How Unbecoming of You: Online Experiments Uncovering Gender Biases in Perceptions of Ridesharing Performance

Abstract

Gender discrimination continues to plague organizations. While the advent of the Internet and the digitization of commerce have provided both a mechanism by which goods and services can be exchanged, as well as an efficient way for consumers to voice their opinions about retailers (i.e., via online rating systems), recent work has begun to uncover significant biases that manifest during the review process. In particular, it has been suggested that the gig-economy’s elimination of previously anonymous arm’s-length transactions may re-introduce bias into perceptions of quality. In this work, we build upon research that has identified biases based on ascriptive characteristics in rating systems of ridesharing platforms. In doing so, we conduct an online experiment using a series of factorial vignettes to consider not only willingness to transact but post-transaction perceptions of quality. Findings suggest that female drivers are not penalized ex ante for taking on gender incongruent roles, i.e., driving, and experience no penalty when a rider reports a high-quality experience. However, conditional upon a lower quality experience, there is a disproportionate penalty for female drivers. Strikingly, we also observe that historic high quality does not serve as a buffer against such penalties, as prior literature would suggest. These findings underscore the challenges platforms face in ensuring their participants are being evaluated in a fair and impartial way, and highlight the difficulties present when non-employees (customers) are rating non-employees (drivers) of the firm.