Bus ridership retention, place-based factors, and COVID-19 in the Boston area

Ridership declines are a major concern for public transport agencies in the United States, especially in the aftermath of the COVID-19 pandemic. This paper examines how bus ridership retention patterns changed in the Boston metro area from 2018 to 2022, i.e. in the years before, during, and after the COVID-19 lockdown. Using spatial regressions, multilevel models, and multi-scale geographically weighted regressions, we analyze how local socioeconomic conditions and built environment characteristics predict ridership retention at the bus stop and census tract levels. The results suggest that high frequency bus routes are an important factor: census tracts and bus stops with high frequency routes were associated with higher ridership retention in the Massachusetts Bay Transportation Authority service area. Bus stops located in census tracts with high percentages of essential workers were also significantly likely to be associated with maintaining ridership during the lockdown and afterward. The results suggest that key bus routes targeting transit dependent populations can be an important strategy for transit agencies in equitably responding to changes in commuting patterns stemming from the COVID-19 pandemic.