What influences Metro station ridership in China? Insights from Nanjing

•China is undertaking one of the most ambitious rail expansions in the world.•Factors affecting Metro ridership at the station level in Nanjing are investigated.•Bicycle P&R spaces are found to be related to Metro station ridership significantly.•Direct demand model for forecasting Metro station ridership is developed. China is undertaking one of the most ambitious rail expansions in the world. This paper investigated the impacts of factors on ridership within Metro stations’ pedestrian catchment area (PCA) in Nanjing, China. Direct ridership model was developed to explain the ridership at 55 Metro stations using a Geographic Information System (GIS) and multiple regression analysis. Independent variables included factors measuring land use, external connectivity, intermodal connection, and station context. Six variables were found to be significantly associated with Metro station ridership at the 0.05 level: population, business/office floor area, CBD dummy variable, number of education buildings, entertainment venues and shop centers. Five variables were proved to be related to station ridership at the 0.01 significance level: employment, road length, feeder bus lines, bicycle park-and-ride (P&R) spaces, and transfer dummy variable. In particular, CBD dummy variable, the number of education buildings, entertainment venues and shop centers, and bicycle P&R spaces were found to be significantly connected to Metro station ridership in the present study. The results not only confirm some findings from previous studies but also show distinct differences regarding some variables specific to the Chinese context.