An integrated assessment of factors affecting modal choice: towards a better understanding of the causal effects of built environment

Ramezani, Pizzo, Deakin, 2018, in Transportation

Location	Rome, Italy
Population	General
Sample size	524
Factor analysis type	exploratory factor analysis, varimax rotation
Stepwise regression	yes
Removal of insignificant variables	yes
Reviewed by	LCM

Abstract

This paper discusses the methodological challenges in understanding causal relationships between urban form and travel behavior and uses a holistic quasi-experimental approach to investigate the separable marginal influence of each of several urban form factors on mode choice as well as the complex relationships between those factors and a wide range of personal traits. Data analysis and models are used to reveal the effect of such interactions on mode choice for both work and non-work trips in Rome, Italy. It is found that population density does not have a significant marginal positive effect on sustainable mode choice for work trips. Conversely, this factor decreases sustainable mode choice for non-work trips. Small scale street design quality alone increases sustainable mode choice for non-work trips. This is while presence of street network integration alone increases automobile use for all trip purposes. The results point to the importance of incorporating all the urban form factors of diversity, design and street network integration if the goal is to increase the use of more sustainable modes of transportation for both work and non-work trips, but also show that attitudes and preferences can modify the response to urban design factors. The findings suggest that thoughtful policies triggering certain attitudes (cost sensitivity, sensitivity to peer pressure regarding the value attributed to sustainable transportation, and transit preference) can be adopted to significantly increase sustainable mode choice even in the neighborhoods with specific physical restrictions. © 2017, Springer Science+Business Media New York.

Factors

Variable	Pattern loading
I need a car to do many of the things I like to do (Convenience)	0.73
I like driving (Mode-liking)	0.63

Variable	Pattern loading
Local shops (shopping areas) within walking distance (Res preference/perception)	0.92
Easy access to a shopping center (Res preference/perception)	0.59

Variable	Pattern loading
Other amenities/services available nearby (Res preference/perception)	0.86
Easy access to city center (Res preference/perception)	0.45
Parks and open spaces available nearby (REs preference/perception)	0.59

Variable	Pattern loading
Easy access to main routes (Res preference/perception)	0.78
Low level of car traffic on neighborhood streets (Res preference/perception)	0.69
Safe neighborhood for walking (low crash risk for pedestrian) (Safety)	0.58
Quiet neighborhood (Res preference/perception)	0.52

Variable	Pattern loading
I prefer to organize my errands so that I make as few trips as possible (Convenience)	0.5
I often use the telephone or the Internet to avoid having to travel somewhere (Lifestyle)	0.87

Variable	Pattern loading
The only good thing about travelling is arriving at your destination (Convenience)	0.77
Travel time is generally wasted time (Convenience)	0.63

Variable	Pattern loading
Traveling by car is safer overall than walking (Safety)	0.79
Traveling by car is safer overall than taking transit (Safety)	0.48
I like to be able to shop or run errands while travelling (Convenience)	0.69

Variable	Pattern loading
I like to be able to leave/collect children at school or similar while travelling (Convenience)	0.98
I like to be able to give children a ride to their leisure time activities while travelling (Convenience)	0.85

Variable	Pattern loading
Fuel price affects my choice of daily travel by car (Cost)	0.77
Transit fare affects my choice of daily travel by metro/bus (Cost)	0.76
Price of parking affects my choice of daily travel by private car (Cost)	0.68

Variable	Pattern loading
When I need to buy something, I usually prefer to get it at the closest store possible (Convenience)	0.8
I don't like travelling long distances (Lifestyle)	0.73

Variable	Pattern loading
Diverse neighbors in terms of ethnicity, race, and/or age (Res preference/perception)	0.7
Neighborhood is frequented by many people from other neighborhoods (perhaps the neighboring districts) for the quality of spaces and services it offers (Res preference/perception)	0.48
Good views (Res preference/perception)	0.79
Variety in housing styles (Res preference/perception)	0.59

Variable	Pattern loading
I try to limit my driving to help improve air quality (Environmentalism)	0.45
Vehicles should be taxed on the basis of the amount of pollution they produce (Transport policy)	0.51
Changing how people travel is a great way to improve the environment (Environmentalism)	0.56

Variable	Pattern loading
I like riding a bike (Mode-liking)	0.46
I prefer to bike rather than drive whenever possible (Mode-liking)	0.52
Biking can sometimes be easier for me than driving (Convenience)	0.96

Variable	Pattern loading
Getting to work without a car is a hassle (Convenience)	0.86
The region needs to build more highways to reduce traffic congestion (Transport policy)	0.81

Variable	Pattern loading
I like taking transit (Mode-liking)	0.65
I prefer to take transit rather then drive whenever possible (Mode-liking)	0.61
I like to avoid queues and congestion while travelling (Convenience)	0.8

Variable	Pattern loading
Walking can sometimes be easier for me than driving (Convenience)	0.52
I like travelling in a calm, non-noisy environment (Convenience)	0.63

Variable	Pattern loading
Safe neighborhood for children to play outdoors (low crash risk for children) (Safety)	0.86
Secure neighborhood (low crime risk) (Safety)	0.92

Variable	Pattern loading
Lots of people out and about within the neighborhood (Res preference/perception)	0.57
Lots of interaction among neighbors (Res preference/perception)	0.7
Economic level of neighbors similar to me (Res preference/perception)	0.5

Variable	Pattern loading
Pavement (easy walking routes) (Res preference/perception)	0.54
Attractive appearance of neighborhood (Res preference/perception)	0.74
Adequate parking space (Res preference/perception)	0.74

Variable	Pattern loading
People in my neighborhood have a positive view of people who drive (Social norms)	0.89
I drive to impress others (Social norms)	0.69

Variable	Pattern loading
People in my neighborhood have a positive view of people who walk (Social norms)	0.78
People in my neighborhood have a positive view of people who bicycle (Social norms)	0.85
People in my neighborhood have a positive view of people who use public transport (Social norms)	0.45

Variable	Pattern loading
I do not like to wait for another travel mode while travelling (Convenience)	0.73
I don't like to have variation in my daily travel time (Lifestyle)	0.69

Variable	Pattern loading
Easy access to public transport service (Res preference/perception)	0.48
Sidewalks throughout the neighborhood (Res preference/perception)	0.58
Good street lighting (Safety)	0.55
Tree lined street (Res preference/perception)	0.71
Flat terrain (Convenience)	0.66

Models

Dependent variable	Mode choice for work trip
Model type	Multinomial logit
Sample size	524.0
R²	0.679
Adjusted R²
Pseudo R² (nan)	nan
AIC	nan
BIC	nan
Log-likelihood at zero	-571.618
Log-likelihood at constants	nan
Log-likelihood at convergence	-141.593

Car
Variable	Coefficient	p-value
Travel time (h)	-4.79	0
Travel cost (€)	-0.472	0.01

Transit
Variable	Coefficient	p-value
Alternative-specific constant	-1.7	0.03
Travel time (h)	-4.79	0
Travel cost (€)	-0.472	0.01
Travel distance by car	0.388	0
Number of stops	0.432	0.14
Peak hour	0.961	0.03
Age 18-34	-0.991	0.01
Age > 64	1.41	0.01
Education	0.484	0.03
Disability	-2.57	0.02
Owning a transit pass	1.73	0
Cost sensitivity	0.359	0.11
Pro-transit	0.731	0
Susceptibility to peer pressure (sustainable transportation)	0.212	0.02
Car safety and flexibility consideration	-0.646	0
Pro-biking	0.915	0
Anti-travel 1	0.66	0
Environmental accountability	0.099	0.15
Susceptibility to peer pressure (driving)	-0.59	0
Distance sensitivity	0.801	0
Absolute driver	-0.937	0
Diversity in building style and socializing opportunities	0.819	0
Speciousness and attractiveness	-0.419	0
Street T1	3.72	0
Street T2	3.67	0.01
Street T3	2.35	0
Street T9	2.95	0
Pro-transit*T1	0.591	0
Speciousness and attractiveness*T2	-3.01	0
Cost sensitivity*T2	3.42	0.01
Car safety and flexibility consideration*T2	1.73	0.03
Speciousness and attractiveness*T3	-0.87	0
Car safety and flexibility consideration*T3	0.565	0
Car safety and flexibility consideration*T4	0.843	0.01
Pro-transit*T9	0.365	0.01

Walk
Variable	Coefficient	p-value
Alternative-specific constant	-0.84	0.39
Travel time (h)	-4.79	0
Female	-1.21	0.02
Education	0.487	0.03
Owning a transit pass	1.73	0
Cost sensitivity	0.359	0.11
Pro-transit	0.731	0
Car safety and flexibility consideration	-0.646	0
Pro-biking	0.915	0
Anti-travel 1	0.66	0
Susceptibility to peer pressure (driving)	-0.497	0
Distance sensitivity	0.801	0
Absolute driver	-0.937	0
Walkability of neighborhood	0.289	0.01
Diversity in building style and socializing opportunities	0.819	0
Speciousness and attractiveness	-0.419	0
Street T1	3.72	0
Street T2	3.67	0
Street T3	2.35	0
Street T9	2.95	0
Pro-transit*T1	0.591	0
Speciousness and attractiveness*T2	-3.01	0
Cost sensitivity*T2	3.42	0.01
Car safety and flexibility consideration*T2	1.73	0.03
Speciousness and attractiveness*T3	-0.87	0
Car safety and flexibility consideration*T3	0.565	0
Car safety and flexibility consideration*T4	0.843	0.01
Walkability of neighborhood*T5	1.33	0
Pro-transit*T5	0.394	0.02
Pro-transit*T9	0.365	0.01

Dependent variable	Mode choice for non-work trip
Model type	Multinomial logit
Sample size	524.0
R²	0.777
Adjusted R²
Pseudo R² (nan)	nan
AIC	nan
BIC	nan
Log-likelihood at zero	-571.618
Log-likelihood at constants	nan
Log-likelihood at convergence	-80.438

Car
Variable	Coefficient	p-value
Travel time (h)	-11.0	0
Travel cost (€)	-0.712	0.01

Transit
Variable	Coefficient	p-value
Alternative-specific constant	-3.71	0.01
Travel time (h)	-11.0	0
Travel cost (€)	-0.712	0.01
Recreational purposes	2.04	0.01
Income < 50,000 €	3.25	0
Single household	-2.12	0.1
Disability	-4.11	0.03
Owning a transit pass	1.68	0
Cost sensitivity	0.502	0
Pro-biking	1.48	0.01
Waiting time/time sensitivity	-0.145	0.16
Susceptibility to peer pressure (driving)	-0.76	0
Speciousness and attractiveness	-0.832	0
Street T1	5.37	0
Street T2	3.02	0.01
Street T3	5.19	0
Street T4	2.31	0.08
Street T7	1.97	0.17
Street T9	5.35	0
Pro-biking*T1	1.47	0.01
Susceptibility to peer pressure (sustainable transportation)*T1	1.16	0.03
Pro-biking*T2	1.71	0.01
Susceptibility to peer pressure (sustainable transportation)*T3	1.67	0
Pro-biking*T3	3.25	0
Pro-biking*T4	3.19	0
Cost sensitivity*T5	2.68	0
Susceptibility to peer pressure (sustainable transportation)*T5	1.52	0.02
Pro-biking*T7	1.2	0.06
Safety and security*T8	1.56	0.04
Pro-biking*T8	1.77	0.06
Pro-biking*T9	2.0	0.01

Walk
Variable	Coefficient	p-value
Alternative-specific constant	-1.07	0.55
Travel time (h)	-11.0	0
Recreational purposes	2.04	0.01
Female	-3.76	0
Age > 64	-2.2	0.11
Income < 50,000 €	3.25	0
Single household	-2.12	0.1
Disability	-4.11	0.03
Owning a transit pass	1.68	0
Pro-biking	1.48	0.01
Susceptibility to peer pressure (driving)	-1.25	0
Walkability of neighborhood	3.56	0
Diversity in building style and socializing opportunities	2.48	0
Accessibility and safety of neighborhood streets	1.6	0
Social interaction	3.48	0
Speciousness and attractiveness	-1.37	0
Safety and security	0.828	0
Street T1	5.37	0
Street T2	3.02	0.01
Street T3	5.19	0
Street T4	2.31	0.08
Street T7	1.97	0.17
Street T9	5.35	0
Pro-biking*T1	1.47	0.01
Pro-biking*T2	1.71	0.01
Pro-biking*T3	3.25	0
Pro-biking*T4	3.19	0
Diversity in building style and socializing opportunities*T5	-5.82	0
Social interaction*T5	-2.76	0
Social interaction*T7	-1.66	0.01
Pro-biking*T7	1.2	0.06
Safety and security*T8	2.98	0
Pro-biking*T8	1.77	0.06
Accessibility and safety of neighborhood streets*T9	5.93	0
Pro-biking*T9	2.0	0.01