Using Structural Equations Modelling to Unravel the Influence of Land Use Patterns on Travel Behaviour of Workers in Montreal

This paper addresses the relations between travel behaviour and land use patterns using a Structural Equations Modelling (SEM) framework. SEM is a multi equation technique which is particularly suited for the study of complex relations, since it allows modelling the effects of land use patterns on travel behaviour while controlling for self selection bias and effects between endogenous variables. The proposed model structure draws on two earlier models developed both for Lisbon and Seattle which concluded for the existence of significant effects of land use patterns in travel behaviour. This paper is part of a research project which aims to compare results from different cities in North America and Europe, using the same modelling framework and similar travel behaviour and land use data sources. The travel behaviour variables included here are multidimensional and include short term, medium term, and relatively long term mobility and related decisions. Regarding long term decisions the model includes variables such as home location. In the medium terms it includes variables such as car ownership. On the shorter term decisions the variables include the amount of mobility by mode (car, transit and soft modes), both in terms of total kilometres travelled and number of trips. The model Using Structural Equations Modelling to unravel the patterns on travel behaviour of workers in Montreal ABREU E SILVA, João; MORENCY, Catherine; DAURIAN, Aurelien; GOULIAS, Konstadinos 12 th WCTR, July 11-15, 2010 – Lisbon, Portugal 2 also includes a trip scheduling variable, which is the total time spent between the first and last trips to reflect daily constraints in time allocation and travel. The modelled land use variables measure the levels of urban intensity and density, diversity, both in terms of types of uses and the mix between jobs and inhabitants/residents, the transport supply levels, transit and road infrastructure, and accessibility ratios. The land use patterns are described both at the residence and employment zones of each individual included in the model by using a factor analysis technique as a data reduction and multicollinearity elimination technique. In order to explicitly account for self selection bias the land use variables are explicitly modelled as functions of socioeconomic attributes of individuals and their households. This makes spatial decision endogenous to the model system. The model results for Montreal are discussed and compared with the results obtained using the Lisbon and Seattle datasets and previously published.