Determinants of ground transport modal choice in long-distance trips in Spain

• Long-distance mode choice is studied using a multilevel multinomial model.
• Spatial heterogeneity in modal choice decision is taken into account.
• Three ground transport alternatives are studied: private car, bus and train.
• The impact of sociodemographic, land-use and trip attributes variables is computed.
• Overnight stays is included as an explanatory variable.

In transport economics, modeling modal choice is a fundamental key for policy makers trying to improve the sustainability of transportation systems. However, existing empirical literature has focused on short-distance travel within urban systems. This paper contributes to the limited number of investigations on mode choice in medium- and long-distance travel. The main objective of this research is to study the impacts of socio-demographic and economic variables, land-use features and trip attributes on long-distance travel mode choice. Using data from 2007 Spanish National Mobility Survey we apply a multilevel multinomial logit model that accounts for the potential problem of spatial heterogeneity in order to explain long-distance travel mode choice. This approach permits us to compute how the probability of choosing among private car, bus and train varies depending on the traveler spatial location at regional level. Results indicate that travelers characteristics, trip features, cost of usage of transport modes and geographical variables have significant impacts on long-distance mode choice.