Traveler preference for park-and-ride facilities: Empirical evidence of generalizability

This paper reports the main findings of a study, conducted in the Netherlands, aimed at testing whether preference functions for park-and-ride facilities, estimated from data collected in a specific Dutch region, can be generalized to a nationwide sample. Preference data in both samples were collected with hierarchical information integration. Contrast parameters were used to test the equality of a set of parameters for decision constructs that were assumed to influence the choice of park-and-ride facilities. Results suggest that the estimated preference functions for the two samples are largely the same within conventional statistical error bounds and provide empirical evidence of generalizability. In addition to the academic importance of this finding, for practitioners it means that no tailored-made research is required to assess the feasibility of such new park-and-ride facilities, especially if the results of this study are further replicated in other contexts and regions.