3 research-based ways to reduce unconscious gender bias in your organization

Insights from Prof. Jennifer Chatman, Prof. Don Moore, and Distinguished Teaching Fellow Kellie McElhaney

Research has long shown that companies with more women on their senior management teams see above-average financial returns.  But with a steady stream of news about workplace sexual harassment, the gender pay gap, and the small percentage of women at the top, it seems the business world has a long way to go in achieving gender equity.

Yet there are many research-proven strategies and practices that organizations can employ to reduce both overt and unconscious bias. For Women’s History Month, we highlight insights from three Berkeley Haas researchers on what organizations can do to reduce gender bias and create fair, equitable, and highly successful workplaces.

Make culture transparent

Political correctness has generally become a term of contempt, used to disparage a culture seen as stifling free expression—as when fired Google engineer James Damore railed against “PC-authoritarians.”

Berkeley Haas Prof. Jennifer Chatman_How to reduce gender bias
Prof. Jennifer Chatman

So Prof. Jennifer Chatman was surprised when she found, in a 2015 study, that mixed-gender groups who talked about rules of “political correctness” were significantly more creative than those who were simply told to be polite or sensitive.

“We were stunned by the results,” says Chatman, an expert on organizational culture who co-authored the study. “Those exposed to the PC norm had noticeably more divergent, novel, and interesting ideas than any of the other groups.”

Chatman’s study challenged both conventional wisdom and prior research on creativity, which presumes that imaginations flourish when constraints are removed. But in today’s diverse workplaces, uncertainty about how to act in mixed settings may actually inhibit creativity in other areas, the researchers concluded.

“Men are likely to worry about looking overbearing or sexist and fear social disapproval, while women concerned about whether their ideas will be accepted might self-censor,” Chatman says. “Spelling out the rules removes the ambiguity.”

For example, specifically stating that sexist language will not be tolerated reduces uncertainty, the researchers noted. In their experiments, the participants who discussed PC norms said they paid more attention to the words they used and tried to avoid offending other group members. Even with those self-restrictions, their groups came up with more ideas—and more innovative ideas—in a brainstorming session then those not given instructions.

“People can simultaneously be thoughtful, reasonable and respectful, while at the same time being wildly creative,” Chatman says.

To build a culture of respect, companies must clearly articulate their values and expected norms, and also be willing to enforce them, Chatman says. 

Make hiring transparent

When the London Symphony Orchestra noticed all their musicians were male, leaders realized gender bias had crept into their hiring. They started asking people to audition behind a screen and barefoot—so they wouldn’t unconsciously be influenced by the click-clack of high heels on the stage.

Berkeley Haas Prof. Don Moore
Prof. Don Moore

“Once they made that decision, they got better musicians and more female representation,” says Prof. Don Moore, an organizational behavior expert who studies decision making, overconfidence, and ethical choice.

Hiring and promoting the most capable people is critical to the success of any organization, yet many still rely on unstructured interviews that vary from person to person and introduce the probability of bias. “We’ve known for decades that interviews are terrible predictors of performance—it’s one of the best-established findings in organizational behavior research—yet somehow the word has not spread,” Moore says.

The simplest solution: Structure interviews so that all candidates are asked the same questions in the same order, and score the responses. Even better: use assessments of intelligence or other relevant skills to rate applicants in a quantifiable way, Moore says.

“It’s possible to get very well-designed tests that have been validated over many years and don’t include factors that we’d rather not consider, such as race, gender, or cultural background,” he says.

What about “fit”?

“I know that people believe they can assess ‘fit’, but when pushed as to what that means, most people have only the vaguest of responses: ‘Well, if I like them,'” Moore says. “But think about where your feelings of liking come from. Is it because the person looks like you? Talks like you? Went to the same schools? Without having a clear definition of ‘fit’ I worry that it opens the door to many prejudices that you don’t want influencing your decision.”

Even though it may be tough to create a perfectly non-biased hiring system, Moore says, “it’s a dream worth reaching for. It will make us better decision makers, it will make us better people, and it will pay dividends.”

Make pay practices transparent

On a national level, progress on closing the gender pay gap has been slow. Yet Gap Inc., proved it’s very much possible to achieve pay equity at the company level.

Adj. Assoc. Prof. Kellie McElhaney
Adj. Assoc. Prof. Kellie McElhaney

In 2014, Gap became the first Fortune 500 company to conduct a robust analysis and announce that it pays female and male employees equally for equal work, on average, across its global workforce of about 135,000. How did Gap do it? Distinguished Teaching Fellow Kellie McElhaney, founding director of the Berkeley Haas Center for Gender, Equity & Leadership, analyzed the company’s pay equity success in a case study published last July.

Gap has a strong history of strong women: It was co-founded by Doris and Don Fisher, who chipped in 50-50 to open their first store; it’s had senior women at the table from the start; and its workforce is 74 percent female. But that doesn’t mean gender parity was a fait accompli—in fact, female-dominated sectors such as retail, tend to have higher-than-average wage disparities since women tend to be clustered at the bottom, McElhaney notes.

Instead, the company’s equity success was the result of a concerted effort to engage and support women and help them rise through the ranks, through flexible work policies, and a culture that’s collaborative and inclusive. It’s also very much due to an intentional approach to minimize unconscious bias in pay, through transparency, accountability, and statistical analysis—practices other organizations can learn from.

Some best practices to make pay more transparent:

  • Create pay equity processes grounded in statistical analysis: Gap analyzed pay rates organization-wide and also by comparing comparable positions, grouping jobs based on responsibilities and experience required, and controlling for variables such as geography. Its analysis was validated externally.
  • Provide managers with data: The company gives managers pay data on their teams annually, including market ranges for each role. It also gives managers resources and discretion to correct pay differences.
  • Don’t require salary history: Women suffer a career-long wage loss when their salaries are anchored to prior inequities. Gap doesn’t require applicants to provide prior salary information during the initial hiring process (although it’s optional).
  • Put salary policies in writing: Gap publishes the company’s compensation practices on GapWeb, which all employees can access (they are not privy to specific pay ranges unless their managers share them).

In the case study, Keith White, Gap’s senior vice president of loss prevention, notes, “When operating eyes wide open, and not just treating people as if they are in a vacuum, [gender equality in pay] becomes a non-issue.”

Back