Do environmental concerns affect commuting choices?: hybrid choice modelling with household survey data
Roberts, Popli, Harris, 2018, in Journal of the Royal Statistical Society. Series A: Statistics in Society
doi:10.1111/rssa.12274
Location |
England |
Population |
Other (specify) |
Sample size |
13139 |
Factor analysis type |
exploratory factor analysis, varimax rotation |
Stepwise regression |
no |
Removal of insignificant variables |
no |
Reviewed by |
LCM |
Abstract
To meet ambitious climate change goals governments must encourage behavioural change alongside technological progress. Designing effective policy requires a thorough understanding of the factors that drive behaviours. In an effort to understand the role of environmental attitudes better we estimate a hybrid choice model (HCM) for commuting mode choice by using a large household survey data set. HCMs combine traditional discrete choice models with a structural equation model to integrate latent variables, such as attitudes, into the choice process. To date HCMs have utilized small bespoke data sets, beset with problems of selection and limited generalizability. To overcome these problems we demonstrate the feasibility of using this valuable modelling approach with nationally representative data. Our results suggest that environmental attitudes have an important influence on commute mode choice, and this can be exploited by governments looking to add to their climate change policy toolbox in an effort to change travel behaviours. © 2017 The Authors Journal of the Royal Statistical Society: Series A (Statistics in Society) Published by John Wiley & Sons Ltd on behalf of the Royal Statistical Society.
Factors
Models
Dependent variable |
Car commuting mode |
Model type |
Hybrid choice model |
Sample size |
6256.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
0.185 |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Choice model |
Variable |
Coefficient |
p-value |
Environmental behaviors (men) |
-0.181
|
<0.001 |
Environmental attitudes (men) |
-0.105
|
<0.001 |
Age |
0.701
|
<0.001 |
Age squared |
-0.563
|
<0.001 |
Children aged 0-2 years |
0.071
|
nan |
Children aged 3-4 years |
0.094
|
nan |
Children aged 5-11 years |
-0.06
|
nan |
Children aged 12-15 years |
-0.006
|
nan |
Education: degree |
-0.033
|
nan |
Education: A level |
-0.036
|
nan |
Education: O level |
0.018
|
nan |
Household income |
-0.088
|
<0.001 |
Married |
0.195
|
<0.001 |
Health status |
0.008
|
nan |
Public transport quality |
-0.16
|
<0.001 |
Average traffic speed |
0.1
|
<0.001 |
Commute time |
-0.166
|
<0.001 |
Environmental attitudes (men) |
Variable |
Coefficient |
p-value |
Age |
0.295
|
<0.001 |
Age squared |
-0.236
|
<0.001 |
Children aged 0-2 years |
0.005
|
nan |
Children aged 3-4 years |
0.006
|
nan |
Children aged 5-11 years |
-0.015
|
<0.05 |
Children aged 12-15 years |
-0.005
|
nan |
Education: degree |
0.127
|
<0.001 |
Education: A level |
0.097
|
<0.001 |
Education: O level |
0.068
|
<0.001 |
Household income |
0.099
|
<0.001 |
Married |
0.01
|
nan |
Health status |
-0.001
|
nan |
Environmental behaviours (men) |
Variable |
Coefficient |
p-value |
Age |
0.517
|
<0.001 |
Age squared |
-0.312
|
<0.05 |
Children aged 0-2 years |
0.023
|
<0.1 |
Children aged 3-4 years |
-0.004
|
nan |
Children aged 5-11 years |
-0.005
|
nan |
Children aged 12-15 years |
0.001
|
nan |
Education: degree |
0.157
|
<0.001 |
Education: A level |
0.079
|
<0.001 |
Education: O level |
0.023
|
nan |
Household income |
-0.058
|
<0.001 |
Married |
0.064
|
<0.001 |
Health status |
0.018
|
<0.001 |
Dependent variable |
Car commuting mode |
Model type |
Hybrid choice model |
Sample size |
6883.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
0.177 |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Choice model |
Variable |
Coefficient |
p-value |
Environmental behaviors (women) |
-0.167
|
<0.001 |
Environmental attitudes (women) |
0.001
|
nan |
Age |
0.778
|
<0.001 |
Age squared |
-0.636
|
<0.001 |
Children aged 0-2 years |
0.152
|
<0.001 |
Children aged 3-4 years |
0.211
|
<0.001 |
Children aged 5-11 years |
0.013
|
nan |
Children aged 12-15 years |
0.009
|
nan |
Education: degree |
0.459
|
<0.001 |
Education: A level |
0.408
|
<0.001 |
Education: O level |
0.265
|
<0.001 |
Household income |
0.054
|
<0.001 |
Married |
-0.199
|
<0.001 |
Health status |
-0.014
|
nan |
Public transport quality |
-0.126
|
<0.001 |
Average traffic speed |
0.147
|
<0.001 |
Commute time |
-0.233
|
<0.001 |
Environmental attitudes (women) |
Variable |
Coefficient |
p-value |
Age |
0.248
|
<0.001 |
Age squared |
-0.238
|
<0.001 |
Children aged 0-2 years |
-0.014
|
nan |
Children aged 3-4 years |
-0.016
|
nan |
Children aged 5-11 years |
-0.023
|
<0.001 |
Children aged 12-15 years |
-0.007
|
nan |
Education: degree |
0.127
|
<0.001 |
Education: A level |
0.09
|
<0.001 |
Education: O level |
0.058
|
<0.001 |
Household income |
0.097
|
<0.001 |
Married |
0.003
|
nan |
Health status |
-0.001
|
nan |
Environmental behaviours (women) |
Variable |
Coefficient |
p-value |
Age |
0.683
|
<0.001 |
Age squared |
-0.448
|
<0.001 |
Children aged 0-2 years |
-0.008
|
nan |
Children aged 3-4 years |
0.011
|
nan |
Children aged 5-11 years |
0.001
|
nan |
Children aged 12-15 years |
-0.003
|
nan |
Education: degree |
0.121
|
<0.001 |
Education: A level |
0.053
|
<0.001 |
Education: O level |
0.021
|
nan |
Household income |
-0.032
|
<0.05 |
Married |
0.031
|
<0.001 |
Health status |
0.009
|
<0.001 |
Dependent variable |
Car commuting mode |
Model type |
Hybrid choice model |
Sample size |
6256.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
0.124 |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Choice model |
Variable |
Coefficient |
p-value |
Age |
0.509
|
<0.001 |
Age squared |
-0.418
|
<0.001 |
Children aged 0-2 years |
0.051
|
nan |
Children aged 3-4 years |
0.089
|
nan |
Children aged 5-11 years |
-0.059
|
nan |
Children aged 12-15 years |
0.021
|
nan |
Education: degree |
-0.268
|
<0.001 |
Education: A level |
-0.178
|
<0.001 |
Education: O level |
-0.059
|
nan |
Household income |
-0.112
|
<0.001 |
Married |
0.149
|
<0.001 |
Health status |
-0.008
|
nan |
Public transport quality |
-0.18
|
<0.001 |
Average traffic speed |
0.106
|
<0.001 |
Dependent variable |
Car commuting mode |
Model type |
Hybrid choice model |
Sample size |
6883.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
0.106 |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Choice model |
Variable |
Coefficient |
p-value |
Age |
0.578
|
<0.001 |
Age squared |
-0.463
|
<0.001 |
Children aged 0-2 years |
0.163
|
<0.001 |
Children aged 3-4 years |
0.176
|
<0.001 |
Children aged 5-11 years |
0.053
|
nan |
Children aged 12-15 years |
0.04
|
nan |
Education: degree |
0.268
|
<0.001 |
Education: A level |
0.324
|
<0.001 |
Education: O level |
0.234
|
<0.001 |
Household income |
0.029
|
<0.1 |
Married |
-0.19
|
<0.001 |
Health status |
-0.025
|
nan |
Public transport quality |
-0.157
|
<0.001 |
Average traffic speed |
0.168
|
<0.001 |
Dependent variable |
Car commuting mode |
Model type |
Hybrid choice model |
Sample size |
6265.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
0.164 |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Choice model |
Variable |
Coefficient |
p-value |
Environmental behaviors (men) |
-0.182
|
<0.001 |
Environmental attitudes (men) |
-0.109
|
<0.001 |
Age |
0.635
|
<0.001 |
Age squared |
-0.501
|
<0.001 |
Children aged 0-2 years |
0.07
|
nan |
Children aged 3-4 years |
0.092
|
nan |
Children aged 5-11 years |
-0.071
|
nan |
Children aged 12-15 years |
0.019
|
nan |
Education: degree |
-0.083
|
nan |
Education: A level |
-0.067
|
nan |
Education: O level |
-0.002
|
nan |
Household income |
-0.112
|
<0.001 |
Married |
0.2
|
<0.001 |
Health status |
0.005
|
nan |
Public transport quality |
-0.18
|
<0.001 |
Average traffic speed |
0.106
|
<0.001 |
Dependent variable |
Car commuting mode |
Model type |
Hybrid choice model |
Sample size |
6883.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
0.132 |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Choice model |
Variable |
Coefficient |
p-value |
Environmental behaviors (women) |
-0.17
|
<0.001 |
Environmental attitudes (women) |
-0.004
|
nan |
Age |
0.695
|
<0.001 |
Age squared |
-0.54
|
<0.001 |
Children aged 0-2 years |
0.156
|
<0.001 |
Children aged 3-4 years |
0.185
|
<0.001 |
Children aged 5-11 years |
0.053
|
nan |
Children aged 12-15 years |
0.037
|
nan |
Education: degree |
0.372
|
<0.001 |
Education: A level |
0.369
|
<0.001 |
Education: O level |
0.253
|
<0.001 |
Household income |
0.024
|
nan |
Married |
-0.164
|
<0.001 |
Health status |
-0.017
|
nan |
Public transport quality |
-0.157
|
<0.001 |
Average traffic speed |
0.168
|
<0.001 |