Factors influencing bicycle use: A binary choice model with panel data

Barberan, De Abreu E Silva, Monzon, 2017, in Transportation Research Procedia

doi:10.1016/j.trpro.2017.12.097
Location Vitoria-Gasteiz, Spain
Population General
Sample size 1451
Factor analysis type exploratory factor analysis, Orthogonal rotation
Stepwise regression nan
Removal of insignificant variables nan
Reviewed by AR

Abstract

Cycling has been commonly neglected in urban transport planning. In the same fashion, there is a shortage of available data on cycling mobility, especially in countries with low rates of bicycle share. Nevertheless, a modal shift towards soft modes such as cycling appears to be one of the keys for progressing towards a sustainable urban mobility paradigm. Understanding the factors that influence bicycle choice is necessary for implementing efficient probike transport policies. This research identifies the main factors affecting bicycle choice for commuting. It analyses an ad-hoc panel survey conducted in Vitoria-Gasteiz, a medium-sized city in northern Spain where cycle rate has rocketed in few years. Data from commuters, either workers or students, were collected in 2012, 2013 and 2014. An unbalanced binary panel model includes both objective - such as gender, age, occupation, car availability, or trip distance - and subjective variables - as attitudinal beliefs towards cycling-. The research confirms the importance of individuaĺs perceptions on cycling for understanding their modal choice and identifies main factors related to higher bicycle use likelihood. © 2017 The Authors. Published by Elsevier B.V.

Factors

Models

Dependent variable Commuting by bicycle
Model type binary logit
Sample size 1451.0
R2 nan
Adjusted R2
Pseudo R2 (nan) 0.301
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants nan
Log-likelihood at convergence nan
Variable Coefficient p-value
Constant 0.4072 0.4287
Age -0.05662 0.0002
Both worker and student 1.85423 0.0018
Travel duration -0.05438 0.0003
subjective norm 0.5126 0.0016
Self-efficacy over unpredictable issues 0.64474 0
Attitude- Comfortable & Safe 0.72681 0
Attitude-Efficient 1.7185 0
Attitude- Pleasant & Suited life-style 0.80547 0
Self-efficacy over predictable issues 0.50904 0.0019
Distance origin-city-centre 0.5km -1.42296 0.0359
Distance origin-city-centre 0.5-1km -1.00915 0.007
Distance origin-city-centre <0.5km 0.92475 0.0663
Travel distance < 1km -3.32238 0
Destination zone mainly industrial -0.75676 0.0349
Car available -0.67262 0.0424
Sigma(u) 1.66518 0
Dependent variable Commuting by bicycle
Model type binary logit
Sample size 1451.0
R2 nan
Adjusted R2
Pseudo R2 nan
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants nan
Log-likelihood at convergence nan
Variable Coefficient p-value
Age -0.01001 0.0002
Both worker and student 0.2877 0
Travel duration -0.00961 0.0003
subjective norm 0.09062 0.0016
Self-efficacy over unpredictable issues 0.11398 0
Attitude- Comfortable & Safe 0.12848 0
Attitude-Efficient 0.30379 0
Attitude- Pleasant & Suited life-style 0.14239 0
Self-efficacy over predictable issues 0.08999 0.0019
Distance origin-city-centre 0.5km -0.23227 0.0135
Distance origin-city-centre 0.5-1km -0.17119 0.0033
Distance origin-city-centre <0.5km 0.15789 0.0488
Travel distance < 1km -0.41288 0
Destination zone mainly industrial -0.13067 0.027
Car available -0.1189 0.0424

The Attitudes and Travel Database is produced with support from the Center for Teaching Old Models New Tricks at Arizona State University, a University Transportation Center sponsored by the US Department of Transportation through Grant No. 69A3551747116.

sha256:a08d9e369743bf7e6d1c40d27347318209b40a7fb1543813fdcf31b898918815