Background
The Transit Cooperative Research Program has undertaken a major study of the choice of neighborhood form and selection of mode of travel, entitled, “Understanding How Individuals Make Travel and Location Decisions: Implications for Public Transportation, TCRP H-31.”  In this project, the research team has applied the Theory of Planned Behavior (TpB) first to the question of how people consider conditions supportive of walking and transit in the selection of their neighborhood; and second, to how such conditions are associated with their propensity to alter their transportation behavior to a lifestyle more dependent on walking and transit.

Objectives
The objective of this paper is to describe some of the early results of TCRP H-31 concerning the role of urban form and individual values in the determination of the propensity to walk and take transit.  

Methods (and Type of Data Collected)
In the first wave of surveying, 865 individuals were surveyed for approximately one half hour concerning many factors related to their choice of both neighborhood type and their mode of transportation. The content of the survey was designed to support the application of the TpB to the question of selection of neighborhood. To this end, the survey examined the role of normative factors (reflected in the subjective norm), the role of attitude formation (reflected in the attitude toward the behavior) and the role of various constraints to adopting the behavior (reflected in the perceived behavioral control.)

To determine the role of “attitude toward the behavior” a series of questions were asked operationalizing the concept of “expectancy/value” theory, which is reflected in the structure of the TpB. In the analysis described in the paper, a composite ‘factor’ was created from a set of these questions concerning the values held by the respondents towards urban conditions, separate from their actual ability to carry out the actions implied by those values. The questions were assembled based on their similarity of response to the base question, “For me, to live within walking distance to stores, restaurants, a public library and a school would be… (desirable/undesirable).” The factor was assembled from 17 questions, selected manually for highest internal consistency and reliability as measured by Cronbach’s alpha.

Based on a cluster analysis resulting from the application of this factor, the full survey sample was clustered into two subgroups, separating those holding “green values” from those holding “non-green values.” 

Results
Basic four-cell matrices were developed to reveal the interaction between attitudes and neighborhood form, as it relates to mode choices including, walk/bike total trip mode share, walk/bike non-work mode share, number of utilitarian walk trips per month, number of non-work utilitarian walk trips per month and transit work trip mode share. All four cells of each matrix performed as expected in each case. Consistent with other research, walk trips for exercise/fun did not vary by neighborhood type or by attitude towards urbanity.  

The results of the exercise help to reveal the extent to which one’s modal behavior is related to various conditions of urban form, arrayed against the two levels of personal values. In most cases, the factor representing values has a higher power of explanation for modal behavior than the variables representing various conditions of urban form.  

Conclusions
The implications of the research suggest that within each neighborhood type, those who hold a set of positive values towards urban conditions have a significantly higher propensity to take transit and undertake utilitarian walking than those who do not hold such pro-urban values. As this exercise was undertaken as an early spin-off from the much larger dataset of the full project, it will be important to re-examine these early conclusions based on the results of the second wave of surveying (that is now underway), which focuses on modal behavior. At the conclusion of the study, all of this will be expressed as part of a developing “three level model” which uses the TpB as the starting point for a construct, which separates out various input variables in terms of:

1) A personal level, at which inclination to behavior is formed;
2) An interpersonal level at which the effects of peer pressure from one’s personal social network are felt;
3) An environmental level at which the physical or other constraints of the real world come into play concerning the possible adoption of the behavior.