The relationships between land use measures and travel behavior: a meta-analytic approach

From a review of the literature on residential self-selection and spatial multicollinearity, this paper identifies the relationships that travel behavior has with three factors: land use, sociodemographics, and attitudes. It then analyzes the magnitude of the land use-travel behavior relationship by testing two meta-regression models. Each model specifies: (1) residential self-selection as a sociodemographic or attitudinal confounding factor in the land use-travel behavior relationship; and (2) spatial multicollinearity as a land use confounding variable. The first model shows that the land use factor is related to travel behavior more strongly than is residential self-selection. The second model finds that, among land use variables, accessibility to regional centers is most strongly related to travel behavior. Depending on the accessibility-travel behavior relationship, other land use variables are conditionally related to travel behavior. This finding implies that accessibility represents the overall effect of land use on travel behavior.