UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Determinants of travel mode choice in urban areas Takla, Emile Fawzy 1974

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1974_A6_7 T33_9.pdf [ 7.07MB ]
Metadata
JSON: 831-1.0093009.json
JSON-LD: 831-1.0093009-ld.json
RDF/XML (Pretty): 831-1.0093009-rdf.xml
RDF/JSON: 831-1.0093009-rdf.json
Turtle: 831-1.0093009-turtle.txt
N-Triples: 831-1.0093009-rdf-ntriples.txt
Original Record: 831-1.0093009-source.json
Full Text
831-1.0093009-fulltext.txt
Citation
831-1.0093009.ris

Full Text

DETERMINANTS OF TRAVEL MODE CHOICE. IN URBAN AREAS by EMILE FAWZY TAKLA B. Arch., Cairo University, 1963 M.A., Unive r s i t y of Calgary, 1971 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n the School of Community and Regional Planning We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH A p r i l , 1974 COLUMBIA In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the Li b r a r y s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. I t i s understood that copying or pu b l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of School of Community and Regional Planning The Uni v e r s i t y of B r i t i s h Columiba Vancouver 8, Canada Date A p r i l , 1974.  ACKNOWLEDGEMENT The author wishes to thank Professor Howard Cherniak f o r his contribution to t h i s work and h i s "long distance" cooperation to supervise t h i s study while the author was off-campus. Thanks are also due to Dr. Gerald Brown f o r h i s guidance and u n f a i l i n g support p a r t i c u l a r l y at the i n i t i a l stages of the study. The assistance of Professor Brahm Wiesman, Acting Director of the School, i n the formulation of the study program and h i s recommendations for f i n a n c i a l support i s g r a t e f u l l y acknowledged. This study was j o i n t l y financed by R. K. Mellon Fellowship and The Transportation Development Agency Fellow-ship. i i i ABSTRACT This study has two major concerns: the i d e n t i f i c a t i o n of mode choice determinants, and the formulation of a mode choice model which accounts f o r the determinants defined. F i r s t , the concepts underlying users' behaviour are expounded. These concepts r e l a t e users' mode choice to several influences, the perceived a t t r i b u t e s of the transportation system, his socio-economic c h a r a c t e r i s t i c s , and s e n s i t i v i t y toward modal at t r i b u t e s . In t h i s study, the l a t t e r i s hypothesized to be a function of the user's time budget and other indic a t i o n s of his l i f e s t y l e . Thus, an i n d i v i d u a l may attach a great s i g n i f i c a n c e to savings i n t r a v e l time as a r e s u l t of h i s engagement i n a c t i v i t i e s which put considerable demand upon his time, although most members of his income group may be s e n s i t i v e p r i m a r i l y to the t r a v e l cost a t t r i b u t e . User's l i f e s t y l e may therefore create divergent s e n s i t i v i t i e s within the same socio-economic group. Previous research findings support the hypothesis that variations i n these s e n s i t i v i t i e s are independent from the. socio-economic c h a r a c t e r i s t i c s . The study's model makes use of t h i s concept. The model i s composed of two parts: the f i r s t i s concerned with the grouping of a l l users according to t h e i r s e n s i t i v i t i e s toward a t t r i b u t e s of the mode choice s i t u a t i o n , and subsequently i v the c a l i b r a t i o n of a stochastic function to explain users' choices i n each group. The second part of the model r e l a t e s the user's time, age and occupation (as i n d i c a t i o n s of some aspects of his l i f e s t y l e ) to these s e n s i t i v i t y , which i s an additional step to substantiate the r a t i o n a l e of the model and i t s p r e d i c t i v e q u a l i t y . The information to be fed into the c a l i b r a t i o n procedure i s to be c o l l e c t e d i n a questionnaire survey on users' behaviour under choice conditions. The model i s therefore a behavioural one; i t s basic function i s to explain the predict users' choices. This approach i s d i f f e r e n t , for example, from the propensity model approach, where users' preferences, rather than behaviour, are the basis for c a l i b r a t i o n . The model proposed i n t h i s study can be applied as a planning t o o l to demonstrate the impact of various trans-portation p o l i c i e s on users' choices. The model i s capable of providing estimates of the number of users that would be attracted to public t r a n s i t as a r e s u l t of, for example, introducing a new t r a n s i t system, improving the e x i s t i n g l e v e l of service, increasing parking charges or gasoline p r i c e s . Other applications include the assessment of the impact of introducing novel transportation modes on r i d e r s h i p under the assumption that t h e i r a t t r i b u t e s are comparable to the e x i s t i n g ones. Also, since the model accounts f o r c e r t a i n aspects of the i n d i v i d u a l ' s l i f e s t y l e , i t i s possible to l i n k changes i n the l a t t e r to h i s mode choice. v TABLE OF CONTENTS ACKNOWLEDGEMENT . . . . ABSTRACT . . . . . . . TABLE OF CONTENTS . LIST OF ILLUSTRATIONS . CHAPTER I - INTRODUCTION 1.1. Introductory Statement 1.2. Statement, of the Problem 1.3. Objectives of the Study 1.4. Hypothesis 1.5. Outline of Research 1.6. Scope and Limitation of the Study 1.7. Significance of Research CHAPTER II - THEORY OF MODE CHOICE U t i l i t a r i a n Theory of Mode Choice 2.4. The Concept 2.5. Evaluation. 2.6. Empirical Evidence, Value of Time 2.7. The D i s u t i l i t y of Comfort Level Perception, Attitudes and Behaviour 2.8. Perception and Attitudes 2.9. E f f e c t on Users Behaviour 2.10. Summary and Conclusions CHAPTER III - THE MODEL STRUCTURE AND APPLICATION . 3.1. Introduction 3.2. Research Orientation 3.3. Conceptual Considerations and 2.1. Introduction Abstract Mode Theory 2.2. 2.3. The Concept Evaluation Applications v i TABLE OF CONTENTS — continued PAGE The Users Socio-Economic C h a r a c t e r i s t i c s ; 3.4. The Household Income 3.5. Occupation 3.6. Age 3.7. Sex The Transportation System A t t r i b u t e s : 3.8. Problems of I d e n t i f i c a t i o n 3.9. Travel Time 3.10. Travel Cost 3.11. The Comfort A t t r i b u t e 3.12. Frequency of Tra n s i t Service The Model Formulation: 3.13. The Problem of C o l i n e a r i t y 3.14. An Overview of the Model 3.15. The Use of Stochastic Models i n Transportation Planning 3.16. The Model Structure 3.17. Summary CHAPTER IV - SURVEY AND QUESTIONNAIRE DESIGN . . . . . 109 4.1. Introduction 4.2. Survey Design 4.3. D e f i n i t i o n of the Sample Space 4.4. The Questionnaire Design CHAPTER V - SUMMARY AND CONCLUSIONS . . . . . . . . . 122 5.1. Introduction 5.2. Summary of the Study 5.3. Planning Implications 5.4. Research P o s s i b i l i t i e s BIBLIOGRAPHY \ . . . . . . . . . . . . . . 135 APPENDICES . . . . . . . 144 A. BUS USERS QUESTIONNAIRE . . . . . . . . . . 145 B. CAR USERS QUESTIONNAIRE . . , ' 149 v i i LIST OF ILLUSTRATIONS Fi g . PAGE 1. P r o b a b i l i t y of Mode Choice as a Function of Saving i n the Generalized Price . 24 2. a. Value of Time as a Function of Total Time Savings (Haney's Scheme) . 34 2. b. Marginal Benefits of Time Saved by T r i p Purpose (Income Group: $8000 - $9999) . 34 3. E f f e c t of Uniform Modal Bias Among Users On Their Mode Choice . . . . . . . 44 4. The Model Structural Relationship . 86 v i i i CHAPTER I INTRODUCTION 1.1. I n t r o d u c t o r y Statement P l a n n i n g i s concerned w i t h the f u t u r e . Hence, p r e d i c t i n g the c h a r a c t e r i s t i c s o f a f u t u r e s i t u a t i o n i s an e s s e n t i a l phase i n the p l a n n i n g p r o c e s s . A major t a s k f o r those engaged i n t r a n s p o r t a t i o n p l a n n i n g i s e s t i m a t i n g f u t u r e demand f o r v a r i o u s t r a n s p o r t a t i o n f a c i l i t i e s ; and based upon these e s t i m a t e s , and p o s s i b l y c o s t o r s o c i a l p r i o r i t i e s , recommendations are put forward f o r the c o n s t r u c t i o n o f the f a c i l i t i e s needed. Gross e r r o r s i n making these e s t i m a t e s l e a d to m i s p r e s e n t a t i o n of f u t u r e needs, and thereby f o s t e r i n g d e c i s i o n s which may d i s t o r t p r i o r i t i e s h e l d i n check by the community. In the e v o l u t i o n o f the p l a n n i n g process over the l a s t two decades, emphasis has been p l a c e d f i r s t on the accommodative f u n c t i o n o f the t r a n s p o r t a t i o n system, t h a t i s , to s a t i s f y the p r o j e c t e d t r a v e l demand by v a r i o u s modes, s u b j e c t to the economic c r i t e r i o n o f b e n e f i t maximization.^" The more r e c e n t t r e n d i s to p l a c e emphasis on the normative aspect of p l a n n i n g , a c c o r d i n g t o which the t r a n s p o r t a t i o n p o l i c y s h o u l d be formulated so as to encourage the use of c e r t a i n modes, w h i l e d i s c o u r a g i n g the use of o t h e r s . By t a k i n g i n t o account the environmental, s o c i a l and economic i m p l i c a t i o n s o f such a p o l i c y , the supply of d i f f e r e n t t r a n s p o r t a t i o n f a c i l i t i e s can be manipulated so as t o a f f e c t the demand f o r d i f f e r e n t modes, and thereby b r i n g about the d e s i r e d change.^ Whether the main concern i s the accommodation o f demand or the m a n i p u l a t i o n o f suppl y , c o n c r e t e knowledge o f mode c h o i c e determinants i s e s s e n t i a l t o e f f e c t i v e t r a n s -p o r t a t i o n p l a n n i n g . In the f i r s t case, the i d e n t i f i c a t i o n o f mode c h o i c e determinants i s necessary t o e x p l a i n and p r e d i c t u s e r s ' c h o i c e s . In the l a t t e r , i n f o r m a t i o n i s needed i on the p o l i c y v a r i a b l e s a f f e c t i n g u s e r s ' c h o i c e s so that° i n c e n t i v e s f o r mode s h i f t can be employed e f f e c t i v e l y . 1.2. Statement o f the Problem Over the l a s t two decades, mode c h o i c e a n a l y s i s has taken d i f f e r e n t o r i e n t a t i o n s . In e a r l i e r t r a n s p o r t a t i o n s t u d i e s , conducted i n the f i f t i e s and through the m i d - s i x t i e s , mode c h o i c e was e x p l a i n e d i n terms of the socio-economic c h a r a c -3 t e r i s t i c s o f the p o p u l a t i o n . The urban area was d i v i d e d i n t o geographic zones, each c o n t a i n i n g a p o p u l a t i o n r e f l e c t i n g a degree o f homogeneity i n terms o f s o c i a l , economic and demographic c h a r a c t e r i s t i c s . In e x p l a i n i n g mode c h o i c e , these c h a r a c t e r i s t i c s were taken i n t o account, and i n many i n s t a n c e s , o t h e r v a r i a b l e s were c o n s i d e r e d : t r i p c h a r a c t e r i s t i c s , such as t r i p l e n g t h and purposes; frequency o f d e p a r t u r e times of t r a n s i t as a measure f o r q u a l i t y o f s e r v i c e ; and p o s s i b l y an index o f l o c a t i o n , u s u a l l y the r e l a t i v e a c c e s s i b i l i t y of v a r i o u s 3 p a r t s of the c i t y by t r a n s i t and highway measured i n time u n i t s . These v a r i a b l e s were e n t e r e d i n t o a r e g r e s s i o n model which produced an es t i m a t e o f t r a n s i t users f o r each geographic zone. Other s t u d i e s u t i l i z e d these i n p u t s to a r r i v e a t the mode s p l i t r a t i o , i . e . , the r a t i o o f t r a n s i t t o c a r us e r s f o r the g i v e n zone. In a l a t e r phase of development, r e s e a r c h e r s have p l a c e d emphasis on the p e r s o n a l c h a r a c t e r i s t i c s o f the u s e r and the c h a r a c t e r i s t i c s o f h i s mode c h o i c e s i t u a t i o n , i . e . , h i s socio-economic c h a r a c t e r i s t i c s , and the d i f f e r e n c e between 4 the a t t r i b u t e s o f a l t e r n a t i v e modes a v a i l a b l e t o him. The argument was advanced t h a t by ac c o u n t i n g f o r the c h a r a c t e r i s t i c s of the i n d i v i d u a l u s e r r a t h e r than the z o n a l p o p u l a t i o n , i . e . , the d i s a g g r e g a t e v e r s u s the aggregate data e n t r y , a more a c c u r a t e 5 d e s c r i p t i o n c o u l d be made f o r the i n d i v i d u a l o b s e r v a t i o n . T h i s seemed t o be an improvement i n the o p e r a t i o n a l c h a r a c t e r i s t i c s o f the model. But t h e r e were c e r t a i n improve-ments i n the c o n c e p t u a l framework o f the a n a l y s i s as w e l l . The d e c i s i o n r o l e o f mode c h o i c e was to take p l a c e a t the i n d i v i d u a l u s e r ' s l e v e l , not by a geographic zone, and thus accounting f o r the i n d i v i d u a l ' s behaviour would r e f l e c t g r e a t e r s i m i l a r i t y t o the r e a l world, which i s an important f e a t u r e of r e l i a b l e b e h a v i o u r a l models.^ Another c o n c e p t u a l advantage a s s o c i a t e d w i t h the use of d i s a g g r e g a t e d models was t h e i r concern w i t h the p r o b a b i l i t y of a person making one mode c h o i c e o r the o t h e r , r a t h e r than the mere p r o d u c t i o n o f an e s t i m a t e of model u s e r s by employing d e t e r m i n i s t i c models. Some r e s e a r c h e r s h e l d the view t h a t p r e d i c t i n g a person's c h o i c e s h o u l d be expressed as a probab-i l i t y (there i s 60 per c e n t chance he would s e l e c t t h i s mode). The p r o b a b i l i t y f u n c t i o n r e l a t i n g c h o i c e determinants t o a s p e c i f i c c h o i c e i s u s u a l l y n o n - l i n e a r , y e t d e t e r m i n i s t i c models, i t was argued, s i m p l i f i e d t h i s r e l a t i o n s h i p by u s i n g r e g r e s s i o n 7 f u n c t i o n s . A completely d i f f e r e n t approach was pursued by r e s e a r c h e r s whose i n t e r e s t was the marketing o f t r a n s i t s e r v i c e s among c a r users who had access t o t r a n s i t . In a q u e s t i o n n a i r e survey, c a r us e r s were asked t o s t a t e how important each of the a t t r i b u t e s o f c a r and t r a n s i t , and the ex t e n t t o which they were s a t i s f i e d w i t h each mode w i t h r e s p e c t t o each 8 i n d i v i d u a l a t t r i b u t e . The us e r ' s a t t i t u d e s , or r a t h e r h i s s e n s i t i v i t y toward modal a t t r i b u t e s , expressed q u a n t i t a t i v e l y along a graduated 5-point L i n k e r t s c a l e , were f e d i n t o a s e t of mathematical f u n c t i o n s t o p r e d i c t u s e r s ' c h o i c e s . The socio-economic c h a r a c t e r i s t i c s , a t l e a s t i n some i n s t a n c e s , were not accounted f o r . The q u e s t i o n which may a r i s e as a r e s u l t of the l i t e r a t u r e review i s whether any o f these models employed a complete s e t of the c h o i c e determinants r e l e v a n t to the problem. In o t h e r words, s i n c e both types o f models have p r o v i d e d some explan-a t i o n f o r u s e r s ' behaviour, would the a d d i t i o n o f one s e t of determinants t o the oth e r c o n t r i b u t e t o the expansion o f the e x p l a n a t o r y power o f the model? I f so, what are the r a m i f i -c a t i o n s o f t h i s to the co n c e p t u a l a n a l y s i s o f mode c h o i c e , and how would the model s t r u c t u r e be a f f e c t e d as a r e s u l t ? These q u e s t i o n s are the main concern of t h i s study. 1.3. O b j e c t i v e s o f the Study Three major o b j e c t i v e s are to be pursued i n t h i s study: (a) To i d e n t i f y mode c h o i c e determinants i n urban areas i n as f a r as the r e i s evidence i n the l i t e r a t u r e t o support t h e i r s i g n i f i c a n c e i n e x p l a i n i n g u s e r s ' b e h a v i o u r . (b) To develop a co n c e p t u a l framework which p r o v i d e s a r a t i o n a l e f o r the r e l a t i o n s h i p between u s e r s ' b e h a v i o u r and i t s determinants. (c) To formulate a mode c h o i c e model which makes use o f t h i s c o n c e p t u a l framework and employs the c h o i c e determinants i d e n t i f i e d i n the study as the i n p u t v a r i a b l e s . I t i s c l e a r from the statement o f o b j e c t i v e s t h a t t h i s study i s a c o n t i n u a t i o n f o r p r e v i o u s r e s e a r c h on mode c h o i c e a n a l y s i s , and t h a t the u l t i m a t e purpose o f t h i s and pr e v i o u s work i s to expand the e x p l a n a t o r y power o f the mode c h o i c e models and t h e i r a b i l i t y t o p r e d i c t u s e r s ' b e h a v i o u r . Thus, the study attempts to improve upon the p r e s e n t s t a t e - o f -t h e - a r t , at l e a s t c o n c e p t u a l l y , the s t a t e - o f - t h e - a r t b e i n g the use o f socio-economic c h a r a c t e r i s t i c s and. c h a r a c t e r i s t i c s o f the mode c h o i c e s i t u a t i o n as model i n p u t s . The p o s s i b l e improvement to be e x p l o r e d here i s the a d d i t i o n o f the user' s 6 s e n s i t i v i t y toward modal a t t r i b u t e s to o t h e r mode c h o i c e d e t e r m i n a n t s . In any case, the use of the mode c h o i c e determin-ants a l r e a d y i d e n t i f i e d i n p r e v i o u s works, t o g e t h e r w i t h those to be advanced i n t h i s study, should be i n c o r p o r a t e d i n a c o n c e p t u a l framework whose r a t i o n a l e can be defended e i t h e r on a p r i o r i ground, or on the b a s i s o f p r e v i o u s r e s e a r c h f i n d i n g s . 1.4. C o n c e p t u a l Framework P r e v i o u s s t u d i e s have a l r e a d y e s t a b l i s h e d t h a t s o c i o -economic c h a r a c t e r i s t i c s are s i g n i f i c a n t i n e x p l a i n i n g u s e r s ' 9 mode c h o i c e s . But th e r e are i n d i c a t i o n s t h a t they a re not the o n l y p e r s o n a l c h a r a c t e r i s t i c s which c o u l d be s i g n i f i c a n t and t h a t o t h e r s should be taken i n t o c o n s i d e r a t i o n . Research has a l r e a d y i d e n t i f i e d the i n d i v i d u a l ' s p e r c e p t i o n and a t t i t u d e s as r e l e v a n t t o the a n a l y s i s . I t can a l s o be e a s i l y demonstrated t h a t the i n d i v i d u a l c ircumstances may a f f e c t h i s mode c h o i c e b e h a v i o u r . As W i l l i a m s o n and Moses' a n a l y s i s has demonstrated, the p r o b a b i l i t y of choo s i n g between two modes, one o f which i s more c o s t l y but takes l e s s time, might depend on whether the i n d i v i d u a l c o u l d c o n v e r t the time d i f f e r e n c e i n t o working time. Or, i f the i n d i v i d u a l had the o p p o r t u n i t y t o spend the time saving i n an agreeabl e a c t i v i t y worthy of the e x t r a c o s t , he may then, even o c c a s i o n a l l y , s h i f t t o the f a s t e r mode. Thus, i n g e n e r a l , the i n d i v i d u a l ' s s e n s i t i v i t y toward modal a t t r i b u t e s may r e f l e c t h i s c i r c u m s t a n c e s , (and p o s s i b l y h i s l i f e s t y l e ) , but not n e c e s s a r i l y h i s socio-economic c h a r -*- • 11 a c t e r i s t i c s . The primary hypothesis of t h i s study can be stated as follows: the user's mode choice behaviour i s affected by four sets of determinants: his socio-economic c h a r a c t e r i s t i c s , the perceived a t t r i b u t e s of the transportation system, the si g n i f i c a n c e he attaches to the i n d i v i d u a l a t t r i b u t e s ( i . e . , his s e n s i t i v i t y toward these a t t r i b u t e s ) , and the circumstances of the t r i p , e.g., the t r i p purpose. I t i s hypothesized here that the user's s e n s i t i v i t y toward modal a t t r i b u t e s i s not strongly correlated with his socio-economic c h a r a c t e r i s t i c s , . except to the extent that the l a t t e r contributes to the explan-ation.of the^individual's l i f e s t y l e . In pursuing t h i s concept further to what might lead to the causal underpinnings of users' behaviour, attention must be given to the possible r e l a t i o n s h i p between the user's s e n s i t i v i t y toward modal attr i b u t e s and his i n d i v i d u a l circumstances. We may also hypothesize that the l a t t e r can be attributed to the user's l i f e s t y l e — l i f e s t y l e being the pattern by which the i n d i v i d u a l user allocates his time and monetary which he i s engaged, here as a r a t i o n a l i -resources among various a c t i v i t i e s i n This set of relationships i s advanced zation f o r the study's conceptual analysis and possibly for the model formulation, but since empirical data i s lacking i n this respect, no attempt i s made i n the study to support these relationships except on a p r i o r i ground. 1.5. Outline of Research In est a b l i s h i n g a conceptual framework for the study's 8 model, i t i s worthwhile t o make use of the e x i s t i n g concepts and t h e o r i e s on mode c h o i c e b e h a v i o u r : (a) the a b s t r a c t mode c h o i c e theory which p l a c e s emphasis on the modal a t t r i b u t e s r a t h e r than i t s i n s t i t u t i o n a l form, e.g., bus, t r a i n , e t c . , (b) the u t i l i t a r i a n t heory o f mode c h o i c e , which r e l a t e s the p r o b a b i l i t y of making a c e r t a i n c h o i c e to the d i s u t i l i t y savings made by such a c h o i c e , (c) the theory of time v a l u e formulated to e x p l a i n the t r a d e - o f f s between time and t r a v e l c o s t s as evidenced i n the u s e r s ' mode c h o i c e b e h a v i o u r , and (d) concepts and t h e o r i e s r e l a t e d t o the i n d i v i d u a l ' s p e r c e p t i o n and a t t i t u d e , a n d - t h e i r e f f e c t on b e h a v i o u r . In a d d i t i o n , the s t a t e d o b j e c t i v e s o f the study c a l l f o r the f o r m u l a t i o n o f a c o n c e p t u a l framework of a model which promises a g r e a t e r e x p l a n a t o r y power than those a l r e a d y i n use. The refinement thought to ac h i e v e t h i s improved c a p a c i t y i s h i n t e d a t p r e v i o u s l y , t h a t i s , by a c c o u n t i n g f o r the us e r ' s s e n s i t i v i t y toward modal a t t r i b u t e s . Yet, f o r t h i s r efinement t o be a worthy u n d e r t a k i n g , s u f f i c i e n t evidence must be brought forward t o i n d i c a t e t h a t no s t r o n g c o r r e l a t i o n e x i s t s between the user' s s e n s i t i v i t y toward modal a t t r i b u t e s and h i s socio-economic c h a r a c t e r i s t i c s . Such evid e n c e w i l l be sought i n p r e v i o u s r e s e a r c h f i n d i n g s . For the purposes o f model f o r m u l a t i o n , o t h e r mode c h o i c e determinants should a l s o be a s s e s s e d i n d i v i d u a l l y : the u s e r ' s socio-economic c h a r a c t e r i s t i c s and t h e ch a r a c -t e r i s t i c s o f h i s mode c h o i c e s i t u a t i o n , i . e . , d i f f e r e n c e s between the a t t r i b u t e s o f the a l t e r n a t i v e modes a v a i l a b l e t o 9 him. The m a n i p u l a t i o n o f these as model i n p u t s r e q u i r e s a c a r e f u l c o n s i d e r a t i o n t o ensure i t s c o m p a t i b i l i t y w i t h the concepts developed i n the study. Again, the i n f o r m a t i o n r e q u i r e d f o r t h i s s t e p can be d e r i v e d from p r e v i o u s r e s e a r c h c o n c l u s i o n s . For reasons d i s c u s s e d i n S e c t i o n 1.2., the model to be formu l a t e d here i s a d i s a g g r e g a t e d one. I n i t i a l l y , a l l u s e r s are t o be c l a s s i f i e d i n t o " s e n s i t i v i t y groups", a c c o r d i n g to the modal a t t r i b u t e t o which they are most s e n s i t i v e . D i s c r i m -i n a n t a n a l y s i s can then be used i n c o n j u n c t i o n w i t h a p r o b a b i l i t y model t o e x p l a i n and p r e d i c t u s e r s ' mode c h o i c e w i t h i n each group. The u s e r ' s socio-economic c h a r a c t e r i s t i c s , and a t t r i b u t e s of the mode c h o i c e s i t u a t i o n are t o be employed i n d e r i v i n g the d i s c r i m i n a n t f u n c t i o n . 1 . 6 . - Scope and L i m i t a t i o n o f the Study Although the c o n c e p t u a l a n a l y s i s o f t h i s study i s r e l e v a n t to any mode c h o i c e s i t u a t i o n , the model to be form-u l a t e d here i s s p e c i f i c a l l y designed f o r a p p l i c a t i o n to urban areas which can be c o n s i d e r e d as medium o r l a r g e s i z e c i t i e s . T h i s p a r t i c u l a r range i s d i c t a t e d by the model c a l i b r a t i o n r equirements. To d e r i v e a b e h a v i o u r a l model o f the k i n d b e i n g c o n s i d e r e d here, t h e r e should be s u f f i c i e n t o b s e r v a t i o n s t o cover a wide range o f mode c h o i c e s i t u a t i o n s . T h i s can be achi e v e d i n a u s e r ' s sample which i s s p a t i a l l y d i s p e r s e d over a r e l a t i v e l y l a r g e geographic a r e a . I t i s suggested t h a t a sample c o l l e c t e d i n a s m a l l c i t y may not s a t i s f y t h i s c r i t e r i o n , 10 and moreover, the t r a n s i t s e r v i c e i n such c i t i e s i s minimal, i f at a l l e x i s t e n t , and hence does not o f f e r a r e a l a l t e r n a t i v e to the c a r . The t h r e s h o l d s e p a r a t i n g s m a l l from medium c i t i e s i s s e t a r b i t r a r i l y by t h i s author at the p o p u l a t i o n o f t h r e e hundred thousand. T h i s r e s e a r c h i s s e t to formulate a mode c h o i c e model, but does not attempt an a c t u a l ' d a t a a n a l y s i s o r c a l i b r a t i o n o f the model i n q u e s t i o n . T h e r e f o r e evidence brought forward t o s u b s t a n t i a t e the s e l e c t i o n o f any of the model i n p u t s , as w e l l as the j u s t i f i c a t i o n f o r v a r i o u s procedures t o be adopted, are to be drawn t o t a l l y from p r e v i o u s r e s e a r c h f i n d i n g s . 1.7. S i g n i f i c a n c e of Research The e s c a l a t i n g p r i c e s o f o i l products as a r e s u l t o f an ever i n c r e a s i n g demand upon a d e p l e t i n g r e s o u r c e , the bl e a k p r o s p e c t o f co n t i n u e d p r i c e e s c a l a t i o n , t o g e t h e r w i t h v a r i o u s environmental c o n s i d e r a t i o n s , have renewed the i n t e r e s t i n p u b l i c t r a n s p o r t a t i o n to meet the community needs f o r m o b i l i t y . P r e v i o u s l y , p l a n n e r s who advocated g r e a t e r r e l i a n c e on p u b l i c t r a n s p o r t a t i o n had d i f f i c u l t y , at l e a s t i n some i n s t a n c e s , i n r e c r u i t i n g support f o r such a p o l i c y . The more r e c e n t trends i n g a s o l i n e p r i c i n g s t r e n g t h e n e d the p l a n n e r s ' argument t h a t a r e a l a l t e r n a t i v e t o the c a r must be o f f e r e d . For such a t r a n s p o r t a t i o n p o l i c y to be e f f e c t i v e , i n f o r m a t i o n i s r e q u i r e d on the impact of a l t e r i n g the mode c h o i c e s i t u a t i o n (by improving the t r a n s i t s e r v i c e , f o r 11 example) on v a r i o u s s o c i a l o r economic groups o f the p o p u l a t i o n . T h i s study, along w i t h o t h e r s on the s u b j e c t of mode c h o i c e , attempts to e x p l a i n and p r e d i c t u s e r s ' behaviour i n response t o changes i n the mode c h o i c e s i t u a t i o n , whether i t i s a planned change, such as i n c r e a s i n g the frequency of t r a n s i t s e r v i c e , o r "unplanned" change, such as the i n c r e a s e i n g a s o l i n e p r i c e s . The refinement proposed here, by a c c o u n t i n g f o r the u s e r ' s s e n s i t i v i t y toward model a t t r i b u t e s , i s thought t o expand the e x p l a n a t o r y power of the model. I f t h i s o b j e c t i v e can be ac h i e v e d , the planner can b e n e f i t from t h i s improved c a p a c i t y i n s e v e r a l ways. F i r s t , the i n i t i a t i o n o r the improvement t o t r a n s i t s e r v i c e can be guided a c c o r d i n g l y so as to meet d i f f e r e n t u s e r ' s c r i t e r i a i n v a r i o u s p a r t s o f the c i t y . L i k e w i s e , the pla n n e r may be b e t t e r equipped t o p r e d i c t the user's response t o g a s o l i n e p r i c e s , f o r example. Secondly, by l i n k i n g the u s e r ' s l i f e s t y l e , o r some measure of i t , t o h i s mode c h o i c e b e h a v i o u r , i t would be p o s s i b l e t o a n t i c i p a t e change i n demand f o r v a r i o u s modes as a r e s u l t o f some s o c i a l or economic changes, e.g., r e d u c t i o n of the weekly working hours and i n c r e a s e i n l e i s u r e time. F i n a l l y , improved knowledge of u s e r s ' e v a l u a t i o n s o f modal a t t r i b u t e s i s progress toward p r e d i c t i n g demand f o r n o v e l t r a n s p o r t a t i o n modes. O b v i o u s l y , c o n v e n t i o n a l methods based on l i n e a r o r curve e x t r a p o l a t i o n would not be h e l p f u l i n e s t i m a t i n g demand f o r such modes.' The precedence upon which e x t r a p o l a t i o n can be based i s n o n - e x i s t e n t . F o r t h i s purpose, the u t i l i t a r i a n 12 theory of mode choice could be more useful because of i t s a p p l i c a b i l i t y to "abstract modes," modes which are described by t h e i r t r a v e l d i s u t i l i t i e s , and not by it's i n s t i t u t i o n a l form. 13 FOOTNOTES 1. A b r i e f a n a l y s i s of t h i s approach i s p r e s e n t e d i n Modal S p l i t : Documentation of Nine Methods f o r E s t i m a t i n g T r a n s i t Usage, by M a r t i n J . F e r t a l e t a l . , U. S. Department o f Commerce, 1966. 2. T h i s t r e n d i s s t r o n g l y expressed by G e r a l d R. Brown's study, Mode Choice Determinants of S e l e c t e d S o c i o - economic Groups: An Investigation o f a P l a n n i n g and  C o n t r o l Mechanism to D i r e c t Automobile D r i v e r s t o  P u b l i c T r a n s p o r t a t i o n , Ph.D. i n the Department of Community and R e g i o n a l P l a n n i n g , The U n i v e r s i t y o f B r i t i s h Columbia, 1971, p u b l i s h e d by the Department of C i v i l E n g i n e e r i n g , 1972. 3. A d e t a i l e d a n a l y s i s o f the r e c e n t e v o l u t i o n of mode use p r e d i c t i o n techniques i s p r o v i d e d i n "Disaggregate S t o c h a s t i c Models of T r a v e l Mode C h o i c e , " by P e t e r R. Stopher and Schalom Reichman, Highway Research  Record, Rep. No. 369, 1971, pp. 91-103. 4. The s u b j e c t s of t h i s study, i . e . , the t r a v e l e r s w i t h whom we a r e concerned i n the a n a l y s i s , are r e f e r r e d to i n two d i f f e r e n t ways. The term " u s e r s " i s mentioned i n c o n n e c t i o n w i t h any group of persons u s i n g a c e r t a i n mode, e.g., car u s e r s , or persons whose behaviour i s r e l e v a n t t o mode c h o i c e a n a l y s i s , e.g., u s e r ' s modal b i a s . In the ensuing d i s c u s s i o n s on l i f e s t y l e , o r s e n s i t i v i t y toward modal a t t r i b u t e s , t h i s term may a l s o be used i f t h e purpose of the d i s c u s s i o n i s d i r e c t l y r e l a t e d t o mode c h o i c e . Other-wise, the study s u b j e c t s a r e r e f e r r e d t o as i n d i v i d u a l s or persons. T h i s d i s t i n c t i o n i s i n harmony w i t h the c u r r e n t usage i n the l i t e r a t u r e , 5. Stopher and Reichman, op. c i t . , p. 94. 6. I b i d . 7. I b i d . , pp. 91-113. 8. See, f o r example, " I n v e s t i g a t i o n o f the E f f e c t of T r a v e l e r A t t i t u d e s i n a Model o f Mode C h o i c e Behaviour," Highway Research Record, Rep. No. 36 9, 1971, pp. 1-14. 14 F e r t a l , Modal.Split: Documentation of Nine Methods. Leon Moses and H. Williamson, "Value of Time, Choice of Mode and the Subsidy Issue i n Urban Transportation, Journal of P o l i t i c a l Economy, Vol. 71, 1963, pp. 247 -264. This concept was hinted to by F. X. De Donnea, The -Determinants of Transport Mode Choice i n Dutch C i t i e s :  Some Disaggregate Stochastic Models, Rotterdam University Press, Rotterdam, 19 71, p. 157. Attributes of the transportation system r e f e r to the a c t i v i t i e s involved i n using the a l t e r n a t i v e transport modes availa b l e to the user, e.g., walking, waiting, in-vehi c l e t r a v e l time, etc. CHAPTER II THEORY OF MODE CHOICE 2.1. Introduction The formulation of a p l a u s i b l e theory explaining users 1 mode choices i s a major task to be undertaken i n the course of th i s study. The "new approach" to the consumer theory as advanced by Lancaster and others, and i t s p a r a l l e l i n the transportation planning f i e l d , the abstract mode choice theory, are introduced i n this chapter. Within the framework estab-l i s h e d by these theories, a transport mode i s "reduced" to a "bundle" of at t r i b u t e s , or a combination of c h a r a c t e r i s t i c s , and users 1 choices are explained as a function of the r e l a t i v e u t i l i t y or d i s u t i l i t y associated with each possible s e l e c t i o n . Attention i s also given to the u t i l i t a r i a n theory of mode choice, since i t places emphasis on how these a t t r i b u t e s are weighted by d i f f e r e n t i n d i v i d u a l s , and the bearing t h i s may have on t h e i r behaviour. The range and complexity of the hypothesized relationships between mode at t r i b u t e s and users' choices are investigated and evidence supporting t h e i r v a l i d i t y i n previous research i s presented below.- F i n a l l y , we may address the problem of perception and at t i t u d e s , and the e f f e c t of these psychological processes on users' behaviour. 15 1 6 ABSTRACT MODE THEORY 2.2. The Concept Major transportation studies conducted i n the f i f t i e s and the s i x t i e s have r e f l e c t e d the influence of the c l a s s i c a l consumer theory i n providing a r a t i o n a l e f o r mode choice analysis.^ According to thi s theory, the user selects a combination of goods which gives him the maximum u t i l i t y under the given budget constraint. Hence, the l i m i t a t i o n i s on the monetary resources and the choice i s among goods. Viewed i n th i s context, mode choice i s equivalent to the s e l e c t i o n of the good which, together with other goods, gives the user the maximum u t i l i t y possible under the given constraint. A departure from this conventional approach was made 2 i n the new approach to consumer theory. This approach placed emphasis on the c h a r a c t e r i s t i c s of the goods rather than the goods themselves. Goods per se do not give r i s e to u t i l i t y nor provide s a t i s f a c t i o n ; rather, the q u a l i t i e s they possess. Further, these q u a l i t i e s are not possessed excl u s i v e l y by one good, but are generally shared among several goods. The v choice i s therefore not among goods, as suggested i n the c l a s s i c a l consumer theory, but among combinations of char-a c t e r i s t i c s . The r a t i o n a l user would then s e l e c t a c o l l e c t i o n of goods whose combined c h a r a c t e r i s t i c s give him a greater s a t i s f a c t i o n than any other possible combination. This simple modification of the consumer theory provides us with an enlightened view of the transportation mode c h o i c e problem. A t r a n s p o r t mode i n t h i s c o n c e p t u a l a n a l y s i s i s viewed as a "bundle" of a t t r i b u t e s : speed, c o s t , l e v e l o f comfort, appearance, e t c . , but the p o p u l a r image o f the mode, whatever t h i s may be, i s i s o l a t e d from the mode c h o i c e problem. The emphasis i s t h e r e f o r e p l a c e d on the u t i l i t a r i a n a spect o f the t r a n s p o r t v e h i c l e , r a t h e r than i t s i n s t i t u t i o n a l form: bus, t r a i n , c a r , e t c . Hence, the c h o i c e problem i s 3 reduced to s e l e c t i o n of combination o f a t t r i b u t e s . T h i s new approach i s a break w i t h the t r a d i t i o n a l approach to mode c h o i c e a n a l y s i s , where the r e s e a r c h e r was concerned w i t h the d i r e c t r e l a t i o n s h i p between the us e r s 1 socio-economic c h a r a c t e r i s t i c s and t h e i r c h o i c e s among two modes, c a r and t r a n s i t . Some s t u d i e s r e c o g n i z e d one o r few mode a t t r i b u t e s , such as t r a v e l time, as v a r i a b l e s a c c o u n t i n g 4 f o r u s e r s ' mode c h o i c e s . But, i n g e n e r a l , the system a t t r i b u t e s were not g i v e n adequate c o n s i d e r a t i o n and the e n t i r e a n a l y s i s was mode-oriented, i . e . , c a r or t r a n s i t , not a t t r i b u t e - o r i e n t e d . 2.3. E v a l u a t i o n Having i n t r o d u c e d the a b s t r a c t mode t h e o r y , and i d e n t i f i e d the d i f f e r e n c e s between t h i s and the c o n v e n t i o n a l mode s p l i t a n a l y s i s , we can proceed to d i s c u s s the p o s s i b l e advantages o f the new approach: (a) The most s i g n i f i c a n t outcome o f any t r a n s p o r t a t i o n p l a n n i n g model i s the i d e n t i f i c a t i o n o f the r e l a t i o n s h i p o f p o l i c y v a r i a b l e s t o u s e r s ' behaviour. O b v i o u s l y t h i s would be u s e f u l i n f o r m a t i o n f o r p l a n n i n g purposes, p a r t i c u l a r l y i f 18 the main o b j e c t i v e i s t o a f f e c t u s e r s ' b e h a v i o u r . 5 T h i s i n f o r m -a t i o n i s a f f o r d e d o n l y by a model i n which the system a t t r i b u t e s are accounted f o r . To the ex t e n t t h a t these a t t r i b u t e s are a l t e r a b l e , and t h a t they a f f e c t u s e r s ' c h o i c e s , such a model i s a proper t o o l which can be used i n the f o r m u l a t i o n o f t r a n s p o r t a t i o n p o l i c y . The more advanced i s our knowledge on the r e l a t i o n s h i p expressed i n the model, the more e f f e c t i v e the t r a n s p o r t a t i o n p o l i c y i s l i k e l y t o be. In c o n t r a s t , a model which e x p l a i n s mode c h o i c e i n terms of the socio-economic v a r i a b l e s e x c l u s i v e l y g i v e s us no guidance as to what measures can be a p p l i e d t o a f f e c t u s e r s ' c h o i c e s - - s i n c e u s e r s 1 socio-economic c h a r a c t e r i s t i c s are not p o l i c y v a r i a b l e s . Due to t h i s shortcoming, such a model would a f f o r d l i m i t e d i n f o r m a t i o n about, f o r example, an e s t i m a t e o f t r a n s i t users i f s e r v i c e i s extended to a g i v e n p a r t o f the c i t y . The assumption under which t h i s e s t i m a t e can be made i s t h a t the l e v e l of s e r v i c e i s the same as o t h e r p a r t s of the c i t y f o r which the model was o r i g i n a l l y c a l i b r a t e d . On the o t h e r hand, a model which accounts f o r the system a t t r i b u t e s would produce an e s t i m a t e o f users who would be a t t r a c t e d to any g i v e n mode as a r e s u l t o f improving i t s s e r v i c e , o r undermining the c h a r a c t e r i s t i c s o f competing modes. That i s , we would be able t o es t i m a t e the i n c r e a s e i n t r a n s i t r i d e r s h i p as a r e s u l t o f i n c r e a s i n g i t s frequency, l o w e r i n g the f a r e , o r r a i s i n g the p a r k i n g charges. (b) The system a t t r i b u t e s express the i n f l u e n c e o f many exogeneous v a r i a b l e s which can not be accounted f o r 19 i n the mode c h o i c e model, but which may have a d e f i n i t e impact on u s e r s ' behaviour. For example, the i n c r e a s e i n g a s o l i n e p r i c e s , as a r e s u l t o f i n c r e a s e d demand on a d e p l e t i n g r e s o u r c e , would have a s t r o n g impact on u s e r s ' b e h a v i o u r which can not be accounted f o r without i n c l u d i n g the c o s t a t t r i b u t e . But there are o t h e r uses f o r modal a t t r i b u t e s which are more s u b t l e . Suppose, f o r example, t h a t government and b u s i n e s s i n s t i t u t i o n s opted f o r s t a g g e r i n g working hours t o save t h e i r employees the problem o f t r a v e l l i n g i n the peak hours c o n g e s t i o n . Would t h i s have an e f f e c t on u s e r s ' mode c h o i c e s ? In some i n s t a n c e s t h e r e would be a s u b s t a n t i a l impact. We may be reminded t h a t t r a n s i t o p e r a t i n g on a s e p a r a t e r i g h t - o f - w a y has a s t r o n g advantage over the c a r i n the peak hours, when t r a f f i c movement i s slow. To spread the peak hour t r a v e l demand over many hours may a l l e v i a t e the peak c o n g e s t i o n , and thereby r e d u c i n g the r e l a t i v e advantage o f t r a n s i t use. Again, without a c c o u n t i n g f o r t r a v e l time by a l t e r n a t i v e modes, the impact of such a change can not be i d e n t i f i e d i n the model's output. I t i s obvious from the f o r e g o i n g d i s c u s s i o n t h a t by t a k i n g modal a t t r i b u t e s i n t o account, we would be b e t t e r equipped t o a n t i c i p a t e the impact o f some s o c i a l o r economic changes on u s e r s ' behaviour. (c) In s t u d y i n g u s e r s ' b e h a v i o u r i n any c h o i c e s i t u a t i o n , the i d e n t i f i c a t i o n o f c a u s e - e f f e c t r e l a t i o n s h i p s i s a h i g h l y d e s i r a b l e o b j e c t i v e . T h i s i s p a r t i c u l a r l y t r u e i n the p l a n n i n g f i e l d where a g r e a t s i g n i f i c a n c e i s a t t a c h e d 20 to the i d e n t i f i c a t i o n o f means by which u s e r s ' b e h a v i o u r can be a f f e c t e d . But, u n f o r t u n a t e l y , t h i s can not be accomplished by any of the s t a t i s t i c a l techniques y e t known. In a mathe-m a t i c a l model, we have no evidence to suggest t h a t the r e l a t i o n -s h i p expressed, no matter how s t r o n g , encounters those v a r i a b l e s d i r e c t l y c a u s i n g these r e l a t i o n s h i p s to come i n t o e f f e c t . In e v a l u a t i n g b e h a v i o u r a l models, equal a t t e n t i o n s h o u l d be g i v e n to t h e i r s t a t i s t i c a l s i g n i f i c a n c e as w e l l as the p l a u s i b i l i t y and l o g i c t h a t i s r e f l e c t e d i n the r e l a t i o n -s h i p expressed. A v a l i d model s h o u l d be s u p p o r t e d on both accounts. For p l a n n i n g purposes, as we have argued p r e v i o u s l y , the u t i l i t y o f the model i n c r e a s e s as i t s c o n t e n t approximates more c l o s e l y the i m p l i e d c a u s e - e f f e c t r e l a t i o n s h i p . Using t h i s as a c r i t e r i o n f o r e v a l u a t i n g mode choice, models, the advantages f o r a c c o u n t i n g f o r mode c h o i c e a t t r i b u t e s become more obvio u s . The a s s o c i a t i o n between u s e r s ' mode c h o i c e s and t h e i r socio-economic c h a r a c t e r i s t i c s does not e x h i b i t a s t r o n g c a u s a l r e l a t i o n s h i p . Does income o r age, f o r example, cause the u s e r s to s e l e c t a c e r t a i n mode and a v o i d the use of o t h e r s ? T r a d i t i o n a l l y , we have accepted t h i s r e l a t i o n s h i p w i t h o u t q u e s t i o n i n g i f s u f f i c i e n t e x p l a n a t i o n i s p r o v i d e d w i t h i n the model. Regardless o f the s t a t i s t i c a l s i g n i f i c a n c e of such a model, i t i s the o p i n i o n o f t h i s author t h a t the l o g i c of such a r e l a t i o n s h i p i s not adequate. I t i s argued here t h a t the. i n c l u s i o n o f the system a t t r i b u t e s would i n c r e a s e the e x p l a n a t o r y c o n t e n t of the model, not o n l y s t a t i s t i c a l l y , b ut l o g i c a l l y as w e l l . T h i s 21 i s not d i f f i c u l t to demonstrate. In the f i r s t case, the model would state that higher income groups would s e l e c t mode A, and lower income groups mode.B. The i n c l u s i o n of the mode attributes would produce a model s t a t i n g , f o r example, that higher income groups are l i k e l y to choose mode A given that i t i s faster than mode B, while lower income groups are l i k e l y 7 to use mode B given that i t s use i s less c o s t l y than mode A. It i s obvious that the l a t t e r account i s more s a t i s f y i n g , and i s closer to express a causal r e l a t i o n s h i p . (d) A model which accounts for the system at t r i b u t e s can possibly be employed to estimate demand for new trans-portation modes. These new modes are now sought by the transportation industry to a l l e v i a t e the pressing problems of street congestion, p o l l u t i o n , and to meet the challenge of the energy c r i s i s . The vehicle or the system which may be designed for this purpose could be r a d i c a l l y d i f f e r e n t from those known to the users. This would cause d i f f i c u l t y to the planner who attempts to pre d i c t users' responses to the new mode. The conventional c a l i b r a t i n g procedure based upon data c o l l a t e d from an e x i s t i n g mode choice s i t u a t i o n i s no longer applicable, simply because such a s i t u a t i o n does not yet e x i s t . This d i f f i c u l t y can be p a r t i a l l y overcome by studying the users' responses to the differences between at t r i b u t e s of alter n a t i v e modes. Such information can be derived from an exis t i n g or a previous mode choice s i t u a t i o n , and can be related to the differences created by the introduction of the new mode. 22 T u r n i n g our a t t e n t i o n now to the l i m i t a t i o n o f the a b s t r a c t mode c h o i c e approach, two problems become immediately apparent. The f i r s t i s the e x t e n t t o which a t r a v e l mode can be reduced to a "bundle" o f a t t r i b u t e s w i t h o u t e l i m i n a t i n g any of the c h a r a c t e r i s t i c s r e l e v a n t t o the problem. Although many r e s e a r c h e r s have i n v e s t i g a t e d u s e r s ' p e r c e p t i o n o f modal a t t r i b u t e s , no s y s t e m a t i c methodology has been developed to i d e n t i f y the p e r c e i v e d r e l e v a n t modal a t t r i b u t e s . The second problem a s s o c i a t e d w i t h the a b s t r a c t mode approach i s the exte n t t o which the 'relevant a t t r i b u t e s are comparable from one mode to the o t h e r . F o r example,, how does the f l e x i b i l i t y o f r o u t i n g , o r the freedom o f choo s i n g the dep a r t u r e time, i n h e r e n t t o c a r use compare w i t h the f i x e d r o u t e and schedules a s s o c i a t e d w i t h the t r a n s i t use? These two problems can be c o n s i d e r e d as the b a s i c l i m i t a t i o n o f the approach, which l i m i t a t i o n i s accentuated f u r t h e r as the study p r o g r e s s e s toward the model d e s i g n . UTILITARIAN THEORY OF MODE CHOICE - 2.4. The Concept In f o r m u l a t i n g a c o n c e p t u a l framework f o r users 1 behaviour i n mode c h o i c e s i t u a t i o n s , the u t i l i t a r i a n theory of mode c h o i c e complements the a b s t r a c t mode th e o r y . Together, they form a u s e f u l s e t o f hypotheses. In the l a t t e r , emphasis i s p l a c e d on the a t t r i b u t e s o f the t r a n s p o r t mode, w h i l e the former r e l a t e s users 1 c h o i c e s to t h e i r p e r c e p t i o n o f such 23 a t t r i b u t e s . The hypothesis entertained i n thi s theory states that the t r a v e l a c t i v i t y has various dimensions which corresponds 8 to the attributes of the t r a v e l mode. These dimensions give r i s e to users' d i s u t i l i t y or d i s s a t i s f a c t i o n , which comprises, i n t o t a l , the "generalized p r i c e " of the t r a v e l a c t i v i t y . In choosing a t r a v e l mode for a given t r i p , the user attempts to minimize his d i s s a t i s f a c t i o n by choosing the mode ass o c i -ated with the l e a s t generalized p r i c e . The theory states further that, each i n d i v i d u a l user perceives the d i s u t i l i t y of t r a v e l by any given mode d i f f e r -ently from other users. Hence, for any group of users, the generalized p r i c e of t r a v e l by any given mode v a r i e s . This v a r i a t i o n , however, occurs i n a ce r t a i n pattern which can be approximated by a normal p r o b a b i l i t y d i s t r i b u t i o n function. Hence, i f the generalized price of the t r a v e l mode and i t s a l t e r n a t i v e are known, i t i s possible to assign a c e r t a i n p r o b a b i l i t y to the user's making one choice and the other. The graph below (Fig. 1) demonstrates t h i s concept c l e a r l y . The horizontal axis refers to diffe r e n c e i n the generalized p r i c e between modes A and B, and the v e r t i c a l axis gives the p r o b a b i l i t y of making one or the other s e l e c t i o n . In a mode choice s i t u a t i o n where the diffe r e n c e i n the generalized price i s zero, i . e . , both modes are equally d i s s a t i s f y i n g , the pr o b a b i l i t y of using e i t h e r mode i s 0.5. In another s i t u a t i o n where the generalized p r i c e of using B i s higher,, the p r o b a b i l i t y of using A increases, and that of B 24 Fig. 1. Probability of Mode Choice as a Function of Savings in the Generalized Price 25 declines as shown i n the graph. We may note that i n t h i s mode choice s i t u a t i o n , there i s s t i l l the p o s s i b i l i t y , however low, that the second-best mode would be selected. But, as the d i s -u t i l i t y of thi s mode increases over i t s a l t e r n a t i v e , the pr o b a b i l i t y of making the "wrong" s e l e c t i o n declines c o n s i s t -ently. This "deviation" can be a t t r i b u t e d mainly to the concept of i n d i v i d u a l perceptual v a r i a t i o n of t r a v e l d i s u t i l i t y . That i s , i f a l l users perceived t r a v e l . d i s u t i l i t i e s i n exactly the same manner, every user w i l l be choosing mode A over mode B as long as a saving i n the generalized p r i c e can be achieved by making such a choice. But, since users' i n d i v i d u a l percep-tions vary, for reasons to be discussed i n Section 2.8., t h i s i s not the case i n both theory and observation. 2.5. Evaluation The u t i l i t a r i a n theory of mode choice puts forward two basic concepts which are useful for the purposes of t h i s study. The f i r s t of these i s the concept of generalized p r i c e , which r e f e r s to the perceived d i s u t i l i t i e s associated with the trav e l a c t i v i t y (for a given t r i p and by a given mode). The I i - second concept i s concerned with the pattern by which user perceptions of the. d i s u t i l i t y savings of using one mode v i s -a-vis i t s a l t e r n a t i v e vary among any group of users. The notion of generalized p r i c e i s extremely useful in explaining users' behaviour. Without r e s o r t i n g to t h i s concept, i t i s impossible to s c i e n t i f i c a l l y explain how users compare one mode against the other, since each i s perceived 26 as a bundle of a t t r i b u t e s . I t i s argued here that the only way to do so i s by expressing a l l the t r a v e l dimensions i n terms of some common units of measurements, be i t time, money, or simply " u t i l e s . " The d i s u t i l i t y of these various dimensions comprises the generalized price of the t r a v e l a c t i v i t y , and, hence, a comparison can be drawn between the use of a mode v i s - a - v i s others. The important question that may a r i s e from t h i s d i s -cussion i s whether the user himself perceives the d i s u t i l i t y of t r a v e l dimensions i n t h i s manner, and whether his behaviour supports the notion of the generalized p r i c e . A d e t a i l e d discussion on t h i s r e l a t i n g time and money i s provided i n Section 2.6. Here, s u f f i c e to mention that there i s evidence to suggest that users do make trade-offs between one t r a v e l d i s u t i l i t y and the other, thus supporting the hypothesis that d i f f e r e n t d i s u t i l i t i e s can be compared i n quant i t a t i v e terms. The second concept to be evaluated her'e i s re l a t e d to the notion that savings i n the generalized p r i c e to be made i n using the best mode over it's a l t e r n a t i v e i s not constant f o r any group of users, but varies from one user to the other. This was attr i b u t e d to v a r i a t i o n s i n value systems, socio-economic c h a r a c t e r i s t i c s , and sampling errors common to a l l s t a t i s t i c a l analyses. These> and other possible explan-ations, are discussed i n Section 2.9. Here, our i n t e r e s t i s confined to the implications of this on model formulation. The hypothesis that savings i n the generalized p r i c e are weighted d i f f e r e n t l y by i n d i v i d u a l users i s by i t s e l f a 27 plausible one. Indeed, i t would be unreasonable to hypothesize the opposite, since users' perception of the "advantages" of one mode over the other can hardly be expected to be i d e n t i c a l . This perceptual v a r i a t i o n , as we may c a l l i t here, has an obvious impact on users' behaviour, and we would be wise to account for i t i n the model formulation. Without information on the pattern of i n d i v i d u a l perceptual v a r i a t i o n s , users' behaviour can not be predicted. But the theory states that perceptual v a r i a t i o n of the d i f f e r -ence i n the generalized.price can be approximated by a normal p r o b a b i l i t y d i s t r i b u t i o n function. This i s assumed i n a l l . s t a t i s t i c a l analyses where no evidence i s encountered to suggest that the shape of the curve i s otherwise. This assumption --here i t i s part of the theory—simply implies that i n any unbiased sample, users' perception of the generalized p r i c e savings i s homogeneous. That i s , there are few who are extreme i n evaluating such saving, but the majority tend to "agree" on some common value. Further, the hypothesis on the shape of the curve, a qu a l i t y usually r e f e r r e d to as the homoscedasticity, i s an important part of the mode choice theory as i t enables the researchers to predict the choices to be made by any group of users. By using the central l i m i t theorem, i t i s possible to predict the user choice given his perceived generalized p r i c e saving and the average perceived saving i n the sample. 28 2.6. Empirical Evidence: Value of Time In searching for evidence on the v a l i d i t y of the u t i l i t a r i a n theory of mode choice, our attention should be directed to the "time value" theory, the main concern of which i s the i d e n t i f i c a t i o n of the trade-off values between time and cost as a mode choice determinant. Our i n t e r e s t i n t h i s theory i s j u s t i f i e d by the suggestion that time and cost are two of the most s i g n i f i c a n t t r a v e l d i s u t i l i t i e s , and furthermore, because the voluminous amount of research on t h i s r e l a t i o n s h i p provides us with empirical data on the trade-offs between these d i s u t i l i t i e s . This theory has developed as a r e s u l t of the i n t e r e s t of transportation economists i n evaluating the benefits of proposed highway improvements. In 19 61, the American Association of State Highway O f f i c i a l s expressed the view that ". . . the d o l l a r value of time savings may vary considerably and no precise method of evaluation has yet been determined. A value of time for passenger cars of $1.55 per hour i s used herein as representative of current opinion for a l o g i c a l and 9 p r a c t i c a l value." In response to thi s statement, Moses and Williamson noted that "no explanation i s given for why $1.55 is more l o g i c a l and p r a c t i c a l than any other figure.""*' 0 Likewise, the use of the average hourly wage, as an equivalent to the value of time saved i n t r a v e l , was c r i t i c i z e d on the ground that "time saving contributes to a tangible reduction i n the cost of transportation only to the extent users are able to make productive, i . e . , g a i n f u l , use of time 29 saved.""1""1" Limited research was conducted i n 1963 to derive two separate values for working and non-working times. But these were also estimates based on what seemed p r a c t i c a l and l o g i c a l , 12 with no evidence to support them. In t h i s study, we are not interested i n the monetary value of time per se, and therefore the dispute on the correct value i s i r r e l e v a n t i n thi s context. Rather, our i n t e r e s t i s focused on the following questions: (1) do users perceive t r a v e l time and cost as two d i s u t i l i t i e s to be traded one against the other? and (2) i s the trade-off value constant or variable? and i f the l a t t e r i s correct, i s t h i s value correlated with, or dependent upon, other variables? and what, are these? One way of inv e s t i g a t i n g t h i s problem i s by examining users' behaviour i n mode choice s i t u a t i o n s , t h e i r personal c h a r a c t e r i s t i c s , and the c h a r a c t e r i s t i c s of the choice s i t u a t i o n i t s e l f . This approach was followed i n many studies, whose findings are discussed below. To s t a r t our discussion on the empirical research findings, we make reference to Beesley who made the f i r s t attempt to derive time value from data c o l l a t e d on users' 13 choices between two t r a n s i t modes i n London, England. These data included t r a v e l time by each mode, the fare, income category and occupational status. Beesley assumed that the l e v e l of comfort i s the same i n the two choices, and, hence, explained mode choice as a trade-off between cost of t r a v e l and t o t a l t r a v e l time savings. By the t r i a l and error technique, he attempted to reach the time value which explains users' choices 30 with a minimum of m i s c l a s s i f i c a t i o n s . The value which yielded the best r e s u l t s was 31% of the average wage of c l e r i c a l o f f i c e workers, and 37%. for executive o f f i c e workers. A t h i r d time value was derived for the high income groups, which varied 14 from 42% to 50% of the hourly wage of t h i s group. The model formulated on the basis of these values c l a s s i f i e d 75% of the observations c o r r e c t l y , which was considered s a t i s f a c t o r y . Although Beesley's work was only exploratory, he established a case for deriving the perceived time value from users' behaviour. He also i d e n t i f i e d the r e l a t i o n s h i p between perceived time value on the one hand, income and occupational status on the other. A reference was also made, without an in-depth a n a l y s i s , to the possible r e l a t i o n s h i p between time value and the circumstances of the t r i p : value of time saved was made dependent, among other things, upon the mode of ,15 t r a v e l . Quarmby pursued this approach i n a more d e t a i l e d study 16 on mode choices i n the work t r i p i n Leeds, England. Three choices were available to the user, car, bus and the t r a i n . This study was more advanced i n that "excess time," i . e . , out-of - v e h i c l e t r a v e l time, walking, waiting, was treated as a d i f f e r e n t d i s u t i l i t y separate from the i n - v e h i c l e t r a v e l time. The hypothesis implied i n t h i s treatment was that the d i s u t i l i t y of excess time was perceived d i f f e r e n t l y from the i n - v e h i c l e time. A further advance made by Quarmby was the use of discrim-inant analysis to calculate the d i s u t i l i t y function of each mode, a method which was at least more s c i e n t i f i c than the t r i a l and error technique. The outcome of Quarmby's work supported the notion that users trade time savings against cost, and that the value of the trade-off was dependent upon the user's income. Quarmby found this value to be between 20% to 25% of the user's wage, a percentage which held true for a wide range of income groups. The hypothesis that the d i s u t i l i t y of excess time was d i f f e r e n t from the i n - v e h i c l e time was also supported. The former was 18 valued at 2.3 times the l a t t e r . This f i n d i n g i s extremely s i g n i f i c a n t for planning purposes. A further f i n d i n g of t h i s study was the r e l a t i o n s h i p between time value and the mode being used: time saved on the car t r i p was valued less by the user than time saved on the bus t r i p , the former being 19 40% to 50% of the l a t t e r . This could possibly be a t t r i b u t e d to the comfort l e v e l which added to the d i s u t i l i t y of time spent i n the bus i n comparison to the time spent i n the car. Later i n Section 2.7., the problem of accounting for the com-f o r t l e v e l as a separate d i s u t i l i t y i s examined. The r e s u l t s of Stopher's study i n County H a l l , London, England, were i n basic agreement with the d e f i n i t i o n of time value i n previous work: i t ranged from 23% to 32% of the hourly wage rate, s l i g h t l y higher than the range defined by 20 Quarmby. A much higher percentage, about 50% of the wage rate, was derived by Lisco i n a Chicago study.. Value of excess time was found to be three times the i n - v e h i c l e time 21 i n this study. Unfortunately, the d e t a i l s of Lisco's study was not published, and therefore no explanation can be given here as to the reason for th i s unusually high value. The above quoted authors, Beesley, Quarmby, Stopher and Lisco, investigated the concept of trade-off between time saved and cost of t r a v e l i n a vari e t y of mode choice s i t u a t i o n s . However, i n a l l these studies, except that by Li s c o and possibly Quarmby, the v a r i a t i o n i n the l e v e l of comfort from one mode to the other was not accounted f o r . Hence, i t may be argued that the trade-off was not merely of money against time (assuming car use y i e l d s time savings), but probably against time plus increased l e v e l of comfort. The value of time derived i n t h i s manner would then be i n f l a t e d . 1 The solu t i o n to thi s problem could be found i n a di f f e r e n t phase of mode choice analysis. Research on the benefits of highway improvements attempted to define how the users perceive the time saved (through such improvements) and the monetary value they attach to thi s saving. Such inform-ation was provided i n the users 1 choices between two highway routes, one of which i s a t o l l road, but a shorter route. In such s i t u a t i o n s , users' choices involved a trade-off between time and cost, yet the l e v e l of comfort i n both cases was nearly the same. No other s i t u a t i o n could provide us with more accurate information on the v a l i d i t y of the time value concept. Thompson and Thomas made use of a t h e o r e t i c a l framework previously advanced by Haney to conduct an extensive analysis 22 of time value. In this framework, i t was hypothesized that the marginal value of time was not a constant, rather a function of the time saved. Haney hypothesized that the time v a l u e was very s m a l l f o r s m a l l amounts o f t o t a l time s a v i n g s , but i t i n c r e a s e d , as shown i n F i g . 2.a., w i t h the i n c r e a s e i n t o t a l time s a v i n g s . The i n c r e a s e d i d not c o n t i n u e ad i n f i n i t u m , however, and beyond c e r t a i n v a l u e , more time s a v i n g s brought l e s s e r i n c r e a s e i n time v a l u e , o b v i o u s l y i n conformance w i t h the law o f d i m i n i s h i n g m a r g i n a l r e t u r n s . T h i s c o n c e p t u a l framework was brought f o r t h by Haney demonstrating what he thought t o be a p l a u s i b l e h y p o t h e s i s on u s e r s ' e v a l u a t i o n o f time s a v i n g s , but he d i d not i n v e s t i g a t e the p o s s i b l e e f f e c t o f the t o t a l t r i p l e n g t h on time v a l u e . However, a c c o r d i n g to Haney's scheme, i t would be improper to a s s i g n a s i n g l e v a l u e to time. Thomas a l s o reached t h i s c o n c l u s i o n as he t r i e d and f a i l e d to d e r i v e a c o n s t a n t 23 time v a l u e . The data he c o l l a t e d f o r t h i s purpose was on highway t o l l s and the time savings o f f e r e d i n r e t u r n . The sample encountered d i f f e r e n t mode s i t u a t i o n s w i t h v a r i o u s t o l l s and time s a v i n g s . Thomas and Thompson made v a r i o u s attempts a t d e f i n i n g , the time v a l u e , and f i n a l l y they produced a model r e l a t i n g t h i s v a l u e t o income and t o t a l time 24 . s a v i n g s . In a subsequent work such a model was c a l i b r a t e d f o r each t r i p purpose: work, p e r s o n a l b u s i n e s s , s o c i a l r e c r e a t i o n a l and v a c a t i o n . Using a s e t of these models, the authors produced s e v e r a l t a b l e s , each g i v i n g the time v a l u e f o r e i g h t income groups (the columns) and t o t a l time sa v i n g s 25 by increment o f one minute (the rows) Up to 20 or 30 minutes. The aim o f these t a b l e s was to f a c i l i t a t e the task o f the highway economist i n e v a l u a t i n g the b e n e f i t s o f highway improvements. 5 10 15 20 Time Savings i n Minutes Fig. 2.a. Value of Time as a Function of Total Time Savings (Haney's Scheme) - 4 - 1 1 5 10 15 Time Saved (Minutes) Fig. 2.b. Marginal Benefits of Time Saved by Trip Purpose (Income Group: $8000 - $9999) To make use o f t h e s e t a b l e s , i n f o r m a t i o n i s needed on the e x p e c t e d number o f u s e r s , t h e i r income c a t e g o r y , t r i p p u r p o s e , and the t ime s a v i n g o f f e r e d by the improvement . The e x p e c t e d monetary r e t u r n s can t h e n be c a l c u l a t e d f rom the t a b l e s . The models c a l i b r a t e d by Thomas and Thompson a f f o r d s us the o p p o r t u n i t y to t e s t the v a l i d i t y o f the t h e o r e t i c a l scheme o f t ime v a l u e advanced by Haney. The graph c o n s t r u c t e d by Reichman, shown i n F i g . 2 . b . , i s h e l p f u l f o r t h i s p u r p o s e . ^ ' Reichman used Thomas and Thompson's t a b l e s to d e m o n s t r a t e the r e l a t i o n s h i p o f m a r g i n a l v a l u e o f t ime t o the t o t a l t i m e s a v i n g s f o r f o u r k i n d s o f t r i p s . E x c e p t f o r v a c a t i o n t r i p s , changes i n the m a r g i n a l v a l u e o f t ime r e f l e c t s a c l e a r p a t t e r n which i s c h a r a c t e r i z e d by t h e f o l l o w i n g z o n e s : (a) f rom one t o f i v e minutes s a v i n g s , the m a r g i n a l v a l u e o f t ime shows l i t t l e or no i n c r e a s e . (b) f rom f i v e t o f i f t e e n m i n u t e s s a v i n g s , i n c r e a s e i n t ime s a v i n g s i s p a r a l l e l e d by the g r e a t e s t i n c r e a s e i n m a r g i n a l v a l u e o f t i m e . (c) beyond f i f t e e n minutes s a v i n g s , w h i c h i s t h e zone o f d i m i n i s h i n g m a r g i n a l r e t u r n s , the m a r g i n a l v a l u e o f t ime i s e i t h e r m a i n t a i n e d at the same v a l u e , o r d e c l i n i n g . These changes i n the c h a r a c t e r i s t i c s o f t h e c u r v e , as i t pro.ceeds from one zone to the o t h e r , a re a l m o s t i d e n t i c a l w i t h those o f the c u r v e c o n s t r u c t e d i n H a n e y ' s scheme a f t e r a c c o u n t i n g f o r the f a c t t h a t the l a t t e r d e p i c t s t h e m a r g i n a l v a l u e o f t ime s a v i n g s , w h i l e the former d e p i c t s t h e a v e r a g e v a l u e o f t i m e . T h i s a n a l y s i s g i v e s us e m p i r i c a l e v i d e n c e t o 36 support Daney's theory of time value, which i s indeed f a r more sophisticated and possibly more pla u s i b l e than the average, single time value approach which was pursued by Quarmby and others. To sum up t h i s section, evidence brought forward i n the l i t e r a t u r e suggest that at le a s t two t r a v e l d i s u t i l i t i e s , cost and time, were perceived i n q u a n t i t a t i v e l y comparable terms by the users, and that t h e i r behaviour c l e a r l y indicated that certain trade-offs were made i n t h e i r mode choice. The trade-off value i d e n t i f i e d was not constant, since the r e l a t i o n -ship between time and cost was found to be non-linear. However, early researchers i d e n t i f i e d the consistency i n users' behaviour i n making these t r a d e - o f f s , and found that the value.of time was dependent mainly upon users' income, and to a les s e r extent upon the circumstances of the t r i p . I t was recognized that the value of time was not uniform since a higher d i s u t i l i t y was assigned to excess time than the in - v e h i c l e time. Hence, i n formulating a mode choice model, these two should be entered separately. This aspect of users' behaviour, the trade-off between time and cost, was expressed more c l e a r l y i n those studies concerned with the choice of t o l l routes. In such s i t u a t i o n s , time savings were achieved at cost, while a l l other variables remained unchanged. Further support for the theory of time value was brought f o r t h , and more importantly, two other dependencies of time value were i d e n t i f i e d : t o t a l time saving and t r i p purpose. These findings were derived from an empirical data a n a l y s i s , and c o n s i d e r i n g the p l a u s i b i l i t y o f the c o n c e p t u a l framework, i t i s the view o f t h i s author t h a t the l i t e r a t u r e bears support f o r at l e a s t t h i s a s p e c t o f the u t i l i t a r i a n theory o f mode c h o i c e . 2.7. The D i s u t i l i t y o f Comfort L e v e l Together w i t h t r a v e l time and c o s t a t t r i b u t e s , the d i s u t i l i t y o f the comfort l e v e l are c o n s i d e r e d to be the most 27 s i g n i f i c a n t mode c h o i c e determinants. However, r e s e a r c h f a i l e d t o d e f i n e the c o n t r i b u t i o n o f the l a t t e r t o the g e n e r a l i z e d p r i c e o f the t r a v e l a c t i v i t y , and hence i t s r e l a t i o n s h i p to time and c o s t has not been expressed i n q u a n t i t a t i v e terms. The d i f f i c u l t y r e s e a r c h e r s have encountered i n ac c o u n t i n g f o r the comfort l e v e l can be a t t r i b u t e d t o two f a c t o r s : f i r s t , the ambiguity o f the term and hence the v a r i e t y o f i n t e r p r e t a t i o n i t may e n t a i l , and s e c o n d l y , the problem o f p r o v i d i n g an o b j e c t i v e , a c c u r a t e measurement f o r comfort. The ambiguity o f the comfort term can be demonstrated as one attempts to d e f i n e the d i f f e r e n c e i n the comfort l e v e l between, f o r example, the p r i v a t e c a r and the bus. The l e v e l o f comfort o f each i s determined by a g r e a t number o f mechanical c h a r a c t e r i s t i c s , the i n t e r i o r and p o s s i b l y the e x t e r i o r o f the v e h i c l e . Few. o f - t h e s e l e n d themselves t o measurement, e.g., l e v e l o f n o i s e and v e h i c l e v i b r a t i o n s . But the l a t t e r are r e l a t i v e l y i n s i g n i f i c a n t r e l a t i v e to the p s y c h o l o g i c a l , more 38 i n f l u e n t i a l f a c t o r s : e.g., the discomfort associated with the vehicle crowding. Furthermore, we should also be concerned with the user's comfort during the entire t r i p , from o r i g i n to des t i n a t i o n . This would include, i n addition to those items mentioned above, the discomfort i n walking, waiting for the v e h i c l e , and exposure to inclement weather conditions. The discomfort associated with these a c t i v i t i e s , which i s l i k e l y to vary seasonally, i s d i f f i c u l t to quantify i n any meaningful manner, which compounded the researcher's problem. This d i f f i c u l t y has lead to a considerable confusion i n accounting f o r the comfort variable i n previous works. Many researchers have accounted for each q u a n t i f i a b l e v ariable separately, e.g., time, cost, but accounted c o l l e c t i v e l y for a l l other q u a n t i f i a b l e attributes under the comfort l e v e l "umbrella." Hence, Lisco's c a l c u l a t i o n , for example, for the value of the working t r i p comfort at $2.00 i s l i k e l y to be i n v a l i d . I t i s the contention of t h i s author that such a monetary value i s equivalent not only to the comfort of car use, but other advantages i n addition, such as the f l e x i b i l i t y of routing and departure t i m e . ^ Other researchers made a p a r t i a l account for the d i s -u t i l i t y of the comfort l e v e l through i t s association with excess 29 t r a v e l time (walking, waiting and transfer time.) The r a t i o n a l e of t h i s approach was that the comfort l e v e l of the t r i p declined with the increase of t h i s time, since i t involved physical e f f o r t and possibly exposure to inclement weather. ' 39 Ob v i o u s l y , t h i s approach p r o v i d e d an incomplete account f o r the t r i p comfort l e v e l i n t h a t the i n - v e h i c l e comfort was n o t i n c l u d e d . R e s e a r c h e r s ' f a i l u r e t o account f o r the comfort a t t r i b u t e s has been d e t r i m e n t a l t o pr o g r e s s i n mode c h o i c e a n a l y s i s . A working d e f i n i t i o n f o r the comfort l e v e l and a r e l i a b l e method o f measurement have not been found, and hence, t h i s a t t r i b u t e can not be i s o l a t e d from o t h e r u n q u a n t i f i a b l e system a t t r i b u t e s . Up t o t h i s s tage o f r e s e a r c h , the c o n t r i -b u t i o n o f t h i s a t t r i b u t e t o the g e n e r a l i z e d p r i c e o f t r a v e l i s l a r g e l y u n d e f i n e d . T h i s problem i s i n v e s t i g a t e d f u r t h e r i n S e c t i o n s 3.1.1., and 3.16. PERCEPTION, ATTITUDES AND BEHAVIOUR T h i s study i d e n t i f i e s the user ' s i n d i v i d u a l p e r c e p t i o n and a t t i t u d e s as mode c h o i c e determinants. P r e v i o u s d i s -. c u s s i o n s demonstrate the e f f e c t o f p e r c e p t u a l v a r i a t i o n s on u s e r s ' b e h a v i o u r , y e t without e x p l a i n i n g why i n d i v i d u a l s ' p e r c e p t i o n vary i n the f i r s t i n s t a n c e . A l s o , d e f i n i t i o n s o f p e r c e p t i o n and a t t i t u d e s were not attempted. These q u e s t i o n s are t o be addressed i n the f o l l o w i n g sections,, which would h e l p us c l a r i f y many aspects o f u s e r s ' behaviour u n e x p l a i n e d so f a r . 2.8. P e r c e p t i o n and A t t i t u d e s In i n v e s t i g a t i n g the p s y c h o l o g i c a l u n d e r p i n n i n g s o f u s e r s ' behaviour, an e x p l a n a t i o n o f some b a s i c concepts becomes i n e v i t a b l e . P s y c h o l o g i s t s , concerned w i t h the f o r m u l a t i o n o f a theory o f human behaviour, d e f i n e d p e r c e p t i o n as the way by which the i n d i v i d u a l e x p e r i e n c e d and i n t e r p r e t e d the w o r l d . 3 0 Thus, r e a l world i s unknown to the i n d i v i d u a l — e x c e p t through p e r c e p t i o n . Yet, i t i s not merely a p r o c e s s o f r e c o g n i t i o n , i t i n v o l v e s f i l t e r i n g , m o d i f i c a t i o n and d i s t o r t i o n o f r e a l i t y : No one experiences the world e x a c t l y as i t i s , and no two persons e x p e r i e n c e i t i n p r e c i s e l y the same way, because knowledge o f t h e w o r l d — o r our experiences w i t h i t — i s f i l t e r e d and m o d i f i e d by p h y s i c a l and p s y c h o l o g i c a l f a c t o r s w i t h i n us.31 T h i s c o n c l u s i o n , t h a t i n d i v i d u a l s vary i n t h e i r p e r c e p t i o n o f events and o b j e c t s , i s u s e f u l i n u n d e r s t a n d i n g some aspects o f the u t i l i t a r i a n theory o f mode c h o i c e . We have more to say about t h i s i n the f o l l o w i n g s e c t i o n . Another b e h a v i o u r a l concept t o be advanced by p s y c h o l -o g i s t s was the i d e n t i f i c a t i o n o f the i n d i v i d u a l ' s p e r c e p t i o n through h i s behaviour. Since p e r c e p t i o n was not d i r e c t l y o b s e r v a b l e , how, then, c o u l d t h i s phenomenon be searc h e d o b j e c t i v e l y ? The answer suggested was t o c o n s i d e r as o b j e c t i v e data: . . . i n f e r r e d events t h a t stand i n a l o g i c a l l y c o n s i s t e n t r e l a t i o n s h i p w i t h the p u b l i c l y o b s e r v a b l e . T h i s means t h a t the o b j e c t i v i t y o f a concept i s determined by the e f f i c i e n c y w i t h which i t y i e l d s r e l i a b l e r e p l i c a t i o n s o f obs e r v a t i o n s . 3 2 This approach was w i d e l y employed by t r a n s p o r t a t i o n r e s e a r c h e r s i n s t u d y i n g the user's p e r c e p t i o n o f the d i s u t i l i t i e s o f time and c o s t o f t r a v e l . To t h i s extent, p s y c h o l o g i s t s ' e f f o r t s have been p a r a l l e l e d by s i m i l a r advances i n r e s e a r c h on mode c h o i c e 41 behaviour. Transportation researchers, however, did not pursue thi s or any other approach to define the modal a t t r i b u t e s perceived as relevant by the user. Although researchers seemed to agree on the relevance of time, cost and comfort, i n t u i t i o n suggests that there are other a t t r i b u t e s whose s i g n i f i c a n c e are s t i l l undetermined. We may now turn our attention to the i n d i v i d u a l ' s attitude, which was also i d e n t i f i e d as a s i g n i f i c a n t factor a f f e c t i n g behaviour. I t was described as predetermined position or bias for or against c e r t a i n objects, which was 33 motivated by an enduring set of perceptual processes. Thus, perception and the formation of attitudes are two mutually e f f e c t i v e phenomena, which leads us to the inference that modal bias, i f i t e x i s t s , i s expressed i n the user's perception of modal a t t r i b u t e s . This inference also suggests that i f the model accounts for the user's perception, t h i s can.be consid-ered as an i n d i r e c t account for his personal biases. 2.9. E f f e c t on Users' Behaviour B a s i c a l l y , there are three ways by which users' choices are affected by the psychological processes mentioned above. The f i r s t of these i s the users' perception of the various attributes of the t r a v e l modes, which i s expressed by the parameters to be derived i n the c a l i b r a t i o n of the d i s u t i l i t y function i n mode choice models employing the discriminant analysis technique. I t i s i n t h i s manner that Beesley, Quarmby, Stopher, Lisco, Thomas and Thompson derived the perceived value of time, and i t i s i n t h i s manner that some of these authors i d e n t i f i e d the d i s u t i l i t y of excess t r a v e l time r e l a t i v e to the i n - v e h i c l e time. As reported previously i n Section 2.6., users' perception was found to be consistent, when various samples are compared, i n regarding the d i s u t i l i t y of these a t t r i b u t e s . Another aspect of users' behaviour which i s of a psychological nature i s modal bias. Attitudes can create a bias for or against any of the modes availa b l e to the user. As a r e s u l t , the user sees the att r i b u t e s of the preferred mode better than what they actually are, but does the opposite with the a l t e r n a t i v e mode (sees i t s att r i b u t e s worse than they are). Ghis tendency i s understandable since the i n d i v i d u a l i s usually anxious to j u s t i f y his choice even through d i s t o r t i o n 34 of f a c t s . Thomas reported such a bias i n a s i m i l a r s i t u a t i o n . In this study of choice.of t o l l routes discussed i n Section 2.6., he noticed that users reported shorter t r a v e l time by the chosen route than the actual t r a v e l time as measured by the researcher. The users also reported longer t r a v e l time f o r the a l t e r n a t i v e mode than the actual time. In other words, those who preferred the t o l l routes exaggerated the time savings, while other users who preferred the free route under-estimated the possible time savings. Obviously, then, users' bias a f f e c t t h e i r perception of t r a v e l times i n such a way that j u s t i f i e d the choices made. Apart from Thomas' observation, the hypothesized r e l a t i o n s h i p between the users' attitude and perception can be useful i n explaining the notion of i n d i v i d u a l perceptual 43 v a r i a t i o n s discussed i n Section 2.4. Attitudes of any group of users toward a given transportation mode can hardly be expected to be uniform, and as they vary, the perception of t r a v e l d i s u t i l i t i e s vary accordingly. T h e o r e t i c a l l y , t h i s v a r i a t i o n i s d i s t r i b u t e d as a normal p r o b a b i l i t y function. That i s to say, i n much s i m p l i f i e d words, for each user who exaggerates the d i s u t i l i t y saving, there i s another who equally under-estimates i t s value. Further, there are few whose perception of d i s u t i l i t y saving i s very d i f f e r e n t from the correct value, while the majority of users are more or less correct i n t h e i r perception. This concept was validated i n the work of a considerable number of researchers. The cumulative frequency diagram of the p r o b a b i l i t y function produced by Warner, Stopher, Pratt, Shunk, and De Donnea, c l e a r l y indicated a strong s i m i l a r i t y between the t h e o r e t i c a l p r o b a b i l i t y d i s t r i b u t i o n function 35 and the observed choices of the users. F i n a l l y , some researchers pointed out that, i n addition to the perceptual v a r i a t i o n caused by the user's personal bias, there was the p o s s i b i l i t y that modal bias was uniform among the users. That i s , the user's personal modal bias was furthered by a bias common to a l l users. I f t h i s were true, the e f f e c t of t h i s uniform bias would be to s h i f t the p r o b a b i l i t y function downward or upward, thus increasing the p r o b a b i l i t y of s e l e c t i n g the preferred mode as shown i n the graph below. Accordingly, even when the generalized prices of t r a v e l by the two a l t e r n -atives were equal, the p r o b a b i l i t y of choosing the preferred 4 4 1.0 + Probability of Choosing Mode A - given ^ user' s bias for A Probability of Choosing Mode A (no bias) foability of Choosing Mode B (no bias) Probability of Choosing Mode B - given user's bias for A -2 0.0 +2 Mode A is better Mode B i s better Fig. 3. Effect of "Uniform" Bias Among Users for Mode A 45 mode would be higher than 0.5. De Donnea and Stopher have found evidence to suggest the presence of car bias i n the users' behaviour examined i n t h e i r sample. However, we should ^ view t h e i r findings with precaution since both authors d i d not account for the unquantifiable at t r i b u t e s of the system. Hence, what they considered a bias for car could merely r e f e r to the influence of the comfort, or f l e x i b i l i t y associated with, the car use. 2.10. Summary and Conclusions This chapter i s devoted to the formulation of a theoret-i c a l framework for users' mode choice, and discussion of evidence brought forward by researchers on the v a l i d i t y of th i s framework. The broad concepts outlined here are derived from two theories: (a) the abstract mode choice theory, and (b) the u t i l i t a r i a n theory of mode choice. A b r i e f account for two psychological processes, perception and att i t u d e s , and t h e i r impact on users' behaviour are also attempted. i n the abstract mode choice theory, emphasis i s placed upon the attributes of the t r a v e l mode rather than i t s - i n s t i t u t i o n a l form. Users' behaviour could be better explained when each mode i s taken as a "bundle" of a t t r i b u t e s , or a combination of c h a r a c t e r i s t i c s . The analysis conducted i n this chapter c l e a r l y indicates that a comprehensive account for mode attributes would.increase the explanatory and p r e d i c t i v e power of the model. More s i g n i f i c a n t l y , i t i s only i n t h i s way i t would be possible to i d e n t i f y the transportation p o l i c y v a r i a b l e s and the i n f l u e n c e o f each on u s e r s ' b e h a v i o u r . The concept o f the " g e n e r a l i z e d p r i c e " to be i n c u r r e d i n any t r a v e l a c t i v i t y i s i n t r o d u c e d as a t o o l to e x p l a i n u s e r s ' behaviour. I t r e f e r s t o the p e r c e i v e d d i s u t i l i t i e s o f t r a v e l by a l t e r n a t i v e modes f o r any g i v e n t r i p . I t s v a l u e i s thus a f f e c t e d by both the mode a t t r i b u t e s and u s e r s ' p e r c e p t i o Furthermore, the user's p e r s o n a l b i a s f o r or a g a i n s t any t r a v e l mode i s expressed i n h i s p e r c e p t i o n o f the d i s u t i l i t y d i f f e r e n c e between the two b e s t modes. I t i s h y p o t h e s i z e d t h a t i n any unbiased sample o f u s e r s , the p e r c e i v e d d i s u t i l i t y s a v i n g s v a r i e s from one user to another, and t h a t t h i s v a r i -a t i o n can be approximated by a normal p r o b a b i l i t y d i s t r i b u t i o n f u n c t i o n . The evidence brought f o r t h i n the l i t e r a t u r e on the v a l i d i t y o f these concepts were mainly concerned w i t h the f o l l o w i n g aspects of u s e r s ' b e h a v i o u r : - u s e r s ' p e r c e p t i o n o f t r a v e l time and c o s t a t t r i b u t e s - the v a r i a b l e s a f f e c t i n g u s e r s ' p e r c e p t i o n o f these a t t r i b u t e s , and - u s e r s ' p e r c e p t i o n of the d i s u t i l i t y s a v i n g s o f f e r e d by the b e s t mode. A d e t a i l e d d i s c u s s i o n o f the time value, theory and . t h e e m p i r i c a l f i n d i n g s o f o t h e r r e s e a r c h e r s on i t s v a l i d i t y produced s u f f i c i e n t evidence to suggest t h a t u s e r s p e r c e i v e the d i s u t i l i t y o f time and c o s t i n q u a n t i t a t i v e l y comparable terms. The parameters d e r i v e d i n s e v e r a l s t a t i s t i c a l a n a l y s e s i n d i c a t e d t h a t u s e r s ' p e r c e p t i o n o f these d i s u t i l i t i e s were 47 c o n s i s t e n t l y expressed i n t h e i r mode c h o i c e b e h a v i o u r . The a n a l y s i s f u r t h e r i n d i c a t e d t h a t the p e r c e p t i o n o f the d i s -u t i l i t y o f time and c o s t was dependent upon the u s e r ' s income, t r i p purpose, l e v e l o f comfort, and t o t a l time s a v i n g s . The i n f l u e n c e o f the d i s u t i l i t y o f the comfort l e v e l on u s e r s ' b e h a v i o u r was i d e n t i f i e d by s e v e r a l a u t h o r s . Yet, due t o the d i f f i c u l t y o f measuring the l e v e l of comfort a s s o c i a t e d w i t h each mode, and i t s c o n f u s i o n w i t h o t h e r non-q u a n t i f i a b l e v a r i a b l e s , the exact c o n t r i b u t i o n o f t h i s a t t r i b u t e t o the d i s u t i l i t y f u n c t i o n remained l a r g e l y un-d e f i n e d . Only a p a r t i a l account was made f o r d i s c o m f o r t through i t s a s s o c i a t i o n w i t h excess t r a v e l time. S e v e r a l w r i t e r s found support to the h y p o t h e s i s t h a t the d i s u t i l i t y s avings gained by u s i n g the b e s t mode over i t s a l t e r n a t i v e was p e r c e i v e d d i f f e r e n t l y by v a r i o u s u s e r s . The p a t t e r n o f t h i s p e r c e p t u a l v a r i a t i o n r e f l e c t e d a s i m i l a r i t y w i t h the normal p r o b a b i l i t y d i s t r i b u t i o n f u n c t i o n as maint a i n e d i n the u t i l i t a r i a n theory o f mode c h o i c e . The c o n c e p t u a l a n a l y s i s and e m p i r i c a l e v i d e n c e i n t r o -duced i n t h i s chapter render the u t i l i t a r i a n t heory a workable h y p o t h e s i s , and v a l i d a t e many o f the concepts i n t r o d u c e d on u s e r s ' b e h a v i o u r . To i n c r e a s e the u s e f u l n e s s o f t h i s theory to mode c h o i c e a n a l y s i s and p r e d i c t i o n , a f u r t h e r advance i s needed on the method of measurement and a c c o u n t i n g f o r the n o n - q u a n t i f i a b l e v a r i a b l e s such as comfort, f l e x i b i l i t y of t r a v e l time, e t c . These a t t r i b u t e s p r e s e n t the r e s e a r c h e r w i t h some "thorny" problems (some o f which w i l l be t a c k l e d i n the f o l l o w i n g 48 c h a p t e r ) . In a d d i t i o n , f u r t h e r r e s e a r c h i s a l s o r e q u i r e d to d e f i n e the u s e r ' s p e r c e p t i o n of modal a t t r i b u t e s , thus p e r m i t t i n g a f u l l d e s c r i p t i o n of the mode c h o i c e s i t u a t i o n . FOOTNOTES A summary a n a l y s i s of v a r i o u s modal s p l i t a n a l y t i c a l techniques f o r t h i s time p e r i o d i s p r o v i d e d i n Modal S p l i t : Documentation of Nine Methods by M a r t i n J . i F e r t a l , e t a l . Thi s approach i s documented by s e v e r a l s o u r c e s , K e l v i n J . L a n c a s t e r , "A New Approach t o Consumer Theory," J o u r n a l o f P o l i t i c a l S c i e n c e s , V o l . 74, 1966, pp. 133-157, R i c h a r d W. Quandt and W i l l i a m Baumol, "The Demand f o r A b s t r a c t T r a n s p o r t Modes: Theory and Measurement," J o u r n a l of R e g i o n a l S c i e n c e , V o l . 6, No. 2, 1966, pp. 13-25, R i c h a r d E. Quandt, " E s t i m a t i o n of Modal S p l i t s , " T r a n s p o r t a t i o n Research V o l . 2, March 1968, Kan Han Young, "An A b s t r a c t Mode Approach to the Demand f o r T r a v e l , " T r a n s p o r t a t i o n  Research, V o l . 3, No. 4, 1969, Thomas F. Golob, "The Survey of User Choice o f A l t e r n a t e T r a s n p o r t a t i o n Modes," High Speed Ground T r a n s p o r t a t i o n J o u r n a l , V o l . 4., No., pp. 103-116, and A l e x i s N. Sommer, "Toward a Theory of T r a v e l e r Mode C h o i c e , " High Spe£d  T r a n s p o r t a t i o n J o u r n a l , V o l . 4, Jan. 1970, pp. 1-8. A c o r o l l a r y t o t h i s h y p o t h e s i s i s t h a t , by a c c o u n t i n g f o r a l l the r e l e v a n t modal a t t r i b u t e s , the use r i s "mode-neutral", having no b i a s f o r or a g a i n s t any p a r t i c u l a r mode. Thomas L i s c o , who accounted f o r the c o s t , time and comfort a t t r i b u t e s r e p o r t e d , " t h e r e was no evidence of an i r r a t i o n a l commuter 'love a f f a i r ' w i t h the automobile," i n "The Value of Commuter * s T r a v e l Time - A study i n Urban Tra n s p o r -t a t i o n , " u n p u b l i s h e d Ph.D. d i s s e r t a t i o n , U n i v e r s i t y of Chicago, 1967, abr i d g e d i n Highway Research Record Rep. No. 245, 1968, p. 36. See, f o r example, the s e c t i o n on t r i p - e n d modal s p l i t E r i e , P e n n s y l v a n i a , i n Modal S p l i t : Documentation of  Nine Methods F e r t a l e t a l . , p. 27. A study which p l a c e d emphasis on the p o l i c y v a r i a b l e s was prepared by Brown, Mode Cho i c e Determinants of  S e l e c t e d Socio-economic Groups. 50 Studies which investigated the manipulation of modal . at t r i b u t e s as a planning t o o l to a f f e c t users'choices are that of Leon Moses and H. Williamson, "The Subsidy Issue", Journal of P o l i t i c a l Economy, "Choice of Travel Mode for the Journey to Work," Journal of  Transport Economics and Po l i c y , Sept., 1967, pp. 273-314, Charles A. Lave, "A Behavioural Approach to Modal S p l i t Forecasting," Transportation Research, V o l . 3, No. 4, 1969, pp. 463-480, Robert -Gf. M c G i l l i v r a y "Demand and Choice Models of Modal S p l i t , " Journal of  Transport Economics and Po l i c y , May 1970, pp. 192-206, G. Brown, Mode Choice Determinants, 1971, David T. Hartgen, George H. Tanner, "Investigations of the E f f e c t of Traveler Attitudes i n a Model of Mode Choice Behaviour," Highway Research Record, Rep. No. 369, 1971 pp. 1-14, and Frederick J . Beier, "Marketing Programs For Mass Transit," T r a f f i c Quarterly, Oct. 1972, pp. 533-545. Such conclusions have been.reached i n the studies conducted by De Donnea, The Determinants of Transport Mode Choice, p. 206, and A. J. Sobey and J . W. Cone, "The Case f o r Personal Rapid T r a n s i t , " Highway  Research Record, Rep. No. 367, 1971, pp. 70-89. The u t i l i t a r i a n theory of mode choice i s presented i n a number of publications: Peter R. Stopher, "A P r o b a b i l i t y Model of Travel Mode Choice f o r the Work Journey," Highway Research Record, Rep. No. 283, 1969, pp. 57-65, Richard A. Pr a t t , "A U t i l i t a r i a n Theory of Travel Mode Choice," Highway Research Record, Rep. No. 322, 1970, pp. 40-53, Sommer, "Mode Choice,"High Speed Transportation Journal, V o l . 4, No. 1, 1970, pp. 1-8, Gordon A. Shunk and Richard J . Bouchard, "An Ap p l i c a t i o n of Marginal U t i l i t y to Travel Mode Choice," Highway Research Record, Rep. No. 322, 1970, pp. 30-41, and Peter R. Stopher and Shalom Reichman, "Disaggregate Stochastic Models," Highway Research Record, Rep. No. 369, 1971, pp. 91-103. Moses and Williamson, "The Subsidy Issue," Journal  of P o l i t i c a l Economy, p. 248. Ibid. United States Secretary of Commerce, F i n a l Report of  the Highway A l l o c a t i o n Study, 1961, p. 205, i n i b i d . M. E. Beesley, "The Value of Time Spent i n T r a v e l l i n g : Some New Evidence," Economica, V o l . 32, 1965, p. 17 4. 51 13. Ibid., pp. 174-185. 14. Ibid., p. 182. 15. Ibid., pp. 182-183. 16. Quarmby, "Choice of Travel Mode," Journal of  Transport Economics and P o l i c y , pp. 273-314. 17. Ibid., pp. 295^297. 18. Ibid., p. 292. 19. Ibid., p. 289: 20. Stopher, " P r o b a b i l i t y Model," Highway Research Record, p. 61. 21. Lisco, "Value of Travel Time," Highway Research Record, p. 36. 22. Dan G. Haney, The Value of Time f o r Passenger Cars: A Theoretical Analysis and Description of Preliminary Experiment, Stanford Research I n s t i t u t e , May 1967, i n Thomas C. Thomas, "Value of Time for Community Motorists," Highway Research Record, Rep. No. 245, 1968, p. 27. 23. Ibid., pp. 17-34. 24. Thomas C. Thomas and Gordon I. Thompson, "The Value of Time f o r Commuting Motorists as a Function of Their Income Level and Amount of Time Saved," Highway  Research Record, Rep. No. 314, 1970, pp. 1-19. 25. Thomas C. Thomas and Gordon I. Thompson, "Value of Time Saved by T r i p Purposes," Highway Research Record, Rep. No. 369, 1971, pp. 104-117. 26. The discussion section by Shalom Reichman i n i b i d , p. 115. 27. Lisco, "Value of Time," Highway Research Record, p. 36. In his study of Chicago commuters behaviour, Lisco i s o l a t e d a l l the observations where the users' t r a v e l time by t r a n s i t was equal that by car. From the analysis of t h i s set of observations, he concluded that the "average" commuter i n h i s sample was paying an a d d i t i o n a l amount of $2.00 (in car operating costs, parking charges, etc.) per day to use his car instead of t r a n s i t . Lave, "A Behavioural Approach, Transportation Research, p. 464. Quarmby, "Choice of Travel Mode," Journal of Transport  Economics and Po l i c y , p. 281, L i s c o , "The Value of Time," Highway Research Record, p. 36, Pratt , " U t i l i t a r i a n Theory," Highway Research Record, p. 48, and Shunk and Bouchard, "Application of Marginal U t i l i t y , " Highway Research Record, p. 32. S. H. Bartley, P r i n c i p l e s of Perception, Harper, New York, 1958, p. 4. Ibid.;, p. 22. William Bevan, "Perception: Evolution of a Concept," Psychological Review, V o l . 65, No. 1, 1958, p. 34. D. Krech and R. S. C r u t c h f i e l d , Theory and Problems  in Experimental Psychology, McGraw-Hill, New York, 1948, p. 152. Thomas, "Value of Time," Highway Research Record, 1968, pp. 22-23. S. L. Warner, Stochastic Choice of Mode i n Urban Travel; A Study i n Binary Choice, Northwestern University Press, 1962, pp. 40-41, Stopher, "Probabilit Model," Highway Research Record, p. 62, Pr a t t , " U t i l i t a r i a n Theory," Highway Research Record, pp. 44-51, Shunk, "Application of Marginal U t i l i t y , " Highway Research Record, p. 36, and De Donnea, Determinants  of Transport Mode Choice, p. 147. CHAPTER I I I MODEL STRUCTURE AND APPLICATION 3 . 1 . I n t r o d u c t i o n While the main concern i n the p r e v i o u s c h a p t e r i s w i t h the l i t e r a t u r e review, and d e r i v a t i o n of a c o n c e p t u a l frame-work upon which the study's model may be based, the p r e s e n t chapter i s concerned w i t h the model s t r u c t u r e and d e f i n i t i o n of i t s o p e r a t i o n a l c h a r a c t e r i s t i c s . This e n t a i l s f o u r s e q u e n t i a l s t e p s : (a) Development of c r i t e r i a f o r i d e n t i f i c a t i o n o f mode c h o i c e determinants. (b) S e l e c t i o n o f a p p r o p r i a t e v a r i a b l e s which r e p r e -s e n t these determinants i n the model. (c) M a n i p u l a t i o n o f these v a r i a b l e s as model i n p u t s . (d) D e f i n i n g the model s t r u c t u r e , and s e l e c t i o n o f the s t a t i s t i c a l techniques t o be used. Another problem to be addressed i n t h i s c h apter i s the assessment of the ex p l a n a t o r y and p r e d i c t i v e v a l u e of the model. A j u s t i f i c a t i o n w i l l be g i v e n as t o why the study's approach would advance our knowledge of u s e r s ' behaviour and improve the r e s e a r c h e r ' s a b i l i t y t o p r e d i c t the users.' response to changes i n the t r a n s p o r t a t i o n system, and to trends i n the s o c i a l , economic or demographic c h a r a c t e r i s t i c s o f the urban p o p u l a t i o n . 53 54 In dealing with these subjects, the chapter i s divided into four basic parts. The f i r s t three parts are devoted to i d e n t i f i c a t i o n and analysis of the model inputs, (a) the users' l i f e s t y l e , Section 3.3., (b) the users' socio-economic charac-t e r i s t i c s , Sections 3.4. to 3.7., and (c) the transportation system a t t r i b u t e s , Sections 3.8. to 3.12. The model formulation, and i t s operational c h a r a c t e r i s t i c s are discussed i n the f i n a l part, Sections 3.13. to 3.16. 3.2. Research Orientation I t might be appropriate at t h i s stage to define the study's basic approach to the problems to be encountered i n th i s chapter. B r i e f l y , these are as follows: (a) the conceptual framework of the model, (b) the s t a t i s t i c a l technique to be employed, and (c) the methods to be used i n c o l l e c t i n g the data required for the model c a l i b r a t i o n . In the conceptual analysis, t h i s study d i f f e r s from others i n that emphasis i s placed upon the user's s e n s i t i v i t y toward modal attr i b u t e s as an i n f l u e n t i a l mode choice deter-- minant. Such s e n s i t i v i t y refers to the s i g n i f i c a n c e the user attaches to various t r a v e l dimensions, cost, time, etc. I t i s hypothesized here that this s e n s i t i v i t y i s a function of two sets of var i a b l e s , the user's time budget, and some rela t e d socio-economic c h a r a c t e r i s t i c s , namely age and occupation. These two sets are employed i n t h i s study to provide an i n d i c a t i o n of some aspects of the user's l i f e s t y l e which may a f f e c t h i s s e n s i t i v i t y t o mode a t t r i b u t e s , which i n t u r n a f f e c t s h i s c h o i c e behaviour."'" In a d d i t i o n t o the i n d i v i d u a l ' s s e n s i t i v i t y , t h i s study i d e n t i f i e s the socio-economic c h a r a c t e r i s t i c s as mode c h o i c e determinants. I t i s the primary h y p o t h e s i s o f t h i s study t h a t the us e r ' s socio-economic c h a r a c t e r i s t i c s are not s t r o n g l y c o r r e l a t e d w i t h h i s s e n s i t i v i t y toward modal a t t r i b u t e s . In a d d r e s s i n g the second i t e m mentioned above (the s t a t i s t i c a l t echnique to be used i n data a n a l y s i s ) , the model to be employed i n t h i s study i s b a s i c a l l y a d i s a g g r e g a t e d s t o c h a s t i c model which e x p l a i n s and p r e d i c t s u s e rs ' mode c h o i c e s on the b a s i s o f t h e i r p e r s o n a l c h a r a c t e r i s t i c s , and c h a r a c t e r i s t i c s o f t h e i r i n d i v i d u a l mode c h o i c e s i t u a t i o n ( i . e . , d i f f e r e n c e s i n the a t t r i b u t e s o f the t r a n s p o r t a t i o n modes a v a i l a b l e to the u s e r ) . Each i n d i v i d u a l user i s t o be accounted f o r s e p a r a t e l y i n the model, and by means of d i s -c r i m i n a n t and p r o b a b i l i s t i c f u n c t i o n , the model produces the p r o b a b i l i t y t h a t a g i v e n user belongs t o one or another group of mode u s e r s , on the b a s i s o f the c h a r a c t e r i s t i c s mentioned above. A l l the necessary i n f o r m a t i o n on the users ' c h a r a c t e r -i s t i c s and a t t r i b u t e s of the mode c h o i c e s i t u a t i o n are to be c o l l e c t e d i n a q u e s t i o n n a i r e survey. This i s a b a s i c c o n s t r a i n t i n d e s i g n i n g t h i s r e s e a r c h , which i s r e c o g n i z e d i n s e l e c t i n g many of the model's procedures. 56 3 . 3 . L i f e s t y l e : Conceptual Considerations  and Application In an attempt to expand the explanatory function of mode choice analysis, this study provides an exploratory treatment of the concept of l i f e s t y l e . I t should be made clear , however, that our i n t e r e s t i n thi s concept i s confined to the possible r e l a t i o n s h i p i t might have to the users' mode choice behaviour, and hence, the study's approach may be d i f f e r e n t from another approach which scholars i n the f i e l d s of sociology or psychology may wish to pursue. The study's d e f i n i t i o n of l i f e s t y l e as the in d i v i d u a l ' s time and money budgets ref e r s s p e c i f i c a l l y to the number of hours and amount of d o l l a r s a l l o c a t e d to each of the a c t i v i t i e s i n which he i s engaged. Assuming that we can i d e n t i f y and describe t h i s pattern, the question we want to investigate at the outset of t h i s inquiry i s whether.such a pattern can be related to the i n d i v i d u a l ' s mode choice, and furthermore, how such a r e l a t i o n s h i p can be r a t i o n a l i z e d , i f i t a l l e x i s t s , and what evidence can be brought to support i t s v a l i d i t y . For the purpose of o u t l i n i n g a conceptual framework for t h i s inquiry, we may wish to consider the i n d i v i d u a l ' s time and money as limi t e d "resources". Various time- and money-consuming a c t i v i t i e s "compete" for these resources, and depending upon the in d i v i d u a l ' s personal values, needs and circumstances, c e r t a i n amounts of time and money are allocated to these a c t i v i t i e s — i n c l u d i n g t r a v e l . But since the choice of the t r a v e l mode e n t a i l s a "commitment" to 57 c e r t a i n expenditures of time and money, which u s u a l l y v a r y from one mode to the other, i t i s t h e r e f o r e a f f e c t e d by oth e r c h o i c e s t o be made co n c e r n i n g other time- and money-consuming a c t i v i t i e s . Thus, the r e l a t i o n s h i p o u t l i n e d above can be d e s c r i b e as f o l l o w s : L = f (m, , m „ m . m , 1 2 k n t l ' t 2 ' ' fck' ' tm ) where L i s the i n d i v i d u a l ' s l i f e s t y l e , t ^ i s the time a l l o c a t e d to a c t i v i t y i , m^ i s the money a l l o c a t e d t o a c t i v i t y i , t^. i s the time a l l o c a t e d t o the t r a v e l a c t i v i t y , m^ . i s the money a l l o c a t e d t o the t r a v e l a c t i v i t y , and n i s the number o f a c t i v i t i e s i n which the i n d i v i d u a l may be i n v o l v e d One o f the c o n s t r a i n t s o f the a l l o c a t i o n problem i s t h a t both time and money, f o r any g i v e n person, are l i m i t e d : T = t, + t„ + . . . . ' . . . . t ! 1 2 n M = m, + m„ + . m 1 2 n where T and M are c o n s t a n t s . There are other c o n s t r a i n t s to the problem of time and money a l l o c a t i o n . Such a l l o c a t i o n s hould r e f l e c t the i n d i v i d u a l ' s v a l u e system, h i s needs and c i r c u m s t a n c e s . Thus we may expect a person who v a l u e s c u l t u r a l o p p o r t u n i t i e s or 58 s o c i a l a c t i v i t i e s which put demand upon h i s time to opt, i n hi s mode choice behaviour, for time savings. Another person who puts extra working hours may behave s i m i l a r l y . In contrast, a person whose circumstances puts a great demand upon his f i n a n c i a l resources (he i s supporting a b i g family) may 2 opt for the le s s expensive mode. Thus, i n broad terms, time and money budgets may r e f l e c t c e r t a i n aspects of the user's l i f e s t y l e , which would, as a p r i o r i consequence, influence his mode choice behaviour. But the problem of time and money a l l o c a t i o n i s far more complex than what i s exhibited i n the conceptual frame-work outlined above i n that the e f f e c t of the i n d i v i d u a l ' s value system, his needs and circumstances on h i s a l l o c a t i o n of time and money budgets i s not s u f f i c i e n t l y c l a r i f i e d . What i s provided here i s a cursory treatment to what might be a complex phenomenon. Also, taking into consideration the methodological problem of c o l l e c t i n g such information i n a questionnaire survey (would a person express his value system adequately on a questionnaire?), we may recognize then that the task of accounting for the i n d i v i d u a l ' s l i f e s t y l e i n a -meaningful manner i s unmanageable for the study purposes. Furthermore, the l i t e r a t u r e provides neither a conceptu nor empirical analysis of the hypothesized r e l a t i o n s h i p between the user's time and money budgets to his mode choice behaviour, except for few attempts and references made by some authors which, while by no means provide the necessary support for the concept, are worth mentioning: 59 (a) Moses and Williamson's hypothesis on the r e l a t i o n -ship between the value of time and the user's a b i l i t y to convert time savings into working hours. In t h i s a n a l y s i s , reference has been made to the e f f e c t of other l e i s u r e a c t i v i t i e s i n increasing the value of time.saved i n trans-portation."^ (b) Brown and De Donnea's reference to the s e n s i t i v i t y toward time savings expressed by members of the managerial and professional occupations i n th e i r mode choice (See Section 3 . 5 . ) . Such s e n s i t i v i t y might be att r i b u t e d to l i f e s t y l e , since members of these groups, by the vi r t u e of t h e i r l e v e l of education, were l i k e l y to be engaged i n a va r i e t y of s o c i a l and c u l t u r a l a c t i v i t i e s . Because of the absence of a vigorous a n a l y t i c a l framework, and support of empirical data, the concept of l i f e s t y l e must be modified to be useful for the study's' purposes. This modification should be aimed at s i m p l i f y i n g the concept, or narrowing i t down.where i t becomes f e a s i b l e for t h i s study to outline a p r a c t i c a l method fo r c o l l e c t i n g the data required to support i t s v a l i d i t y . Furthermore, the attempt should be made to bring the modified concept under the "umbrella" of previous research findings. That i s , the model structure which i s to be formulated accordingly should be made to stand the scrutiny of empirical v e r i f i c a t i o n when compared with the findings of other researchers. The proposed modification i s as follows: i n as f a r as mode choice i s concerned, the user's l i f e s t y l e i s l i k e l y 60 to be expressed as a s e n s i t i v i t y , or a set of s e n s i t i v i t i e s , toward the relevant modal a t t r i b u t e s . S e n s i t i v i t y i s defined here as the s i g n i f i c a n c e the user attaches to any given a t t r i b u t e , be i t t r a v e l time, cost, etc. Thus, persons who are pressed f o r time because of t h e i r c u l t u r a l and s o c i a l engagements, or other time-demanding duties, would be s e n s i t i v e to the time a t t r i b u t e i n t h e i r choice behaviour. In comparison, those whose f i n a n c i a l resources are burdened by c e r t a i n obligations would be s e n s i t i v e to the cost a t t r i b u t e , a l l other factors being equal. Thus, for the s p e c i f i c purpose of formulating the 7 study's model, the user's s e n s i t i v i t y toward modal at t r i b u t e s can be used as a surrogate for his l i f e s t y l e i n that i t expresses those aspects of his l i f e s t y l e which a f f e c t h i s mode choice behaviour. More s i g n i f i c a n t l y , the l i t e r a t u r e provides s u f f i c i e n t evidence to i n d i c a t e that these s e n s i t i v -i t i e s may vary independently from h i s socio-economic charac-t e r i s t i c s (See Section 3.13.), and hence, for the purpose of model formulation, there i s a strong j u s t i f i c a t i o n to incor-porate both kinds of variables as model inputs; (A possible i n t e r p r e t a t i o n for this independence i s that v a r i a t i o n i n l i f e s t y l e may create divergent s e n s i t i v i t i e s within the same socio-economic group. We have more to say about s i m i l a r findings and their interpretations i n Section 3.13.). The question remains as to what are the a t t r i b u t e s to which the user may be s e n s i t i v e , and how can t h i s s e n s i t i v i t y be expressed i n a manner which allows for t h e i r incorporation i n the model. On the basis of the work conducted by Hartgen and Tanner, who attempted to i d e n t i f y the s i g n i f i c a n c e of some t h i r t y transportation system at t r i b u t e s , the following can be i d e n t i f i e d as the most relevant, and are to be used f o r the study purposes: - t o t a l t r a v e l cost, - t o t a l t r a v e l time, - excess t r a v e l time, - f l e x i b i l i t y of departure time, and - comfort l e v e l of the t r i p . Turning our attention to the problem of expressing these s e n s i t i v i t i e s , two methods can be i d e n t i f i e d : (a) to ask the user to rank these attributes by t h e i r order of s i g n i f i c a n c e to h i s mode choice, and (b) to ask the user to express t h e i r s i g n i f i c a n c e along a Linkert scale. Both methods could prove operational, but for the purposes of t h i s study, an a r b i t r a r y choice i s made for the l a t t e r . Having placed emphasis oh the user's s e n s i t i v i t y toward modal a t t r i b u t e s , this study goes a step further by attempting to explain this s e n s i t i v i t y , or rather, set of s e n s i t i v i t i e s by r e l a t i n g these to a s i m p l i f i e d account for the user's time budget and some of his r e l a t e d socio-economic 4 c h a r a c t e r i s t i c s , namely age and occupation. A simple account for the user's time budget i s proposed here as the number of hours a l l o c a t e d by the user to work, family (or at home, with r e l a t i v e s ) , s o c i a l and c u l t u r a l a c t i v i t i e s and outdoor recreation. I t i s unfortunate that the l i t e r a t u r e does not provide any useful analysis of the rel a t i o n s h i p of time budget and mode choice behaviour, and thus we have to accept the above mentioned c l a s s i f i c a t i o n which i s developed on the basis of the author's subjective evaluation. The addition of the user's time budget, age and occupation (as an explanatory v a r i a b l e to the user's sensi-t i v i t y toward modal attributes) to the model structure i s an experimental step which i s hoped to provide some badly needed evidence on a plausible hypothesis, should t h i s proposed model be implemented. This r e l a t i o n s h i p i s expected to improve the rati o n a l e of the model and possibly give the researcher some hints as to the causal r e l a t i o n s h i p s underlying users' mode choice behaviour. However, since t h i s i s an experimental undertaking, the model should be formulated so as to be capable of explaining users' choices on the basis of these sets of c h a r a c t e r i s t i c s alone ( i f necessary): (a) s e n s i t i v i t y toward modal a t t r i b u t e s , (b) the users ' socio-economic c h a r a c t e r i s t i c s , and (c) at t r i b u t e s of the mode choice s i t u a t i o n . In other words, the r e l a t i o n s h i p hypothe-sized here to explain the user's s e n s i t i v i t y should be "attached" to the model structure as an " o f f - l i n e " procedure whose success or f a i l u r e should not a f f e c t the explanatory power of the model, at l e a s t not i n the s t a t i s t i c a l sense. To sum up this section, the user's s e n s i t i v i t y toward modal attr i b u t e s i s selected as a surrogate for h i s l i f e s t y l e for the purpose of model formulation. I t i s argued here that such s e n s i t i v i t y i s not strongly correlated with his socio-economic c h a r a c t e r i s t i c s (except f o r , perhaps, age and occupation). An additional step i s developed i n the model to explain this s e n s i t i v i t y on the basis of some in d i c a t i o n s of the user's l i f e s t y i e : ^ time budget a l l o c a t i o n , age and occupation. In the following sections, each of the socio-economic c h a r a c t e r i s t i c s and at t r i b u t e s of the model choice s i t u a t i o n i s analysed separately. In the f i n a l part of t h i s chapter, the attempt i s made to use the information and concepts developed on mode choice determinants i n s t r u c t u r i n g the study's model, and that i s where much of the study's concepts are to become considerably clearer to the reader. THE USERS' SOCIO-ECONOMIC CHARACTERISTICS The following four sections are devoted to analysis of the user's socio-economic c h a r a c t e r i s t i c s , income, occupation, age and sex, i n r e l a t i o n s h i p to his mode choice behaviour. These variables have been dealt with i n d e t a i l by most studies i n t h i s f i e l d , and only a b r i e f reference i s made here to demonstrate the extent to which the study's approach i s s i m i l a r . to or d i f f e r e n t from others. 3.4. The Household Income The influence of the income variable on the user's behaviour can be f e l t i n several ways: (a) A v a i l a b i l i t y of f i n a n c i a l resources i s l i k e l y to induce the user to attach less s i g n i f i c a n c e to the cost a t t r i b u t e , yet a greater s i g n i f i c a n c e to other a t t r i b u t e s : 64 t r a v e l time, l e v e l of comfort, etc. (b) Income, i n general, determines the s o c i a l class of the user, and hence the biases to which he may be sub-jected i n his s o c i a l environment. (c) The income l e v e l also determines the a v a i l a b i l i t y of a l t e r n a t i v e a c t i v i t i e s i n which the user and his family can be engaged. The a v a i l a b i l i t y of such al t e r n a t i v e s i s l i k e l y to increase the user's s a t i s f a c t i o n to be derived from his out-of-work time, since he has the resources to s a t i s f y his personal taste. Accordingly, the demand upon his time i s greater, and his time value i s l i k e l y to be higher, which, i n ' turn, a f f e c t s his d i s u t i l i t y of the time spent i n the t r a v e l a c t i v i t y . The use of the household income as an input v a r i a b l e r a i s e s a d i f f i c u l t y which should be recognized here. Since those members of the household who contribute to i t s t o t a l earnings are l i k e l y to have more l i b e r t y i n u t i l i z i n g these f i n a n c i a l resources to s a t i s f y t h e i r personal taste, t h e i r t r a v e l d i s u t i l i t y i s l i k e l y to be d i f f e r e n t from other members of the household. The use of the household income variable ignores t h i s d i f f e r e n c e . To solve t h i s problem, De Donnea entered a dummy variable i n the mathematical function to indicate whether or not the user i s the head of the household, 5 i . e . , the main earner i n the family. Although t h i s variable, was found to be s i g n i f i c a n t , i t should be remembered that t h i s i s only a p a r t i a l solution since no account i s made yet for the p o s s i b i l i t y that a second income earner (the wife, the son, . . .) could be i n s i m i l a r p o s i t i o n to u t i l i z e these f i n a n c i a l resources to s a t i s f y his taste i n his mode choice behaviour. This study pursues a d i f f e r e n t approach to the above mentioned problems. While i t accounts for the household income, i t also accounts for other r e l a t e d c h a r a c t e r i s t i c s : (a) Whether the user has an access to an al t e r n a t i v e mode, and i f so (b) His s e n s i t i v i t y toward modal a t t r i b u t e s . Thus, information i s obtained on whether the user i s i n the po s i t i o n to s a t i s f y his taste (does he have a choice?), and i f so, whether he i s se n s i t i v e to the cost a t t r i b u t e (can he u t i l i z e the f i n a n c i a l resources of the family to s a t i s f y his tast e ? ) . Such information provides s u f f i c i e n t d e s c r i p t i o n of the influences which varies for d i f f e r e n t members of the household with respect to th e i r mode choice. 3.5. Occupation Since occupation i s not a continuous v a r i a b l e , i t s influence on the user's mode choice can only be i d e n t i f i e d i n c r o s s - c l a s s i f i c a t i o n s t a t i s t i c a l analysis, where i t i s pos s i b l to compare the percentage of any given group of mode users from one occupational group to the other. Because of thi s constraint, most of the studies employing a mathematical function did not account for t h i s v a r i a b l e . Among the few who explored t h i s variable was De Donnea who found that the percentage of car use was s i g n i f i c a n t l y higher i n two occu-pational groups, executive professions and blue c o l l a r workers He considered t h i s to be a bias for car use and accounted for i t s influence by entering a dichotomous variable i n the model to indicate whether or not the user belonged to any of the two occupations. Brown's study revealed some unexpected r e s u l t s on the e f f e c t of the occupation variable on users' behaviour. In his sample, he found a higher percentage of bus users i n managerial and professional employees and secretaries than 7 expected.. C l e a r l y , t h i s behaviour was not expected from the former two groups. In a further analysis based upon what was considered users' stated preferences, Brown found that managerial and professional employees, more than any other groups, were se n s i t i v e toward t r a v e l cost, and that t h i s s e n s i t i v i t y was 8 not shared by the s e c r e t a r i a l professions. Again, these were unexpected findings. Yet, the more i n t e r e s t i n g observation encountered i n this study was that professional employees were more time-sensitive than other users i n the same income category. This s e n s i t i v i t y might be att r i b u t e d to the p o s s i b i l i t y that professional employees were engaged i n more c u l t u r a l or s o c i a l a c t i v i t i e s , or that they have to put more hours i n t o work to meet the r e s p o n s i b i l i t i e s of the job. Thus, a l l o c a t i o n of time budget could have possibly provided an explanation to these p a r t i c u l a r observations. The observations made by Brown and De Donnea are s i g n i f i c a n t empirical findings. Yet, on a p r i o r i ground, there appears to be no p l a u s i b l e r e l a t i o n s h i p between the user's occupation and his mode choice, except, perhaps, through 67 the association of the former with the user's income and his l i f e s t y l e . Since both these c h a r a c t e r i s t i c s are accounted for i n the.study's model/ the use of the occupation variable as a d i r e c t input to the model would probably be redundant. Information on the occupational status of the user, however, could be useful i f i t can be r e l a t e d to the user's l i f e s t y l e , or h i s s e n s i t i v i t y toward modal a t t r i b u t e s . Since future demand fo r various occupations for any given region can roughly be predicted, i t would be usef u l f o r the researcher to i d e n t i f y the influence Of occupation on the user's mode choice behaviour. This approach i s adopted i n the present study, and as mentioned i n Section 3.3., the attempt i s to be made to incorporate such a r e l a t i o n s h i p into the model structure. (See Sections 3.14, and 3.16.). 3.6. Age The r e l a t i o n s h i p between the user's age and his mode choice has been i d e n t i f i e d i n the studies conducted by Brown, De Donnea, Warner, Lisco and Lave."*"0 Although these authors agreed on the s i g n i f i c a n c e of the age c h a r a c t e r i s t i c , they employed d i f f e r e n t methods to account for t h i s v a r i a b l e . These differences r e f l e c t a discrepancy among these authors i n understanding how age affected the users' behaviour. B r i e f l y , three methods were i n use: (a) Age was accounted for as a continuous variable to be entered i n the model without any transformation. Roughly • speaking, age and car use were found to be p o s i t i v e l y correlated. 68 (b) Warner used a d i f f e r e n t technique: users were grouped into nine categories and were given a code number, with code number 1 given to the younger group (16 to 19 years) and 9 f o r the older group (65 years and o l d e r ) . He entered the natural logarithm of the code, number of the user's 12 age group into the model to account for t h i s c h a r a c t e r i s t i c . (c) By c r o s s - c l a s s i f i c a t i o n analysis of mode users, Brown found that the bias for car use was evident i n the age 13 group 40 to 60, while De Donnea contended that such a bias 14 was common among the 25 to 55 age group. De Donnea accounted ' ' • • ! for the e f f e c t of age as a dichotomous v a r i a b l e i n the same manner he accounted for the occupational b i a s . For the purposes of t h i s study, however, the use of the age c h a r a c t e r i s t i c as an input to the model ra i s e s a d i f f i c u l t y s i m i l a r to that of using the user's occupational status i n that there i s no i d e n t i f i a b l e cause-effect r e l a t i o n s h i p between this variable and mode choice, except through the e f f e c t of the user's time budget, s e n s i t i v i t y to modal attributes or l i f e s t y l e . Indeed, one expects the user to change his l i f e s t y l e as he grows i n age. Lave suggested that, " i t i s possible that there i s a systematic r e l a t i o n s h i p between the shape of a commuter preference function ( d i s u t i l i t y 15 function?) and h i s age or sex." Hence, i f l i f e s t y l e , or some measure of i t , i s accounted for , the use of the age variable would be redundant. But, again, t h i s i s not completely true. Along with the occupational status, the age c h a r a c t e r i s t i c of the user 69 can be employed to explain and eventually p r e d i c t the user's s e n s i t i v i t y to the modal a t t r i b u t e s . The age c h a r a c t e r i s t i c s of a given regional population can be roughly predicted for future years by demographic analysis, and hence i t may be possible to foresee i t s e f f e c t upon the s e n s i t i v i t y of the user and his mode choice behaviour. 3.7. Sex At t h i s early stage of research on the r e l a t i o n s h i p between the user's l i f e s t y l e and his mode choice, i t i s not conceptually clear how the sex va r i a b l e would a f f e c t the user's behaviour, although such a r e l a t i o n s h i p i s already documented i n the l i t e r a t u r e . The notion that males are "car oriented" was supported by Brown, Warner, Lisc o and L a v e . ^ This r e l a t i o n s h i p can be explained i n two ways: (a) Female members of the household have "less access" to the car. For example, i f the husband needs the car to commute to work, he. i s l i k e l y to have a p r i o r i t y i n using the car. To take t h i s p o s s i b l i t y i n t o consideration, Lisco accounted for two c h a r a c t e r i s t i c s of the mode choice s i t u a t i o n , (i) whether the husband was the only person i n the household . having access to the car use, and ( i i ) whether the wife needed to drive to work independently. An affirmative answer to each of these questions was entered separately as a dichotomous v a r i a b l e . In t h i s study, we need not be concerned with t h i s problem, since the data analysis i s confined i n i t i a l l y to those 70 users having access to the car, i n addition, of course, to t r a n s i t . The model does not make a d i s t i n c t i o n between "degrees" of a c c e s s i b i l i t y to car, and simply assumes that i f the user has access to the car, i t i s an undeterred access. (b) Females have generally an aversion to car d r i v i n g . This i s obviously an anti-modal bias, and indeed could be the only argument encountered i n the l i t e r a t u r e to support the notion that such a bias e x i s t s . I t i s for t h i s reason that the sex v a r i a b l e should be accounted for as a dichotomous var i a b l e i n the study's model. THE TRANSPORTATION SYSTEM ATTRIBUTES 3.8. Problems of I d e n t i f i c a t i o n In accounting for the transportation system a t t r i b u t e s , several s i g n i f i c a n t research problems a r i s e for inquiry and s o l u t i o n : (a) what attributes should the model i d e a l l y account for? ( (b) what attributes i s i t possible and p r a c t i c a l to account for? (c) should modal attr i b u t e s be accounted for as they are, or as they are perceived by the user? In addressing the f i r s t two problems, the argument previously advanced i n Section 2.2. that a transportation mode i s to be considered, at l e a s t conceptually, as a bundle of a t t r i b u t e s , should be brought into discussion. Now, with 71 emphasis being placed on the model formulation rather than the conceptual analysis, we may question what such a "bundle" contains. Is the colour of the veh i c l e , for example,.an attribute? Obviously, c e r t a i n c r i t e r i a must be developed to as s i s t i n i d e n t i f y i n g these att r i b u t e s f or the purposes of the study: (1) There should be s u f f i c i e n t evidence, to suggest that such a t t r i b u t e s are s i g n i f i c a n t i n a f f e c t i n g users 1 behaviour. Since t h i s study i s not concerned with c o l l e c t i n g such information d i r e c t l y from the users, evidence should be sought i n previous research findings. (2) For any given mode choice s i t u a t i o n , a t t r i b u t e s which are i n f l u e n t i a l i n a f f e c t i n g the users 1 behaviour are those which d i f f e r s i g n i f i c a n t l y from one mode to the other. As ah example, Hartgen found that users i n his sample ranked "a r r i v i n g without an accident" as a highly s i g n i f i c a n t 17 att r i b u t e of the transportation mode they would use.. But, i n the mean time, the users 1 s a t i s f a c t i o n with the two modes available, car and t r a n s i t , did not d i f f e r s u b s t a n t i a l l y with respect to thi s p a r t i c u l a r a t t r i b u t e . Thus, the a t t r i b u t e , a l b e i t s i g n i f i c a n t , evidently would have no e f f e c t on the users ' behaviour i n this or a s i m i l a r mode choice s i t u a t i o n . With the choice between car and t r a n s i t being the main concern of t h i s study, attention should be given to those system attributes which have been i d e n t i f i e d by the users as sighif i c a n t , : a n d c l e a r l y d i f f e r between these two modes. To i d e n t i f y these variables, researchers have taken two d i r e c t i o n s : the f i r s t , by analysing the users' actual mode choices, using discriminant analysis or s i m i l a r s t a t i s t i c a l techniques; the second, by asking the users to i d e n t i f y the relevant a t t r i b u t e s i n a questionnaire survey. By pursuing the f i r s t approach, the following attributes were i d e n t i f i e d : (a) t o t a l t r a v e l time. (b) excess t r a v e l time. (c) t r a v e l cost. These findings c l e a r l y conform to research expectations and i no s u b s t a n t i a l argument i s required at t h i s stage to j u s t i f y t h e i r i n c l u s i o n . Researchers who pursued the second approach a r r i v e d at re s u l t s which confirmed these fi n d i n g s , and i d e n t i f i e d , i n addition, two further a t t r i b u t e s : (d) a r r i v a l at intended time. (e) comfort l e v e l of the v e h i c l e . These two attributes require some elaboration. The former i s p a r t i c u l a r l y important i n circumstances where, f o r example, the t r a f f i c congestion i s such that the a r r i v a l time becomes uncertain, or that the scheduling of the t r a n s i t s e r v i c e does not match the user's intended departure time. Hartgen found that users i n h i s sample were s a t i s f i e d with the car i n t h i s respect, but less so with the a v a i l a b l e t r a n s i t 18 s e r v i c e . In essence, the user's s a t i s f a c t i o n or d i s s a t i s f a c t i o n with the scheduling of t r a n s i t service i s obviously dependent upon the frequency of t r a n s i t s e r v i c e f o r the intended t r i p time. Frequent service, say at three minute i n t e r v a l s , can provide the user with the same degree of f l e x i b i l i t y as that provided i n car use. The longer the i n t e r v a l , the greater the i n f l e x i b i l i t y and d i s s a t i s f a c t i o n . Thus, i n mode choice situations i n v o l v i n g t r a n s i t , frequency of t r a n s i t service can be used as a proxy f o r f l e x i b i l i t y of departure time. Accounting for the comfort l e v e l a t t r i b u t e s r a i s e s several d i f f i c u l t i e s which have been discussed i n Section 2.7. Hartgen and Tanner, for example, did not i d e n t i f y the comfort l e v e l as a sing l e v a r i a b l e , rather, as a "composite" of many 19 at t r i b u t e s of the t r a v e l a c t i v i t y : (a) avoid walking less than a block, (b) v e h i c l e unaffected by weather, (c) r i d e i n a clean v e h i c l e , and (d) r i d e i n a modern v e h i c l e . The f i r s t of these i s obviously correlated with the excess t r a v e l time, which i s accounted f o r separately. The other a t t r i b u t e s .are e s s e n t i a l l y a de s c r i p t i o n of the comfort l e v e l of the v e h i c l e i t s e l f . To account for t h i s v a r i a b l e , a simpler, more economic method i s proposed i n Section 3.11., to allow for one-dimensional measure of t h i s complex a t t r i b u t e . The studies consulted by th i s author provide strong evidence to i n d i c a t e that the attributes discussed so far. are s i g n i f i c a n t determinants of users' mode choice behaviour. Hartgen's study c l a r i f i e d this question further by i d e n t i f y i n g several a t t r i b u t e s which are not s i g n i f i c a n t : a v a i l a b i l i t y 20 of package space, and the provision of bus s h e l t e r . But i n either case, we have no assurance that the l i s t i s conclusive, since many other at t r i b u t e s have not been explored as to t h e i r e f f e c t on users' behaviour, e.g., privacy. Indeed, t h i s element of uncertainty i s inherent i n the conceptual framework of t h i s type of model, and to the abstract mode choice approach, where the universe of the relevant system a t t r i b u t e s i s i n i t i a l l y defined by the researcher. Thus, the exercise of a c e r t a i n degree of s u b j e c t i v i t y i s i n e v i t a b l e . The other problem to be addressed i n t h i s section i s whether an account should be made for these a t t r i b u t e s as they ac t u a l l y e x i s t (and, hence, the measurements are to be taken by the researcher to ensure the o b j e c t i v i t y of reporting), or, al t e r n a t i v e l y , attributes should be accounted for as they are perceived by the user (and thus users' reported data are to be employed instead). This problem was addressed by few researchers, Thomas and Thompson, and Watson, and the discrepancy between the users' reported a t t r i b u t e s and actual a t t r i b u t e s was i d e n t i f i e d and att r i b u t e d to several 21 reasons: (a) personal bias: i n route choice s i t u a t i o n s , users tended to exaggerate the benefits they,receive from the chosen route, and under-value the p o t e n t i a l benefits of using the alter n a t i v e route. This was obviously an e f f o r t to j u s t i f y a choice already made. In mode choice s i t u a t i o n s , users are very l i k e l y to behave i n s i m i l a r manner. (b) rounding errors: most users tended to define t r a v e l time i n "lumps" of f i v e minutes, thus expressing t h e i r personal bias by rounding t r a v e l times upward or downward to multiples of f i v e minutes, so that t h e i r mode choice appears j u s t i f i a b l e . (c) perceptual errors: the users' perception of time and cost of tr a v e l could be far from correct, p a r t i c u l a r l y when onerous a c t i v i t i e s , l i k e walking or waiting i n inclement weather conditions, are involved. Despite the v a l i d i t y of these observations, i t i s obvious that the user's mode choice i s affected by what he perceives as modal a t t r i b u t e s , and d e f i n i t e l y not by any of the objective measurements made by the researcher. Therefore, a l l researchers who investigated t h i s problem unanimously employed the users' reported data, rather than o b j e c t i v e l y measured a t t r i b u t e s . The purpose of thi s section was to discuss the problems associated with the i d e n t i f i c a t i o n of the modal at t r i b u t e s which are relevant to users' behaviour. In the following sections, attention i s given to methods of manipulating these at t r i b u t e s as model inputs. It should be remembered i n examining these.attributes that our i n t e r e s t i s not with the system attributes per se, rather, with the comparison between attributes of al t e r n a t i v e modes. The method of making such a comparison may vary from one modal a t t r i b u t e to the other, and, hence, each of these i s discussed separately i n the following sections. 3 . 9 . Travel Time B a s i c a l l y , there are two methods to account f o r t r a v e l time i n any mode choice s i t u a t i o n : (a) As absolute savings i n the t r a v e l time by using 22 the f a s t e s t mode, that i s , T^ - T^. (b) As the r a t i o of t r a v e l time by one mode to the other, T 2 / T ] [. 2 3 The diff e r e n c e between the two approaches i s not t r i v i a l . In the f i r s t case, a time savings of, say, f i v e minutes i n a t r i p of ten minutes would have the same entry to the model as f i v e minutes savings i n a f i f t y minute t r i p . In the second case, the t o t a l length of the t r i p by each mode i s accounted for i n the r a t i o . Quarmby tested the two approaches and found that, i n his sample, the former approach y i e l d e d a greater explanatory 24 power for users' behaviour. His findings may not be applicable to other samples: i t i s possible, one can argue, that the l a t t e r approach i s more appropriate i n samples where there i s considerable variations i n the t r a v e l times. But since no evidence was encountered to support t h i s argument, i t i s suggested that the former approach i s to be adopted for the study's purposes—subject to further t e s t i n g to confirm 25 Quarmby's fin d i n g s . The other problem to be addressed i n t h i s section i s whether excess t r a v e l time should be accounted for d i f f e r e n t l y from the i n - v e h i c l e time. As demonstrated i n Section 2.6., there i s evidence to suggest that the d i s u t i l i t y of excess time i s considerably higher than the i n - v e h i c l e time, presumably 2.5 to 3 times the l a t t e r . To account for t h i s d i f f e r e n c e , 77 two methods can be employed: (a) to enter savings i n excess time and i n - v e h i c l e time separately as two var i a b l e s . (b) to convert excess time fo r each mode to i n -v e h i c l e time (by multiplying the former by a fact o r of 2.5 or 3) and enter savings i n the t o t a l t r a v e l time, as computed a f t e r the conversion, as a single v a r i a b l e . I n t u i t i v e judgement may indicate that the d i s u t i l i t y of excess t r a v e l time could be considerably higher than 2 « 5 or 3 times the in- v e h i c l e time i n t r a v e l conditions where the user i s exposed to inclement weather, for example. Thus, i t would be inappropriate to accept a standard conversion factor i n p r i n c i p l e . This suggests the f i r s t approach i s more, useful for the study. 3.10. Travel Cost Accounting for the cost a t t r i b u t e d i f f e r e n c e among t r a n s i t a l t e r n a t i v e s i s a simple procedure by which the t r a n s i t fares are compared from one mode to the other. The d i f f i c u l t y appears i n mode choice situations i n v o l v i n g the car a l t e r n a t i v e , - since t r a v e l cost by car i s not c l e a r l y defined. Conceptually, t r a v e l cost i n this case can be accounted for i n two d i f f e r e n t ways : (a) The average cost approach, that i s , the t r a v e l cost for the t r i p including parking charges; operating cost, gasoline, o i l , l u b r i c a t i o n , maintenance, and re p a i r , annual depreciation i n car value, insurance cost, and the cost of the committed investment (interest to be paid). By summing 78 these expense items for say, a year, and d i v i d i n g these costs by the annual mileage t r a v e l l e d , i t i s then possible to derive the average t r a v e l cost per mile. (b) The marginal cost approach, which e n t a i l s the additional or out-of-pocket cost of any given t r i p : gasoline, o i l , parking charges, and road t o l l s i f any. A substantial argument can be developed to support each of these approaches. In the f i r s t case, the r e a l costs of t r a v e l by car are f u l l y expressed, i n the l a t t e r , only costs which are paid s p e c i f i c a l l y for making t h i s t r i p . Lave argued that i f the user purchased the car for the purpose of commuting to work, then the average cost of t r a v e l to work should be accounted f o r . But, i f the car was purchased to serve mainly for r e c r e a t i o n a l purposes, the cost of commuting to work should :be accounted for as the marginal cost (the car i s already i n the user's possession, what i s the cost of making a work t r i p ? ) . The argument for the use of the f i r s t approach i n one case, and the second approach i n another case was obviously tenuous since i t was d i f f i c u l t to define the user's motivation i n purchasing the car, a f a c t which Lave recognized i n his a r t i c l e . 2 ^ This problem was addressed i n Quarmby's work, who stressed that the user's perception of t r a v e l cost should be recognized as the appropriate entry to account for the cost . a t t r i b u t e . i n the discriminant analysis he conducted on the users' choices, he entered the average and marginal cost of t r a v e l , each at a time, and concluded that the l a t t e r y i e l d ed the model a higher explanatory power.^1 i t can be i n f e r r e d from t h i s conclusion that the users' behaviour was affected by the marginal, rather than the average cost of t r a v e l . Since this inference i s drawn from the users 1 behaviour, i t can also be stated that i t defines the users' perception of t r a v e l cost by car. Accounting for the t r a v e l cost d i f f e r e n c e , rather than the cost r a t i o , seemed to be the approach most researchers 2 8 agreed upon, except for Warner. The cost d i f f e r e n c e approach, i n the opinion of th i s author, i s more p l a u s i b l e , since the user's perception of monetary units i s uniform, and i s l i k e l y to be independent from the t o t a l cost of t r a v e l . 3.11. The Comfort Level The discussion introduced i n Section 2.7. i d e n t i f i e s some of the research d i f f i c u l t i e s to be faced i n accounting • for t h i s a t t r i b u t e . In addition to these conceptual d i f f i -c u l t i e s , empirical research findings have been incon s i s t e n t i n d efining the r o l e of the comfort a t t r i b u t e i n a f f e c t i n g . users' behaviour. Brown and Hartgen seemed to i n d i c a t e that the comfort l e v e l of the vehicle was not a very s i g n i f i c a n t . 29 factor, while Lisco's study concluded the opposite. Since no d e f i n i t i v e evidences were encountered i n the l i t e r a t u r e to suggest that the comfort l e v e l can be s a f e l y ignored, the model to be formulated i n t h i s study should i n i t i a l l y account for i t . The method suggested here i s to. enter this a t t r i b u t e as a continuous v a r i a b l e . The inform-80 ation required on the user's evaluation of the comfort l e v e l of a l t e r n a t i v e modes can be obtained i n a questionnaire survey, where the user i s asked to rank the comfort of each v e h i c l e along a Li n k e r t scale. The difference, expressed as abstract 3( units of measurement, can be entered as a continuous v a r i a b l e . For the part of the t r i p spent outside the vehicle i n walking, waiting or during the transfer time, since i t i s considered by most users as sheer discomfort, the account i s made through i t s association with the length of excess t r a v e l time. I t i s p a r t i c u l a r l y d i f f i c u l t , and possibly redundant, to provide a separate account f o r the user's comfort l e v e l during t h i s time. 3.12. Frequency of Transit Service This a t t r i b u t e expresses the user's f l e x i b i l i t y i n se l e c t i n g his departure time, and subsequently, his a r r i v a l time at destination. In comparing various modes with respect to t h i s a t t r i b u t e , the difference i n the frequency of service would be a v a l i d measure, with the increase i n frequency of one mode r e l a t i v e to the other i n d i c a t i n g a greater degree - of f l e x i b i l i t y . For example, i f the frequency of service of mode A i s ten minutes, and that of mode B i s f i v e minutes, the differe n c e indicates the advantage of B over A. The same method i s also applicable to mode choice s i t u a t i o n s i n v o l v i n g car and t r a n s i t . But, since the car i s usually immediately a v a i l a b l e , the difference between departure times can, i n t h i s case, be set equal to zero, while such a 81 diff e r e n c e f o r the t r a n s i t service, whether small or large, would give an i n d i c a t i o n of i t s r e l a t i v e disadvantage r e l a t i v e to the car.. Thus, instead of c a l c u l a t i n g the difference i n the frequency of the two modes, we need only to enter the time period between departures of the t r a n s i t service, as a continuous v a r i a b l e . THE MODEL FORMULATION 3.13. The Problem of C o l i n e a r i t y Having defined the universe of mode choice determin-ants, we may now address the problem of c o l i n e a r i t y , which has a considerable bearing on the model performance. By scanning the choice determinants i d e n t i f i e d so f a r , one may suspect that the user's l i f e s t y l e , or i t s surrogate, the s e n s i t i v i t y toward modal a t t r i b u t e , may be correlated with the user's socio-economic c h a r a c t e r i s t i c s . This problem should be investigated to ensure that the i n c l u s i o n of some measure of the user's l i f e s t y l e would not be redundant. The study's hypothesis postulates, however, that the socio-economic variables do not provide an adequate des c r i p t i o n for the user's personal c h a r a c t e r i s t i c s , and that by accounting for some indicat i o n s for his l i f e s t y l e , the explanatory power of the model would be expanded. The purpose of t h i s section i s to discuss previous research findings to t h i s e f f e c t . From previous research conclusions, some inferences can be drawn to suggest that the i n c l u s i o n of l i f e s t y l e as mode choice determinant could have possibly been useful i n 82 explaining some aspects of users' behaviour. Further, a firm assertion can be made from these findings that the user's s e n s i t i v i t y toward modal attributes i s not, i n general, strongly correlated with his socio-economic c h a r a c t e r i s t i c s : (a) De Donnea reported that bias f o r car use i n his sample was i d e n t i f i e d among the executive professions and 3 1 blue c o l l a r workers. Since these two groups have apparently no s o c i a l or economic c h a r a c t e r i s t i c s i n common, an adequate explanation f o r such a bias cannot be derived from the socio-economic c h a r a c t e r i s t i c s alone. Likewise, Brown f i n d i n g that the share of managerial and s e c r e t a r i a l professions i n t r a n s i t 3 2 usage was unexpectedly high i s d i f f i c u l t to i n t e r p r e t . (b) Further s i g n i f i c a n t findings were reported i n Brown's study as an outcome of the applications of the propensity model. The purpose of this model was to i d e n t i f y the changes i n the system attributes so that a given proportion of the users would s h i f t to t r a n s i t use. By r e l a t i n g the users' socio-economic c h a r a c t e r i s t i c s to t h e i r s e n s i t i v i t i e s to changes i n the modal a t t r i b u t e s , Brown reported some unexpected findings i n that the income influence was inconsistent, at l e a s t i n some instances, with what one would expect on i n t u i t i v e grounds. In examining t h e i r propensity to s h i f t to t r a n s i t use, higher and middle income groups demanded lower o v e r a l l out-of-pocket cost (marginal cost?) while the lower income group seemed to 33 be s a t i s f i e d with the l e v e l of cost. Further to t h i s , i t i s the medium and high income groups who would t o l e r a t e longer walking distance at the r e s i d e n t i a l end, while the lower 83 income group would not. 3^ The occupation v a r i a b l e seemed to show s i m i l a r influences (which are also d i f f i c u l t to explain) i n that managerial and professional groups demanded lower fares, and tolerated longer walking distance at the r e s i d e n t i a l end of the t r i p ( i n s e n s i t i v i t y to t r a v e l time?), while labourers and craftsmen were unwilling to t o l e r a t e longer walking 35 distances. Furthermore, labourers, sales people and managers were above the average i n t h e i r s e n s i t i v i t y to 36 t r a v e l cost. The t e s t conducted to i d e n t i f y the influence of age indicated that those under for t y expressed higher 37 propensity to s h i f t to t r a n s i t . A l l of these observations are apparently inconsistent with what one would i n t u i t i v e l y expect on the r e l a t i o n s h i p between the user's socio-economic c h a r a c t e r i s t i c s and his s e n s i t i v i t y toward changes i n various t r a v e l a t t r i b u t e s . They are also i n d i r e c t contrast with empirical research findings on the r e l a t i o n s h i p between the user's income and his evaluation of t r a v e l time savings (See Section 2.6.). Thus, there i s s u f f i c i e n t ground to suspect that the socio-economic c h a r a c t e r i s t i c s do not provide an adequate d e s c r i p t i o n for the user's c h a r a c t e r i s t i c s . This f i n d i n g supports the argument introduced i n Section 3.2., thus, p l a c i n g the concept of l i f e s t y l e as an explanatory t o o l i n more favourable l i g h t . I t i s conceivable that l i f e s t y l e might have provided an explan-ation for the observations mentioned above, f o r example, that Brown's sample included a substantial number of lower income 84 users whose l i f e s t y l e put a great demand upon t h e i r time. Hence, the s e n s i t i v i t y of these users to time savings. (c) Beier's work provided by far the strongest evidence to indicate that there i s no c o r r e l a t i o n between the user's income or age, and his sensitiveness toward modal a t t r i b u t e s . The data c o l l a t e d by Beier provided information on the user's income, age, his ranking of the modal attributes and the improvements i n t r a n s i t service which would motivate 3 8 a s h i f t to t r a n s i t use. By grouping users according to t h e i r age and income categories, Beier found that the user's ranking of the r e l a t i v e s i g n i f i c a n c e of modal a t t r i b u t e s , as w e l l as t h e i r i n d i v i d u a l propensity to s h i f t to t r a n s i t use, did not s i g n i f i c a n t l y d i f f e r among the defined groups. (dj Hartgen's work gave further i n d i c a t i o n s that the c o r r e l a t i o n between the user's socio-economic c h a r a c t e r i s t i c s and h i s a t t i t u d i n a l bias i s not a strong one. He devised an a t t i t u d i n a l bias index which expressed the user's s a t i s f a c t i o n with car r e l a t i v e to t r a n s i t with respect to the modal at t r i b u t e s considered s i g n i f i c a n t by the user himself. By using t h i s index as a measure for the user's a t t i t u d e s , Hartge found that i t exhibited no r e l a t i o n s h i p to the user's income 39 or occupation. Although these findings did not produce d e f i n i t i v e evidence on the s i g n i f i c a n c e of the user's l i f e s t y l e as -a mode choice determinant, they provide us with the assurance that user's s e n s i t i v i t y (or a t t i t u d i n a l b i a s , using Hartgen's terminology), i s not, i n general, correlated with his socio-85 economic c h a r a c t e r i s t i c s , and hence,, the problem of c o l i n e a r i t y i s not expected i n the study's model. 3.14. An Overview of the Mode Choice Model In the previous sections/ we have already i d e n t i f i e d some mode choice determinants and grouped these into three sets of v a r i a b l e s : (a) the user's socio-economic charac-t e r i s t i c s , (b) perceived att r i b u t e s of the transportation system, and (c) the user's s e n s i t i v i t y toward modal a t t r i b u t e s . Another determinant to be added to the above i s the t r i p purpose. This addition i s j u s t i f i e d on the account that the value of time saved for any given t r i p i s dependent, among other things, upon the t r i p purpose, as established i n Section 2.6. A further j u s t i f i c a t i o n i s that the t r i p purpose usually gives an i n d i c a t i o n , although i n d i r e c t , of some of the circum-stances of the t r a v e l a c t i v i t y which may a f f e c t the user's behaviour. For example, work t r i p s are usually associated with s t r e e t congestion, crowded t r a n s i t v e h i c l e s , and so on. On the other hand, no attempt has been made i n thi s study to demonstrate how these determinants are to be manip-ulated as model inputs. This i s discussed i n d e t a i l i n Sections 3.15. and 3.16. But, before t h i s task i s attempted, a b r i e f review of the model structure (which i s demonstrated graphically i n F i g . 4), would be appropriate at t h i s stage of the study. The i n i t i a l step i n the model operation e n t a i l s the " f i l t e r i n g " of raw data to omit a l l captive users from the data OMIT CAPTIVE USERS RAW DATA CHOICE USERS WORK TRIPS PHASE A ' sample selection PHASE B> grouping users by trip purpose B C D PHASE E> predicting users sensitivities toward modal attributes P A S -PHASE C> grouping users by their sensitivities toward modal attributes car transit car transit car transit PHASE D> developing a stochastic model for each group to predict mode choice (car or transit) oo FIG.4 THE MODEL STRUCTURAL RELATIONSHIPS 87 set. Thus, subsequent analysis i s to be conducted on users i n mode choice situations only. A l l users are then grouped according to the t r i p purpose, that i s , work and non-work t r i p s . (If the sample s i z e allows, the l a t t e r may be divided in t o personal business, shopping, s o c i a l and r e c r e a t i o n a l t r i p s ) . For each of these groups, users are c l a s s i f i e d further into sub-groups according to t h e i r s e n s i t i v i t y toward modal at t r i b u t e s . For example, one of these sub-groups would include users who expressed (in a questionnaire survey) t r a v e l cost as the most s i g n i f i c a n t a t t r i b u t e a f f e c t i n g t h e i r choice. Another group would include comfort-sensitive users, and so on. Having created these sub-groups, which are presum-ably homogeneous with respect to the s e n s i t i v i t y v a r i a b l e , the following step would be the c a l i b r a t i o n of a set of stochastic functions (for each sub-group) to explain and pr e d i c t users' mode choices on the basis of some of the socio-economic c h a r a c t e r i s t i c s (income and sex), and attributes of t h e i r i n d i v i d u a l mode choice s i t u a t i o n . Another step i s added to the model structure (Phase E i n F i g . 4 ) , to explain and eventually p r e d i c t the user's s e n s i t i v i t y toward modal a t t r i b u t e s . In this step, the user's age, occupation and time budget are employed as independent variables i n a set of discriminant functions to p r e d i c t such s e n s i t i v i t y . This step i s meant to enhance the p r e d i c t i v e q u a l i t y of the model by providing an account for some aspects of the user's l i f e s t y l e which are thought to influence h i s mode 88 choice behaviour. In as fa r as the operational c h a r a c t e r i s t i c s of the model are concerned, t h i s step can also be considered as an a l t e r n a t i v e to phase C, since i t r e l a t e s the user's s e n s i t i v i t i e s toward modal at t r i b u t e s to some of t h e i r personal c h a r a c t e r i s t i c s . The l a t t e r step may not contribute to the expansion of the explanatory power of the mode choice model i n the s t a t i s t i c a l sense, but i t may improve the r a t i o n a l e of the model i n that i t defines the r e l a t i o n s h i p between the user's c h a r a c t e r i s t i c s and his behaviour i n a c l e a r e r , more under-standable manner. Furthermore, t h i s set of functions may prove us e f u l f o r planning purposes, as mentioned previously i n Section 3 . 2 . , i n that they demonstrate the impact of s o c i a l or economic change, or new trends i n the l i f e s t y l e of the population on t h e i r s e n s i t i v i t y toward modal a t t r i b u t e s , and i n turn, upon t h e i r mode choice behaviour. On the other hand, should t h i s set of discriminant functions f a i l , for any reason, to explain the users' se n s i -t i v i t i e s i n the actual data analysis, t h i s step can be omitted from the model structure without a f f e c t i n g i t s explanatory power i n the s t a t i s t i c a l sense, although the r a t i o n a l e of the model and i t s predictive q u a l i t y might be affected. The purpose of the foregoing discussion i s to b r i e f l y o u t l i n e the various phases of the mode choice model. More d e t a i l e d analysis of these steps i s provided i n the remainder of t h i s chapter, which includes, i n addition, a d e s c r i p t i o n of the s t o c h a s t i c models and an assessment of t h e i r usefulness 89 for planning purposes. 3.15. The Use of Stochastic Models i n  Trans port a t i o n Planning In mode choice analysis, stochastic models are commonly used i n conjunction with discriminant, l o g i t , or probit analysis. In these models, the main concern i s to define the p r o b a b i l i t y that a user would make a c e r t a i n mode choice given his c h a r a c t e r i s t i c s and the c h a r a c t e r i s t i c s and choices of a l l users i n the sample. Usually, these c h a r a c t e r i s t i c s provide a d e s c r i p t i o n for his mode choice s i t u a t i o n , that i s , attributes of modes a v a i l a b l e to him, his socio-economic status, or i n some instances a combination of these two groups of v a r i a b l e s . Mathematically, the stochastic model i s formulated 40 as a c o n d i t i o n a l p r o b a b i l i t y problem. Given the charac-t e r i s t i c s of two users' groups, -and the t o t a l number of users i n each group, the problem here i s to define the p r o b a b i l i t y that a c e r t a i n user, whose c h a r a c t e r i s t i c s are known, belongs to any given group. The Bayesian Theorem provides an answer to t h i s problem i n the following mathematical function, which i s a commonly used formula for the stochastic models: " 1 + e * where p (x) i s the p r o b a b i l i t y of membership i n a given users group, z i s the discriminant score of the observation In discriminant analysis, the discriminant score can be derived from the following mathematical function: n z = a + £ a. X. (2) u i = l 1 1 where a Q and a^ are parameters to be derived i n the c a l i b r a t i o n process, and X^ i s the input variable describing the users' c h a r a c t e r i s t i c s . In the f i r s t run of the model, users' c h a r a c t e r i s t i c s and t h e i r mode choices are employed irt the c a l i b r a t i o n process, so that the values of the c o e f f i c i e n t s a Q / a^ can be derived. One mathematical c r i t e r i o n for deriving the above mentioned function i s to maximize the "variance between" the populations r e l a t i v e to the variance within the populations. The model can be refined by omitting the variables which do not contribute s i g n i f i c a n t l y to the i d e n t i f i c a t i o n of users' groups, or variables which are highly correlated with others. Once the model i s c a l i b r a t e d and r e f i n e d , the researcher can assess i t s explanatory power by examining the appropriate s t a t i s t i c a l parameters, which also enables him to recognize the r e l a t i v e s i g n i f i c a n c e of each of the explanatory v a r i a b l e s . For any new observation—the choice of which .is not known, but i t s c h a r a c t e r i s t i c s X^ are g i v e n — t h e model produces the p r o b a b i l i t y p(x) that t h i s observation belongs to a c e r t a i n users group. Each observation i s to be assigned to the group which the model associates with the highest p r o b a b i l i t y . One measure of the model performance i s the percentage of correct c l a s s i f i c a t i o n i n the i t e r a t i o n procedure. In transportation planning research, the d i s u t i l i t y function was construed as a s p e c i a l case of the discriminant function. In t h i s case the input variables (X^) are confined to the modal at t r i b u t e s , e.g., time, cost, comfort, etc. The discriminant score z can then be interpreted as a measure of the d i s u t i l i t y of t r a v e l by the given mode, with higher score i n d i c a t i n g increasing modal d i s u t i l i t y , and vic e versa. In other models, where the input variables are measures for the att r i b u t e s differences between a l t e r n a t i v e modes, the d i s u t i l i t y score can be interpreted as the d i s u t i l i t y savings i n the generalized p r i c e of t r a v e l mode using one mode as opposed to i t s a l t e r n a t i v e . The d i s u t i l i t y savings concept, r e f e r r e d to i n Section 2.4., was employed by Stopher, Thomas and Thompson, 41 Pratt, Shunk and De Donnea, but t h i s study does not make use of t h i s concept, and employs discriminant analysis i n the conventional manner to i d e n t i f y users' groups on the basis of modal attributes and the users' socio-economic c h a r a c t e r i s t i c s . Hence, the discriminant score z cannot be interpreted here as a d i s u t i l i t y index. 3.16. The Model Structure The s t a t i s t i c a l technique to be employed i n th i s model i s the discriminant analysis as described i n the previous section, except for few modifications to take i n t o account 4 2 multiple users' groups instead of the binary case. The discriminant score can be derived as follows: 92 Z = " Ck + v i k X i k ( 3 ) where z i s the discriminant score i s a parameter to be derived i n the c a l i b r a t i o n procedure, v ^ i s a vector of parameters to be derived i n the c a l i b r a t i o n process f o r v a r i a b l e i and group k. X^ k i s a vector of variables selected by the researcher to d i f f e r e n t i a t e between users' groups. Having derived the discriminant score f o r the user (for each group)^ the p r o b a b i l i t y that such a user belongs to group k i s given by: P k <z £) - , <n£ /rt /2) exp (- \ y\) • • • g k 9 (n k / C r k Si) exp (-§ X 2 ) =1 vz„ k where i s the p r o b a b i l i t y that a user with discriminant score z^ belongs- to.group k, ^ and k are the standard deviations i n users' groups £ and k r e s p e c t i v e l y , n^ , and n^ . are the number of users i n groups I and k respectively, 93 The above formula- i s the general case to be used for a p r i o r i c l a s s i f i c a t i o n of users, which takes into account the s i z e of each group. The preliminary steps of the data analysis i n t h i s study are as follows: (a) grouping users according to t h e i r t r i p purpose; work and non-work t r i p s , (b) within each group, f i v e sub-groups are to be created to accommodate users whose s e n s i t i v i t y toward modal at t r i b u t e s are s i m i l a r ; that i s , users who specify t r a v e l cost, t o t a l t r a v e l time, excess t r a v e l time, frequency of t r a n s i t s e r v i c e , or comfort l e v e l of the t r i p as the most s i g n i f i c a n t 43 a t t r i b u t e should be grouped accordingly. In these two steps, ten sub-groups of users are created f o r . f u r t h e r analysis. For each of these sub-groups, a discriminant function i s to be derived to explain and predict users 1 choices (car or t r a n s i t ) on the basis of the following c h a r a c t e r i s t i c s of the mode choice s i t u a t i o n : X^, the.household income, X^r the user's sex (a dichotomous v a r i a b l e ) , X 0, t o t a l t r a v e l time difference (T - T, - . ) , 3 car t r a n s i t X^, excess t r a v e l time difference, that i s , out of vehicle time, walking, waiting, etc., X^, t r a v e l cost difference, X c, comfort l e v e l difference, and 94 X^, frequency of t r a n s i t s ervice. The use of the cost difference variable and the comfort difference v a r i a b l e s , X^ and Xg, require some elaboration. Travel cost by car, as argued i n Section 3.10., should include only the marginal cost of making the t r i p , i . e . , parking, gasoline and o i l , and road t o l l s i f applicable. For the purposes of t h i s study, a fig u r e of $0.04 per mile can be 44 used as an estimate for the marginal cost of operating a car. This fi g u r e , together with the estimated length of the t r i p and parking charges are to be used to produce an estimate of the car t r i p . Comparing th i s estimate with the t r a n s i t fare w i l l give us the t r a v e l cost d i f f e r e n c e . The comfort a t t r i b u t e i s also to be entered as a continuous v a r i a b l e . I t i s suggested here that the user be requested i n the questionnaire to assess, along a Linkert scale, the comfort l e v e l of both his car and the t r a n s i t v e h i c l e . The difference i n his r a t i n g of the two v e h i c l e s , expressed as abstract units of measurement, can then be entered as X,. b The above mentioned var i a b l e s , together with the user's s e n s i t i v i t y toward model attr i b u t e s are thought to provide an adequate explanation for user's choices. I t i s argued here that the user's s e n s i t i v i t y toward modal attr i b u t e s provides some account for his l i f e s t y l e , at l e a s t i n s o f a r as mode choice behaviour i s concerned. However, such an account would y i e l d l i m i t e d benefits unless a user's s e n s i t i v i t y toward modal a t t r i b u t e can be r a t i o n a l l y and s t a t i s t i c a l l y explained within the model structure. 95 To explain and predict the user's s e n s i t i v i t y toward modal a t t r i b u t e s , i t i s suggested here that other va r i a b l e s , which may also be rel a t e d to his l i f e s t y l e , should be con-sidered. One set of variables i s the user's time budget. Other i n d i c a t i o n s which may be associated with l i f e s t y l e are the user's age and occupation (See Sections 3.5., and 3.6.). By incl u d i n g a dichotomous variable i n d i c a t i n g whether the user i s over or under 40, the influence of age on l i f e s t y l e , (and hence on the user's s e n s i t i v i t y ) , might be accounted f o r . I t might also be desirable to include another dichotomous variable to account for the occupational status of users i n management, professional or executive groups. Thus, an additional set of discriminant functions i s required to explain and pre d i c t the user's s e n s i t i v i t y toward the f i v e system a t t r i b u t e s , X^ to X^, (for each of the two t r i p purpose categories) on the basis of the following variables: • = Y^, the user's age, a dichotomous v a r i a b l e , Y^r the user's occupation, a dichotomous v a r i a b l e , Y^, the number of weekly hours the user puts into work, . •... : Y^, the number of weekly hours spent.in s o c i a l and c u l t u r a l a c t i v i t i e s , Yj., the number of weekly hours spent i n outdoor recreation, and Xg, the number of hours spent at home, with the family, etc. 96 The purpose of t h i s a d d i t i o n a l set i s to improve the r a t i o n a l e of the model and enhance i t s p r e d i c t i v e q u a l i t y , but should these variables f a i l to explain the user's s e n s i t i v i t y toward modal a t t r i b u t e s , t h i s step can be omitted without a f f e c t i n g the s t a t i s t i c a l parameter of the model which r e f l e c t s i t s explanatory power. This precautionary measure f u l f i l l s a requirement set i n Section 3.3., as a r e s u l t of the lack of empirical data to support the v a l i d i t y of t h i s r e l a t i o n s h i p . The model structure outlined above f u l f i l l s the study's basic objectives. However, i f the researcher i s inte r e s t e d i n creating more homogeneous groups, (for which the discriminant, functions to be derived), so that the model y i e l d s a greater explanatory power, several p o s s i b i l i t i e s e x i s t for his con-si d e r a t i o n . One method of achieving t h i s i s by d i v i d i n g non-work t r i p s i n t o shopping., s o c i a l , r e c r e a t i o n a l , personal business and other t r i p s . Furthermore, an improved d e s c r i p t i o n of the user's s e n s i t i v i t i e s may be achieved by accounting f o r the second most important a t t r i b u t e , the t h i r d most important a t t r i b u t e , and so on t i l l a l l attributes are accounted f o r . This method of c l a s s i f i c a t i o n increases the number of possible users' groups up to 120 categories (the f a c t o r i a l of the t o t a l groups number, that i s 5 X 4 X 3 X 2 X 1 = 120) for each t r i p purpose. Obviously, such a detailed c l a s s i f i c a t i o n would make , the research task unmanageable, and the researcher i s required to exercise his judgement as to the most appropriate c l a s s i f i -cation to serve the research purposes, taking i n t o consider-ation the t o t a l sample siz e as well as the minimum s i z e of the 97 sub-groups to be created. 3.17. Summary The design of the study's model i s aimed, through data analysis and " f i l t e r i n g , " at i s o l a t i n g choice users, and d i v i d i n g them into sub-groups each of which r e f l e c t s a c e r t a i n degree of homogeneity with respect to some of the i n f l u e n t i a l factors a f f e c t i n g t h e i r choice behaviour. The c r i t e r i a employed for c l a s s i f i c a t i o n are: (a) t r i p purpose, and (b) the users' s e n s i t i v i t i e s toward modal a t t r i b u t e s . This procedure allows for the model to account for c e r t a i n aspects of the users' l i f e s t y l e which may a f f e c t t h e i r choice behaviour. In addition, i t ensures that further analysis i s conducted on samples of users which are homogeneous i n c e r t a i n respects. For each of the user's sub-groups defined, a stoch a s t i c p r o b a b i l i s t i c function i s to be derived to explain and p r e d i c t the users' choices: car or t r a n s i t . The dis c r i m i n a t i o n i s to be based upon two sets of va r i a b l e s : (a) c h a r a c t e r i s t i c s of the mode choice s i t u a t i o n , that i s , differences between the car and t r a n s i t with respect to t r a v e l cost, t o t a l t r a v e l time, excess t r a v e l time, the comfort l e v e l , and the frequency of t r a n s i t service, and. (b) some of the user's socio-economic c h a r a c t e r i s t i c s , age and sex. These inputs are believed to be adequate to explain the users' mode choices. Yet, for the model to be of p r e d i c t i v e value, the re l a t i o n s h i p between these variables and future 98 changes i n the s o c i a l , economic, or demographic c h a r a c t e r i s t i c s of the population should be i d e n t i f i e d . Thus, f o r any given region, changes which may a f f e c t the users' behaviour can be i d e n t i f i e d as follows: . (a) changes i n the c h a r a c t e r i s t i c s of the mode choice s i t u a t i o n : cost of t r a v e l (increasing cost of gasoline, parking), t r a v e l time (street congestion, new t r a n s i t tech-nology) , comfort l e v e l , etc. . . . (b) changes i n the socio-economic c h a r a c t e r i s t i c s of the population, age structure, income d i s t r i b u t i o n , demand for various occupations, etc. (c) changes i n the population l i f e s t y l e , higher l i v i n g standards, a v a i l a b i l i t y of l e i s u r e time, new work or l e i s u r e e t h i c s , etc. These, changes may or may not be p r e d i c t a b l e . However, i n as f a r as the trends can be i d e n t i f i e d , and the researcher i s w i l l i n g to attach a degree of c r e d i b i l i t y to these pre-d i c t i o n s , t h e i r impact on the users' expected behaviour would also be i d e n t i f i a b l e . The study's model i s formulated so that the users' socio-economic c h a r a c t e r i s t i c s and his time budget are r e l a t e d to his s e n s i t i v i t y toward modal a t t r i b u t e s . This r e l a t i o n s h i p i s expressed i n another set of discriminant functions to explain and predict the user's s e n s i t i v i t y toward modal a t t r i b u t e s on the basis: of his age, occupation and time budget. This additional set obviously does not add to the explanatory power of the model i n a s t r i c t l y s t a t i s t i c a l sense, but i t enhances i t s rationale as well as i t s p r e d i c t i v e and 99 p r a c t i c a l value. The model inputs so far seem to cover many of the changes i n the population socio-economic c h a r a c t e r i s t i c s , and some i n d i c a t i o n of t h e i r l i f e s t y l e s . The researcher may be w i l l i n g to p r e d i c t that such changes w i l l take place, and. the r e s u l t of his analysis would be to define t h e i r impact on the users' mode choices. A l t e r n a t i v e l y , he may be merely seeking information on what the impact of these changes would be on the users' behaviour, regardless of whether the changes i n question are expected with any degree of c e r t a i n t y . 100 FOOTNOTES 1. Such a s e n s i t i v i t y can be i d e n t i f i e d i n the person's statement i n a questionnaire survey for example, or a l t e r n a t i v e l y , i n his mode choice behaviour. The s e n s i t i v i t y of users toward excess t r a v e l time was i d e n t i f i e d by the r e l a t i v e l y high d i s u t i l i t y derived from users'behaviour (see Sec. 2.6). 2. The notion that some users are "pressed" f o r time, and that t h i s may have an a f f e c t on t h e i r mode choice i s re f e r r e d to by De Donnea, Determinants of  Transport Mode Choice, p. 157, and Brown, Mode Choice  Determinants of Selected Socio-economic Groups, p. 103. 3. Moses and Williamson, "The Subsidy Issue," Journal of  P o l i t i c a l Economy," pp. 247-264. 4. We should note that, although the users' socio-economic c h a r a c t e r i s t i c s and t h e i r s e n s i t i v i t y toward modal a t t r i b u t e s are not, i n general, strongly correlated, i t i s might possible, however, that some users groups, enjoying, say, a high occupational status (e.g. managers, professionals) would express similar s e n s i t i v i t i e s toward modal a t t r i b u t e s . But t h i s i s only because such groups may have s i m i l i a r l i f e s t y l e s . . Likewise, age may not be related to the user's s e n s i t i v i t y toward modal a t t r i b u t e s , except to the extent that i t indicates a trend i n l i f e s t y l e . 5. De Donnea, Determinants of Transport Mode Choice, p. 130. 6. Ibid., p. 193. 7. Brown, Mode Choice Determinants of Selected Socio- economic Groups, p. 101. 8. Ibid., p. 223. 9. Ibid., p. 103. 10. Ibid., p. 222, De Donnea, Determinants of Transport  Mode Choice, p. 137, Warner, Stochastic Choice of Mode, p. 35, Lave, "Behavioural Approach," Transportation  Research, p. 472. 11. Brown, i n i b i d . , pp. 123-127. 12. Warner, Stochastic Choice of Mode, p. 85. 101 Brown, Mode Choice Determinants of S e l e c t e d S o c i o - economic Groups, p. 101. De Donnea, Determinants of T r a n s p o r t Mode C h o i c e , p. 137. Lave, " B e h a v i o u r a l Approach," T r a n s p o r t a t i o n Research, p. 471. Brown, Mode Choice Determinants of S e l e c t e d S o c i o - economic -Groups, p. 101. Warner, S t o c h a s t i c Choice of  Mode, p. 35, Lave i n I b i d . , p. 47 2. Hartgen and Tanner, " E f f e c t of T r a v e l e r A t t i t u d e s , " Highway Research Record., p. 5. I b i d . I b i d . I b i d . i Thomas, "Value of Time," Highway.Research Record, p. 24, and Peter. L. Watson, "Problems A s s o c i a t e d w i t h Time and Cost Data Used i n T r a v e l C h o i c e Modeling and V a l u a t i o n of Time," Highway Research Record, Rep. No. 369, 1971, pp. 148-158. T h i s approach was pursued by De Donnea, Determinants  o f Mode Choice, pp. 7 7-8 8, and Quarmby, "Choice of T r a v e l Mode," J o u r n a l of T r a n s p o r t Economics and P o l i c y , pp. 280-281. . Warner, S t o c h a s t i c C h o i c e of Mode, p. 35. Quarmby, "Choice of T r a v e l Mode," J o u r n a l o f Transport.  Economics and P o l i c y , p. 297. I f the r e s e a r c h e r concluded from t e s t i n g approaches a & b i n t h i s s e c t i o n t h a t t h e r a t i o y i e l d s the model a g r e a t e r e x p l a n a t o r y power, he may wish to use t h i s approach. T h i s procedure i s a l s o v a l i d f o r a l l a t t r i b u t e s , y e t a c e r t a i n degree o f c o n s i s t e n c y i n t r e a t i n g these a t t i t u d e s ( e i t h e r by a c c o u n t i n g f o r the d i f f e r e n c e or the r a t i o ) may a l s o be d e s i r a b l e . Lave, "A B e h a v i o u r a l Approach," T r a n s p o r t a t i o n  Research, pp. 467-468. Quarmby, "Choice of T r a v e l Mode," J o u r n a l of T r a n s p o r t  Economics and P o l i c y , p. 290. 10 2 28. Warner, Stochastic Choice of Modes, p. 3 5. 29. Brown, Mode Choice Determinants of Selected Socio-economic Groups, p. 105, Hartgen and Tanner, " E f f e c t ot Traveler Attitudes," Highway Research Record, p. 5, and Lisco, "Value of Travel Time," Highway  Research Record, p. 36. 30. The Linkert scale, as used i n t h i s study, i s a graduated f i v e points scale, which allows the user to express his subjective evaluation of any given a t t r i b u t e s . For example, the user may be asked to state how he evaluates the comfort l e v e l of the t r i p (a) very comfortable . . . ., (b) comfrotable, . . . ., (c) somewhat comfortable . . . ., (d) uncomfortable . . . ., or (e) very uncomfortable . . . . 31. De Donnea, Determinants of Transport Mode Choice, p. 131, and Brown i n i b i d . , p. 101. 32. Brown, Mode Choice Determinants of Selected Socio-economic Groups, p. 101. 33. Ibid., P- 207 and 213. 34. Ibid., P- 207. 35. Ibid., P- 213. 36. Ibid. , P- 216. 37. Ibid., P.- 200. 38. Beier, Marketing T r a n s i t , T r a f f i c Quarterly, pp. 538 39. David T. Hartgen, The Influence of A t t i t u d i n a l and Si t u a t i o n a l Variables on Urban Mode Choice, Department of Transportation, New York State, Rep. No. PRR. 41, March, 1973, p. 40. 40. For de t a i l e d discussion of the mathematical formulation of t h i s model, see M u l t i v a r i a t e Data Analysis, by William W. Cooley and Paul R. Lohnes, John Wiley, New York, 1971, pp. 243-286, and Stochastic Choice of  Mode, by Warner, pp. 5-16. 103 Stopher, " P r o b a b i l i t y Model," Highway Research Record, Thomas and Thompson, "The V a l u e of Time," Highway  Research Record, Rep. No. 314, 1970, P r a t t , " U t i l i t a r i a n Theory," Highway Research Record, Shunk and Bouchard, " A p p l i c a t i o n of M a r g i n a l U t i l i t y , " Highway Research Record, and De Donnea, Determinants  of T r a n s p o r t Mode. G e r a l d R. Brown, " P o l i c y Model f o r M u l t i m o d a l T r a n s - . p o r t a t i o n System P l a n n i n g , " T r a n s p o r t a t i o n Research S e r i e s , Rep. No. 2, The U n i v e r s i t y of B r i t i s h Columbia, A p r i l , 1973, pp. 2-6. T h i s i d e a should be a t t r i b u t e d t o R u s s e l L. A c k o f f , " I n d i v i d u a l P r e f e r e n c e s f o r V a r i o u s Means of T r a n s p o r t a t i o n , " I n t e r i m Report of Management S c i e n c e C e n t r e , U n i v e r s i t y of. P e n n s y l v a n n i a , May 1965, p. 13. T h i s f i g u r e was estimated on the b a s i s of g a s o l i n e consumption a t the r a t e of 13 m i l e s per g a l l o n , f i f t y cents per g a l l o n . Perhaps a more a p p r o p r i a t e . .. approach i s to experiment w i t h s i m i l a r v a l u e s of c a r o p e r a t i n g c o s t and adopt the v a l u e which y i e l d s the model a g r e a t e r e x p l a n a t o r y . T h i s i s , i n essence, Quarmby's approach. CHAPTER IV SURVEY AND QUESTIONNAIRE DESIGN 4.1. Introduction This chapter i s devoted to problems associated with sample s e l e c t i o n and questionnaire design. This i s to be the f i n a l part of the study's course of i n v e s t i g a t i o n , since no attempt i s made to c o l l e c t and analyse raw data on users' behaviour. Evaluation of the study's progress and i t s pertinence to•planning i s expounded i n the following chapter. 4.2. Survey Design The discussion put forward i n Section 3.16., concerning the data requirements for the model c a l i b r a t i o n procedure c l e a r l y indicates that such information can be drawn from only a sample of users whose behaviour i s monitored i n choice s i t u a t i o n s . Information on captive users i s not useful for the model c a l i b r a t i o n and should be excluded from the data analysis p r i o r to t h i s stage. To s a t i s f y this condition, i t i s suggested here that the questionnaire survey be conducted i n transportation corridors where car use i s p r a c t i c a l and t r a n s i t service i s av a i l a b l e . In the meantime, i t i s also d e s i r a b l e to extend the sample space so as to include modes choice s i t u a t i o n s which shows a wide range of v a r i a t i o n i n the d i s u t i l i t y savings 104 105 between car and t r a n s i t . Thus, the survey would not include a random sample of users; rather, the attempt i s made to c o l l e c t s u f f i c i e n t information s p e c i f i c a l l y for the purpose of model calibration." 1" I t follows that the data should include the average as well as the "extreme" mode choice s i t u a t i o n s , i . e . , varying degrees of "competition" between car and t r a n s i t . Thus, the sample space to'be defined for the study purposes should s a t i s f y c e r t a i n c r i t e r i a : (a) Total time dif f e r e n c e . The sample should include choice si t u a t i o n s with time savings between car and t r a n s i t use varying from small to large, and should i d e a l l y r e f l e c t the advantage of car i n some cases and the advantage of t r a n s i t i n others. I t i s very l i k e l y that i f car use has an advantage over t r a n s i t use with respect to time savings, such an advantage would increase with the distance t r a v e l l e d . Thus, by accounting for choice si t u a t i o n s at d i f f e r e n t locations along the transportation corridor, the sample would include choices made under varying degrees.of competition between car and transit.with respect to t h i s p a r t i c u l a r a t t r i b u t e . On the other hand, choice s i t u a t i o n s where t r a n s i t i s more advantageous with respect to t o t a l t r a v e l time are l i k e l y to be somewhat l i m i t e d . Such si t u a t i o n s may be encountered i n c i t i e s where t r a n s i t operates on i t s own right-of-way, while other vehicles operate on congested s t r e e t s . Otherwise, the use of t r a n s i t v e h i c l e s , such as buses, may only be advantageous (with respect to the time attribute) i n " i n t e r n a l " downtown f r i p s , i . e . , t r i p s whose 106 o r i g i n and destination are i n the downtown area, since walking to and from the parking space at both the t r i p ends are l i k e l y to be inproportionate r e l a t i v e to the i n - v e h i c l e time. (b) Travel cost difference. The sample should also include mode choices with a wide range of cost d i f f e r e n c e s . . But since t r a v e l cost by t r a n s i t does not vary s i g n i f i c a n t l y (regular bus fare, monthly pass f a r e ) , the required range can only be created by v a r i a t i o n i n the t r a v e l cost by car: operating cost, parking charges, and t o l l roads i f applicable. Operating cost, i t should be noted, may not always be s i g n i f i -cant ($0.04 per mile), and t o l l roads are not common i n thi s country. Thus, to include choices with s u b s t a n t i a l cost differences, the sample should i d e a l l y include a number of car users who are carrying commercial parking charges, and others who are provided with free or subsidized parking (employee parking). (c) Differences i n level s of t r a n s i t s ervice. Choices involving t r a n s i t use should also account for d i f f e r e n t l e v e l s of service. Generally speaking, the l e v e l of t r a n s i t service i s a function of two a t t r i b u t e s : frequency of departure times at various hours of the day, and the "density" of t r a n s i t l i n e s i n the c i t y sector, i . e . , the length of routes divided by the area serviced. The former accounts for the f l e x i b i l i t y the user may have i n adjusting his departure time with other d a i l y a c t i v i t i e s , while the l a t t e r expresses the convenience of t r a n s i t use. To account for this v a r i a t i o n , the sample should include users i n areas which are provided with lev e l s 107 of t r a n s i t service varying from poor to good with respect to these two a t t r i b u t e s . 4.3. D e f i n i t i o n of the Sample Space The question to be addressed i n t h i s section i s how to define the sample space so as to meet the c r i t e r i a established above. I t i s suggested here that a delibe r a t e e f f o r t be made so that the study sample would account for a var i e t y of mode choice s i t u a t i o n s . This objective can be achieved by several a l t e r n a t i v e procedures, one of which i s expounded below. As a f i r s t step, the researcher may s e l e c t several r e s i d e n t i a l areas at d i f f e r e n t locations i n the c i t y which r e f l e c t a substantial v a r i a t i o n i n the q u a l i t y of t r a n s i t service provided (e.g., frequency of departure times, route mile per serviced area) at d i f f e r e n t distances from the downtown area. In s e l e c t i n g these areas, i t may be desirable to assess t h e i r s u i t a b i l i t y for the analysis intended with respect to the socio-economic c h a r a c t e r i s t i c s , and possibly other c h a r a c t e r i s t i c s which may seem relevant, e.g., r e s i d e n t i a density. Having i d e n t i f i e d the r e s i d e n t i a l areas to be included i n the analysis, the researcher may want to consider p a r t i t i o n -ing the sample into three sub-spaces which correspond i n i t i a l l y to the number of questionnaires to be d i s t r i b u t e d i n various parts of the c i t y . These sub-samples, to be named here A, B, and C, can be described as follows: 108 (a) sample space A, which i s to include car and t r a n s i t users t r a v e l l i n g from these areas to downtown, (b) sample space B, to be devoted to observations concerning users s t a r t i n g t h e i r t r i p from downtown to the selected r e s i d e n t i a l areas, (c) sample space C, to include downtown " i n t e r n a l " t r i p s , i . e . , t r i p s originated and ended i n the downtown area. The method of achieving t h i s p a r t i t i o n i n g i n the survey and the purpose served i n each case i s discussed below. The survey proposed here i s designed as a mail questionnaire survey. Thus, to " f i l l - i n " sample space A, the questionnaire forms are to be d i s t r i b u t e d to users s t a r t i n g t h e i r t r i p s i n the selected r e s i d e n t i a l areas and are destined to downtown. The t r a n s i t users' questionnaire forms are to be handed to users as they step into the t r a n s i t v e h i c l e , i f they answer a f f i r m a t i v e l y to the question whether they are headed to the downtown area. (It may be advisable to avoid the d i s t r i b u t i o n of these forms at major transfer points, so that the sample be confined to residents of the areas selected f o r the study). Likewise, the car users' questionnaires are to be d i s t r i b u t e d to car drivers i n the area who s a t i s f y the same condition (again, avoiding major a r t e r i a l s so that through t r a f f i c i s excluded). To formulate sample space B, a number of bus users' questionnaires are to be d i s t r i b u t e d i n the downtown area to users destined to the selected r e s i d e n t i a l areas i n the same manner as above. However, the d i s t r i b u t i o n of car users' 109 questionnaires i n t h i s sample space i s d i f f e r e n t i n that a number of these are to be d i s t r i b u t e d i n commercial and employee parking l o t s , so that the sample space would include users carrying parking charges, as well as others who are provided with free or subsidized parking. Sample space C, as mentioned previously, i s created for i n t e r n a l t r i p s i n the downtown areas by car. and t r a n s i t . The questionnaire i s to be handed to users s t a r t i n g and terminating t h e i r t r i p s i n the downtown area. The purpose of creating t h i s space i s to ensure the sample also accounts for time savings which may be achieved by t r a n s i t use. The r a t i o n a l e for p a r t i t i o n i n g the sample space into three d i v i s i o n s i s now becoming c l e a r : for sample space A, i t i s to ensure that a s u f f i c i e n t v a r i a t i o n i n t r a v e l time and l e v e l of t r a n s i t service have been included i n the data base; for example space B, i t i s to ensure that a wide range of cost v a r i a t i o n i s accounted f o r . Sample space C i s devised to include a wide range of time v a r i a t i o n by accounting for "negative" time savings to be made by car use i n some mode choice s i t u a t i o n s . Although there are s u f f i c i e n t grounds to suggest the survey design as discussed above, i t should be remembered that there i s a p o s s i b i l i t y that the s e l e c t i o n of a biased sample may not prove useful as anticipated i n t h i s study. Taking these extreme mode choice s i t u a t i o n s i n t o account may possibly weaken the explanatory power of the model, i f the extreme cases a c t u a l l y r e f l e c t d i f f e r e n t kinds of trade-offs 110 from the bulk of the sample. Should t h i s be the case, the r e s e a r c h e r may wish to exclude such extreme cases, so t h a t the data a n a l y s i s would be c o n f i n e d t o users whose behaviour i s more homogeneous. The r e s e a r c h e r may a l s o wish to c o n s i d e r o t h e r m o d i f i -c a t i o n s i n the r e s e a r c h d e s i g n . For example, a random sample, (r a t h e r than a b i a s e d sample), c o u l d be more u s e f u l i f the r e s e a r c h e r i s i n t e r e s t e d i n a n a l y s i n g the behaviour o f u s e r s i n a geographic area. Another m o d i f i c a t i o n i s p o s s i b l e i f the r e s e a r c h i s conducted i n a c i t y where t r a n s i t v e h i c l e s operate on a separate r i g h t - o f - w a y , and thus, t r a n s i t use would y i e l d time-savings even i n l o n g e r t r i p s , e.g., home-to-work t r i p s . In t h i s case, sample space C would not be r e q u i r e d . In o t h e r s i t u a t i o n s , i t may be d e s i r a b l e to c o n s i d e r sample spaces A and B as mutu a l l y e x c l u s i v e . That i s , the s e l e c t e d r e s i d e n t i a l areas w i l l be d i v i d e d i n t o two groups, a number o f q u e s t i o n n a i r e forms w i l l be d i s t r i b u t e d t o users i n the f i r s t group to form sample space A, and another number of q u e s t i o n n a i r e s w i l l be d i s t r i b u t e d to users i n the downtown area d e s t i n e d to the second group of r e s i d e n t i a l areas t o form sample space B. The purpose o f t h i s d i s t i n c t i o n i s to a v o i d the p o s s i b i l i t y t h a t the same user may be g i v e n the q u e s t i o n n a i r e form twice, once i n h i s r e s i d e n t i a l area, and a second time i n the downtown on h i s way home. The r e s e a r c h e r should e x e r c i s e h i s judgement as t o the number of q u e s t i o n n a i r e forms to be d i s t r i b u t e d i n each of the sub-samples mentioned above, the s p e c i f i c l o c a t i o n and time o f I l l the d i s t r i b u t i o n . I t should be remembered that according to the model designed for the study purposes, ten stochastic functions are to be formulated (See Sections 3.14., and 3.16.). Since each of these would require a minimum of eighty observations for the c a l i b r a t i o n procedure, i t may be necessary to d i s t r i b u t e some 4000 questionnaire forms so that the researcher may have a reasonable chance to receive at l e a s t the minimum number of observations required f o r the c a l i b r a t i o n of each function. 4.4. The Questionnaire Design Two questionnaire forms are designed for the study's survey, one for car and the other for t r a n s i t users. The questionnaire text i s put forward i n Appendices A and B. The questionnaire should, i f answered f u l l y , provide the researcher with a l l the information required for the model formulation phase as outlined i n Sections 3.14., and 3.16. The t r a n s i t users questionnaire i s phrased s p e c i f i c a l l y for bus use. I t provides the following information: (a) o r i g i n , destination, and s t a r t i n g time of the t r i p . (b) t r i p purpose. (c) t r a v e l time by the chosen mode, as perceived by the user, broken down to walking, waiting and i n - v e h i c l e time. (d) frequency of t r a n s i t service at the t r i p time. (e) t r i p cost. 112 (f) the user's assessment of the r e l a t i v e s i g n i f i c a n c e of the following a t t r i b u t e s : - t r a v e l cost, - t o t a l t r a v e l time, - excess t r a v e l time, - frequency of bus service, and - comfort l e v e l of the bus t r i p . (g) whether the user could have t r a v e l l e d by the car instead, and i f so, the following data are requested: - walking and in-vehicle time, - parking charges i f any. (h) the user's socio-economic c h a r a c t e r i s t i c s : age, sex, income l e v e l , occupation, and number of persons i n his household. (i) the user's approximate time budget, i . e . , a l l o c a t i o n of his time among f i v e basic categories: work, at home or with family, c u l t u r a l or s o c i a l engagements, and out-door re c r e a t i o n . (j) whether the user can, at his w i l l , work overtime. The use of these data as inputs into the study's model i s -discussed i n d e t a i l i n the previous chapter, except for the i n c l u s i o n of some addit i o n a l information which the researcher may wish to explore i t s value for the a n a l y s i s : (a) non-work t r i p s are c l a s s i f i e d further into f i v e categories, (b) assessment of the system a t t r i b u t e s i s extended to f i v e important att r i b u t e s (instead of being concerned with 113 the most important a t t r i b u t e only), and (c) the user's a b i l i t y to work overtime at w i l l , (d) the number of persons i n the household. The researcher may wish to a l t e r h i s research design to account for any of these va r i a b l e s , should his investigations indicate i t s s i g n i f i c a n c e . The car user's questionnaire provides s i m i l a r data to those itemized above, and, i n addition, information on whether the user i s motivated to use the car because: (a) he needed the car at the end of the t r i p to make other t r i p s , (b) he has to take along another passenger, (c) he i s sharing a car pool with others, (d) he needs the package space i n the car for .shopping bags, working tools, etc., (e) using the bus i s a f f e c t i n g his s o c i a l status. This information, together with other data requested i n item 5 i n the questionnaire (Appendix B), allow for " f i l t e r i n g " the sample p r i o r to the c a l i b r a t i o n procedure. Those1 haying no access to an a l t e r n a t i v e mode, need the car at the end of the. t r i p , or use the car for the benefits of someone else are p r a c t i c a l l y captive users and should be omitted from the analysis. Likewise, users who opted for the car because they are sharing a car pool (door-to-door service?), need the package space i n the car, or using the car as a status symbol should also be excluded from the data base. Such users are influenced by attr i b u t e s which are not 114 c h a r a c t e r i s t i c to t r a n s i t , and they are also, i n a sense, captive users. As a r e s u l t of t h i s f i l t e r i n g process, only users who have a r e a l choice and are affected by a t t r i b u t e s common to the two modes ava i l a b l e may be accounted f o r . A s i m i l a r problem i s raised by users who q u a l i f y as choice users (according to the c r i t e r i a mentioned above), but are not aware of the attri b u t e s of the a l t e r n a t i v e mode. For example, a user who has a f u l l access to t r a n s i t use and has no compelling reason to use the car, may s t i l l choose i t for tr a v e l without acquiring any information on the at t r i b u t e s of the t r a n s i t system. The problem-raised here i s two-fold. F i r s t , such a user made his choice on the basis of incomplete information on his mode choice s i t u a t i o n , and hence, the constraints a f f e c t i n g his behaviour would be d i f f e r e n t from other "informed" users. Secondly, for the purposes of model formulation, a t t r i b u t e s of the transportation modes av a i l a b l e to the user are necessary information without which the d i s -criminant functions cannot, be derived. The l a t t e r d i f f i c u l t y can be resolved, i f the researcher chooses, by estimating the attributes of the t r a v e l a c t i v i t i e s by the al t e r n a t i v e mode for such users, and feeding these into the c a l i b r a t i o n procedure. The information on the user's o r i g i n , destination, t r i p time, and attri b u t e s of the e x i s t i n g transportation system can be u t i l i z e d to provide these -estimates. However, one may suspect that the behaviour of "uninformed" users could be d i f f e r e n t i n some respects from 115 other users. For example, modal bias may be more pronounced i n the behaviour of the former group. In handling t h i s d i f f i c u l t y , the researcher has two a l t e r n a t i v e s to follow: (a) to drop such observations altogether from the data base, 2 thus confining his analysis to informed, choice users, or (b) to include such observations (complemented by the estimated attributes) subject to further analysis to ensure that t h e i r i n c l u s i o n w i l l not reduce the explanatory power of the model. The researcher may want to arrange for two runs f o r the model, with and without observations on the."uninformed" users. I t can then be determined whether t h e i r i n c l u s i o n would be to the deteriment of the model performance.. In any case, the researcher would be well advised to conduct a further analysis on t h i s group of users to i d e n t i f y any differences which may be of relevance to the inquiry on users 1 behaviour. 116 FOOTNOTES 1. The problem of incomplete coverage of data to a s u f f i c i e n t range of mode choice s i t u a t i o n can be i d e n t i f i e d i n the following works: Beesley, "The Value of Time, Economica, Shunk, "Application of Marginal U t i l i t y , "Highway Research Record, and De Donnea, Determinants of Transport Mode Choice. Only Thomas and Thompson seemed to have had a s u f f i c i e n t data base, "Value of Time," Highway  Research Record, 1971. 2. If the researcher finds that a sub s t a n t i a l proportion of choice users are not aware of the a t t r i b u t e s of the t r a n s i t system, he may i n f e r from t h i s observation that either the t r a n s i t system i s providing a poor l e v e l of service, and thus very few consider i t a choice, or the information on the t r a n s i t system i s not r e a d i l y accessible to the pu b l i c , i n which case he may wish to i n i t i a t e an advertising campaign to remedy the s i t u a t i o n . CHAPTER V SUMMARY AND CONCLUSIONS 5.1. Introduction This chapter/ being the l a s t , i s a summary and a c r i t i c a l analysis of the study's course of in v e s t i g a t i o n s . The study approach and conclusions are appraised, and i t s pertinence to planning i s discussed. F i n a l l y , the study's l i m i t a t i o n s are expounded, and as a r e s u l t of acknowledging these, the attempt i s made to out l i n e further research p o s s i b i l i t i e s . 5.2. Summary and Evaluation of the Study The dual concern of t h i s study i s with the i d e n t i f i -cation of mode choice determinants, and the formulation of a model which promises greater explanatory c a p a b i l i t i e s than what i s afforded by the current state-of-the-art i n t h i s f i e l d . This required an evaluation of the e x i s t i n g theories and concepts pertinent to mode choice behaviour, namely: the abstract mode choice theory, the u t i l i t a r i a n theory of mode choice, theory of time value, and concepts developed oh perception, a t t i t u d e s , and behaviour. In summary, the t h e o r e t i c a l framework which can be established from the l i t e r a t u r e review and empirical research findings defines the p r o b a b i l i t y of making a c e r t a i n mode 117 118 choice as a function of two sets of var i a b l e s : differences between modal at t r i b u t e s of a l t e r n a t i v e choices, and the users' socio-economic c h a r a c t e r i s t i c s . The concept of the generalized p r i c e of the t r a v e l a c t i v i t y i d e n t i f i e s the user's perception also as a mode choice determinant. The generalized p r i c e , as employed i n thi s study, i s a combined measure for modal attr i b u t e s and the user's perception of these. I t i s implied i n t h i s concept that the d i s u t i l i t y of various t r a v e l a t t r i b u t e s are comparable i n quantitative terms. Supportive evidence to substantiate this concept was encountered i n the l i t e r a t u r e . The conceptual framework established f o r the study's purposes makes use of the concept of the i n d i v i d u a l ' s l i f e s t y l e , the l a t t e r being i n i t i a l l y defined as the pattern by which the i n d i v i d u a l allocates his time and monetary resources. Further analysis leads to ce r t a i n modifications i n thi s concept: i n as f a r as mode choice behaviour i s concerned, the i n d i v i d u a l ' s l i f e s t y l e i s l i k e l y to be expressed i n his s e n s i t i v i t y to modal a t t r i b u t e s , which i n turn, can be traced to a v a i l a b i l i t y and the demand upon his time and monetary resources. This concept can be useful i n formulating the operational procedures of the model as follows: - users can i n i t i a l l y be c l a s s i f i e d into sub-groups on the basis of the emphasis they place on various system a t t r i b u t e s , - for each of these sub-groups, a stochastic model i s to be formulated to explain users' choices on the basis of 119 t h e i r socio-economic c h a r a c t e r i s t i c s and c h a r a c t e r i s t i c s of the mode choice s i t u a t i o n . - another set of p r o b a b i l i s t i c functions i s to be formulated as an " o f f - l i n e " procedure i n the operation of the model. That i s , the model can be operational without this set although i t s potential uses and value would be r e s t r i c t e d . The purpose of this a d d i t i o n a l set i s to explain the user's s e n s i t i v i t y toward modal at t r i b u t e s on the basis of his time budget and some related socio-economic character-i s t i c s . This step i s introduced merely to improve the rationale of the model and enhance i t s p r e d i c t i v e q u a l i t y rather than i t s explanatory function i n a s t r i c t l y s t a t i s t i c a l sense. The model i s to be cal i b r a t e d from information on users' behaviour i n actual mode choice s i t u a t i o n s . Users are asked to provide information on the mode selected, charac-t e r i s t i c s of the choice s i t u a t i o n a v a i l a b l e to them, t h e i r s e n s i t i v i t y to modal at t r i b u t e s , t h e i r socio-economic char-a c t e r i s t i c s , and an approximate account for t h e i r time budget. The model i s , therefore, a behavioural one, i t s basic function being to explain and predict users' choices. This approach i s d i f f e r e n t from the approach pursued i n the propensity models where the user i s asked to specify the conditions under which he would s h i f t modes.''' In the l a t t e r case, i t can be re a d i l y argued that the pr e d i c t i o n 120 made i s v a l i d only.to the extent that users act i n the manner they previously s p e c i f i e d as t h e i r preference, or "would-be" behaviour. Yet, the r e l a t i o n s h i p between the users' prefer-ences and t h e i r actual behaviour has never been established i n the previous research i n t h i s f i e l d , and the issue i s s t i l l considered, at best, a thorny one. This problem i s avoided here i n i t s ent i r e t y by pursuing the behavioural approach. In evaluating the model s t r u c t u r a l r e l a t i o n s h i p s , i t may be noted that some of the concepts employed seem to be. supported on a p r i o r i grounds, while others were empi r i c a l l y v e r i f i e d i n previous research. The primary hypothesis of the study, that users' s e n s i t i v i t i e s toward modal a t t r i b u t e s are not strongly correlated with t h e i r socio-economic character-i s t i c s , i s supported by evidence i n the l i t e r a t u r e which are considered by thi s author to be adequate, and hence, a strong case can be made to account f o r both sets of determinants i n model choice an a l y s i s . On the other hand, no substantial evidence was encountered to suggest that such s e n s i t i v i t i e s are caused by or related to the user's time budget. This p l a u s i b l e r e l a t i o n s h i p remains unproven hypothesis. - (Indeed, this subject i n i t s e n t i r e t y i s s t i l l unexplored). Other elements of uncertainty i n the a p p l i c a t i o n of the time budget concept i s whether the d e s c r i p t i o n provided for various a c t i v i t i e s (works, family, s o c i a l , r e c r e a t i o n a l ) , w i l l prove useful and reveal s i m i l a r r e l a t i o n s h i p s to those discussed i n Section 3.4. However, i n addressing the l a t t e r 121 two c r i t i c i s m s , i t should be pointed out that the use of the time budget concept i n the model i s confined to the " o f f - l i n e " procedure which i s added to the basic structure of the model (phase E i n Section 3.14., and F i g . 5) to explain and p r e d i c t user's s e n s i t i v i t i e s toward modal a t t r i b u t e s . Should the time budget scheme prove unsuccessful i n the model implementation, the explanatory function of the model would not be affected (at l e a s t not i n the s t a t i s t i c a l sense), although i t s p r e d i c t i v e q u a l i t y would be. The researcher may then wish to seek, a d i f f e r e n t explanation for the users' s e n s i t i v i t i e s toward modal a t t r i b u t e s , possibly i n conjunction with a d i f f e r e n t approach to the i n d i v i d u a l ' s l i f e s t y l e . Another aspect of the model operation which should be viewed cautiously i s the user's assessment of the r e l a t i v e s i g n i f i c a n c e of modal a t t r i b u t e s . One may advance a p r i o r i argument that a user may consider, f o r example, the comfort a t t r i b u t e to be the most s i g n i f i c a n t for a given t r i p ; but t h i s i s only true as long as the t r i p cost or time does not exceed a c e r t a i n l i m i t (beyond which cost or time becomes the most s i g n i f i c a n t a t t r i b u t e ) . Thus, s u b s t a n t i a l change i n the system a t t r i b u t e may lead to d i f f e r e n t s e n s i t i v i t i e s among users. This poses further l i m i t a t i o n s on the use of the model for p r e d i c t i v e purposes. F i n a l l y , there are other shortcomings inherent to the t h e o r e t i c a l framework of the study which have been hinted at previously. The application of the abstract mode choice 122 theory has p r a c t i c a l l i m i t a t i o n s i n that no systematic or objective method i s developed to i d e n t i f y the "relevant" system a t t r i b u t e s . Should the researcher apply h i s judgement to s e l e c t these a t t r i b u t e s , or the universe from which these attributes are to be s e l e c t e d — a step which he cannot escape— he i s under no conditions c e r t a i n that he accounted f o r a l l the system a t t r i b u t e s perceived by the user to be pertinent to the problem of mode choice. Another shortcoming of the a p p l i c a t i o n of the abstract mode approach i s that some of the modal a t t r i b u t e s are hot comparable from one mode to the other. For example, how can one compare the f l e x i b i l i t y of routing provided by car use (thus permitting the user to avoid unpleasant routes, pick up or take someone along, etc.) with the f i x e d t r a n s i t route which does not o f f e r such a f l e x i b i l i t y . This d i f f i c u l t y i s inherent to the use of q u a l i t a t i v e data i n quant i t a t i v e models. But, needless to say, that many of the system att r i b u t e s cannot be expressed q u a n t i t a t i v e l y i n a s c i e n t i f i c a l l y acceptable manner. I i 5 . 3 . Planning Implications In i t s concern with conceptual analysis.and model formulation, t h i s study c l a r i f i e s several problems, i d e n t i f i e s others, and present some ideas which can be of be n e f i t to the planner. The model formulated i n the course of t h i s study, i f implemented, would expand our knowledge on the r e l a t i o n s h i p 1 2 3 between users' mode choices and the i r s e n s i t i v i t i e s toward modal a t t r i b u t e s , and furthermore, on the r e l a t i o n s h i p between the l a t t e r and users' time budget (which i s considered i n t h i s study to be an expression of ce r t a i n aspects of the i n d i v i d u a l ' s l i f e s t y l e ) . The i d e n t i f i c a t i o n of these complex and i n t e r r e l a t e d r elationships may possibly lead to the d e f i n i t i o n of the causal r e l a t i o n s h i p underlying users' behaviour. Since the model also accounts f o r the relevant attributes of the transportation system, i t provides the planner with means of assessing the impact of a l t e r n a t i v e transportation p o l i c i e s on users'mode c h o i c e s — p o s s i b l y i n quantitative terms. 3 The following are examples of such a l t e r n a t i v e s : (a) increase the cost of car t r a v e l by r a i s i n g the parking charges, road t o l l s , etc. (b) increase the t r a v e l time by car by reducing the parking space a v a i l a b l e i n the proximity of major destinations, thus prolonging the walking time to and from parking, or by allowing s t r e e t congestion to occur i n parts of the c i t y where t r a n s i t vehicles operate on a right-of-way. (c) lower the comfort l e v e l of car trips . b y neglecting road maintenance or improvements, and (d) reduce t r a n s i t t r a v e l time by providing a greater areal coverage of t r a n s i t service, increasing i t s frequency, es t a b l i s h i n g a separate right-of-way, etc. The planner may also wish to consider the impact of some of these p o l i c i e s on various mode users i n d i f f e r e n t . 1 2 4 socio-economic groups. This can possibly give him some indi c a t i o n s on the effectiveness of p o l i c y changes i n areas i n the c i t y characterized by, for example, high income or older age residents. With c e r t a i n modifications i n i t s s t r u c t u r a l r e l a t i o n s h i p s , the model may also be useful for th i s purpose: a l l observations are to be i n i t i a l l y c l a s s i f i e d into sub-groups, according to the socio-economic c r i t e r i a s p e c i f i e d (income l e v e l , age, e t c . ) . Within each group, users ' mode choices can be explained, by means of stochastic functions, on the basis of at t r i b u t e s of the mode choice s i t u a t i o n and the socio-economic c h a r a c t e r i s t i c s , excluding, of course, the variable used as a c r i t e r i o n f or the i n i t i a l c l a s s i f i c a t i o n . A comparison can then be drawn between the behaviour of various groups, and the differ e n c e can be attr i b u t e d to the e f f e c t of the variable employed as a c l a s s i f i c a t i o n c r i t e r i o n . Furthermore, the study's model can also be u t i l i z e d as a planning t o o l i n introducing a new t r a n s i t system or improving an e x i s t i n g one. This would require a s i m i l a r survey design to be conducted i n areas for which t r a n s i t s ervice i s planned. In t h i s survey, the planner should seek, information on the number or percentage of captive car users who would not s h i f t to t r a n s i t f or any of the reasons discussed i n Section 4.3. In addition, other information on the users 1 c h a r a c t e r i s t i c s and t h e i r t r a v e l pattern, as l i s t e d i n the study's questionnaire, should also be c o l l e c t e d . The planner 125 can then proceed to lay. the t r a n s i t routes so as to include a great number of poten t i a l users i n the "catchment" area of these routes. By estimating walking and t r a v e l time for the p o t e n t i a l users, a model s i m i l a r to that introduced here can be employed to estimate the number of users who would s h i f t to the new mode. Further to these, the mention i s made of other applications i n the course of the study. In Section 2.3., the use of the abstract mode choice theory for p r e d i c t i n g the use of new transportation modes i s expounded. Such modes can be reduced to their most s i g n i f i c a n t a t t r i b u t e s , and to the extent that these are comparable with a t t r i b u t e s of e x i s t i n g modes, the researcher can predict the impact of introducing a new mode on users' behaviour. Likewise, reference i s also made to the increased c a p a b i l i t y of the researcher to invest-igate or pr e d i c t the impact of s o c i a l or economic change, or new trends i n l i f e s t y l e s (as defined i n t h i s study). Since the model takes account of the user's personal c h a r a c t e r i s t i c s , (including h i s l i f e s t y l e ) and modal a t t r i b u t e s , such changes can be "entered" into the model and i t s impact on users' behaviour can be i d e n t i f i e d . Admittedly, however, major s o c i a l or economic changes, or r a d i c a l l y d i f f e r e n t orientations i n the population's l i f e s t y l e s have behavioural implications which may well produce new patterns of behaviour beyond the pr e d i c t i v e range of t h i s or any other model. The researcher, therefore, should exercise his judgement as to the degree of c r e d i b i l i t y he may be w i l l i n g to attach to the model performance 126 under these conditions of uncertainty. 5.4. Research P o s s i b i l i t i e s This study's approach i s d i f f e r e n t from others i n that i t places emphasis upon two aspects of user's behaviour: (a) the re l a t i o n s h i p between the user's perception of t r a v e l d i s u t i l i t y and mode choice, and (b) the re l a t i o n s h i p between the user's l i f e s t y l e and the s i g n i f i c a n c e he attaches to i n d i v i d u a l , a t t r i b u t e s of the t r a v e l mode. Previous works investigated the f i r s t aspect of user's behaviour mentioned above and s i g n i f i c a n t findings are reported i n the study. However, as the study points out, there i s s t i l l a degree of uncertainty i n the researcher's mind as to what attributes constitute a t r a v e l mode. In other words, we are back to the question posed at the outset of the study (Section 2.3.), and that i s how to reduce a t r a v e l mode to a "bundle" of attributes i n a manner which f u l l y accounts for the user's perception of these modes. This d i f f i c u l t y i s compounded when one considers the operational obstacles of formulating a model to account for both quantitative and q u a l i t a t i v e data. Yet, improving the state- o f - t h e - a r t i n mode choice analysis requires an appropriate s o l u t i o n to these problems. In addition, the concept of user's l i f e s t y l e , introduced here, although i t appears a promising conceptual t o o l , needs further refinement and support of empirical data. To improve 127 i t s r a t i o n a l e , perhaps a more elaborate conceptual framework i s required to r e l a t e the ind i v i d u a l ' s time and money budget to his personal values. A f r u i t f u l avenue of research would be to consider the pattern of money and time expenditures as an expression of the indiv i d u a l ' s personal values, and that for the i n d i v i d u a l to a r r i v e at the equilibrium state d e f i n i n g h i s pattern of time and money d i s t r i b u t i o n , c e r t a i n trade-offs must be made between various competing a c t i v i t i e s , and between goods and a c t i v i t i e s . As an example, an i n d i v i d u a l may choose, depending upon his personal values, to spend more time i n work and less i n t r a v e l . Also, an i n d i v i d u a l may choose to give up a vacation to buy some fu r n i t u r e , for example, thus trading a c t i v i t i e s f o r goods. In thi s context, money and time may then be considered a common resource, which the i n d i v i d u a l dispenses of i n the form of goods and a c t i v i t i e s i n such a manner which maximizes h i s s a t i s f a c t i o n . This general conceptual framework, being applicable to a c t i v i t i e s and commodities, may be construed as an elaboration on the e x i s t i n g consumer theory where consideration i s given ex c l u s i v e l y to the demand and supply of goods, and the concern i s with maximizing the consumer's s a t i s f a c t i o n under the constraint of f i n a n c i a l resources a v a i l a b l e . Adding to these considerations time as a li m i t e d resource, and a c t i v i t i e s as goods (time and/or time consuming?), the r a t i o n a l i z a t i o n of the i n d i v i d u a l ' s behaviour under these conditions poses challenging conceptual problems which are worthy of the researcher's attention. Since t r a v e l i s a time- and money-128 consuming a c t i v i t y , resolving these problems may enhance the t h e o r e t i c a l underpinnings of mode choice analysis. 129 FOOTNOTES 1. A detailed discussion of t h i s approach i s documented i n Mode Choice Determinants of Selected Socio- Economic Groups, by Brown, pp. 186-223. 2. The discrepancy between predicted mode use on the basis of stated preferences and actual use was i d e n t i f i e d by David T. Hartgen, Forecasting Remote  Park-and-Ride Transit Usage, Research and Applied Systems Section, New York State Department of Transportation, Dec. 72, p. 41. 3. This approach was attempted by Brown, Mode Choice  Determinants of Selected Socio-Economic Groups, pp. 166-223, and Hartgen, Forecasting T r a n s i t Usage, 1972, as well as many other authors. BIBLIOGRAPHY 130 BIBLIOGRAPHY A. BOOKS Bartley, S. H., P r i n c i p l e s of Perception, Harper, New York, 1958. Cooley, William W., and Paul R. Lohnes, M u l t i v a r i a t e Data  Anal y s i s , John Wiley, New York, 1971. Bevan, William, "Perception: Evolution of a Concept," Psychological Review, V o l . 65, No. 1. 1958, pp. 34-55. De Donnea, F. X., The Determinants of Transport Mode Choice  i n Dutch C i t i e s : Some Disaggregate Stochastic Models, Rotterdam U n i v e r s i t y Press, Rotterdam, 1971. Hope, Keith, Methods of Mu l t i v a r i a t e Analysis, U n i v e r s i t y of London Press, 1968. Horaburger, Wolfgang, ed., Urban Mass Tr a n s i t Planning, U n i v e r s i t y of C a l i f o r n i a , Berkeley, 1967. Krech, D. and R. S. C r u t c h f i e l d , Theory and Problems i n Experimental Psychology, McGraw-Hill, New York, 1948. Meyer, J . R., J . F. Kain and M. Wohl, The Urban Transportation  Problem, Harvard Univ e r s i t y Press, Cambridge, 3rd. ed., 196 M i l l e r , Delbert C., Handbook of Research Design and S o c i a l  Measurement, McKay Comp, New York, 1970. Smerk, George M., ed., Readings i n Urban Transportation, Indiana U n i v e r s i t y Press, Bloomington, 1968. The Universal Encyclopedia i n Mathematics, Mentor, 1964. Warner, S. L., Stochastic Choice of Mode i n Urban Travel; A Study i n Binary Choice, Northwestern Univ e r s i t y Press, 1962 B. PERIODICALS Adams, Warren T., "Factors Influencing Mass Transit and Automobile Use i n Urban Area", Public Roads, V o l . 30, 1959, pp. 256-260. 131 132 Beesley, M. E., "The Value of Time Spent i n T r a v e l l i n g : Some New Evidence," Economica, V o l . 32, 1965, pp. 174-185. Beeseley, M. E. and J. F. Kain, "Forecasting Car Ownership and Use," Urban Studies, V o l . 2., No. 2, 1965, pp. 163-185. Beier, Frederick J . , "Marketing Programs f o r Mass T r a n s i t / " T r a f f i c Quarterly, Oct., 1972, pp. 533-545. Botzow, Herman, "An Empirical Method f o r Estimating Auto Commuting Costs," Highway Research Record, Rep. No. 197, pp. 56-69. Domenich, T. A., G. Kraft, and J . P. V a l e t t e , "Estimation of Urban Passenger Travel Behaviour: An Economic Demand Model," Highway Research Record, Rep. No. 238, 1968, pp. 64-78. F e r r e r i , Michael G., et a l , "Choice and Captive Modal S p l i t Models," Highway Research Record, Rep. No. 369, 1971, pp. 80-90. j Golob, Thomas F., "The Survey of User Choice of A l t e r n a t i v e Transportation Modes," High Speed Ground Transportation  Journal, V o l . 4, No. 1, pp. 103-116. Goodman, Joseph M. "Evaluation of a Bus T r a n s i t System i n a Selected Urban Area," Highway Research Record, Rep. No. 314, 1970, pp. 14-122. Gustafson, Richard L., Harriet N. Curd, and Thomas F. Golob, "Users Preferences for a Demand-Responsive Transportation System," Highway Research Record, Rep. No. 467, 1971, pp. 31-45. Hadden, J. K.,"The Use of Public Transportation i n Milwaukee, Wisconsin," T r a f f i c Quarterly, V o l . 18, 1964, pp. 219-232. Haney, Dan G., "Consistency i n Transportation Modal S p l i t and Evaluation Models," Highway Research Record, Rep. No. 348, 1971, pp. 61-72. Harris, B r i t t o n , "New Tools f o r Planning," Journal of American  I n s t i t u t e of Planners, V o l . 31, May 1965, pp., 90-95. Harris, B r i t t o n , "Plan or Projection, An Examination of the Use of Models i n Planning," Journal of the American I n s t i t u t e  of Planners, Vol. 26, Nov. 1960, pp. 265-272. 133 Hartgen, David T., and George H. Tanner, "Investigations of the E f f e c t of Traveler Attitudes i n a Model of Mode Choice Behaviour," Highway Research Record, Rep. No. 369, 1971, pp. 1-14. Hartgen, D. T. and G. H. Tanner, "Individual A t t i t u d e s and Family A c t i v i t i e s : A Behavioural Model of Mode Choice," High Speed Ground Transportation Journal, V o l . 4, No. 2, Sept. 1970, pp. 439-467. Hedges, Charles, "An Evaluation of Commuter Transportation A l t e r n a t i v e s , " Highway Research Record, Rep. No. 269, 1969, pp. 12-45. H i l l , D. M., and H. G. von Cube, "Development of a Model for Forecasting Travel Mode Choice i n Urban Areias," Highway  Research Record, Rep. No. 38, pp. 7 8-96. H i l l e , Stanley and Theodore K. Martin, "Consumer Preference i n Transportation," Highway Research Record, Rep. No. 197, pp. 3 6-41. Lancaster, Kelvin J . , "A New Approach to Consumer Theory," Journal of P o l i t i c a l Economy, Vo l . 74, 1966, pp. 133-157. Lansing, John and Gary Hendricks, "How People Perceive the Cost of Journey to Work," Highway Research Record, Rep. No. 197, 1967, pp. 44-55. Lisco, T. W. "The Value of Commuter's Travel Time - A Study i n Urban Transportation," Ph.D., Un i v e r s i t y of Chicago, 1967, abridged i n Highway Research Record, Rep. No. 245, 1968, p. 36. Lave, Charles A., "A Behavioral Approach to-Modal S p l i t Forecasting, Transportation Research, V o l . 3, No. 4., 1969, pp. 463-480. Le Boulanger, "Research Into the Urban T r a v e l l e r ' s Behaviour," Transportation Research, V o l . 3, 1971, pp. 113-125. Leathers, Nancy J . , "Residential Location and Mode of Transportation to Work: A Model of Choice," Transportation  Research, V o l . 1, 1967, pp. 129-155. Lowry, Ira S., "A Short Course - Model Design," Journal of the  American I n s t i t u t e of Planners, V o l . 31, No. 2, May 1965, pp. 158-166. 134 M c G i l l i v r a y Robert G. "Demand and Choice Models of Modal S p l i t , " Journal of Transport Economics and P o l i c y , May 1970, pp. 192-206. Moses, Leon and H. Williamson, "Value of Time, Choice of Mode, and the Subsidy Issue i n Urban Transportation," Journal of P o l i t i c a l Economy, V o l . 71, 1963, pp. 247-264. Nash, A l l a n N. and Stanley H i l l e , "Public A t i t t u d e s Toward Transport Modes: A Summary of Two P i l o t Studies," Highway  Research Record, Rep. No. 233, 1968, pp. 33-46. Pratt, Richard H., "A U t i l i t a r i a n Theory of Travel Mode Choice," Highway Research Record, Rep. No. 322, 1970, pp. 40-53. Pratt, Richard H., and Thomas B. Deen, "Estimation of Sub-Modal S p l i t Within the Transit Mode," Highway Research  Record, Rep. No. 205, 1967, pp. 20-30. Quandt, Richard E.» "Estimation of Modal S p l i t s , " Transportation  Research, Vol. 2, March, 1968, pp. 41-50. Quandt, Richard E. and William J . Baumol, "The Demand f o r Abstract Transport Modes: Theory and Measurement," Journal of Regional Science, V o l . No. 2, 1966, pp. 13-25. Quarmby, David A., "Choice of Travel Mode for the Journey to Work," Journal of. Transport Economics and P o l i c y , Sept. 1967, pp. 273-314. Quarmby, David A., "Relating Public Transport to Urban Planning," Journal of the I n s t i t u t e of Transport, V o l . 30, No. 12, Sept. 64, pp. 435-440. Shunk, Gordon A., and Richard J . Bouchard, "An A p p l i c a t i o n of Marginal U t i l i t y to Travel Mode Choice," Highway Research  Record, Rep. No., 322, 1970, pp. 30-41. Schnore, Leo F., "The Use of Public Transportation i n Urban Areas," T r a f f i c Quarterly, Oct. 1962, pp. 488-499. Shaffer, Margaret T., "Attitudes Techniques i n Action," Highway Research Record, Rep. No. 305, 1970, pp. 112-122. Sharp, C. H., "The Choice Between Cars and Buses on Urban Roads," Journal of Transport Economics and P o l i c y , V o l . 1, No. 1, Jan. 1967, p. 104-111. 135 Sobey, A. J . , and J . W. Cone, "The Case f o r Personal Rapid Tra n s i t , " Highway Research Record, Rep. No. 367, 1971, pp. 70-89. Sommers, A l e x i s N. "Toward a Theory of Traveler Mode Choice," High Speed Transportation Journal, V o l . 4, No. 1, pp.' 1-8. Stopher, Peter R., "A P r o b a b i l i t y Model of Travel Mode Choice for the Work Journey," Highway Research Record, Rep. No. 283, 1969, pp. 57-65. Stopher, Peter R., and Shalom Reichman, "Disaggregate Stochastic Models of Travel Mode Choice, Highway Research  Record, Rep. No. 369, 1971, pp. 91-103. Thomas, Thomas C , and Gordon I. Thompson, "Value of Time Saved by T r i p Purpose," Highway Research Record, Rep. No. 369, 1971, pp. 104-117. • "The Value of Time f o r Commuting Motorists as a Function of Their Income Level and Amount of Time Saved, Highway Research Record, Rep. No. 314, 1970, pp. 1-19. Thomas, Thomas C , "Value of Time f o r Commuting Motorists, Highway Research Record, Rep. No. 245, 1968, pp. 17-34. Tomazinis, Anthony R., "Objectives and Obstacles i n Mass Tra n s i t System Development," High Speed Ground Transporta- t i o n Journal, V o l . 1, March 1967, pp. 364-377. Tomazinis, Anthony R., "Modal S p l i t Model i n the Penn-Jersey Transportation Study Area," Highway Research Record, Rep. No. 1967, pp. 41-75. Thuett, Bruce and Arthur J . Balek, "On the Need for a D e f i n i t i o n of Demand f o r Transportation," High Speed Ground Trans- portation Journal, Vol. 2, No. 3, Sept. 1968, pp. 576-592. i " ' Voorhees, Alan M. and S. Bellomo, "Urban Travel and C i t y Structure," Highway Research Record, Rep. No. 322, 1970, pp. 121-135. Watson, Peter L., "Problems Associated with Time and Cost Data Used i n Travel Choice Modeling and Valuation of Time," Highway Research Record, Rep. No. 369, 1971, pp. 148-158. Wynn, Houston F. "Short Modal_Split Formula," Highway Research  Record, Rep. No. 283, 1969, pp. 48-57. 136 Young, Kan Han, "An Abstract Mode Approach to the Demand for Travel," Transportation Research, V o l . 3. No. 4, 1969, pp. 443-461. Zupan, J e f f r e y M., "Mode Choice: Implications for Planning," Highway Research Record, Rep. No. 251, 1968, p. 25. C. SPECIAL REPORTS Ackoff, Russell L., "Individual Preferences for Varoius Means of Transportation," Interim Report of Management Science Centre. U n i v e r s i t y of Pennsylvania, May, 1965. Brown, Gerald R., "The Cor r e l a t i o n of Socio-economic Factors with Corridor Travel Demand," paper presented at the 53rd annual meeting of the Highway Research Board, Washington, D.C., Jan. 1974. . "A P o l i c y Model f o r Multimodal Transportation System Planning," Transportation Research Series, Report No. 2, The U n i v e r s i t y of B r i t i s h Columbia, A p r i l , 1973. . "An Analysis of User Preferences f o r System Characteri-s t i c s to Cause a Mode S h i f t , " paper presented at the 51st annual meeting of the Highway Research Board, Washington, D.C., Jan. 197 2. . Mode Choice Determinants of Selected Socio-economic Groups, Ph.Di, , Department of Community and Regional Planning, The U n i v e r s i t y of B r i t i s h Columbia, 1971. F e r t a l , Martin J . , et a l . Modal S p l i t , Documentation of Nine  Methods f o r Estimating Transit Usage, U.S. Department of Commerce, 196 6. Golob, Thomas F., and Richard L. Gustafson, "Economic Analysis of a Demand-Responsive Public Transportation System," a paper presented at the 50th Annual Meeting of the Highway Research Board, Washington, January, 1971. Hartgen, David T., Forecasting Demand fo r Improved Quality T r a n s i t Service with Small Sample Surveys, Research and Applied Systems Section, New York State Department of Transportation, A p r i l , 1973. ' "The influences of A t t i t u d i n a l and S i t u a t i o n a l Variables on Urban Mode Choice," Research and Applied Systems Section, New York State Department of Transportation, March, 1973. 137 Hartgen, David T., Forecasting Remote Park-and-Ride T r a n s i t  Usage, Research and Applied Systems Section, New York State Department of Transportation, Dec. 1972. . Mode Choice and Atti t u d e s : A L i t e r a t u r e Review. Research and Applied Systems Section, New York State Department of Transportation, July, 197 0. Hartgen, David T., George H. Tanner, Richard A. Maitino, An Empirical Approach to Estimating Mode Choice i n Small Urban Areas, Research and Applied Systems Section, New York State Department of Transportations, A p r i l , 1971. . Behavioural Model of Mode Choice, Research and Applied Systems Section; New York, Department of Transportation, March, 1970. . Individual Attitudes and Family A c t i v i t i e s : A Behavioural Model of Traveler Mode Choice. Research arid Applied Systems Section, New York State Department of Transportation, Aug., 1970. Lussi, John K., A R e l i a b i l i t y Analysis of Our Modal S p l i t Simulation Process, Research and Applied Systems Section, New York State Department of Transportation, No. 1969. Robinson, John D. and P h i l l i p R. Shaver, Measures of S o c i a l  Psychological A t t i t u d e s , I n s t i t u t e f o r S o c i a l Research, University of Michigan, 1973. Schaefer, Richard J . , Goerge H. Tanner, David T. Hartgen, Survey of Park and Ride Tr a n s i t Demand, Research and Applied Systems,Section, New York Department of Transportation, March, 1973. Schneider, Norman, R., Operating Costs f o r Passenger Cars on  Three F a c i l i t y Types; A r t e r i a l s - Junior Expressways- Expressways, Research and Applied Systems Section, New York State Department of Transportation, August, 1967. Tanner, George H., and Rosa Barba, Park-and-Tide Tr a n s i t Service: Some Guidelines and Considerations f o r Service  Implementation, Research and Applied Systems Section, New York State Department of Transportation, A p r i l , 1973. 138 D. UNPUBLISHED DOCUMENTS Cook, Norman A., "Perceptual V a r i a t i o n s of R e t a i l i n g i n Edmonton," Ph.D. Thesis, Department of Geography, Uni v e r s i t y of Alberta, 1972. Marvin, Francis P., Richard L. Gustafson, "User Preferences Dial-A-Bus: A Comparison of Two C i t i e s , " a working paper, General Motors Research Laboratories, n.d. Sheung-Ling Chan, " M u l t i c o l i n e a r i t y i n Transportation Models," M.A. Thesis, School of Community and Regional Planning, The University of B r i t i s h Columbia, 1970. 139 APPENDIX A EXAMPLE OF BUS USERS1 QUESTIONNAIRE Transit Authority X i s . conducting a survey on the t r a v e l l i n g habits of the residents of t h i s area. Would you please f i l l i n t h i s questionnaire and mail i t i n the attached envelope. Your cooperation may provide us with the information necessary to improve the transportation system serving t h i s c i t y . 1. Where d i d you s t a r t t h i s t r i p ? Address: Where i s the intended destination of t h i s t r i p ? Address: What i s the purpose of t h i s t r i p ? (a) work (b) shopping (c) personal business (d) s o c i a l , r e c r e a t i o n a l (e) school or u n i v e r s i t y • (f) others (specify) . At which time d i d the t r i p start? time: a.m. p.m. 5. How long d i d you have to walk to the bus stop? minutes 6. How long d i d you have to wait for the bus? minutes 7. How frequent i s the bus service at t h i s time? every - minutes 8. How long i s t h i s bus ride? minutes 9. How would you describe t h i s bus ride? (a) very comfortable (b) comfortable (c) somewhat comfortable (d) uncomfortable (e) very uncomfortable 140 10. Is transfer to another bus l i n e necessary i n t h i s t r i p ? Yes no If the answer i s yes, please continue, otherwise, go to question number 12. 11. (a) Do you have to walk to get to the next bus stop? Yes no If yes, how long: . minutes (b) how long do you have to wait at t h i s t r a n s f e r point? minutes (c) how frequent i s t h i s bus service? every minutes (d) how long i s t h i s bus ride? minutes 12. How long do you have to walk from the l a s t bus stop to your destination?: minutes 13. Do you use a monthly bus pass? Yes no 14. Could you have taken the car for t h i s t r i p instead of the bus? yes ' no If the answer i s yes, please continue, otherwise, go to question number 17. 15. In choosing the bus instead of the car, please indicate how important each of the following c h a r a c t e r i s t i c s of your t r i p : i (a) Cost of t r a v e l : • ; - very important - important • -. - somewhat important . - unimportant • • . - very unimportant - . (b) Total t r a v e l time: - very important - important - somewhat impprtant - unimportant - very unimportant 141 (c) The length of time spent i n walking, waiting and tr a n s f e r : - very important - important - somewhat important - unimportant - very unimportant (d) Frequency of the bus service: - very important - important - somewhat important - unimportant - very unimportant (e) Comfort l e v e l of the t r i p : - very important - important - somewhat important - unimportant - very unimportant _ 16. Suppose now that you took the car instead of the bus, please t e l l us, to the best of your knowledge, about your t r i p : (a) would you have to pay parking charges or road t o l l s ? . . . If yes, how much: d o l l a r s (b) how long the car t r i p would be: minutes (c) would you have to walk to or from the parking place? If yes, how long: minutes (d) How would you describe the car r i d e i n t h i s case: j (i) very comfortable I ( i i ) comfortable j ( i i i ) somewhat comfortable (iv) uncomfortable . (v) very uncomfortable • In order to have complete information on you as a bus r i d e r , please answer the following questions: 17. Your sex: male female 18. Your age: (a) 16 - 25 (d) 46 - 55 (b) 26 - 35 (e) 56 - 65 (c) 36 - 45 (f) over 65 142 19. Which of the following categories would describe your occupation best? (a) c l e r i c a l , s e c r e t a r i a l (b) sales personnel (c) managerial (d) professional ______ (e) labour, trade (f) self-employed (e) student (g) housewife _ (h) others, (specify) -20. Please indicate the annual income of your household: (a) le s s than $6000 (b) $ 6000 to $ 8000 _____ (c) $ 8000 to $10000 (d) $10000 to $12000 (e) $12000 to $14000 (f) $14000 to $16000 (g) over $16000 21. Can you work overtime i f you wish? Yes no 22. How many persons are i n your household? persons 23. Now, we would l i k e to know about your way of l i f e and how you generally spend your time, since we think that t h i s may have an e f f e c t on your t r a v e l habits. Please give us an estimate f o r the number of weekly hours you spent i n the following a c t i v i t i e s ? (a) at work (b) at home, with family or r e l a t i v e s (c) p r a c t i s i n g your f a v o r i t e hobby, pursuing c u l t u r a l i n t e r e s t s , associating with f r i e n d s _ (d) sports and outdoor recreation 143 APPENDIX B CAR USERS QUESTIONNAIRE The d e s i g n . o f t h i s q u e s t i o n n a i r e can be made s i m i l a r to t h a t o f the bus u s e r s ' q u e s t i o n n a i r e , except f o r the q u e s t i o n s concerning the a t t r i b u t e s o f the t r a n s p o r t a t i o n system. In t h i s appendix o n l y these q u e s t i o n s are mentioned. 5. How long i s t h i s c a r r i d e ? minutes 6. Do you have t o walk to or from the p a r k i n g space? I f y e s, how lo n g : ' minutes 7. Do you pay p a r k i n g charges or road t o l l s ? I f yes, how much: • d o l l a r s 8. How would you d e s c r i b e your c a r r i d e ? (a) v e r y c o m f o r t a b l e (b) comfortable . (c.) somewhat c o m f o r t a b l e ._ (d) uncomfortable (e) v e r y uncomfortable 9. D i d you choose the car because: (a) you are s h a r i n g a c a r p o o l , or have t o take someone e l s e a l o n g , • •(b)- you needed the package space t o c a r r y t o o l s , shopping bags, e t c . (c) you needed the c a r a t the end of the t r i p t o make oth e r t r i p s , (d) you would be embarrassed t o be seen r i d i n g the bus, (e) t h e r e i s no a c c e p t a b l e bus s e r v i c e t o g e t you to your d e s t i n a t i o n i n time. I f the answer to any of the q u e s t i o n s a, b, c, d, o r e i s yes, p l e a s e go t o q u e s t i o n 12, o t h e r w i s e c o n t i n u e . 10. Suppose now t h a t you have taken the bus i n s t e a d of the c a r , p l e a s e e s t i m a t e , t o the b e s t of your knowledge, the f o l l o w i n g 144 (a) how long would you have to walk to the bus stop: minutes (b) how long do you have to wait f o r the bus stop: minutes (c) how frequent i s the bus service at t h i s time of the day? every minutes (d) how long would the bus r i d e be? minutes (e) how would you describe t h i s ride? (i) very comfortable • ( i i ) comfortable ( i i i ) somewhat comfortable • (iv) uncomfortable (v) very uncomfortable (f) would a transfer to another bus l i n e be necessary to get to your destination? If yes, please continue, otherwise go to question k below. (g) would you have to walk to another bus stop f o r t h i s transfer? If yes, how long minutes (h) how long would you have to wait for the bus at t h i s stop: • minutes (i) how frequent i s the bus service f o r t h i s l i n e at th i s time of the day: every minutes (k) how long would you have to walk to your f i n a l destination a f t e r you leave the bus: minutes 11. In choosing the car instead of the bus, please indicate how important each of the following c h a r a c t e r i s t i c s of your t r i p : (a) Cost of t r a v e l : - very important - important - somewhat important - unimportant - very unimportant : • (b) Total Travel time: - very important - important ' - somewhat important - unimportant - very unimportant 145 (c) The length of time spent i n walking, or waiting ( if a p p l i c a b l e ) : - very important - important - somewhat important ' - unimportant • - very unimportant (d) Frequency of the bus service ( i f you consider taking the bus): - very important - important - somewhat important - unimportant - very unimportant (e) Comfort l e v e l of the t r i p : - very important -- important - somewhat important - unimportant [ - very unimportant 12. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0093009/manifest

Comment

Related Items