Modeling Commuting Behaviour in Canadian Metropolitan Areas: Investigation of Factors and Methods with Micro Data
This thesis investigates the factors which affect commuting distance in Canadian Metropolitan Areas. The commuting behavior in Windsor, Halifax, Calgary and Ottawa-Gatineau is studied using confidential micro-data from the 2006 Canadian Census survey. Land use datasets from Statistics Canada, Desktop Mapping Technology Inc. and Natural Resource Canada are also utilized. The nature of urban form in these CMAs is explored. Regression and ordered choice models are employed to study the similarities and differences in commuting behavior. The impact of factors such as sex, age and employment status is line with previous studies. However, the use of new variables, namely mortgage, mobility and occupational types, provide new insights about the modeled process. Accessibility to jobs and mixed land use are the most significant land use variables in all CMAs. The recommendations from this thesis support philosophies that promote multinucleated and mixed land use development within the context of smart growth strategies.