First Advisor

Anthony M. Rufolo

Term of Graduation

Spring 2006

Date of Publication


Document Type


Degree Name

Doctor of Philosophy (Ph.D.) in Urban Studies


Urban Studies




Land use -- Oregon -- Portland Metropolitan Area, Local transit -- Oregon -- Portland Metropolitan Area, Transportation -- Oregon -- Portland Metropolitan Area, Social surveys -- Oregon -- Portland Metropolitan Area



Physical Description

1 online resource (2, viii, 198 pages)


Empirical studies on the link between land use characteristics and travel behavior may generate biased estimates when the choice of residential locations is influenced by mode preference (i.e., residential self-selection). Without appropriately addressing this matter, the effects of land use characteristics on travel behavior may be overestimated. This study aims specifically to address the residential self-selection problem related to transit service. It also aims to take into account the effects of density and transit service at the origin and destination on mode choice because the level of transit service is often correlated with density. Mode choice effects associated with density may actually be due to levels of transit service.

The two-day activity data from the 1994/1995 Portland Household Activity and Travel Behavior survey are used to analyze discrete choice models of commuting mode and residential location choices. Generally omitted in many studies, different transit service characteristics around both residential and employment locations are identified at a micro scale level. A cluster analysis is employed as a way to classify areas that have similar transit service characteristics into distinct groupings, or levels of transit service. Levels of transit service derived from the cluster analysis characterize different dimensions of transit service.

No evidence was found that the residential choice regarding the levels of transit service and commuting mode choice were jointly made. People do not appear to choose residential locations with regard to levels of transit service to match their preferred mode. Although this relationship (the self-selection) could not be found, the findings suggest that it is necessary to include transit service characteristics in mode choice models because they help to separate transit service effects from density.

With the exception of the self-selection question, the findings suggest that the levels of transit service at the employment locations (not the ones at the residential locations) encourage transit use and non-motorized modes. High levels of transit service at the residential location do not appear to influence transit use. As a result, land use policies and improvements of transit service emphasizing employment locations may be more effective than those that target residential locations.


In Copyright. URI:

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).


If you are the rightful copyright holder of this dissertation or thesis and wish to have it removed from the Open Access Collection, please submit a request to and include clear identification of the work, preferably with URL.

Persistent Identifier