Numerous horror stories have emerged from LGBTQ rideshare employees and passengers facing discrimination and violence from homophobic and transphobic customers and drivers. Now a new study of 3,200 riders indicates that inherent bias is common among rideshare drivers, with African-American and LGBTQ customers -- as well as LGBTQ allies -- much more likely to be canceled on than white and straight people.
The study, "When Transparency Fails: Bias and Financial Incentives in Ridesharing Platforms," observed the habits of drivers in Washington, D.C., and how riders fare when they're perceived as African-American, queer, or an LGBTQ ally.
Study authors Chris Parker and Jorge Meija created passenger profiles with pictures of Black riders and names often perceived as Black, like Keisha, Latoya, Rasheed, and Jamal; they also created profiles that included a rainbow flag, connoting a rider was LGBTQ or an ally. In looking at how drivers responded to ride requests from such passengers, Parker and Meija discovered the Black "riders" were canceled on nearly three times as much as white riders, while the LGBTQ or ally "customers" were canceled on almost twice as much as those without the rainbow symbol.
"We know that LGBTQ riders face discrimination with these rideshare apps, but we thought that it was an interesting little twist, that even just signaling your support for the LGBTQ community could result in a canceled ride," Parker told NBC News.
The study did not look at how Black LGBTQ riders fare.
When rideshare apps became popular, many saw them as ending biases inherent in taxis, where drivers were known to zoom by people they thought of as undesirable. Since Uber and Lyft enable demographic information on riders, it looks like that same prejudice has seeped into 21st-century technology. Parker suggests apps can combat this by penalizing drivers who exhibit bias in their cancellations and incentivizing drivers who exhibit low cancellation rates among minority groups.