The role of the built environment in travel behavior is one of the major areas of research in urban planning. Previous experiments in built environment studies have shown that dense, well-connected and diverse environments influence people's travel behaviour, especially when walking. This study aims to further investigate the impact of such built environments in an attempt to find the level of the built environment known to influence travel behaviour.
The travel pattern of the elderly and non-elderly by the travel distance and the travel mode choice by the neighborhood walkability are also studied. The 2012 Seoul Metropolitan Household Travel Survey was used to analyze the travel pattern of individuals in their travel-origin neighborhood. The findings suggest that (1) there is marginal effect of built environments on travel behaviour; (2) elderly and non-elderly have different travel mode choices for commuting and shopping trips; (3) the elderly are more sensitive to the neighborhood's walkability, that the elderly in the walkable neighborhood are more likely to walk and take public transport than the non-elderly in the walkable neighborhood.
My thesis suggests that the role of built environments on travel behavior actually has a positive impact on individuals' travel behavior by promoting the likelihood of active travel. However, contrary to the previous beliefs that the increased level of built environment would lead to correspondingly increased active travel, the built environment has its limitations on the ability to encourage active travel.
INTRODUCTION
Exploring Thresholds of Built Environment Characteristics for Walkable Communities: Empirical
- INTRODUCTION
- METHODS
- ANALYSIS
- RESULTS
- DISCUSSION
- CONCLUSIONS
In one of the few studies that examined the effects of the built environment on walking, Frank and Pivo (1994) showed that the relationship between population and employment density and preferred modes of travel (ie, private car, public transit, or walking) is non-linear. Several studies conducted in densely populated Asian cities, meanwhile, have provided somewhat mixed support for the association between walking and characteristics of the built environment. Overall, previous findings suggest that the effects of the built environment still apply to the walkability of highly densely populated Asian cities (Jun et al., 2013; Sung et al., 2013), but may not be as apparent as they are in North American cities.
In short, the objectives of this study were (1) to empirically test whether a nonlinear relationship exists between the built environment and walking and (2) to identify the building threshold. Our study region, the Seoul metropolitan area, has one of the highest densities in the world and, we believe, sufficient variation in density and diversity to explore the optimal density of the built environment. To quantify the built environment characteristics of respondents' neighborhoods, each individual's home location was identified using the neighborhood code reported during home visits.
To measure each dimension of the built environment, we used the variables of (1) population density, (2) the Herfindahl-Hirschman Index (HHI), and (3) road connectivity. To examine the nonlinear effects of the built environment, 15 factors of the built environment were presented in the regression analysis with the introduction of 14 dummy regressors. Three multilevel logistic models were developed to account for three travel modes—walking, public transport, and private vehicle use—and the impact of the built environment on the occurrence of each mode was compared.
The average gross population density of the neighborhoods was 18,614.6 persons/km2 (standard deviation = 13,496 persons/km2). This figure shows that the population of the Seoul metropolitan area is one of the densest in the world, comparable to that of Mumbai and Calcutta in India, Shenzhen in China and Taipei in Taiwan. We attempted to determine the threshold at which features of the built environment promote walking based on the assumption that an optimal level of density, road connectivity, and land use mix can exist within the context of the Seoul Metropolitan Area.
Consistent with previous findings, the findings of our statistical modeling revealed that the three dimensions of the built environment and a walkability measure combining the three dimensions are positively associated with the number and duration of walking trips and a reduction in private vehicle rides. The density of the densest area in the USA—Manhattan—is approximately 27,135 persons/km2, and that of Brooklyn and the Bronx in New York City are 13,168 and 12,853 persons/km2, respectively. Thus, the threshold of population density that we have identified is similar to the density of the densest.
A previous study reported that Korean elderly's walking distance is not shorter or smaller than non-elderly's (Choo et al., 2013). Perhaps the small proportion of the population engaged in recreation is a unique feature of the Seoul metropolitan area.
Impact of community walkability on travel mode choice for commuting and shopping
- METHODS
- RESULTS
- DISCUSSION
- CONCLUSION
Form of travel between elderly and non-elderly Table 3.2 Share of each form of travel by purpose of travel. The travel mode choices of the elderly and non-elderly show that the propensity to walk and use a private vehicle is very different depending on the purpose of travel and age group. The elderly were more likely to walk shorter distances compared to the non-elderly.
While the odds of the elderly taking public transport were highest for D10, the non-elderly showed a higher propensity to take public transport for shorter distances (D8, D9) than D10. As can be seen, the elderly and non-elderly showed different propensities for private vehicle use. The travel patterns of the elderly and non-elderly are generally similar for walking and taking public transport, but different for taking private vehicles.
The choice of travel mode between walking and private vehicle use and between public transport and private vehicle use by the elderly and non-elderly is presented in Table 3.4. It shows that in all distance groups, both elderly and non-elderly have a trend towards walking over private vehicle use compared to D10. This finding shows that for trips of the same distance, the elderly are more likely to choose walking than the use of private vehicles compared to the non-elderly.
None of the non-elderly groups showed a higher tendency towards walking or public transport compared to the highly walkable group (Y4). In terms of destination neighborhood walkability for commuting trips, older adults in all neighborhood groups have higher odds of walking and taking public transportation compared to non-elderly in highly walkable neighborhoods (Y4 ). While seniors in non-walkable (Y2) and walkable (Y3) neighborhoods have lower odds of walking than those in highly walkable neighborhoods (Y4), the odds are similar.
Analysis of the walkability of the destination neighborhood for shopping trips revealed that except for the elderly and non-elderly in non-walkable (Y1 and O1) neighborhoods, all other groups were likely to walk more than the group in the very walkable (Y4) ) neighborhood. While the odds of taking public transport are higher for the elderly, the travel patterns of the elderly and non-elderly are similar, especially for walking. Both the elderly and non-elderly groups in very non-walkable (Y1 and O1) and non-walkable (Y2 and O2) neighborhoods have a lower chance of taking public transport.
The research results show that elderly and non-elderly people have different travel patterns for the same travel distances. This study examined the mode of transport choice of elderly and non-elderly people based on travel distances and walkability level in the neighborhood.
CONCLUSION
International variation in neighborhood walkability, transit, and recreational environments using geographic information systems: The IPEN adult study. Aging in an ultra-dense metropolis: perceived neighborhood characteristics and utilitarian walking among elderly people in Hong Kong. Analysis of changes in elderly people's travel behavior using travel diary survey data in Seoul metropolitan area.
Small area and individual-level predictors of physical activity in urban communities: a multilevel study in Stoke on Trent, England. Impacts of mixed use and density on the use of three travel modes: single-occupant vehicle, transit, and walking. Neighborhood built environment and physical activity of Japanese elderly: results from the Aichi Gerontological Evaluation Study (AGES).
Minnesota Department of Transportation Research Services Division, Report #4 in the series: Access to Destinations Study. Public open space and walking The role of proximity, perceptual characteristics of the surrounding built environment, and street configuration. A Longitudinal Analysis of the Influence of the Neighborhood Built Environment on Walking for Transport The RESIDE Study.
The study of the problem and the way to improve the development of the new city of the capital. Transit-oriented development in a high-density city: Identifying its relationship with transit transportation in Seoul, Korea. Neighborhood walkability, physical activity and walking behaviour: the Swedish Neighborhood and Physical Activity (SNAP) study.
Perceived neighborhood environmental attributes associated with adults' transportation-related walking and cycling: Findings from the USA, Australia, and Belgium. Assessing personal and built environment attributes to physical activity: a multilevel analysis on multiple sources of population-based data.
