Modelling Work-Trip Mode Choice Decisions in Two-Worker Households

Abstract

This paper proposes that the theoretically more appropriate approach for modelling mode choice decisions for the journey to work by workers residing in two-worker households is through the development of a choice model at the household level that takes into account the decisions of both workers. There are at least two reasons for this. First, in the majority of cases the model estimation data comes from household travel surveys, which have the household as the sampling unit and not the individual worker. The majority of households in urban areas these days have two or more workers. Thus once a household is selected into the sample all its working members are necessarily selected as well. The idea of a random sample therefore holds at the level of the household and not the individual worker. Second, workers from the same household in the long run select modes for their journeys to work so as to minimize their combined disutility to travel as opposed to independently minimizing their respective individual disutilities. The household model thus treats the alternatives available to the household workers as a "modal bundle" with the two household workers, after intra-household bargaining, selecting the modal bundle that minimizes their combined total disutility to travel. An empirical test of both household and individual choice models is conducted and the household model found to give more precise predictions.