Partially supported by a NULab Seedling Grant.
Online dating profiles are the digital analogue of personal ads that were traditionally published in newspapers and other print media. Although these services have long been studied by researchers from the relationship sciences, their transition to the digital world has only recently produced a surge of interest from computational social scientists. However, the majority of projects being conducted in this topic area have focused on the experiences of, and potential benefits to the average user; little is known about the experiences of, and harm faced by women and minorities on these platforms.
Due to the lack of diversity in tech industries—from the design process to testing—technologies like online dating are often built with one target audience in mind: White users. Despite the limited scope of this target audience, these technologies are often adopted by diverse masses, which can result in adverse and even harmful effects for some users.
To better understand how users’ implicit (or explicit) racial biases interact with the algorithms that govern online dating platforms, we plan to conduct a series of experiments that examine how a user’s experience is shaped by their race and gender. We approach this study with an algorithm auditing framework, focus on one of the most popular online dating platforms in the US (Tinder), and hypothesize that online dating algorithms may reinforce existing racial hierarchies rather than create new inequalities.
Collaborators: Apryl Williams, Hanyu Chwe
Ronald Robertson, Network Science PhD Student