Abstract
This study focuses on understanding the commute modal choice of university students. The methodological novelty lies in using students commuting from their parents’ homes to school as a quasi-experimental group to control for residential self-selection bias; these students face exogenous residential constraints that largely decouple their home location from travel preferences. The objective is to highlight the relative influence of public transport accessibility on students' propensity to choose alternative modes over the car, while contrasting the effects of residential neighborhoods versus campus areas. Utilizing data from the MobiCampus-UdL Travel Surveys (2017–2019), two separate binary logit models were estimated: one for the entire student sample and one specifically for students living with their parents to rigorously contrast and validate the behavioral determinants. The study analyzes responses from 8,783 students across 17 educational institutions of different urban settings within the University of Lyon, France. The results show that beyond car access and travel characteristics, spatial factors (parking policy and transit accessibility) strongly influence modal choice. Notably, the availability of a metro or tram station within 500 meters matters for all students, regardless of their living arrangements. Furthermore, public transport accessibility impacts students’ car non-use more significantly in campus areas than in residential neighborhoods. These findings suggest that to effectively reduce car use among higher education students, policy measures should focus primarily on improving public transit accessibility in and around study areas.

This publication has 35 references indexed in Scilit: