As other blogs have noted, the Notice of Proposed Rulemaking (NPRM) for the long-awaited revisions to the Sex Discrimination Guidelines (RIN 1250-AA05) included some very interesting ideas. Some of those relate to the role of performance measurement systems, which, when used to make employment decisions like promotion, merit increases, bonuses, and termination, can be considered a selection procedure under the Uniform Guidelines on Employee Selection Procures (1978). The new regulations cite the Supreme Court ruling in Lewis v City of Chicago to support this notion, but it is an intuitive one; performance ratings (or objective performance metrics if they are available) used as part of a promotion or compensation decision process are no different than a test, interview or experience/education screen used as part of a hiring process.
Interestingly, performance ratings are mentioned in the proposed Sex Discrimination Guidelines both in the context of disparate treatment and disparate impact. Related to disparate treatment, the proposed regulations prohibit:
“Distinguishing on the basis of sex in apprenticeship or other formal or informal training programs; in other opportunities such as networking, mentoring, sponsorship, individual development plans, rotational assignments, and succession planning programs; or in performance appraisals that may provide the basis of subsequent opportunities”
Related to disparate impact, the proposed regulations note that:
“Contractors may not implement compensation practices, including performance review systems that have an adverse impact on the basis of sex and are not shown to be job related and consistent with business necessity.”
Performance ratings are probably not a new topic for readers that have been conducting regression-based EEO pay analyses. In this situation, it is often the case that performance is related to compensation outcomes, and as such should be included in an EEO pay analysis as a legitimate factor predicting pay. However, those of you that have been audited are also likely familiar with the allegation that performance ratings can be “tainted” by discrimination, and as such may not be appropriate as a legitimate factor explaining pay in a regression equation. This is often a complicated issue requiring sophisticated statistical analyses.
We suggest that two additional points are worth noting:
- Performance appraisals themselves can be assessed for adverse impact. Similar to hiring data, there are various statistical significance tests and practical significance measures that can be used to determine (a) whether a difference in performance across EEO group (e.g., sex, race/ethnicity, age) is likely due to chance, and (b) if it isn’t likely due to chance, how large the difference is. This is often a useful analysis to conduct, particularly when performance measures are used in part to make employment decisions like promotion, merit increases, bonuses, and terminations. In the last year, DCI has conducted adverse impact analyses on performance measures for a wide range of federal agency and private sector clients. It appears that employers are realizing the usefulness of such an analysis.
- Should statistically significant and practically meaningful disparities in performance measures exist between EEO protected groups, the next question is whether the performance measurement system is defensible. This is often a function of job-relatedness. Having an Industrial/Organizational Psychologist objectively review the performance appraisal system can help you understand the likelihood that your system would survive EEO scrutiny. In this context the following considerations are often important:
- Are the performance dimensions supported by some type of job analysis?
- Is the system structured such that dimensions are clearly defined, a quantitative and/or qualitative scale is used, and behavioral benchmarks are available to help the rater?
- Are raters trained?
- Is there a system in place for a high-level management review of performance ratings to determine if there are any patterns/inconsistencies that need to be reviewed?
- Is there some structured guidance on how to use performance ratings in employment decisions (e.g., merit increases, promotions)?
- Is there an appeal process for employees who believe their performance ratings are not accurate?
- Is there a well-developed feedback system through which employees can receive information about their performance that will promote their future development and enable them to improve job performance?
If you have any questions about the above issues please feel free to contact us. We have a feeling that EEO analyses of performance rating systems will become an important piece of your EEO/AA compliance puzzle……….if they aren't already.
By Emilee Tison, Ph.D., Consultant and Eric Dunleavy, Ph.D., Principal Consultant, DCI Consulting Group
