Joint analysis of the spatial impacts of built environment on car ownership and travel mode choice

Concerns over transportation energy consumption and emissions have prompted more studies into the impacts of built environment on driving-related behavior, especially on car ownership and travel mode choice. This study contributes to examine the impacts of the built environment on commuter’s driving behavior at both spatial zone and individual levels. The aim of this study is threefold. First, a multilevel integrated multinomial logit (MNL) and structural equation model (SEM) approach was employed to jointly explore the impacts of the built environment on car ownership and travel mode choice. Second, the spatial context in which individuals make the travel decisions was accommodated, and spatial heterogeneities of car ownership and travel mode choice across traffic analysis zones (TAZs) were recognized. Third, the indirect effects of the built environment on travel mode choice through the mediating variable car ownership were calculated, in other words, the intermediary nature of car ownership was considered. Using the Washington metropolitan area as the study case, the built environment measures were calculated for each TAZ, and the commuting trips were drawn from the household travel survey in this area. To estimate the model parameters, the robust maximum likelihood (MLR) method was used. Meanwhile, a comparison among different model structures was conducted. The model results suggest that application of the multilevel integrated MNL and SEM approach obtains significant improvements over other models. This study give transportation planners a better understanding on how the built environment influences car ownership and commuting mode choice, and consequently develop effective and targeted countermeasures.     

Transportation carbon dioxide emissions by built environment and family lifecycle: Case study of the Osaka metropolitan area

How a city grows and changes, along with where people choose to live likely affects travel behavior, and thus the amount of transportation CO₂ emissions that they produce. People generally go through different stages in their life, and different travel needs are associated with each. The impact of the built environment may vary depending on the lifecycle stage, and the years spent at each stage will differ. A family with children may last for twenty to thirty years, while the time spent without dependents might be short in comparison. Over a family’s lifecycle, how big of a difference might the built environment, through household location choice, have on the amount of transportation CO₂ emissions produced? From a climate change perspective, how significant is residential location on the CO₂ produced by transportation use? This paper uses data from the Osaka metropolitan area to compare the direct transportation CO₂ emissions produced over a family’s lifecycle across five different built environments to determine whether any are sustainable and which lifecycle stage has the greatest overall emissions. This understanding would enable the design of a targeted policy based on household lifecycle to reduce overall transportation CO₂ of individuals throughout one’s lifecycle. The yearly average per-capita family lifetime transportation CO₂ emissions were 0.25, 0.35, 0.58, 0.78, and 0.79 metric tonnes for the commercial, mixed-commercial, mixed-residential, autonomous, and rural areas respectively. The results show that only the commercial and mixed-commercial areas were considered to be sustainable from a climate change and transportation perspective.     

Walking frequency, cars, dogs, and the built environment

To explain walking propensity or frequency, empirical studies have generally used two sets of explanatory variables, namely, socio-demographic variables and built environment variables. They have generally shown that both socio-demographic characteristics and built environment characteristics are associated with walking propensity. We examine the traditional walkability variables that encompass density, mix of uses, and network connectivity in New Jersey, using a statewide sample including an oversample of Jersey City. We estimate a two-stage least squares model using a conditional mixed process that combines an ordered probit model of walking frequency in the second stage based on a truncated regression of car ownership in the first stage. Our results show that built environment variables have some small effects, mainly from better network connectivity associated with increased walking frequency. One of our key findings is that built environment features also work indirectly via how they influence car ownership. In general, we find sufficient evidence that suggests fewer cars are owned in areas with more walkable built environment features. The other key variable that we control for is whether a household owns a dog. This also proved to be strongly associated with walking suggesting that dog ownership is a necessary control variable to understand the frequency of walking.     

Modeling residential sorting effects to understand the impact of the built environment on commute mode choice

This paper presents an examination of the significance of residential sorting or self selection effects in understanding the impacts of the built environment on travel choices. Land use and transportation system attributes are often treated as exogenous variables in models of travel behavior. Such models ignore the potential self selection processes that may be at play wherein households and individuals choose to locate in areas or built environments that are consistent with their lifestyle and transportation preferences, attitudes, and values. In this paper, a simultaneous model of residential location choice and commute mode choice that accounts for both observed and unobserved taste variations that may contribute to residential self selection is estimated on a survey sample extracted from the 2000 San Francisco Bay Area household travel survey. Model results show that both observed and unobserved residential self selection effects do exist; however, even after accounting for these effects, it is found that built environment attributes can indeed significantly impact commute mode choice behavior. The paper concludes with a discussion of the implications of the model findings for policy planning.

Exploring the influence of built environment on tour-based commuter mode choice: A cross-classified multilevel modeling approach

Understanding travel behavior and its relationship to urban form is vital for the sustainable planning strategies aimed at automobile dependency reduction. The objective of this study is twofold. First, this research provides additional insights to examine the effects of built environment factors measured at both home location and workplace on tour-based mode choice behavior. Second, a cross-classified multilevel probit model using Bayesian approach is employed to accommodate the spatial context in which individuals make travel decisions. Using Washington, D.C. as our study area, the home-based work (Home-work) tour in the AM peak hours is used as the analysis unit. The empirical data was gathered from the Washington-Baltimore Regional Household Travel Survey 2007–2008. For parameter estimation, Bayesian estimation method integrating Markov Chain Monte Carlo (MCMC) sampling is adopted. Our findings confirmed the important role that the built environment at both home location and work ends plays in affecting commuter mode choice behavior. Meanwhile, a comparison of different model results shows that the cross-classified multilevel probit model offers significant improvements over the traditional probit model. The results are expected to give a better understanding on the relationship between the built environment and commuter mode choice behavior.     

Breaking car use habits: The effectiveness of a free one-month travelcard

Based on calls for innovative ways of reducing car traffic and research indicating that car driving is often the result of habitual decision-making and choice processes, this paper reports on a field experiment designed to test a tool aimed to entice drivers to skip the habitual choice of the car and consider using—or at least trying—public transport instead. About 1,000 car drivers participated in the experiment either as experimental subjects, receiving a free one-month travelcard, or as control subjects. As predicted, the intervention had a significant impact on drivers’ use of public transport and it also neutralized the impact of car driving habits on mode choice. However, in the longer run (i.e., four months after the experiment) experimental subjects did not use public transport more than control subjects. Hence, it seems that although many car drivers choose travel mode habitually, their final choice is consistent with their informed preferences, given the current price–quality relationships of the various options.     

The effect of prices and income on car travel in the UK

The objective of this paper is to analyse the factors determining household car travel, and specifically the effects of household income and the prices of cars and motor fuels, and to explore the intertemporal pattern of adjustment. The question of asymmetry in the response to rising and falling income is also addressed. Such asymmetry may be caused by habit or resistance to change or the tendency to acquire habits to consume more easily than to abandon them. The impact of prices, the speed of adjustment and the resistance to change will be important in determining the possibility of influencing travel behaviour and specifically car use. The study utilises repeated cross-section data from the annual UK Family Expenditure Surveys and employs a pseudo-panel methodology. The results are compared with those for car ownership estimated on the basis of similar models.     

Exploring built environment impacts on transit use – an updated meta-analysis

ABSTRACT

The built environment (BE) is widely accepted to influence transit use (TU). Evidence to date suggests the relationship is dependent on many factors which can be difficult to account for in quantitative studies. This creates barriers to transferring research into practice. Considering many studies together can be useful for accounting for more of the factors impacting transit use. Yet, meta-analysis of research measuring these influences was last undertaken in 2010 based on 18 studies. Since then 90 new quantitative studies have been published. These recent studies use improved methodologies and are conducted in more diverse geographies. This paper reports an improved and updated meta-analysis of built environment impacts on transit use. It compares elasticity estimates from research published pre-and post-2010 and explores the impact of new methods and a more diverse geographical representation on findings. Updated meta-elasticities range from <0.01 to 0.26; a similar range to the 2010 study. However, at the individual indicator levels, more recent results are different. Elasticities for urban density, including population, employment and commercial density, have increased significantly in studies published since 2010, as did that of land use mix. However, measures of local access, design and jobs-housing balance decreased in post-2010 studies. These results confirm the small but imprecise relationship between the BE and TU. Results also suggest that while the range of elasticity impacts is relatively consistent, new study methodologies, notably those that control for regional accessibility and self-selection, and the increasing geographical diversity in study applications, is acting to change BE-TU findings at the indicator level. Research setting and context are important to consider when using empirical results to design BE strategies to promote transit use.     

The influences of past and present residential locations on vehicle ownership decisions

This study explores the relationship between historical exposure to the built environment and current vehicle ownership patterns. The influence of past exposure to the built environment on current vehicle ownership decisions may be causal, but there are alternative explanations. Households may primarily select to live in neighborhoods that facilitate their vehicle ownership preferences, or they may retain preferences that they have developed in the past, irrespective of their current situations. This study seeks to control for these alternative explanations by including the built environment attributes of households’ past residences as an influence on vehicle ownership choices. We use a dataset from a credit reporting firm that contains up to nine previous residential ZIP codes for households currently living in the 13-county Atlanta, Georgia, metropolitan area. Results show that past location is significant, but of marginal influence relative to the attributes of the current location. From a practical perspective, our results suggest that models that include current but not past neighborhood attributes (also controlling for standard socioeconomic variables) can forecast vehicle ownership decisions reasonably well. However, models that include both current and past neighborhood attributes can provide a more nuanced understanding of the built environment’s potentially causal influences on vehicle ownership decisions. This better understanding may provide more realistic forecasts of responses to densification or other travel demand management strategies.     

Examining the effects of the built environment and residential self-selection on commuting trips and the related CO2 emissions: An empirical study in Guangzhou,China

Numerous studies have established the link between the built environment and travel behavior. However, fewer studies have focused on environmental costs of travel (such as CO₂ emissions) with respect to residential self-selection. Combined with the application of TIQS (Travel Intelligent Query System), this study develops a structural equations model (SEM) to examine the effects of the built environment and residential self-selection on commuting trips and their related CO₂ emissions using data from 2015 in Guangzhou, China. The results demonstrate that the effect of residential self-selection also exists in Chinese cities, influencing residents’ choice of living environments and ultimately affecting their commute trip CO₂ emissions. After controlling for the effect of residential self-selection, built environment variables still have significant effects on CO₂ emissions from commuting although some are indirect effects that work through mediating variables (car ownership and commuting trip distance). Specifically, CO₂ emissions are negatively affected by land-use mix, residential density, metro station density and road network density. Conversely, bus stop density, distance to city centers and parking availability near the workplace have positive effects on CO₂ emissions. To promote low carbon travel, intervention on the built environment would be effective and necessary.     

Spatio-temporal analysis of car distance,greenhouse gases and the effect of built environment: A latent class regression analysis

This work examines the temporal–spatial variations of daily automobile distance traveled and greenhouse gas emissions (GHGs) and their association with built environment attributes and household socio-demographics. A GHGs household inventory is determined using link-level average speeds for a large and representative sample of households in three origin–destination surveys (1998, 2003 and 2008) in Montreal, Canada. For the emission inventories, different sources of data are combined including link-level average speeds in the network, vehicle occupancy levels and fuel consumption characteristics of the vehicle fleet. Urban form indicators over time such as population density, land use mix and transit accessibility are generated for each household in each of the three waves. A latent class (LC) regression modeling framework is then implemented to investigate the association of built environment and socio-demographics with GHGs and automobile distance traveled. Among other results, it is found that population density, transit accessibility and land-use mix have small but statistically significant negative impact on GHGs and car usage. Despite that this is in accordance with past studies, the estimated elasticities are greater than those reported in the literature for North American cities. Moreover, different household subpopulations are identified in which the effect of built environment varies significantly. Also, a reduction of the average GHGs at the household level is observed over time. According to our estimates, households produced 15% and 10% more GHGs in 1998 and 2003 respectively, compared to 2008. This reduction can be associated to the improvement of the fuel economy of vehicle fleet and the decrease of motor-vehicle usage – e.g., a decrease of 4% is observed for fuel efficiency rates and 12% for distance according to the raw average estimates from 1998 with respect to 2008. A strong link is also observed between socio-demographics and the two travel outcomes. While number of workers is positively associated with car distance and GHGs, low and medium income households pollute less than high-income households.     

The impacts of built environment on home-based work and non-work trips: An empirical study from Iran

This paper aims to explore the impact of built environment attributes in the scale of one quarter-mile buffers on individuals’ travel behaviors in the metropolitan of Shiraz, Iran. In order to develop this topic, the present research is developed through the analysis of a dataset collected from residents of 22 neighborhoods with variety of land use features. Using household survey on daily activities, this study investigates home-based work and non-work (HBW and HBN) trips. Structural equation models are utilized to examine the relationships between land use attributes and travel behavior while taking into account socio-economic characteristics as the residential self-selection. Results from models indicate that individuals residing in areas with high residential and job density, and shorter distance to sub-centers are more interested in using transit and non-motorized modes. Moreover, residents of neighborhoods with mixed land uses tend to travel less by car and more by transit and non-motorized modes to non-work destinations. Nevertheless, the influences of design measurements such as street density and internal connectivity are mixed in our models. Although higher internal connectivity leads to more transit and non-motorized trips in HBW model, the impacts of design measurements on individuals travel behavior in HBN model are significantly in contrast with research hypothesis. Our study also shows the importance of individuals’ self-selection impacts on travel behaviors; individuals with special socio-demographic attributes live in the neighborhoods with regard to their transportation patterns. The findings of this paper reveal that the effects of built environment attributes on travel behavior in origins of trips do not exactly correspond with the expected predictions, when it comes in practice in a various study context. This study displays the necessity of regarding local conditions of urban areas and the inherent differences between travel destinations in integrating land use and transportation planning.     

Built environment effects on the integration of dockless bike-sharing and the metro

Bike-sharing provides a convenient feeder mode for connecting to a metro and is believed to be an efficient way to solve the first- and last-mile problem. Despite the increasing attention paid on the use of bike-sharing, few studies have investigated how built environment factors affect the integrated use of dockless bike-sharing (DBS) and the metro. Using data from one of the largest DBS operators in China (Ofo), this paper employed a series of negative binomial regressions to examine the effect of the built environment on the integrated use of DBS and the metro, using Shenzhen as a case study. The findings show that mixed land use is positively related to integrated use. Residential areas have higher access-integrated rates during the morning peak hours, while industrial areas are associated with more integrated uses, connecting factories and metro stations. Furthermore, parks and public squares encourage both access- and egress-integrated use during peak times. Transportation facility features, including bus stops and dedicated bike lanes, are positively related to integrated use, while areas with dense metro distribution and main streets with many intersections are negatively related. Transfer distance also plays a crucial and negative role in integrated use. In addition, metro stations that are closer to the city center with a higher number of passengers are more likely to be integrated with bike-sharing. These findings can be used to collectively facilitate a connection between cycling and metro transit by creating a bicycle-friendly environment.     

The relationship between instrumental, symbolic and affective factors as predictors of car use: A structural equation modeling approach

Traditionally, urban mobility has been studied from the utilitarian or practical viewpoint, focusing on instrumental motivations and ignoring symbolic and affective aspects that may play a relevant role. The purpose of this work is to analyze from a psychosocial perspective the influence of symbolic, affective, and instrumental motivations on the frequency of car use, taking into account diverse reasons for traveling. From a sample of the Spanish population, participants were 284 people (50.3% female), with a driver’s license, car owners and residents in cities of various sizes, who completed an anonymous questionnaire. The effect of each type of variable was estimated by a structural equation model. Results indicate that people’s affective link with their private vehicle explains 12% of frequency of car use, as a latent variable of different kinds of trips: visiting friends or relatives, going to work or to a study center, going shopping, or to leisure areas. The instrumental advantages associated with cars and thinking that it expresses one’s status predict the affective link with the car. These findings corroborate the relevance of the non-instrumental aspects involved in the selection of the means of transportation.     

Influences of built environment characteristics and individual factors on commuting distance: A multilevel mixture hazard modeling approach

Concerns over transportation energy consumption and green-household gas (GHG) emissions have prompted a growing body of research into the influence of built environment on travel behavior. Studies on the relationship between land use and travel behavior are often at a certain aggregated spatial unit such as traffic analysis zone (TAZ), spatial issues occur among individuals clustered within a zone because of the locational effects. However, recognition of the spatial issues in travel modeling was not sufficiently investigated yet. The object of this study is twofold. First, a multilevel hazard model was applied to accommodate the spatial context in which individuals generate commuting distance. Second, this research provides additional insights into examine the effects of socio-demographics and built environment on commuting distance. Using Washington metropolitan area as the case, the built environment measures were calculated for each TAZ. To estimate the model parameters, the robust maximum likelihood estimation method for a partial function was used, and the model results confirmed the important roles that played by the TAZ and individual level factors in influencing commuting distance. Meanwhile, a comparison among the general multilevel model, single level and multilevel hazard models was conducted. The results suggest that application of the multilevel hazard-based model obtains significant improvements over traditional model. The significant spatial heterogeneity parameter indicates that it is necessary to accommodate the spatial issues in the context of commuting distance. The results are expected to give urban planners a better understanding on how the TAZ and individual level factors influence the commuting distance, and consequently develop targeted countermeasures.     

The residential parking rent price elasticity of car ownership in Japan

By using household-level micro data captured through the National Survey of Family Income and Expenditure for 2004, this study evaluates the residential parking rent price elasticity of car ownership in Japan. It analyzes the number of cars owned by a household, using various attributes including expenditure for renting a parking space on a monthly basis. The estimation results derived from the IV-ordered probit model show that the absolute value of parking rent price elasticity of car ownership is, at most, 0.48, which is fairly small (i.e., inelastic). The elasticity value varies depending on city size; for megacities, elasticity is always negative for car ownership, whereas for middle-sized or small cities, towns, and villages, elasticity is positive for one-car ownership and negative for the ownership of more than one car. Hence, when the price of parking increases, some people may switch from more than one car to one car and some people in megacities may switch from one to zero cars. Indeed, the net effect of a price increase may be that non-car ownership increases in megacities and one-car ownership increases in other cities.     

Young people and the private car: A love-hate relationship

Urban mobility is one of the main concerns of the public authorities in developed countries. In France, household travel surveys are conducted every ten years in major cities to gather weekday mobility data. They enable decision-makers to better understand travel patterns, their change and their determinants, in order to adapt transport infrastructures to the population′s needs. While the automobile has allowed the level of mobility to increase since 1950, an unexpected finding has emerged from recent surveys in most developed countries, namely that there has been a marked decline in car use. Analyses show that this trend is mainly because young adults (18–34 years old) are less likely to acquire a driver′s license. This paper tries to better understand the decrease in the rate of driver′s license holding among young adults in the Lyon conurbation and to quantify the impact of the main explanatory factors in a temporal perspective. It also aims to analyze the consequences of this trend on private car use as a driver for daily trips. It quantifies the influence of economic, socio-demographic and spatial factors on driver ′ s license holding and car use by considering the responses to the last three household travels surveys conducted in the Lyon conurbation area (1995, 2006 and 2015). The temporal dimension allows us to highlight a change in the relationship between young adults and the private car in the French context.     

Built environment and mode choice relationship for commute travel in the city of Rajkot,India

Metropolitan areas around the world are looking for sustainable strategies to reduce use of private automobiles, energy consumption and emissions, often achieved by built environment interventions that encourage use of sustainable modes of transport. This study contributes by providing the empirical evidence on the relation between built environment and mode choice in context of Indian city of Rajkot. Using personal interview data and data available from Rajkot Municipal Corporation it is observed that there is a strong tendency among Rajkot residents to preselect their residential location to suit their modal preferences. This is especially true for non-motorized transport users. Among the built environment variables, access to destination and land use related indicators also have significant influence on mode choice. The study Infers that the land use policy should focus on accessibility and mixing of diverse uses, and transport supply will have to be location based to support non-motorized and public transport travel.     

Individual transport emissions and the built environment: A structural equation modelling approach

Increasing CO₂ emissions from the transport sector have raised substantial concerns among researchers and policy makers. This research examines the impact of the built environment on individual transport emissions through two mediate variables, vehicle usage and vehicle type choice, within a structural equation modelling (SEM) framework. We find that new-urbanism-type built environment characteristics, including high density, mixed land use, good connectivity, and easy access to public transport systems help reduce transport CO₂ emissions. Such mitigating effect is achieved largely through the reduced vehicle miles travelled (VMT) and is enhanced slightly by the more efficient vehicles owned by individuals living in denser and more diverse neighborhoods, all else being equal. Our research findings provide some new evidence that supports land use policies as an effective strategy to reduce transport CO₂ emissions.     

The effect of location,mobility and socio-demographic factors on task and time allocation of households

This paper investigates the role of location factors in task and time allocation at the household level. It is hypothesized that, if time constraints are less binding as a result of living in an urban area or owning more cars, spouses engage more often and longer in out-of-home activities and schedule their activities more independently. The hypotheses are tested with logistic and Cox regression models of activity participation and time allocation on a data set collected in the Amsterdam–Utrecht region in the Netherlands. Results suggest that the hypotheses are supported with respect to specific household activity scheduling decisions.