A Model of Spatial Structure, Activity Participation and Travel Behavior

In recent years many studies have addressed the relationship between land use and travel behavior. At the same time, the notion that travel demand stems from activity demand has been explicitly incorporated into activity-based models. The role of spatial context in the activity-based approach has, however, been largely neglected. This paper has explored the influence of spatial variables on personal activity and travel behavior, controlled for personal and household variables. For this purpose a structural equation model was developed to simultaneously estimate direct and indirect causal relationships. The model was based on two-day diary data that was collected in the Netherlands. The model fit was good, as was the explained variance between the activity and the travel variables. Significant relationships emerged between the activities themselves and with trip generation and travel time demand. The diverse socio-demographics exercised direct and indirect influences on the endogenous variables. The spatial variables exercised only a weak but nonetheless significant influence on the activity and travel variables. One (tentative) conclusion is that indirect effects can steer a total effect in another direction, for example, though density has no direct effect on travel time, the total daily travel time still increases because of the extra trips in high-density areas.