Ethics in the Age of Algorithms: Unravelling the Impact of Algorithmic Unfairness on Data Analytics Recommendation Acceptance

Abstract

Algorithms used in data analytics (DA) tools, particularly in high-stakes contexts such as hiring and promotion, may yield unfair recommendations that deviate from merit-based standards and adversely affect individuals. While significant research from fields such as machine learning and human–computer interaction (HCI) has advanced our understanding of algorithmic fairness, less is known about how managers in organisational contexts perceive and respond to unfair algorithmic recommendations, particularly in terms of individual-level distributive fairness. This study focuses on job promotions to uncover how algorithmic unfairness impacts managers' perceived fairness and their subsequent acceptance of DA recommendations. Through an experimental study, we find that (1) algorithmic unfairness (against women) in promotion recommendations reduces managers' perceived distributive fairness, influencing their acceptance of these recommendations; (2) managers' trust in DA competency moderates the relationship between perceived fairness and DA recommendation acceptance; and (3) managers' moral identity moderates the impact of algorithmic unfairness on perceived fairness. These insights contribute to the existing literature by elucidating how perceived distributive fairness plays a critical role in managers' acceptance of unfair algorithmic outputs in job promotion contexts, highlighting the importance of trust and moral identity in these processes.