Increasingly, bias and adverse impact are thought of as being important considerations in the selection process, which is somewhat at odds with the more traditional approach of being primarily concerned with predictive validity. While there is now a large amount of research on this topic coming out of the United States, there is limited (if any) research in the Australian context.
Consequently, the first aim of this research was to investigate whether adverse impact in selection may occur for Indigenous-Australian candidates, and if so, whether there were any explanatory variables.
A second aim of this research was to explore the use of multi-objective optimisation to arrive at selection procedure weights which simultaneously maximise validity while minimising adverse impact.
The presentation will conclude with some brief general recommendations for reducing adverse impact in practice.

This research will explain the practical and legal differences between bias and adverse impact. Furthermore, this research will demonstrate pareto-optimisation, a method which can be used by researchers and practitioners to find optimal weightings for combinations of assessments (e.g., personality, cognitive ability, structured interviews), so as to arrive at weights which optimise both validity and adverse impact. It will conclude with brief, general recommendations for decreasing adverse impact in practice.

Cummings, Daniel J