Using hierarchical tree-based regression model to examine university student travel frequency and mode choice patterns in China

This paper applies a nonparametric statistical method, hierarchical tree-based regression (HTBR) model, to explore university student travel frequency and mode choice patterns in China, using the data collected by a web-based travel survey. In this study, HTBR models were constructed to predict student travel frequency and classify student mode choice. It was found that student grade, school location city, public transit station coverage ratio (PTSCR) and family income have impacts on student travel frequency, and travel distance, bicycle ownership, school location city, PTSCR and student gender are significantly correlated to student mode choice. The study results reveal travel characteristics of university students in China at a disaggregate level and provide information to better understand their travel behaviors.