Not all algorithmic controls are equal: the double-edged impact of algorithmic control dimensions on mental health and risky riding behavior among food delivery riders

Abstract Online labor platforms rely on big data-driven algorithms to implement precise, immediate, and fine-grained control over the labor service process of food delivery riders, called algorithmic control. While algorithmic control enhances the overall efficiency of platform-based organizations, it also exerts multiple psychological and behavioral impacts on food delivery riders, though the underlying mechanisms remain unclear. This study subdivided the three dimensions of perceived algorithmic control (i.e., tracking evaluation, behavioral constraint, and standardized guidance), and considered work pressure as a mediator based on the Job Demands-Resources (JD-R) model, in order to explore whether and how perceived algorithmic control affects the mental health and risky riding behavior of food delivery riders. The study also investigates the moderating role of perceived autonomy. Data from 466 Chinese food delivery riders were analyzed using a structural equation model and bootstrapping procedure. The results showed that both perceived algorithmic tracking evaluation and behavioral constraint impaired riders’ mental health and promoted risky riding behavior through work pressure, while perceived algorithmic standardized guidance improved their mental health and reduced risky riding behavior through work pressure. Perceived autonomy mitigates the negative effects of algorithmic tracking evaluation and enhances the positive effects of algorithmic standardized guidance, but it amplifies the negative effects of algorithmic behavioral constraint. These findings provide a clear explanation for understanding the “double-edged sword” effect of perceived algorithmic control on the food delivery riders’ psychology and behavior, which not only enriches the research on algorithmic control in the context of the gig economy, but also sheds light on the managerial practices of platform-based organizations to improve their algorithmic systems and enhance the sustainable development of gig workers.