This study began with the hypothesis that public transportation interacts with land use patterns, changing travel patterns in neighborhoods served by transit. Importantly, this effect would apply not just to transit riders, who make an exchange of automobile use for transit, but also for people who do not use transit. These people, who live in places shaped by transit, would tend to drive less, reducing their overall petroleum use and their carbon footprint. In order to test this hypothesis, the authors began with a survey of the literature on the interaction of land use and travel patterns. The literature focuses on three major categories of influences on travel: land use/urban environment, socio-demographic factors, and cost of travel. For the purposes of this study, land use/urban environment variables were further broken down to include a separate category for transportation infrastructure. Many past studies have found a significant correlation between land use variables and travel behavior, though results vary depending on how the problem and the variables are defined. Boarnet and Crane (2001) emphasized that without accounting for social characteristics, like age and education, land use-transportation models are incomplete. They also discussed the importance of economic measures, such as household or personal income, as a measure of the cost of travel time. Other studies evaluated the relative importance of these and other variables, informing this model. After evaluating possible variables for this model, the authors formed a statistical model that would allow them to tease apart the relationship between land use, transit availability, and travel behavior.
Stored only when both papers are in the corpus — 60 citations exist upstream.