BIAS INTERRUPTERS

Tools for Organizations

Incremental steps that improve diversity in your organization can yield large gains. Diverse work groups perform better and are more committed, innovative and loyal. Gender diverse workgroups have better collective intelligence, which improves performance by the group and its members, leading to better financial performance. Racially diverse workgroups consider a broader range of alternatives, make better decisions, and are better at solving problems. Bias, if unchecked, affects many different groups: modest or introverted men, LGBT+, individuals with disabilities, class migrants (professionals from non-professional backgrounds), women, and people of color. We’ve distilled the huge literature on bias into simple steps that help you and your company perform better.

Bias interrupters are small adjustments to your existing systems. They should not require you to abandon systems currently in place.

THE CHALLENGE

A recent study of performance evaluations in tech found that 66% of women’s performance reviews contained negative personality criticism (“You come off as abrasive”) whereas only 1% of men’s reviews did.(1) We know now that workplaces that view themselves as being highly meritocratic often are, in fact, more biased than other organizations(2)  and that the usual responses—one-shot diversity trainings, mentoring and networking programs—typically don’t work.(3) 

THE SOLUTION: A 3-Step Approach

Bias interrupters are tweaks to basic business systems that can yield large gains for your business, using a 3-step process.

1. Use Metrics

Data and metrics help you spot problems—and assess the effectiveness of the measures you’ve taken. Businesses use metrics to help them achieve any strategic goal. Key metrics:

  • Do your performance evaluations show consistently higher ratings for majority men than for women, people of color, or other relevant groups?
  • Do women’s ratings fall after they have children? Do employees’ ratings fall after they take parental leave or adopt flexible work arrangements?
  • Do the same performance ratings result in different promotion or compensation rates for different groups?

Keep metrics by: 1) individual supervisor; 2) department; and 3) the organization as a whole.

2. Implement Bias Interrupters

All bias interrupters should apply both to written evaluations and in meetings, where relevant. Because every organization is different, not all interrupters will be relevant. Consider this a menu.

To understand the research and rationale behind the suggested bias interrupters, read our Identifying Bias in Performance Evaluations Worksheet (2-page version or with citations) which summarizes hundreds of studies.

  • Empower people involved in the evaluation process to spot and interrupt bias by reading our Identifying Bias in Performance Evaluations Worksheet. Read and distribute the Worksheet to help you understand the rationale behind the steps suggested below.
  • Appoint Bias Interrupters.
    Have team members or HR business partners who have been trained to spot bias involved at every step of the evaluation process.
  • Begin with clear and specific performance criteria directly related to job requirements.
    Try: “He is able to write an effective summary judgment motion under strict deadlines,” instead of: “He writes well.”
  • Require evidence from the evaluation period that justifies the rating.
    Try: “In March, she gave X presentation in front of Y client on Z project, answered his questions effectively, and was successful in making the sale,” instead of: “She’s quick on her feet.”
  • Consider performance and potential separately for each candidate.
    Performance and potential should be appraised separately, given the tendency for majority men to be judged on potential; others on performance.
  • Separate personality issues from skill sets for each candidate.
    Personal style should be appraised separately from skills, because a narrower range of behavior often is accepted from women and people of color. For example, women may be labeled “difficult” for doing things that are accepted in majority men.
  • Level the playing field by ensuring everyone knows how to promote themselves effectively and sending the message they are expected to do so. Distribute our Writing an Effective Self-Evaluation Worksheet, which can help.
  • Offer alternatives to self-promotion.
    Encourage or require managers to set up more formal systems for sharing successes, such as a monthly email that lists employees’ accomplishments.
  • Provide a bounceback
    Managers whose performance evaluations show persistent bias should receive a bounceback (i.e. someone should talk through the evidence with them).
    What’s a bounceback? An example: in one organization, when a supervisor’s ratings of an underrepresented group deviate dramatically from the mean, the evaluations are returned to the supervisor with the message: either you have an undiagnosed performance problem that requires a Performance Improvement Plan (PIP), or you need to take another look at your evaluations as a group. The organization found that a few people were put on PIPs– but that over time supervisors’ ratings of underrepresented groups converged with those of majority men. The organization that used this found that all groups found performance evaluations equally fair. 
  • Have Bias Interrupters play an active role in calibration meetings.
    In many organizations, managers meet to produce a target distribution of ratings or cross-calibrate rankings. Have managers read our Identifying Bias in Performance Evaluations Worksheet before they meet. Have a trained Bias Interrupter in the room.
  • Don’t eliminate your performance appraisal system.
    Eliminating formal performance evaluation systems and replacing them with feedback-on-the-fly creates conditions for bias to flourish.

3. Repeat as Needed

  • Return to your key metrics. Did the bias interrupters produce any change?
  • If you don’t see change, you may need to implement a stronger bias interrupter, or you may be targeting the wrong place in the performance evaluation process.
  • Use an iterative process until your metrics improve.

RESOURCES

Footnotes

1. Snyder, 2014
2. Castilla, 2015
3. Kalev, Dobbin, & Kelly, 2006For the full citations, see our bibliography.

THE CHALLENGE

When comparing identical resumes, “Jamal” needed eight additional years of experiences to be considered as qualified as “Greg,” and “Jennifer” was offered $4,000 less in starting salary than “John.” (1) We know now that workplaces that view themselves as being highly meritocratic often are, in fact, more biased than other organizations (2) and that the usual responses—one-shot diversity trainings, mentoring and networking programs—typically don’t work. (3)

THE SOLUTION: A 3-Step Approach

Bias interrupters are tweaks to basic business systems that can yield large gains for your business, using a 3-step process:

1. Use Metrics

Businesses use metrics to assess whether they have progressed towards any strategic goal. Metrics can help you pinpoint where bias exists, and assess the effectiveness of the measures you’ve taken. (Whether metrics are made public will vary from company to company, and from metric to metric.)

  • Track the demography of the candidate pool through the entire hiring process, from initial contact, to resume review, to interviews, to hiring. Break down the demography by under-represented groups: women, people of color, people with disabilities, veterans, members of the LGBT community, etc.
  • Track whether hiring qualifications are waived more often for people from certain groups than other groups.

Keep metrics by: 1) individual supervisor; 2) department; and 3) the organization as a whole.

2. Implement Bias Interrupters

All bias interrupters should apply both to written evaluations and in meetings, where relevant. Because every organization is different, not all interrupters will be relevant. Consider this a menu. To understand the research and rationale behind the suggested bias interrupters, read our Identifying Bias in Hiring Worksheet (2-page version or with citations) which summarizes hundreds of studies.

A. Empower and Appoint
  • Empower people involved in the hiring process to spot and interrupt bias by using our Identifying Bias in Hiring Worksheet. Read and distribute.
  • Appoint Bias Interrupters—HR professionals or team members trained to spot bias, and involve them at every step of the hiring process.
B. Assembling a Diverse Pool
  • Limit referral hiring (“friends of friends”) If your existing organization is not diverse, hiring from your current employees’ social networks will replicate the lack of diversity.
  • Tap diverse networks Reach out to diverse candidates where they are. Identify job fairs, affinity networks, conferences and training programs that are aimed at women and people of color in your field and send recruiters.
  • Getting the word out If diverse candidates are not applying to your jobs, get the word out that your company is a great place to work for women and people of color. One company offers public talks by women at their company and writes blog posts, white papers, and social media articles highlighting the women who work there.
  • Change the wording of your job postings Using words like “leader” and “competitive” in your ads will tend to reduce the number of women who apply.(4) Explicitly stating that the salary is negotiable can reduce the gender gap in applicants.(5)
  • Insist on a diverse pool If you use a headhunter, tell him or her you expect a diverse pool, not just one or two diverse candidates. If the initial pool is largely homogenous, it is statistically unlikely that you will hire a diverse candidate. The odds of hiring a woman were 79 times greater if there were at least two women in the finalist pool; the odds of hiring a person of color were 194 times greater.(6)
C. Resume Review
  • Pre-commit to what’s important—and require accountability Pre-commit in writing to what qualifications are important, both in entry-level and in lateral hiring. When qualifications are waived for a specific candidate, require an explanation of why they are no longer important—and keep track to see for whom requirements are waived. (7)
  • Ensure resumes are graded on the same scale Establish clear grading rubrics and ensure that everyone grades on the same scale. Consider having each resume reviewed by two different managers and averaging the score.
  • Avoid inferring family obligations A candidate’s family situation should have nothing to do with your job search. Avoid inferring family obligations from a candidate’s resume and don’t count “gaps in a resume” as an automatic negative. Give candidates an opportunity to explain gaps by asking about them directly during the interview stage.
  • Consider candidates from multi-tier schools Don’t limit your search to candidates from Ivy League and top-tier schools. This favors majority candidates from elite backgrounds and hurts people of color and class migrants (professionals from blue-collar backgrounds).(8) Studies show that top students from lower ranked schools are often similarly successful.(9)
  • Try using “blind auditions” (unless you’re actively recruiting diverse candidates) If women and candidates of color are dropping out of the pool at the resume review stage, consider removing demographic info from resumes before review. This way, candidates can be evaluated based solely on their qualifications. Downside: if you’re actively trying to recruit women and candidates of color, this strategy will not apply.
D. Interviews
  • Use structured interviews Ask the same list of questions to every person who is interviewed. Ask questions that are directly relevant to the job the candidate is applying for.(10)
  • Ask performance-based questions Performance-based questions, or behavioral interview questions, (“Tell me about a time you had too many things to do and had to prioritize”) are a strong predictor of how successful a candidate will be at the job.(11)
  • Administer skills-based screening If applicable, ask candidates to take a skills-based assessment (for example, if part of the job is analyzing data sets and making recommendations, ask the candidate to do that.)
  • Develop a consistent rating scale and discount outliers Candidate’s answers (or skills-based assessments) should be rated on a consistent scale and backed up by evidence. Average the scores granted on each relevant criterion and discount outliers.(12)
  • If “culture fit” is a criterion for hiring, provide a specific definition Does someone fit? Culture fit can be important but when it’s misused, it can disadvantage people of color, class migrants, and women.(13) Culture fit should not mean the “lunch test” (who you would like to have lunch with). Questions about sports may feel exclusionary to women, and questions about hobbies may make class migrants feel less at home if they did not grow up, for example, playing golf.
  • “Gaps in a resume” should not be automatic disqualification Mothers are 79% less likely to be hired than an identical candidate without children.(14) Train people not to make inferences about whether someone is committed to their job due to parental status.
  • Provide candidates and interviewers with a handout detailing expectations Develop an interview protocol sheet that explains to everyone what’s expected from candidates in an interview. Distribute it to candidates and interviewers for review.

3. Repeat as needed

  • Return to your key metrics. Did the bias interrupters produce any change?
  • If you don’t see change, you may need to implement a stronger bias interrupter, or you may be targeting the wrong place in the hiring process.
  • Use an iterative process until your metrics improve.

RESOURCES

Footnotes

1. Bertrand & Mullainathan, 2004; Moss-Racusin, Dovidio, Brescoll, Graham, & Handelsman, 2012
2. Castilla, 2015
3. Kalev, Dobbin, & Kelly, 2006
4. Gaucher, Friesen, Kay, 2011
5. Leibbrandt & List, 2012
6. Johnson, Hekman, & Chan 2016
7. Norton et al, 2004; Brewer, 1996; Tetlock & Mellers, 2011
8. Rivera, 2015
9. Dale & Krueger, 2002; 2011
10. Thorngate, 2009
11. Bock, 2015
12. Bock, 2015; Thorngate, 2009
13. Rivera, 2015
14. Correl & Paik, 2007For the full citations, download our bibliography.
"Creative

Need Assistance Implementing Bias Interrupters? We can help.

Need Assistance Implementing Bias Interrupters? We Can Help.

200 McAllister St., San Francisco, CA 94102

The Center for WorkLife Law's website and its content are provided for general information purposes only and may not reflect current legal developments, verdicts or settlements. Any information contained in the website should not be construed as legal advice and is not intended to be a substitute for legal counsel. No attorney-client relationship will be formed based on your use of this website or any service provided.

Spread the Word

Please take a moment to spread the word about Bias Interrupters.