Abstract: Based on conservation of resources theory and social identity theory, this paper builds a model to explore the mechanism of how perceived algorithm control affects the level of gig workers’ work engagement. By analyzing and testing 385 valid empirical data collected from takeout riders, online ride-hailing drivers and instant delivery workers, this paper examines the “double-edged sword” effect of perceived algorithmic control on the work engagement of gig workers. The results indicate that perceived algorithmic control has a “double-edged sword” effect on the work engagement of gig workers, perceived algorithmic control may, on the one hand, improve the role clarity of gig workers and, on the other hand, lower work engagement by causing a loss of work autonomy. Furthermore, perceived algorithmic procedure fairness may also strengthen the ability of perceived algorithmic control to improve gig workers’ work engagement through role clarity and mitigate its ability to weaken their work engagement through job autonomy, i.e., the higher the perceived algorithmic procedure fairness is, the better able perceived algorithmic control is to facilitate gig workers’ work engagement by increasing their role clarity, while the lower the perceived algorithmic procedure fairness is, the more likely perceived algorithmic control is to dampen gig workers’ work engagement by diminishing their job autonomy.

Key words: perceived algorithmic control, work engagement, role clarity, job autonomy, perceived algorithm procedure fairness