Modelling bicycle use intention: the role of perceptions
Fernández-Heredia, Jara-Díaz Monzón, 2016, in Transportation
doi:10.1007/s11116-014-9559-9
| Location |
Madrid, Spain |
| Population |
Other (specify) |
| Sample size |
3048 |
| Factor analysis type |
exploratory factor analysis, unknown rotation |
| Stepwise regression |
no |
| Removal of insignificant variables |
yes |
| Reviewed by |
LCM |
Abstract
Users’ perceptions are identified as key elements to understand bicycle use, whose election cannot be explained with usual mobility variables and socio-economic characteristics. A hybrid model is proposed to model the intention of bicycle use; it combines a structural equations model that captures intentions and a choice model. The framework is applied to a case of a university campus in Madrid that is studying a new internal bike system. Results show that four latent variables (convenience, pro-bike, physical determinants and external restrictions) help explaining intention to use bike, representing a number of factors that are linked to individual perceptions. © 2014, Springer Science+Business Media New York.
Factors
Models
| Dependent variable |
Bike use intention |
| Model type |
Logit |
| Sample size |
3048 |
| R2 |
nan |
| Adjusted R2 |
|
|
Pseudo R2
(McFadden)
|
0.1559 |
| AIC |
1.095 |
| BIC |
nan |
| Log-likelihood at zero |
nan |
| Log-likelihood at constants |
nan |
| Log-likelihood at convergence |
-12.468 |
| Variable |
Coefficient |
p-value |
| Intercept |
1.103
|
0.0 |
| Male |
0.131
|
0.0 |
| Income |
-0.386
|
0.0 |
| Level of graduate |
0.084
|
0.025 |
| Student |
0.265
|
0.0 |
| Home with more than 3 members |
-0.16
|
0.0 |
| Spanish nationality |
0.232
|
0.003 |
| Study/work at UPM |
-0.254
|
0.0 |
| Study/work at non university center |
-0.203
|
0.009 |
| Car user |
-0.199
|
0.0 |
| Walker |
0.403
|
0.0 |
| Distance traveled in CU |
0.497
|
0.0 |
| Total travel time |
3.72
|
0.0 |
| Inner trip with origin in CU |
0.795
|
0.0 |
| Inner trip with origin in CU |
0.278
|
0.0 |
| Trip purpose: leisure |
-0.872
|
0.0 |
| Distance among system points |
-0.365
|
0.0 |
| Possibility of traffic restrictions |
-0.558
|
0.0 |
| Possibility of bike lane |
0.114
|
0.013 |
| Fee |
-2.002
|
0.0 |
| Non bike availability |
0.132
|
0.0 |
| Frequency of bike use |
0.892
|
0.0 |
| Public bike system familiarity |
0.204
|
0.0 |
| Mandatory bike use |
-0.294
|
0.0 |
| Leisure bike use |
0.087
|
0.001 |
| Dependent variable |
Bike use intention |
| Model type |
Hybrid |
| Sample size |
3048 |
| R2 |
nan |
| Adjusted R2 |
|
|
Pseudo R2
(McFadden)
|
0.1872 |
| AIC |
1.054 |
| BIC |
nan |
| Log-likelihood at zero |
nan |
| Log-likelihood at constants |
nan |
| Log-likelihood at convergence |
-12.468 |
| Structural equations: Age |
Variable |
Coefficient |
p-value |
| Convenience |
-0.3
|
0.0 |
| Structural equations: Car availability |
Variable |
Coefficient |
p-value |
| Convenience |
0.1
|
0.066 |
| Structural equations: Faculty member |
Variable |
Coefficient |
p-value |
| External restrictions |
-0.31
|
0.022 |
| Structural equations: Frequency of bike use |
Variable |
Coefficient |
p-value |
| Convenience |
0.49
|
0.0 |
| Probike |
0.5
|
0.0 |
| Physical determinants |
-0.58
|
0.0 |
| External restrictions |
0.84
|
0.0 |
| Structural equations: Income |
Variable |
Coefficient |
p-value |
| External restrictions |
-0.29
|
0.03 |
| Structural equations: Leisure bike use |
Variable |
Coefficient |
p-value |
| Convenience |
-0.19
|
0.0 |
| Probike |
-0.19
|
0.0 |
| Physical determinants |
0.06
|
0.084 |
| External restrictions |
-0.31
|
0.001 |
| Structural equations: Level of graduate studies |
Variable |
Coefficient |
p-value |
| Probike |
-0.2
|
0.0 |
| Structural equations: Male |
Variable |
Coefficient |
p-value |
| Probike |
-0.27
|
0.0 |
| Structural equations: Mandatory bike use |
Variable |
Coefficient |
p-value |
| External restrictions |
-0.63
|
0.0 |
| Structural equations: Sport bike use |
Variable |
Coefficient |
p-value |
| Probike |
0.18
|
0.0 |
| Structural equations: Student |
Variable |
Coefficient |
p-value |
| External restrictions |
-0.47
|
0.01 |
| Structural equations: Study/work at UPM |
Variable |
Coefficient |
p-value |
| Physical determinants |
-0.09
|
0.005 |
| Structural equations: Willingness to use UNIBICI |
Variable |
Coefficient |
p-value |
| Physical determinants |
-0.25
|
0.0 |
| Variable |
Coefficient |
p-value |
| Intercept |
-1.423
|
0.0 |
| Male |
0.106
|
0.002 |
| Level of graduate |
0.08
|
0.029 |
| Student |
0.289
|
0.0 |
| Home with more than 3 members |
-0.133
|
0.0 |
| Study/work at UPM |
-0.199
|
0.0 |
| Study/work at non university center |
-0.144
|
0.074 |
| Car user |
-0.094
|
0.039 |
| Walker |
0.285
|
0.0 |
| Distance traveled in CU |
0.535
|
0.0 |
| Total travel time |
3.551
|
0.0 |
| Inner trip with origin in CU |
0.752
|
0.0 |
| Inner trip with origin in CU |
0.277
|
0.0 |
| Trip purpose: leisure |
-1.201
|
0.0 |
| Distance among system points |
-0.381
|
0.0 |
| Possibility of traffic restrictions |
-0.585
|
0.0 |
| Possibility of bike lane |
0.118
|
0.011 |
| Fee |
-2.101
|
0.0 |
| Convenience |
1.186
|
0.0 |
| Probike |
2.157
|
0.0 |
| External restrictions |
-1.236
|
0.0 |
| Physical determinants |
0.452
|
0.0 |
| Non bike availability |
0.097
|
0.011 |
| Frequency of bike use |
0.55
|
0.0 |
| Public bike system familiarity |
0.144
|
0.0 |
| Mandatory bike use |
-0.344
|
0.0 |
| Leisure bike use |
0.068
|
0.01 |