Originally posted on the Mattingly Solutions Blog.
Performance management systems should be focused on performance. Unfortunately, all too often bias plays a role in these evaluations. What kinds of bias affect performance evaluation?
Similarity bias. This type of bias occurs when people are attracted to or respond more favorably to people like them. Research has shown that this type of bias leads to racial differences in performance evaluations, where white managers are more likely to rate white employees higher than Black employees.
The “prove it again” bias. This type of bias occurs when groups stereotyped as less competent have to prove themselves more so than the majority of white men. These groups, including racial minorities, women, and LGBTQ+ individuals, are more likely to have their mistakes noticed and remembered as compared to white men in performance reviews (HBR).
The tightrope. There is also evidence that there is a narrower range of workplace behavior accepted from women and POC. This results in these individuals’ personalities being mentioned more often in their performance reviews. Further, women, especially women of color, are more likely to have gendered office housework tasks brought up in their review (HBR).
These are just some examples of the ways in which bias can have an impact on performance evaluations (see this article for more examples). These instances of bias can have profound impacts.
For example, a midsize law firm was auditing their performance evaluation system for bias and found that only 9.5% of people of color in their firm had mentions of leadership in their performance evaluations, which was more than 70% lower than white women. These leadership ratings were predictive of higher competency ratings in the following year.
Based on this evidence, it’s important to consider how performance evaluations can leave room for bias. Often the place that most often allows bias to interfere is the open-ended box that asks managers to provide broad feedback about their employees. When there are no criteria or clear instructions, we are more likely to fall victim to bias. We see that this is the case for open-ended responses, as men are more likely to receive longer reviews focused on technical skills while women are more likely to receive reviews concerning their communication skills (HBR).
How can performance management systems be changed or redesigned to minimize bias and increase effectiveness?
Focus on competencies and clear objectives.
Evaluations should focus on competencies that are a priority to the organization and in alignment with organizational values. Shaping performance management around the desired competencies reduces bias by specifying what managers are evaluating while also encouraging specific behaviors and skills that are desired by the organization.
Require evidence for any open-ended feedback.
When requesting open-ended feedback from managers, mandate that they provide clear evidence for any claim they make in reference to an employee’s performance, either positive or negative. This evidence requires managers to think twice about any statement they are making about an individual’s performance.
Create a rubric for evaluations.
Similarly, if you are using an open-ended style of question in performance reviews, provide managers a rubric to use that defines the specific criteria employee’s performance should be evaluated by.
Use more descriptive prompts.
Providing a clearer prompt to managers can also reduce bias. For example, rather than something like “Tell us about your employee’s performance over the last year,” instead, prompt with “Describe the ways your employee's performance met your expectations.”
Run a consistency check.
Lastly, encourage managers to run a consistency check after they complete their evaluations across different employees. Note places where they may have allowed bias to influence their feedback, referencing some of the common biases mentioned above.
Looking to advance DEI through data-driven insights in your organization? Contact us at Plan to Action today to drive meaningful change, together.
Comments