Date of Award
Doctor of Philosophy
Chris Cherry, Greg Erhardt, Luiz Renato Lima
Public transit ridership in the United States was declining even before the COVID-19 pandemic. Many different factors affect transit ridership levels, and the recent declines in ridership across the country make it critical to assess which factors are causing declines and which may help to increase ridership. Therefore, this dissertation seeks to quantify the impact of three emerging factors on bus ridership: shared electric scooters (e-scooters), new fare payment technologies and policies, and the COVID-19 pandemic.
This dissertation analyzed the impacts of shared e-scooters on bus ridership. This was investigated at the route level in Louisville, KY, and Nashville, TN, using fixed effects regression. The findings showed that the impacts on bus ridership vary by shared e-scooter trip purpose but the net impact was minimal.
This dissertation also explored the ridership impacts of mobile fare payment applications and fare capping policies using staggered difference-in-difference techniques. The results revealed that mobile fare payment applications did not yield significant ridership gains, but monthly fare capping policies did. Moreover, the impact of monthly fare capping policies was heterogenous, and agencies that adopted monthly fare capping policies for more than one year experienced an average ridership increase ranging from 3.6% to 4.1%.
The last factor considered was the COVID-19 pandemic. This dissertation proposed a multiple mediation framework to untangle the direct and indirect impacts of COVID-19. The direct impact refers to a change in travel behavior (i.e., people stop riding transit because they are sick or fear getting sick), while the indirect impact refers to reduced ridership due to changes in factors like employment. The findings suggest that employment, telework, and population relocation mediated 13% to 38% of bus ridership declines.
This dissertation makes a number of important contributions that include developing a method to estimate the impacts of shared e-scooters on bus ridership using route-level data and applying a staggered difference-in-difference approach to transportation policy analysis that considers heterogeneity in treatment effects. The findings of this dissertation can help inform transit agencies and policymakers as they plan for a post-COVID-19 world.
Ziedan, Abubakr, "Emerging Trends in Bus Ridership in the United States. " PhD diss., University of Tennessee, 2022.
Available for download on Friday, December 15, 2023