Understanding transport mode choice for commuting: the role of affect
Ababio-Donker, Saleh, Fonzone, 2020, in Transportation Planning and Technology
doi:10.1080/03081060.2020.1747203
| Location |
Edinburgh, Scotland, United Kingdom |
| Population |
General |
| Sample size |
500 |
| Factor analysis type |
confirmatory factor analysis, none rotation |
| Stepwise regression |
no |
| Removal of insignificant variables |
yes |
| Reviewed by |
LCM |
Abstract
This study examines the relationship between positive and negative user valence and transport mode choice behaviour. We integrate latent attitudes ‘affect’ and ‘salience’ into transport mode choice models using the framework of integrated choice and latent variable modelling and simultaneous maximum likelihood estimation methods. The results are consistent with findings in similar travel behaviour and behavioural economics literature. The study extends the findings of previous research and has demonstrated that user sentiments about public transport mode and salient public transport experiences have a significant impact on travel mode choice behaviour. It was found that private motorised users are more sensitive to overcrowding and anti-social behaviours on PT than active and PT travellers. Key attitudinal indicators influencing individual transport choice behaviour are established to guide public policy. The key indicators of Affect and Salience must be analysed and addressed through public policy to enhance PT user experience and develop services and facilities to increase the utility of PT in-vehicle travel time. © 2020, © 2020 Informa UK Limited, trading as Taylor & Francis Group.
Factors
Models
| Dependent variable |
Mode choice |
| Model type |
MNL |
| Sample size |
500 |
| R2 |
0.285 |
| Adjusted R2 |
|
|
Pseudo R2
(nan)
|
nan |
| AIC |
nan |
| BIC |
nan |
| Log-likelihood at zero |
nan |
| Log-likelihood at constants |
nan |
| Log-likelihood at convergence |
-82.42 |
| Car |
Variable |
Coefficient |
p-value |
| ASC |
-2.72
|
0.007 |
| Cost (car) |
-0.1
|
0.31 |
| Gender |
0.5
|
0.06 |
| Number of cars |
1.87
|
0 |
| Travel time (car) |
-0.4
|
0.016 |
| Trip frequency |
-0.23
|
0.028 |
| Walking time to bus stop |
0.22
|
0.002 |
| Work trip |
-0.86
|
0.007 |
| Non-motorized travel |
Variable |
Coefficient |
p-value |
| ASC |
-3.9
|
0 |
| Age |
-0.17
|
0.06 |
| Distance |
-0.16
|
0 |
| Education |
0.51
|
0 |
| Public transport |
Variable |
Coefficient |
p-value |
| Age |
-0.37
|
0.145 |
| Cost (PT) |
-3.1
|
0.015 |
| Income |
-0.17
|
0.08 |
| Travel time (PT) |
-0.2
|
0.039 |
| Dependent variable |
Mode choice |
| Model type |
HCM |
| Sample size |
500 |
| R2 |
0.362 |
| Adjusted R2 |
|
|
Pseudo R2
(nan)
|
nan |
| AIC |
nan |
| BIC |
nan |
| Log-likelihood at zero |
nan |
| Log-likelihood at constants |
nan |
| Log-likelihood at convergence |
-78.1 |
| Affect |
Variable |
Coefficient |
p-value |
| ASC |
0.34
|
0.007 |
| Age |
0.38
|
0.001 |
| Travel pass |
0.42
|
0 |
| Car |
Variable |
Coefficient |
p-value |
| ASC |
-3.18
|
0.004 |
| Cost (car) |
-0.21
|
0.225 |
| Gender |
0.53
|
0.048 |
| Number of cars |
1.82
|
0 |
| Travel time (car) |
-0.49
|
0.005 |
| Trip frequency |
-0.2
|
0.071 |
| Walking time to bus stop |
0.2
|
0.01 |
| Work trip |
-0.18
|
0.032 |
| Affect |
nan
|
nan |
| Non-motorized travel |
Variable |
Coefficient |
p-value |
| ASC |
-3.85
|
0 |
| Age |
-0.23
|
0.013 |
| Distance |
-0.18
|
0 |
| Education |
0.48
|
0 |
| Public transport |
Variable |
Coefficient |
p-value |
| Age |
-0.74
|
0.016 |
| Cost (PT) |
-0.24
|
0.008 |
| Income |
-0.18
|
0.08 |
| Travel time (PT) |
-0.2
|
0.043 |
| Affect |
0.62
|
0.006 |
| Salient experience |
-0.29
|
0.005 |
| Salient experience |
Variable |
Coefficient |
p-value |
| Age |
0.42
|
0.003 |
| Number of cars |
0.15
|
0.094 |