Abstract
Many important decisions in societies such as school admissions, hiring or elections are based on the selection of top-ranking individuals from a larger pool of candidates. This process is often subject to biases, which typically manifest as an under-representation of certain groups among the selected or accepted individuals. The most common approach to this issue is debiasing, for example, via the introduction of quotas that ensure a proportional representation of groups with respect to a certain, often binary attribute. This, however, has the potential to induce changes in representation with respect to other attributes. For the case of two correlated binary attributes, we show that quota-based debiasing based on a single attribute can worsen the representation of the most under-represented intersectional groups and decrease the overall fairness of selection. Our results demonstrate the importance of including all relevant attributes in debiasing procedures and that more efforts need to be put into eliminating the root causes of inequalities as purely numerical solutions such as quota-based debiasing might lead to unintended consequences.
Highlights
Selection of top-ranked individuals from a larger pool of candidates is a ubiquitous mechanism for decision-making
We present a theoretical model with correlated binary attributes to demonstrate that debiasing can paradoxically worsen the representation of the most disadvantaged group if a second hidden attribute is taken into account
If only colour is considered, debiasing appears to work as intended by successfully eliminating under-representation: since the perceived quality of green entities is lower than that of the orange entities they are under-represented in top k% if the selection is blind towards all attributes, but quota-based debiasing on colour successfully corrects for that
Summary
Selection of top-ranked individuals from a larger pool of candidates is a ubiquitous mechanism for decision-making. Hiring and promotion are essentially processes of choosing top individuals from a limited 2 pool of candidates based on an implicit ranking of their skills. Such processes are known to be affected by biases. While successfully eliminating under-representation with respect to one attribute, quotas typically ignore changes in the representation with respect to other attributes. This can lead to unintended consequences and can even decrease the representation of already under-represented groups. It was found that the introduction of minority quotas in elections—while increasing the representation of minority women—could simultaneously lead to lower representation levels of women in the majority even though this group was already under-represented [8]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
