chevron-down Created with Sketch Beta.
July 10, 2017

Can Bias Interrupters Succeed Where Diversity Efforts Have Stalled?

Cynthia L. Cooper

You might want to get to know this term: “bias interrupters.” In brief, they are new guidance—“toolkits”—aimed at improving diversity in the workplace. If the hopes for them bear out, bias interrupters may accomplish change in the legal profession where other efforts have disappointed.

The ABA Commission on Women in the Profession (CWP) is taking a leading role in pioneering this fresh, targeted approach to dealing with the damaging and stubborn effects of implicit bias in the legal workplace. But the idea for bias interrupters, first described by diversity expert and law professor Joan C. Williams, came about only after she took two left turns on the well-trod path of diversity programming.

Five years ago, Williams, a Distinguished Professor of Law at the University of California, Hastings College of the Law in San Francisco, and the founding director of the Center for WorkLife Law at Hastings, all but threw in the towel on organizational change on diversity. It was, she says, “progressing at the rate of an extremely lazy snail.”

Despite studies and trainings, the parity needle was barely moving in the legal field for women or people of color. The number of women equity partners in law firms, for example, grew by only 2 percent in a decade, reaching a mere 18 percent in 2016. “There would be a ‘women’s initiative’ or a ‘diversity initiative’ and ‘let’s have a budget and get some speakers in here.’ That kind of accepted toolbox has now been shown not to be very effective,” Williams says.

Implicit bias embedded in cultural messaging and expectations, it turns out, was more difficult to root out than the efforts to quell it. “If you have implicit bias constantly transmitted through your basic business systems, it doesn’t matter how many programs you put on,” Williams adds.

With outcomes so limited, Williams shifted her attention from organizational change to individual solutions. “I said, whatever, let’s just give women the tools to navigate through the bias that’s not changing.”

In 2014, Williams released a book, What Works for Women at Work: Four Patterns Working Women Need to Know, cowritten with her daughter, Rachel Dempsey, and identifying specific types of bias-tinged behavior—such as women having to work harder to get recognition or losing traction because of motherhood—and proffering advice on how to circumvent these career trip wires.

Shortly afterward, in probing bias in the technology field for Harvard Business Review, Williams hit upon a new approach and recovered her grounding in organizational change—but with a twist. She realized the patterns of bias her book had described could be systematically addressed from within an organization. She dubbed these actions “bias interrupters.”

“A bias interrupter is a tweak to one of the basic organizational systems—hiring, assignments, performance evaluations are examples,” Williams says. “It’s a tweak that, in an evidence-based way, interrupts the constant transmission of subtle forms of implicit bias.”

Williams created self-guiding steps for organizations to measure and identify stumbling blocks to bias-free decision making in their internal operations, adopt small custom-designed changes to the areas that pose problems, gauge the success of these interventions, and return to deploy adjustments where outcomes are lagging.

The idea of bias interrupters piqued the interest of Michele Coleman Mayes, elected in 2014 to a three-year term as chair of the CWP. “It’s like Ghostbusters,” she quips.

Mayes, too, was dissatisfied by the approach to diversity—it had devolved into “take a seminar and congratulate yourself,” she says. New tools would offer something tangible that could assist people committed to tackling diversity issues. “If we could do research around that, it might truly advance this area,” she thought.

Soon, the CWP, joined by the Minority Corporate Counsel Association in Washington, D.C., embarked on an initiative to develop bias interrupters for the legal field, partnering with Williams and researchers at the WorkLife Law Center.

Laying the Groundwork

The use of bias interrupters is keyed, first, to metrics and analysis. This applies at a law firm or in-house legal department, but also to the macro design of the bias interrupters tools.

To unearth exactly where hidden bias seeps into day-to-day legal practice, Williams and her team developed a survey, “Understanding In-house and Law Firm Lawyers’ Workplace Experiences Survey,” drawing on the extensive literature on implicit bias. Social psychology experiments show, for example, that lawyers will rate an identical memo differently if the name attached to it sounds white or African American. But how do these indicators and studies from other professions play out in the real world of law? “This is the first survey like this,” Mayes explains. (See sidebar at the end of this article.)

The survey of lawyers created questions designed to tease out four patterns of behavior that might hold back women and people of color (and, by extension, other groups who experience bias, such as people with disabilities, LGBTQ, “introverted” men, and people with working-class backgrounds). Williams identifies the patterns as “prove-it-again,” “tightrope,” “maternal wall,” and “tug of war.”

“Prove-it-again” is a pattern of implicit bias by which women and men of color must repeatedly prove themselves in order to be seen as equally capable to white males. In the survey of lawyers, “prove-it-again” bias proved to be extensive. A substantial majority of women lawyers of color (63 percent) and more than half of men of color and white women (54 percent each) reported “prove-it-again” demands, compared to only 28 percent of white men.

Women (67 percent of women of color and 54 percent of white women) and men of color (58 percent) said they are held to higher standards than peers. “I am clearly held to a different standard and treated in a different way, although others may not do this intentionally,” wrote one female survey participant in a comment. Only 35 percent of white men said they are held to higher standards than peers.

Women also experienced others getting credit for their ideas—half of women reported this, compared to less than a third of men, suggesting “that women consistently feel as though their contributions are being discounted,” according to survey analysis.

Tightrope” describes situations in which people find themselves measured against certain stereotypes and must walk a narrow line to balance cultural expectations. Women, for example, are expected to be helpful, sensitive, modest, and nice, while men are expected to be direct, assertive, competitive, and ambitious, the survey analysis explains. Many behaviors are implicated in “tightrope” bias—interruptions while speaking; negative responses to assertiveness, self-promotion, or expressions of anger; unequal divisions of office “housework” and “glamour” assignments; and limited opportunities for leadership.

In the survey of lawyers, women report that they are often interrupted, less able to show assertiveness, and more likely than white men to be engaged in administrative tasks. Women and men of color feel less free to express anger and are less likely than white men to feel that they are treated as leaders. “We asked whether they had an opportunity to be considered a leader. Men would go, ‘Oh yeah.’ Women would say, ‘No, not really,’” Mayes observes.

“Maternal wall” refers to bias triggered by parenting and the tradeoffs between career and family. Of the surveyed lawyers, 51 percent of women of color and 44 percent of white women reported that having children changed their colleagues’ perceptions of them. They said that they were denied access to high-quality assignments and promotions, paid less, and disadvantaged for working part time. Only 20 percent of men reported that having children affected their work.

One woman at a law firm commented, “I was passed over for partner because I had a child. The two male attorneys who were hired at the exact same time as me, who had comparable prior experience and the same job responsibilities, were made partner, but I was not. When I asked why, I was told it was because I had given birth to a child.”

“Tug of war” occurs when bias creates conflict within a group, for example, stoking competition if there is only one slot for a woman or a person of color. In the survey of lawyers, one-third of white women and 28 percent of women of color said that other women lawyers did not understand the level of commitment it takes to be successful. But, overall, “tug of war” proved less pervasive than other patterns of bias, and the substantial majority of women said that their female colleagues support each other.

Because these identified types of bias can quietly poke their way into basic systems in law firms or in-house legal departments, bias interrupters are intended to break the patterns with modest modifications. “Small steps, big change,” the toolkits announce.

Putting Bias Interrupters into Motion

The bias interrupters toolkit for law firms provides a process for addressing potential bias in hiring, assignments, performance evaluations, and compensation. (In-house departments align with separate toolkits for corporations that reflect their organizational structure and are found at biasinterrupters.org.)

To create “interrupters” for the legal field, the sponsoring organizations convened groups of in-house counsel and managing partners. They were invited to look at the research and brainstorm on effective adjustments to business practices.

Using the bias interrupters is a three-step process, and that’s important. The steps are (1) using metrics to pinpoint where bias exists, (2) introducing specific bias interrupters, and (3) measuring levels of bias again and reconfiguring as necessary.

“Businesses have never applied to this problem the techniques they apply to any problem they are deeply committed to solving,” Williams says. “When they are deeply committed to solving a problem, they go into the evidence, they analyze the problem, they develop the metrics, and they keep trying until they reach their goals.”

First, law firms or in-house legal departments should use metrics to pinpoint bias in areas where internal feedback leads it to suspect that problems of bias may exist. (The Center for WorkLife Law also can conduct an internal survey, if requested, according to Jamie Dolkas, the center’s director of women’s leadership.) If, for example, a skewed division seems to exist between “glamour work” and “office housework,” the bias interrupters tool can help. The tool for law firms is a four-page guidance with additional templates to facilitate the process.

In explaining the challenge, the bias interrupter on assignments states that “Misallocation of glamour work and the office housework is a key reason why leadership across the legal profession is still male-dominated.”

Williams notes that bias in assignments was “reported robustly” in the lawyer survey. More than 80 percent of white male lawyers report some access to desirable assignments, but only 53 to 63 percent of women or men of color report the same access. On office housework (planning parties, ordering lunch, scheduling, providing emotional support), only 20 to 26 percent of men report that they play such roles more often than peers, compared to 43 to 50 percent of women who do.

The assignments toolkit sets out a process for collecting firm-centric data to create a clear picture of how the management of assignments plays out. One template asks the lawyers to self-evaluate how much time they spend on “housework” tasks, while another creates a guide for defining what constitutes “glamour work.” Worksheets then dig into the higher-profile work and ask managers to describe what opportunities they offer—access to business development resources or gaining subject matter expertise, for example—and to identify the recipients of these assignments.

“Then you analyze the metrics—who’s doing the glamour work, who’s doing the office housework, who’s doing the low-profile work. And you analyze the metrics by individual supervising attorney to see if there are particular people who are presenting a problem,” Williams explains. “It is very, very focused on how bias plays out in very fine-grained ways in today’s workplace. We hope, with surgical precision, to interrupt it.”

Second, if patterns of bias emerge, a number of open-source bias interrupters tools are suggested on a pick-and-choose basis to spur organizational change. For example, interrupters suggest assigning “office housework” in rotation or giving it to administrative assistants instead of lawyers. One bias interrupter recommends ending practices of seeking volunteers for administrative tasks to ease the “tightrope” expectations of women that they should step forward. To ensure a more equitable spread of “glamour work,” bias interrupters suggest, among other ideas, formalizing the pool of available talent and individual consultations with particular supervisors.

The bias interrupters’ process does not end there. The third step is to measure levels of bias again and “return to your metrics,” Williams says, to determine if the interventions have produced any change and to continue to further tweaks if they have not.

“We basically set up an incredibly detailed process that you just have to walk through,” Williams says. “This is an attempt to invent a system, whereby we can say to organizations, ‘If you ain’t keeping score, you’re only practicing.’ Because this is what people do if they are serious.”

Bias interrupters in hiring, performance evaluations, and assignments follow the same process of measuring the contours of the problem in a specific way and applying targeted tactics to cut through bias.

If bias affects hiring, candidate interviewers are asked to read a two-pager that lists 19 common types of bias that slip into the process. One warns about the pattern of “He’ll go far,” “She’s not ready,” and points out that “prove-it-again” bias results in white males being judged on their potential when others are judged on what they have accomplished. Some of the nearly two dozen bias interrupters on hiring are familiar—expanding the pool of applicants or tapping diverse networks, while others are more far-reaching—limiting referral hiring, conducting blind screening of resumes, or adding a bias monitor to hiring teams.

“Some of them are easy; some of them will be harder,” Williams says. “That’s why we assume that people will treat these as a menu and do the easier ones first and see if that’s enough.”

In the performance evaluation area, a bias interrupter helps supervisors double-check their assumptions with a reminder that assertiveness is often judged as a positive for men and a negative for women. One tweak for evaluations proposes separating personality from skill sets in performance reviews; another suggests that to counter a gender divide on self-promotion, the firm sets up monthly e-mails to share successes across the board.

The partner compensation toolkit sets out seven important metrics to review, such as who gets “pitch credit” and whether lateral partners are paid differently. Among the 13 bias interrupters on partner compensation is a suggestion to create a low-risk way to settle disputes over credit. Another proposes giving credit for nonbillable work that is vital to the firm and addressing parental leave by annualizing income based on the average of months the lawyer worked.

Time Is Ripe

The concept of bias interrupters is still awaiting real-world responses. “We’re trying to create a new system for organizations that don’t just want to engage in what’s called ‘symbolic compliance’ to diversity,” Williams says.

Mayes is hopeful, believing that the tough political climate is leading more people to step up their own commitments to change. “This is something you can do,” she says. “If you can label bias and talk about it in a concrete way, you have more impetus to address it.

“There is no one-stop shopping that solves everything, but it allows you to see if certain bias interrupters cause a different dynamic in the workplace,” she continues. “It’s not just ‘admiring’ that all of us are inherently wired with bias. Why I remain hopeful is there are concrete steps that you can take to fix the problem.”

========================================

The Workplace Experiences Survey for Bias Interrupters in the Legal Field

Research for the development of bias interrupters for the legal field was gathered through a survey conducted between April and June of 2016. The “Understanding In-house and Law Firm Lawyers’ Workplace Experiences Survey” elicited responses from 2,857 lawyers, of which 63 percent were women and 28 percent people of color. Approximately three-quarters of the respondents work at law firms, while the rest are employed at in-house legal departments.

Drawing on social psychology studies to measure whether lawyers experience patterns of gender or racial bias, the survey asks respondents to rate on a six-point scale their reactions to 52 statements about hiring, assignments, business development, performance evaluations, compensation, and support or mentorship.

Sample statements say:

  • “I have to prove myself over and over again to get the same level of respect and recognition as my colleagues.”
  • “I am satisfied with the extent to which I have access to high-profile assignments.”
  • “I am seldom interrupted at meetings.”
  • “As compared to my colleagues with comparable seniority, experience and ethic, I get paid less.”
  • “Taking family leave would be a detriment to my career.”

Respondents also answered 25 demographic questions on race, age, workplace type, organization size, and geographic location, and they were able to add comments. “This is the first comprehensive attempt to measure whether lawyers report having experienced the kinds of bias documented in experimental studies,” as noted in “Racial and Gender Bias in the Profession,” a report on the outcomes.

Cynthia L. Cooper

Cynthia L. Cooper is an independent journalist in New York City with a background as a practicing lawyer.