UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The Peripheral journey to work in Vancouver Hickman, Richard Michael 1968

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

Item Metadata

Download

Media
831-UBC_1968_A8 H53.pdf [ 3.83MB ]
Metadata
JSON: 831-1.0104409.json
JSON-LD: 831-1.0104409-ld.json
RDF/XML (Pretty): 831-1.0104409-rdf.xml
RDF/JSON: 831-1.0104409-rdf.json
Turtle: 831-1.0104409-turtle.txt
N-Triples: 831-1.0104409-rdf-ntriples.txt
Original Record: 831-1.0104409-source.json
Full Text
831-1.0104409-fulltext.txt
Citation
831-1.0104409.ris

Full Text

THE PERIPHERAL JOURNEY TO WORK IN VANCOUVER by RICHARD MICHAEL HICKMAN B.A., University of London, I963 A THESIS SUBMITTED IN THE REQUIREMENTS MASTER PARTIAL FULFILMENT OF FOR THE DEGREE OF OF ARTS in the School of Community and Regional Planning We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1968 In presenting this thesis i n pa r t i a l fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my School or by his representatives. It i s understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. School of Community and Regional Planning, University of British Columbia, Vancouver 8 , British Columbia. May 1968 ( i i ) ABSTRACT The hypothesis of this study i s that commuter journeys to employment in the central business d i s t r i c t s of large c i t i e s are not representative, i n terms of t r i p length and dispersion, of commuter journeys to employment in the suburbs. It i s argued that i n some larger c i t i e s , journeys to suburban or peripheral employment form an important and growing proportion of a l l work journeys, and that i f these are significantly different from the journey to work to the central business d i s t r i c t , this w i l l have important implications i n future transportation planning, and indirectly i n planning the d i s t r i -bution of residences and employment. A short review of existing journey to work literature i s presented. The majority of previous studies of the journey to work have been concerned primarily with the commuter journey to the downtown area, or are i n such general terms that, without further analysis, i t i s not possible to identify the patterns and characteristics of the peripheral journey to work. A random sample of employed residents of the City of Vancouver and the Municipality of Burnaby i s used to document the characteristics of peripheral work journeys i n the Vancouver Metropolitan area, and to compare them with downtown work trips. Vancouver forms a suitable c i t y for a study of the peripheral journey to work as i t shows low development densities, a high degree of dependence on travel by car, and a reasonable proportion of employment located i n the suburbs. The sample drawn i s not large enough and the information not varied enough to conduct a detailed explanatory ( i i i ) investigation of the factors influencing the pattern of peripheral work t r i p s . However the descriptive material indicates that peripheral work trips are significantly shorter i n length than commuter trips to the central area of Vancouver, and that they show a much greater variety of t r i p length and t r i p direction. The results suggest that peripheral work trips are composed of a large number of very small zone to zone volumes, forming a relatively even multidirectional network of trips throughout the suburban area. The present pattern of trips does not appear to be suited to the provision of high or medium volume transit f a c i l i t i e s for suburban journeys, and this i n turn i s a constraint upon the formation of large concentrations of jobs i n suburban areas. The interrelationships of urban structure i n terms of the distribution of homes and employment, and the s u i t a b i l i t i e s of alternative transportation modes are discussed, and the need for explicit policy objectives and coordinated land use and transportation plans i s stressed. In addition, the evidence suggests that a large proportion of persons employed i n the suburbs appear to prefer a more specialised choice:of residential location, rather than attempting to minimise the journey to work. It i s suggested that the descriptive evidence i s sufficient to indicate the distinctiveness of peripheral work journeys from a transport-ation point of view, and that they are important enough to merit more detailed explanatory studies and special attention i n transportation planning. (iv) TABLE OF CONTENTS CHAPTER PAG-E I INTRODUCTION 1 Scope and significance of study 1 Objectives and methods 2 limitations of study 4 II WORK JOURNEY PATTERNS 6 The separation of workplaces and residences 6 Descriptive studies 12 Explanatory studies 15 Studies of the peripheral journey to work 19 III THE PERIPHERAL JOURNEY TO WORK IN VANCOUVER 23 Method of investigation 23 Results and findings 26 IV CONCLUSIONS AND IMPLICATIONS 49 Comparison with other studies ,49 Choices of residential location for peripheral employment 51 Applicability of gravity model to peripheral work journeys 53 Implications for transportation planning 53 BIBLIOGRAPHY 58 APPENDIX Sampling procedure 60 (v) LIST OF TABLES TABLE PAGE I Distribution of jobs i n London Metropolitan Region, 1951, 1961, 1981 8 II Estimated changes i n employment in Metropolitan Vancouver, 1955 - 1985 10 III Land use at origin and destination for urban travel i n f i f t y c i t i e s 13 IV Comparison of employment characteristics for West Suburban Area and Central Business District of Chicago 21 V Characteristics of sample of work journeys in Vancouver 25 VI Mean work-residence separation for two Vancouver samples 26 VII Variation i n work t r i p length by place of employment 29 VIII Direction of peripheral work journeys in Vancouver 30 (vi) LIST OF FIGURES FIGURE PAGE 1 Dependence on downtown employment by distance from downtown for component clusters of sample 27 2 Percentage frequency distributions of peripheral and downtown work journeys by t r i p length 28 3 Key map showing location of residential clusters forming work journey sample 33 4-13 Cartographic plots of work journeys originating at each of ten residential clusters forming work journey sample 35-44 14 Employment by t r a f f i c zones, Vancouver 1965> (simplified). 46 ( v i i ) ACKNOWLEDGEMENT I wish to record my appreciation of the advice and suggestions of Professors V Setty Pendakur and Brahm Wiesman, and Mr John Wolforth during the preparation of this study. CHAPTER I INTRODUCTION Scope and significance of study The planning for and provision of transportation f a c i l i t i e s for the journey between home and workplace has become an integral element i n the preparation and execution of cit y development plans. For this purpose comprehensive descriptive studies of travel patterns have been carried out i n many major c i t i e s . Usually, most emphasis has been given to the measure-ment and analysis of commuter trips to downtown workplaces, for i n the large c i t i e s these trips have proved the most d i f f i o u l t to cater for adequately, largely due to their concentration into certain areas at particular times of day. With a very few exceptions, the commuter journey to workplaces situated i n the suburbs has been treated only i n a general manner, as a subsidiary component of urban travel. There has been l i t t l e attempt to investigate the underlying factors influencing this type of t r i p , or to compare the characteristics of downtown and non-downtown oriented commuter journeys. It i s these commuter journeys to workplaces i n the suburbs, the peripheral journey to work, that form the main focus of this study, although some consideration of downtown journeys^" has been necessary for comparitive 2 purposes. There i s increasing evidence i n some larger c i t i e s that these ^ For the purposes of this study, 'work t r i p ' means the commuter journey between the home and the place of work, and does not include travel for business purposes to other locations or premises. This i s discussed i n Chapter II. - 2 -peripheral or suburban work journeys form not only the larger proportion of a l l commuter tri p s , but also the fastest growing sector of work journeys, and that this type of t r i p w i l l be of increasing importance i n the future. If this type of journey exhibits significantly different characteristics from the downtown commuter journey, this w i l l have implications i n terms of transportation requirements, and i n wider planning considerations of urban structure. Objectives and methods This study attempts to satisfy: three major objectives: ( l ) to review selected literature on the journey to work and to examine the extent to which i t can be adapted to the study of the peripheral journey to work; (2) to examine the changing distribution of employment i n two major c i t i e s , London, England, and New York, i n order to assess the importance of the peripheral journey to work, and to evaluate the degree to which similar tendencies may be affecting the metropolitan area of Vancouver, British Columbia; (3) to collect information on the peripheral journey to work i n Vancouver, and to compare some of the characteristics of peripheral work journeys with those of commuter tripsto downtown Vancouver. The latter part of the study i s based on information from the Vancouver City Directory. A random sample of residential areas i n the City of Vancouver and the Municipality of Burnaby was used as the basis for build-ing up a picture of the destinations of work trips originating i n the sample areas. Each of the ten residential tracts that formed the sample ^ For a f u l l description of the sampling procedure see Appendix, and for the location of the ten residential areas making up the sample see the key map, Figure 3. consisted of four or six street "blocks, depending on the density of develop-ment. The street section of the directory was used to identify the names of a l l employed adults l i v i n g i n the areas sampled, and the alphabetical section of the directory provided their occupations and the names of their employers. The name of the employer could then be looked up to give the location of the workplace. This procedure was used to obtain details of the origins and destinations, of about 700 commuter journeys, and this information was then used to compute the direction, length, and dispersion of downtown and peripheral trips. For this study of the peripheral journey to work, no detailed attempt was made to relate origins and destinations to the social class of residents or to the number and composition of job opportunites at the place of work. The objectives are limited to a comparison of work trips to downtown and suburban employment on the basis of simple, easily measured characteristics such as length and dispersion of tr i p s . If basic differences are revealed, this would indicate the value of more intensive studies, both descriptive and explanatory, based on larger samples. It would also point to the importance of attention to the peripheral journey to work i n the planning of land use and transportation systems. For this study, two hypotheses were adopted: (1) Commuter journeys to suburban locations are l i k e l y to be significantly  shorter than commuter journeys to downtown employment; (2) Commuter journeys to suburban locations, are l i k e l y to show greater  variety, i n terms of tr i p length, dispersion, and direction, than  commuter journeys to downtown employment. Data for I963 were used so that the study would be directly comparable to another study , chiefly relating to the journey to work to downtown Vancouver, based on material for the same year. Vancouver forms a suitable subject for a study of the peripheral journey to work because the metropolitan area already exhibits a f a i r l y wide distribution of places of employment. Present patterns of residential growth and transportation f a c i l i t i e s , i f unchanged, seem l i k e l y to reinforce the growth of suburban employment. Limitations of the study Restrictions of time and resources were the chief constraint on the size and detail of the investigation that has been undertaken. It was found that there was l i t t l e published literature on the peripheral journey to work, and that there was no s t a t i s t i c a l material on the journey to work i n Vancouver available i n a suitable form for analysis. The sample study that has been carried out i s sufficiently large for broad generalisations, but i s not adequate for repeated division and microanalysis. Thus no attempt has been made to investigate the type of employment , the social class of residents, the mode of transport used, or other variables that would probably form important aspects of a more comprehensive explanatory study of the peripheral journey to work. The primary objective of this study has been to make a preliminary appraisal of the subject of the peripheral journey to work; to gather some of the more basic data; and to make some simple comparisons with downtown journeys to work to show the distinctive-ness and importance of this type of journey. Chapter II contains a review of existing literature on the journey to John R. Wolforth, Residential Location and the Place of Work, B.C. Geographical Series No. 4- (Vancouver: Tantalus Research Ltd., 1965) work, particularly the few studies that specifically relate to the peripheral journey to work. The chapter also includes a brief evaluation of the changing distribution of employment i n London, New York, and Vancouver. In chapter III, the data and findings of the work journey sample from the Vancouver City Directory are presented i n both s t a t i s t i c a l and graphical form. In chapter IV, the most significant material contained i n chapters II and III i s drawn together, and the implications for transportation planning are discussed. CHAPTER II WORK JOURNEY PATTERNS The separation of workplaces and residences The emergence of the journey to work as a significant aspect of the functioning of c i t i e s i s the result of the increasing separation of work-places and residences. In mediaeval times, a large proportion of the population lived at, or very close to, the place of work; for example i n the upper stories of buildings containing shops or workshops at lower levels. The speed-distance capability of transportation modes available placed a very real constraint on the distance that could separate the dwelling and the workplace. The emergence of specialised workplaces i n the form of larger workshops, facto'ries, and offices during the 19th century necessitated a complementary specialisation of buildings primarily for residential use. As more efficient means of transport became available, i t was possible for an increasingly greater distance to separate homes, and workplaces. At f i r s t , only the higher income groups could afford the luxury of homes distant from the noise and d i r t of the factories and close-packed workmens' dwellings of the industrial revolution. However, with the advent of the tramway during the late 19th century and the institution of cheap workmens' fares, a degree of work residence separation became possible for nearly 5 a l l social groups. Rasmussen has traced the rapid areal growth of London that resulted from this phenomenon and ill u s t r a t e s by a series of c Steen E i l e r Rasmussen, London, the Unique City (London: Jonathan Cape, 1937), Pelican Edition, pp. 126-132 - 6 -maps the rapid extension of the housing areas, particularly during the 1880's and 1890*s. The development of commuter r a i l services and automobile travel f a c i l i t a t e d a further dispersion of residential areas, and an increas-ingly specialised differentiation of residential areas serving different socio-economic groups. This differentiation and specialisation was recognised i n the stylised urban structure theory developed by the Chicago, School of Ecologists i n the mid twenties, and has been well documented. During the Victorian period, the majority of jobs were s t i l l located at or near the oity centre, but the dominance of the central c i t y i n the distribution of employment has since been progressively weakened. In New York i n 1889, the core area (as defined for the New York Metropolitan Region Study, and covering a much larger area than the central business d i s t r i c t of Manhattan), contained just under threequarters of a l l jobs i n the region^. In London at this period, maps prepared for Charles Booth's survey show that jobs were located mainly i n the centre, and i n the adjacent dockside g areas of factories and warehousing . However from the turn of the century onwards i n New York, and from the f i r s t world war onwards in London, suburban locations began to gain an increasing though relatively small share of the growth in employment. By the outbreak of the second world war, the core of New York accounted for only 6o$ of jobs i n the region^, while i n London, the inner area covered by the London County Council contained 6 R. E. Park (Ed.), The City (Chicago: 1925) 7 Edgar M. Hoover and Raymond Vernon, Anatomy of a Metropolis (Harvard University Press, 1959) Anchor Edition, p 22 Charles Booth, Life and Labour of the People of London (London: 1903) -Q Hoover and Vernon, loc. c i t . only 45$ of jobs i n the metropolitan region by 1951 • The post war period has witnessed continued growth of jobs i n the suburban and peripheral areas of both c i t i e s , while the inner zones show a stable employment situation, with very l i t t l e growth of jobs i n inner London, and an absolute decline i n the core of New York 1 1. TABLE I DISTRIBUTION OF JOBS IN LONDON METROPOLITAN REG-ION', 1951,1961,1981 1951 a 1961 b Change Estimated 1951-61 Change I96I-8I absolute $ absolute $ London County ^ 42.0$ +11,985 +1.2$ +87,000 +3.4$ Council area 45-4$ Rest of Greater London Council area 29.6$ 29.6$ +lM-,658 +8.7$ +145,000 +8.0$ Inner and Outer Outer part of Country Rings, 28.5$ L.T.S. area including New Towns 24.9$ +333,528 +22.9$ +71,000 +24.0$ LONDON METROPOLITAN REGION 100$ 100$ +490,171 +8.7$ +303,000 +6.0$ (Total for (5,633,413) (6,123,584) L.T.S. area) ' ' Great Britain, Registrar General, Census 1951, Census 1961 c Greater London Council, "New 1981 Planning Estimates for Phase III of the London Transportation Study", London; May, I967 (Mimeographed) ^ The London County Council area covers the inner part of London developed prior to 1914« The area i s considerably larger than the central business d i s t r i c t . Great' Britain, Registrar General, Census, 1951 Hoover and Vernon, loc. c i t . - 9 -Table I shows the changes i n the distribution of employment i n London between 1951 and 1961, and the changes expected to occur by 1981. The rapid growth of employment i n the outer rings, and the stable situation at the centre i s evident. The zone of the region lying outside the Greater London Council area gained 68$ of a l l growth of jobs i n the region during the decade, while the London County Council area accounted for only 2$ of a l l growth. These figures drawn from two large and well documented ci t i e s clearly show the growing importance of jobs situated i n non-central locations-It i s the commuter journeys attracted to these suburban jobs that form the primary focus of the present study. It i s d i f f i c u l t to test the extent to which similar trends are affect-ing Vancouver, for detailed employment statistics are not available. However, estimates prepared for the Vancouver Transportation Study hased on census material for 1951 and 1961, and other sources, indicatea-.rapid increase i n employment insuburban locations, and a progressive decline i n 12 the relative importance of jobs i n the ci t y of Vancouver . Table II, developed from these estimates, shows that for the decade 1955-65* Vancouver City experienced the lowest rate of employment growth of a l l : the municipalities i n the metropolitan area except for New Westminster, another old-established centre. For the period 1965-85» the rate of employment growth in Vancouver City i s expected to drop substantially, while Burnaby i s expected to maintain a high growth rate. The municipalities of the south shore, Richmond, Delta, and Surrey, are expected to experience even higher rates of growth than during the decade 1955-65' Thus, while the City of Vancouver City Planning Department, Metropolitan Vancouver, 19551 1965, and 1985: Selected Data from the Vancouver Transportation  Study (Vancouver City Council, March 1967), p 28 - 10 -Vancouver secured over half the growth of jobs i n the metropolitan area during the period 1955-65* i t s share i s predicted to be only about one quarter of a l l employment growth for the period 1965-85. The share of this growth secured by the south shore i s expected to tr i p l e compared with i t s share of metropolitan growth i n the period 1955-65' TABLE II ESTIMATED CHANGES IN EMPLOYMENT IN METROPOLITAN VANCOUVER 1955-1985 Share of Metropolitan „ , , Rate of Growth Share of Growth Employment 1955 1965 1985 1955-65 1965-85 1955-65 1965- i85 Vancouver City 68.1$ 64.9$ 51-7$ 31$ 23$ 56$ 27$ Burnaby 7-9$ 8.6$ 10.5$ 49$ 89$ 10$ 14$ Rest of Burrard Peninsula 12.1$ 12.1$ 12.3$ 36$ 57$ 12$ 12$ North Shore ^ 4 . 6 $ 5.8$ 6.5$ 73$ 73$ 9$ 8$ South Shore 7-3$ 8.7$ 19.0$ 61$ 238$ 12$ 38$ METROPOLITAN REGION 100$ 100$ 100$ 100$ 100$ 100$ 100$ Source: Adapted from Metropolitan Vancouver, 1955* 1965» and 1985 (Vancouver: 1967) a University Endowment Lands, New Westminster, Coquitlam, Port Coquitlam, Port Moody, Fraser Mi l l s , and adjacent unorganised territory 1 3 West Vancouver, City of North Vancouver, District of North Vancouver ° Richmond, Delta, Surrey - 11 -The overall effect of these predicted changes i s that while nearly 70$ of a l l jobs i n the metropolitan region were located i n Vancouver in 1955, only a l i t t l e over half of them are l i k e l y to be within the c i t y limits by 1985 and an even smaller share within the central business d i s t r i c t . On the other hand, the peripheral communities of the south shore are expected to increase their share of a l l jobs i n the region from about 7$ to 19$ during the same period. Two points must be emphasized: (a) the figures for the City of Vancouver are for the entire area of the ci t y , and employment in the central business d i s t r i c t represents only a portion of these figures; and (b) the s t a t i s t i c a l sources and processing that support these estimates are open to some criticism i n detail, but the major trends affecting the distribution of employment i n the Vancouver area are clear. Prior to a detailed examination of the peripheral journey to work in Vancouver, a selected review of the existing literature on the journey to work would be useful. Comprehensive accounts of the f u l l range of journey to work studies and surveys of particular c i t i e s are available elsewhere 1^' For present purposes, emphasis i s placed on sources summarising the present state of knowledge, both descriptive and explanatory, and on the few studies that relate to peripheral work t r i p s , so that an overall perspective can be obtained. ^ Wolforth, Op. Cit. ^ Howard S. Lapin, Structuring the Journey to Work (Philadelphia: University of Pennsylvania Press, 19&t-) - 12 -Descriptive Studies Descriptive studies of the journey to work have appeared for many years. Much of the early work consisted of a purely verbal description and concentrated on the social and economic costs of the increasing length of the journey to work. This was one of the considerations that led to the decision to form a green belt round London during the 1930*s and to the commencement of the o f f i c i a l B ritish New Towns programme i n 194-6. The post war period saw the beginning of urban transportation studies based on methodical survey and analysis techniques and, for the f i r s t time, i t became possible to describe urban travel patterns both quantitatively and comprehensively. 15 Curran and Stegmeier have amalgamated and summarized travel data derived from transporation surveys in f i f t y North American c i t i e s . Table III, adapted from their study, shows the land use at origin and destination for nearly 28 million motorised person t r i p s , including travel by truck, car, taxi, and public transit, but excluding travel by bicycle or on foot. Trips from home to work and business, and i n the reverse direction, represent just under 4-0$ of a l l t r i p s , and this gives an indication of the overall importance of the commuter journey i n the total pattern of urban motorised t r i p s . In 16 a more detailed summarizing study, Lapin finds that the length of work t r i p increases with the size of the c i t y , and that variations i n modal choice are associated with location characteristics of residential areas. 15 Prank B. Curran and Joseph T. Stegmeier, 'Traffic Patterns i n F i f t y Cities', Public Roads - A Journal of Highway Research, Vol. 3 0 , No. 5» December, 1958, (Washington: U.S. Department of Commerce, Bureau of Public Roads). Lapin, Op. Cit. - 13 -TABLE III LAND USE AT ORI&IN AND DESTINATION FOR URBAN TRAVEL IM FIFTY CITIES LAND USE AT ORI&IN LAND USE AT DESTINATION Work & Business Social & Recreation Shop Miscellaneous Home Total Work and Business 4 .9$ 0.2$ 0.3$ 2.3$ 20.2$ 27-9$ Social and Recreation 0.5$ 1.7$ 0.4$ 0.7$ 8.7$ 12.0$ Shop 0.7$ 0.3$ 0.8$ 0.5$ 5.2$ 7-5$ Miscellaneous 2.3$ 0.6$ 0.4$ 1.6$ 6.9$ 11.8$ Home 19.5$ 9-2$ 5.6$ 6.5$ - 40.8$ TOTAL 27.9$ 12.0$ 7.5$ 11.6$ 41.0$ 100$ a Source: Curran and Stegmeier, Op. Citi. 27*7 million motorised person trips He also notes the tendency to peaking i n the timing of work journeys, and the low passenger occupancy of automobiles. In particular, he quotes a 1947 study of the Philadelphia-Camden area which found that the proportion of persons travelling to work by car increased with the distance of the workplace from the cit y centre. Although Lapin's discussion focusses primarily on work trips to the central business d i s t r i c t , he emphasises the importance of the distribution of employment throughout the urban area as a key influence on the pattern of work trips. This theme i s taken up by - 14 -Louis K. Loewenstein^ who analyses the distribution of jobs and residences i n a number of American c i t i e s . He examines the varying distribution of jobs of different types i n concentric zones superimposed on each urban area, and the corresponding residential distribution of persons employed i n each of the selected categories. In each concentric zone, the discrepancy between the number of resident employees and the number of jobs i n each category i s used to compute the estimated zone to zone movements for work purposes i n general figures. Employment i n manufacturing i s found to be widely d i s t r i -buted throughout the urban area of the c i t i e s examined, and the residences of persons employed i n manufacturing show a similarly wide dispersion. Loewenstein infers from this that work tr i p s to employment i n manufacturing occur mainly i n the suburbs, but he does not have adequate information to make a detailed investigation of the resulting pattern of peripheral work 18 journeys. Beverly Duncan , using Chicago material, also finds a close association between the degree of work/residence separation and occupation, but again there i s no detailed description of journey patterns, or any discussion of peripheral work tr i p s . The most comprehensive descriptive coverage of suburban work travel patterns can be derived from the large scale transportation surveys conducted i n many major oities i n recent years. These are usually based on sample information collected i n various t r a f f i c zones which i s expanded and projected to estimate zone to zone travel desires. Often these zone to Louis: K. Loewenstein, The Location of Residences and Work Places  i n Urban Areas, (New York"! The Scarecrow Press, I965) 18 Beverly Duncan, 'Factors i n Work-Residence Separation: Wage and Salary Workers, Chicago, 1951'» American Sociological Review, (Vol. 21, February, 1956), pp 48-56 - 15 -zone estimates represent trips for a l l purposes and do not distinguish work tr i p s . Such surveys do not normally require detailed descriptions or explanations of suburban work journey patterns, and although the original data, i f reprocessed, can provide valuable descriptive material, the usual form of presentation i s too generalised to be of assistance i n the present study of the peripheral journey to work. Explanatory Studies The studies showing more emphasis on the explanation than the description of journey to work patterns f a l l into two main groups: those that attempt a direct explanation of the travel movements from zone to zone;; and the 'indirect' studies that are more concerned with the factors underlying the location of residences. An important element i n the latter i s the costs of the journey to work i n money, time, and inconvenience as a locational factor affecting choice of residence. Few of these studies investigate the factors affecting the location of employment, which tends to be studied as an issue separate and distinct from the study of the journey to work. The gravity model, or the modified form of potential model, has been used as an explanatory tool i n many t r a f f i c studies. In essence, this form of model assumes that the interaction between any two zones i s a function of the population i n each zone, which may be measured i n a variety of ways, and i s inversely related to the distance separating the centroids of the two zones, measured by airline distances, route miles, or time. However, i t has been found that the levels of explanation afforded by the gravity model are often not sufficiently accurate when applied to individual small zones, or when situations are particularly complex. In these cases i t has been necessary to introduce such modifications as frictionless zones and - 16 -an element of randomness i n the process of distributing trips according to 19 population and distance. Bevis has used a basic gravity model to predict zone to zone travel desires for the Chicago Area Transportation Study, and 20 has obtained good results, while Hansen's modified gravity model incorpor-ating an exponential function of distance provided reasonably accurate predictions of residential land use changes for Washington D. C. A further example of the use of a potential model for explaining and predicting work 21 journeys i s the work done by Beverly Duncan and Otis Dudley Duncan i n Chicago. 'Workplace potential' isoline maps were prepared for various categories of employment, and these were used to estimate the expected number of employees i n the various categories for each of 211 census tracts i n a random sample. Estimates were compared with census data, and a good level of explanation was obtained i n about half of the cases. A more sophisticated model for predicting the distribution of residential 22 growth was prepared by Herbert and Stevens . This involved a number of variables incorporated into a complex regression programme. It was hypo-thesised that individual households endeavour to satisfy a wide range of 19 Howard W. Bevis, 'A Model for Predicting Urban Travel Patterns', Journal of the American Institute of Planners, Volume 25, No. 2, (May 1959), PP 87-89 20 Walter G. Hansen, 'How Accessibility Shapes Land Use', Journal of  the American Institute of Planners, Vol. 25, No. 2, (May 1959) PP 73-76 21 Beverly Duncan and Otis Dudley Duncan, 'The Measurement of Intra-City Ideational and Residential Patterns', Journal of Regional  Science. Vol. 2, No. 2, ( P a l l , i 9 6 0 ) , pp 37-54 22 John D. Herbert and Benjamin H. Stevens, 'A Model for the D i s t r i -bution of Residential Activity i n Urban Areas', Journal of  Regional Science, Vol. 2, No. 2, (Pall i 9 6 0 ) , pp 21-36 - 17 -requirements, some of which are mutually exclusive or at best only partly compatible. A linear programme was used that enabled each household to maximise benefits, subject to such exogamous variables as the av a i l a b i l i t y of land and the number of households to be allocated, and the resulting distribution was iterated, on an incremental basis, for short time periods. Once again, a satisfactory level of explanation was obtained, but the results contribute only indirectly to an analysis of the peripheral journey to work. Kain , Alonso , and Wingo have worked separately on similar investigations of factors influencing residential location. A oommon theme i n their work i s the interchangability of site rents and transportation costs. Alonso's 'bid rent curve' reflects this equilibrium, with an equal level of satisfaction at a l l points on the curve even though this satis-faction i s based on a varying mix of residential space and job accessibility. Kain's work, developing the same concept, shows that family size and income are the most important variables i n determining where on the equilibrium curve a household's preference w i l l be located. Wingo attempts to place a marginal value on leisure time as a subtler benefit of a shorter work t r i p . In a l l these cases, employment i n suburban locations allows an individual ox John F. Kain, 'The Journey to Work as a Determinant of Residential Location', Paper read before the meeting of the Regional Science Association, December, 1961 William Alonso, 'A Theory of the Urban Land Market', Papers and  Proceedings of the Regional Science Association Sixth Annual  Meeting, (Washington, 1959), Vol. VI, I960 pp 149-157 Lowdon Wingo Jnr., Transportation and Urban Land (Washington: Resources for the Future Inc., - 18 -household a wider range of choice i n the allocation of resources for resid-ential space and job accessibility, permitting greater accessibility with the same amount of residential space, or conversely a greater amount of space for the same degree of accessibility. The kinds of choices that are made w i l l influence the length and direction of peripheral work journeys, and the relative weakening of the space and accessibility constraints w i l l make way for other variables, such as recreational opportunities, to exert a greater and more complex influence on residential location. 26 Wolforth's investigation of work/residence separation i n Vancouver reveals that the distorted pattern of worktrips to the central business d i s t r i c t of Vancouver i s largely due to the concentration of employment of particular types at the centre, and the social differentiation of residential areas, relecting historical and topographical influences. In addition, he finds that workers employed i n the suburbs were more l i k e l y to liv e closer to their place of employment than those with downtown jobs. In particular, Wolforth emphasises (p 76) that workers with high imcomes tend to live i n high cost housing while workers with low incomes live i n low cost residential areas, irrespective of the distances of these areas from the place of work. This suggests that the spatial differentiation of different kinds of jobs and different kinds of housing w i l l be important determinants i n the pattern of peripheral work journeys as well as for downtown journeys. A large proportion of the descriptive and explanatory studies of the journey to work, of which the preceding review i s a selection, concentrate primarily on theconmuter journeys to central business d i s t r i c t s , or are 2 ^ John R. Wolforth, Residential Location and the Place of Work, B.C. Geographical Series No. 4 (Vancouver: Tantalus Research Ltd., 1965) - 1 9 -summary studies of a general or abstract nature. The review provides an indication of the state of current knowledge of the journey to work, and of the scope and depth of the studies that are being undertaken. Many of the factors that influence the characteristics of the commuter journey to the central business d i s t r i c t seem l i k e l y to affect the peripheral work journey in a similar or modified manner, and can form the basis of futher testing. However these studies are of l i t t l e assistance i n the more immediate task of documenting and describing the peripheral journey to work, and i n making some simple basic comparisons with the downtown work journey. Only when this kind of work has been done can the wider applications of other studies be evaluated. Studies of the peripheral journey to work 27 Carroll, i n a paper published i n 1 9 5 2 , tests two hypotheses relating to the journey to work: (a) that the homes of persons working i n the central business d i s t r i c t are distributed throughout the c i t y i n a manner similar to the t o t a l population, and (b) that the homes of employees working i n suburban areas are clustered closely round the place of work. Using data from several c i t i e s , notably information for Baltimore for the year 1 9 2 5 , Carroll i s able to substantiate these hypotheses. However, 28 Wolforth's work on Vancouver while endorsing hypothesis (b) shows that, i n the case of Vancouver, hypothesis (a) i s not valid, and that the d i s t r i -bution of downtown employees i s subject to major distorting influences. 27 J. Douglas Carroll Jnr., 'The Relation of Homes to Work Places and the Spatial Pattern of Cities', Social Forces, Vol. 3 0 , No. 3 , (March, 1952), pp 271-282 Wolforth, Op. Cit. - 20 -A study of employment and commuting patterns i n the West Suburban 29 area of Chicago, based on data from the Chicago Area Transportation Study , provides further evidence of the distinctive characteristics of the periph-eral journey to work, compared with downtown commuter journeys. This work confirms the relatively closer association of workplaces and residences for suburban employees suggested by Carroll, and the greater importance of automobile travel for peripheral work trips found i n Philadelphia. The frequency distribution of journeys by t r i p length for the Chicago central business d i s t r i c t approximated a normal curve, but for employees travelling to jobs i n the West Suburban area, the frequency distribution showed a very pronounced positive skew. In addition, i t was found that employment in the suburbs showed a larger proportion of male employees than downtown employment, and a larger proportion of employment i n the categories of craftsmen, operatives, and labourers. Employment in the West Suburban area showed a lower proportion of professional, managerial, c l e r i c a l , and sales employees than the central business d i s t r i c t of Chicago. (See Table IV) Employees working in the West Suburban area were l i k e l y to li v e i n the same area, or i n the same sector of the urban area either closer to the centre or further out towards the periphery. There were few long distance commuter journeys crossing tangentially from one suburb to another. A series of modified gravity models was used to test the effectiveness of these conclusions as a basis for predicting commuting patterns for peripheral employment centres. The simple gravity model gave only a low level of explanation, but the introduction of a frictionless zone four 29 Edward J. Taaffe, Barry J. Garner, and Maurice H. Yeates, The  Peripheral Journey to Work (Evanston, I l l i n o i s : Northwestern University Press, 1963) - 21 -miles i n radius from the place of employment and an element of randomness i n the hypothetical distribution of employees' homes, together with preferential weighting for the adjacent suburban areas provided a greatly improved level of explanation. Further disaggregation showed the influence of sex, race, social class, income group, and alternative employment oppor-tunities on the pattern of suburban work journeys. TABLE IV COMPARISON OF EMPLOYMENT CHARACTERISTICS FOR WEST SUBURBAN AREA AND CENTRAL BUSINESS DISTRICT OF CHICAG-0 Chicago CBD West Suburban Area Mean length of commuter journey 6.7 miles 5.2 miles Proportion of employees travelling to work by car 2U% Proportion of female employees 32$ 23$ Proportion of professional and managerial employees 25$ :19$ Proportion of c l e r i c a l employees 31$ 15$ Proportion of sales employees 8fo Proportion of craftsmen, operative, and labourer employees 29$ Source: Taaffe, Sarner, and Area Transportation Study Yeates, Op. Cit. adapted from Chicago The Chicago study i s an intensive investigation of the peripheral journey to work, and shows both the complexity and distinctiveness of this type of journey within the t o t a l travel pattern. This i s an isolated study i n one cit y , and an additional examination of the peripheral journey to work would provide further information, and would serve to confirm or - 22 -modify the Chicago findings. It was with this objective that a sample study of the journey to work i n Vancouver was undertaken. In addition, i t was hoped that a description of the peripheral journey to work i n Vancouver would complement Wolforth's study relating primarily to downtown Vancouver, and that,taken together, the two studies would provide an overall description and partial explanation of commuting patterns i n the cit y for a particular time period. CHAPTER III THE PERIPHERAL JOURNEY TO WORK IN VANCOUVER Method of investigation A detailed investigation of the peripheral journey to work in Vancouver was undertaken to provide data on journey length and direction. Vancouver forms a particularly suitable c i t y for a study of the peripheral work journey for there i s a generally low density of urban development, and a large measure of dependence on automobile travel. Mass transit f a c i l i t i e s play a relatively minor role i n the t o t a l travel patterns throughout the metropolitan area although bus services are of local importance. Conditions are therefore favourable for a dispersion of employment, at least:in certain categories, into peripheral suburban areas. High car ownership and easy travel in the suburbs should enable places of employment i n the suburbs to attract employees from wide catchment areas, unrestricted by such constraints as the location and frequency of bus services. The d i f f i c u l t i e s of car commuting and parking i n downtown Vancouver are an added incentive for employment decentralisation where possible. On the other hand certain kinds of jobs, involving specialised business linkages, w i l l be firmly tied to the central area. The congestion of t r a f f i c on some of the bridge bottlenecks acts as an additional constraint on long distance car travel i n certain directions. An examination of the peripheral journey to work in Vancouver i s timely - 23 -- 24 -from a planning point of view, for the importance and characteristics of the peripheral work journey have significant implications i n evaluating the potential role of mass transit f a c i l i t i e s i n transportation planning for Vancouver, and the c i t y i s nearing the size when major decisions and invest-ments w i l l be necessary to provide a transportation system that w i l l permit further growth. The information on the journey to work i n Vancouver i s based on a sample of residential areas drawn randomly from detailed land use maps of the metropolitan area^. The sample consists of ten clusters, each made up of several street blocks primarily i n residential use. The Vancouver City  Directory for 1 9 6 3 ^ was used to establish the identity of a l l employed adults l i v i n g i n each cluster from the l i s t s of persons by address. The alphabetical l i s t of persons and their occupations was then used to obtain the name of the employer of each employed person l i v i n g i n the relevant street blocks. Finally, by looking up the name of the employer, i t was possible to identify the location of the workplace of each person. This lengthy cross-reference procedure was the chief limitation on the size of the sample. In approximately 1 7 $ of a l l cases, i t was not possible to pinpoint the location of the workplace, either because no information was given i n the directory, or because the description or nature of the employ-ment prevented the identification of a precise workplace, as i n the case of schoolteachers, where the particular school was not named, or persons such as sales representatives whose work was essentially peripatetic. The ^ For description of sampling method and processing see Appendix The sample was based on 1 9 & 3 data so that direct comparisons could be made with Wolforth's work for the same year. - 25 -TABLE V CHARACTERISTICS OF SAMPLE OF WORK JOURNEYS IN VANCOUVER Residential Cluster 1 2 3 4 5 6 7 8 9 10 Total Number of Commuter Journeys to Central Area of Vancouver 27 15 26 38 9 21 24 19 24 8 211 Commuter Journeys to Central Area , w as $ of a l l W journeys identified 38$ 38$ 46$ 19$ 28$ 26$ 28$ 25$ 11$ 30$ Mean Distance from Central Area Mean 4.4m 3'0m 4'3m 4«6m 4.9m 4.7m 4.3m 4.3m 6.0m 9*2m Length . 4«lm Number of Commuter Journeys to Peri-pheral Workplaces 41 25 42 ' 45 39 54 69 49 71 63 498 Mean Length of Peripheral Com-muter Journeys 3-8m 2.7m 3-8m 3.8m 3.5m 3.2m 3.1m 3«3m 3«9m 4.3m 3.6m Cases where Workplace not Identifiable 15 9 14 13 7 24 21 10 23 7 143 TOTAL EMPLOYED ADULTS 852 Source: Vancouver City Directory, I963 See Figure 3 for the location of residential clusters that make up the sample. workplace was successfully identified for a to t a l of 709 persons, about one 32 third of whom worked i n the central area, of Vancouver . The effective As defined by Vancouver City Engineering Department for Traffic Survey purposes: see Figure 3-- 26 -sample of persons making peripheral journeys to work was 498. This figure i s not large enough to permit complex breakdowns and cross analysis.: , but i s adequate for general conclusions, preliminary analysis, and as an i n d i -cation of the value of the method and approach used. Results and findings Table V shows the major characteristics of the sample, including separate information for the central area and peripheral work journeys. The mean length of commuter journeys to the central area of Vancouver was found to be approximately 15$ longer than the mean length of peripheral work journeys. These figures compare very favourably with Wolforth's sample for the same year (Table VT). There i s , therefore, a degree of clustering of homes closer to suburban workplaces than to downtown employment. A T-test was applied to the data and this showed conclusively that, i n spite of internal variations i n each of the distributions, the two groups of trips are significantly different i n mean length, and that the difference could not have arisen by chance. Hypothesis ( l ) , that peripheral work journeys are l i k e l y to be significantly shorter than downtown commter journeys can therefore be regarded as substantiated. TABLE VI MEAN WORK-RESIDENCE"SEPARATION FOR TWO VANCOUVER SAMPLES Wolforth sample Hickman sample Employed i n Central Area 4 . 0 miles 4*1 miles Employed outside Central Area 3»4 miles 3.6 miles 825 persons 709 persons Source for both samples: Vancouver City Directory, I963 - 2 7 -A comparison of the proportion of persons travelling to the central area and the distance to the central area from each of the ten individual clusters provides further confirmation of the inadequacy of distance alone i n account-ing for the varying dependence on employment in the central area. Figure 1 shows the scatter of the sample clusters i n terms of distance from the central area of Vancouver and the proportion of persons working i n the central area. The line of best f i t by the least squares method has negative slope, and indicates that there i s a general decrease i n dependence on the central area for employment with increasing distance. However the scatter shows a good deal of variation around the line of best f i t , and large residuals remain unexplained. FIGURE 1 DEPENDENCE ON DOWNTOWN EMPLOYMENT BY DISTANCE FROM DOWNTOWN FOR COMPONENT CLUSTERS OF SAMPLE employed persons working in central area of Vancouver slope of line of best f i t calculated by method of least squares y = - 5 . 2 x + 5 7 - 3 Distance of residential clusters i n sample from central business d i s t r i c t of Vancouver (miles) In addition, a chi-squsared test was used to compare the proportion of downtown employees l i v i n g i n the clusters to the west and east of Main Street lying a comparable distance from the central area. This demonstrated - 28 -that the differentials could not have arisen through chance alone, and that other variables must be present. Wolforth's conclusion^ that a dispropor-tionate number of central area employees are drawn from the area of the Point Grey peninsula west of Main Street i s therefore endorsed. A detailed breakdown of commuter trips by distance was prepared, distinguishing downtown and peripheral journeys. The two types of work t r i p show distinct differences i n frequency distribution (Figure 2 ) . While work trips to downtown are clustered f a i r l y closely and symmetrically around the mean figure, peripheral work trips show a much greater variety of t r i p length, forming a low, dispersed frequency curve, with positive skew. FIGURE 2 PERCENTAGE FREQUENCY DISTRIBUTIONS OF PERIPHERAL AMD DOWNTOWN WORK JOURNEYS BY TRIP LENGTH Length of commuter trips i n miles Wolforth, Op. Cit. p 62 - 29 -Despite the lower mean distance for peripheral work journeys, the effect .of the skewed curve i s to produce a marginally higher proportion of longer journeys (over six miles) to peripheral workplaces than to the central area. The differences i n the distributions are reflected i n the resulting standard deviations, shown i n Table VTI. Peripheral work journeys show a larger standard deviation than central area work journeys, and when this i s expressed on a percentage basis as a coefficient of variation, peripheral journeys show about twice as much variation i n t r i p length as downtown journeys. The size and structure of the sample may tend to par t i a l l y reduce the variety of t r i p length exhibited by work trips to the central area, but the great variety of t r i p lengths for peripheral work journeys i s unmistake-able, and the hypothesis that this class of t r i p shows more variation i n length than downtown journeys can be accepted as valid. TABLE 711 VARIATION IN WORK TRIP LENGTH BY PLACE OF EMPLOYMENT Mean work-residence Standard Coefficient Total separation Deviation of variation trips 27$ 211 trips 53$ 498 trips Source: Vancouver City Directory, 1963 In addition to the above analysis of the variation i n t r i p length, the data also confirm that there i s a good deal of variety i n t r i p direction. In most of the ten clusters that make up the sample, one directional quadrant accounts for the majority of peripheral journeys. Normally, this i s i n the Employed i n Central Area - 3 0 -same direction as the central area, reflecting the large number of peripheral job opportunities that are located i n the industrial and service zones i n the inner urban ring surrounding the central area. However i n nearly a l l the clusters, the remaining quadrants account for a substantial share of peripheral work tri p s , usually about 40$. Thus, the peripheral work t r i p , although showing a preponderance of trips i n the general direction of the central area, exhibits a variety of t r i p direction, and there i s a s i g n i f i -cant proportion of these trips that are i n other directional quadrants. The proportion of peripheral trips from each of the ten clusters occuring i n each quadrant i s summarised i n Table VIII. TABLE VIII DIRECTION OF PERIPHERAL WORK JOURNEYS IN VANCOUVER Residential Cluster 1 2 3 4 5 6 7 8 9 1 0 Total Proportion of Peripheral Work Trips by Quadrant Northwest 3 2 $ 1 2 $ 1 7 $ 1 3 $ 5 1 $ 5 2 $ 5 4 $ 2 9 $ 5 5 $ 4 3 $ 3 9 $ Northeast Wo 5 6 $ 6 9 $ 7 0 $ 1 5 $ 9 $ 1 3 $ 0 $ 1 0 $ 3 $ 2 5 $ Southwest 5 $ 0 $ 0 $ 4 $ 2 1 $ 2 6 $ 34$ 5 7 $ 24$ 3 2 $ 2 0 $ Southeast 1 7 $ 3 6 $ 1 4 $ 1 3 $ 1 3 $ 1 3 $ 1 6 $ 14$ 1 1 $ 2 2 $ 1 6 $ Total trips 4 9 8 Source: Vancouver City Directory, I 9 6 3 - 31 -These s t a t i s t i c a l summaries have demonstrated the general variety of t r i p length and t r i p direction of the peripheral work journeys documented i n the sample. To examine this variety i n greater detail, cartographic plots of the documented commuter journeys originating at each residential cluster have been prepared, and these are reproduced i n Figures 4 to 13, relating to the locations shown in key map, Figure 3* A careful inspection of the cartographic plots, and a comparison with the estimated distribution of employment for 1965^" (Figure 14) have resulted i n a number of tentative supplementary conclusions about the spatial distribution of peripheral work journey destinations: (a) Even where there are large numbers of employment opportunities close at hand i n the suburbs, there i s l i t t l e tendency for there to be a large number of short trips to these local destinations. It has not been possible to make a detailed analysis of the location of job opportunities, but the evidence suggests that many nearby opportunities have been ignored. The variety of length and dispersion of peripheral work tr i p s i s evidently a persistent feature, although the number of trips from cluster 1 to the University Endowment Lands i s a notice-able and understandable exception to this generalisation. This i s partly due to the location of the University of British Columbia campus where i t can only be approached from one side. (b) Where there i s clustering of the destinations of peripheral trips into particular suburban work zones, for example the Broadway/Sranville Street area or the Clark Drive/East Hastings Street industrial area, Vancouver City Planning Department, Metropolitan Vancouver, 1955  I965, and 1985, (Vancouver, 19^7), based on estimates, Table 14, p 30. FIGURE 3 KEY MAP SHOWING- LOCATION OF RESIDENTIAL CLUSTERS FORMING WORK JOURNEY SAMPLE, VANCOUVER, 1 9 6 3 FIGURES 4 - 1 3 CARTOGRAPHIC PLOTS OF WORK JOURNEYS ORIGINATING AT. EACH OF TEN RESIDENTIAL CLUSTERS FORMING WORK JOURNEY SAMPLE, VANCOUVER, 1 9 6 3 (unless otherwise indicated, each line represents one work journey) FIGURE 5 : RESIDENTIAL CLUSTER 2 FIGURE 10: RESIDENTIAL CLUSTER 7 FI&URE 14 EMPLOYMENT BY TRAFFIC ZONES, VANCOUVER, 1965 (simplified) - 47 -destinations are s t i l l f a i r l y scattered within the zone, for suburban employment opportunities tend to occur i n elongated linear areas along main roads, railways, and water frontages. (c) The western residential clusters on Point Grey peninsula have a limited range of alternative employment zones within easy access. The dispersion of peripheral work journeys from these clusters i s notice-ably more grouped in pattern than peripheral trips from the eastern clusters where employment opportunities are available in a number of accessible l o c a l i t i e s . This, together with the specialised employment structure discussed by Wblforth, accounts for the greater dependence of these western clusters on downtown employment, and the more random distribution of peripheral work t r i p destinations relating to the eastern clusters. In summary, i t can be said that, on the basis of this sample, trips to peripheral employment in Vancouver tend to be a l i t t l e shorter than trips to downtown employment in terms of mean length, and show a great deal more variety of t r i p length and t r i p direction. Movement by car i s f a i r l y easy in the suburbs i n most directions, and distance does not seem to be a serious constraint on work journey patterns. Thus, the spatial distribu-tion of employment i s not an . important factor i n directly determining journey patterns, and other factors, such as job or residence specialisation, are l i k e l y to be more significant influences. The resulting pattern of peripheral work trips i s a random, criss cross network, not suited to easy generalisation or description, comprising many small zone to zone movements from numerous residential areas to a variety of employment opportunities. It i s l i k e l y that special features of the road system, such as the embryo - 48 -freeway network, and the restricted number of water crossings, attract large t r a f f i c volumes, but otherwise the pattern of peripheral work journeys i s an extensive, multidirectional spread of f a i r l y even t r a f f i c volumes. This t r a f f i c probably makes relatively even use of the main elements of the road system throughout the metropolitan area. CHAPTER IV CONCLUSIONS AND IMPLICATIONS Comparison with other studies The sample of work journeys i n Vancouver i s not large, and i t i s emphasised that preliminary conclusions resulting from i t should be regarded as tentative, and should form the basis for more detailed studies. The data presented i n Chapter III indicates that the peripheral journey to work i n Vancouver i s shorter i n length than commuter trips to the central business d i s t r i c t . It also shows a large degree of dispersion i n terms of t r i p length and direction. The difference i n the mean length of downtown and peripheral work trips i s less than was found i n the West Suburban area: of 35 Chicago , partly because the commuter trips to downtown Vancouver are shorter i n mean length than those to the central business d i s t r i c t of Chicago. This difference i s probably largely due to the difference i n the size of the two c i t i e s : Lapin has noted the increase i n the mean length of the work journey with increasing c i t y s i z e ^ . The mean length of the peripheral journey to work indicated by the Vancouver sample i s almost identical to the median^ length of peripheral work trips found i n the Chicago study. 35 Taaffe, Garner, and Yeates, Op. Cit., p 17 ^ Lapin, Op. Cit., p 44 ^ In the Chicago study, the frequency distribution by distance of peripheral work journeys was highly skewed, and the median i s a more useful and r e a l i s t i c indicator of the length of trips than the mean. - 49 -- 50 -However i n relation to the size of the c i t y , and the mean length of downtown commuter journeys, peripheral work trips i n Vancouver are longer than would have been expected on the basis of the Chicago study. The Vanvouver peripheral work journeys show the same spatial pattern as was found in Chicago, with emphasis on the same sector of the c i t y both closer towards the centre and out towards the periphery, and on the adjacent sectors. In addition, an examination of the cartographic plots of t r i p destinations suggests that peripheral trips are relatively more dispersed i n Vancouver than i n Chicago. There appears to be a higher proportion of trips that are not to destinations i n the same or adjoining sectors, but which cut across two or three sectors, and form tangential suburban trips of considerable length. This more dispersed pattern of peripheral journeys i s expected to J O occur i n Chicago i n the future . It may be that for various reasons, Vancouver i s at a more advanoed stage i n the dispersal process, showing relatively larger and more divergent peripheral work t r i p s . F i r s t , because Vancouver i s a newer c i t y that has grown mainly i n the automobile era, i t shows lower development densities, and the dispersed pattern of travel may be a response to the ease of and dependence on car travel. Second, due to i t s relatively rapid and recent growth, the c i t y shows far less of the radial-concentric ring structure that i s so noticeable i n Chicago. It i s less l i k e l y that t r i p patterns w i l l have been conditioned by a long period of dependence on the central area for employment and on radial routes for transportation. Third, a variety of topography i n Vancouver, due to i t s mountain and coastal site, may have produced more powerful influences on residential location than are present Taafe, Garner, and Yeates, Op. Cit., Figure VI - 2, p 117 - 51 -i n Chicago. The attraction of residential sites i n the west of the city may attract some residents who have to make long compensating work trips to employment i n other parts of the area. The social differentiation of housing, as indicated by the median value of dwellings, i s a basic east/west division, rather than the fine textured mosaic of heterogeneous housing groups typical of older c i t i e s where topography has exerted less influence. In general, the Vancouver findings substantiate the results of the Chicago study of the peripheral journey to work. Suburban work journeys form a distinctive and growing element i n the total urban travel pattern. If the Chicago and Vancouver patterns of peripheral trips also occur on the fringes of London and New York, where very large increases i n employment are pre-dicted, the result w i l l be complex and unusual travel patterns involving very large numbers of t r i p s . Specialised knowledge and techniques w i l l be required to plan for the necessary transportation systems. Choices of residential location for peripheral employment . The Vancouver data have shown that even when persons are employed in the suburbs, and have many employment opportunities available close to the place of residence, the f a i r l y long journey to work i s a persistent feature of travel patterns. Transportation planners have tended to assume that households endeavour to minimise the separation of residence and workplace, subject to constraints of space preferences and purchasing power, especially i n the larger c i t i e s where very long commuting distances are involved. It i s argued that the spatial separation of workplaces and residences i s the result of these factors, together with the time lag i n the adjustment of residences to workplaces. Thus, i f persons tend to change jobs more often - 5 2 -than they change homes, the separation of the home and workplace may vary i n a random manner. If persons change homes more often than they change jobs, i t might be assumed that the new home w i l l be chosen closer to the- place of work. In theory, suburban employment widens the choice of housing and accessibility available to an employee by enabling him to purchase greater space with the same degree of accessibility or greater accessibility. In addition, i t i s d i f f i c u l t to establish for what proportion of the population there i s a genuine 'choice' of housing, and to what extent many groups are constrained by purchasing power and space requirements. However the present data suggest that a peripheral journey to work of four or five miles, probably by car, i s a common undertaking. In at least some of these cases, an element of choice i n housing location must exist, and journeys of this length are apparently not regarded as overly arduous. Some persons choose to continue to make journeys of this length even when they are employed i n the suburbs and many work opportunities are available close at hand. These individuals appear to prefer to realise the benefits of suburban employment i n terms of a wider and more specialised choice of residential location rather than i n minimising the journey to work. Benefits of residential location such as space, amenity, and social prestige may be emphasised even more in the future as the amount of leisure time increases. Conversely, the drawbacks of a long journey to work may be reduced as transportation improvements are made, and when, eventually, the working week f a l l s from five to four days. City development plans should therefore ensure that aspirations concerning the style and location of residence can be achieved, rather than attempting to sacrifice residence advantages so that supposed transportation problems can be minimised. - 53 -Applicability of gravity model to peripheral work journeys The gravity model assumes that the f r i c t i o n of travel time or distance i s minimised, but i t has been shown that i n the case of suburban employment, other considerations besides the length of the work t r i p appear to influence residential location. Thus i t has been found necessary to introduce such devices as frictionless zones and Monte Carlo distribution methods in other studies. To a certain extent, the distorting effect of other locational factors can be reduced by refined adjustments to the f r i c t i o n factors applied to each zone to zone component in the gravity model. These adjust-ments must be based on detailed survey material, and become increasingly cumbersome. In addition, these rigorous survey requirements restri c t the usefulness of the gravity model when attempting to predict zone to zone travel volumes i n new urban areas that do not yet exist, or i n areas where a great deal of change i s expected. With the growing significance and complexity of the peripheral journey to work, the simple gravity model w i l l be less and less able to provide an adequate prediction of travel patterns. In i t s place, a series of more sophisticated explanatory models w i l l be required. These must be based on a more thorough knowledge of the motiv-ations that underlie t r i p making so that assumptions and predictions w i l l be based on structural relationships rather than on measures of association and other more complex empirical methods. Implications for transportation planning The combination of three factors, extensive low density residential development, a wide and spatially scattered range of job opportunities, and complex residential aspirations, has produced in Vancouver a dispersed - 54 -pattern of peripheral work tr i p s . Apart from the freeways and bridges, there appears to be a relatively even t i d a l spread of t r a f f i c onto a l l the main components of the road system. Nearly a l l zone to zone travel desires are probably of modest dimensions, and i t i s only where large concentrations occur that high corridor volumes are accumulated. This dispersed pattern of small flows of trips to peripheral destinations does not appear to be 39 suited to high capacity urban travel modes such as mass transit r a i l systems , for both the number of t r i p origins i n any residential zone and t r i p destin-ations i n any employment zone i s low. If the present pattern continues, and certain peripheral zones emerge as large concentrations of employment, the flows to these zones can best be served by a flexible mediumvolume transit system, such as buses. A greater degree of residential and employment dispersion, producing a more random multidirectional flow of work trips to peripheral employment would result i n a more efficient use of the road system as t i d a l inefficiencies would be reduced. On the other hand, i f i t i s thought desirable that there should be large concentrations of jobs, or high density 'urban' centres, or the provision of a public transit system for other reasons, the most economical arrangement would be to have a medium or high capacity mass transit system of some kind, with large concentrations of t r i p attractions, such as employment nodes or shopping centres, at intervals along the route, set i n a corridor of f a i r l y high density residential development. Clearly a 'concentrated' transportation system would be an inefficient 39 These statements do not apply to the central buiness d i s t r i c t of Vancouver or other c i t i e s : transportation f a c i l i t i e s for these areas must be adjusted to the volume, concentration, and growth prospects of downtown employment. - 55 -way of serving a 'dispersed' residential or employment pattern, and a 'dispersed' transportation pattern would be an equally inefficient method of serving a 'concentrated' employment pattern. It i s important that the kind of employment/residential density pattern developed i s appropriate to the c i t y , and that the transportation pattern i s appropriate to the employment/ residential pattern. Both of these requirements involve an element of foresight, planning, and control. A system of policy formulation and implementation i s required to ensure that the distribution of employment and residences, and the transport-ation f a c i l i t i e s that link them are consistent with the long term require-ments of the community. At present, the pace and location of employment growth i n the Vancouver area i s not related to the planning of transportation f a c i l i t i e s , nor e x p l i c i t l y to an overall view of the future form and function of the ci t y . Present trends i n the distribution of residential and employment growth, and i n transportation planning appear to be partly inconsistent with the Vancouver City Council's emphasis on the long term importance and growth of the downtown area. The provision of transportation f a c i l i t i e s to serve the commuter flows to the downtown area must be related to the volume and concentration of trips attracted to the area, and must reflect the long term intentions of the community. In certain circumstances, the absence of mass transit f a c i l i t i e s to serve the downtown area may act as a very real constraint on the growth of the entire metropolitan area as i t may be impossible to accommodate growth of certain kinds of employment for which a downtown location i s essential. On the other hand, much of the new employment growth may not demand :. the close business linkages and face to face contact that characterises - 5 6 -central area employment, or can substitute telephone contact i n i t s place. In this case, the restriction of suburban employment, and the forcing of an unrealistic proportion of businesses to locate within or adjacent to the central area w i l l perhaps impose unnecessary building, rental, servicing, and transportation costs on the community as a whole. Yet i f the growth of suburban employment continues on an unplanned basis, and very large concen-trations of jobs build up i n particularly favourable peripheral locations, fresh transportation problems w i l l be created which would perhaps be more d i f f i c u l t to cope with than i f the growth were in the central business d i s t r i c t . The study of the peripheral journey to work i n Vancouver has confirmed the Chicago work. The commuter journey between suburban homes and suburban employment i s a large and growing element i n the total travel pattern of the city. These trips show a variety of t r i p length and t r i p direction, and produce a dispersed, multidirectional network of peak hour t r i p s , usually by car. This system of many low zone to zone flows in many directions i s d i f f i c u l t to serve by mass transit f a c i l i t i e s , and this i n turn impedes the growth of large job concentrations, either at central or peripheral locations. If present unplanned trends continue, the distinctive transportation require-ments of the peripheral work journey w i l l impose themselves on the structure of the c i t y , and, for better or for worse, w i l l be a powerful influence on the form and function of the city. These trends are not necessarily consistent with the most beneficial long term growth of Vancouver, or with the requirements of the citizens. Specific and explicit policies are required: f i r s t , to identify the broad form and functions that the c i t y should f u l f i l l ; second, to work out - 57 -the pattern of residential and employment distribution that these functions demand; and third, to devise a transportation system that w i l l adequately serve these distributions. It i s one of the key responsibilities of c i t y government and the planning profession to formulate such policies so that the f u l l development potential of the c i t y i s realised, and so that the aspirations of citizens can be achieved. - 5 8 -SELECTED BIBLIOGRAPHY Alonso, William. 1 Theory of the Urban Land Use Market', Papers and Proceed- ings of the Regional Science Association, Vol. VI, i960, pp 1 4 9 - 1 5 7 . Bevis, Howard W. 'A Model for Predicting Urban Travel Patterns', Journal  of the American Institute of Planners, Vol. 2 5 , No. 2 , May, 1 9 5 9 , PP 8 7 - 8 9 -Carroll, J Douglas, Jnr. 'The Relation of Homes to Work Places and the Spatial Pattern of Cities', Social Forces, Vol. 3 0 , No. 3 , March, 1 9 5 2 , pp 271-282. Curran, Frank B, and Stegmeier, Joseph T. 'Traffic Patterns i n F i f t y Cities', Public Roads - A Journal of Highway Research, Vol. 3 0 , No. 5 , December, 1 9 5 8 . Duncan, Beverly. 'Factors i n Work-Residence Separation: Wage and Salary Workers, Chicago, 1 9 5 1 , American Sociological Review, Vol. 21, February, 1 9 5 6 , pp 48-56. Duncan, Beverly, and Duncan, Otis Dudley. 'The Measurement of Intra-City Locational and Residential Patterns', Journal of Regional Science, Vol. 2 , No. 2 , F a l l , i960, pp 3 7 - 5 4 -Hansen, Walter G. 'How Accessibility Shapes Land Use', Journal of the American Institute of Planners, Vol. 2 5 , No. 2 , May, 1 9 5 9 » PP 7 3 - 7 6 . Herbert, John D, and Stevens, Benjamin H. 'A Model for the Distribution of Residential Activity i n Urban Areas', Journal of Regional Science, Vol. 2 , No. 2 , F a l l , i960, pp 2 1 - 3 6 . Hoover, Edgar M, and Vernon, Raymond. Anatomy of a Metropolis, Harvard University Press, 1 9 5 9 * Kain, John F. 'The Journey to Work as a Determinant of Residential Location', Paper read before the meeting of the Regional Science Association, December, I96I. Lapin, Howard S. Structuring the Journey to Work, Philadelphia: University of Pennsylvania Press, 1 9 6 4 . Loewenstein, Louis K. The Location of Residencesr.and Work Places i n Urban  Areas, New York: The Scarecrow Press, I965. Park, Robert E. (Ed.). The City, Chicago, 1 9 2 5 -- 5 9 -Rasmussen, Steen E i l e r . London, The Unique City, London: Jonathan Cape, 1 9 3 . 7 . Taaffe, Edward J, Garner, Barry J, and Yeates, Maurice H. The Peripheral Journey to Work, Evanston, I l l i n o i s : Northwestern University Press, 1 9 6 3 . Vancouver City Planning Department. Metropolitan Vancouver, 1955, 1965. and 1 9 8 5 , Vancouver: Vancouver City Council, 1 9 6 7 . ' Wingo, Lowdon, Jnr.' Transportation and Urban Land, Washington: Resources for the Future Inc., 1 9 6 1 . Wolforth, John R. Residential Location and the Place of Work, (B.C. Geographical Series No. 4), Vancouver:Tantalus Research Ltd., 1 9 6 5 . APPENDIX SAMPLING- METHOD AND PROCEDURE The objective was to draw a random sample of employed persons l i v i n g i n the suburban areas of the City of Vancouver and the Municipality of Burnaby in order to document work trips to suburban employment. It was decided to exclude the West End residential area from the sampling frame as i t was close to the central business d i s t r i c t of Vancouver, and previous studies indicated that only a small proportion of West End residents could be expected to work in the suburbs. With this exception, a l l areas primarily i n resident-i a l use within the study area were included i n the sampling frame. A cluster sample was used so that the process of documentation would be simplified, and so that a picture of commuting patterns at each of the clusters could be b u i l t up. A detailed land use map of the study area was used as an i n i t i a l sampling frame, and a l l grid squares primarily i n residential use were numbered. A l l grid squares with more than 2 0 $ of their land area i n non-residential use were excluded. Using random numbers tables, ten residential grid squares were identified. Each street block mainly i n residential use within these squares was numbered, and one street block i n each of the ten squares identified by random number methods. In each square, this street block was used as the origin to assemble a, group of four or six residential street blocks clustered as closely as possible round the origin block. -The number of street blocks used depended on the density of residential development. - 6 0 -- 6 1 -It was hoped to obtain approximately 1 0 0 employed persons i n each of the ten clusters. A sample of about 1 0 0 0 employed persons would represent a l i t t l e under 1 $ of the t o t a l labour force l i v i n g i n the study area at that time. A sample of this size, after loss through non-response, i s too small for detailed disaggregation, but provides adequate figures suitable for broad generalisations. Having obtained the addresses of the dwellings that formed the basis of the sample, the employed residents at these dwellings were identified , 4 0 from the street section of the Vancouver City Directory for 1 9 6 3 • The alphabetical index of persons and employers i n the directory was then used to obtain the name of each person's employer. The: alphabetical index also provided the address of the workplace under the name of the employer. In approximately 1 7 $ of a l l cases, i t was not possible to locate the workplace, either because no information was given i n the directory or because the description or nature of the occupation did not permit the identification of a precise workplace. The workplace was successfully identified for a, total of 7 0 9 persons, of whom 2 1 1 worked in the central area of Vancouver as defined by the City of Vancouver Engineering Department. The remaining 4 9 8 persons worked i n locations outside the central area, and formed the sample for the analysis of the peripheral journey to work. The locations of a l l residences and workplaces were plotted on a series of maps, and the airline distance for each work t r i p was measured. A duration of journey measure would have been Data for 1 9 6 3 were used so that the study would be directly comparable to a study of the journey to work to the central area of Vancouver based on material for the same year. - 62 -more desirable, but i t was not possible i n the time available to measure peak hour travel times i n a l l directions from each of the ten residential clusters. The characteristics of downtown work journeys were then compared with those of peripheral work journeys. 

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}]}"
                            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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0104409/manifest

Comment

Related Items