Diversity is a hot topic for executives. Inspired by research showing that demographically diverse teams and organizations can improve performance, many managers are eager to capitalize on the benefits of diversity. One survey found that 72% of companies offer diversity initiatives. Yet, progress in the corporate realm has been slower than expected. And in the wake of several high-profile gender discrimination lawsuits and ongoing debates about racial profiling, it is clear that gender and race continue to bias how we evaluate people in the workplace and beyond. What is less obvious is what can be done about it.
Managers frequently explain a lack of diversity in organizations or industries as a problem of the pipeline, insisting there are simply not enough qualified or interested women or minorities to fill available positions, especially high-level ones. However, biased hiring decisions help keep diversity artificially low. Some of employers’ preferred recruiting sources, such as prestigious universities and personal networks, stack the deck against diversity. But crucially, the way we evaluate those candidates who do end up in the pipeline varies systematically based on applicants’ demographic characteristics.
Economists, psychologists, and sociologists have shown that whites and males are significantly more likely to receive invitations to interview than are racial minorities and women with identical résumés (the same goes for nonmothers v. mothers and, in certain cities, straight v. openly gay men). Once in the interview room, the same behaviors that result in job offers for majority groups can lead to rejection letters for under-represented ones. Even in the startup world, men are significantly more likely to be funded by investors than are women, even when they pitch the exact same idea.
What is going on? Is talk of diversity just talk? Not necessarily. While overt discrimination still happens, in the 21st century, discrimination has largely moved underground. It often operates through stereotypes: mental models deeply ingrained in our culture about what particular groups are like or should be like—who is smarter, kinder, funnier, more creative, better with clients, more technologically savvy, or a more effective leader. Stereotypes systematically distort how we evaluate job applicants, whether or not we believe these associations to be true, or are even conscious of them.
Because stereotypes often operate in the background of human cognition, they are difficult to combat. Complicating matters, one of the most popular tools managers use to overcome them—diversity training—has limited effectiveness. When it comes to combatting bias, awareness is simply not sufficient. So what can be done? Social science research points to two vital clues: structure and accountability.
In my own research on corporate hiring practices, I have found that managers often insist that hiring is “an art, not a science” and resist attempts to formalize it. Likewise, unstructured interviews are the most common tools American managers use to make hiring decisions. However, open-ended interviews not only are notoriously poor predictors of future on-the-job performance but also set up ideal conditions for stereotypes—whether based on perceptions of competence or “fit”— to bias hiring decisions. Adopting standardized, data-driven questions and tests aimed at probing (and ideally, simulating) job-relevant behaviors and skills can help keep job interviews more closely tied to performance. But crucially, firms also need to establish systematic rubrics for judging the quality of responses so evaluation isn’t left to the eyes—and stereotypes—of the interviewer.
But structure shouldn’t end with job interviews. In many companies, candidates’ fates are sealed in the course of oral deliberations, where interviewers or hiring committee members gather to decide collectively who gets the offer. Few candidates are perfect, and research shows that we are especially likely to rely on stereotypes when candidate quality is ambiguous or decisions are close calls. Providing evaluators with formalized methods to compare candidates in post-interview decision meetings apart from interviewers’ “gut” feelings of who would make a good candidate (or which employee has the highest status or loudest voice in the room) can help increase both the quality and diversity of new hires.
In the most comprehensive study of the effectiveness of diversity practices to date, sociologists Alexandra Kalev, Frank Dobbin, and Erin Kelly found a second ingredient for successful diversity initiatives: accountability. This can take several forms. Psychological research shows that, although we all hold stereotypes, we are less likely to act on them when we can be personally identified with a decision. In hiring, something as simple as having people who review résumés or interview candidates write their names on candidate evaluation sheets before turning them in to hiring committees or HR can help induce a sense of personal responsibility. Likewise, providing committees with the demographic composition of the applicant pool and asking them to submit the demographic composition of the candidates they interviewed and hired can be a subtle, but effective nudge.
On an organizational level, having a person or group—such as a diversity manager, diversity council, or both—that is officially vested with increasing diversity and is held accountable for reaching diversity goals is one of the most effective practices. But crucially, these persons need to have sufficient authority and status within an organization to make decisions. In one company I observed, the firm had a diversity manager and several diversity recruiting initiatives, but when it came time to make final hiring decisions, the diversity manager was asked to leave the room. Diversity personnel are only as effective as the amount of real power they wield within an organization.
Finally, the most dramatic strategy is making diversity statistics public. Large law firms now do so annually. Clients and future employees now compare the quality of law firms based not only on revenues and salaries but also diversity. Recently, Google (GOOG) and Facebook (FB) have released their gender statistics. Time will tell how such external transparency—especially if it continues from year to year—will affect hiring processes in Silicon Valley.
When it comes to diversity initiatives, too often the types of programs heralded as best practices fail to produce real change. Focusing on implementing structured evaluations in résumé screening, job interviews, and hiring committee deliberations as well as increasing accountability provides data-driven pathways for combatting bias and increasing diversity.
Lauren Rivera is an associate professor of management and organizations at Kellogg School of Management at Northwestern University. She is also the author of Pedigree: How Elite Students Get Elite Jobs.