Towards more legitimate algorithms: A model of algorithmic ethical perception, legitimacy, and continuous usage intentions of e-commerce platforms

By constructing an integrated model that delves into the intricate relationship between algorithmic ethical perceptions (i.e., fairness, accountability, and transparency), algorithmic legitimacy, and continuous usage in-tentions of e-commerce platforms, this study aims to investigate the antecedents and consequences of algorithmic legitimacy. Through the utilization of structural equation modeling, we analyzed 488 questionnaires, revealing that the fairness, accountability, and transparency of algorithmic processes within e-commerce platforms posi-tively affected algorithmic legitimacy. Additionally, algorithmic legitimacy exerted a positive influence on the continuous usage intention of e-commerce platforms. Algorithmic legitimacy mediated the effect of algorithmic fairness, accountability, and transparency on the continuous usage intention of e-commerce platforms. Furthermore, this study unearthed the moderating role of user innovativeness and found it weakened the positive impact of algorithmic legitimacy on the continuous usage intention of e-commerce platforms. This study provides valuable insights into legitimacy theory in the domain of algorithms by introducing the conceptualization of algorithmic legitimacy and examining its antecedents and consequences. Moreover, this study enhances our current understanding of ethical considerations related to algorithms in the context of digital intelligence.

Automated summary

By constructing an integrated model, this study investigates the impact of algorithmic ethical perceptions and algorithmic legitimacy on the continuous usage intention of e-commerce platforms. The results show that the fairness, accountability, and transparency of algorithmic processes positively affect algorithmic legitimacy, which in turn positively influences the continuous usage intention of e-commerce platforms. Additionally, user innovativeness moderates the relationship between algorithmic legitimacy and continuous usage intention.