Big data and Artificial Intelligence use in job selection processes: An analysis of candidates' perceptions of organizational attractiveness

The adoption of ‘algorithmic' human resource management (HRM) has been expanding rapidly, with algorithmic selection tools proliferating. This is associated with the belief that algorithms enable efficient, optimized, and data-driven decision-making. However, the fact that organizations use algorithms and run them on different sources of information may influence candidates’ perceptions of the organization. While the “war for talent” remains fierce, and most companies are struggling to meet talent demands, understanding candidates’ reactions and concerns towards algorithms during selection becomes timely and needed, particularly since it is an under-researched area. This study aims to analyze how candidates’ perceptions and reactions to an organizational attractiveness can be affected by the use of professional and personal data in algorithmic selection. We conducted a 3x2 scenario-based design experiment with business and engineering students (n=265) from Italy and Spain. The experiment comprised three different hiring scenarios based on algorithms and different sources of information. The experiment evaluated how candidates’ perceptions varied through various established measures. This study is the first to analyze how candidates perceive different dimensions of organizational attractiveness depending on the use of algorithms, and professional and personal data in algorithmic hiring. Against tendencies to present algorithmic hiring as having adverse outcomes, the results reflect minor variations in candidates’ decision to proceed with their job application. However, we noted differences in candidates’ perceptions of algorithmic selection depending on their academic backgrounds and trust in technology. The field of algorithmic hiring is developing rapidly, and recruiters must create new hiring strategies integrating such algorithmic tools. The study's results make it possible to predict candidates' attitudes to an organization that uses algorithms and different kinds of information during their hiring. The study also proposes an agenda for future research and discusses theoretical, methodological, practical, and policy implications.