organizational needs (Roberson et al., 2013), framing training in a way that does not create resistance (Dobbin & Kalev, 2022), devoting effective length of training (Ameri et al., 2021), and embedding training within a broader diversity management plan and accountability structures (Bezrukova et al., 2016; Kalev et al., 2006). Other factors, such as the hardwiring of cognitive biases and people’s complex reactions to explicit attempts to change them, as well as the deeper systemic biases rooted in everyday organizational routines, mean that diversity training may simply not be an efficient tool for reducing bias and for pursuing the organizational changes needed to increase workplace diversity. Factors are listed in an order reflecting the potential to improve training if taking them into account.
Lack of Self-Study and Planning
Diversity training can often be likened to an over-the-counter painkiller. Rarely is it preceded by an organizational needs assessment conducted by a local diversity taskforce. Such an assessment can explore what the diversity challenges are in different functions and jobs, what kind of training is needed, who should be trained, and whether organizational conditions will support training so that learning can be implemented in everyday routines (Roberson et al., 2013). A needs assessment would also gather baseline data about workforce diversity and workers’ feelings of inclusion and readiness for diversity training. For example, given that attitudes toward diversity can influence receptivity to training, an assessment of these attitudes can guide the choice of training content and framing (Roberson et al., 2013). Absent this baseline, training programs are not tailored to the specific organization, barriers, and audience, and the assessment of training’s effectiveness is impossible. Having a working diversity taskforce, charged with data collection and analysis of local barriers to diversity, is thus a prerequisite for effectively tailored diversity training. But it is rarely part of the process.
Short-Term Sessions Make a Poor Platform for Learning
Most diversity training programs are short-term interventions, rarely spanning more than a single day. These make for poor learning platforms. Thus, even if the primary aim is education, typical diversity training programs do not even deliver on that goal. Studies show that learning over time is more effective than learning all at once (Ameri et al., 2021; Litman & Davachi, 2008). Memory works better when input is presented in smaller segments, ideally from different voices (Greene, 2008; Zajonc, 1980). Furthermore, people tend to resist ideas at first encounter but accept ideas that have had a chance to settle in. As advertisers have long realized, repeated exposure leads people to buy into ideas, even if they resisted them initially (Ameri et al., 2021). The idea of repeated messaging led some companies (such as GitLab, Husch Blackwell) and many universities (such as Harvard, Ohio State) and government agencies (such as the National Institutes of Health) to institute a speaker series around diversity, while human resources associations and diversity consultants (such as SHRM) have started offering such services.
Focusing on Individual Biases, Ignoring Biased Career Systems
Even if bias could be reduced in training, broader biased career systems will remain intact (Green, 2016). Modern career systems often look universal and unbiased but in fact create inequalities. Addressing such biases entails changing career systems to ensure that opportunities are open to all. This means, for example, recruiting at Historically Black colleges and professional associations of underrepresented groups, rather than only in majority-white colleges. It means creating formal mentoring programs to ensure that every junior employee is offered a mentor, rather than allowing mentoring to evolve naturally, which occurs less frequently for workers from underrepresented groups. It means opening skill and management training to all employees and reaching out to them, rather than limiting training to full-timers or to workers in certain jobs, and letting managers handpick their favorites. It means making the work-life interface manageable for all employees and not penalizing workers who juggle their work and family commitments. Addressing these biases in organizations’ career systems, research shows, increases workforce diversity in good jobs significantly and more effectively than diversity training (Dobbin & Kalev, 2022), and when diversity training is accompanied by other diversity initiatives, it may show more positive effects (Bezrukova et al., 2016).
Resistance to Control
Most diversity training programs are mandatory. Some argue that mandatory training signals organizational commitment to diversity, and this may increase trainees’ motivation and assures that training reaches those who need it most (Bezrukova et al., 2016; Roberson et al., 2013; Rynes & Rosen, 1995). Others argue that mandatory training signals control, which can trigger reactance—the need to maintain psychological autonomy (Brehm & Brehm, 1981; Rudman et al., 2001)—and backlash against the program’s goals. Plant and Devine (2001) found that when subjects were externally pressured to write a pro-Black essay, they complied but experienced high levels of anger, threat, and resentment, which then resulted in backlash against a pro-Black policy. When Legault et al. (2011) showed anti-prejudice materials to subjects, prejudice declined among those primed to think that they chose the goal of reducing prejudice, but rose among those primed to think the goal was imposed (see also Devine et al., 2002; Legault et al., 2007; Plant & Devine, 1998). Sociological research on job autonomy shows similar patterns: people resist controls regulating their work and act to undermine management goals (Gouldner, 1954; Hodson, 1996). Hence, the most common type of diversity training, mandatory training, sends a message that is likely to lead to backlash.
Biases Are Hardwired
Implicit biases have been embedded in us during our life course, through countless implicit and explicit societal messages about stereotypical traits of different groups (Devine, 1989). Implicit bias can be likened to a bad habit, and “the elimination of a bad habit requires essentially the same steps as the formation of a new habit,” which include the decision to stop old behavior, a memory of that decision, and repeated attempts to actually do so (Devine, 1989, p. 15). It also requires developing a new cognitive structure that is consistent with the new behavior. It is hard to expect such deep wiring to be untangled by a diversity training session.
False Sense of Confidence
Diversity training can foster a false sense of confidence among participants that their workplace is fair. Castilla and Benard (2010) argue that when decision makers believe their workplace is nondiscriminatory, they let their cognitive guards down and do not censor their biases. In their study, women received lower bonuses for the same performance scores as men when decision makers were led to think that the organization was meritocratic, but not in the control group. Others have found that workers react more harshly to discrimination claims in workplaces with diversity training (Kaiser et al., 2013) and that the presence of training can also blind women to unfair treatment against them (Brady et al., 2015). Thus, the presence of diversity training may strengthen bias due to its symbolic value.
Attempts to Suppress Bias May Increase Bias
Studies show that asking subjects to suppress bias actually increases it (Galinsky & Moskowitz, 2000). Stereotypes rebound after the suppression requirement is removed (Kulik et al., 2000; Macrae et al., 1994; Wolsko et al., 2000). Ironically, stereotypes can become more accessible as a result of interventions that ask participants to ignore them, as is often the case in diversity training.
Negative Reactions to Messages of Multiculturalism
While messages of bias suppression and color blindness may lead to a rebound of stereotypes, the message of multiculturalism may make whites feel excluded. Field and laboratory studies on diversity management (Ditomaso et al., 2003; Dover et al., 2016; Konrad & Linnehan, 1995; Plaut et al., 2011) show that whites are less supportive of and may resist programs that emphasize multiculturalism and diversity of identities. Plaut et al. (2011) show that whites associate messages of multiculturalism with feelings of exclusion (the “what about me” effect), which leads to resistance to diversity efforts. In the case of diversity training on topics related to sexual orientation, workers may feel training violates their religious beliefs (Kaplan, 2006). Whites tend to be more supportive of programs that promote color blindness or identity-neutral topics. Members of underrepresented groups react in the opposite way. They feel that the message of color blindness excludes them (Plaut et al., 2009, 2011). Taken together, training that emphasizes either color blindness or multiculturalism runs a high risk of alienating participants.
In sum, some of the factors that cause diversity training to be ineffective can be changed to improve training outcomes, while others may be more endemic to the idea of reducing bias through training. Executives can assemble a taskforce for a deep needs-assessment analysis and baseline data collection before training, offer voluntary rather than mandatory training, discuss cultural and business-specific content rather than legal content, design long-term diversity training rather than single-session interventions, and integrate diversity training within a broader change that opens up career systems to all. More research is needed to determine whether and how these changes will improve training. Other factors that impede training’s success may be more difficult to change. The facts that training can create a false sense of confidence, that cognitive biases are hardwired and that messages of either color blindness or multiculturalism may reinforce bias and cause workers from different groups to feel excluded—all these mean that diversity training may be a limited tool for reducing bias and promoting organizational change.
Reducing Bias by Working Together
Diversity training has been the most prominent managerial approach for bias reduction since the 1960s. Another means for reducing bias and stereotypes at work is by changing simple work routines that perpetuate inequality (Kanter, 1977; Reskin, 2000). Biases are shaped by the very way work is organized. In most workplaces today, men and women, whites and nonwhites, abled and disabled workers, work in separate jobs and spaces. The segregated structure of modern work harbors gender, ethnic, and racial segregation, creating a standing mechanism that perpetuates group boundaries, biases, and stereotypes and blocks diversity. The section “Working Apart Perpetuates Bias, Collaboration Reduces It” explains why most work structures increase bias and how collaboration can reduce it. The section “Self-Managed Teams and Cross-Training Reduce Bias and Increase Diversity” discusses two management innovations, self-directed work teams and cross-training programs, that improve collaboration, reduce bias, and open new opportunities for underrepresented groups.
Working Apart Perpetuates Bias, Collaboration Reduces It
The modern workplace is usually organized around a hierarchical and horizontal division of tasks that runs mostly along gender, racial, and ethnic lines, wherein women and people of color are concentrated in low-level, undervalued, dead-end functions. In law or financial firms, women and people of color are more likely to be passed over for large commercial clients, and in manufacturing firms they tend to hold the most menial jobs and receive the least training. Finance departments have mostly male employees and human resources mostly women; R&D has mostly white employees, production mostly nonwhites. In management, women and people of color are heavily represented in support and service functions, such as human resources, legal compliance, and public relations. In short, men and women, whites and nonwhites, rarely work together on equal grounds (Stainback & Tomaskovic-Devey, 2012).
The fact that men and whites at work mostly meet women, people of color, and workers from other underrepresented groups in lower-status positions perpetuates negative stereotypes about their competence and commitment; and workers in marginalized positions usually do not receive opportunities for proving stereotypes wrong, such as high-profile assignments, skill training, or networking with the higher-ups (Burt, 1998; Ibarra, 1992; Kanter, 1977; Knoke & Ishio, 1998; McGuire, 2000; Meyerson & Fletcher, 2000; Osterman, 2022). Furthermore, routine everyday interactions between workers in high- and low-status jobs perpetuate status beliefs and strengthen stereotypes (Ridgeway, 1997). For example, men are more likely to interrupt when interacting with women, and whites are more likely to feel nervous and avoid eye contact with Blacks (Green & Kalev, 2008). The segregated structure of work then serves as a bedrock for cognitive bias, stereotypes, and inequality.
Changing the segregated way in which work is done can expand opportunities for stereotype-negating interactions and reduce cognitive bias. Studies show that when interactions among men and women, whites and nonwhites, take place under equal and collaborative conditions, they are less likely to evoke stereotypes (Ridgeway & Smith-Lovin, 1999). Collaboration with peers encourages an investment of greater cognitive resources than working with subordinates, and this leads to prejudice reduction (Pettigrew & Tropp, 2006), lowers group boundaries, and fosters common identity (Reskin, 2000; Sherif et al., 1961; Tajfel et al., 1979; Turner et al., 1987). When it comes to older workers, for example, studies find that intergenerational contact facilitates common identity between younger and older employees (Iweins et al., 2013) and reduces age bias (Henry et al., 2015).
An early natural experiment brings these points home most convincingly. During the Second World War, the U.S. army was still segregated, but high numbers of causalities on the European front forced integration of some of the white companies. When sociologist Samuel Stouffer and his research team surveyed troops on their racial attitudes, they found that white troops in the integrated companies expressed significantly lower racial bias and higher willingness to work with Black soldiers. This was not the case for troops in nonintegrated companies. Fighting together as peers improved racial biases and attitudes. Hundreds of studies since have repeated this finding for different groups, in different countries and settings: working together as equals reduces prejudice and weakens group boundaries (Dobbin & Kalev, 2022; Pettigrew & Tropp, 2006).
Self-Managed Teams and Cross-Training Reduce Bias and Increase Diversity
Two popular managerial innovations adopted in the 1980s increase collaboration between groups at work: self-managed teams and cross-training. These programs break down task segregation and the group boundaries and stereotyping that come with them. The increased collaboration provides women and people of color more opportunities to negate stereotypes, reduces bias, and leads to increases in the share of women and people of color in managerial jobs in firms that adopt them (Kalev, 2009).
The ideas of teams and job enrichment have been around at least since the 1920s, promoted in Fritz Roethlisberger and Elton Mayo’s work at Western Electric’s Hawthorne Works in Cicero, Illinois, and, later, in Douglas McGregor’s model of humanistic management. These programs gained a renewed popularity during the 1980s in the context of rapid technological changes and intense international competition. Inspired by Japanese, German, and Swedish experiences, managers and scholars viewed moving away from the Fordist model of production, with its detailed division of labor and task segregation, toward cross-functional structures of teams work and skill development. These were viewed as an effective way to improve quality and competitiveness (Piore & Sabel, 1984). Studies have since shown that these management models outperform traditional, hierarchical management in a range of industries, from steelmaking to tech to retail.
Self-managed cross-functional teams typically bring together workers from different jobs, with different educational and occupational backgrounds, to hold frequent meetings, assume joint responsibilities on work tasks, share knowledge, and participate in decision making. They encourage members to contribute ideas, make decisions, and take responsibility for results. At least 4 in 10 firms use self-managed teams to perform some of their core tasks (Kalev, 2009). For example, in a high-tech company, engineers, technicians, and administrative assistants are members of a self-directed work team that meets a few times a week to design new products (Daday & Burris, 2002). In a bank, team members collaborate on phone services and technical tasks (Ollilainen & Rothschild, 2001), and workers in a paper mill plan key activities and tasks collectively, rotate jobs among themselves, and assume greater responsibility for production, quality, and safety (Vallas, 2003b).
The visibility and tie-building afforded by self-managed teams help to negate stereotypes and can translate into new career opportunities. These can arise, for example, from an offer by someone on the team for a lateral move to a job with better promotion prospects, from learning from a team member about a new training opportunity, or from the visibility gained when higher-ups look at the teams’ output. In a cross-firm study of more than 2,000 scientists in the biotech industry, Laurel Smith-Doerr (2004) found that female scientists are much more likely to be promoted to management in firms with self-managed teams than in traditional firms. The women she interviewed explained that teams offered them the opportunity to collaborate with a wide range of people who saw their contributions, and this translated into more visible projects that led to promotions. By contrast, in traditional firms, the department head often gets credit for his group’s work.
The team environment can also alleviate group differences that stem from status differences. In a high-tech company, the team structure mitigates the exempt/non-exempt divide, which is often also a gender and racial divide. An administrative assistant working on such teams attests, “non-exempt can now feel like they are not demeaned; they are treated as an equal part of the team” (Daday & Burris, 2002, p. 17). Teams may reproduce categorization and boundaries if their members expect women, for example, to do the menial tasks just like in a segregated job structure. But the team environment provides justification for fighting that. A team member is better able to push back when people ignore her suggestions or expect her to do the photocopying. In teams, all are equal by definition. Leaderless teams can provide new opportunities for hourly workers who are seldom seen by HR or even their own supervisors as “management potential.”
Unintentionally, then, self-managed teams reduce cognitive and structural biases and act as vehicles of diversity; they increase visibility, reduce stereotypes, and open new opportunities for women and people of color that are otherwise trapped behind a metaphorical curtain of undervalued jobs and negative stereotypes.
Cross-training consists of multiskilling programs that provide workers with knowledge of and experience in different jobs across the company. These are popular programs, adopted by at least 70% of mid- to large-size firms in the United States (Kalev, 2009). The content of cross-training varies widely. Some cross-training programs enrich workers’ skills and increase their motivation and job satisfaction (Adler, 1992; Campion et al., 1994; Ollilainen & Rothschild, 2001); others are exploitative methods in disguise (Smith, 1997, p. 322) where workers are pressured to perform more de-skilled work at a higher pace (Handel & Levine, 2004; Taplin, 1995).
Like self-directed teams, cross-training programs are not part of most firms’ diversity management toolkit, but by blurring boundaries between jobs, and creating a platform for workers to meet and learn about each other’s positions and jobs, cross-training can undermine the negative implications of job segregation on decision makers’ bias and on the careers of women and people of color (Roberge et al., 2011). Through rotating across jobs, women, people of color, and other marginalized workers can reach out beyond their job boundaries, gain exposure to new people, tasks, and skills, and demonstrate their capabilities and management potential. This is exemplified nicely in Ollilainen and Rothschild’s (2001) study of a cross-training program in a bank. Men refused to learn women’s phone service jobs, but women were encouraged—and took the opportunity—to gain new skills and learn how to perform multiple functions. Cross-training, the authors conclude, provides “an opportunity for lower-status women workers to outlearn, and perhaps even move into some of the higher-status tasks formerly reserved for men” (Ollilainen & Rothschild, 2001, p. 161; see Vallas, 2003a, p. 235 for a similar scenario on a production line). Ian Taplin (1995, p. 35) recognized that job rotation in a textile mill is no more than a “sweat method,” but observed that supervisors came to better appreciate the abilities of their low-skill workers after they saw their performance training across jobs.
Not surprisingly, then, self-directed cross-functional teams and cross-training have shown positive effects on the careers of women and people of color. In a study that analyzed data on more than 800 companies over 30 years, the adoption of these two programs has led to increases in the diversity of managerial ranks. Using a statistical analysis that took into account the presence of diversity programs and other organizational features, Kalev (2009) found that the share of white women and Black men and women in management grew by an average of 5% solely due to the adoption of self-directed work teams. Similar changes were observed after adoption of cross-training programs, which also led to increases in Asian men and women in management (Dobbin & Kalev, 2022; Kalev, 2009). These improvements in managerial diversity occurred without management treating such innovations as diversity programs. Implementing cross-functional teams and cross-training programs with diversity considerations in mind will enhance their impact. For example, managers can make sure that women and workers from underrepresented groups get an equal chance to be on cross-functional teams and training, that team deliverables are showcased, and that cross-training spans a wide variety of jobs.
Taken together, looking at research on managerial innovations and bias reduction over the last several decades, it is apparent that reducing bias by expanding opportunities for collaboration across functions and groups has been more effective in improving diversity than the average diversity training. Some diversity training programs incorporate this idea and include contact with others as part of training (Avery & Thomas, 2004; Rubenzer & Pierce, 2023). These are relatively rare and are mostly found in education systems rather than corporations, but they may serve as test cases and models to follow.
Conclusion
Corporations often start their diversity journey with talking the talk. They require managers to gather in a conference room for half a day and learn about unlawful discrimination, bias, culture, and business needs. They often spend significant amounts of dollars on annual diversity training sessions. Whatever their educational benefits, most of these training sessions will not reduce bias or improve attitudes, behavior, or workforce diversity.
How could executives maximize the potential of diversity training? Diversity training has been institutionalized as an industry and as a concept, and it is probably here to stay. Evidence reviewed in this article suggests that if designed and implemented differently, training could be effective. In particular, long-term voluntary training programs that avoid legal threats, including collaborative contact between different identity groups, tailored to specific situations, barriers, and audiences, and integrated within broader organizational efforts and accountability structures, are expected to be more effective (Avery & Thomas, 2004; Bezrukova et al., 2016; Devine et al., 2012; Fitzgerald et al., 2019; Kalev et al., 2006; Pettigrew & Tropp, 2006; Roberson et al., 2013).
More research evidence is needed about diversity training and the effectiveness of improved formats suggested by the literature. Additional key research needed on diversity training includes a wider range of results that extend beyond the individual level, over longer time ranges, and in real settings (Devine & Ash, 2022). This includes looking at intersectional categories and categories common to contexts outside the United States. Our stereotypes vary across gender, race, ethnic, immigrant, age, and other categories, and their intersection. Each disadvantaged group experiences different sets of barriers, and these vary across cultural and national contexts (Cuddy et al., 2009). Research on diversity training has been relatively focused on the more measurable categories of gender, race, and ethnicity, and on the U.S. context (Jonsen et al., 2011; Peretz et al., 2015).
Diversity trainers and scholars also have much to learn from research on anti-harassment training. Research on sexual harassment training has been largely divorced from research on diversity training, despite the similarities between the two. Harassment training, too, has not shown positive effects and, in some cases can lead to negative effects. For example, laboratory studies suggest that men who score high on “likely harasser” and “gender role conflict” scales frequently have adverse reactions (Kearney et al., 2004; Robb & Doverspike, 2001; Tinkler et al., 2015). Thus, harassment training, like diversity training, may do more harm than good if it antagonizes the very men it hopes to reform. Research also shows, however, that bystander harassment training, which treats trainees as allies of victims rather than as offenders, has been effective in increasing managers’ willingness to intervene to stop harassment (Katz & Moore, 2013; Potter & Moynihan, 2011) and in increasing the share of women among managers (Dobbin & Kalev, 2019). Diversity trainers and researchers may have something to learn from sexual harassment bystander training. It is possible that framing diversity trainees as bystanders rather than as biased decision makers will be more effective in mobilizing them to action.
The lack of effectiveness of diversity training is not due to lack of effort. But these efforts are too often doomed to failure, because the cards are stacked against training people out of their hardwired biases. In the words of Bromley and Powell (2012), diversity training is a clear case of means-ends decoupling. The outcomes of this decoupling are, first, that diversity training and research on diversity training are in a state of perpetual reform, manufacturing variations in training and numerous studies and meta-studies that lead to similar conclusions; and second, that diversity management resources are not invested where they could be most effective.
The research shared in this article suggests that to effectively reduce bias and increase diversity at work, corporations need to focus on increasing opportunities for collaboration at work, not merely on educating biased individuals. This is not because individuals are no longer biased. Many still are (Smith & Hunt, 2021). It is because biases are affected by the context in which these individuals operate, and they can be effectively ameliorated when collaboration increases.
The modern division of labor has quickly become a gendered and racialized division of labor as well, with women, people of color, and other underrepresented groups often relegated to the most marginalized, least rewarding tasks. Management models that undermine task segregation and increase the collaboration of workers from different functions were adopted by corporations since the 1980s for the purpose of increasing productivity. These innovations grew in popularity and improved productivity. They also had the unintended consequence of increasing peer-like contact between men and women, whites and nonwhites, weakening stereotypes and group boundaries. These led to improved career opportunities for workers from underrepresented groups and to higher managerial diversity.
This article thus has pointed to popular programs that ameliorate the segregated division of labor as an additional, far more effective way to reduce decision makers’ bias. There is certainly a need to know more about how to maximize the diversity benefits of cross-functional teams and cross-training. Furthermore, research has shown that diversity innovations that change career systems, such as targeted recruitment programs, formal mentoring programs, universal skill and management training, and formal work-life supports are effective in reducing structural bias increasing diversity (see Dobbin & Kalev, 2022; Kalev et al., 2006). But there is little research on how to maximize these benefits. There is little knowledge, for example, about what are effective clusters of programs, what implementation pitfalls exist for each of them, and what are relevant adjustments for specific industries or occupations. Compared to the hundreds of studies on diversity training, little attention has been paid to all other diversity initiatives, including those that show strong, longer-lasting effects.
References
- Adler, P. S. (1992). The learning bureaucracy: New United Motor Manufacturing Inc. In B. M. Staw & L. L. Cummings (Eds.), Research in organizational behavior (pp. 94–111). JAI Press.
- Ameri, M., Amoroso, L. M., & Kurtzberg, T. R. (2021). Advancing diversity training. Rutgers Business Review, 6(2), 154–160.
- Avery, D. R., & Thomas, K. M. (2004). Blending content and contact: The role of diversity curriculum and campus heterogeneity in fostering diversity management competency. Academy of Management Learning & Education, 3(4), 380–396.
- Bezrukova, K., Spell, C. S., Perry, J. L., & Jehn, K. A. (2016). A meta-analytical integration of over 40 years of research on diversity training evaluation. Psychological Bulletin, 142(11), 1227–1274.
- Blair, I. V. (2002). The malleability of automatic stereotypes and prejudice. Personality and Social Psychology Review, 6(3), 242–261.
- Brady, L. M., Kaiser, C. R., Major, B., & Kirby, T. A. (2015). It’s fair for us: Diversity structures cause women to legitimize discrimination. Journal of Experimental Social Psychology, 57, 100–110.
- Brehm, S. S., & Brehm J. W. (1981). Psychological reactance: A theory of freedom and control. Academic Press.
- Bromley, P., & Powell, W. W. (2012). From smoke and mirrors to walking the talk: Decoupling in the contemporary world. The Academy of Management Annals, 6(1), 483–530.
- Bromwich, J. E. (2019, June 4). Sephora will shut down for an hour of diversity training tomorrow. The New York Times.
- Brooks, R. R. W., & Purdue-Vaughns, V. (2007). Supermodular architecture of inclusion. Harvard Journal of Law & Gender, 30(2), 379–388.
- Burt, R. S. (1998). The gender of social capital. Rationality and Society, 10(1), 5–46.
- Campion, M. A., Cheraskin, L., & Stevens, M. J. (1994). *Career-related antecedents and outcomes of job rotation. Academy of Management Journal, 37(6), 1518–1542.
- Canduela, J., Dutton, M., Johnson, S., Lindsay, C., McQuaid, R. W., & Raeside, R. (2012). Ageing, skills and participation in work-related training in Britain: Assessing the position of older workers. Work, Employment and Society, 26(1), 42–60.
- Castilla, E. J. (2012). Gender, race, and the new (merit-based) employment relationship. Industrial Relations, 51, 528–562.
- Castilla, E. J., & Benard, S. (2010). The paradox of meritocracy in organizations. Administrative Science Quarterly, 55(4), 543–676.
- Chang, E. H., Milkman, K. L., Gromet, D. M., Rebele, R. W., Massey, C., Duckworth, A. L., & Grant, A. M. (2019). The mixed effects of online diversity training. Proceedings of the National Academy of Sciences of the United States of America, 116(16), 7778–7783.
- Chen, M., & Bargh, J. A. (1997). Nonconscious behavioral confirmation processes: The self-fulfilling nature of automatically activated stereotypes. Journal of Experimental Social Psychology, 33, 541–560.
- Correll, S. J., Benard, S., & Paik, I. (2007). Getting a job: Is there a motherhood penalty? American Journal of Sociology, 112(5), 1297–1339.
- Crespo, G. (2018). Doctor says she was racially profiled while trying to help fellow passenger. CNN.
- Cuddy, A. J. C., Fiske, S. T., Kwan, V. S. Y., Glick, P., Demoulin, S., Leyens, J. P., Bond, M. H., Croizet, J. C., Ellemers, N., Sleebos, E., Htun, T. T., Kim, H. J., Maio, G., Perry, J., Petkova, K., Todorov, V., Rodríguez-Bailón, R., Morales, E., Moya, M., . . . Ziegler, R. (2009). Stereotype content model across cultures: Towards universal similarities and some differences. British Journal of Social Psychology, 48, 1–33.
- Daday, G., & Burris, B. (2002, August). The effects of teaming-structures on race, ethnicity, and gender differences in a high-tech corporation: A case study [Paper presentation]. Annual Meeting of the American Sociological Association, Chicago, USA.
- Dasgupta, N. (2004). Implicit ingroup favoritism, outgroup favoritism, and their behavioral manifestations. Social Justice Research, 17(2), 143–169.
- Dencker, J. C. (2008). Corporate restructuring and sex differences in managerial promotion. American Sociological Review, 73(3), 455–476.
- Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56(1), 5–18.
- Devine, P. G., & Ash, T. L. (2022). Diversity training goals, limitations, and promise: A review of the multidisciplinary literature. Annual Review of Psychology, 73, 403–429.
- Devine, P. G., Forscher, P. S., Austin, A. J., & Cox, W. T. L. (2012). Long-term reduction in implicit race bias: A prejudice habit-breaking intervention. Journal of Experimental Social Psychology, 48(6), 1267–1278.
- Devine, P. G., Plant, E. A., Amodio, D. M., Harmon-Jones, E., & Vance, S. L. (2002). The regulation of explicit and implicit race bias: The role of motivations to respond without prejudice. Journal of Personality and Social Psychology, 82(5), 835–848.
- Ditomaso, N., Parks-Yancy, R., & Post, C. (2003). Structure, relationships, and community responsibility. Management Communication Quarterly, 17(1), 143–150.
- Dobbin, F. (2009). Inventing equal opportunity. Princeton University Press.
- Dobbin, F., & Kalev, A. (2019). The promise and peril of sexual harassment programs. Proceedings of the National Academy of Sciences of the United States of America, 116(25), 12255–12260.
- Dobbin, F., & Kalev, A. (2022). Getting to diversity—What works and what doesn’t. Harvard University Press.
- Dover, T. L., Major, B., & Kaiser, C. R. (2016). Members of high-status groups are threatened by pro-diversity organizational messages. Journal of Experimental Social Psychology, 62, 58–67.
- Elvira, M., & Town, R. (2001). The effects of race and worker productivity on performance evaluations. Industrial Relations, 40(4), 571–590.
- Fiske, S. T. (1998). Stereotyping, prejudice and discrimination. In S. T. Fiske & D. T. Gilbert (Eds.), Handbook of social psychology (Vol. 2, pp. 357–411). McGraw-Hill.
- FitzGerald, C., Martin, A., Berner, D., & Hurst, S. (2019). Interventions designed to reduce implicit prejudices and implicit stereotypes in real world contexts: A systematic review. BMC Psychology, 7(1), 29.
- Forscher, P. S., Lai, C. K., Axt, J. R., Ebersole, C. R., Herman, M., Devine, P. G., & Nosek, B. A. (2019). A meta-analysis of procedures to change implicit measures. Journal of Personality and Social Psychology, 117(3), 522–559.
- Galinsky, A. D., & Moskowitz, G. B. (2000). Perspective-taking: Decreasing stereotype expression, stereotype accessibility, and in-group favoritism. Journal of Personality and Social Psychology, 78(4), 708–724.
- Goffman, E. (1967). Interaction ritual: Essays in face to face behavior. AldineTransaction.
- Gouldner, A. W. (1954). Patterns of industrial democracy: A case study of modern factory administration. Free Press.
- Green, T. K. (2016). Discrimination laundering: The rise of organizational innocence and the crisis of equal opportunity law. Cambridge University Press.
- Green, T. K., & Kalev, A. (2008). Discrimination-reducing measures at the relational level. Hastings Law Journal, 59(6), 1435–1461.
- Greene, R. L. (2008). Repetition and spacing effects. In J. Byrne (Ed.), Learning and memory: A comprehensive reference (pp. 65–78). Elsevier.
- Greenwald, A. G., & Lai, C. K. (2020). Implicit social cognition. Annual Review of Psychology, 71(1), 419–445.
- Hahn, A., & Gawronski, B. (2019). Facing one’s implicit biases: From awareness to acknowledgment. Journal of Personality and Social Psychology, 116(5), 769–794.
- Handel, M. J., & Levine, D. I. (2004). Editors’ introduction: The effects of new work practices on workers. Industrial Relations, 43(1), 1–43.
- Henry, H., Zacher, H., & Desmette, D. (2015). Reducing age bias and turnover intentions by enhancing intergenerational contact quality in the workplace: The role of opportunities for generativity and development. Work, Aging and Retirement, 1(3), 243–253.
- Hodson, R. (1996). Dignity in the workplace under participative management: Alienation and freedom revisited. American Sociological Review, 61(5), 719–738.
- Ibarra, H. (1992). Homophily and differential returns: Sex differences in network structure and access in an advertising firm. Administrative Science Quarterly, 37(3), 422–447.
- Iweins, C., Desmette, D., Yzerbyt, V., & Stinglhamber, F. (2013). Ageism at work: The impact of intergenerational contact and organizational multi-age perspective. European Journal of Work and Organizational Psychology, 22(3), 331–346.
- Jonsen, K., Maznevski, M. L., & Schneider, S. C. (2011). Special review article: Diversity and its not so diverse literature; an international perspective. International Journal of Cross Cultural Management, 11, 35–62.
- Kaiser, C. R., Major, B., Jurcevic, I., Dover, T. L., Brady, L. M., & Shapiro, J. R. (2013). Presumed fair: Ironic effects of organizational diversity structures. Journal of Personality and Social Psychology, 104(3), 504–519.
- Kalev, A. (2009). Cracking the glass cages? Restructuring and ascriptive inequality at work. American Journal of Sociology, 114(6), 1591–1643.
- Kalev, A. (2014). How you downsize is who you downsize. American Sociological Review, 79(1), 109–135.
- Kalev, A., & Dobbin, F. (2020). Companies need to think bigger than diversity training. Harvard Business Review, 2–4.
- Kalev, A., Dobbin, F., & Kelly, E. L. (2006). Best practices or best guesses? Assessing the efficacy of corporate affirmative action and diversity policies. American Sociological Review, 71(4), 589–617.
- Kalinoski, Z. T., Steele-Johnson, D., Peyton, E. J., Leas, K. A., Steinke, J., & Bowling, N. A. (2013). A meta-analytic evaluation of diversity training outcomes. Journal of Organizational Behavior, 34, 1076–1104.
- Kanter, R. M. (1977). Men and women of the corporation. Basic Books.
- Kaplan, D. M. (2006). Can diversity training discriminate? Backlash to lesbian, gay, and bisexual diversity initiatives. Employee Responsibilities and Rights Journal, 18, 61–72.
- Katz, J., & Moore, J. (2013). Bystander education training for campus sexual assault prevention: An initial meta-analysis. Violence and Victims, 28(6), 1054–1067.
- Kearney, L. K., Rochlen, A. B., & King, E. B. (2004). Male gender role conflict, sexual harassment tolerance, and the efficacy of a psychoeducative training program. Psychology of Men & Masculinity, 5(1), 72–82.
- Kellough, J. E., & Naff, K. C. (2004). Responding to a wake-up call: An examination of federal agency diversity management programs. Administration & Society, 36(1), 62–90.
- Kelly, E., & Dobbin, F. (1998). How affirmative action became diversity management. American Behavioral Scientist, 41(7), 960–984.
- Kim, K. W., Kalev, A., Dobbin, F., & Deutsch, G. (2021). Crisis and uncertainty: Did the great recession reduce the diversity of new faculty? Sociological Science, 8, 308–324.
- Knoke, D., & Ishio, Y. (1998). The gender gap in company job training. Work and Occupations, 25(2), 141–167.
- Konrad, A. M., & Linnehan, F. (1995). *Formalized HRM structures: Coordinating equal employment opportunity or concealing organizational practices? Academy of Management Journal, 38(3), 787–820.
- Krieger, L. H. (1995). The content of our categories: A cognitive bias approach to discrimination and equal employment opportunity. Stanford Law Review, 47(6), 1161–1248.
- Kulik, C. T., Perry, E. L., & Bourhis, A. C. (2000). Ironic evaluation processes: Effects of thought suppression on evaluations of older job applicants. Journal of Organizational Behavior, 21(6), 689–711.
- Kulik, C. T., & Roberson, L. (2008). Common goals and golden opportunities: Evaluations of diversity education in academic and organizational settings. Academy of Management Learning & Education, 7(3), 309–331.
- Lai, C. K., Marini, M., Lehr, S. A., Cerruti, C., Shin, J.-E. L., Joy-Gaba, J. A., Ho, A. K., Teachman, B. A., Wojcik, S. P., Koleva, S. P., Frazier, R. S., Heiphetz, L., Chen, E. E., Turner, R. N., Haidt, J., Kesebir, S., Hawkins, C. B., Schaefer, H. S., Rubichi, S., . . . Nosek, B. A. (2014). Reducing implicit racial preferences: I. A comparative investigation of 17 interventions. Journal of Experimental Psychology: General, 143(4), 1765–1785.
- Lai, C. K., Skinner, A. L., Cooley, E., Murrar, S., Brauer, M., Devos, T., Calanchini, J., Xiao, Y. J., Pedram, C., Marshburn, C. K., Simon, S., Blanchar, J. C., Joy-Gaba, J. A., Conway, J., Redford, L., Klein, R. A., Roussos, G., Schellhaas, F. M. H., Burns, M., . . . Nosek, B. A. (2016). Reducing implicit racial preferences: II. Intervention effectiveness across time. Journal of Experimental Psychology: General, 145(8), 1001–1016.
- Legault, L., Green-Demers, I., Grant, P., & Chung, J. (2007). On the self-regulation of implicit and explicit prejudice. Personality and Social Psychology Bulletin, 33(5), 732–749.
- Legault, L., Gutsell, J. N., & Inzlicht, M. (2011). Ironic effects of antiprejudice messages: How motivational interventions can reduce (but also increase) prejudice. Psychological Science, 22(12), 1472–1477.
- Litman, L., & Davachi, L. (2008). Distributed learning enhances relational memory consolidation. Learning & Memory, 15(9), 711–716.
- Macrae, C. N., Bodenhausen, G. V., Milne, A. B., & Jetten, J. (1994). Out of mind but back in sight: Stereotypes on the rebound. Journal of Personality and Social Psychology, 67(5), 808–817.
- McGuire, G. M. (2000). Gender, race, ethnicity, and networks. Work and Occupations, 27(4), 501–524.
- Meyerson, D. E., & Fletcher, J. K. (2000). A modest manifesto for shattering the glass ceiling. Harvard Business Review, 78(1), 126–136.
- Monteith, M. J., & Mark, A. Y. (2005). Changing one’s prejudiced ways: Awareness, affect, and self-regulation. European Review of Social Psychology, 16(1), 113–154.
- Naff, K. C., & Kellough, J. E. (2003). Ensuring employment equity: Are federal diversity programs making a difference? International Journal of Public Administration, 26(12), 1307–1336.
- Ollilainen, M., & Rothschild, J. (2001). Can self-managing teams be truly cross-functional? Gender barriers to a “new” division of labor. In S. P. Vallas (Ed.), The transformation of work (Vol. 10, pp. 141–164). Emerald Group.
- Osterman, P. (2022). How American adults obtain work skills: Results of a new national survey. Industrial and Labor Relations Review, 75(3), 578–607.
- Pager, D., Bonikowski, B., & Western, B. (2009). Discrimination in a low-wage labor market. American Sociological Review, 74(5), 777–799.
- Paluck, E. L., & Green, D. P. (2009). Prejudice reduction: What works? A review and assessment of research and practice. Annual Review of Psychology, 60(1), 339–367.
- Peretz, H., Levi, A., & Fried, Y. (2015). Organizational diversity programs across cultures: Effects on absenteeism, turnover, performance and innovation. The International Journal of Human Resource Management, 26(6), 875–903.
- Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90(5), 751–783.
- Piore, M. J., & Sabel, C. F. (1984). The second industrial divide—Possibilities for prosperity. Basic Books.
- Plant, E. A., & Devine, P. G. (1998). Internal and external motivation to respond without prejudice. Journal of Personality and Social Psychology, 75(3), 811–832.
- Plant, E. A., & Devine, P. G. (2001). Responses to other-imposed pro-Black pressure: Acceptance or backlash? Journal of Experimental Social Psychology, 37(6), 486–501.
- Plaut, V. C., Garnett, F. G., Buffardi, L. E., & Sanchez-Burks, J. (2011). “What about me?” Perceptions of exclusion and whites’ reactions to multiculturalism. Journal of Personality and Social Psychology, 101(2), 337–353.
- Plaut, V. C., Thomas, K. M., & Goren, M. J. (2009). Is multiculturalism or color blindness better for minorities? Psychological Science, 20(4), 444–446.
- Potter, S. J., & Moynihan, M. M. (2011). Bringing in the bystander in-person prevention program to a United States military installation: Results from a pilot study. Military Medicine, 176(8), 870–875.
- Reskin, B. F. (2000). The proximate causes of employment discrimination. Contemporary Sociology, 29(2), 319–328.
- Ridgeway, C. L. (1997). Interaction and the conservation of gender inequality: Considering employment. American Sociological Review, 62(2), 218–235.
- Ridgeway, C. L., & Smith-Lovin, L. (1999). The gender system and interaction. Annual Review of Sociology, 25(1), 191–216.
- Rivera, L. A. (2016). Pedigree: How elite students get elite jobs. Princeton University Press.
- Robb, L. A., & Doverspike, D. (2001). Self-reported proclivity to harass as a moderator of the effectiveness of sexual harassment-prevention training. Psychological Reports, 88(1), 85–88.
- Roberge, M. E., Lewicki, R. J., Hietapelto, A., & Abdyldaeva, A. (2011). From theory to practice: Recommending supportive diversity practices. Journal of Diversity Management, 6(2), 1–20.
- Roberson, L., Kulik, C. T., & Tan, R. Y. (2013). Effective diversity training. In Q. M. Roberson (Ed.), The Oxford handbook of diversity and work (Vol. 1, January Issue, pp. 341–365). Oxford University Press.
- Roscigno, V. J., Mong, S., Byron, R., & Griff, T. (2007). Age discrimination, social closure and employment. Social Forces, 86(1), 313–334.
- Rubenzer, K. N., & Pierce, J. T. (2023). Experiential diversity training and science learning for college students alongside peers with intellectual and developmental disabilities. bioRxiv.
- Rudman, L. A., Greenwald, A. G., & McGhee, D. E. (2001). Implicit self-concept and evaluative implicit gender stereotypes: Self and ingroup share desirable traits. Personality and Social Psychology Bulletin, 27(9), 1164–1178.
- Rynes, S., & Rosen, B. (1995). A field survey of factors affecting the adoption and perceived success of diversity training. Personnel Psychology, 48(2), 247–270.
- Schlanger, M., & Kim, P. T. (2014). The equal employment opportunity commission and structural reform of the American workplace. Washington University Law Review, 6, 1519–1590.
- Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. (1961). Intergroup cooperation and conflict: The Robbers Cave experiment. University of Oklahoma Book Exchange.
- Smith, R. A., & Hunt, M. O. (2021). White supervisor and subordinate beliefs about Black/White inequality: Implications for understanding and reducing workplace racial disparities. Social Problems, 68(3), 720–739.
- Smith, V. (1997). New forms of work organization. Annual Review of Sociology, 23, 315–339.
- Smith-Doerr, L. (2004). Flexibility and fairness: Effects of the network form of organization on gender equity in life science careers. Sociological Perspectives, 47(1), 25–54.
- Stainback, K., & Tomaskovic-Devey, D. (2012). Documenting desegregation: Racial and gender segregation in private sector employment since the Civil Rights Act. Russell Sage Foundation.
- Sturm, S. (2001). Second generation employment discrimination: A structural approach. Columbia Law Review, 101(3), 458.
- Tajfel, H., Turner, J. C., Austin, W. G., & Worchel, S. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Brooks/Cole.
- Taplin, I. M. (1995). Flexible production, rigid jobs: Lessons from the clothing industry. Work and Occupations, 22(4), 412–438.
- Tinkler, J., Gremillion, S., & Arthurs, K. (2015). Perceptions of legitimacy: The sex of the legal messenger and reactions to sexual harassment training. Law & Social Inquiry, 40(01), 152–174.
- Truxillo, D. M., Finkelstein, L. M., Pytlovany, A. C., & Jade, S. J. (2015). Age discrimination at work: A review of the research and recommendations for the future. The Oxford Handbook of Workplace Discrimination, 1, 129–142.
- Tsui, A. S., & O’Reilly, C. A. (1989). Beyond simple demographic effects: The importance of relational demography in superior-subordinate dyads. Academy of Management Journal, 32(2), 402–423.
- Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Basil Blackwell.
- U.S. Department of Labor. (n.d.). History of Executive Order 11246. Office of Federal Contract Compliance Programs
- Vallas, S. P. (2003a). Rediscovering the color line within work organizations. Work and Occupations, 30(4), 379–400.
- Vallas, S. P. (2003b). Why teamwork fails: Obstacles to workplace change in four manufacturing plants. American Sociological Review, 68(2), 223–250.
- Ward, L. M., & Grower, P. (2020). Media and the development of gender role stereotypes. Annual Review of Developmental Psychology, 2(1), 177–199.
- Welch, K. (2007). Black criminal stereotypes and racial profiling. Journal of Contemporary Criminal Justice, 23(3), 276–288.
- Williams, R. M., Jr. (1947). The reduction of intergroup tensions: A survey of research on problems of ethnic, racial, and religious group relations. Social Science Research Council Bulletin, 57(xi), 153.
- Wolsko, C., Park, B., Judd, C. M., & Wittenbrink, B. (2000). Framing interethnic ideology: Effects of multicultural and color-blind perspectives on judgments of groups and individuals. Journal of Personality and Social Psychology, 78(4), 635–654.
- Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35(2), 151–175.