Taking account of the role of safety on vehicle choice using a new generation of discrete choice models

Abstract

In attitudinal studies safety often appears as an important attribute desired by consumers when buying a new car ( Ben-Akiva and Lerman, 1985). However, economic models of vehicle choice usually neglect the role of safety. On the one hand, to capture the qualitative nature of safety and variables related to safety one should consider safety as an underlying construct in a context of latent variable models. The problem is that psychometric models that make use of latent variables do not necessarily provide a complete understanding of agent behavior and may lead to poor predictive power. On the other hand, discrete choice models – although a powerful tool to explain decision making based on utility maximization behavior – fail to include qualitative factors as explanatory variables of the decision process. In this paper, we explore how to model safety through a new generation of discrete choice models which simultaneously consider both a standard discrete choice model and latent causal variables. Using stated preference data concerning purchase intentions of low-emission vehicles in Canada, we test a hybrid choice model to explain consumers’ preferences for safety. Based on the results as well as on the hybrid choice modeling approach, we outline a general framework for the correct modeling of the adoption of safer vehicles and appreciation of safety equipment.