Analyzing the relation between commuting satisfaction and residential choices using discrete choice theory and structural equation modeling

The concept of travelling satisfaction is gaining more and more interest in the transportation field. While increasing travellers’ satisfaction should be a goal of policy makers and practitioners, a drop in commuting satisfaction might lead to switch from one mode to another. Objective trip characteristics (time, cost, mode) as well as other elements such as mode attitude, lifestyle, etc. affect travel satisfaction rating. Despite an extensive literature on travel satisfaction determinants, often, the interaction between the studied determinants is overlooked. The main aim of this paper is to quantify the impact (both direct and indirect) of residential choices on the home-to-work stated travelling satisfaction. Methodologically, a Discrete Choice Theory approach (via the well-known concept of the utility Logsum) and a Partial Least Square Structural Equation Modelling (PLS-SEM) approach have been used and compared. Results of both modelling approaches show that the direct effect of residential choices on commuting satisfaction is negligible compared to individuals’ external factors such as trip characteristics. However, using the PLS-SEM approach, indirect effects of residential choices on commuting satisfaction can be quantified