Search for a command to run...
Uber is a popular ride-sharing application that matches people who need a ride (or riders) with drivers who are willing to provide it using their personal vehicles. Despite its growing popularity, there exist few studies that examine large-scale Uber data, or in general the factors affecting user participation in the sharing economy. We address this gap through a study of the Uber market that analyzes large-scale data covering 59 million rides which spans a period of 7 months. The data were extracted from email receipts sent by Uber collected on Yahoo servers, allowing us to examine the role of demographics (e.g., age and gender) on participation in the ride-sharing economy. In addition, we evaluate the impact of dynamic pricing (i.e., surge pricing) and income on both rider and driver behavior. We find that the surge pricing does not bias Uber use towards higher income riders. Moreover, we show that more homophilous matches (e.g., riders to drivers of a similar age) can result in higher driver ratings. Finally, we focus on factors that affect retention and use information from earlier rides to accurately predict which riders or drivers will become active Uber users.