Information impact on quality of multimodal travel choices: conceptualizations and empirical analyses

This paper investigates the impact of a variety of travel information types on the quality of travel choices. Choice quality is measured by comparing observed choices made under conditions of incomplete knowledge with predicted choice probabilities under complete knowledge. Furthermore, the potential impact of travel information is considered along multiple attribute-dimensions of alternatives, rather than in terms of travel time reductions only. Data is obtained from a choice experiment in a multimodal travel simulator in combination with a web-based mode-choice experiment. A Structural Equation Model is estimated to test a series of hypothesized direct and indirect relations between a traveler’s knowledge levels, information acquisition behavior and the resulting travel-choice quality. The estimation results support the hypothesized relations, which provides evidence of validity and applicability of the developed measure of travel-choice quality. Furthermore, found relations in general provide some careful support for the often expected impact of information on the quality of travel choices. The effects are largest for information services that generate previously unknown alternatives, and lowest for services that provide warnings in case of high travel times only.