What affects commute mode choice: neighborhood physical structure or preferences toward neighborhoods?

https://doi.org/10.1016/j.jtrangeo.2004.11.001Get rights and content

Abstract

The academic literature on the impact of urban form on travel behavior has increasingly recognized that residential location choice and travel choices may be interconnected. We contribute to the understanding of this interrelation by studying to what extent commute mode choice differs by residential neighborhood and by neighborhood type dissonance—the mismatch between a commuter’s current neighborhood type and her preferences regarding physical attributes of the residential neighborhood. Using data from the San Francisco Bay Area, we find that neighborhood type dissonance is statistically significantly associated with commute mode choice: dissonant urban residents are more likely to commute by private vehicle than consonant urbanites but not quite as likely as true suburbanites. However, differences between neighborhoods tend to be larger than between consonant and dissonant residents within a neighborhood. Physical neighborhood structure thus appears to have an autonomous impact on commute mode choice. The analysis also shows that the impact of neighborhood type dissonance interacts with that of commuters’ beliefs about automobile use, suggesting that these are to be reckoned with when studying the joint choices of residential location and commute mode.

Introduction

In the USA and Europe land use based solutions to transportation problems have rapidly gained in popularity over the past decade. The principles of New Urbanism (in the USA) or the Compact City (Europe) have found a solid place in the profession’s thinking. This popularity is not least the result of numerous empirical studies demonstrating that living in higher density, mixed use neighborhoods is associated with less car use compared to living in low density, suburban environments (for example, Frank and Pivo, 1994, Næss et al., 1995).

The academic literature is, however, equivocal about the effect of neighborhood characteristics on reducing car use in general and for commuting in particular. It is not completely clear, for instance, how important land use characteristics are in the explanation of commute behavior. While urban form dimensions at the neighborhood level affect commute mode choice or commute length (Cervero, 1996a, Cervero, 2002), other variables appear to be more important. This is particularly true for sociodemographic factors, such as gender, household composition, and income. However, travel-related attitudes or lifestyle variables may also be more important. Although not specifically focusing on commute travel, Kitamura et al. (1997) found that attitudes are more strongly associated with travel than are land use characteristics.

In addition, the relationship between urban form and commute behavior may not be a direct one. Recently a number of authors have claimed that residential location choice is not exogenous to the association between land use variables and travel behavior (Boarnet and Crane, 2001, Cervero and Duncan, 2002, Handy, 1996, Sermons and Seredich, 2001, Srinivasan and Ferreira, 2002, Van Wee et al., 2003). They argue that a household with a predisposition toward a certain type of travel “self-selects” a residential location enabling the pursuit of that preferred type of travel. For example, households whose members prefer to travel by public transit choose to reside for that very reason in a location providing easy access to transit infrastructure. If this is true, the commonly observed correlations between land use configuration and travel behavior do not so much reflect direct causality but complex relationships of these factors with others, such as attitudes toward travel. This suggests a need for studies of the interdependence of different types of residential locations, commute behavior characteristics, and attitudes toward travel and land use. To the best of our knowledge little work has thus far been done on this subject.

This paper is positioned in a series of studies designed to enhance our understanding of the complex relationships among residential location, commute behavior, and attitudes toward travel and land use. We focus on the concept of residential neighborhood type dissonance, or mismatch between preferred and actual type of residential location, as a way to assess the comparative roles of the built environment and residential self-selection in travel behavior choices. Our basic question is simple: do mismatched individuals travel more like the matched residents of the neighborhood they actually live in, or more like the matched residents of the kind of neighborhood they prefer to live in? The former outcome suggests that the effects of the built environment outweigh personal predispositions; the latter outcome suggests the converse.

Schwanen and Mokhtarian (submitted for publication) begins this series of studies by exploring the role of attitudes toward travel and land use in residential location choice. In Schwanen and Mokhtarian (2004), we model dissonance itself as a function of demographic and attitudinal characteristics. Three papers follow to evaluate the impact of dissonance on travel behavior: Schwanen and Mokhtarian (2003) compares non-commute trip frequencies of matched and mismatched urban and suburban residents; in the present paper, we compare the commute mode choice of consonant and dissonant workers; and Schwanen and Mokhtarian (in press) completes the picture by examining the role of dissonance in mode-specific distances traveled for all purposes. Each study uses data from the same survey of commuters in three neighborhoods in the San Francisco Bay Area: the urban neighborhood of North San Francisco and the (different types of) suburban communities of Concord and Pleasant Hill. Because land use preferences and realities are taken into account as well as sociodemographics, mobility limitations, personality traits, and lifestyle factors, the empirical analyses presented in these studies offer deeper insights into the nature of the influence of land use characteristics on travel behavior.

The study background and hypotheses tested here are detailed in the following section. The paper then proceeds to a description of the data available for this study, as well as definitions of the variables used in the empirical analysis. Section 4 investigates the impact of neighborhood dissonance on commute mode choice through descriptive analysis, and Section 5 presents a multinomial logit analysis in which sociodemographic factors, mobility limitations, personality factors and lifestyle types are included as control variables. The paper concludes with a summary of the results and a discussion of the implications for land use and transport policy.

Section snippets

Study background

Mode choice for commute trips is probably the dimension of travel behavior that has been studied most thoroughly. Conventional wisdom holds that workers act as rational consumers and choose the mode providing the highest utility. This utility is typically a function of objective price or level of service factors—travel time and travel cost—and taste variables, usually represented by socioeconomic and demographic characteristics of households and sometimes supplemented by locational variables (

Data

The data used for this study comprise responses to a 14-page questionnaire that collected information on a variety of travel and related issues. The survey was mailed in May 1998 to 8000 randomly selected households of three neighborhoods in the San Francisco Bay Area. Half were mailed to the urban neighborhood of North San Francisco (Fig. 2); the other half were split evenly between the contiguous suburbs of Concord and Pleasant Hill (Fig. 3). A randomly selected adult member of the household

Descriptive analysis

Not surprisingly, the driver/passenger in a personal vehicle category dominates the mode split: 76.9% of the respondents in the whole sample commute by this alternative. The second largest share is that for bus/ferry (9.7%), which is due to the large number of North San Francisco residents in the total sample. Segmented by neighborhood, we see that the personal vehicle class is even more dominant for suburban respondents (88.5%). The differences with North San Francisco are marked: although the

Multinomial logit analysis

Although the descriptive analysis suggests that the level of neighborhood type dissonance affects commute mode choice, the question remains whether this holds true after other factors—sociodemographics, mobility limitations, personality and lifestyle types, and travel attitudes (Table 2)—are taken into account.

A convenient and common functional form for analyzing the influence of potential explanatory variables on a categorical dependent variable is the multinomial logit (MNL) model. The MNL

Discussion and conclusion

This paper has sought to enhance our understanding of the complicated relationships among residential location, commute behavior, and attitudes toward land use and travel. We have investigated to what extent commute mode choice differs not only by residential neighborhood but also by the presence and level of mismatch between a commuter’s current and preferred type of neighborhood. Hence, the work reported here provides insights into the question of the relative importance of individuals’

Acknowledgments

The University of California Transportation Center funded the collection of the data used for this study, with further analysis also supported by the Daimler-Chrysler Corporation. Part of the research reported here was carried out during a visit of the first author to the Institute of Transportation Studies of the University of California, Davis. This visit was sponsored by a SYLFF grant from the Tokyo Foundation with additional subsidy from the Travel Fund of the Netherlands National Science

References (39)

  • S. Srinivasan et al.

    Travel behavior at the household level: understanding linkages with residential choice

    Transportation Research D

    (2002)
  • M.N. Bagley et al.

    A methodology for the disaggregate, multidimensional measurement of residential neighbourhood type

    Urban Studies

    (2002)
  • M. Ben-Akiva et al.

    Discrete Choice Analysis: Theory and Application to Travel Demand

    (1985)
  • M.G. Boarnet et al.

    Travel by Design: The Influence of Urban Form on Travel

    (2001)
  • R. Cervero

    Traditional neighborhoods and commuting in the San Francisco Bay Area

    Transportation

    (1996)
  • R. Cervero et al.

    Residential Self Selection and Rail Commuting: A Nested Logit Analysis

    (2002)
  • R. Crane

    The influence of urban form on travel: an interpretive review

    Journal of Planning Literature

    (2000)
  • Curry, R.W., Jr., 2000. Attitudes toward travel: the relationships among perceived mobility, travel liking, and...
  • R.M. Feldman

    Settlement-identity: psychological bonds with home places in a mobile society

    Environment and Behavior

    (1990)
  • Cited by (394)

    • A gendered perspective on ride-hail use in Los Angeles, USA

      2024, Transportation Research Interdisciplinary Perspectives
    • Commuting and its spillover effects on subjective well-being: Evidence from China

      2024, Transportation Research Part D: Transport and Environment
    • Role of income on travel behavior in polluted air

      2023, Journal of Transport and Health
    View all citing articles on Scopus
    View full text