Adverse impact refers to the selection process as a whole and occurs if members of a protected group (such as race/ethnicity or gender) are disproportionately screened out from the hiring process. If the difference in selection rates is over 20%, then there is evidence for adverse impact. From an assessment perspective, to reduce the risk of adverse impact in the hiring process, tools are tested for any significant differences in how such groups score. With cognitive assessments, in particular, there are significant group differences in how certain racioethnic groups score (due to socioeconomic reasons). Therefore, if cognitive assessments are used as the sole means in a hiring decision and the cut-off scores are stringent, then there is a risk of adverse impact in the hiring process. This is why it is recommended to combine cognitive assessments with other tools and selection methods as well as ensure that the cut-off scores are reasonable for the role.
The cut-off or “”passing”” score is important when assessing the risk of adverse impact for cognitive assessments and this is something that we help clients determine. We have done this multiple times with large organizations for different positions to ensure that there is less than a 20% difference in pass rates. The risk of adverse impact with cognitive assessments also depends on the difficulty of the assessment in addition to the cut-off scores. Since the Cognitive Quotient (CQ) is a basic abilities assessment, there is a lower risk of adverse impact compared to more advanced assessments (such as the Business Reasoning assessment for leaders). Our adverse impact analysis shows that there are no significant group differences on the CQ with the 25th percentile as a cut-off score, and often find on a client-by-client basis that the 50th percentile is an acceptable range.