Urban Public Transport Mode Choice Behavior with Different Income Levels During the Controlling Stage of the COVID-19 Pandemic

Public transport ridership has been hit hard by the COVID-19 pandemic in many cities. Studying passengers' public transport choice behavior during the pandemic can uncover the major influential factors and help propose policies reduce the negative impacts of the pandemic on public transport revenue. In this study, we propose integrating the structural equation model (SEM) and the Binary Logit (BL) model to study urban public transport choice behavior by income level specifically during the controlling stage of the COVID-19 pandemic. Taking Beijing, China as an example, we collected survey data including passengers’ socioeconomic attributes, travel attributes, and attitude-perception attributes considering the subjective risk perception, and analyzed their impacts on mode choice preferences. The results show that risk perception has a significant impact on public transport choice preference, indicating that the government should take measures to reduce the epidemic transmission risks on public transport, so as to improve the passengers' loyalty, recognition, and satisfaction with public transport during the controlling stage of the COVID-19 pandemic. Furthermore, the high-income, the middle-income, and the low-income groups value the quickness of subway, the convenience of subway, low cost of bus, respectively, which contributes to proposing targeted strategies to improve the public transport ridership. Finally, compared with the BL model, the SEM-BL model can achieve higher goodness-of-fit, which help better explain the actual travel behavior. The methodology proposed in this paper provides a theoretical framework for modeling mode choice behavior and some policies to speed up public transport ridership and economic recovery.