Changing travel behaviour and attitudes following a residential relocation

Jonas De Vos, Dick Ettema, Frank Witlox, 2018, in The Journal of Transport Geography

doi:10.1016/j.jtrangeo.2018.10.013
Location Ghent, Belgium
Population General
Sample size 1539
Factor analysis type exploratory factor analysis, promax rotation
Stepwise regression no
Removal of insignificant variables no
Reviewed by AR

Factors

Variable Pattern loading
I need a car to feel free (Mode-liking) 0.83
I like driving a car (Mode-liking) 0.78
City centers should be/should remain well accessible by car (Transport policy) 0.71
Car use should be taxed more heavily (Transport policy) -0.56
Cities should be low-traffic areas (Transport policy) -0.41
I like using public transport (Mode-liking) -0.37
Public transport is unreliable (Convenience) 0.29
Variable Pattern loading
Destinations should be well accessible by bike (Transport policy) 0.77
Streets should be walk friendly (Transport policy) 0.75
Destinations should be well accessible by foot (Transport policy) 0.72
Neighbourhoods should be bike friendly (Transport policy) 0.51
Cities should be low-traffic areas (Transport policy) 0.41
More investment in public transport is needed (Transport policy) 0.35
I like to cycle (Mode-liking) 0.29
I like to walk (Mode-liking) 0.28

Models

Dependent variable Car Use
Model type multinomial logit
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox and Snell) 0.27
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2946.82
Log-likelihood at convergence 2482.55
Decrease
Variable Coefficient p-value
Intercept -4.62 <0.01
Change in urbanisation 0.31 <0.01
Frequent car use -0.75 <0.01
frequent PT use 0.5 <0.05
Frequent cycling -0.29 <0.1
Pro sustainable travel 0.18 <0.1
Change in car attitude 0.21 <0.05
Change in PT attitude 0.29 <0.01
Change in cycling attitude 0.28 <0.01
Change in walking attitude 0.24 <0.05
Change in comm. Distance -0.31 <0.01
Change in car possession -0.45 <0.01
Age 0.01 <0.05
Car ownership 0.23 <0.1
Increase
Variable Coefficient p-value
Change in PT attitude -0.36 <0.01
Change in comm. Distance 0.34 <0.01
Change in car possession 0.41 <0.01
Age -0.04 <0.01
Driving license 0.82 <0.05
Dependent variable PT use
Model type multinomial logit
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox and Snell) 0.21
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2942.03
Log-likelihood at convergence 2594.03
Decrease
Variable Coefficient p-value
Frequent PT use -0.75 <0.05
Change in PT attitude -0.46 <0.01
Change in walking attitude 0.25 <0.05
Gender 0.26 <0.1
Increase
Variable Coefficient p-value
Intercept -6.07 <0.01
Change in urbanisation 0.32 <0.01
Frequent PT use 1.11 <0.01
Change in PT attitude 0.81 <0.01
Change in walking attitude 0.32 <0.01
Gender 0.26 <0.1
Single -0.4 <0.05
Children 0.34 <0.1
Dependent variable Cycling
Model type multinomial logit
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox and Snell) 0.28
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2979.88
Log-likelihood at convergence 2496.15
Decrease
Variable Coefficient p-value
Frequent PT use 0.68 <0.01
Frequent cycling -1.1 <0.01
Change in cycling attitude -0.43 <0.01
Change in walking attitude 0.26 <0.05
Driving license -0.44 <0.1
Car ownership 0.27 <0.1
Increase
Variable Coefficient p-value
Intercept -4.72 <0.01
Change in urbanisation 0.22 <0.01
Frequent PT use 0.37 <0.1
Frequent cycling 0.29 <0.1
Change in cycling attitude 1.03 <0.01
Change in comm. Distance -0.23 <0.01
Change in car possession -0.39 <0.01
Children 0.42 <0.05
Car ownership 0.45 <0.01
Dependent variable Walking
Model type multinomial
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox and Snell) 0.31
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2945.18
Log-likelihood at convergence 2402.24
Decrease
Variable Coefficient p-value
Frequent walking -0.38 <0.1
Change in PT attitude -0.3 <0.05
Change in comm. Distance 0.22 <0.05
Children -0.38 <0.1
Increase
Variable Coefficient p-value
Intercept -6.09 <0.01
Urban 0.46 <0.05
Change in urbanisation 0.34 <0.01
Frequent walking 0.28 <0.1
Change in PT attitude 0.21 <0.1
Change in walking attitude 1.11 <0.01
Change in comm. Distance -0.24 <0.01
Change in car possession -0.35 <0.01
Car ownership 0.37 <0.01
Dependent variable Car Attitude
Model type multinomial logit
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox and Snell) 0.1
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2234.44
Log-likelihood at convergence 2092.05
Improved
Variable Coefficient p-value
Intercept -3.16 <0.01
Change in urbanisation 0.17 <0.05
Frequent walking 0.29 <0.1
Pro sustainable travel 0.18 <0.1
Change in cycling 0.17 <0.1
Change in walking 0.2 <0.05
Education -0.42 <0.05
Income -0.52 <0.01
Car ownership 0.54 <0.01
Worsened
Variable Coefficient p-value
Intercept -1.69 <0.01
Pro car -0.71 <0.01
Change in car possesion -0.4 <0.01
Age -0.02 <0.05
Education -0.57 <0.05
Car ownership 0.46 <0.01
Dependent variable PT attitude
Model type multinomial logit
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox & Snell) 0.2
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2461.84
Log-likelihood at convergence 2142.58
Improved
Variable Coefficient p-value
Intercept -3.28 <0.01
Change in urbanisation 0.22 <0.01
Frequent PT use 0.45 <0.05
Change in car use -0.18 <0.05
Change in PT use 0.65 <0.01
Worsened
Variable Coefficient p-value
Pro car 0.64 <0.01
Change in PT use -0.4 <0.01
Change in cycling 0.2 <0.1
Change in walking -0.18 <0.1
Age -0.02 <0.05
Income -0.42 <0.05
Dependent variable Cycling Attitude
Model type multinomial logit
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox & Snell) 0.25
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2601.17
Log-likelihood at convergence 2182.66
Improved
Variable Coefficient p-value
Intercept -4.0 <0.01
Change in urbanisation 0.29 <0.01
Frequent car use -0.37 <0.05
Frequent cycling 0.29 <0.1
Pro car 0.2 <0.1
Change in cycling 0.91 <0.01
Change in walking -0.14 <0.1
Change in car possesion -0.25 <0.05
Gender 0.34 <0.01
Children 0.3 <0.1
Driving license 0.22 <0.1
Worsened
Variable Coefficient p-value
Intercept -1.83 <0.05
Frequent cycling -1.38 <0.01
Change in cycling -0.34 <0.01
Change in comm. Distance 0.28 <0.1
Gender 0.51 <0.05
Single -0.48 <0.1
Dependent variable Walking attitude
Model type multinomial logit
Sample size 1539.0
R2 nan
Adjusted R2
Pseudo R2 (Cox & Snell) 0.3
AIC nan
BIC nan
Log-likelihood at zero nan
Log-likelihood at constants 2535.19
Log-likelihood at convergence 2024.98
Improved
Variable Coefficient p-value
Intercept -3.23 <0.01
Urban 0.74 <0.01
Change in urbanisation 0.26 <0.01
Frequent walking 0.45 <0.01
Change in walking 0.67 <0.01
Change in comm. Distance -0.25 <0.01
Change in car possession -0.19 <0.1
Income -0.29 <0.1
Worsened
Variable Coefficient p-value
Change in urbanisation -0.24 <0.1
Change in walking -0.7 <0.01
Change in comm. Distance 0.44 <0.01
Age 0.02 <0.05
Children 0.5 <0.1

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