Understanding the demand for travel: it's not purely 'derived'
Mokhtarian, Salomon, and Redmond, 2001, in Innovation
doi:10.1080/13511610120106147
Location |
San Francisco Bay Area, CA |
Population |
General |
Sample size |
1904 |
Factor analysis type |
exploratory factor analysis, Oblique rotation |
Stepwise regression |
nan |
Removal of insignificant variables |
yes |
Reviewed by |
MWC |
Abstract
We contest the derived demand paradigm for travel as a behavioural absolute. To the contrary, we suggest that travel has an intrinsic positive utility and is valued for its own sake, not just as a means of reaching a destination. We argue that the same positive characteristics that lead people to engage in travel as a recreational activity in itself are likely to motivate them to engage in apparently excess travel in the context of their mandatory and maintenance activities as well. This paper explores the conceptual basis of a positive utility for travel, and presents some results from an ongoing empirical study of attitudes toward travel. In modelling distance travelled (in each of 11 categories), we found that subjective variables such as Travel Liking, the adventure-seeker Personality trait, the travel stress Attitudinal factor, and the Excess Travel indicator added considerable explanatory power to the Demographic variables traditionally used in such models. It appears that, far from being completely determined by demographically based needs, the amount of travel demanded is heavily in uenced by one’s attitudes toward travel. This is not only true for discretionary (entertainment) purposes, as would be expected, but for more ‘mandatory’ purposes such as work/school-related activities as well. We are convinced that the demand for travel arises from a fundamental human need for mobility and other subjective characteristics, as well as from the external causes typically measured. To more accurately forecast travel demand and policy response, the role of those subjective characteristics needs to be understood much better than it is at present.
Factors
Models
Dependent variable |
ln(Short-distance total miles per week + 1) |
Model type |
linear regression |
Sample size |
1308.0 |
R2 |
0.391 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
3.732
|
0.0 |
Commute speed (objective) |
0.0215
|
0.0 |
Liking for short-distance travel by bus |
0.0525
|
0.01 |
Adventure seeking |
0.123
|
0.0 |
Percent of time a vehicle is available |
0.00252
|
0.002 |
Female |
-0.167
|
0.0 |
Personal income category (1–6) |
0.0895
|
0.0 |
Suburban |
0.364
|
0.0 |
Dependent variable |
ln(Short-distance commute miles per week + 1) |
Model type |
Linear regression |
Sample size |
1313.0 |
R2 |
0.33 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
3.421
|
0.0 |
Commute speed (objective) |
0.0303
|
0.0 |
Female |
-0.201
|
0.0 |
Age category (1-5) |
-0.212
|
0.0 |
Personal income category |
0.123
|
0.0 |
Suburban |
0.482
|
0.0 |
Dependent variable |
ln(Short distance non-commute work or school related miles per week + 1) |
Model type |
linear regression |
Sample size |
1313.0 |
R2 |
0.113 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
-0.52
|
0.041 |
Commute speed (objective) |
0.00672
|
0.014 |
Frequency of short-distance trips to eat a meal (objective) |
0.102
|
0.024 |
Frequency of short-distance trips taking others where they need to go (objective) |
0.218
|
0.0 |
Liking for long-distance work or school related travel |
0.132
|
0.005 |
Workaholic |
0.189
|
0.002 |
Adventure seeking |
0.157
|
0.002 |
Personal income category (1-6) |
0.106
|
0.001 |
Number of people in household |
0.114
|
0.006 |
Dependent variable |
ln(Short distance entertainment miles per week + 1) |
Model type |
linear regression |
Sample size |
1301.0 |
R2 |
0.139 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
0.917
|
0.003 |
Frequency of short-distance trips to eat a meal (objective) |
0.147
|
0.0 |
Liking of short-distance entertainment/recreational/social travel |
0.144
|
0.001 |
Travel stress |
-0.125
|
0.005 |
Feel attached to neighborhood |
0.139
|
0.007 |
Frustrated |
-0.109
|
0.007 |
Adventure seeking |
0.132
|
0.002 |
Excess travel indicator |
0.0214
|
0.013 |
Percent of time a vehicle is available |
0.00494
|
0.0 |
Personal income category (1-6) |
0.0751
|
0.002 |
Age category (1-5) |
-0.171
|
0.001 |
Number of people in the household |
-0.0862
|
0.001 |
Pleasant Hill |
0.219
|
0.003 |
Dependent variable |
ln(Short-distance personal vehicle miles per week + 1) |
Model type |
linear regression |
Sample size |
1308.0 |
R2 |
0.523 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
1.301
|
0.0 |
Commute speed (objective) |
0.0271
|
0.0 |
Pro-environmental policy |
-0.133
|
0.0 |
Travel freedom |
0.106
|
0.011 |
Adventure seeking |
0.144
|
0.0 |
Percent of time a vehicle is available |
0.0204
|
0.0 |
Number of personal vehicles in the household |
0.0698
|
0.018 |
Personal income category (1-6) |
0.0934
|
0.0 |
Suburban |
0.388
|
0.0 |
Dependent variable |
ln(Short distance walk miles per week + 1) |
Model type |
linear regression |
Sample size |
1351.0 |
R2 |
0.265 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
0.842
|
0.0 |
Liking of short-distance personal vehicle travel |
-0.12
|
0.0 |
Liking of short-distance walking/jogging/cycling travel |
0.424
|
0.0 |
Family/community oriented |
-0.0781
|
0.027 |
Adventure seeking |
0.114
|
0.0 |
Excess travel indicator |
0.0339
|
0.0 |
Percent of time a vehicle is available |
-0.00297
|
0.007 |
Concord |
-0.327
|
0.0 |
Pleasant Hill |
-0.152
|
0.019 |
Dependent variable |
ln(total long-distance miles per year + 1) |
Model type |
linear regression |
Sample size |
1263.0 |
R2 |
0.278 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
7.714
|
0.0 |
Frequency of short-distance trips to eat a meal (objective) |
0.192
|
0.001 |
Frequency of short-distance commute trips (objective) |
-0.258
|
0.004 |
Frequency of short-distance entertainment/recreational/social trips |
0.153
|
0.012 |
Liking for personal vehicle travel |
-0.195
|
0.001 |
Travel stress |
-0.213
|
0.002 |
Commute benefit |
-0.191
|
0.002 |
Frustrated |
-0.197
|
0.003 |
Status seeking |
-0.142
|
0.025 |
Adventure seeking |
0.23
|
0.0 |
Excess travel indicator |
0.0594
|
0.0 |
Limitations on flying |
-0.726
|
0.002 |
Respondent has a driver's license |
1.0218
|
0.013 |
Number of others in household with driver's license |
0.291
|
0.0 |
Personal income category (1-6) |
0.336
|
0.0 |
Number of people in the household |
-0.212
|
0.0 |
Concord |
-0.653
|
0.0 |
Pleasant Hill |
-0.292
|
0.014 |
Dependent variable |
ln(long distance work or school miles traveled per year + 1) |
Model type |
linear regression |
Sample size |
1307.0 |
R2 |
0.22 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
0.222
|
0.761 |
Frequency of short-distance trips to eat a meal (objective) |
0.228
|
0.03 |
Frequency of short-distance trips for work or school-related activities (objective) |
0.197
|
0.004 |
Liking for long-distance personal vehicle travel |
-0.382
|
0.001 |
Liking for long-distance work/school-related travel |
0.874
|
0.0 |
Liking for long-distance travel overall |
-0.429
|
0.002 |
Adventure seeking |
0.624
|
0.0 |
Organizer |
0.317
|
0.014 |
Excess travel indicator |
0.0567
|
0.036 |
Female |
-0.635
|
0.003 |
Personal income category (1-6) |
0.756
|
0.0 |
Dependent variable |
ln(Long distance entertainment miles traveled per year + 1) |
Model type |
linear regression |
Sample size |
1298.0 |
R2 |
0.201 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
4.827
|
0.0 |
Short-distance weekly miles traveled to eat a meal (objective) |
0.0149
|
0.001 |
Frequency of short-distance trips for entertainment/recreational/social purposes (objective) |
0.334
|
0.0 |
Liking for short-distance personal vehicle travel |
-0.222
|
0.002 |
Travel stress |
-0.322
|
0.0 |
Commute benefit |
-0.256
|
0.001 |
Frustrated |
-0.276
|
0.002 |
Workaholic |
-0.237
|
0.012 |
Excess travel indicator |
0.09897
|
0.0 |
Limitations on flying |
-0.623
|
0.037 |
Female |
0.425
|
0.001 |
Respondent has driver's license |
1.27
|
0.016 |
Personal income category (1-6) |
0.264
|
0.0 |
Concord |
-0.946
|
0.0 |
Pleasant Hill |
-0.437
|
0.003 |
Dependent variable |
ln(long distance miles by personal vehicle per year + 1) |
Model type |
linear regression |
Sample size |
1335.0 |
R2 |
0.104 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
2.972
|
0.0 |
Frequency of short-distance trips for work/school-related activities (objective) |
0.116
|
0.029 |
Frequency of short-distance trips for entertainment purposes (objective) |
0.195
|
0.021 |
Weekly miles traveled by personal vehicle (objective) |
0.00207
|
0.0 |
Liking for short-distance personal vehicle travel |
-0.267
|
0.01 |
Liking for short-distance bus travel |
-0.281
|
0.001 |
Liking for short-distance travel to eat a meal |
-0.257
|
0.028 |
Liking for long-distance travel by personal vehicle |
0.462
|
0.0 |
Frustrated |
-0.287
|
0.003 |
Excess travel indicator |
0.0737
|
0.0 |
Percent of time a vehicle is available |
0.0133
|
0.0 |
Concord |
0.467
|
0.015 |
Dependent variable |
ln(Long distance air miles traveled per year + 1) |
Model type |
linear regression |
Sample size |
1302.0 |
R2 |
0.294 |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
nan |
Variable |
Coefficient |
p-value |
Constant |
7.573
|
0.0 |
Frequency of short-distance trips to eat a meal (objective) |
0.345
|
0.0 |
Frequency of short-distance commute trips (objective) |
-0.488
|
0.002 |
Frequency of short-distance trips for work/school-related activities (objective) |
0.139
|
0.014 |
Liking for personal vehicle travel |
-0.364
|
0.0 |
Liking for work/school-related travel |
0.215
|
0.018 |
Travel stress |
-0.265
|
0.03 |
Commute benefit |
-0.237
|
0.028 |
Pro-hi density |
0.59
|
0.0 |
Frustrated |
-0.406
|
0.0 |
Adventure seeking |
0.391
|
0.0 |
Female |
0.399
|
0.025 |
Age category (1-5) |
-0.367
|
0.008 |
Number of people 6-15 years old in household |
-0.798
|
0.0 |
Personal income category (1-6) |
0.647
|
0.0 |
Concord |
-1.277
|
0.0 |