Attitudes towards mode choice in Switzerland
Atasoy, Glerum, Bierlaire, 2013, in DISP
doi:10.1080/02513625.2013.827518
Location |
Switzerland |
Population |
General |
Sample size |
1763 |
Factor analysis type |
exploratory factor analysis, nan rotation |
Stepwise regression |
no |
Removal of insignificant variables |
no |
Reviewed by |
LCM |
Abstract
We integrate latent attitudes of the individuals into a transport mode choice model through latent variable and latent class models. Psychometric indicators are used to measure these attitudes. The aim of the inclusion of attitudes is to better understand the underlying choice preferences of travelers and therefore increase the forecasting power of the choice model. We first present an integrated choice and latent variable model, where we include attitudes towards public transport and environmental issues, explaining the utility of public transport. Secondly, we present an integrated choice and latent class model, where we identify two segments of individuals having different sensitivities to the attributes of the alternatives, resulting from their individual characteristics. The calibration of these types of advanced models on our sample has demonstrated the importance of attitudinal variables in the characterization of heterogeneity of mode preferences within the population. © 2013 Copyright Taylor and Francis Group, LLC.
Factors
Models
Dependent variable |
Mode choice |
Model type |
ICLV |
Sample size |
1763.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
-1069.8 |
Car |
Variable |
Coefficient |
p-value |
Constant |
-0.599
|
0.418 |
Travel cost |
-0.0559
|
0.0 |
Travel time |
-0.0294
|
0.0 |
Number of cars |
0.97
|
0.0 |
Number of children |
0.215
|
0.001 |
Region (French-speaking=1) |
1.06
|
0.0 |
Work trip (yes =1) |
-0.583
|
0.0 |
Environmental concern |
Variable |
Coefficient |
p-value |
Constant |
3.23
|
0.0 |
Education |
0.235
|
0.0 |
Number of bikes |
0.0845
|
0.0 |
Age*Age>45 |
0.00445
|
0.027 |
Pro-car |
Variable |
Coefficient |
p-value |
Constant |
3.02
|
0.0 |
Number of cars |
0.104
|
0.0 |
Education |
0.235
|
0.0 |
Region (Valais=1) |
-0.223
|
0.005 |
Region (Bern=1) |
-0.361
|
0.0 |
Region (Basel-Zurich=1) |
-0.256
|
0.0 |
Region (East Switzerland) |
-0.228
|
0.001 |
Region (Graubünden) |
-0.303
|
0.001 |
Public transport |
Variable |
Coefficient |
p-value |
Travel cost |
-0.0559
|
0.0 |
Travel time |
-0.0119
|
0.0 |
Urban (yes=1) |
0.283
|
0.025 |
Student (yes=1) |
3.26
|
0.0 |
Pro-car |
-0.574
|
0.0 |
Environmental concern |
0.393
|
0.003 |
Soft mode |
Variable |
Coefficient |
p-value |
Constant |
-0.772
|
0.353 |
Distance traveled |
-0.224
|
0.0 |
Number of bikes |
0.385
|
0.0 |
Dependent variable |
Mode choice |
Model type |
Latent class model |
Sample size |
1763.0 |
R2 |
0.507 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
-1032.5 |
Dependent, car |
Variable |
Coefficient |
p-value |
Constant |
-0.936
|
0.001 |
Travel cost |
-0.302
|
0.0 |
Travel time |
-0.111
|
0.0 |
Number of cars |
1.23
|
0.0 |
Number of children |
-1.03
|
0.234 |
Region (French-speaking=1) |
1.2
|
0.0 |
Work trip (yes =1) |
-0.13
|
0.682 |
Dependent, public transport |
Variable |
Coefficient |
p-value |
Travel cost |
-0.302
|
0.0 |
Travel time |
-0.0445
|
0.0 |
Urban (yes=1) |
0.39
|
0.005 |
Student (yes=1) |
3.7
|
0.0 |
Independent, car |
Variable |
Coefficient |
p-value |
Constant |
-0.945
|
0.0 |
Travel cost |
-0.027
|
0.006 |
Travel time |
-0.0161
|
0.01 |
Number of cars |
1.23
|
0.0 |
Number of children |
0.404
|
0.0 |
Region (French-speaking=1) |
1.2
|
0.0 |
Work trip (yes =1) |
-0.785
|
0.0 |
Independent, public transport |
Variable |
Coefficient |
p-value |
Travel cost |
-0.027
|
0.006 |
Travel time |
-0.00692
|
0.013 |
Urban (yes=1) |
0.39
|
0.005 |
Student (yes=1) |
3.7
|
0.0 |
Independent, soft mode |
Variable |
Coefficient |
p-value |
Constant |
0.512
|
0.19 |
Distance traveled |
-0.199
|
0.0 |
Number of bikes |
0.205
|
0.001 |
Dependent variable |
Mode choice |
Model type |
Multinomial logit model |
Sample size |
1763.0 |
R2 |
0.49 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
-1067.4 |
Car |
Variable |
Coefficient |
p-value |
Constant |
-0.413
|
0.017 |
Travel cost |
-0.0592
|
0.0 |
Travel time |
-0.0299
|
0.0 |
Number of cars |
1.0
|
0.0 |
Number of children |
0.154
|
0.018 |
Region (French-speaking=1) |
1.09
|
0.0 |
Work trip (yes =1) |
-0.582
|
0.0 |
Public transport |
Variable |
Coefficient |
p-value |
Travel cost |
-0.0592
|
0.0 |
Travel time |
-0.0121
|
0.0 |
Urban (yes=1) |
0.286
|
0.02 |
Student (yes=1) |
3.21
|
0.0 |
Soft mode |
Variable |
Coefficient |
p-value |
Constant |
-0.47
|
0.204 |
Distance traveled |
-0.227
|
0.0 |
Number of bikes |
0.347
|
0.0 |