Citation

Ziedan A, Darling W, Brakewood C, Erhardt G, Watkins K. Transp. Res. A Policy Pract. 2021; 153: 20-34.

Copyright

(Copyright © 2021, Elsevier Publishing)

DOI

10.1016/j.tra.2021.08.019

PMID

unavailable

Abstract

Shared electric scooters (shared e-scooters) are rapidly growing in popularity across the United States. In 2019, more than 88 million shared e-scooter trips occurred nationwide. However, the impact of this growing mode on transit ridership is still largely unknown and limited to the findings of user surveys. The overarching goal of this study is to conduct an empirical analysis to quantify the impact of shared e-scooters on bus ridership using Louisville, Kentucky, as a case study conducted prior to the COVID-19 pandemic. To assess if shared e-scooters decrease or increase bus ridership, this study evaluated more than half a million shared e-scooters trips, and estimated several daily, weekly, and monthly fixed effects regression models of route level bus ridership. The results of the preferred fixed effect regression models indicate that shared e-scooters do not have a significant impact on local bus ridership. However, the model results suggest that shared e-scooters could potentially complement express bus routes as they serve the first/last mile of a trip, but further research on express routes is recommended. Most critically, the results of this study indicate that shared e-scooters are not one of the primary causes of bus ridership decline in Louisville. This finding is important for cities across the United States as they explore the causes of the recent declines in bus ridership.

Language: en

Keywords

Bus ridership; Fixed effects regression; Shared electric scooters