Adverse Impact and Cognisess

By Nick Jarman

Recognise Talent – Enable Potential – Encourage Diversity

The Cognisess Approach

Adverse impact refers to employment practices that appear neutral but have a discriminatory effect on a protected group. Adverse impact may occur in hiring, promotion, training and development, transfer, layoff and even performance appraisals.* 

People analytics offer many advantages when it comes to managing a hiring process, maximising the quality of a company’s new hires and reducing any Adverse Impact found in a process.  

The merits of removing human elements of a selection process are many, not least in saving a huge amount of time that would normally be spent sifting through CVs, reading cover letters and processing hundreds and thousands of candidates manually.  

As well as this, reducing manual processes in selection can help alleviate a lot of potential bias, from an unconscious lean towards candidates with similar interests and backgrounds to preferring candidates with similar work experiences and ignoring those demonstrating high potential but lacking in past opportunities to demonstrate it in the workplace.  

Making use of data-driven analytic processes can help the best candidates to be identified, regardless of who they are, where they have come from and what they have previously achieved in their career.  

Cognitive assessments are the gold standard for how people analytics methodology can be used in predicting future job potential. However, this form of assessment has been associated with differences between demographic groups.  

Personality assessments provide fantastic data to enrich an overall view of a candidate’s suitability for a particular type of job. These assessments don’t have particularly high validity in predicting future performance when used in isolation but in combination with cognitive assessments can be invaluable in providing a holistic view of a candidate. Another big advantage of this form of assessment is in the small differences found in personality attributes across ethnicity, gender and age.  

At Cognisess we don’t rate candidates based solely on their attribute scores. We make use of an algorithm which takes Attributes judged to be important in the role and assigning them weightings used to predict future success in a role. By using these ‘Profilers’ we are able to minimise any differences known to affect candidates of demographics at the level of each attribute, thus facilitating the reduction of bias in any hiring decisions. Further to this, we can reduce a data set comprised of several Attribute scores and reduce this into a percentage match score, making the sifting of large candidate cohorts as simple as the click of a button.  

The use of different cut-off scores for accepting/rejecting candidates can influence the demographics of candidates sent through to the next stage of a selection process. Ultimately this is the more salient end-measure of Adverse Impact in a process and so it is important to be able to be able to discover the effects of different Profilers and cut-off scores before finalising any decision in candidate selection. This is easy to do with a Profiler, ensuring the best and most-informed decision is always made and the least possible amount of Adverse Impact is found within the selection process.  

Harnessing statistics, psychometrics and flexibility, Cognisess can aid the fair selection of the most capable, and the most diverse range of candidates for your role.  

*Society for Human Resource Management 2020