Not seeing the (moral) forest for the trees? How task complexity and employees’ expertise affect moral disengagement with discriminatory data analytics recommendations

Data analytics provides versatile decision support to help employees tackle the rising complexity of today’s business decisions. Notwithstanding benefits these systems, research has shown their potential for provoking discriminatory While technical causes have been studied, human side mostly neglected, albeit still need decide turn recommendations into actions. Drawing upon theories technology dominance and moral disengagement, we investigate how task employees’ expertise affect approval data recommendations. Through two online experiments, confirm important role advantageous comparison, displacement responsibility, dehumanization, as cognitive disengagement mechanisms that facilitate such approvals. generally enhances mechanisms, retains a critical in analytics-supported decision-making processes. Importantly, find complexity’s effects on users’ dehumanization vary: more subjects increase whereas richer information opposite effect. By identifying approvals recommendations, this study contributes toward designing tools, methods, practices combat unethical consequences using systems.