Hopefully by now you have recognized the benefits of workplace analytics and are becoming more comfortable understanding any associated risk. Does the law provide any guidance? As we discussed in a prior blog entry, some government agencies have issued reports, weighing in on the use of workplace analytics. The Equal Employment Opportunity Commission also held a public meeting last Fall regarding the use of workplace analytics. To be sure, the law around the use of analytics is still developing and it is likely that we will see additional guidance in the future.
For now, employers must look to the existing body of law when designing analytics platforms. They must consider issues of disparate treatment, disparate impact, prohibitions against certain pre-employment inquiries, and data security concerns, to name a few. There has been concern expressed by some, including the EEOC, about the presence of bias in employee selection algorithms. But there has been little recent guidance provided to employers about how to address this issue. And there may not need to be.
In 1978, a group of federal agencies, including the EEOC and Department of Labor joined to issue the Uniform Guidelines on Employee Selection Procedures (UGESP) to “incorporate a single set of principles which are designed to assist employers, labor organizations, employment agencies, and licensing and certification boards to comply with requirements of Federal law prohibiting employment practices which discriminate on grounds of race, color, religion, sex, and national origin. They are designed to provide a framework for determining the proper use of tests and other selection procedures.”
While UGESP may not have been drafted with the use of modern analytics platforms in mind, they provide a roadmap for the employer using modern analytics programs to effectuate employee selection decisions. Use of analytics platforms to make selection decisions likely fall under UGESP and its application to “other selection procedures.” Generally, under UGESP, when a selection procedure has an adverse impact against any protected group, that selection procedure must be validated. While there are a few ways to properly validate a selection procedure (here, an algorithm), generally, to be valid, a selection procedure must be shown to be legitimate and consistent with performance of the job. And employers must ensure that there are no other selection mechanisms that may be used and cause less impact.
So, while UGESP was drafted more than thirty years ago, it may just be the roadmap employers have been looking for when designing cutting-edge analytics platforms. In a sense UGESP is timeless.