New approach to reducing gender inequality at work

At a time when many companies are feeling pressured to report on and improve gender inequality within the workforce, a Stanford sociologist is finding success with a new method for reducing the kind of bias that leads to these inequalities.

In a recently published paper in Gender & Society, Shelley Correll, director of the Clayman Institute for Gender Research, explains the method, which she and her team piloted and found successful while working with several technology companies over the last three years.

The method, which Correll dubs "a small wins model," focuses on educating managers and workers about bias, diagnosing where gender bias could enter their company's hiring, promotion or other evaluation practices and working with the company's leaders to develop tools that help measurably reduce bias and inequality.

"The change we can realistically expect to produce in any one instance will be small, imperfect and incomplete," Correll wrote. "Step by step, I believe that these small wins are the path to achieving our larger goal, which is the transformation of our organizations."

Methods of recruitment, evaluation

Over the past 30 years, research has shown that stereotypes about what men and women are capable of and how they should behave cause people to evaluate men and women differently, especially when the criteria for evaluation are ambiguous. This bias puts women at a disadvantage in workplaces, where they get hired and promoted at lower rates than men.

Women are usually subjected to a higher bar, requiring more evidence than men to be seen as qualified. In addition, if people judge a woman to be competent, they often judge her as less likeable, a correlation that is not true for men.

Because of this research, many companies in recent years have invested resources into reducing the kinds of bias that lead to gender imbalance. The two most common approaches have been unconscious bias training and formalizing of methods for hiring and evaluating employees so that they are based on achievement-related criteria.

Correll said that although the two approaches help companies with diversity, they are not complete solutions.

While unconscious bias training is important and helps improve short-term behavior during hiring and promotion decisions, the effects wear off over time and can be threatening to people in power, leading to more bias rather than less, Correll said.

Formal procedures for hiring and promotions also haven't been entirely successful. For example, Correll wrote, fire departments previously used height as a criterion for screening applicants. Despite height being an objective benchmark, its requirement screens out more women than men and a person's height isn't directly related to one's ability to perform the job of a firefighter.

"Bias training can backfire, increasing bias; and formal procedures can be misused by decision makers or, worse, have gender biases built into their design," Correll wrote. "In spite of these limitations, I argue that educating about stereotyping and bias and formalizing evaluation processes are two key building blocks crucial for producing sustainable change."

Small wins add up

Although neither approach alone has been entirely successful, Correll argues for combining the two in a way that also includes measuring small incremental changes within the organization.

In this model, a company would provide workers with bias training and experts would analyze the company's procedures to understand where bias might be creeping in. The experts would then work with managers of the company to develop better procedures or tools, implement them and evaluate what changes they produce.

The entire process focuses on creating objective performance checklists and other tools that eliminate ambiguity and the chance for bias from people using them.

"To create sustainable change, we need to shift the target of change from the individual decision maker to organizational processes," Correll wrote.

Correll and her team tried this model with several companies over the past three years as part of Stanford's Center for the Advancement of Women's Leadership. They found measurable improvements in gender inequalities at those companies.

For example, at a mid-size technology company in the Western U.S., Correll said her research showed that after putting in place a new employee scorecard, managers were less likely to discuss an employee's personality rather than his or her work. Before the new scorecard, comments during these discussions criticized 14 percent of women for being "too aggressive" and 8 percent of men for being "too soft." A year later, after rolling out the new scorecard, those figures dropped to zero percent and 1 percent, respectively.

"The small-wins approach gives people results and something small they can do daily," Correll said. "Those small wins start to add up. They are the building blocks to larger change."

Correll said she has been inspired by the number of managers and leaders dedicated to equality in the workplace whom she has met over the past three years. Although reducing bias and inequality completely will take a long time, Correll said she feels optimistic.

"It's encouraging to see that many people genuinely want to be more inclusive," Correll said. "These companies didn't have to open up to us researchers. And I can't stress enough how valuable working with them and evaluating their data has been."