Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The North American skyway systems : a post implementation evaluation Pati, Debajyoti 1991

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

Item Metadata

Download

Media
831-UBC_1991_A7_4 P37.pdf [ 23.38MB ]
Metadata
JSON: 831-1.0098624.json
JSON-LD: 831-1.0098624-ld.json
RDF/XML (Pretty): 831-1.0098624-rdf.xml
RDF/JSON: 831-1.0098624-rdf.json
Turtle: 831-1.0098624-turtle.txt
N-Triples: 831-1.0098624-rdf-ntriples.txt
Original Record: 831-1.0098624-source.json
Full Text
831-1.0098624-fulltext.txt
Citation
831-1.0098624.ris

Full Text

THE NORTH AMERICAN SKYWAY SYSTEMS: A POST IMPLEMENTATION EVALUATION by DEBAJYOTI PATI B.Arch., University of Bombay, India, 1987 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ADVANCED STUDIES IN ARCHITECTURE in THE FACULTY OF GRADUATE STUDIES (School of Architecture) We accept this thesis as confirming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1991 (c) Debajyoti Pati, 1991 in presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it 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 department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of CM I V? The University of British Columbia Vancouver, Canada Date I 6TH m \ DE-6 (2/88) 11 ABSTRACT Towards the beginning of the twentieth century technological innovations in the transportation sector generated strong centrifugal forces in the settlement pattern of North American urban agglomerations. The birth of the electric Street Cars led to rapid suburbanization of cities. The subsequent invention of private Automobiles resulted in a quantum jump in intra-city travel, and associated problems involving traffic congestion, atmospheric pollution, and accident rates. A rapid deterioration in the vitality of Central City areas followed. Downtowns across the continent gradually lost their position as the principal commercial, entertainment, cultural, and retail center of large tracks of residential areas. While cities were experimenting with hosts of ideas and strategies to rejuvenate the central city vitality, a revolutionary idea evolved. City planners and architects, through the participation of the public and business groups, linked extensive areas of downtown through bridges and corridors, which got referred to as the Skyway System. Through the development of the system, cities hoped to solve an array of downtown problems involving transportation, urban redevelopment, and commerce. The system seemed to work well. However, most of the evaluations were conducted within the realms of transportation planning, and little information was available regarding the system's performance as a public place and a downtown attraction, in the realms of urban redevelopment. A survey conducted in 1988 hinted at the possibility of the system not performing as an urban public place. But without any conclusive evidence the question remained unanswered. As an initial step towards evaluating the system's performance in areas of urban development, the system was compared with another contemporaneous strategy; the pedestrian mall which was implemented with identical objectives, and the conventional sidewalks on city streets which have been a part of the urban landscape from the I l l beginning of human civilization. Two areas were emphasized in the evaluation process. The first area pertained to the usage pattern, and the user population of the facilities. And the second area involved an examination of the physical and behavioral set up of the environment with an intention of establishing, wherever possible, a relationship between the similarities/differences in the physical environment and corresponding similarities/differences in usage pattern. Data on user activity was recorded through simple observation on behavioral data recording sheets, on weekdays and weekends on all the facilities. The survey was conducted on the Skyway System, and Nicollet Mall in downtown Minneapolis, over a period of twenty days during summer 1990. Statistical analysis of the collected data demonstrated a striking similarity in usage population and usage pattern between the Skyway System and the Nicollet Mall. But, the Skyway System and the conventional street had many differences. Together, the Skyway System and the mall, represented a marked progress in the design of urban public spaces, from the conventional sidewalks. However, many problems persisted on the skyway segments, the foremost being the acute lack of furnishing elements on the system, and improper spatial configuration of those upper level areas. Besides, certain degree of change to the management strategy of skyway spaces proved essential. The thesis concludes with a few suggestions concerning modification to the physical environment, and the management strategy in order to extract maximum benefit out of the system development. IV TABLE OF CONTENTS Abstract ii Table of Contents iv List of Tables vi List of Figures ix Acknowledgement xi Chapter One: Grade-separated Pedestrian Facilities : Origin and Development; Socioeconomic and Political Factors 1 Discarding the Automobile 1 Grade Separated Pedestrian Systems Prior to Twentieth Century 3 Utopian Thoughts of Early Twentieth Century 18 Development of Integrated, Grade Separated Pedestrian Networks 25 Pedestrianization and the Urban Activity System 28 Chapter Two: The Skyway Systems in North America 33 Twentieth Century Cities: the Transition of City Centers 33 The Urban Redevelopment Programs 34 Revitalization Strategies during the 70's and 80's 35 Pedestrianization Strategies in North America: Objectives and Impacts 38 Grade-separated Pedestrian Systems as Urban Revitalization Strategy 44 Forms of Implementation 47 Elevated Pedestrian Networks in the Realm of Overall CBD Pedestrianization 50 Issues and Controversies 55 Skyway System and System Usage 64 V Chapter Three: Laying Down the Ground Rules 65 Research Endeavors During the Past Two Decades 65 The question 77 The Scope of the Study 78 Survey Method and Instrument 87 The city, the Survey Points, and the Users 96 Chapter Four: The Skyway, the Mall, and the Street 103 Downtown Minneapolis: A Bird's Eye View 103 The People Downtown 109 Activities on the Facilities 144 Differing Usage 170 Chapter Five: City Core and the Perceptual Scenario 175 Starting Out 175 The Urban Forms and Social Activities 177 A Better Environment 218 A public Place ? 222 Going Downtown 224 Bibliography 228 Appendix I: Behavioral Data Recording Sheets 236 Appendix II: Chapter Four Tables 238 Appendix III: Bar Charts 287 vi LIST O F T A B L E S Table 3.1 Summary of skyway establishments 67 Table 3.2 Distribution of Skyway establishments 68 Table 3.3 Types of Skyway establishments 68 Table 3.4 Percentage distribution of Person Trips by purpose 74 Table 3.5 Perceived profiles of Skywalk users 74 Table 3.6 Monthly usage of skyway versus at-grade pedestrian crossing 76 Table 3.7 Final survey schedule 100 Table 4.1 Crosstabulation of Age by Site for weekdays 114 Table 4.2 Crosstabulation of Age by Site for weekend 115 Table 4.3 Pairwise comparison of facilities for Age distribution 116 Table 4.4 Mean ranking of Age groups during weekdays 116 Table 4.5 Crosstabulation of Gender by Site for weekdays 121 Table 4.6 Crosstabulation of Gender by Site for Weekend 122 Table 4.7 Mean ranking of Gender groups during weekdays 123 Table 4.8 Mean ranking of Gender groups during weekend 123 Table 4.9 Pairwise comparison of facilities for Gender distribution 124 Table 4.10 Crosstabulation of Gender by Role on Mall(F) for weekdays 124 Table 4.11 Crosstabulation of Purpose by Gender on Mall(F) for weekdays 125 Table 4.12 Crosstabulation of Role by Gender on Mall(R) for weekdays 125 Table 4.13 Crosstabulation of Purpose by Gender on Mall(R) for weekdays 126 Table 4.14 Crosstabulation of Gender by Role on Skyways for weekdays 126 Table 4.15 Crosstabulation of Purpose by Gender on Skyways for weekdays 127 Table 4.16 Crosstabulation of Role by Site for weekdays 131 Table 4.17 Crosstabulation of Role by Site for weekend 132 vii Table 4.18 Pairwise comparison of facilities for Role distribution 133 Table 4.19 Mean ranking of Role groups on the facilities during weekdays 133 Table 4.20 Mean ranking of Role groups on the facilities during weekend 134 Table 4.21 Crosstabulation of Purpose by Role for Marquette Ave. for weekday 134 Table 4.22 Crosstabulation of Grouping by Site for weekdays 139 Table 4.23 Crosstabulation of Grouping by Site for weekend 140 Table 4.24 Pairwise comparison of facilities for Grouping distribution 141 Table 4.25 Mean ranking of Grouping types on the facilities during weekdays 141 Table 4.26 Mean ranking of Grouping types on the facilities during weekend 142 o Table 4.27 Crosstabulation of Grouping by Role for Marq. Ave. for weekday 142 Table 4.28 Crosstabulation of Grouping by Purpose for Workers on weekdays 143 Table 4.29 Crosstabulation of Grouping by Site for Workers observed Shopping 143 Tables4.30 Crosstabulation of Size by Site for weekdays 148 Table 4.31 Crosstabulation of Size by Site for weekend 149 Table 4.32 Crosstabulation of Purpose by Site for weekdays 150 Table 4.33 Crosstabulation of Purpose by Site for weekend 151 Table 4.34 Pairwise comparison of the facilities for Purpose distribution 152 Table 4.35 Crosstabulation of Purpose by Age for all facilities, ignoring day 152 Table 4.36 Crosstabulation of Purpose by Role for all facilities, ignoring day 153 Table 4.37 Crosstabulation of Purpose by Role for Marq. Ave. for weekdays 153 Table 4.38 Crosstabulation of Activity by Site for weekdays 158 Table 4.39 Crosstabulation of Activity by Site for weekend 159 Table 4.40 Pairwise comparison of the facilities for Activity distribution 160 Table 4.41 Mean ranking of Activity types on the facilities during weekdays 160 Table 4.42 Mean ranking of Activity types on the facilities during weekend 161 Table 4.43 Crosstabulation of Activity by Role for Skyway System for weekday 161 Table 4.44 Crosstabulation of Activity by Posture on facilities, ignoring day 162 V l l l Table 4.45 Crosstabulation of Activity by Role for Marq. Ave. for weekdays 163 Table 4.46 Crosstabulation of Role by Grouping for Skyway for weekdays 164 Table 4.47 Crosstabulation of Role by Grouping for Skyway for weekend 164 Table 4.48 Crosstabulation of Activity by Role for Skyway for weekend 165 Table 4.49 Crosstabulation of Posture by Site for weekdays 167 Table 4.50 Crosstabulation of Posture by Site for weekend 168 Table 4.51 Pairwise comparison of the facilities for Posture distribution 169 Table 4.52 Crosstabulation of Activity by Role for Marq. Ave. for weekday 169 Table 5.1 Temporal distribution of users by activity on Skyway on weekdays 185 Table 5.2 Temporal distribution of users by Role on Skyway during weekdays 188 Table 5.3 Temporal distribution of users by Role on Nicollet Mall on weekdays 189 Table 5.4 Temporal distribution of users by Purpose on Skyway on weekdays 197 IX LIST OF FIGURES Figure 1.1 The elevated street system at Camacchio 5 Figure 1.2 Brisighella's Via Degli Asini 6 Figure 1.3 Leonardo's proposed grade-separated street system at Vigevano 10 Figure 1.4 Genoa's Ponte Carignano 11 Figure 1.5 Perugia's Via Appia 12 Figure 1.6 Via Dell' acqedotto at Perugia 13 Figure 1.7 Royal bridge in Venice 14 Figure 1.8 The Loew Bridge. New York 16 Figure 1.9 The Brooklyn Bridge, New York 17 Figure 1.10 Ferriss's grade-separated pedestrian networks 20 Figure 1.11 Multilevel automobile roadways in Ferriss's predictions 20 Figure 1.12 Sketch view of "Contemporary City" 22 Figure 1.13 Grade-separated traffic in the "Radiant City" 23 Figure 1.14 Yona Friedman's "Spatial City" 24 Figure 1.15 Jellicoe's "Motopia" 25 Figure 1.16 Multilevel traffic system in British New Towns 27 Figure 1.17 Pedestrian/car and bus segregation at Evry 29 Figure 1.18 Deck level access to residential buildings at Le Vaudreuil 29 Figure 3.1 Skywalk system usage during weekdays 70 Figure 3.2 Percent using skyway on a typical weekday 70 Figure 3.3 Primary trip purpose of Skyway users 75 Figure 3.4 Sex profile of Skyway users 75 Figure 3.5 Race profile of Skyway users 75 Figure 3.6 Map of downtown Minneapolis core area 101 Figure 4.1 Proposed restructuring of Minneapolis in the M E T R O 2000 P L A N 108 Figure 5.1 Skyway segment inside the IDS Center atrium lobby 178 Figure 5.2 View of skyway segment located deep inside buildings 179 Figure 5.3 View of skyway segments located on periphery of buildings 179 Figure 5.4 Free standing skyway corridors in downtown Minneapolis 180 Figure 5.5 Skyway segments with retail establishments defining the edges 180 Figure 5.6 Skyway segment through the Dayton's in downtown Minneapolis 181 Figure 5.7 Skyway corridor inside the North Star Building 181 Figure 5.8 Skyway gallery inside the IDS Center overlooking the atrium lobby 183 Figure 5.9 Skyway gallery inside City Center building: festive environment 183 Figure 5.10 Skyway segment comparison for Stationary Activity types 186 Figure 5.11 Proportional distribution of Stationary Activities on skyways 186 Figure 5.12 Skyway segments with transparent edges on building periphery 190 Figure 5.13 Temporal distribution of users by Role on Skyways on weekdays 190 Figure 5.14 Temporal distribution of users by Role on Nicollet Mall on weekdays 191 Figure 5.15 Workers on the Skyway System 191 Figure 5.16 Clustering effect of furnishing elements on Nicollet mall 193 Figure 5.17 Stationary activity in the core during weekdays 195 Figure 5.18 Skyway bridge as space dividing elements 195 Figure 5.19 Cross-sectional organization of Nicollet Mall 196 Figure 5.20 Comparison of skyway segments by Trip Purpose 196 Figure 5.21 Facade of City Center building on Nicollet Mall 203 Figure 5.22 Stationary user distribution on Mall(F) North sidewalk 212 Figure 5.23 Stationary user distribution on Mall(F) North sidewalk 212 Figure 5.24 Distribution of stationary users on Marquette Ave. South sidewalk 213 Figure 5.25 Temporal distribution of users by Stationary Activity On Mall(F) 213 Figure 5.26 Stationary user distribution on Mall(F) South pavement 214 XI ACKNOWLEDGEMENT I wish to express my sincerest gratitude to Professor Dino Rapanos, my thesis advisor, for his invaluable guidance and unfailing enthusiasm during the course of my study. I also wish to acknowledge the support provided by the Management of IDS Center, City Center and North Star Center, Minneapolis; by John Cummings in Minneapolis City Planning Dept.; by Tom Duffee, Downtown Council of Minneapolis; by the 1989-90 class of UBC students in the School of Planning for their contribution to Chapter Four of this thesis; by Anthony Parker; and by Prof. Henry Hightower of UBC Planning School in the data analysis. Lastly, I wish to acknowledge the continuous moral support offered by my parents during my stay in Canada. 1 CHAPTER ONE GRADE-SEPARATED PEDESTRIAN FACILITIES : ORIGIN AND  DEVELOPMENT ; SOCIOECONOMIC AND POLITICAL FACTORS. 1.1.0 Discarding the Automobile During the mid-twentieth century a radical transformation was occurring to the central areas of several North American cities. Numerous downtowns across the continent were falling back on the provision of pedestrian mode of transport for inter-city travel. Large tracks of city roads were gradually cut off from automobile traffic, and redesigned for the walking public. Streets were re-paved, sidewalks widened, street side retail areas encouraged, and many amenities and facilities from seats to dancing fountains introduced to make the outdoor public areas more welcoming to the city dwellers. Academicians as well as the masses proudly referred to it as the pedestrianization of American cities. Somewhere in this process, of physical restructuring of the public domain, came a revolutionary concept; the development of the Skyway System. The system, for the first time in the post-modern era, offered the urban pedestrian a choice of walking either indoor or outdoor. Initially developed as small scale commercial ventures, it steadily grew, through one of the unprecedented demonstrations of public-private cooperation, to cover extensive areas of downtown development. While the urban transportation planners were busy demonstrating the success of the concepts in areas of traffic congestion, pollution, and road safety, social scientists were carefully monitoring another critical area of impact of the new found development tool; their catalytic action in areas of social interaction. The 2 key word to the ensuing debate was success. The last two decades witnessed conflicting opinion regarding the success of the facilities. An assessment of that nature requires a well defined basis for judgement, and to create a base for evaluation it would be improper to ignore the numerous instances in history where special off-grade facilities for pedestrians existed. The first few grade-separated pedestrian facilities were built centuries ago during the Middle Ages in Europe, and since then European cities have continued to favour the idea of providing special facilities for the walking public. Development of grade-separated pedestrian networks during the Middle Ages and down to the beginning of the twentieth century came about owing to socioeconomic and political factors which have varied significantly with the context. Not only were the forms of these older pedestrian systems different from their counterparts in the twentieth century but also, in every event of their occurrence in the urban landscape, they germinated out of widely varying social conditions, catering to different social values, and aimed at solving problems typical to their own period of development. The implementation of the proposals and the comparative success or failure of each in achieving its objectives was greatly influenced by a number of contextual variables. Formation of a basis for evaluation warrants a close scrutiny of those variables, and an assessment of their influence in the current socioeconomic environment. A sequential enumeration of the development of such systems in history starts from the -war ravaged periods of medieval Europe, where transportation efficiency seem to have had influenced such urban development strategies. 3 1.2.0 Grade Separated Pedestrian Systems Prior to Twentieth Century 1.2.1 MEDIEVAL PERIOD (9TH TO 14TH CENTURY) : DEVELOPMENTS IN BRISIGHELLA AND CAMACCHIO The history of medieval Europe is one of war and ravage. Beginning with the tenth century, feudal lords built themselves castles on hills outside the towns and cities, and commanded considerable power over the city trading routes. They were powerful enough to virtually control the entire economy of the nearby cities. That was the primary factor leading to the growth and development of Traders Guilds, which slowly gathered power and civic influence inside city limits. Gradually, cities became independent centers of power, generally grouped together, which fueled the centuries long war and strife, devastating the social life of the people. Defence became the primary issue and town design and the internal settlement pattern was dictated by the design of city walls, ramparts, moats and the controlled, protected entrance gatewaj'S. With such heavy investments on defence, it became physically difficult to increase the area of the cities. Fear of enemy warriors and ravagers and to some extent the commercial attractiveness of the cities led scores of people, living in smaller urban or rural settlements, into the protected areas of larger cities. As the population increased, the accepted method was to increase the number of cities, by building new towns for the migrating population. In the perpetual struggle for survival, social activities receded to unprecedented limits. The role of piazzas and squares in city life reduced drastically as compared to the corresponding role of agoras and forums of the preceding eras. Considerations for security and defence, manifested in the development of narrow and tortuous street patterns, hemmed in by walls. The resulting difficulty in maintaining and cleaning the streets made life severely unhygienic. The more powerful citizens lived in good houses, pushing the common man and workers to oppressed and starved living conditions. Social conditions of low and middle 4 class localities were rendered with disease, poverty, and a filthy environment. Sanitary neglect led to large scale epidemics and untold misery (Hiorns, 1956). As the number of new towns grew, the Europeans, and more specifically the Italians demonstrated tremendous ingenuity in designing their new towns. This was the phase in history when Europe experienced the maximum growth of urban areas, and the planners of these towns exploited all the advantages of the topographical features of land, especially in the coastal areas and the hill towns (Hiorns, 1956), which in many instances led to the development of elevated street patterns. One of the systems was built in Camacchio (Fig. 1.1), a small urban area fifty miles south of Venice, where a widespread network of elevated streets was built, probably to separate the service routes from the pedestrian circulation routes. The waterfront location of the city must have helped in developing large scale network of canals, thus securing transport and amenity advantages. The canal system and the road system were in the form of overlapping grids with the road system at the upper level. The canals seem to have served as an efficient conduit for goods transport throughout the city, and wide pavements along the canals, serving as market areas, pulled commercial activities down to the canal level. The upper level were most likely reserved exclusively for the pedestrians. The two levels were connected by stairs at intervals. Due to the low level of social activities during that time, the creation of these network of elevated streets for social interaction could be a highly improbable interpretation. Moreover, there was probably no serious transport problem, and the land area of the cities were not large enough to necessitate any non-pedestrian mode of inter-city travel. Most of the streets were generally narrow, thus restricting the passage of vehicular traffic. It may be safer to infer that the elevated street system in Camacchio were chiefly designed for transportation service efficiency. Fig.1.1: The elevated street system at Camacchio Source: Rudofsky, Streets for People,, 1969, p.189 Camacchio was not the only instance. Similar reasons, but entirely different contextual situations led to the construction of service streets at the second floor level at Brisighella (Fig. 1.2), a hill town, leaving the lower levels for the pedestrians. In Brisighella's Via Degli Asini, the upper level streets run at the second floor level, linking each house on the street. These elevated streets were an integral part of the building form, quite inconspicuous to the lower level public areas. The lower level streets were lined with shops and business establishments where as the upper level streets were used for carrying grains on asses to the mills (Rudofsky, 1969). Here too, transport efficiency seem to have been the motivating factor. 6 Fig. 1.2: Brisighella's Via Degli Asini. Source: Rudifsky, Streets for People, 1969, p.195 1.2.2 EARLY RENAISSANCE (15TH TO 17TH CENTURY) : SYSTEMS AT VIGEVANO, GENOA AND PERUGIA The on-set of the Renaissance in Europe in the fifteenth and sixteenth centuries encouraged radical transformation in the cultural, social and religious facets of urban life. A new conception of the essential values of nature and history resulted in the foundation of a new class of humanistic culture. The City started to be regarded as the heart of an organized society. This view influenced the urban transformation brought about by social, economic and political needs of the period. Cities changed from being mere economic entities to political forces. The old urban middle class split and a new special class of artists, close to the political and religious leaders, emerged, leading to an acceleration in 7 the level and magnitude of patronage to art and architecture. The incidents of warfare declined and the urban residents began to turn their attention towards social activities (Argan, 1969). The revival of Vitruvian theory of forms and proportions in architecture and its extension to the design of urban areas had a profound influence on the design and modification of streets and squares. Urban areas begun to be conceived as stages for human activities. The end of four centuries of war and famine diverted the attention of the urban residents towards serious restructuring of their cities. Piazzas and squares were re-instated in large numbers and with more detailed design and architectural features. They dominated the internal road planning strategies and were the focal point of socioeconomic activities. The exterior spaces of the city were conceived as an 'interior', and, more precisely, as the interior of a palace in which the squares were the rooms and the streets were the corridors and staircases. Cities were modified and founded, based on the principle that the perfect architectural and urban form corresponds to the perfection of the city's political and social arrangements. Considerable effort was expended to provide uniqueness to cities through urban form and architecture. The position of building in the urban spatial fabric and the creation of the correct perspective became essential to create the appropriate kind of physical environment. Street network pattern and design was modified considerably to raise the architectural tone of urban areas. Streets, open spaces and buildings became interlinked in a unified spatial configuration which provided the city with the platform for social action with all its uniqueness (Hiorns, 1956; Zucker, 1959). Possibly streets were elevated at a few locations inside the city, during that period, to provide the city residents with vantage viewing points to relish the architectural splendors. In spite of the radical restructuring of physical environment, the social and sanitary conditions were not very different from the previous eras. Even as late as the eighteenth century, the lavish living conditions of the upper strata of city residents coexisted with the poverty, filth, disease and insanitary conditions of the lower class 8 localities. Plague and epidemic were frequent and the housing quarters and streets were damp, dirty and crowded. The luxurious life of the rich contrasted with the privation and discomfort of the middle class and the extreme misery of the poor. An important fact befitting consideration was that the development and urban restructuring programs were most often conceived of, and implemented by the political and religious leaders of the cities, and the problem was aggravated by the dependence of the cities on the initiatives of despotic leaders for whom the cities were stages for both whim and neglect (Hiorns, 1956). In such a socioeconomic scenario, development of elevated pedestrian facilities could have had several reasons. One of the first ideas for elevating pedestrian circulation routes, during this period, came from Leonardo Da Vinci. Historians guess that some of his sketches, belonging to the period 1487-90, were those of the proposed urban center at Vivegano, a town south-west of Milan, for which Leonardo was entrusted by Sforza to prepare an urban design proposal (Fig. 1.3) (Gibbs-smith, 1978). The sketches show two separate systems of streets, perpendicular to each other and at different levels; the lower street maintaining their continuity through wide archways below the upper-level streets. The two systems were connected through spiral staircases at each archway. The idea could have been an innovative vision of an efficient transportation service system in a city-wide scale where the lower level of the system was exclusively reserved for vehicular and goods movement, and movement of provisions including wood and wine into the residences. All the charms of street life were probably taken up to the upper level, which was exclusively meant for pedestrians. Both the systems were entirely public in nature and the fact that the rear of every house was shown facing the lower streets shows the importance efficient transport could have played in these initial ideas. That service was accorded high priority by Leonardo can be seen in some of his other sketches belonging to the same period (Gibbs-smith, 1978), where a wide network of canals, quite similar to the system at Camacchio, 9 linked every house in the town at the cellar level. This was more likely conceived of to introduce efficient goods delivery by boats directly into the cellar of each house. From yet another viewpoint, the proposed grade separated pedestrian system at Vivegano probably took its form from the then changing attitude of the society towards the public domain. As discussed before, for the people of renaissance Italy public areas were stages for social activities. A bright example could be the Galleria at Milan, which served as a huge, multipurpose, multi-activity social space for the citizens (Rudofsky, 1969). That could be probably what the upper-level street system at Vigevano was meant for. All forms of vehicular traffic were probably planned to be excluded from the upper level to allow the urban dwellers to engage in social activities without interference. Further, the elevated street system at the city square could have possibly been created to give an elevated viewing platform to the city residents and visitors to enjoy and appreciate the marvellous architecture of the numerous piazzas. In contrast, Genoa and Perugia, both hill towns, had different contextual situations as compared to plain area cities like Vigevano. During the Middle Ages, the nobility and feudal lords slowly began to shift from their isolated residences to within the city boundaries, largely out of fear for nomadic ravagers. During the Renaissance, the achievement of political and economic stability led to an intense wave of construction. The nobility moved up to the higher parts of the city and built their palatial residences on terraced gardens, on the mountain slopes (Argan, 1969). Probably the convenience in travelling between places located at higher levels encouraged the princes to build the elevated walkways. Another possible reason for the development of the elevated streets may have been to achieve some degree of segregation from the lower and middle class people as well as to avoid the dirt and Filth of the streets in their localities. 10 Fig.1.3: Leonardo's proposed grade-separated street system at Vigevano. Source: Gibbs-smith, The Inventions of Leonardo Da Vinci, 1978 p.98 At Genoa, the harbor accelerated the city growth and development. The city acquired a reputation as the city of palaces, gardens and churches. Its most famous streets were lined with palatial mansions of architectural distinction including the Cambiaso, and the Parodi. During the Renaissance, the streets and squares in Genoa underwent substantial modification to accentuate the beauty of the physical environment (Hiorns, 1956). Building elevated streets to provide platforms for viewing the newly modified squares, piazzas and palaces with their lofty facades is a distinct possibility. An example of Genoa's grade separated streets can be seen in the engravings by Giovanni Lorenzo Guidotti (1780) (Rudofsky, 1969), which shows an elevated street, Ponte Carignano, 11 passing over low lying districts (Fig. 1.4). Convenience of travel between high lying areas, and segregation from filth seem to be the motivating factor in this example. Fig.1.4: Genoa's Ponte Carignano Source: Rudofsky, Streets for people, 1969, p.187 The system in Perugia highlights other possible reasons for development, which originated long after the Perugians built perhaps some of the most beautiful on-grade street system in Medieval Europe, clearly demonstrating the skill of the artist and technicians of their age. The whole town was designed and detailed for pedestrians which is best reflected in the fine pavements of Via Appia (Fig. 1.5), where the texture, width, height and length of every step vary, breaking the monotony, and giving the pedestrians different choice of surfaces to walk on. All these facilities catered to the deep rooted passion in the Perugian residents for walking and promenading, which exists within the Perugians even today (Rudofsky, 1969). 12 Fig. 1.5: Perugia's Via Appia Source: Rudofsky, Streets for People, 1969, p.192 Furthermore, the undulating topographical features of perugian town probably gave the people additional reasons to build grade-separated pedestrian facilities. Walking probably became easier by connecting places at higher level together. Perugia's Via dell'acqedotto (Fig. 1.6) was one example of such elevated streets where obsolete aqueducts were converted to pedestrian access ways, and merged with the existing grade level street system. It was converted into a foot bridge in 1812 to serve as a safe and comfortable walkway for the professors and students of the university. The converted aqueduct connected the university on the hill to the neighboring areas (Rudofsky, 1969). With house entrances opening on to the upper level streets at many places, these walkways served the three purposes of promenading, viewing, and convenience for travelling between the high lying areas. 13 Fig.1.6: Via Dell'acqedotto at Perugia. Source: Rudofsky, Streets for people, 1969 Another instance of elevated streets in history is a covered passageway, primarily for royal use, built in 1564 by Michelangelo's pupil Giorgio Vasari for the wedding ceremony of the Grand Duke Francesco to an Austrian princess (Fig. 1.7). The passageway started at the Uffizi, proceeded along the right bank of Arno, crossed the Ponte Vecchio on the top floor of its houses, continued along Via Guicciardini on the left bank, and terminated at the Pitti Palace. This walkway was not built for the people, though later it served as an excellent weather proof passageway for visitors, offering splendid view of the town (Rudofsky, 1969). Many other such examples may have disappeared since then. No other major facility was built until the nineteenth century, when transportation and commercial problems led to the implementation of pedestrian facilities in North America. 14 Fig.1.7: Royal bridge in Venice. Source: Rowe, Design Quarterly, Vol.129, 1985, p.10 1.2.3 DEVELOPMENT OF PEDESTRIANIZATION SCHEMES IN NINETEENTH CENTURY NEW YORK The use of streets in nineteenth century New York have been thoroughly depicted in John Grafton's 'New York in the Nineteenth Century'. With predominantly two classes of people; the rich and the poor, streets also served two distinctly separate functions. Large section of the poor population conducted their daily socioeconomic activities on the streets. The poor slept on the streets, in squatter settlements, and used the street corners for their daily activities. The economic activities on those streets were wide ranging, 15 including large open air market places, where food, clothes and many cheap, new and second hand, household goods were sold. Vendors and auctioneers used these spaces regularly for their daily transactions and, quite frequently, the middle class people too patronized these small time operations. Corners and niches on busy streets served well for art exhibitions, and many a times social workers used the same platforms for spreading their idealistic messages. Consequences of such usage pattern were many. Traffic conditions were the most affected. Neither could the vehicular traffic be orderly nor could the pedestrians get their right of way, which fueled thoughts for separating transportation modes. The problems were further aggravated by the scores of people who used the streets as their dwelling areas. Wandering unemployed children, thieves and, at many places, unattended cattle contributed to the lack of safety and poor hygienic conditions in the city's busiest thoroughfares. Ideas for developing separate pedestrian facilities also originated to some extent due to the rich New Yorker's passion for promenading and leisure activities. Waterfront areas witnessed some of the most popular leisure activities, representing in a way the vital importance attached to the beach fronts and to the view by the affluent section of the society. Transportation problems as well as the passions of the upper crust society resulted in the implementation of two prominent pedestrianization strategies during the late nineteenth century: the Loew Bridge (1867) and the Brooklyn Bridge (1870). The Loew Bridge (Fig. 1.8) was probably the first instance of vertical separation of transportation modes in North America, which was proposed and implemented due to the commercial interests of the retail establishments on Broadway. For most part of the nineteenth century, Broadway was the center stage of New York's life. Large retail establishments on both sides of the streets led to a considerable degree of pedestrian-vehicular conflict, due to frequent crossing of Broadway by shoppers. It was also one of New York's busiest thoroughfares, and traffic problems were further aggravated probably due to its diagonal 16 orientation amongst the regular rectangular road grids of Manhattan. In addition, the proximity of the commercial establishments on Wall Street, rendered Broadway one of the major trip generators and destinations of the city. For the shoppers, crossing Broadway was not only difficult but was also dangerous. To save the life of his potential customers and to attract more customers from the opposite side of the street, Genin, one of the major retailer of Broadway, built the Loew Bridge in 1867 (Grafton, 1977), which represented the first such development from purely commercial considerations. Fig.1.8: The Loew Bridge , New York Source: Grafton, New York in the Nineteenth Century, 1977, p. 70 The Brooklyn Bridge (1870), designed by John Roebling (Fig. 1.9), on the other hand solved two major urban problems. First, to increase the mobility of all modes of traffic and second, to provide attractive facilities for the leisure activities of the pedestrians. Although, this bridge was not an example of grade separated pedestrian facilities, it however demonstrated a clear case of planning extensive facilities for the pedestrians in the modern 17 era. The pedestrian facility on this bridge provided the right kind of environment for the most valued social activity of the residents of New York; promenading. The designer's main idea was to design it for strolling and leisure activities, where New Yorkers could enjoy both the waterfront setting and the magnificent view of the city (Grafton, 1977; Rudofsky, 1969). Fig 1.9: The Brooklyn Bridge, New York Source: Grafton, New York in the Nineteenth Century, 1977, p.31 Traffic in New York was undoubtedly one of the major concerns during the late nineteenth century, but partial solutions to traffic congestion came about in the form of grade separation of automated transportation modes at a later stage, when rail lines were lifted up to the second floor level (Grafton, 1977). Pedestrianization or innovative 18 pedestrian facilities, however, were probably never considered as solutions to the center-city traffic problems. 1.3.0 Utopian Thoughts of Early Twentieth Century The appearance of modern day grade separated pedestrian walkways, possibly, occurred with the changing attitude of the people to the activities on the streets. For a long time, streets were designed and modified for the convenience of the automobiles, and vehicles were the most important actors on the street. "Elevated streets for the pedestrians might seem absurd to the Americans and altogether unfair to vehicles. It is the essence of American chivalry to give precedence to the automobiles" (Rudofsky, 1969). The changing streetscape of the twentieth century may be associated with the changing role of pedestrians and pedestrian activities in the urban activity fabric. The changing streetscape were the results of experimentation, based to some extent on a host of Utopian concepts which emerged during the later part of nineteenth century and the beginning of twentieth century. The visions differed widely from one another, with some of them offering detailed workable alternatives. In spite of the wide variation in those concepts, almost all forwarded the vision of a city with easy, convenient transportation and communication facilities, in whatever mode they occurred. The more technologically oriented ones had elaborate descriptions and sketches of grade separated traffic systems. Among the notable ideas of that period, the ones with large scale provisions of off-grade pedestrian systems include Le Corbusier's Contemporary City and Radiant City, Hugh Ferriss's visions of Metropolis of Tomorrow, and similar such ideas of Yona Friedman, and G.A.Jellicoe. 19 1.3.1 VISIONS OF HUGH FERRISS Hugh Ferriss's visions were not entirely motivated by the deteriorating socioeconomic and political conditions. His ideas flowed from a close scrutiny and interpretation of the then present trend in urban design and technological innovations. His visions were mostly in the form of predictions of the future potential development of the urban form and some of those ideas closely matches the form urban areas are taking today. The most prominent among his predictions include the grade-separated pedestrian systems (Fig. 1.10). Ferriss cited the increasing traffic congestion which occurred from the proliferation of skyscrapers, as the main reason for going above grade. No other alternative, including widening the avenues, cutting through new streets or even revising traffic regulations seemed possible to him to overcome the problem. He suggested grade separated traffic systems as the best possible alternative. Not surprisingly, he included a radical change in the retailing pattern along with this change. His predictions sketched shops with two main entrances. The upper entrance to cater to the pedestrian flow in the upper level streets and the lower level entrance for the customers in the cars. Ferriss did not stop at this point. In another illustration of his ideas, he predicted multilevel automobile roadways (Fig. 1.11), at all possible floor levels, running across the city through the buildings, with the buildings large enough to cover a couple of blocks. " One could drive at will across the facades of the buildings at the fifth, tenth or the twentieth storey" (Ferriss, 1929). Here too transportation problems fueled the thought process. 20 Fig.1.11: Multilevel automobile roadways in Ferriss's predictions Source: Ferriss, Metropolis of Tomorrow, 1962, p.65 21 1.3.2 L E CORBUSIER'S IDEAL CITIES Unlike Ferriss, Corbusier's ideas were largely motivated by the unhealthy social, political and economic situation prevalent at that time. His ideas were revolutionary in nature and he believed strongly in bringing change to the social set-up through radical restructuring of the urban form. He endeavored to effect transformation in the socioeconomic and political set-up, entirely through spatial reforms. Unlike the other pioneers of Utopian theories of his period, including F.L.Wright and Howard Ebenezer most of whom strongly advocated decentralization of the urban settlement pattern, Corbusier's vision idealized further concentration of urban activities and the appropriate exploitation o f the benefits of the industrialized age. His ideas are most often associated with high technology and urbanism, and were based on the idea of employing all the resources of advanced technology to enable architecture to achieve a form suitable to the spirit of the modern age (Fishman, 1977). His ideal cities thus took the form of geometrically arranged megastructures of glass and steel, super highways and multilevel traffic systems. His concept of a Contemporary City (Fig. 1.12) was futuristic conception of speed, movement, and mechanization, with a Garden City emphasis on greeneries and open spaces. "The transportation system preserves the very life of the city. The city that achieves speed, achieves success",(Quoting Corbusier, Fishman, 1977). His idea of speedy travel manifested in the form of an elaborately coordinated transportation system. Every mode of transport, including pedestrians and bicycles, had their own separate speedy accessways in his city for the modern age, linking the megastructures in both, horizontal and vertical circulation networks. He sought a total separation of motor freeways from building lines and pedestrian ways, with the pedestrian circulation paths cutting through landscaped spaces and passing underneath buildings raised on pilotis. 22 Fig.1.12: Sketch view of "Contemporary City". Source: Evenson, Le Corbusier, 1969 Unlike the Contemporary City, in which Corbusier advocated a horizontal separation of pedestrian routes, his plans of the Radiant City (Fig. 1.13) went for a vertical separation of pedestrian pathways. Here, the road systems were designed at two levels for automobile traffic on a grid of 400m by 400m, the ground level for heavy traffic and the upper level for rapid motor traffic. Pedestrian pathways were provided below the road level (Evenson, 1969). 23 Fig.1.13: Grade separated traffic in the "Radiant City". Source: Evenson, Le Corbusier, 1969 1.3.3 YONA FRIEDMAN'S CONCEPT OF SPATIAL CITY As late as 1958, Yona Friedman, a Paris-based architect, forwarded vision of a Spatial City (Fig. 1.14). Friedman's concepts demonstrated striking similarity with the 'machine model' of Corbusier's Radiant City, motivated largely by idealistic arguments of equity of allocation, good access facility, smooth technical function, productive efficiency, autonomy of parts, physical health, and material well being. His ideas took the form of a tiered city where the new upper levels were proposed to be constructed above the older city, thus providing for growth without uprooting the older areas. In this city, he preserved the privilege of the pedestrians by lifting the pedestrian areas above the automobile routes (Lynch, 1981). Fig.1.14: Yona Friedman's "Spatial City". Source: Rudofsky, Streets for People, 1969, p.197 1.3.4 MOTOPIA OF G.A .JELLICOE A more explicitly expressed and intricately detailed idea of pedestrian-vehicular separation came from G.A.Jellicoe's model of Motopia (Fig. 1.15). All automated traffic was lifted up to the building terrace level, in his vision, and the form of the building as well as the city structure was determined by the lay-out of the roads and traffic circles. The entire area at ground level was proposed as parks and pedestrian areas, in this city (Lynch, 1981). It is difficult to state definitively the workability of this proposal. Nevertheless, the idea was one of the First in proposing the provision of pedestrian facility in a city wide scale. Fig.1.15: Jellicoe's "Motopia". Source: Lynch, Good City Form, 1981, p.166 1.4.0 Development of Integrated, Grade Separated Pedestrian Networks The first appearance of modern day grade-separated pedestrian systems occurred in the European New Towns. Initial development of New Towns concentrated on creating healthy, pleasant, efficient, safe and beautiful environment for the city inhabitants. Most of the concepts were similar to the ideas of Garden City of Howard Ebenezer, including among others the idea of decentralization, low density development, restricted population, and community participation. To some extent, Corbusier's ideas of speedy transportation, increased accessibility, modal separation and multi-activity urban centers were also tried out. The New Towns seemingly played a major role in the experimentation of these ideas and innovations, where the integration of varying activities in megastructures, idealistic city-organization structures, and new transportation systems and techniques were experimented. It is interesting to note that development of grade-separated pedestrian 26 systems came largely as innovation in the traffic management system. Such systems were conducive to the growing conception among urban dwellers, of providing priority for the pedestrians on urban roads. Although the system changed in its physical form during the process of these experimentations, it was never abandoned altogether. 1.4.1 EXPERIMENTATION IN BRITISH NEW TOWNS Grade separated pedestrian systems came in two different phases of development in British New Towns. In the fourteen first generation New Towns which were built during the 50's, including Stevenage, Crawley and Harlow, they appeared at only a few major locations of the town. Such a fact can be attributed to the relatively low density spread of these towns where, the primary design criteria was efficient vehicular movement. In such widely spread out, low density developments, with strong neighborhood, highly decentralized commercial cores, and large travelling distances in the core city areas, grade separated pedestrian systems, probably, could not justify large scale applications. The second generation New Towns in Britain, which were built during the 60's, witnessed for the first time the most widespread incorporation of such systems into the urban fabric as an integral element of town design and function. Here again, there were a host of factors supporting such development. The relatively denser settlement pattern, weaker neighborhood, strong central-city areas, shorter walking distances in the core areas, and increased attention to the mobility of the pedestrians helped in reenforcing the idea of incorporation. The vehicular routes in these New Towns were longer and more indirect than the walking routes. The first town with these characteristics was Cumbernauld and subsequently eight other towns including Skelmersdale near Liverpool, Livingston near Edinburgh, and Hook in Hampshire (which was never built), had elaborate networks of off-27 grade pedestrian systems built in their core areas (Fig. 1.16) (Galantay, 1985: from articles by Thomas, William, and Carter, Christopher). Fig.1.16: The multilevel traffic system in the second generation British New Towns Source: Galantay, New Towns World Wide, 1985, p. 169 1.4.2 NEW TOWNS IN NETHERLANDS AND FRANCE Other examples of such experimentation can be found in Netherlands, in the city of Lelystad and Almere, built during the 1960's. The pattern of development of these towns were quite similar to those of British New Towns. In Lelystad, the original plan had the pedestrian routes on ground level and the vehicular traffic on elevated streets. Subsequently, the plans were revised, and the vehicular accessways were taken slightly below ground level with pedestrian overpasses at necessary locations. A vast vehicle-free zone in the central-city area supplemented the entire network. The system allowed the pedestrians to enter the central zone at an elevated level, after crossing the bordering vehicular road. Ideas had begun to develop during the planning phases to provide multilevel access to buildings in the core city areas which, however, were never 28 implemented. The resulting urban spatial pattern took the form of galleries at the second floor level in every building, connected to one another through a system of elevated walkways. Most of the shopping activities remained on the ground level pedestrian areas, due to the incapability of the weak economy to support a two-tier retailing activity. The system in Lelystad soon lost its popularity. Because of the limited number of overpasses, the pedestrian had to make detours to go from one place to another where as the vehicles enjoyed unhindered mobility. The elevated walkways in the core areas suffered too, due to the weak economy. The second New Town, hence, saw neither of these two systems. Unlike Lelystad, Almere was a multi-nuclear development, where the core areas were not dense enough, nor was the economic activity level high, to support the concept of elevated pedestrian walkways and upper level galleries. The notable modification to the pedestrian system came up in the form of pedestrian crossings slightly below ground level with the vehicular roadways elevated slightly above grade (Galantay, 1985: from article by Constandse, A.K.) . Similar examples can be found in France, where the New Town movement came late. In all the nine New Towns planned and implemented in France during the late sixties, pedestrian systems were accorded high priority and most of them saw the application of some form of grade-separated pedestrian circulation systems for the first time in modern France (Fig. 1.17 and 1.18) (Galantay, 1985). 1.5.0 Pedestrianization and the Urban Activity System The next major occurrence of elevated pedestrian systems and facilities took place in the mid-twentieth century. Before the tewntieth century pedestrian schemes were 29 Fig.1.17: Pedestrian/car and bus segregation at Evry. Source: Galantay, New Towns World Wide, 1985, p.124 Fig.1.18: Deck level access to residential buildings at Le Vaudreuil Source: Galantay, New Towns World Wide, 1985, p.123 implemented, urban areas across Europe and North America had numerous examples of pedestrianization schemes some of which achieved success and some failed. The brief 30 analysis of various schemes carried out in this chapter, makes it amply clear that provision of pedestrian facilities do not necessarily lead to their successful and efficient utilization, since many of those described here either failed to attract users or were closed soon after their implementation. Instances of failure includes the system at Lelystad (section 1.4.2), the Brooklyn Bridge and the Loew Bridge at New York (section 1.2.3). Moreover, as the contextual situation varied, the factors supporting the implementation of such schemes changed. To look from the correct perspective, pedestrianization and to that matter any urban development schemes are but small entities in a complicated network of the urban activity system. As the co-ordinates of time change, the actors as well as the activity pattern change significantly. The changing contextual variables had tremendous impact not only on the implementation of the schemes but also on the post implementation performance of the facilities. Beside the ever changing contextual situations, all the past instances of off-grade pedestrian system development highlight a few significant issues. First, a majority of the systems were implemented as transportation planning tools. The importance transportation planning considerations played in the development can also be observed in the Utopian theories of Ferriss and Corbusier. Loew Bridge was the only instance which came about as a commercial venture, but there too the commercial interests originated out of the deteriorating traffic condition on Broadway. However, transportation planning was not always the sole objective behind such developments, and the second point originates from this fact. A close scrutiny of the successes and failures of various systems points to a distinct probability. Most of the systems which failed during the course of history were implemented mainly as transportation/commerce tools, and not as urban public places. Not much is known regarding the performance of the systems in Brisighella and Camacchio, where objectives 31 were possibly limited to providing efficient transportation system. The objectives of subsequent developments in Genoa, Vivegano and Perugia influenced a much larger area of the urban activity system. The motivating factors behind building the systems in these Renaissance towns, besides transportation efficiency, were many including convenience, pleasure, view, social activities and segregation. The promotion of the facility objectives from serving a purely utilitarian purpose during the Middle Ages to softer and more subjective areas of human actions like promenading and social interaction during Renaissance amply reflects the difference in the socioeconomic scenario of the two periods. Since, the systems developed during those periods were proposed and implemented mostly by autocratic decrees, post-implementation performance evaluations were seldom carried out, leaving little conclusive evidence for assessment. Nevertheless, the fact that many large scale systems were implemented like the one in Perugia, and conceived of as the scheme in Vivegano, points to the probable success of such ventures. Also, this period in history witnessed the largest degree of incorporation of off-grade pedestrian systems prior to the twentieth century. The system in Perugia, in particular, experiences an optimum degree of use even today, and the high level of social activities prevalent in Renaissance Europe is no secret (Zucker, 1959; Rudofsky, 1969; Argan, 1969; Hiorns, 1956). The varieties of outdoor social activities which took place in the urban spaces and Piazzas must have attracted large number of users to the facilities, and it is notable that the Perugian system took birth as a public place for social activities including the Perugian's passion for walking and promenading. During the next phase of development, in nineteenth century New York, Loew Bridge was removed only two years after its implementation, in 1868 (Grafton, 1977). The closing of the Loew Bridge establishes further support to the current notion, since it was chiefly a commercial/transportation venture. Similar reasons led to the failure of the upper level galleries in Lelystad. The system developed to segregate the pedestrians from the automobiles probably could not provide the urban pedestrian with a proper environment, although the automobiles enjoyed unhindered access throughout the 32 core area. Such ideas were never incorporated in the subsequent New Town proposals in Netherlands. All the facts hints at the possibility that grade separation of transportation modes, especially between the pedestrians and the automobiles, is less likely to succeed unless the entire development is oriented to address several human issues for the walking public. A third issue qualifies consideration in the context of such development in the twentieth century. The examples considered in this chapter clearly demonstrate that changes in the socioeconomic-political variables could affect facility performance to a significant extent. The forces leading to the development, and factors affecting the ultimate success of any urban development scheme are many, and their interrelationship, highly complicated. Pre-implementation analysis, and the envisaged degree of success of any project objectives, as a result, cannot be based on achievements experienced in a different contextual situation, without making a thorough analysis of the variables and their interrelationships in the context under consideration for development. Copying of off-grade pedestrian facility programs and designs from one city to another, thus, could have dangerous implications, considering the fact that the cold-winter cities, specifically Minneapolis and Montreal, are the most copied models of off-street systems ( Whyte, 1988). Off-grade pedestrian networks, however, have already been built in nine medium size North American cities. Due consideration of the issues discussed above, in the present •^ sas.-- ... context, would include two major study areas; the contextual variables prevalent during the development of the North American System, and the chief objectives behind their implementation, which have to be thoroughly analyzed before any post-implementation evaluation can be made. That sets the stage at the beginning of the twentieth century North America, when the introduction of Street Cars revolutionized urban settlement pattern. C H A P T E R TWO THE SKYWAY SYSTEMS IN NORTH A M E R I C A 2.1.0 Twentieth Century Cities: the Transition of City Centers Towards the beginning of the twentieth century, technological innovations in the transportation sector generated strong centrifugal forces in the urban settlement pattern of the North American cities. However, most of the associated transportation strategies were probably wrongly timed and designed, since, they led to drastic, cumulative deterioration of the vitality of the central-city areas. The process began with the inception of trains and electric Street Cars in the urban mass transportation programs. The city centers suffered a gradual reduction of their residential population, and the emigrating population relocated themselves along the mass transportation routes. With the loss of the residential population the downtown assumed a redefined role as the work, cultural, retail and entertainment center for extensive areas of residential developments around it. The subsequent birth of the automobile, for the first time, led to dramatic increases in the mobility of urban residents, leading to a quantum jump in the inter-city travel volume. With the increased accessibility to far flung areas, the pace of suburbanization accelerated. Towards the end of the second decade of this century, traffic volumes on North American roads far exceeded the designed capacity. First, the depression and then the second World War prevented the local, state, and federal authorities from allocating any large scale resource towards mitigating the problems in the transportation sector. Immediately after the war, in 1956, the U.S. federal government launched the Interstate Highway System 3 4 Program as a solution to transportation related problems in urban areas in the United States, but severe urban congestion, traffic congestion, noise and environmental deterioration followed due to the program, especially in the downtowns. The downtowns steadily lost their capacity to support the expansion needs of the industries and employment bases. Congested traffic around industrial sites and service areas caused increasing disruption to deliveries and employee access. Land prices shot up and downtown retail corridors were no longer safe, enjoyable places for pedestrians and shoppers. Industries, offices and retail businesses looked towards the suburbs for greener areas. With the subsequent rapid suburbanization of retail and employment bases, the city-centers lost their position as the principal cultural, employment, financial and retail center of the metropolitan regions, to large suburban establishments. The resulting reduction of tax base, the level of city services, level of consumer spending and the deteriorating overall quality of life of the city residents was a potential threat to the economic vitality of the central cities and the economic stability of the region as a whole (Carpenter, 1975; Levinson, 1986; Robertson, 1985). 2.2.0 The Urban Redevelopment Programs The first large scale efforts to revitalize the city cores came in the form of the Urban Renewal Program of the U.S. federal government, which were mainly implemented between 1949 and 1974. Funds for the program came mainly from the federal government, and the state and local governments decided on the project location and other administrative and judicial matters. The implementation was left largely to the developers and as a result, the program resulted in large scale destruction of low-income neighborhood, displacement of commercial and residential areas and the stalling of 35 developmental activities on large tracks of cleared urban lands. The Urban Renewal Program, nevertheless, produced a few successful redevelopment projects including the Charles Center in Baltimore, Quincy Market in Boston, and the Golden Triangle in Pittsburgh (Robertson 1985). During the same period numerous other strategies were experimented to bring about some change in the overall health of the core-city areas. Most of the strategies came as incentives to developers in the form of tax breaks, land write-offs and other special service provisions, funded by the federal government through schemes such as the Community Development Block Grants and the Urban Development Action Grants, for state approved, coordinated redevelopment projects. None of the strategies demonstrated any tangible success in revitalizing the city cores (Robertson, 1985). American cities had to wait until the 1970's to experience success for the first time in decades of experimentation. 2.3.0 Revitalization Strategies during the 70's and 80's Failure of the revitalization strategies of the 50's and 60's to accelerate the recovery of CBD vitality led to the adoption of eight, often overlapping, major planning strategies by the U.S. federal and state authorities, which included the coordinated development of a variety of renewal action plans (Robertson, 1985). The eight areas of concentration, viz. pedestrian malls, indoor shopping centers, special use activity generators, historic district preservation, waterfront areas and activity development, open spaces, downtown accessibility, and the commercial office sector, are briefly discussed in the following subsections. 2.3.1 PEDESTRIAN MALL 36 The most widely adopted strategy was the pedestrian mall. Frequently complemented be some form of automobile restriction, they were aimed to create a more walkable and pleasurable urban setting for the downtown shoppers and pedestrians. By eliminating automobile traffic the cities hoped to attract more shoppers from the suburbs to the downtown retail and commercial bases and thereby rejuvenate the core area commercial environment, and the tax base. 2.3.2 INDOOR SHOPPING MALL The primary objectives of creating indoor shopping malls was to directly compete with the suburban shopping centers, which offered climate-controlled, indoor, landscaped, automobile free environments, with easy vehicular access to the facilities. Since most of the downtown retail activities had shifted to the suburban centers, the cities hoped to attract the patrons back by creating similar environments inside the Central Business Districts. 2.3.3 SPECIAL U S E ACTIVITY GENERATORS By creating special function public areas like convention centers and large sports facilities, the cities expected to attract more people from the suburban areas, and out of town residents, to the downtown. The basic idea was to re-establish the City-Center as the principal multi-activity center of the metropolitan area. In addition, planners expected some of the people visiting those facilities for the special programs to patronize the core 37 area retail establishments before and after the events, thereby improving the sagging economy. 2.3.4 PRESERVATION OF HISTORIC DISTRICTS Well preserved historic districts make the environment lively and unique. The authorities aimed at taking advantage of the old buildings and thereby encourage activities including tourism, antique shopping, boutiques, night clubs, affluent residences inside the downtown. Such districts were anticipated to serve as major urban hubs or focal points, not available in the suburbs. 2.3.5 WATERFRONT REDEVELOPMENT Waterfront activities have been a major passion for most of the North American urban dwellers and this scheme aimed at exploiting this facet of the urban residents lifestyle to bring more people into the central city. Since most of the North American cities developed around water bodies, this was considered a potential strategy for urban development. Plans included the replacement of old railyards and warehouses with modern recreational facilities. 2.3.6 OPEN SPACES This scheme was developed to alleviate the negative impacts of congestion and high density development prior to the 70's. The realization that open spaces, street corners, 3 8 squares, malls, and plazas go a long way in providing urban platforms for social activities, and providing a human scale to the CBD public domain led to the implementation of this scheme. 2.3.7 ACCESSIBILITY Accessibility was accorded high importance in the urban planning strategies. Increased accessibility to the downtown retail and business establishments from the suburbs and from the remote areas of downtown was considered important to attract more customers and increase the level of commercial activities in the downtown. The scheme often included improvement in the public transit, traffic pattern and parking facilities. 2.3.8 OFFICE SECTOR Multi-use, large scale developments for commercial office spaces, through mixed-use facility development, were encouraged to create competitive facilities in the core areas for the increasing orientation of the urban economy towards the financial and service sectors. 2.4.0 Pedestrianization Strategies in North America: Objectives, and Impacts Out of all the varieties of planning strategy experimented after the 60's, pedestrianization of downtown areas, through the implementation of pedestrian malls and 39 precincts, witnessed the widest application. In spite of frequent opposition to the idea from various groups, it was considered so innovative and effective that almost every medium and large urban area in the continent had implemented such a scheme by the end of the 70's. Initial development of pedestrian areas and precincts were limited to single streets, due to the apprehensions of the opposition groups. However, within a few years, pedestrian areas and Automobile-Restricted-Zones were starting to be considered as one of the most effective measures for reviving the economic vitality of central business districts. By 1977, at least 90 U.S. cities had implemented some form of ARZs (Loukissas 1984), including several major cities like Boston, Chicago, Baltimore and Philadelphia, where large scale pedestrianization schemes were implemented during the same period. 2.4.1 OBJECTIVES OF PEDESTRIANIZATION/ARZ SCHEMES The scope of pedestrianization schemes were not limited to any single facet of the urban activity system. For a long period of time, it was considered a panacea for downtown problems and its final application was associated with something as complicated as reviving the vitality of city centers. In the process, pedestrianization schemes involved five main objectives for achievement, which together were envisaged to bring about a resurgence of socioeconomic activities in the core areas. They were (compiled from articles by Loukissas, 1984; Loukissas, and Mann, 1985; Weisbrod, 1982; Karash, 1983; Dueker, Pendleton, and Rao, 1983; Cottle, 1972): (a) Improve traffic conditions : Reduction of automobile congestion, pedestrian congestion, and pedestrian-automobile conflict. To increase the safety of the pedestrians. 40 (b) To achieve modal shift in urban travel : to encourage the use of public transit and non-automated modes of transport. To discourage the use of private automobile for downtown trips. (c) Better pedestrian environment : To attract more pedestrian activities in the CBD by improving the physical environment. To encourage people in the downtown to stay longer. To make the downtown environment attractive. (d) Improve the environmental quality : Reduction of noise, fumes, and air pollution. To save energy through reduction of travel volume as well as travel time. (e) Increase commercial activity : To attract more businesses to the downtown area, and to encourage existing business to stay. Support and expand the market for downtown retail activities. To attract more shoppers to the downtown shopping areas. 2.4.2 FORMS OF IMPLEMENTATION Not all facilities were alike. Although, almost every facility implemented was based on all the five objectives, their final form varied depending on the contextual dissimilarities, and numerous other variables affecting implementation (Loukissas, and Mann, 1985). Also, the primary objective for development differed, from urban planning to transportation planning, and commerce, from city to city. They however can be grouped under three major categories. (a) Pedestrian Precincts : total automobile ban with road/parking control policies, and associated improvements to the physical environment. 41 (b) Partial/Selective Automobile Ban : in some cases with preferential transit service, road/parking control policies, and an improvement to the malls physical environment. (c) Grade Separated Pedestrian Systems : with associated, large scale, road/parking control policies, and physical restructuring of the internal environment. 2.4.2.1 PEDESTRIAN PRECINCTS Pedestrian precincts were associated with a total elimination of all types of vehicular traffic from the designated area. However, emergency and service access for fire vehicles, postal vehicles and ambulances was provided for in the schemes. In several instances time restricted access to service vehicles for goods delivery was provided for from the front, in the absence of back alleys for such activities. But in a majority of the facilities, the precincts were exclusively reserved for the walking public. The central vehicular carriageways and the sidewalks were leveled and re-paved for the pedestrians, and the architectural tone of the entire area elevated by means of various pedestrian amenities and landscaping features (Cottle, 1972). 2.4.2.2 PARTlALlSELECTIVE A UTOMOBILE BAN The most widely adopted form of all pedestrianization schemes was a combination of a ban on all private automobiles, and preferential routes for public transit services. Many a times referred to as transit mall, here again, the essential services including post, ambulance, fire, and police were allowed access with restrictions on the time and routes. Primary objective was to achieve better mobility for the pedestrians, increase accessibility to the central city commercial areas from the suburbs, and to induce modal shift in inter-42 CBD trips. Operational measures to improve transit service including priority signaling, preferential lanes, improved loading facility and scheduling, and rate rationalization were effected. In all cases, the central carriageway was reduced in width, sidewalks were widened, and exclusive transit lanes established on the mall. In a similar fashion to the precinct, the physical environment was improved for the pedestrians (compiled from articles by Dueker, Pendleton, and Rao, 1983; Karash, 1983; Weisbrod, 1982; Gomez-Ibanez, and Fauth, 1980; Loukissas, and Mann, 1985; Loukissas, 1984). 2.4.2.3 GRADE SEPARATED PEDESTRIAN SYSTEMS With a much larger scope and impact area, in several instances, separate precincts for the pedestrians was provided for by a vertical separation of the pedestrian and automobile transport modes. These systems fall under a different category of pedestrianization scheme and hence will be covered later in this chapter. 2.4.3 IMPLEMENTATION AND IMPACTS OF MALLS AND PRECINCTS Pedestrianization schemes were successfully implemented in a large number of towns and cities in Canada, U.S.A. , and Europe. The success rates have varied, although most of the impact assessments have been carried out in the realms of transportation planning. Most of the malls were perceived as successful, by the public and the authorities alike. The Downtown Crossing project in Boston, Nicollet Mall in Minneapolis, the Kalamazoo in Michigan (one of the first examples of permanent pedestrian precincts), Lincoln Road Mall in Miami, Fulton Mall in Fresno, Pamona Mall and Santa Monica Mall in California, the 5th and 6th Avenue Transit Mall in Portland, Spark Street Mall in Ottawa, and more recently, the Granville Mall in Vancouver, and many other malls have 43 , been regarded as highry successful pedestrianization ventures in North America. Although pedestrian malls and precincts have not proved to be the panacea of all central city problems (Loukissas, 1984), most of the implemented malls seem to have had a positive impact on CBD health. They demonstrated a snowballing effect by providing stimulus to the economic, social, cultural, and aesthetic life of the city core. The image of a declining downtown was reversed in most cases, and optimism rejuvenated (Onibokun, 1975). In spite of the variation in success rates, majority of the malls demonstrated positive impacts in several significant areas. During the post-implementation period malls experienced considerable increase in retail oriented businesses, and an improvement in commercial office development on parallel streets (Loukissas, 1984). Property values rose, store vacancy rates reduced, capital investments increased, retail establishments experienced a surge in sale, and a marked increase was experienced in the number of pedestrians visiting the mall establishments and facilities. There was some marked reduction in the rate of deterioration of downtown economic activities, and the projects successfully stalled the suburbanization of CBD businesses to certain extent. Transportation related benefits were numerous. Traffic conditions improved, reducing vehicular congestion, and the volume of private automobiles on downtown streets. Marginal transit benefits were achieved, and some degree of tangible modal shift occurred through increased ride sharing. Air and noise pollution decreased, and significant reduction in traffic accidents experienced. Social activities on many malls increased too, with car displays, boat shows, craft exhibits, festivals, community sales, and special events occurring throughout the year. In the most unsuccessful cases, little negative impact was recorded (Onibokun, 1975; Weisbrod, 1982; Carlson, and Carlson, 1974; Dueker, Pendleton, and Rao, 1983; Smith, 1985; Cottle, 1972). 44 Efforts at establishing the success of the pedestrianization experiments, however, remained more or less restricted to the first two categories of implementation. Having established the general positive impacts of the mall experiments, only a few research works were dedicated for the assessment of the third major type of pedestrianization experiment; the Grade-Separated Systems. Implemented almost simultaneously with the other two types of revitalization strategies, the off-grade Systems came about with a substantially broader objective and envisaged area of impact. Detail investigation on the issues related to the role of grade-separated systems in the urban revitalization strategies is conducted through the rest of this chapter. 2.5.0 Grade-Separated Pedestrian Systems as Urban Revitalization Strategy The grade-separated pedestrian systems offered the planners an extremely powerful tool in itself for urban development. It was expected to successfully handle, at least in theory, a large number of complicated issues simultaneously. The3' offer a physical alternative, fully or partially, to at least six of the eight planning strategies for center-city revitalization forwarded during that period (section. 2.3.0). Out of the eight strategies, waterfront development and development of historic districts have no relations with off-grade pedestrian systems. The pedestrian malls, special activity generators and open spaces have some relationship with the implementation of these systems. The last three strategies- indoor malls, accessibility and orientation towards office sector-, however, have a direct relationship with off-grade pedestrian systems (Robertson, 1988). Moreover, unlike other pedestrianization schemes, grade-separated pedestrian systems have tremendous potential in generating pedestrian areas without affecting vehicular accessibility to downtown activity areas. The objective for the implementation of such a powerful and 45 expensive system, as a result, covered more than a single facet of the urban activity system, and had a wider impact area in comparison to the malls. The main objective areas, which are discussed in the subsequent, subsections, can be categorized under three major classes (compiled from articles by Shostack, 1978; Carpenter, 1975; Goldfield, 1976; Ponte, 1986; Reis, 1982; Podolske and Heglund, 1976) : (a) Transportation planning objectives. (b) Urban planning objectives. (c) Business and commercial objectives. 2.5.1 OBJECTIVES OF GRADE-SEPARATED PEDESTRIAN SYSTEMS 2.5.1.1 TRANSPORTATION OBJECTIVES (a) To relieve traffic congestion : To relieve the sidewalks from large volume of pedestrian traffic. To reduce pedestrian-vehicular conflict at grade level, and in the process, save valuable resources in terms of vehicular travel time and pedestrian walking time. To make appropriate arrangements for goods delivery and, in some cases, reduce conflicts on sidewalks between pedestrian flow and goods flow. To increase the vehicular capacity of streets and thereby effect further reduction in travel time. (b) To improve circulation pattern : Traffic flow in and out of the central business district; parking; automobile movement; and pedestrian movement. (c) To increase transit ridership and encourage modal shift for downtown travel. (d) To make the downtown areas and facilities easily accessible to the suburban residents. 46 (e) To create a complete network of connections between the major activity hubs, facilities, focal points, parking areas, transit terminals and transfer points. (f) To relief future traffic pressure through provision of rapid transit and to encourage modal shift by making rapid transit more attractive through the creation of network connections. (g) Safety : To improve pedestrian safety by enabling them to cross streets without having to buck the traffic and thus reducing pedestrian-vehicular conflict. (h) Environmental improvement : To reduce noise and air pollution by decreasing automobile volume and automobile travel time. To reduce automobile fuel consumption. 2.5.1.2 URBAN PLANNING OBJECTIVES (a) To provide for further concentration of activities in the downtown without creating congestion. (b) To increase the downtown residential population density. (c) To increase the downtown after-dark activities. (d) To regroup basic activities (office, retail, entertainment, etc.) into compact areas with special considerations for pedestrians. (e) To restore downtown's economic, cultural and residential importance and make it the major center of financial, business, social, entertainment and cultural activities in the region. (f) To elevate property values in downtown and establish a strong tax and service base. To attract more people to visit, shop, work, and reside in the central city. To make use of the same scarce and valuable real estate several times over by stacking different circulation modes at different levels. (g) To raise the general tone and ambiance of the CBD. 47 (h) To stem the suburbanization of business, employment and retail bases. To attract more businesses to the downtown establishments. 2.5.1.3 BUSINESS AND COMMERCIAL OBJECTIVES (a) To provide shopping environments to the urban residents, as good as the suburban shopping centers and malls. (b) To attract more retail activities and shoppers to downtown establishments and thus bring about an upswing in the downtown economy. (c) To make the central city's overall business atmosphere attractive and conducive to economic development. To recover the businesses lost to the suburban establishments. 2.6.0 Forms of Implementation Similar to the development of the other pedestrianization strategies, grade-separated systems also varied significantly in their forms of implementation. The variations were of two types. The first originated from the range of objective associated with the implementation, and hence their role in the revitalization process. The second variation arise from their conjunction with other pedestrianization strategies in the overall process of downtown renewal. Depending on the role they played, two major categories can be made; those which were developed as an integral part of broader urban development strategies, and isolated systems with micro-level influence. 48 2.6.1 DEVELOPMENT AS PART OF BROADER URBAN DEVELOPMENT STRATEGIES The first major difference between the two forms of development lied in the envisaged area of impact. The systems developed as part of broader urban development strategies were extensive in form, and influenced wide areas of the urban activity system. However, all the large-scale integrated grade-separated pedestrian circulation systems in North America built during the last three decades did not necessarily had a single primary objective for development. The priorities differed from city to city and as a result there are minor variations in the systems. An excellent example is the system in Dallas where it was developed essentially as an urban planning strategy (Ponte, 1986) where as in Cincinnati, transportation benefits fueled the development (Shostack, 1978). The system at Minneapolis initiated by private sector assumed, in the later phases of development, a wider role as a global revitalization strategy for the Central Business District (Shostack, 1978). Systems also differ according to the location of the pedestrian circulation facility in the vertical grid. In many systems, walkways were built above grade at the second, third, and in some cases even at the fourth floor levels. Best examples of such developments can be found in Calgary in Canada, and Minneapolis in the U.S.A. . In many other cities, systems have been constructed below ground level, the largest of which exists in Montreal, Canada. Most of the above-grade systems have buildings connected by elevated walkways roughly located at mid-blocks. Some of the systems, in harsh climatic zones, developed fully enclosed and climate-controlled walkways for pedestrians. In Minneapolis, the climate controlled walkways lead directly through large indoor courts and atria and extensive shop lined corridors. A majority of these elevated walkways are at the second floor level with some exceptions in Calgary and Minneapolis where the walkways have been built further up at the third, fourth and fifth floor levels. In most of the walkways built at a later stage, 49 mechanical vertical connectors including elevators and escalators connect the upper level passages with the grade level network, to increase the system's accessibility to old and handicapped users. Finally, almost every off-grade pedestrian network have had elaborate physical improvement programs associated with the implementation, including provision of carpeted floors, designer graphics and paving tiles, attractive signs, orientation maps, security systems, landscaping elements, public rest areas, shops and viewing points. Another factor which differentiate these systems from the other form of development, is the system size. All these systems were large and covered a number of blocks. The largest network was planned in Minneapolis where the system was initially planned to connect 64 downtown blocks with 76 bridges (Shostack, 1978). Calgary, with plans to connect 60 CBD blocks is by and far the second largest planned system (Shostack, 1978). The other cities with extensive systems include St Paul with 27 planned bridges connecting 25 blocks (Morphew, 1984), Duluth with 15 planned bridges connecting 15 blocks, and Cincinnati with 17 planned bridges connecting 14 blocks (Robertson, 1988), Des Moines with 23 planned bridges connecting 19 blocks (Heglund, 1982), and one of the smallest systems at Edmonton with plans to connect 7 downtown blocks (Podolske, and Heglund, 1976). The building works for most of the systems started during the 60s with a few starts made in the 70s. 2.6.2 ISOLATED, ELEVATED PEDESTRIAN ACCESS CORRIDORS Although, similar in their physical form to the large scale integrated developments, the major difference of these off-grade pedestrian corridors from the large-scale ones is that they do not form a part of any major urban development strategy. They are rarely 50 interconnected and the objectives for their implementation cover extremely limited activity areas. Instances of isolated developments are numerous. Practically, in every North American city such isolated walkways have emerged in the denser commercial and large institutional areas. In most of the cases they link large shopping centers to other shopping and commercial office buildings, parking garages, or other active pedestrian corridors. The main objectives are limited to private commercial interests, the primary motive being to make commercial establishments attractive and easily accessible to customers in the highly competitive urban commercial market. In some instances, they have been built in institutional campuses only for the convenience of the users. The influence of these isolated systems on urban transport and commerce are limited to micro-level urban activities with negligible influence on the urban activity system at macro level. 2.7.0 Elevated Pedestrian Networks in the Realm of overall C B D Pedestrianization The second basis for variation in grade-separated systems originated from the conjunctive role the system played with other pedestrianization strategies. Instances are rare where the Skyway networks or the sub-grade systems constituted the only pedestrianization strategy for the core city areas, since, most of the North American cities had experimented with one of the three pedestrian systems, more commonly the grade-level malls, before adopting the off-grade ones. 51 There are various reasons for such a pattern of development. The planning decisions were affected to some extent by the comparative advantages and disadvantages of each type. In comparison to the off-grade ones, the grade-level pedestrian precincts and malls involve considerably less effort and finance for implementation. They are the cheapest of all the pedestrianization options available to the planners, and since they basically involve street closures, revising traffic plans, modifying the physical environment and in some cases, minor re-routing of traffic and transit services, they do not require long time for their implementation (Fruin, 1973). In addition, in case the project impacts turn negative, the original street configuration and traffic pattern is easier to restore, without much expenditure of finance and time. The grade level pedestrian systems, however, do not enjoy the advantage of providing weather proof environment to the pedestrians. Experimentations have been carried out in this area, the brightest example being the Nicollet Mall at Minneapolis. The mall pavements were provided with heating elements to keep it snow free during the winter months, and the bus shelters were heated with infrared radiation systems for the convenience of the transit users (Goldfield, 1976). Few other efforts have been made in this direction in the North American pedestrian malls, and the grade level malls continue to be disadvantaged during the cold winter months. The below grade pedestrian networks are the most expensive in financial terms. But they are easy to route and do not depend on the street and building locations for their layout. These systems can be easily tied to the subway terminals and hence effectively serve as pedestrian access corridors. They often provide a great amount of leasable space, allowing more retail opportunities for the developers. They do not have to abide to the set-back laws laid down by the city municipalities and can extend right up to the plot boundaries. Large scale underground people space can be created and linked to these systems. The underground corridor do not have the problem of negative aesthetical impacts on the downtown architecture, which the elevated systems have in terms of 52 compatibility to different architectural styles and blocking landmark views. They, however, completely cut off the pedestrians from the exterior environment. The elevated pedestrian networks, on the other hand, have the advantage of costing considerably less compared to the underground corridors. They also offer a secure and psychologically satisfying environment to the pedestrians by providing clear linkages between the exterior and interior environments. They permit exposure to natural light and air, and provide better orientation than the underground system. They also have the potential of accommodating multi-level retailing in the core areas with minimum financial investments (Fruin, 1973; Reis, 1982). The grade level pedestrian areas were most often implemented due to the lesser cost involved in the implementation, and the sub-grade level networks were built mainly supplementing the subway transit networks. Owing to a host of other variables affecting the implementation of pedestrianization schemes, North American cities had no common pattern of pedestrian system development. However, the type of combination of the overall CBD pedestrian system, the extent of the system at each level and the level of integration at the interface could affect the level of pedestrian activity in the system as a whole. The variations in CBD off-grade pedestrian systems can be grouped into four major categories; skyway with grade-level mall, skyway with grade-level mall and sub-grade system, skyway only, and sub-grade system with grade-level mall. 2.7.1 SKYWAY WITH PEDESTRIAN MALL The best example of such a combination exists in Minneapolis, Calgary and St.Paul. However in most of the cases, the grade level mall does not extend beyond a couple of blocks where as the skyway networks essentially spread over a large number of 53 blocks in the Central Business Districts. At Minneapolis the Nicollet Mall, eight blocks in length, was implemented before the Skyway System was incorporated into the CBD revitalization scheme. Special transit services were introduced in this two lane mall for trips within the CBD as well as for trips terminating in the suburbs. A system of skyway at the upper levels of the mall extend up to the parking ramps on the fringe of the mall. The skyway supplement the mall through out its length thereby integrating the grade level and upper level pedestrian areas. A similar system exists in St Paul where the ground level shopping arcades compliment the overhead walkway network. In Calgary too, the 8th Avenue Transit Mall coexist with the multilevel skyway network thus providing ample opportunities to the core area pedestrians to have a choice of routes depending on trip purpose (Shostack, 1978). 2.7.2 SKYWAY WITH SUB-WALK AND GRADE-LEVEL MALL Dallas and Edmonton offer the best example of elevated, grade level and below grade integrated pedestrian networks. After the completion of its network of elevated walkways, the system at Dallas would have nearly half of the total pedestrian system below and half above grade. The whole system has been very efficiently designed with below grade connections located near subway terminals to cater to the daily flow of transit users. In the historic districts, too, the system has been built below grade to avoid any incompatibility with the architectural character of the area. With the other areas having networks above grade level, the entire CBD pedestrian facility went well with the planning strategy to develop a multi-level City Center (Ponte, 1986). 54 2.7.3 PEDESTRIAN SYSTEMS ONLY ABOVE-GRADE Such systems are rare and exists in very few North American cities. Cincinnati and Hamilton are two of the examples where no large scale grade-level pedestrianization schemes have been implemented and the off-grade systems are restricted to the above grade network only (Podolske, and Heglund, 1976). The upper level networks take the entire burden of providing transport, commercial and other benefits. 2.7.4 SUBWAYS WITH GRADE-LEVEL M A L L Montreal has the largest network of this category of pedestrian system. The below grade pedestrian corridor network in Montreal is the largest in the world, connecting the subway rail transit terminals and transfer points to the commercial centers in the core area. Mostly privately owned, the shop lined corridors, which initially started from Place Ville Marie, were planned to cover almost 100 acres of the downtowns total 185 acres by 1990. The entire system consists of three principal axis; Place Ville Marie-Bonaventure, Place Victoria, and Les Terrasses. These climate controlled, enclosed malls and connecting passages supplement the pedestrian shopping malls at grade level (Brown, and Sijpkes, 1985; Shostack, 1978). Another good example of such networks is the subway network in midtown Manhattan, located in the heart of New York city's business, commercial and entertainment district (Benz, and Lutin, 1983). 55 2.8.0 Issues and Controversies Investigation of the development of the off-grade pedestrian networks would remain incomplete without a report on the numerous controversies which originated during the past two decades. Irrespective of their form of development the systems have come under severe criticism from various architects and planners. Important as they are, the issues could possibly have had considerable bearing on the route choice of pedestrians in the downtowns, which have a direct impact on the level of pedestrian activities inside the off-grade systems. Wherever the grade-separated systems coexist with pedestrian areas at other levels, and since most of the off-grade systems coexisted with grade-level malls, these issues assume even greater importance, due to the existence of some degree of competition between the facilities. The following subsections describe each issue in detail. 2.8.1 ORIENTATION The issue of orientation inside the off-grade pedestrian systems has been a major controversy over the last decade. It has been argued that the systems are not properly designed, that directional signs are not provided to the right extent and at appropriate locations, and that the systems are closed off from the outside environment because of which the users easily loose their orientation inside the system. People, as a result, may have to take longer routes to their destinations, thereby rendering the system inefficient. The problem is more serious in the privately built and controlled systems like the ones at Minneapolis and Montreal, where the private developers manipulate the physical design of the system to route the users through the maximum possible length of retail corridor for commercial benefits (Brown, and Sijpkes, 1985; Morphew, 1984; Robertson, 1988). Circuitous routes aggravate the problem even further. Such deficiencies could demonstrate 56 potential negative impacts on the usage level of the systems as the users would tend to remain out at grade level where loss of orientation is not experienced. Several research works, however, contradict the above assumptions. In a survey of elevated pedestrian systems in five U.S. cities- Minneapolis, St Paul, Cincinnati, Des Moines and Duluth- Robertson (1988) discovered that the percentage of users experiencing disorientation ranged from 7% in Duluth to nearly 30% in Minneapolis and Des Moines. However, such low figures can be explained by the fact that majority of the skywalk users interviewed were downtown workers, who, most of the time follow a regular path inside the system out of habit (Robertson, 1988). Effects of disorientation on the other categories of users including shoppers, promenaders and tourists have not been widely reported. 2.8.2 CONTROL AND REGULATIONS This issue has the potential of not only affecting the level of pedestrian usage of the system but also effecting the use of the system by selective user group. The magnitude and type of problem depends, to a large extent, on the type of system ownership and control. Most of the elevated pedestrian networks, except the one at Minneapolis, are built with varying degree of public-private cooperation, and the controlling rules and regulations are made by public authorities. On the other hand, the system in Minneapolis and Montreal are entirely private systems and the controls are mostly in the hands of private bodies. The problems addressed are essentially of two types. They are : (a) Those arising out of uncoordinated operating hours. (b) Those arising out of restrictive legislations. 57 2.8.2.1 OPERATING HOURS Uncoordinated operating hours, especially in the privately owned systems lead to considerable confusion. No where is it more evident than the system at Minneapolis where opening hours vary between 6.00 am and 9.30 am, and the closing hours between 5.00 pm to 10.00 pm (Robertson 1988). Such inconsistencies typically affect the core area residential population, who are affected by the inconsistent timing especially during morning hours, evening hours and during week ends. Night time activities in the central areas are strongly affected too. The system, as a result, gets efficiently used only during the daily office hours, and the retail areas inside the system, the most important commercial strategy, remain closed during the rest of the time (Morphew, 1984; Robertson, 1988). Even with properly coordinated operating hours nighttime activities can be affected as in the Dallas multilevel system, where the system shuts down after 6.30 pm (Ponte, 1986). In such cases, the system is basically reduced down to a 9.00 to 5.00 pedestrian corridor catering primarily to the downtown workers, rather than being the Indoor Town Center which was the original idea behind developing such facilities. 2.8.2.2 RESTRICTIVE LEGISLATIONS Restriction of activities on the off-grade pedestrian systems could lead to possible advantages as well as disadvantages. Restrictive ordinances have virtually been issued in all publicly controlled systems. In the St Paul skyway sitting, kneeling, lounging or otherwise reclining on any skywalk floor and step is strictly banned. Even playing of radios or tape recorders are prohibited. Similarly loitering is prohibited inside the Des Moines system in addition to a complete prohibition on persons to collect, assemble or group together. Activities like peaceful picketing, public speaking and leafleting are allowed, but limited to maximum of two persons at a time (Robertson 1988; Morphew, 58 1984). Such legislations could go a long way in making the systems safe, and such perceived safety could encourage more people to use the system networks. On the other hand, these regulations may limit the type of users to only workers and high income shoppers. The prohibition of such activities which are the essence of urban social life and in which people reflect their support for freedom and democracy, could lead to a draining in the perceived level of individual freedom inside public domains. The problems become more acute in the privately controlled systems where the rules are determined by each private property owner, leading to severe confusion regarding the degree of freedom in different segments of the system. Such a process allegedly leads to a selective adoption and shift of activities from the grade level and suburban activity system into the, so called, indoor city-center. Such selective and partial shift of activities could generate serious handicaps in making the off-grade, indoor city a true urban public space, since to be public, any facility should provide for the most important public desire; freedom, and be conducive to all activity types no matter how undesirable they are in the commercial sense. "The new atriums and enclosed malls have the potential to become social places which make a significant contribution to the quality of urban life, especially for those with limited mobility. The level to which this potential is met, however, depends on the extent to which they are truly public environments" (David, Michael and Pieter, 1985). The teenagers in the street corners, the vendors, old aged promenaders, and for that matter every other commercially undesirable urban actor and activity are a legitimate part of the entire activity system, and have equal participation in making the urban spaces truly public. The issue, which have been debated for long in academic circles (Brown, and Sijpkes, 1985; Morphew, 1984), could lead to possible negative impacts on system performance. There are several other legal complications which aggravate the problem in privately owned systems. The owners of privately owned domain could legally withdraw, at any time and at their own discretions, their invitation to the public (Krushelnicki, 1985). 59 This could lead to some degree of voluntary curtailment of freedom by the users. The private security guards of these areas could also prove to be detrimental to free use of the facility, since they tend to discourage all categories of non-commercial use of the facilities. Quite frequently, the question of public areas in private domains has been raised (Krushelnickie, 1985; Brown, Maclean, and Sijpkes, 1985; Robertson, 1985; Warner, 1985). The questions arising out of these controversies directly addresses the impact of such legislations on the level of system use, and the system's performance as a public place. 2.8.3 SOCIAL STRATIFICATION A large number of factors including the regulations, control, and physical configuration of the systems lead to some degree of social stratification based on income, profession, age, profession and racial origin. In Minneapolis, St.Paul, Cincinnati, Duluth and Des Moines there are more women, white-collar workers with high income using the system as compared to men, blue-collar workers, shoppers and low income users (Robertson, 1988). This suggests a possible stratification of users in the multilevel systems, where the upper-level system seems to cater mainly to the upper income strata and white-collar workers. Commenting on social stratification, Whyte (1988) too refers to various systems where the grade levels remain left out for students, minority groups, and other low income residents who especially cannot afford to go to the upper level commercial establishments. Restriction of activities (section 2.8.2.2) intensify the problems even further. 60 2.8.4 PHYSICAL ENVIRONMENT : PEDESTRIAN FACILITIES AND AMENITIES Though not as controversial an issue as the other issues, the importance of physical planning has been reinforced by many researchers. The issues are wide ranging and almost completely cover the urban design of public indoor spaces. As an example, among the design elements crucial for creating such environments, the quantity, comfort and placement of seats in public areas is important. Seats provided in circulation areas could deter the users from using it for a long time. Similarly, seats located near entrances are mostly favoured by the users. Movable seats with backrests serve better as compared to fixed seats without one. These are but a few of the host of other design considerations which determine the degree of facility usage. Vegetation, daylight, fountains, public washrooms, telephones, entertainment areas, festivities, information boards, directional signs, view corridors, shop fronts, advertisement windows, eating areas, and many other similar elements determine the potential of the environment to host urban social activities (Levinson, 1986; Brown, Maclean, and Sijpkes, 1985; Brown, Maclean, and Sijpkes, 1986; Patricios, 1979). An excellent example is the St.Paul Skyway System where the design of pavings and graphics as well as the overall design of the elevated bridges are kept uniform throughout the upper level network, and match with the grade-level paving. Such uniformity gives a public appearance to the entire system, thereby guaranteeing, passively, freedom to the users (Jacob, 1985). Similarly the presence of temporary private facilities like push carts selling drinks and hot dogs could introduce some degree of publicness even in privately controlled environments. Other such desirable elements include open areas at nodes, indoor courts or atria, and parks and gardens (Brown, Maclean, and Sijpkes, 1985). Although, effects of physical design of the indoor environment on system usage is not widely reported, researchers have been critical about the environment in privately controlled domains. 61 2.8.5 ACCESSIBILITY The accessibility, of the system from the grade level could be an important determinant of the level of pedestrian activity. People tend to shy away from vertical connectors which are located deep inside buildings, since they perceive it as an intrusion into private territories (Robertson, 1988). Similarly, easy access to the upper levels by the old and handicapped users depend on the availability of mechanical vertical connections including lifts and escalators to other levels, and the provision of other facilities like ramps and leveled corridors. The number and location of such vertical connections at the interface could affect the usage pattern of the off-grade systems. The location of the vertical connectors in relation to transit transfer points and parking lots is vital to system usage. Again, a contradictory finding that very few of the skyway users face difficulty in locating entrances, the maximum being at Minneapolis and Cincinnati where only 16% of the surveyed users reported facing difficulty (Robertson, 1988), has been explained by the fact that possibly most of the users in these cities were workers with regular fixed trip route patterns. 2.8.6 SAFETY AND SECURITY The perceived safety of pedestrians inside the systems, as opposed to actual safety, could also affect the pedestrian volume. The restrictive ordinances (section 2.8.2.2) could provide a certain degree of perceived safety to the system users. However, a more visible system of safety operations including uniformed policemen, special security staff, visible security booth at intervals, closed-circuit T V monitoring, straight corridors and passages, appropriate daylight levels, and system transparency to other levels, among others, could possibly affect the perceived level of safety inside the system. To what extent this 62 perception affect system usage is not known. However, the issue has been addressed by various authors (Morphew, 1984; Wiedenhoeft, 1975), and such security systems, if efficiently provided, could help women, old aged and handicapped users in using the system more often, and also could encourage general public to use the system during off hours. 2.8.7 AESTHETICS: VISUAL COHESIVENESS It has been frequently argued that the architectural character of the downtown buildings is destroyed and violated by the skyway bridges, due to the difference in material and design of the bridges and the buildings. This problem is more evident in historic districts. In addition, the skyway bridges have been reported to have blocked the view of landmarks and vistas from the street level, though the elevated bridges provided vantage viewing points for the skywalk users (Morphew, 1984; Robertson, 1988;, Robertson, 1987). The problem has been partially tackled in St.Paul where the uniform design of the bridges for the entire system, relegates the skyway network to the background (Jacob, 1985) thereby reducing the visual damage to a minimum. In Dallas, the Historic districts and view corridors have been provided with well connected underground networks of pedestrian corridors to eliminate both the problems in the downtown pedestrian system (Reis, 1982). 63 2.8.8 COST - BENEFIT An important argument by Whyte (1988) pertained to the level of pedestrian activities inside the system, which allegedly remain exclusively limited to a few segments around the focal points, hubs, high density corridors connecting transit terminals to work areas, and retail areas. The pedestrian volume and activity recedes drastically as one moves away from these areas. However, the mechanical systems including air-conditioning, heating, escalators and lighting facilities of the entire system run throughout the official operating hours, thereby making the cost of the system considerably more than the benefits derived. The cost-benefit issue is crucial in the sense, that the benefits derived are directly related to the level of pedestrian activity inside the system. Under-use of certain segments are related to the route choice made by the users, which is a function of the trip purpose. And the route choice is affected by all the other issues described in this section. 2.8.9 COMMERCIAL SHIFT The last major issue has been the rise in commercial activities inside the system, after the implementation of the scheme. There has been no study to ascertain whether the economic upswing is experienced in the entire central business district or whether such upswings are a result of a shift of commercial activities from other commercial corridors in the downtown not connected to the system, or from the areas inside the system with lesser degree of pedestrian activities to the densely used segments of the system (Whyte, 1988; Robertson, 1988; Brown, and Sijpkes, 1985; Ponte, 1986). 64 2.9.0 Skyway System and System Usage If the observations on performance of the grade-separated pedestrian systems in history (especially the Perugian and the nineteenth-early twentieth century systems) could gain some ground for validation, consequences of the various issues raised by academicians on the mid-twentieth century developments could prove to be nothing but detrimental, since most of the issues seemingly have a direct negative impact on system usage. In turn, the issues as well as their impacts address the question of the Skyway System performance as an urban public space. But unlike the systems in history, debate regarding the relationship between system performance as a public space, and its success as a transportation planning tool, does not prove to be necessary in this context, since the mid-twentieth century systems originated with objectives encompassing both the spheres of the urban activity system. Past research endeavors provide evidence in support of the success of the Skyway System as a transportation planning tool. Other studies reported little negative impact of the system on the commercial climate of the core area (chapter3). That brings into focus the impact of the issues on the third objective area; facility usage pattern, and the system performance as a public place. Few research works have been published demonstrating the true presence/impact of any of the issues in todays Skyway System, nor has any assessment been made to evaluate the system as an urban design tool. Although most of the research works have been carried out in the realms of transportation planning, two major surveys conducted during the past two decades, where indirect assesments have been made in the areas of urban design, warrants further attention for the purpose of this thesis. The next chapter begins with a discussion on the two surveys; the first conducted by Barton-Aschman Associates in Minneapolis and St.Paul, and the second by Kent A. Robertson in five major U.S. cities. 65 CHAPTER THREE LAYING DOWN THE GROUND RULES 3.1.0 Research Endeavors During the Past Two Decades Scientific research works on the North American Skyway System have not been widely published. A majority of published literature on the Skyway System, some of which were referred to in Chapter Two, have had their foundations based on general observations and phenomenological studies, or were reports of system development/developmental strategies and urban development policies. Efforts at scientific assessment of the system during the past two decades is limited to two major surveys for the purpose of this study. The first survey was carried out in December 1974 by Barton-Aschman Associates in the central city areas of metropolitan Minneapolis and St.Paul, U.S.A. (Podolske and Heglund, 1976; Heglund 1982). They undertook a one-year long pedestrian counting program on the Skyway Systems of the two neighboring cities to gauge their overall effectiveness. No new skyway link in either system opened during that twelve month period, as a result providing an accurate picture of various aspects of the System's usage pattern. The second survey was conducted during Summer'1988 by Kent A. Robertson (Robertson, 1988). In order to examine a wide range of economic, social, planning, and design issues, Robertson carried out a comparative evaluation of the Skyway Systems in five North American cities; Cincinnati, Des Moines, Duluth, Minneapolis, and St. Paul. The assessment was made on the basis of system usage, access and orientation, safety, economic activities, design and aesthetics, and the system's potential as a downtown redevelopment tool. The 66 entire survey included three major phases of data collection. The first phase included several days of observation of the system in each city with particular focus on usage, design, directional signs,, -obstacles, activities, and any unique characteristics. It also included several interviews with city officials involved in skyway planning, design, and implementation. The second phase involved the creation of an inventory of major land uses present on each block, and of every retail and service establishments located on the skyway corridors. In the last phase, approximately one hundred skyway users in each city were surveyed through a close-ended questionnaire to ascertain their diurnal usage pattern, their perception concerning a wide spectrum of skyway features, and problems encountered by them. The survey was carried out between 8:30 a.m. and 6:30 p.m. on summer weekdays. 3.1.1 SURVEY RESULTS 3.1.1.1 LAND USE AND ECONOMIC IMPACTS The result of the surveys rendered a positive image of system performance. The system's impact on land use and economic development proved nothing but positive. The Barton-Aschman Associates survey of 1974 revealed the existence of a dramatic increase in the number of financially successful establishments totally dependent on skyway patronage. As of 1973-74 there were as many as 120 establishments directly fronting on the system, in the two cities (Table 3.1). The success of the system can be gauged by the relative rental levels of that period. During the formative stages of system development, the rental charges on the second level (skyway level) were marginally to considerably below the corresponding rates at grade level. By 1974 the rental charges on the Skyway System were equal or higher than those of the grade level spaces (Heglund, 67 Number of Establishment Percent of Establishment Mnsp. St.Paul T o t a l Mnsp. St.Paul T o t a l RETAIL : 25 10 35 37 18 28 Departmental Store Mens & Womens Cloth G i f t /Card /News Shop Other R e t a i l Sales 3 (3) 11 5 6 1 (1) 2 4 3 4 (4) 13 9 9 FOOD : 9 7 16 14 12 13 Restaurants Food & Beverage Sales 7 2 2 5 9 7 SERVICE : 29 37 66 44 65 54 Banking Stockbroker Insurance Barber & Beauty P r i n t e r Real Estate T r a v e l Agency Other Services 14 (7) 5 1 1 3 1 1 3 7 (6) 5 5 6 2 3 3 6 21 (13) 10 6 7 5 4 4 9 OTHER : 3 3 6 5 5 5 Apartment Bldgs. Federal Court Bldg Clubs Hotel 1 2 (2) 1 1 (1) (1) 1 1 (1) 1 (1) 2 (2) TOTAL 66 (12) 57 (9) 123 (21) 100% * NOTE : Number i n brackets indicate " l a r g e " establishments defined as covering 1/4 block or more, or most of one b u i l d i n g . Survey taken June, 1975 Table 3.1: Summary of Skyway Establishments Connected Directly to Skyway System. Data Source: Podolske and Heglund, 1976. 68 C i n c i n a t t i DesMoines Duluth Minneapolis St.Paul To t a l establishments* 62 36 49 179 162 Total contiguous Blocks 14 19 15 27 25 Establishments Per Block 4.4 1.9 3.3 6.3 6.5 Blocks with (%) 5+ Establishments 6 (42 •9) 3 (15.8) 2 (13.3) 14 (51.9) 10 (40.0) Blocks with (%) 10+ Establishments 1 (7 •1) 0 1 (6.7) 8 (29.6) 5 (20.0) Blocks with (%) 20+ Establishments 0 0 1 (6.7) 1 (3.7) 3 (12.0) * For mall-type settings, only skywalk l e v e l establishments counted Table 3.2: Distribution of Skyway Establishments. Data Source: Robertson, 1988. C i n c i n a t t i DesMoines Duluth Minneapolis St.Paul R e t a i l Goods 26 (41. 9) 9 (25.0) 22 (44.9) 75 (41. 9) 68 (42. 4) F i n a n c i a l and Business Services 8 (12. 9) 13 (36.1) 8 (16.3) 42 (23. 5) 36 (22. 2) Food 16 (25. 8) 7 (19.4) 10 (20.4) 44 (24. 6) 25 (15. 4) Personal Services 6 (9. 7) 5 (13.9) 6 (12.2) 18 (10. 1) 27 (16. 7) Other 6 (9. 7) 2 (5.6) 3 (6.1) 0 6 (3. 7) Table 3.3: Types of Skyway Establishments (In Percentage). Data Source: Robertson, 1988. 69 1982; Podolske and Heglund, 1976). The findings of the 1988 survey by Robertson adds to the previous finding. By 1988, the total number of establishments on the Minneapolis and St. Paul Skywaj' Systems had climbed up to 341 (Table 3.2, Table 3.3). In the twin cities, ten or more establishments could be found on at least twenty percent of the blocks, suggesting that the establishments not only served the passing pedestrians, but also, probably, generated additional traffic flow as well, resembling in the process a multi-block urban shopping mall. While retail sale in downtown St.Paul continued to increase, the share of the street level establishments decreased to a point where over three-fourth of the downtown retailing in 1988 took place on the skyway level. In addition, the average annual lease rate was $10.58/ft^ as compared to the grade level rate of $8.90/ft^, suggesting considerable commercial success of the Skyway System at St.Paul. Similar trends were found in the other cities, although to a lesser degree (Robertson, 1988). 3.1.1.2. SYSTEM USAGE The system usage pattern raises a few significant issues. In the survey conducted by Barton-Aschman Associates, it was revealed that the systems at Minneapolis and St.Paul received their heaviest usage during the mid-day period. On a typical spring day, approximately l/4th (23% in Minneapolis and 28% in St.Paul) of all daily skyway bridge crossings occurred during the noon hour (Fig.3.1) and nearly half of all crossings occurred in the three hour period between 11:00 a.m. and 2:00 p.m.. This heavy use of the Skyway System during the mid-day period generally reflected overall inter-block pedestrian activity in the CBD. A subsequent day long monitoring of a typical skywalk bridge (Fig.3.2) showed that the attractiveness of the skywalks as opposed to at-grade crossings remained reasonably constant during an entire working day. Furthermore the skywalk monitored attracted the lowest percentage of the total pedestrian population (not less than 40%) 70 during the morning and evening rush hours, when most pedestrian activity was between place of work and motor vehicle intercept points (parking garages and bus stops) (Heglund, 1982). Also, the greatest relative skywalk attractiveness occurred during mid-morning (9:30-10:30) and mid-afternoon (3:00-4:00) when it could be assumed that inter-building travel constitutes a greater share of total pedestrian activity (Fig.3.2) (Podolske and Heglund, 1976). 2,000 1,800 1,600 1,400 ui i 1.200 S 1.000 £ 800 c 600 UJ 4 0 0 CD S 200 z I V \ M l 'IrSl 1 I J \ A J I V / r.p V 7 8 9 10 11 12 1 2 3 4 5 6 A M P M T I M E OF DAY Fig 3.1: Skywalk system usage during weekdays. Data Source: Heglund, 1982. 100 90 BO 70 60 50 40 30 20 10 0 7 a AM 3 : 9 M t l 12 1 PM TIME OF DAY 3 4 5 6 Fig 3.2: Percent using skyway on Typical Weekday Data Source: Heglund, 1982. It is quite clear from the above data that maximum pedestrian activity in and around the survey area occurred during the mid-day period, since skywalk attractiveness 71 as opposed to at-grade crossings remained more or less constant. It is also clear that more pedestrians did not prefer skywalks for their trip from modal interchange points to the work places, probably due to inadequate accessibility to. the system from the modal interchange points. Information regarding this was not collected in the survey. Also, no information is available on the then skj'walk user profile. Without any information on the skywalk user profile, two distinct possibilities arise out of the above mentioned data. First, all kinds of pedestrian activity including eating, shopping, meeting, visiting, promenading, and conducting business, by all categories of people, occurred mostly during the midday period. Second, and more probably, most of the pedestrians generating the mid-day traffic surge in the survey area were downtown office workers, on their way to lunch hour eating and shopping (11:00 a.m. to 2:00 p.m.), and the rest of the type of activities, by all other categories of people, on the skyway system, remained low during other hours of the day. If the general temporal pedestrian activity pattern for the entire central city were to be considered in comparison to the usage pattern of the surveyed skyway segment, two distinct possibilities can be pictured. (a) Case I: General pedestrian activity pattern of downtown coincides with the pedestrian activity pattern on skyways (Fig.3.1): People other than downtown workers did not prefer downtown establishments and hence did not constitute any significant proportion of the total downtown pedestrian trips. This, in turn, would imply that the downtown revitalization policies were not exactly successful. (b) Case II: General pedestrian activity pattern of downtown does not coincide with the pedestrian activity pattern on skyways: People other than downtown workers did come into the downtown for shopping, leisure, business etc. but did not prefer the upper level establishments, wherever the alternative 72 existed. In other words, a majority of the daily users of the skywalk level establishments were downtown workers, using the system for access. A pedestrian survey was conducted in 1972 in the central area of St.Paul, Minnesota (Heglund, 1982). The survey included all pedestrian trips in the downtown, and not just those occurring on the Skyway System. The result of the survey (Table 3.4) has more conformity with the second implication (Case II) since, it revealed that quite a substantial percentage of people (approximately 50% of all surveyed) did came into the downtown, both, during noon hour and P.M. hours for purposes totally unrelated to work, though workers constituted a major user group. The poor degree of usage of the Skyway System during P.M. hours hints at the possibility that users other than workers may not have been using the skyways for their activities (Fig.3.1). The above two surveys did not provide any conclusive evidence, but did hint at the possibility that the Skyway System might have had been host to a relatively selective user group (workers) for a selective activity type (lunch hour traffic) as compared to other grade level public areas. After the 1988 survey, Robertson raised identical doubts about skyway performance. When asked for their primary purpose downtown that day, it was found that the total 502 respondents, in five cities, could be categorized as 51 percent workers, 20.9 percent shoppers, 21.5 percent visitors/tourists, 1 percent residents, and 5.6 percent other purposes. Also 46.4 percent of the total respondents were females and 26.3 percent were non-whites (Fig.3.3, Fig.3.4, Fig.3.5) (Robertson, 1988). The survey had, but, severe limitations, questioning the objectivity of the assessment. The limitations arise out of the non-random character of the survey and the absence of survey results for weekends and seasons other than summer (Robertson confesses about these limitations in his article, "Pedestrian Skyway Systems: Downtown's Great Hope or Pathway to Ruin", 1988). However, it does suggest a massive overrepresentation of downtown workers among the 73 skywalk users, on any typical work day. The handicaps arising out of the non-random character got reduced by another supporting data assessment. Robertson, through his -questionnaires, gauged the perceived profiles of skywalk users in the five cities (Table 3.5). The highlight of the results was that among the perceived skywalk users, workers outnumbered shoppers, white-collar workers outnumbered blue-collar workers, and high income users outnumbered modest income users, thus limiting the categories of user groups unlike any true public place. The results were more or less consistent in all the five cities surveyed. The two results, due to the non-randomness of survey procedure does not scientifically suggest any true differences, but, quoting Robertson, "predicted behavior and perceived users....are only surrogates for actual behavior and users". A conclusion can be drawn, thus, that among the system users during summer week days, between 8:30 a.m. and 6:30 p.m., workers probably outnumber shoppers by a large margin, though the true figure cannot be ascertained. This raises some questions regarding the use of the system as a public place, its attractiveness as a downtown facility, and its potential as a central city revitalization tool. Another piece of survey result from the 1974 survey by Barton-Aschman Associates, justifies attention. The survey established that under the most favorable conditions, a well designed Skyway System could be expected to attract at least l/3rd of all inter-block pedestrian travel. During typical summer conditions, an extensive Skyway System might be expected to attract approximately one-half of all inter-block pedestrian activity. And under freezing conditions (Nov.-March), it appears that an extensive Skyway System could be expected to at least attract 2/3rd and 3/4th of all inter-block pedestrian flow (Table 3.6). This implies the possibility of the Skyway System deriving a large part of its attractiveness from the external weather condition, as compared to the system's planning and design of the physical environment and facilities. 74 NOON HOUR P.M.PEAK HOUR DESTINATION WORK HOME SHOP WORK HOME SHOP (1) (2) (3) (4) (5) (6) (7) Work 9 30 Home 6 10 Shop 15 0 Business 21 20 Eat 43 40 Other 6 0 14 7 50 17 9 77 0 42 43 3 38 25 9 9 0 4 18 1 0 4 7 3 12 18 Table 3.4: Percentage Distribution of Person Trips by Purpose. Data Source: Heglund, 1982 (Percentage indicating type of person more lik e l y to use Skywalks) Cincinatti Des Moines Duluth Minneapolis St.Paul WOMEN (as opposed to men) (54.2) 57. 4 56.3 50.0 49.5 57.6 WORKERS (as opposed to shoppers)(78.5) 69. 3 94.2 90.0 69.7 68.7 WHITE-COLLAR WORKERS (as opposed to Blue-Collar workers)(76.9 72. 3 94.2 66.0 77.8 73.7 HIGH INCOME (as oppo -sed to Modest Income)(74.3) 73. 3 92.2 68.0 69.7 67.7 a. Total survey = 502 (99-102 per city) b. High income=$50,000 or more; modest income=under $10,000. (The figures in parenthesis in the f i r s t column are the overall figures) Table 3.5: Perceived Profiles of Skywalk Users. Data Source: Robertson, 1988. PROFILE OF SYSTEM USERS SKYWAY SYSTEM WORKERS 61 61% SHOPPERS 20.9 21% OTHER PURPOSES 5.6 6% RESIDENTS 1 1% VISITORS/TOURISTS 21.5 22% PRIMARY TRIP PURPOSES Fig.3.3: Primary Trip Purpose of Skyway Users. Data Source: Robertson. 1988. PROFILE OF SYSTEM USERS SITWAT SYSTEM PROFILE OF SYSTEM USERS S I T WAT SYSTEM n i u i i 44.4 44% SIX raSTRllDTlON RAM DHTJUIDYIOH Fig.3.4: Sex Profile of Skyway Users. Fig.3.5: Race Profile of Skyway User. Data Source: Robertson, 1988. Data Source: Robertson, 1988. 76 PERCENTAGE USING SKYWALKS MONTH HIGHEST LOWEST AVERAGE (1) (2) (3) (4) January 90 62 76 February 86 48 72 March 85 50 71 A p r i l 78 62 66 May 75 36 56 June 76 25 46 Ju l y 79 26 47 August 66 30 47 September 76 36 52 October 76 32 60 November 95 24 68 December 82 51 67 Based on t o t a l crossing of skywalk and crosswalks Table 3.6: Monthly Usage of Skyway Versus At-Grade Pedestrian Crossing. Data Source: Heglund, 1982. From all the previous works several conclusions can be clearly drawn. From the transportation planning viewpoint the Skyway System could be termed as an innovative solution to many problems, including traffic congestion, pedestrian safety, reduction in accident rate, vehicle delay savings, and vehicle fuel savings. The system would probably rate high in any cost-benifit analysis (Ponte, 1986; Heglund, 1982; Podolske and Heglund, 1976). However, as a public place many questions remain unanswered. Most of the published literatures provide hints, but no conclusive evidences. How does the system rate in comparison to other urban public places? Is there any true overrepresentation of certain user category or activity among the entire population? How does the system affect or improve the life of urban dwellers? What type of usage pattern does the system experience during weekends and seasons other than Summer? Do people use the system more for access than any other activity type? Are all the negative attributes of the system, 77 addressed by various authors (section 2.8.0), truly existent? Questions regarding the Skyway System could be limitless and scientific research endeavors to ascertain true usage pattern are not widely published. This thesis, in the subsequent chapters, would attempt to address some of the above issues for an initial scientific assessment of the Skyway System in the North American context, and to build a statistical foundation for future research endeavors. 3.2.0 The question Without any reliable data from previous research efforts, the primary question, directly or indirectly raised by authors during the last decade, pertains to the role of the upper level urban area created by the skyway network, and its performance in comparison to other similar urban landforms. Although issues have been raised concerning inappropriate management strategies and user amenities, they remain more or less invalid without any supporting evidence regarding their effects on the usage pattern of the Skyway System. Robertson's research suggests only the possibility of the Skyway System not truly performing as urban public places, but without any reliable evidences, the question remains unanswered. This research studied skyway performance as a public place, in isolation from the transportation planning and commercial viewpoints, which in turn led to two basic questions forming the objectives of the study : (1) Considering the fact that the Skyway System in most of the cities have climate controlled, automobile free, multi-block urban mall type physical environment with extensive categories of establishment on the network, and moreover, since the network was incorporated into the urban fabric through incentive zoning and ordinances with a clear objective of creating public 78 places, how effectively arc these areas performing as urban public places in comparison to other similar public spaces; the pedestrian mall which were implemented ivith identical objectives (section 2.4.1) and were regarded as successful urban design strategies (section 2.4.3), and the conventional pedestrian sidewalks on city streets? (2) How far does the physical design of the skyway environment explain the similarities or differences arising out of the above comparison? 3.3.0 The Scope of the Study Three areas of concentration were derived from the first part of the stated objective. They pertained to the two most important indicators of performance of any public place; viz. the type of people using it, and the type of usage they are putting it to. Four areas were derived from the second. The areas from the first part included : (1) The types of people using each of the facilities under survey. (2) The temporal usage pattern of each user type on the facilities. (3) The activity (trip purpose) of the users, or their primary reason for using the facility. The second part of the study concentration included the assessment of the spatial use message rendered by the elements of the physical environment of each facility, and explanations relating the physical environment with the usage pattern derived from the first part of the study. The four areas of concentration were : 79 (1) The spatial configuration of each facility. (2) Physical characteristics of the defining surfaces and the edges. (3) Level and type of facility furnishing. (4) Level and type of social activities on each of the surveyed facilities. 3.3.1 VARIABLES IN THE DESIGN Nine variables were identified to be included in the design; user age group, sex, ethnicity, role, level of physical impairment, grouping pattern, purpose, stationary activity type, and body posture. To form a basis for the formation of categories within each variable, the major issues identified by previous authors (section 2.8.0) were adopted. The issues included accessibility, control and regulation, safety and security, orientation, and user amenities and facilities. Classifications were made as long as at least one of the issues introduced the possibility of potential interaction within the unclassified category. The details of each categories will be discussed in the subsequent sub-subsections (the final classification was made after two pilot surveys carried out by the author in Vancouver, B.C. , before the collection of actual data. The pilot surveys were conducted on Granville Mall with its outdoor public areas, and the City Square Mall with the indoor, climate controlled spaces, during summer weekdays. The changes made to the classification after the pilot surveys were rechecked for possible exclusion of any variable category and/or duplication, during an initial informal observation in downtown Minneapolis, U.S.A. , where the final data was collected). Beside the nine user variables there were four other variables in the design; day, site, location, and time, which will also be discussed below. 80 3.3.1.1 USER AGE GROUP . On the basis of age the users were classified into four major categories; Child (less than 13 years), teenagers (13 to 19 years), adult (20 to 64 years), and the elderly (above 64 years). One of the basis of classification is the major change in lifestyle among the four groups. The teenagers, with their visible presence in malls and street corners, usually hanging out, represent a marked difference in lifestyle from the younger age groups who remain most of the time under family influence. Similarly, the elderly and retired also maintain a regular presence on the mall, either grooming or shopping around. Beside the marked difference in lifestyle, each of the age groups have theoretical potential interaction with some of the issues. The elderly could be affected by improper accessibility (including vertical transport modes), control and regulation (over stationary activities on the skyway networks), safety and security (from muggers on isolated segments), orientation (improper directional signs), and the user amenities (possible lack of furnishing elements or rest areas). Regulations on the network could affect the teenagers too, with prohibition on hanging out, sitting, and loitering on the skyway segments. The use message of the environment could also affect the presence of teenagers. The impacts of user amenities and facilities, and accessibility on the child could be as much as on the elderly. Since none of the issues seem to have any impact on persons within 20 and 64 years, no further classifications were made. 3.3.1.2 SEX CATEGORIES The natural dichotomy of the sexual groups led to the classification of users into two basic categories; male and female. The females, unlike the males are significantly more susceptible to physical danger on remote areas of the network, and are more likely to feel 81 safer due to strict regulations on the segments which discourages antisocial elements from using the system for their activities. 3.3.1.3 ETHNIC GROUPS The two major visible ethnic groups, the White and the Black, constituted two of the three ethnic groups. The third group included all the other visible minorities. Classification of ethnic groups was based on White's (1988) observation that the Skyways act as social barriers where the upper level areas are frequented more by the Whites (the dominant and relatively wealthier group), with the other minority groups mainly using the grade level areas. Owing to the proportionately insignificant number of each visible ethnic minority groups, all those groups were included in the 'others' category. 3.3.1.4 ROLE (USER TYPE) After considerable modification and rearrangement to the initial classification, the users were divided into six categories of role type; workers, visitors, tourists, security staff, mother/father with babies, and others. The dominant categories are the workers, visitors, tourists and others. The first three group categories are self explanatory. The 'others' category constituted all the people who did not fall in the rest of the categories. That includes, among others, the vagrants, beggars, low class hangouts (people more probably on social welfare schemes), pimps and prostitutes, and the punks. Worker as a dominant group was included due to the expressed notions of a few authors, including Robertson 1988, that they constitute the most frequent users of the system. Visitors and tourists are potentially affected by all the issues, from accessibility to orientation, and more importantly, their proportional representation is a valid clue in the assessment of the system's success as a downtown revitalization measure. The visible presence of the 82 security staff provides an apparent notion of safety to the system users, especially the female, elderly, and tourists, and hence was made a special group. The group 'mother/father with babies', although a duplication of the visitors and tourists categories, were made into a special group to assess some important aspects on the surveyed systems including accessibility, and user amenities & facilities. Later, during the data analysis, all users in this category were properly reinstated into their original groups; visitors, tourists, or any other group type. 3.3.1.5 THE PHYSICALLY IMPAIRED For exactly identical reasons as the 'mother/father with baby' grouping, all physically impaired users were recorded in their proper category to assess the accessibility aspect of all systems, as well as user amenities and facilities. According to the degree of physical impairment, the users were categorized into user with cane/crutches, and the ones on wheel chair. 3.3.1.6 GROUPING PATTERN OF USERS Categorization of grouping types of the users were mainly based on user needs and amenities, and consideration of perceived safety. The grouping patterns also implicate the relative usage type and the success of the development objectives. Accordingly five major groups were identified; alone, couple of the same sex, couple of the opposite sex, people in groups of three or more, and people with families. Consideration of safety could likely affect the route choice decisions of people who were alone, same sex female couples, and visitors/ tourists with families. Similarly, people with families/children are more likely to need user amenities more than other groups. The couples of opposite sex groups, especially young lovers also, at certain instances, are likely to get affected by the physical environment. 83 Moreover, a urban public place as a downtown attraction should be able to attract people in groups, especially visitors with families and large tourist groups. All such factors led to the final classification. . 3.3.1.7 PURPOSE (MOBILE ACTIVITIES) One of the main indicator of system performance is the activity of the users. A public place should be conducive to large varieties and degree of user activity types. The type of activity occurring is largely dependent on physical design and provisions as well as the management strategies. After initial categorization and subsequent modifications the activity types were categorized into seven major groups; access/walking, shopping, promenading, to work/working, recreating, searching, and hanging out. The 'searching' category constituted all the people disoriented, and was included to test the notions of some authors (Robertson, 1988) that the skyways seriously disorient the users. The 'to work/ working' category included all the people found working within the survey area. 3.3.1.8 ACTIVITY (STATIONARY ACTIVITIES) Another major type of indicator of system performance is the level and variety of stationary activities. The arguments for the classification are the same as those for the mobile activity types. Seven categories were identified after initial studies; watching/grooming, talking, eating, reading, performing, waiting, and meeting. 3.3.1.9 BODY POSTURE Body posture goes hand in hand with stationary activity types, and is a good indicator of the deficiencies in the physical design of the environment, the environment's 8 4 attractiveness to mobile and stationary activity types, and the impacts of facility management strategies on user activity level. Five distinct postures were identified; sitting, standing, lying, leaning, and kneeling/squatting. 3.3.1.10 DAY OF THE WEEK The variable 'day' had two distinct categories; weekday and weekend. The usage pattern of the systems are likely to differ for weekdays and weekends. For instance, during weekend, street side parking in most of the central city areas remain free of charges. This factor coupled with the reduction in automobile traffic on downtown streets have the potential to attract large number of suburban dwellers to downtown establishments. In addition, special recreational activities and functions are mostly arranged during weekends. The absence of workers on streets and sidewalks constitute another deviation from weekday scenario. The composition of user types and activity types, thus, is likely to experience marked variations between weekdays and weekends, and hence the categorization was made to avoid any potential statistical interaction. 3.3.1.11 SITE As per the stated objectives, three different site types were identified, each representing a different type of pedestrian facility. The first type was the Skyway System including the skyway bridges, corridors, passages and links. The second category was the pedestrian mall, and the last category was the conventional pedestrian sidewalk along streets. Two important criteria were considered for site selection. First was equal accessibility. The survey points were designed to be located at such places only, where alternative of using any of the three systems would be available to the user. The second criteria was the land use. Since most of the mall edges are lined with retail establishments, 85 and since retail/business establishments constitute a major attraction to the users, the survey points were designed to be located only in those areas where a considerable portion .-- of the edges would be defined by retail/business establishments. 3.3.1.12 LOCATION The variability in survey location within each system was designed mainly for time sampling and location sampling considerations to strengthen the external validity of the survey results. In the true sense, it does not constitute a variable, and discussions on variability of location will be made in detail in the section on sampling procedure. 3.3.1.13 TIME The variability of time period of observation was also designed to ensure external validity of survey results, and hence will be discussed later in this chapter. 3.3.2 VARIABLES K E P T CONSTANT Many of the variables were designed not to vary, and their variability was not considered. They include the city of survey, weather, the season, the form of system development, and the system's composition with other pedestrian schemes. 3.3.2.1 THE CITY The city for carrying out the survey work was limited to one, mainly due to two considerations. The first factor was the limited funds available for undertaking the project. 86 The second was the additional amount of time, and manpower needed for including variability in city and carrying out the survey work. 3.3.2.2 THE WEATHER Rough weather increases the attractiveness of indoor public places, which could affect the usage pattern of the systems temporarily. Ignoring such an impact area could affect the objective of the study. To avoid the interaction between weather condition and system usage, the data recording exercise was designed to be conducted only during fair weather conditions. Fair weather condition excluded rain, high winds or storm, and excessive heat. 3.3.2.3 THE SEASON Since the primary objective of the study was the comparative analysis of the Skyway System with two other systems both of which are outdoor, it was essential to exclude all Skyway System relative attractiveness arising out of cold external weather conditions (section 3.1.1.2). The climate controlled environment of the skyway segments should probably be most attractive during cold winter periods, and least attractive during summer, in comparison to the attractiveness of the pedestrian mall. Variations in the Skyway System's attractiveness against the sidewalks during different seasons has already been demonstrated in the Barton-Aschman survey (Table 3.6). It was therefore necessary to avoid such interaction. Secondly, the objective was to study the comparative attractiveness arising out of the physical design of the environment, and not out of rough weather conditions. The survey was hence designed to be conducted when the impact of the climate on system usage remained the least; during summer. 87 3.3.2.4 FORM OF SYSTEM DEVELOPMENT - Out of the two variations in the form of system development (section 2.6.0), the isolated form of development (section 2.6.2) does not represent any large scale system. The objectives and impact areas of the first category (section 2.6.1) are considerably more extensive than the isolated one, and systems in the first category are more often referred to as the Skyway System, universally. Segments inside the systems developed as part of broader urban design strategies (section 2.6.1), hence, were chosen for conducting the survey. 3.3.2.5 SYSTEM COMPOSITION WITH OTHER PEDESTRIANIZATION MEASURES Presence of an underground pedestrian network has a theoretical potential of rivalling the above-grade network in attractiveness and degree of usage. In addition, considering the objectives of the study, the presence of a pedestrian facility at grade level, in the vicinity of the Skyway Sj'Stem proved mandatory. As a result, out of the four variations in system composition (section 2.7.0), the Skyway System with grade level pedestrian mall (section 2.7.1) was considered appropriate for the survey. 3.4.0 The Survey Method and Instrument Numerous methods were available to study the addressed question. Questionnaire and interviews have, for a long period of time, been serving as the backbone of sociological and behavioral research. In addition, recent developments in survey technology have witnessed the formation of some innovative survey techniques. Foremost among them is contrived observation through time-lapse photography. During the decade long survey of 88 New York's plazas, Whyte amply demonstrated the efficiency of such modern research methods (The Social Life of Small Urban Spaces, 1980). Other researchers have proved the efficiency and efficacy of simple unobtrusive observation through data recording on behavioral maps and data recording sheets (Francis, 1984; Preiser, 1972; Sands and Ciolek, 1979; Hall, 1974 unpublished). The method adopted for the survey was selected after careful consideration of numerous potential difficulties. Considering the type of user informations which was required to be gathered, and the active variables in the design (section 3.3.0), a questionnaire survey or an interview offered the best option. However, such methods also get accompanied by several difficulties. Prime among those was the characteristics of the surveyor (the author). The surveyor, being non-white, having a foreign accent, and partially limited communicating skill, was not ideal for an interview survey, and to some extent, even a questionnaire survey. In addition to the above negative attributes, the surveyor's age (25 years), sex (male), and social class (somewhere in the lower middle), were also potential difficulties (difficulties arising out of physical/social characteristics of the surveyor including his/her race, age, sex, social class, and speaking abilities have been addressed by Webb, 1966, where in he has clearly demonstrated their negative impacts, and the possible interaction of surveyor variables with the survey outcome). Other difficulties included the scarcity in manpower, with the author as the only person available for data collection. A questionnaire or interview survey, hence, required more time in comparison to unobtrusive observation technique. Beside the time factor, the expense factor exercised considerable bearing too. With limited funds available for the survey work, simple unobtrusive observation proved to be financially most attractive. Owing to the extremely high cost of contrived observation (time-lapse photography) technique, the method was excluded from any consideration. Furthermore, questionnaire or interview surveys lead to smaller sample size, and without call back facilities could be unrepresentative of the entire population (Webb, 1966). With due consideration to all the impact areas, simple unobtrusive observation seemed the most appropriate procedure for .carrying out the survey of facility, usage pattern. _ . _ . Without Cameras, Questionnaires, or Tape Recorders, the mapping of users and activities on the facilities proved challenging. A simple note book and a pen was the only alternative left for the recording exercise. Moreover, the main task was to design a recording system, where detail characteristics of the users and their activities could be recorded easily and speedily, by a single surveyor. In addition, facility had to be made so that the recorded data could be easily transferred on to a Computer database for subsequent computing and analysis. From such a viewpoint elaborate descriptive recording of each sample user separately was not considered the best solution. Further exercise in building a recording instrument provided the alternative of keeping separate recording sheets for each user with all the variables listed sequentially. The idea was to circle the characteristics and activities of each user quickly with a pencil, to be fed in later to the Computer data base. However, it involved carrying a huge volume of printed recording forms (one for each sample) which had the potential of reaching up to 300 sheets per day, and around 2000 recording sheets for the entire data collection exercise. Not only was it too voluminous, but also uncomfortable to be carried by a single surveyor, and relatively expensive recording option. The final design of the recording sheet was made to resemble a conventional Computer Spread-Sheet, with all variables placed in consecutive rows in the first column, and the subsequent columns reserved for the sample users (Appendix I). The data as a result was not only easy to transfer on to a Spread Sheet, but also each to record by ticking off the correct category in each variable for each user sampled. With sufficient space to record thirty observations on each sheet, the volume and weight of the recording device also proved each to be handled by a single surveyor. Actual recording of the data was done on two 'Behavioral Observation Data Recording Sheet' (appendix I) to record the 90 user characteristics for mobile activities and stationary activities separately. Each of the sheet, with space to record thirty users, was appropriate to record observations up to approximately one hour.. Besides, space was .allocated to record the date, time, and location of the survey area, as well as the weather conditions. The bottom of each sheet was provided with four to five rows of blank spaces. This space was intended to be used to note all such observations or informations which could be useful or essential, but could not be recorded within the observation columns. The size of the recording sheet was kept limited to the size of any magazine to increase the unobtrusiveness of the recording process. The recording device was refined several times during and after the pilot surveys in Vancouver in order to remove the possibilities of any potential error and difficulties during the data collection exercise through simple observation. Simple observation, however, does not represent an error proof research procedure. Many areas of potential impact on survey data remains. Explanation of those areas as well as the precautionary measures adopted would be dealt with later in this section, before which, another important area deserving attention would be discussed; identification of variable categories through simple observation. 3.4.1 IDENTIFICATION OF VARIABLE CATEGORIES THROUGH SIMPLE OBSERVATION Identification and allocation of each sample user to the correct category constituted the most sensitive issue in the current adopted methodology. Identification of users on the basis of age (the variable 'age' stands for the approximate age group of the user, and hence can be reliably determined through observation), ethnic group, sex, grouping, and posture is relatively easy and reliable through observation. Other variables including the role , purpose (mobile activity), or activity may be more complicated to judge through simple 91 observation. The method adopted included the identification of physical and social cues, and classification based on predetermined social/behavioral parameters. Beside the variables -'role'-'purpose', .and 'activity', all other, user variables .can be reliably determined through simple observation, and hence will not be discussed in the subsequent sections. 3.4.1.1 ROLE (USER TYPE) The final decision regarding the role of a sample user rested on a variety of factor, the combination of which, and not any single cue, resulted in the allocation. Cues were many, ranging from the user's belongings and dressing style to their grouping pattern, including, to name a few, briefcases, tourist maps, shopping bags, guide books, cameras, camcorders, pagers, keychains, walkie talkies, luggages, formal/informal dressing patterns, summer dresses, and even the style of walking. The downtown workers were mostly 'adults', and dressed in formal office wear. There were a few exceptions, however, and some of the high income, upper class executives could be seen dressed in casual summer wear, where other cues were identified to ascertain the user category. In most of the cases workers would carry briefcases/ office bags with them. The time of the day also mattered. During morning and evening hours they would mostly be found walking fast (faster than the rest of the users) not noticing the ambient environment, and would be mostly alone. During lunch and coffee breaks, one could observe workers in groups, mostly chatting with one another. Relatively higher level executives could be found accompanied by their office secretaries. They rarely carried their briefcases during lunch breaks, but their grouping pattern and walking style disclosed their identity. It was common to observe workers shopping for food, but rarely for any other goods. Irrespective of their dressing or grouping pattern and walking style, the workers were always immaculately dressed. 92 In contrast to the workers, the visitors were more often found in groups. Their casual summer wear, which in many cases would be limited to shorts and T-shirts, made up the major difference between them and the workers. It was common to find visitors with families and hence make up a unique grouping category. The elderly, teenagers and young lovers made up the other distinct categories. Visitors and shopping bags go together, and it was rare to find a group without one. One visible visitor type were the young housewifes. Mostly from the upper class, their presence on the systems with shopping bags in their hands was a common scene. In addition, this class of users were rarely in any hurry, and more often observed promenading or shopping on the system. They constituted most of the users engaged in stationary activities, and made maximum use of the system furnishings. Unlike the workers, and unless they came with families, they rarely shopped for food, but they helped themselves to light snacks and coffee some of the time. The tourists and the visitors resembled one another except for some physical cues. Some distinct physical possession of most of the tourists included a city map, and cameras/camcorder. Tourists were mostly observed in large groups and sometimes accompanied with guides. They were also more likely to shop, and watch every activity and element of the physical environment around them. One could frequently observe them taking photographs or posing for one. Most of them possessed backpacks and dressed with some disregard to the present day fashion. International tourists were relatively easy to identify due to the visible difference in clothing style, and sometimes even the brand of cigarette they smoked or the language they spoke. Some of the richer tourists were relatively more formally dressed, and were difficult to identify from the visitors, but again other cues gave their identity away. Lastly, tourists were more likely to be found searching their way or reading maps than any other user category. 9 3 Security Staff were easily identified through their uniforms, and further explanation is not necessary. Similarly, the category 'mother/father with baby' was only a duplication and will not be discussed. The users in the 'others' category were the people not included in the rest of the categories. They included the punks, hookers, pimps, hippies, low class vagrants, and people on social welfare. All such users were visually distinguishable with relative ease. The above descriptions were valid for most of the users sampled. However, in some cases the physical cues of the sampled user proved conflicting where the more likely category was chosen over the less likely ones, according to the surveyor's judgement. 3.4.1.2 PURPOSE (MOBILE ACTIVITY TYPE) The classification of users into mobile activity types was carried out according to previously established parameters. Users were judged on the basis of their observed activity within the limits of a physical boundary, which was half a block long on each systems. On the Skyway System, segments of approximately half-a-block length was chosen for the survey. The portion of the systems chosen for survey, was done so after an initial phenomenological observation to ascertain the relative degree of activities on each. The busiest portions of the system was chosen for the survey. 'Access/Walking' as an activity was described as an act where the user traversed the entire length of the physical boundary without looking around or promenading, in a relatively fast pace and stiff posture. With the observation of any single action where users tended to slow down or look around or promenade, they were included in the 'promenading' category. Every segment surveyed had shopping establishments along the edges. Entry into any shop, even for a small period of time qualified the users for the 'shopping' category. Many of the users sampled were service personnel, including janitors, and sweepers working within the survey boundary. Such users were categorized under the 94 'to work/working' category. 'Recreating' as a purpose type included active recreation only. Those included roller skating, cycling, jogging, playing, and speed walking. Users were classified as 'searching' only when they were observed to have totally lost their way. When users were observed repeatedly crossing the survey line, and restricting their activities to a particular area, they were categorized as 'hanging out'. 3.4.1.3 ACTIVITY (STATIONARY ACTIVITY TYPES) An area of potential conflict and possible duplication was in the categorization of the variable 'activity'. It is very unlikely to observe, in a public place, any user engaged in only a single category of the range of activity types identified. People could watch/groom while waiting, or read while eating, or even talk while meeting. The range of possibilities are numerous. The problem was tackled by recording those activities, which the users were observed engaged in for a longer duration, in comparison to other activities, of the observation time. Users were recorded as watching/grooming when they were observed to be watching only, without being simultaneously engaged in some other activity type. Similar argument was established for the rest of the categories. People were recorded as talking, or eating, or reading when they were observed to be involved in those activities for a longer period of time in comparison to other activities. Performing as a category was more or less restricted to street performers, and hence did not encounter the problem of duplication. Users were categorizes as waiting, only when they were observed later (within the time period of observation) with the person they were waiting for. In general they were observed in the standing posture, and looking most of the time in the direction of approach of their friends/spouses. Meeting mostly involved spontaneous meeting. Most of the people in this activity type were also observed talking, and again mostly standing. 95 3.4.2 ERROR SOURCES AND DEFICIENCIES Potential error sources in the adopted methodology can be of two types. The first, and the most important type source pertains to the possibility of erroneous categorization of users. Such an error could arise due to the difference in cultural base of the surveyor and the surveyed sample. In such a context, the potential of wrong categorization especially in the users 'role' variable stands high. Even without any cultural differences, error could persist because of the limited reliability of simple observation as a survey instrument. The issue was partly tackled through the process of two pilot surveys conducted in Vancouver, B.C. , by the author. During the pilot surveys, recording was carried out simultaneously by three surveyors (including the author) where the other two surveyors belonged to the same cultural base as the sampled population. Cross checking of the surveyed data showed a maximum incongruity of two percentage points between the three separately recorded data, which was considered to be within tolerable limits of incongruity. Furthermore, only a single surveyor carried out the final survey, and any possible error in judgement would be relatively constant in intensity and probably repeated through all the three systems surveyed. Since the objective of the survey was a comparative analysis, and not an absolute measure of system performance, some impact of the error was probably effectively reduced. In spite of all the above arguments, error sources still remain. The ideal solution to such a problem is triangulation of measurement process, preferably in this context, through a simultaneous user interview. However, incorporation of multiple measurement processes was made difficult due to the limited financial availability and manpower, as well as the negative attributes of the surveyor as an interviewer, which has been discussed before. The inferences drawn from the collected data in the subsequent chapters of this thesis, hence, possess limitations in the reliability of the measurement instrument. 96 Another area of potential error is the possibility of increased efficiency of the surveyor through time. It was partly controlled by prior exposer of the surveyor to the survey scene through the two pilot surveys. Beside the two pilot studies, the author carried out an informal observation exercise in downtown Minneapolis before starting the data collection process. Finally, precaution was necessary to make the data collection process as unobtrusive as possible to avoid any reactive measurement effect form the surveyed population. It was taken by limiting the size of the data recording sheets to that of a magazine, and the sheet was constantly kept outside the vision of the user traffic. Care was also taken not to look directly at any user sample, nor to focus camera lenses on any system user for an unusual duration. 3.5.0 The City, the Survey Points, and the Users 3.5.1 T H E CITY Nine North American cities possess some form of Skyway System belonging to the category described in section 3.3.2.4; Minneapolis, Calgary, St.Paul, Dallas, Cincinnati, Edmonton, Hamilton, Duluth, and Des Moines. Out of the nine systems, only two satisfy the criteria laid down in section 3.3.2.5; Minneapolis, and Calgary are the only two cities where the Skyway Systems coexist with large, successful pedestrian malls in the vicinity. Many reasons led to the selection of Minneapolis Skyway System for undertaking the survey. The most important factor was the system size. Minneapolis has one of the largest Skyway System in North America linking 32 contiguous downtown blocks (Metro 2000 Plan), with a potential of expanding over a total of 64 blocks as envisaged in the original plan (Shostack, 1978). Unlike Cincinnati or Des Moines, Minneapolis is a large metropolis, 97 with all the associated urban development problems. It experiences extreme climatic conditions, with severely cold Winter and relatively hot Summer. In addition, the Minneapolis Skyway System is the most copied system in other cities (Whyte, 1988). Furthermore, the system is totally privately owned, unlike the system at St.Paul, and thus raises numerous controversial issues related to public/private environment, safety and security, control and regulation, and operating hours of various system segments. It also has a large user base which includes the residents of the central city, and suburban areas of both Minneapolis and St.Paul. The last reason for selecting Minneapolis for the survey was the availability of relatively large number of research publications on its system, unavailable for the other cities, and which helped establish a base point for initiating the study. 3.5.2 ESTABLISHING SURVEY POINTS Two main criteria were identified for the selection of survey points. First, it was decided that the survey points be located at such areas where the sample user would have the alternative of using any of the three systems under consideration. The second criterion was that all survey areas should have similar, if not identical, physical environment, especially in the areas of facility furnishing, and degree of retail activities. The areas around Nicollet Mall represent the core area of downtown Minneapolis. Keeping the stated objectives in mind, and after a visual survey of the Skyway System in the core, nine buildings qualified in terms of accessibility to other systems, activity level, and environment for the survey. Three of the nine buildings were finally selected as survey sites (the selection of survey sites on the Skyway System was entirely dependent on the granting of permission by the building management for carrying out the survey. 98 Permission could be obtained from only three of the nine buildings, and hence the final selection of survey sites cannot be technically termed as a selection process); IDS Crystal Court, City Center, and North Star Building (Fig.3.6). Separate survey sites were established on the portions of Nicollet Mall undergoing renovation and the portions not demolished (before arriving in Minneapolis, the author was ignorant about the ongoing renovation work on Nicollet Mall. Since the process of renovation work apparently effected some degree of change to the mall's usage pattern, the author deemed it proper to collect data from both the portions of the mall; the part under renovation, and the furnished portion). For the portion under renovation, three blocks were selected on the basis of activity level, edge structure, availability of skyway alternative, and the physical environment. The three blocks constituted contiguous blocks between 6th Street and 9th Street of downtown Minneapolis. On the basis of the same criteria, only one block on the furnished portion of the mall qualified for the survey; the block between 11th Street and 12th Street abutting Peavey Plaza (Fig.3.6). Beside the aspect of retail availability, the remote location of the furnished portion of the mall, in comparison to the survey points on other facilities, was another reason for choosing only a single block for observation. The block between 11th and 12th street was the closest to the core area of downtown Minneapolis. Moreover, the presence of the Peavey Plaza on its edge kept the mall pavements lively and festive, and it was an ideal area for system comparison as urban public spaces. However, data from the furnished portion, due to its remote location, unavailability of skyway alternative, and relatively lesser retail activity, was intended to be used more to supplement the data from the mall under renovation, rather than to be used as an independent system for comparison. The search for survey locations on conventional streets was based on identical criteria including availability of retail/business establishments on the edges, level of street 99 furnishing, and accessibility to the Skyway System. Marquette Avenue, running parallel to Nicollet Mall was selected for the survey, and the three selected survey blocks were located between 6th Street and 9th Street (the number of survey points on the other systems were limited to three to match the corresponding number on the Skyway System, and for the purpose of convenience in time sampling), one on each block (Fig.3.6). 3.5.3 SAMPLING THE USERS A three stage sampling procedure was adopted to ensure validity of the survey findings. In the first two stages, which were conducted simultaneously, random sampling of time period of observation, and the survey location on all the facilities together, was conducted through a simple random sampling procedure, for strengthening the external validity of the research results. Each location was allocated four hours from the total data collection time period. After an initial investigation of usage level, twelve hours (8:00 a.m. to 8:00 p.m.) was determined as the total survey time during weekdays, and eight hours (10:00 a.m. to 6:00 p.m.) during weekends. Beyond these established time limits, downtown activity level remained at the minimum, and hence was not considered appropriate for the survey. The allocation of the various locations on the four facilities to the three time periods of observation on each day, after the random selection process, is enumerated in detail in Table 3.7. The third random selection procedure was adopted to select the sample users from among all the people on the facilities. On each survey segment, a survey line was established depending on the convenience of the surveyor. Two separate methods were adopted to collect data on the mobile and stationary activities. Fifty minutes of each hour was allocated for the mobile activity portion of the collected data. Observations were 100 recorded each second minute (the two minute time interval was adopted after the pilot surveys established that the interval was convenient enough to record one observation and have some spare time for relaxation) of the hour, and the user crossing the survey line at that instance was selected as the sample user. In cases where no user reached the survey line on the second minute, observations were not taken. For the people in groups, or at instances where two or more users crossed the line almost simultaneously on the second minute, observations were always recorded for the person nearest to the observer. Data on stationary activity was recorded at the beginning of each hour. No data was collected in between the hourly intervals, and the users observed engaged in stationary activities at the very beginning of each hour were recorded. SITE DAY TIME SURVEY NUMBER DATE/DAY LOCATION Minneapolis Weekday 8:00—12:00 7 Aug.8 /Wed NorthStar Center Skyway 12:00—4:00 1 Aug.3 / F r i C i t y Center System : 4:00 8:00 5 Aug.6 /Mon IDS Center Weekend 10:00-12:00 11 Aug.ll/Sat IDS Center 12:00—4:00 2 Aug.4 /Sat C i t y Center 4:00 6:00 13 Aug.11/Sat NorthStar Center N i c o l l e t Weekday 8:00—12:00 8 Aug.9 /Thu 7th-8th Street Mall(R) : 12:00—4:00 19 Aug.14/Tue 6th-7th Street 4:00 8:00 21 Aug.l5/Wed 8th-9th Street Weekend 10:00-12:00 22 Aug.18/Sat 8th-9th Street 12:00—4:00 12 Aug.11/Sat 7th-8th Street 4:00 6:00 24 Aug.18/Sat 6th-7th Street N i c o l l e t Weekday 8:00—12:00 10 Aug.10/Fri l l t h - 1 2 t h Street Mall(F) : 12:00—4:00 17 Aug.13/Mon l l t h - 1 2 t h Street 4:00 8:00 6 Aug.7 /Tue l l t h - 1 2 t h Street Weekend 10:00-12:00 14 Aug.12/Sun l l t h - 1 2 t h Street 12:00—4:00 23 Aug.18/Sat l l t h - 1 2 t h Street 4:00 6:00 16 Aug.12/Sun l l t h - 1 2 t h Street Marquette Weekday 8:00—12:00 18 Aug.14/Tue 7th-8th Street Avenue : 12:00—4:00 20 Aug.l5/Wed 6th-7th Street 4:00 8:00 9 Aug.9 /Thu 8th-9th Street Weekend 10:00-12:00 3 Aug.5 /Sun 7th-8th Street 12:00—4:00 15 Aug.12/Sun 8th-9th Street 4:00 6:00 4 Aug.5 /Sun 6th-7th Street Table 3.7: Final Survey Schedule (Survey Conducted During Summer, 1990). 101 Fig.3.6: Map of Downtown Minneapolis Core Area. Source: City of Minneapolis, Planning Department, 1990. 102 Along with the survey line, a survey area was identified for the purpose of observation, on each survey location. The survey area was approximately half a block in length. The main purpose of establishing the survey area was to track the sample user down to a reasonable distance, up to which his activities could be easily and clearly monitored. In recording mobile activity types (purpose) the area acted as the boundary within which the observation was to be conducted and a decision made regarding the purpose type of the sample user. For stationary activities and postures, only users found engaged in stationary activities within the survey area qualified as sample users. In short, the recording of stationary activities was almost identical to taking a snap shot of the survey area after an interval of sixt37 minutes on each location. With such a data collection strategy twenty five or less users were sampled each hour of the day, on each system, for the mobile activity types, and the number of people surveyed for stationary activities varied depending on the time of the day and the location. Before discussing in detail the results of the data analysis, however, it is essential to provide a comprehensive picture of downtown Minneapolis, its present form, and the planned development during the next decade, in order to understand the micro and macro-level role of the Skyway System, and the Nicollet Mall. The next chapter begins with a brief description of downtown Minneapolis and its future growth, and subsequently leads on to the description of inferences drawn from the data collected in the central areas of downtown Minneapolis. 103 C H A P T E R F O U R T H E S K Y W A Y , T H E M A L L , A N D T H E S T R E E T 4.1.0 Downtown Minneapolis : A Bird's Eye View Like the rest of the major cities in North America, Minneapolis, during the fifties and sixties, went through its own share of decay of the central city area. Today, however, Minneapolis, with a central city area of 1800 acres, boasts of one of the unique downtowns in the North American continent, and more specifically in the upper Midwest region of United States. The beginning of such a transformation occurred with the inception of the 1959-60 'Central Area Plan', into the city development agenda. The plan have been updated thrice since then, subsequently being termed as the 'Metro Center 1985\ 'Metro Center 1990\ and the 'Metro Center 2000'' plans. Pedestrianization, as a major solution to center city revitalization, received an opportunity for implementation in this city. During the early sixties, the plan began to materialize with the beginning of the implementation of the Skyway System, the Nicollet Mall, and the interstate innerbelt. The first Skyway bridge opened in 1962, connecting the Cargill and Roanoke Building across Seventh Street in downtown Minneapolis. By 1978, 14 Skyway bridges, connecting 11 downtown contiguous blocks, represented one of the most extensive Skyway networks in the U.S.A. (Downtown Council of Minneapolis, 1988; Irvin and Groy, 1982). However, neither the Skyway network, nor the Nicollet Mall stand in isolation. Along with other planned measures, they form a part of an ambitious plan to develop the central city as the primary corporate, commercial, entertainment, education, and retail center. As a result, an isolated discussion 104 of these pedestrian measures may not project the true image, and a brief description of the macro level plans and strategies is warranted. Before discussing the macro level strategies planned for in the Metro 2000 Plan, it is essential to provide a brief description of three of its principal elements; the central city, the Nicollet Mall, and the Skyway System. 4.1.1 T H E CENTRAL CITY Bound by the interstate highways (1-35 and 1-94) on the East and West, the Plymouth Avenue on the North, and the Mississippi on the South, a roughly triangular shaped area describes the Minneapolis central city. Not many people reside within these boundaries. A few scattered apartments, residential hotels, and rooming houses, in severely deteriorated and run down areas of Loring and Elliot Park neighborhoods, were the only available housing during the fifties. Since 1959, a series of renewal plans, including the Gateway district renewal plan of 1966, Loring Park project of 1972, and Elliot Park neighborhood of 1978, have added a few decent living areas, including around 5,200 units since the completion of the Gateway Towers project. The total residential population of the center city in 1987 was 20,000 living in slightly over 13,500 units. Commercial offices constitute the major building type in the downtown. Especially, during the last thirty years, it has become the home to the headquarters of a number of international companies and institutions, thus even further reinforcing its historical role as the economic hub of the entire upper Midwest region. Eighty percent of the downtown jobs are provided by these offices, office buildings generate a third of the city's revenue, and the large employee population constitute a major customer base for the downtown retail establishments. Downtown retailing, however, is restricted to the Nicollet Mall and a few segments of the Skyway network (Fig.3.6). Although, the downtown is not the major retail base in the metropolitan area, its customer base remains strong, with nearly forty percent 105 being regional shoppers (Downtown Council of Minneapolis, 1988; Irvin and Groy, 1982; Wiedenhoeft, 1975; Goldfield, 1976; Carpenter, 1975). 4.1.2 T H E NICOLLET MALL Completed in 1968, the eight block long Nicollet Mall was the result of the labour and cooperation between the business community and the public government. With a winding central motorway, wide sidewalks, extensive street furnishing elements, lines of retail outlets on either side, the mall is the downtown's principal pedestrian area. It is situated right in the middle of the core area, and supported by minibus public transit system, to cater to the needs of scores of workers, shoppers and visitors, for whom, the mall represents the heart of Minneapolis Downtown. A few, small and large, plazas serve as open spaces, the largest being the Peavey Plaza on the intersection of the Nicollet Mall and 11th street. The mall is also the primary retail base at grade level. Outside the mall, retail outlets are scarce in the entire downtown. Buildings on three consecutive blocks across the mall are connected by skywalks; the blocks between 5th street and 8th street (Shastock, 1978; Downtown Council of Minneapolis, 1988; Irvin and Groy, 1982). 4.1.3 T H E SKYWAY SYSTEM Initially developed for purely commercial motives, the system started with the building of two aerial walkways in 1962. Subsequently, it was integrated into the city development agenda, and was made an integral part of "Metro Center 1985" plan. The Minneapolis Skyway System, which links around 32 contiguous downtown blocks today, was an important part of the overall transportation planning strategy. As a result, they 106 have been designed, and managed as a route exclusively for pedestrian movement. Uses such as retailing, seating, vending, and display are confined to spaces off the skyway corridor. Development criteria for the new segments parallel those for the sidewalks; to be continuous, and free of obstructions to the pedestrians (Minneapolis Skyway Advisory Committee, Minneapolis Skyway System design standards, 1985). Through the decades after its inception, the system developed as a result of public-private cooperation, and most of the segments are owned, maintained, and operated by private organizations. It is funded entirely by private developers with technical assistance and planning coordination of the City Planning Department. The entire system has been separated into two; the primary network, and the secondary network, although all skyway links have not yet been implemented. The primary network, consisting of two main axes, has been designed to handle the pedestrian traffic from peripheral parking ramps and residential housing areas to the core. The secondary network, consisting of smaller feeder segments, has been designed to link the remainder of the downtown, and in due course will ease circulation between major downtown centers. The system links the buildings with skybridges roughly located at mid-block, and constitute of a series of interior walkways, corridors, atria spaces, courts, arcades, and lobbies. Many of the segments are lined with retail establishments, and a few of the corridors overlook the atria spaces in large office buildings. However, far from being a public activity space, the system serve as a complimentary pedestrian facility to the downtown's sidewalks which, after Nicollet Mall, constitute the downtown's most important pedestrian facility. As in every other city, the design criteria for the sidewalks emphasize efficient pedestrian movement over other social activities. To limit obstruction of pedestrian traffic flow on the sidewalks, the design criteria suggest limitations to street furniture including trash cans, telephone booths, newspaper boxes, and bicycle racks, on the walkway (Irvin and Groy, 1982). All the elements of the pedestrian system, however, form a part of the overall development strategy; the Metro Center 2000 Plan, which is explained in the next section. 4.1.4 T H E METRO 2000 PLAN In its thirtieth year of evolution, this ambitious plan aims at restructuring the downtown into the primary retail, commerce, employment, education, and corporate center of the upper Midwest region. The emphasis is on developing the central city with a sense of place, a sense of unity, a sense of time, a sense of encounter and a sense of theater (Downtown Council of Minneapolis, 1988). The downtown has been envisioned as a number of identifiable districts; retail, financial, housing, medical, and entertainment (from the 1985 plan), with the present plan highlighting a few emphasis areas for entertainment, government, convention, education, technological research, and health care. The areas and districts, however are not visioned as separate entities. The plan aims for organizing the downtown into a single, large multi-use project, with a compact retail core from 5th to 11th street on Nicollet Mall surrounded by a densely built, horseshoe shaped high rise office development, a secondary ring of less densely built offices, peripheral parking to cater to the daily needs of the workers, and major residential neighborhoods on the periphery (Fig.4.1). And all these areas are to be connected by the skyways to form one single cohesive whole, visually contrasted by a system of open spaces. One major policy plans for limited vehicular movement within the core. In effect, pedestrian facilities will be the major circulating system inside the downtown, supported by other mass transport facilities including the mini bus, and the HOV's. Besides, preferential parking for shoppers and visitors close to the retail and office core is being planned for, to attract more regional shoppers to the downtown. The Nicollet Mall, and the Skyway System, as a result have a dominant role to play in the future development of downtown Minneapolis. And even today, they do not go unnoticed by the large majority of workers and downtown visitors. Beside catering to the travel needs of the workers today, the skyway's attractiveness to other downtown users, 108 retail tall office buildings less dense office perpheral parking garage Fig 4.1: Plan of proposed restructuring of downtown Minneapolis during the next decade. Source : Downtown Council of Minneapolis; METRO 2000 PLAN, 1988. 109 conduciveness to other usage types, and its acceptance by the city dwellers as a useful and attractive public space, could project the actual potential of the these pedestrian systems in realizing the ambitious plan for the next decade. The subsequent sections in this chapter studies the performance of the Skyway System in comparison to the mall, and the ordinary street, which has been a part of the cityscape since the development of first urban settlement. 4.2.0 The People Downtown Among the many areas of possible comparison, one vital aspect was the type of people who used each of the four facilities. Many authors, including Robertson (1988) and Whyte (1988) have raised the issue of the possible variability in the user types, between the Skyway and the street. More specifically as discussed earlier (Chapter Three), Robertson surveyed such areas as the age of the people on the network, the gender of the users, their ethnic background, their primary role downtown (workers as against shoppers, etc), and the socioeconomic status of the users. He came up with findings, which portrayed a significantly restricted role of the skyway network as a public place in the overall urban setting. This section deals, specifically, with the type of users, who use each facility, during weekdays and weekends. Besides, the major issues addressed by previous authors (section 2.8.0) have also been included in the analysis. They include, Accessibility, Control and Regulation, Safety and Security, Orientation, and the Physical Environment. While dealing with these issues, the four facilities have been measured on an ordinal categorical scale; from least easily accessible to most easily accessible, most controlled to least controlled environment, most perceived safety to least perceived safety, and most 110 disorienting to least disorienting. Before discussing the analysis, it is necessary to short list the potential user and usage types which each of the issues were assumed to affect, and explain the statistical procedure adopted for the analysis of the collected data. Ease of accessibility pertains to the discussions in the literature regarding the difficulty of locating skyway entrances. Many authors have expressed their concern over the normal location of skyway entrances, deep inside the building lobby (section 2.8.5). Accessibility also includes the provision of mechanical vertical connectors to the upper levels. And lastly, it could be explained in terms of accessibility by motor vehicles. From any of the above points of view, skyway seemed to be the least easily accessible, and the street, most easily accessible. In any case, the worst affected user type should be the elderly, disabled, and shoppers, for whom, ease of accessibility could be a decisive factor in the route choice. The four facilities have varying degrees of control. The skyways, predominantly privately owned, have a more controlled environment than the other facilities (section 2.8.2). Examples include, the City Center skyway, which has a dress code, and prohibition of photography, only to name a few. The mall does not exercise such levels of control on its users, though, some degree of rules and discipline is maintained. The street offers the most freedom. Theoretically, a controlled environment could affect such users like the teenagers on the skyway, elderly/retired hangouts, and most of the users in the 'others' category in user role types. Safety and Security (section 2.8.6) was one of the most important factors. The term safety relates here to perceived safety rather than the real safety. With larger control, and presence of security personnel on the skyway, and to a lesser degree on the mall, these areas are perceived safer by user groups including women (Robertson, 1988, I l l about skyway safety). The other type of safety factor was generated from the absence of automobiles, thus reducing the chances of accident. The mall was considered comparatively less safe than the skyway because of the occasional vehicular traffic occurring within it. The street was presumed to be the most unsafe place from these points of view. Again, such differences could seriously affect certain user groups including the female, elderly, occasional visitors, and people who are alone. It is difficult to measure the degree of physical environment. Hence, discussions regarding it have been included contextually in this chapter, wherever needed, and most part of the explanation have been included in Chapter Five. Statistical tests conducted on the collected data included three separate levels of examination. The first range of tests were conducted on all variables to examine and provide evidence for any variation in proportional distribution of user variables on the four sites surveyed. The independent variables (user variables) in the analysis were measured during the survey either on the categorical scale or ordinal categorical scale. Accordingly, Chi-square test for independence, and the Lambda measure of association were conducted on all variables. To increase the precision of text results ordinal measures of association including Kendall's Tau-B, Kendall's Tau-C, Gamma, and Somers'D were conducted on the ordinal categorical variables. Instances where the first test demonstrated significant result a second range of test was carried out to isolate all the pairs of facilities where the significant difference in facility usage pattern existed. These tests were similar to the first range of tests, but were conducted between pairs of site and included the Chi-square and Lambda test measures. The number of pair-wise comparison for each variable was limited to three; viz. Skyway System with Nicollet Mall(R), Skyway System with Nicollet Mall(F), and Skyway System with Marquette Avenue, in order not to exceed the probability of committing a Type-I error. To examine the impacts of the major issues on the facility 112 performance a third range of test was conducted. The tests were either ordinal measures of association or the Kruskal-Wallis one-way ANOVA, conducted wherever the scale of measurement of the variables permitted for one. (Note: In the succeeding text of this thesis 'street' would mean 'Marquette Avenue'; 'Mall(F)' would mean 'the furnished portion of Nicollet Mall at Minneapolis; 'Mall(R)' would mean 'the portion of Nicollet Mall under renovation work'; and 'skyway' would mean 'the Minneapolis Skyway System'). 4.2.1 T H E USER'S A G E GROUPS A Chi-square test for independence suggested that there was enough evidence at 0.0294 significance level that during weekdays, the various age groups varied in their proportional distribution on the four survey sites. The association was, however, weak with a lambda value = 0.0215, meaning that the association was not highly predictable (Table 4.1). The weekend results also suggested enough evidence at 0.0146 significance level (the variable 'age' was re-categorized for the weekend data since the original categories in Table 4.2 did not confirm to the Chi-square test assumptions. The reclassified cross-tab table is included in appendix II, Table 1), but again with a weak association, with a lambda value = 0.04167 (Table 4.2). On the whole, although there was enough evidence to believe that the proportional distribution of the age groups on the four site differed, the degree of predictability was not high. A pair-wise comparison of the four survey sites made the differences clearer. During weekdays, there was enough evidence, at 95% confidence level, to believe that a significant difference existed between the skyway and Marquette Ave.(Chi-square sig. value 0.002). Although the corresponding lambda measure (0.0459) was not large, the finding gave a clearer picture (Table 4.3). Most of the variability in the proportional distribution was explained by the differences between the Skyway and Marquette Ave., during weekdays, and the Skyway and the mall did not have any 113 significant differences in the proportional distribution of various age groups. Unlike during weekdays, a pair-wise comparison of the weekend data did not provide enough evidence to believe that a difference in usage pattern between the sky and Marquette Ave. existed. Explanation of such a difference could be many. All routes on the skywa}' network does not remain open during weekend, nor are the skyway retail establishments. Furthermore, coupled with the reduction in worker population, a considerable reduction in automobile traffic on downtown streets attract some visitors and tourists on to the streets from the core area and thus probably effecting a more uniform age distribution on the two facilities. Unlike, again, during weekdays, the skyway and Mall(R) showed enough evidence of disproportionate representation. One possibility is that where as the mall did not experience significant change between weekday and weekend, the street usage changed significantly, and approached the usage pattern on skyway due to a larger representation of visitors and tourists on its sidewalks, and a reduction in worker population. As regards to the major issues, none of the those seem to have had any significant impact on the route choice made by each group. All the ordinal measures including Kendall's Tau B and C gave insignificant results (Table 4.1 & 4.2), suggesting that ease of accessibility, control & regulation, perceived safety level or feeling of disorientation did not affect in any significant way the predictability of route choices of the various age groups. However, the user groups differed significantly in their average distribution on the four facilities. The Kruskal-Wallis one way A N O V A , gave a Chi-square at 0.0008 significance level for weekdays, although not for the weekend data. Table 4.4 illustrates the mean rank of each age group for weekdays. On an average teenagers were more likely to be on the skyways than the mall or the street. Clearly, non of the presumptions, concerning the various issues, fit into this finding. Ease of accessibility may not have been a deciding factors for the teenagers, nor the perceived safety level, and the skyways with stricter control did not prove to be incompatible to their needs. A control on social status AGE C h i Id Teen A d u l t E l d e r l y COUNT EXP VAL ROW PCT COL PCT S ITE Skyway N i c o l l e t N i c o l l e t Marque t t M a l l R M a l l F e Avenue 1.00| 2 . 0 0 | 3 .00 | 4 . 0 0 | ---- + + + • 1.00 <13 2 .00 13 to 3 .00 20 to 4 .00 >64 COLUMN TOTAL 7 4 .8 36. 8X 2 . 3 * ' 6 4 .8 31 .6X . 2. OX 2 4 .8 10. 5X .7X 4 4.5 21 . 1X 1 .4X 13 7.6 43 . 3X 4 .3X 10 7.6 33. 3X 3.3X 6 7.6 20 . OX 2. OX 1 7.2 3.3X .4X 246 259 .0 24. OX 82 . 3X 253 259. 9 24. 7X 84. 3X 266 259 .9 26. OX 88 . 7X 259 245.2 25. 3X 91 .5X 33 27 .6 30. 3X 11 .OX 31 27 .7 28. 4X 10. 3X 26 27 .7 23 .9X 8.7X 19 26. 1 17.4X 6.7X 299 300 300 283 25.3X 2 5 . 4 % 25 .4X 23.9X ROW TOTAL 19 1 .6X 30 2.5X 1024 86 . 6X 109 9.2X 1162 100.OX CHI-SQUARE D . F . 18 .53978 STAT IST IC LAMBDA SOMERS' D STAT IST IC KENDALL ' S TAU B KENDALL ' S TAU C GAMMA SIGNIFICANCE 0 .0294 SYMMETRIC 0 .01827 0 .00450 VALUE 0 .00525 0 .00297 0 .01236 MIN E . F . 4 .549 WITH AGE DEPENDENT CELLS WITH E . F . < 5 4 OF 16 < 25.OX) WITH S ITE DEPENDENT 0 .00000 0 .00297 SIGNIFICANCE 0.4204 0 .4204 0 .02154 0 .00928 Table 4.1: Crosstabulation of Age by Site for the weekday data. 115 AGE COUNT EXP VAL ROW PCT COL PCT S ITE Skyway N i c o l l e t N i c o l l e t M a l l R M a l l F 1.00| 2.00| 3.00| Marque t t ROW e Avenue TOTAL 4.001 1.00 7 5 2 1 C h i l d <13 4.2 4.7 4.7 1.4 46.7% 33.3% 13.3% 6.7% 4.0X 2.5% 1 .0% 1.8% 2.00 7 9 7 3 Teen 13 t o 7.2 8.2 8.2 2.3 26.9% 34 .6% 26.9% 11.5% 4.0% 4.5% 3.5% 5.3% 3.00 133 174 166 51 A d u l t 20 t o 145. 1 165.8 165.8 47.3 25.4% 33.2% 31 . 7% 9.7% 76.0% 87.0% 83.0% 89.5% 4.00 26 12 25 2 E l d e r l y >64 18.6 21.2 21.2 6.0 41.8% 17.9% 37.3% 3.0% 16.0% 6.0% 12.5% 3.5% COLUMN 175 200 200 57 TOTAL 27.7% 31.6% 31.6% 9.0% 15 2.4% 26 4.1% 524 82.9% 67 10.6% 632 100.0% CHI-SQUARE D . F . S IGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 17.98809 STAT IST IC 0.0353 SYMMETRIC 1 . 353 WITH AGE DEPENDENT 5 OF 16 ( 31.3%) WITH S ITE DEPENDENT LAMBDA SOMERS' D 0.03333 -0 .02101 0.00000 -0.01490 0.04167 -0.03561 STAT IST IC VALUE SIGNIFICANCE KENDALL ' S TAU 8 KENDALL ' S TAU C GAMMA -0.02303 -0.01420 -0.04974 0.2608 0.2608 Table 4.2: Crosstabulation of Age by Site for the weekend data. 116 P A I R S OF S I T E C H I - S Q WEEKDAY S I G LAMBDA C H I - S Q WEEKEND S I G LAMBDA SKYWAY V S M A L L ( R ) : 0 . 6 2 7 2 6 O . 8 9 0 2 0 . 0 2 0 0 7 1 0 . 8 4 0 4 0 . 0 1 2 6 0 . 1 0 2 8 6 SKYWAY V S M A L L ( F ) : 6 . 9 6 6 8 4 0 . 0 7 3 0 0 . 0 6 3 5 5 4 . 9 4 5 0 0 . 1 7 5 9 0 . 0 4 5 7 SKYWAY V S M A R Q . A V E . : 1 4 . 7 7 9 0 . 0 0 2 0 0 . 0 4 5 9 6 . 9 6 0 1 0 . 0 7 3 2 0 . 0 0 0 0 ( N O T E : T h e u n d e r l i n e d s i g n i f i c a n c e v a l u e s i n t h e t a b l e d e m o n s t r a t e s e v i d e n c e o f s i g n i f i c a n t d i f f e r e n c e a t 9 5 % c o n f i d e n c e l e v e l ) . Table 4.3: Pairwise comparison of facilities for the proportional distribution of Age groups. Data Source: Appendix II; Table 2 to 7. S ITE BY AGE KRUSKAL-WALLIS 1-WAY ANOVA F a c i l i t y O b s e r v e d MEAN RANK CASES 495 26 19 AGE = 1 C h i l d <13 399 43 30 AGE = 2 Teen 13 to 19 605 09 1024 AGE = 3 A d u l t 20 to 64 533 49 109 AGE = 4 E l d e r l y >64 CASES 1182 1182 TOTAL CHI-SQUARE 15.7776 SIGNIFICANCE 0.0013 CORRECTED FOR T IES CHI - SQUARE SIGNIFICANCE 16.8314 0 ,0008 Table 4.4: Mean ranking of Age groups on the facilities, during weekdays. 117 could probably explain such a phenomena, where most of the teenagers on the skyway probably belonged to the upper class, and thus readily accepted on it. For a great majority of teenagers, peer visibility and privacy from others may be the determining issue, which only further surveys could answer. Considering the issue of accessibility, the lesser average distribution of the elderly and children in comparison to teenagers and adults on the skyway was consistent with prior notion that the skyways suffer from deficient design. However, in proportional terms, out of all the elderly sampled during weekdays, there proportional representation was as much on the skyway as the mall (Table 4.1; 11.0% and 10.3% respectively). The difference of +0.7%, with a 95% confidence interval of +5.64% to -4.2%, did not provide enough evidence to represent a true difference. If accessibility were an impact area implying defective physical design of the facility, then a reduction in elderly population should be expected. In addition to the problems of accessibility, the skyways lack in one of the most important furnishing elements; the seats and for the scores of elderly users on the mall, who make it a daily habit to visit the core area, the availability of rest areas was a determining factor in route choice (gathered from personal interview of some of the elderly mall users). Moreover, the elderly are more likely to be affected by the orientation factor, and normally should tend to prefer the mall, which is less disorienting and better furnished that the skyways. Their equal or more representation on the skyway network in comparison to the mall in contrast to the aforesaid presumptions may be explained by several factors. Orientation may not have been a decisive factor, after all. Most of the regular elderly visitors stick to familiar areas (from interview) and probably do not experience disorientation. Besides, during the time period when the observations were taken, a large portion of the Nicollet Mall, in the core area, was undergoing renovation, which had left the mall without any furnishing, and made walking more difficult, with more obstructions and circuitous travel paths. According to a few of the elderly interviewed, a large number of the regular elderly users had temporarily shifted their routes on to the skyway level, which offered some, although not 118 comparable to the mall, degree of furnishing. In addition, downtown Minneapolis, and especially the core area, was the venue for the 1990 convention of Hnd World War veterans during the period of survey, which had brought in a large number of elderly population from all parts of Minnesota and other states, who resided in hotels hooked on to the skyway network. The convention may also explain the significant increase in elderly representation on skyway during weekend, as compared to mall(R), where the difference of 10% provided enough evidence for a true difference in population with corresponding 95% confidence interval being +3.64% to +16.35%. During normal periods, the presence of elderly on the skyway could be expected to be comparatively smaller than the present recorded value. To conclude, the equal or more presence of elderly on the system points at the possible absence of impact of any problem in accessibility and orientation on skyway usage. Secondly, along with the elderly (mostly retired hangouts), the comparatively large proportion of teenagers on the skyway segments again proves that the private control of the spaces probably have little impact on the usage of the system as a public place (Appendix III, Fig. 1 & 2). Consistent with previously held notion adults constituted the largest category of users inside the core area (86.6% of users were between 19 to 64 years), and children below 13 years were virtually absent. 4.2.2 GENDER DISTRIBUTION OF THE USERS The four sites differed significantly in the proportional distribution of the gender groups. A Chi-square test provided enough evidence, at 0.0000 significance level for weekdays that such differences existed. However, the strength of association was not high, with a lambda value of 0.0714 for weekdays, implying that there was not a large degree of predictability in the distribution of the two gender groups on the four facilities. During weekend, the gender distribution did not provide enough evidence of a variation in 119 distribution at the 95% confidence level as can be inferred from the large significance value (0.0749) for the Chi-square test. However, ordinal measures showed significant variability. Among the possible issues affecting route choice for the gender groups, the perceived safety factor generates primary interest. Control and regulations, in certain ways increase the perceived safety level, due to higher visibility of security staff and lesser possibility of unwanted users being present on the facility. According to the survey conducted by Robertson in 1988, people, especially the female users, perceived the skyways as safer areas as compared to the streets below, in all the five different cities where the survey was conducted. Besides, the sheer density of people on a system could add to the perceived safety level. More specifically, the density of people on the streets outside the core was so less in comparison to the mall and the skyway that there was a distinct possibility of those streets being perceived as relatively unsafe. Perceived feeling of lesser safety due to automobiles on the streets is self explained. All the ordinal measures of association showed a negative association, although only moderately strong, with the value of Somers' D being -0.136 for weekdays and -0.104 for weekends (Table 4.5, Table 4.6). In spite of the weak association, it is fairly evident that as the perceived security level decreased (from the Skyways to the Street), the proportion of females in the user population decreased (Appendix III, Fig.3 & 4). This is consistent with Robertson's finding in 1988. Safety may not have been the only factor, but the female users were more likely to use the Skyway than the Mall or the Street, irrespective of the day of the week. A Kruskal-Wallis test reinforces the assumption (Table 4.7, Table 4.8). Apart from the perceived safety factor, non of the other issues explain the larger proportional distribution of female population on the Skyway System, and inside the core area. The finding differs, however, from Robertson's survey, which inferred that a typical Skyway user, during weekdays, would be a female (Robertson, 1989). The present data did not suggest so since the difference of 9% between the proportional distribution of male and female users on the skyway system, during weekdays, did not provide enough evidence of a true difference. Testing the 120 alternate hypothesis that the true proportion of male on the system p m (0.455) N E p 0 (0.5) against the null hypothesis: p m = p 0 , the attained significance level of Z-score was 0.1188, implying that at the 95% significance level there was not enough evidence of a true difference. Even tests for weekend data demonstrated similar trend, with a significance level of 0.5962 for the Z-score leading to identical conclusions that the proportion of male and female on the skyway was not significantly different. Another important aspect of the variation in gender distribution was explained by the pair-wise comparison (Table 4.9). The Skyway and mall(R) did not have any significant difference in the proportional distribution of gender groups, either during weekdays or during weekend. Most of the variation during weekdays could be explained by the difference between the Skyway and Marquette Ave., and Skyway and Mall(F). From the weekday data, once again the at-grade street proved to be significantly different from the Skyway. During weekend most of the variation was explained by the difference in gender distribution between skyway and Mall(F). The large difference between the skyway and mall(F) both, during weekday and weekend, may be explained by the remote location of mall(F) from the core area where the rest of the survey points were located. In addition, the comparatively larger proportion of 'not determined' users (particularly, lower class hangouts and vagrants), and fewer shopping facility on mall(F) in comparison to mall(R) and the skyway, probably led to lesser attraction of mall(F) to female visitors and shoppers. This can be observed, for instance, from the cross classification tables (Table 4.10 to 4.15) which show that only 34% of the female users on mall(F) during weekdays were visitors, of which only 29.4% were engaged in shopping activities (Table 4.10 & 4.11). In contrast, 60.8% of all female users on mall(R) were visitors, 43.7% of which were observed shopping (Table 4.12 & 4.13). Similarly, 40.5% of all female users on the skyway were visitors, 53% of which were observed shopping (Table 4.14 & 4.15). 12 GENDER COUNT EXP VAL ROW PCT COL PCT S ITE Skyway N i c o l l e t N i c o l l e t Ma rque t t M a l l R M a l l F e Avenue 1.00| 2 . 0 0 | 3 . 0 0 | 4 . 0 0 | ----+ . . . + - . . . . . . . + . . . . . - . - + ROW TOTAL 1.00 136 157 200 165 658 Ma le 166.4 167.0 167.0 157.5 5 5 . 7 % 2 0 . 7 % 2 3 . 9 % 30 .4% 25. 1% 4 5 . 5 % 5 2 . 3 % 6 6 . 7 % 5 8 . 3 % 2 .00 163 143 100 118 524 Fema le 132.6 133.0 133.0 125.5 4 4 . 3 % 31. 1% 2 7 . 3 % 19. 1% 2 2 . 5 % 5 4 . 5 % 4 7 . 7 % 33 . 3% 4 1 . 7 % COLUMN 299 300 300 283 1182 TOTAL 2 5 . 3 % 2 5 . 4 % 2 5 . 4 % 2 3 . 9 % 100.0% CHI-SQUARE D . F . 29 .41731 3 STAT IST IC LAMBDA SOMERS' D SIGNIFICANCE 0 .0000 SYMMETRIC 0.06401 •0.10789 MIN E . F . CELLS WITH E . F . < 5 125.459 NONE STAT IST IC VALUE WITH GENDER DEPENDENT 0 .05153 -0.08946 SIGNIFICANCE WITH S ITE DEPENDENT 0 .07143 -0.13591 KENDALL ' S TAU B KENDALL ' S TAU C GAMMA -0.11026 -0 .13416 -0.17974 0 .0000 0 .0000 Table 4.5: Crosstabulation of Gender groups by Site for the weekday data. SITE COUNT GENDER Male EXP ROW COL VAL PCT PCT Skyway 1.00 N l c o l l e t Mall R 2.00 N l c o l l e t Mall F . 3.00 Marquett e Avenue 4.00 ROW TOTAL 1 OO 84 94.7 24.6% 48.0% 104 108.2 30.4% 52.0% 122 108.2 35.7% 61.0% 32 30.8 9.4% 56. 1% 342 54. 1% Female 2 OO J 91 80.3 31.4% 52.0% 96 91.8 33. 1% 48.0% 78 91.8 26.9% 39.0% 25 26.2 8.6% 43.9% 290 45.9% COLUMN TOTAL 175 27.7% 200 31.6% 200 31.6% 57 9.0% 632 100.0% CHI-SQUARE D.F. 6.90793 3 STATISTIC LAMBDA SOMERS' STATISTIC KENDALL'S TAU B KENDALL'S TAU C GAMMA SIGNIFICANCE MIN E.F 0.0749 SYMMETRIC 0.03463 -0.08497 VALUE •0.08639 -0.10295 -0.14431 26. 155 CELLS WITH E .F .< 5 NONE WITH GENDER DEPENDENT 0.02414 -0.07200 SIGNIFICANCE 0.0094 0.0094 WITH SITE DEPENDENT 0.04167 -0.10365 Table 4.6: Crosstabulation of Gender groups by Site for the weekend data. KRUSKAL-WALLIS 1-WAY ANOVA S ITE F a c i l i t y O b s e r v e d BY GENDER MEAN RANK CASES 627 . 1 1 658 GENDER = 1 , Male 5 4 6 . 7 9 524 GENDER = 2 Female 1182 TOTAL CASES 1182 CHI-SQUARE 16. 1497 SIGNIFICANCE 0.0001 CORRECTED FOR TIES CHI-SQUARE S IGNIFICANCE 17.2284 0 . 0 0 0 0 Table 4.7: Mean ranking of Gender groups on the facilities during weekdays. KRUSKAL-WALLIS 1-WAY ANOVA S ITE F a c i l i t y O b s e r v e d BY GENDER MEAN RANK CASES 331 .53 342 GENDER = 1 Male 2 9 8 . 7 8 290 GENDER = 2 Female 632 TOTAL CORRECTED FOR TIES CASES CHI-SQUARE SIGNIFICANCE CHI-SQUARE SIGNIFICANCE 632 5.0499 0 .0246 5.5210 0 .0188 Table 4.8: Mean ranking of Gender groups on the facilities during weekend. PAIRS 1 WEEKDAY WEEKEND OF SITE | CHI-SQ SIG LAMBDA CHI-SQ SIG LAMBDA SKYWAY VS MALL(R) : 2.5432 0.1108 0.06689 0.4480 0.5033 0.0000 SKYWAY VS MALL(F) : 26.4268 0.0000 0.2107 5.8570 0.0155 0.07429 SKYWAY VS MARQ.AVE.: 9.0613 0.0026 0.1025 0.8373 0.3602 0.0000 (NOTE : The underlined significance values in the table demonstrates evidence of significant difference at 95% confidence level). Table 4.9: Pairwise comparison of facilities for the proportional distribution of Gender categories. Data Source: Appendix II; Tables 8 to 13. ROLE GENDER M a l e Fema le COUNT ROW PCT V i s i t o r Tour 1st S e c u r I t y Other Wo Not Oete ROW COL PCT Guard r k e r m i n e d TOTAL 1.00| 2 .00 3.00 4 . 0 0 | 5.00 1 .00 59 17 5 98 21 200 2 9 . 5 8.5 2 .5 4 9 . 0 10.5 66 .7 6 3 . 4 6 8 . 0 106*.0 64. 1 87 .5 2 .00 34 8 55 3 100 3 4 . 0 8 .0 55 .0 3.0 33 .3 36 .6 32 .0 35 .9 12.5 COLUMN 93 25 5 153 24 300 TOTAL 3 1 . 0 8.3 1 .7 51 .0 8 .0 100.0 Table 4.10: Crosstabulation of Gender by Role for Nicollet Mall(F), during weekdays. GENDER COUNT ROW PCT COL PCT PURPOSE 1 .00 A c c e s s , W a l k i n g 2.00 S h o p p i n g P romenad ing 3.00 5.00 R e c r e a t I n g Male 1.00| Female ROW TOTAL 2 . 0 0 | + COLUMN TOTAL 13 11 54 .2 4 5 . 8 2 2 . 0 32 .4 9 10 47 .4 5 2 . 6 15.3 29 .4 21 12 6 3 . 6 36 .4 35 .6 3 5 . 3 16 1 94. 1 5.9 27 . 1 2.9 59 34 63 .4 36 .6 24 25 .8 19 20 .4 33 35 .5 17 18.3 93 100.0 Table 4.11: Crosstabulation of Purpose by Gender for Nicollet Mall(F), during weekdays, for all Visitors observed. ROLE GENDER Ma l e Fema le COUNT ROW PCT V i s i t o r Tour 1st S e c u r i t y O the r Wo Not De te ROW COL PCT Guard r k e r rm1ned TOTAL 1 .00 2 .00 3.00 4 .00 5.001 1 .00 39 22 1 82 13 157 24 .8 14.0 .6 52 .2 8 .3 5 2 . 3 31 .0 68 .8 100.0 65 . 1 8 6 . 7 2 .00 87 10 44 2 143 60 .8 7.0 30 .8 1 .4 47 . 7 6 9 . 0 31 .3 34 .9 13.3 COLUMN 128 32 1 126 15 300 TOTAL 4 2 . 0 10. 7 . 3 4 2 . 0 5.0 100.0 Table 4.12: Crosstabulation of Role by Gender for Nicollet Mall(R), during weekdays. COUNT ROW PCT COL PCT PURPOSE 1 .00 A c c e s s , W a l k i n g GENDER Ma le Female ROW TOTAL 1.00| 2 . 0 0 | 2 .00 Shopp 1ng 3.00 PromenadIng 5.00 R e c r e a t 1 n g COLUMN TOTAL 10 33 . 3 2 5 . 6 20 6 6 . 7 2 3 . 0 9 19. 1 2 3 . 1 12 31 .6 30 .8 38 80 .9 43 .7 - - + + 26 6 8 . 4 29 .9 8 72 .7 2 0 . 5 3 2 7 . 3 3.4 39 3 1 . 0 87 6 9 . 0 30 23 .8 47 37 .3 38 30.2 11 8.7 126 100.0 Table 4.13: Crosstabulation of Purpose by Gender for Nicollet Mall(R), during weekdays, for all Visitors observed. ROLE COUNT GENDER M a l e Fema le ROW PCT V i s i t o r Tour 1st S e c u r i t y Other Wo Not Dete COL PCT Guard r k e r rmIned 1.001 2 . 0 0 | 3.00 4 .00 5 .00 1 .00 37 15 4 66 14 27 . 2 11.0 2.9 48 .5 10. 3 35 .9 46 .9 100.0 4 6 . 5 77 .8 2 .00 66 17 76 4 40 .5 10.4 46 .6 2.5 64 . 1 53 . 1 53 .5 22 .2 COLUMN 103 32 4 142 18 TOTAL 34 .4 10. 7 1. 3 47 .5 6 .0 ROW TOTAL 136 45 .5 163 54 .5 299 100.0 Table 4.14: Crosstabulation of Gender by Role for Skyway System, during weekdays. 127 GENDER COUNT ROW PCT COL PCT Male Female PURPOSE 1.00| 2 . 0 0 | 6 .00 S e a r c h I n g 3 5 0 . 0 8.1 7 .00 Hang ing Out COLUMN TOTAL 37 35 .9 1 .00 5 7 A c c e s s , W a l k i n g 41 .7 58 .3 13.5 10.6 2 .00 11 35 Shopp ing 23 .9 76. 1 29 .7 5 3 . 0 3.00 18 20 P romenad ing 47.4 52 .6 48 .6 30 .3 3 5 0 . 0 4 .5 1 100.0 1.5 66 64 . 1 ROW TOTAL 12 11.7 46 44. 7 38 36 .9 6 5.8 1 1.0 103 100.0 Table.4.15: Crosstabulation of Purpose by Gender on Skyway, on weekdays, for all visitors observed. 4.2.3 DISTRIBUTION OF ETHNIC GROUPS The Chi-square test for the proportional distribution of ethnic groups on the four facilities, did not provide enough evidence to believe at the 95% confidence level that a significant difference existed, either during weekdays or during weekend (appendix II, Table 14 & 15). This inference goes against the notions held by various authors (Whyte 1988, Robertson 1988) that the Skyways act as social barriers. They argued that several minority groups are under-represented on the facility, and White users are more likely to use the system. Non of the issues raised seem to have had any bearing on the route choice decision of the ethnic minorities which can be inferred from the high significance value of B-Tau and Gamma in Table 14 and 15, appendix II. Hence it may be concluded that the proportional distribution of ethnic categories on all the four facilities was not significantly different. 128 4.2.4 VARIATIONS IN USER ROLE CATEGORIES The proportional distribution of users by their trip purpose (role) on the facilities was significantly different, which was inferred from the evidence provided by the low significance value for Chi-square test both, during weekdays (sig. 0.0002; Table 4.16) and weekends (sig. 0.0005, for reordered role categories to satisfy the Chi-square test assumptions; Table 16, appendix II). However a low value of the lambda, 0.0725 for weekdays and 0.03 for the weekends (Table 4.16 & 4.17), suggested that there was a weak predictable association between the proportional distribution of users by role, and the four facilities. This does not imply, though, that the users on all the facilities were the same. A pair-wise comparison of the four facilities showed that a significant difference existed between the role distribution on the skyway and Marquette Avenue, although the weekend difference could be established only after reordering the role categories to satisfy the Chi-square test assumptions, with a Chi-square significance of 0.0000, and Lambda value equalling 0.01754. Little evidence of any significant difference between the skyway, and Mall(R) and Mall(F) could be found, either during weekdays or during weekend (Table 4.18). Depending on the category of role of the users, their trip purpose could be theoretically affected by some of the issues discussed earlier. Ease of accessibility, and the perceived safety level, to some extent could affect the trip pattern of occasional visitors and tourists. Control and regulations, too, could have a potential effect on the degree of presence of the lower class hangouts, destitute, punks, etc. Similarly, disorientation could have some bearing on the route choice of the occasional visitors. Ordering the sites in such ordinal scale, a Kruskal-Wallis one way A N O V A reveals some significant differences in the average distribution of user role types on the facilities (Table 4.19, Table 4.20). During weekdays Marquette Avenue was more likely to have workers (with the highest mean 129 rank; Table 4.19) as users than any other user type ( 62.5% of all users; Table 4.16), where as visitors and tourists were more likely to be on the skyway or the mall, as compared to the street. In contrast to Marquette Ave, only around 42% to 51% of all users constituted the workers categorj' on all the other facilities. The tourists as a user group were confined to the Skyway and Mall(R), as compared to Mall(F) and the street. For the tourists accessibility does not seem to have been the determinant, nor was orientation. However, the higher degree of perceived safety seem to have led to a larger distribution of tourists within the core area. Since only 14.4% of all tourists sampled were observed on Marquette avenue, and tourists constituted only 5.3% of all users on Marquette Avenue, it may be possible that the streets outside the core were perceived as unsafe. Besides, the physical environment may have been another factor. The core area of Minneapolis offers a considerably wider variety of activities and facilities than the street. Especially, for the tourists, the Skyway System itself probably becomes a point of attraction, considering there larger presence on the skyway. In addition, the wider variety of shops, retail establishments and eating places on the Skyway and the Mall, which the street does not offer, probably adds to the negative feeling about the street. Even the proportion of visitors on Marquette Avenue was far less than the corresponding percentage on the other facilities (27.9% on the Street as opposed to 42%, of all surveyed, on Mall(R)). As a result, the streets turned out to be used chiefly by workers, travelling between their work places and transport destinations, or the Nicollet Mall for lunch and lunch hour shopping. The above argument is supported by the fact that 68.9% of all the people on Marquette Ave. were found using the system for access/walking, during weekdays (Table 4.32), 76.4% of all the users walking on Marquette Ave. were workers, and 84.2% of all the workers .on the Street were walking (Table 4.21). The street outside the core area, including Marquette Ave, severely lacked in other usage types (Table 4.32). Among other role types, the average distribution of security staff on the facilities was consistent with many of the findings already discussed, and which will be discussed later. Security staff, as a user 130 category, were more likely to be on the Skyway than the mall or the street. This partly explains the larger proportional presence of female users (Section 4.2.2) and is also consistent with Robertson's finding that Skyways are perceived as safer places (including the Skyways in Minneapolis). For the visitors, accessibility and orientation may partly have been responsible for their lesser average distribution on the Skyway as against the Mall(R). The scene changed to a certain degree during the weekend. Workers were observed more on the two portions of the Mall and the Skyway, as compared to the streets (Table 4.20). This can be explained by the fact that most of the offices, except the Skyway and the mall establishments remain closed during the weekends. Hence, the few workers who enter downtown during the weekend, remain within the core area. Workers as a group constituted only 6.6% of all users downtown, during weekend (Table 4.17). The average distribution of the visitors did not change. However, a change in the average distribution of the tourists do not comply with the notions of perceived safety, nor the physical environment. They showed a larger average presence on the mall than the Skyway. This may partially be explained by the fact that the operating hours on the skyway reduces during weekend, and the virtual absence of users on some of the segments reduce the attractiveness of the Skyway. Most of the tourists outside the core area were early morning walkers and promenaders, and very few actually shopped, due to the severe lack of shopping facilities outside the core area, at grade level. Beside the early morning tourists, and a few visitors, the streets outside the core area remained virtually empty during weekend, as evident from the significantly fewer number of users who fell inside the sample (57 as against 200 for the Malls). A crowd time and again gathered near bus stops on the avenue, and the street got deserted once the bus left. The considerable reduction of automobile traffic on the roads imply that a majority of people residing outside downtown probably do not visit downtown during weekend. 131 S ITE COUNT ROLE V i s i t o r EXP ROW COL VAL PCT PCT Skyway 1.00 N i c o l l e t N i c o l l e t Marque t t ROW M a l l R M a l l F e Avenue TOTAL 2 . 0 0 | 3 .00 | 4 . 0 0 | 1 00 103 101 .4 2 5 . 7 % 34 .4% 126 101.8 31 .4% 4 2 . 0 % 93 101 .8 2 3 . 2 % 31 .0% 79 9 6 . 0 19 .7% 27 . 9 % 401 33 .9% Tour 1st 2 00 32 26 .3 30 .8% 10 .7% 32 26 .4 30 .8% 10. 7% 25 26 .4 2 4 . 0 % 8 .3% 15 2 4 . 9 14.4% 5.3% 104 8.8% 3.00 S e c u r i t y Guard 4 2.8 36 .4% 1. 3% 1 2.8 9. 1% . 3 % 5 2 .8 4 5 . 5 % 1.7% 1 2.6 9 . 1 % .4% 11 .9% 4 .00 O t h e r Worker 142 151 .3 2 3 . 7 % 4 7 . 5 % 126 151.8 21 . 1% 4 2 . 0 % 153 151.8 2 5 . 6 % 5 1 . 0 % 177 143. 2 2 9 . 6 % 6 2 . 5 % 598 50 .6% 5.00 Not D e t e r m i n e d 18 17.2 26 .5% 6 .0% 15 17.3 22 . 1% 5.0% 24 17.3 3 5 . 3 % 8 .0% 1 1 16.3 16.2% 3.9% 68 5.8% COLUMN TOTAL 299 2 5 . 3 % 300 2 5 . 4 % 300 2 5 . 4 % 283 2 3 . 9 % 1182 100.0% CHI-SQUARE O . F . 38 .05852 12 STAT IST IC LAMBDA SOMERS' D SIGNIFICANCE 0 .0002 SYMMETRIC 0 .04366 0 .07555 MIN E . F . CELLS WITH E . F . < 5 2.634 4 OF 20 ( 20 .0%) WITH S ITE DEPENDENT WITH ROLE DEPENDENT 0 .00000 0 .06890 0 .07256 0 .08363 STAT IST IC VALUE SIGNIFICANCE KENDALL ' S TAU B KENDALL ' S TAU C GAMMA 0.07591 0 .06889 0. 11105 0 .0013 0 .0013 Table 4.16: Crosstabulation of Role by Site for the weekday data. 132 COUNT EXP VAL ROW PCT COL PCT ROLE V i s i t o r Tour 1st S ITE Skyway 1 .OOl N i c o l l e t Mai 1 R 2.001 N i c o l l e t M a l l F 3.001 1 .00 2 .00 3.00 S e c u r i t y Guard 4 .00 O the r Worker 5.00 Not D e t e r m i n e d COLUMN TOTAL 131 119. 1 30. 5% 74. 9% 140 136 . 1 32. 6% 7 0 . 0 % 130 136. 1 30 .2% 6 5 . 0 % Marquet t e Avenue 4.001 - - + 29 38. 8 6 .7% 50 .9% 21 37 38 19 31 .8 36 .4 36 .4 10.4 18.3% 3 2 . 2 % 33 .0% 16.5% 12.0% 18 .5% 19 .0% 33. 3% 3 0 0 0 .8 .9 .9 .3 100.0% . 0 % .0% .0% 1.7% . 0 % .0% .0% 12 13 17 0 11.6 13.3 13.3 3.8 28 .6% 3 1 . 0 % 4 0 . 5 % .0% 6 . 9% 6 . 5 % 8 .5% .0% 8 10 15 9 11.6 13.3 13.3 3.8 19.0% 2 3 . 8 % 3 5 . 7 % 21 .4% 4 .6% 5.0% 7.5% 15.8% 175 200 200 57 2 7 . 7 % 3 1 . 6 % 31 .6% 9 .0% ROW TOTAL 430 6 8 . 0 % 115 18.2% 3 .5% 42 6 .6% 42 6 .6% 632 100.0% CHI-SQUARE 37 .01318 D . F . 12 STAT IST IC SIGNIFICANCE 0 .0002 SYMMETRIC MIN E . F . 0.271 WITH ROLE DEPENDENT CELLS WITH E . F . < 5 6 OF 20 ( 30 .0%) WITH S ITE DEPENDENT LAMBDA SOMERS' D 0 . 0 2 0 5 0 0 .10849 0 .00000 0.09182 0 .03009 0 .13257 STAT IST IC VALUE SIGNIFICANCE KENDALL ' S TAU B KENDALL 'S TAU C GAMMA 0 .11033 0 .08752 0 . 18323 0.0008 0 .0008 Table 4.17: Crosstabulation of Role by Site for the weekend data. 133 PAIRS WEEKDAY WEEKEND OF SITE CHI-SQ SIG LAMBDA CHI-SQ SIG LAMBDA SKYWAY VS MALL(R) : 5.33634 0.2545 0.07358 6.3364 0.1754 0.01714 SKYWAY VS MALL(F) : 2.74662 0.6011 0.05686 9.26917 0.0547 0.02286 SKYWAY VS MARQ.AVE.: 16.21594 0.0027 0.12367 27.20415 0.0000 0.01754 (NOTE : The underlined s i g n i f i c a n c e values i n the t a b l e demonstrates evidence of s i g n i f i c a n t d i f f e r e n c e at 95% confidence l e v e l ) . Table 4.18: Pairwise comparison of facilities for the proportional distribution of user Role types. Data Source: Appendix II; Tables 17 to 22. - - - - KRUSKAL-WALLIS 1-WAY ANOVA S ITE F a c i l i t y O b s e r v e d BY ROLE MEAN RANK CASES 558 68 401 ROLE = 1 V i s i t o r 514 77 104 ROLE = 2 Tour 1st 530 73 11 ROLE = 3 S e c u r i t y Guard 630 21 598 ROLE 4 Other Worker 571 79 68 ROLE = 5 Not D e t e r m i n e d 1182 TOTAL CORRECTED FOR TIES CASES CHI-SQUARE SIGNIFICANCE CHI-SQUARE SIGNIFICANCE 1182 17.2282 0 .0017 18.3790 0 .0010 Table 4.19: Mean ranking of user Role categories on the facilities, during weekdays. KRUSKAL-WALLIS 1-WAY ANOVA S ITE BY ROLE F a c i l i t y O b s e r v e d MEAN RANK CASES 301 .00 430 ROLE = 1 V i s i t o r 361.82 115 ROLE = 2 Tour 1st 8 8 . 0 0 3 ROLE = 3 S e c u r i t y Guard 302 .88 42 ROLE = 4 Other Worker 381 .61 42 ROLE = 5 Not De te rmined 632 TOTAL CORRECTED FOR TIES CASES CHI-SQUARE SIGNIFICANCE CHI-SQUARE SIGNIFICANCE 632 20 .3952 0 .0004 22 .2983 0 .0002 Table 4.20: Mean ranking of user Role categories on the facilities, during weekend. COUNT ROW PCT COL PCT PURPOSE 1.00 A c c e s s , Wa l k i ng 2.00 Shopp Ing 3.00 5.00 R e c r e a t i n g 6 .00 S e a r c h 1 n g 7 .00 H a n g i n g Out ROLE V i s i t o r P romenad ing 4 .00 To Work, Work ing COLUMN TOTAL T o u r i s t S e c u r i t y Other Wo Not De te Guard r k e r rm lned 1.00| 2 . 0 0 | 3 .00 | 4 . 0 0 | 5 . 0 0 | 100.0 1 . 3 79 27 .9 15 5.3 177 6 2 . 5 35 17.9 4 4 . 3 8 4. 1 53 .3 149 76 .4 84 .2 3 1.5 27 .3 12 4 2 . 9 15.2 1 3.6 6 .7 14 5 0 . 0 7.9 1 3.6 9.1 29 76 .3 36 .7 6 15.8 4 0 . 0 3 7.9 1.7 1 8.3 100.0 11 91 .7 6.2 2 4 0 . 0 2.5 3 6 0 . 0 2 7 . 3 4 100.0 36.4 1 1 3.9 ROW TOTAL 195 6 8 . 9 28 9 .9 38 13.4 12 4.2 5 1.8 4 1 .4 283 100.0 Table 4.21: Crosstabulation of Purpose by Role for Marquette Avenue, for weekday data. 135 The number of people in the 'not determined' category is an area worth analyzing. They include the people who are generally not accepted into the average society, including destitute, vagrants, hangouts, hippies, punks. Prostitutes and pimps did not have any observable presence on any of the facility. Neither of the facilities had any significant presence of the 'not defined' category of users (3.9% to 15.8% of the entire population during weekdays and weekend). On an average they were more likely to be on Mall(F) than on any other facility, with the average distribution on the other facilities remaining fairly similar. The remote location of Mall(F) as compared to the other facilities may be one of the factors. The relative^' similar degree of distribution of this user type on the other three facilities do not satisfy the notion of control and regulation. In general, however, they did not have a conspicuous presence on any facility. Most of the people in this category used Hennepin Avenue (a block North-West of Nicollet Mall; (Fig.3.6) for their activities. Probably, the core area, with the skyway and the mall, do not fit into their cognitive conception of an ideal environment for their activities. Some of the Skyway segments would not accept them due to the strict dress code required by the facility, and the Mall(R) does not seem to be conducive to their usage needs. A few of them occasionally wander into the core area, but they remain inconspicuous and also exit inconspicuously. Once again, private control of skyway spaces did not seem to have an impact on the usage pattern, as is evident from the more or less similar distribution of users in the 'Not Determined' category. Among other issues, orientation and perceived safety may have had a marginal impact on the presence of visitors, considering their lesser proportional presence on the Skyway System as against the mall, but the difference is so neglegible that it may not be erronous to assume a total absence of impact of such issues (Appendix III, Fig.5 & 6). Lastly, there was no indication of impact of accessibility on any category of user type on the facilities. 136 4.2.5 GROUPING ON THE FACILITIES There was sufficient evidence to believe that the proportional distribution of grouping pattern on the four facilities were different. The Chi-square test for independence had a low significance value both, during weekdays (sig. 0.0007) and weekend, (sig. 0.0035) to support such a belief. However, the predictability was weak, as can be inferred from the low value of lambda; 0.0408 for weekdays, and 0.0949 during weekend (Table 4.22, Table 4.23). Most of the differences in the test was explained by the difference between the proportional distribution of groups on the Skyway as against Mall(F), which varied both, during weekdays and weekend (Table 4.24). Once again, Marquette Avenue differed significantly from the Skyway during weekdays, although not during weekend. However, the degree of association between the Skyway and Mall(F) was more (lambda 0.110 for weekdays and 0.1657 during weekends) than that between Skyway and Marquette Ave. (lambda 0.0918). A possible explanation may be that the differences in grouping pattern increased with the distance from the core are, since Mall(F) lies farther from the core than any other facility. One of the factors with a theoretical probability of affecting the group size is the perceived safety level. Users with families may tend to restrict themselves to safer areas downtown. The other factor could be the availability of retail establishments. Here again, people with families should tend to be in areas with more activities, and larger retail facilities. In either case, the skyways seem to be the best place for larger groups, with a larger degree of perceived safety and more retailing opportunities. The Street on the other hand lack in both, perceived safety (section 4.2.2 & 4.2.4) and retail facilities . A Kruskal-Wallis test for mean rank of the distribution of user groups showed a similar trend (Table 4.25, Table 4.26). During the weekdays as well as weekend, larger groups including three or more adults, and families with children (especially visitors and tourists) were more likely to be on the skyway as compared to the street or the mall (see also Table 4.22, Table 4.23). Contrary to the notion of perceived 137 safety, people observed alone were more likely to be on the street (Marquette Ave.). However, most of the people observed alone on Marquette Avenue were workers (65.4%, Table 4.27) who used, the streets mainly for access, regularly, and hence may fall outside the general influence of perceived safety levels. This finding is consistent with the previous inferences drawn that most of the people on Marquette Ave. were downtown workers, who used the facility primarily as an access route from their work place to other destinations (Section 4.2.4). All larger groups constituted of 3 or more workers, were more engage in shopping and promenading than access/walking (Table 4.28), and hence used the skyway for their needs (60% of all observed, Table 4.29) rather than the street. For such worker groups, skyway was also probably the preferred direct access route from other areas of downtown, to the shopping areas in the core. Beside the perceived safety factor the quality of physical environment also had a potential determining effect on the route choice of visitors with families. The mall and, to a lesser extent, the skyway provide comparatively more facilities and amenities catering to the needs of such user groups including the facilities for resting, larger variety of visual input, and a comparatively more festive atmosphere. The lesser proportion of larger groups on Mall(F) in spite of better furnishing can be explained by the considerably lesser degree of retailing facility on the mall. The larger average distribution of the 'not determined' user category, and Mall(F)'s comparatively remote location could also have been some of the other factors. The average distribution did not change considerably during weekend, and the same inferences can be drawn. An aspect worth noticing is the significantly large number of users who were alone. On the whole, single persons constituted the largest group type on all the facilities (75.4% during weekdays and 56% during weekend, of all users in the surveyed facilities; Table 4.22 & 4.23). Most of the policies regarding urban design of public areas are tuned towards the users in groups, or with families, where as in fact most of the users in public 138 areas remain alone. In contrast to notions of perceived safety and physical environment, the large observed turn out of couples on the street (table 4.22) may be explained by the fact that a majority of them were workers (Table 4.27) during lunch hours, who constituted 59.3% of the same sex couples and 50% of the man woman couple, during weekdays, on Marquette avenue. With the absence of workers in downtown during weekend, the number of couples on the street reduced considerably. The number of visitors and tourists in couples, on the street, was considerably less. Perceived safety, out of all the issues, seem to have had some impact on system usage considering the larger presence of bigger groups on the Skyway and Mall(R) as compared to the street. The physical environment might have been another factor. The relatively lesser presence of larger groups on Mall(R), as compared to the Skyway, may have been due to the lack of furnishing elements during the ongoing renovation work. After completion of the mall work, it will probably experience an increase in the proportional visits of larger groups, which were mostly constituted of occasional visitors with families, and tourists (Appendix III, Fig.7 & 8). Nevertheless, the larger average distribution of bigger groups on the Skyways indicates its positive role as a public place, although under private control. 4.2.6 GROUP SIZE A clearer picture of the grouping pattern can be observed from the relative sizes of each group on the facilities. Consistent with the findings on grouping pattern, the Chi-square test for proportional distribution of group sizes showed sufficient evidence to believe that the proportional distribution on the four facilities differed (sig. 0.0001 for weekdays and 0.0006 for weekend; Table 4.30, Table 4.31). Considering the facilities as increasingly 139 COUNT EXP VAL ROW PCT COL PCT GROUPING A lone S ITE Sk yway N i c o l l e t M a l l R 1.00| 2 .00 | N i c o l l e t M a l l F 3.001 Marque t t e Avenue 4.001 1 .00 2 .00 Two, Same Sex 3.00 Two, Man & Woman 4 .00 3 or More A d u l t s 5 .00 F a m i l y , C h i l d r e n COLUMN TOTAL 204 225 .4 2 2 . 9 % 6 8 . 2 % 224 226. 1 25 . 1% 74. 7% 235 226. 1 26 .4% 78. 3% 228 213 .3 25 .6% 8 0 . 6 % 36 31 .4 2 9 . 0 % 12 .0% 31 31 .5 2 5 . 0 % 10.3% 30 31 .5 2 4 . 2 % 10 .0% 27 29 .7 2 1 . 8 % 9 .5% 16 19.0 2 1 . 3 % 5.4% 22 19.0 2 9 . 3 % 7.3% 19 19.0 25 . 3% 6 .3% 18 18.0 2 4 . 0 % 6 .4% 20 11.1 4 5 . 5 % 6 . 7 % 9 11.2 2 0 . 5 % 3.0% 11 11.2 2 5 . 0 % 3 .7% 4 10.5 9. 1% 1.4% 23 12. 1 4 7 . 9 % 7.7% 14 12.2 2 9 . 2 % 4 .7% 5 12.2 10.4% 1.7% 6 11.5 12 .5% 2. 1% 299 300 2 5 . 3 % 2 5 . 4 % 300 2 5 . 4 % 283 2 3 . 9 % ROW TOTAL 891 7 5 . 4 % 124 10 .5% 75 6 .3% 44 3.7% 48 4. 1% 1182 100.0% CHI-SQUARE D . F . 33 .72237 12 STAT IST IC SIGNIFICANCE 0 .0007 SYMMETRIC MIN E . F . CELLS WITH E . F . < 5 10.535 NONE WITH GROUPING DEPENDENT WITH S ITE DEPENDENT LAMBDA SOMERS' D 0 .03069 -0.09781 0 .00000 -0.07589 0 .04082 -0.13755 STATIST IC VALUE SIGNIFICANCE KENDALL ' S TAU B KENDALL ' S TAU C GAMMA -0.10217 -0.07587 -0.18309 0 .0000 0 .0000 Table 4.22: Crosstabulation of Grouping by Site, for the weekday data. 140 COUNT EXP VAL ROW PCT COL PCT GROUPING A lone S ITE Skyway 1.001 N i c o l l e t N i c o l l e t M a l l R M a l l F 2 . 0 0 | 3 .00 | Marquet t e Avenue 4 . 0 0 | 1 .00 2 .00 Two, Same Sex 3.00 Two, Man & Woman 4 .00 3 or More A d u l t s 5.00 F a m i l y , C h i l d r e n COLUMN TOTAL 87 9 8 . 0 24 .6% 4 9 . 7 % 102 112.0 28 .8% 5 1 . 0 % 137 112.0 38 .7% 6 8 . 5 % 28 31.9 7.9% 49. 1% 21 17.4 33. 3% 12 .0% 23 19.9 3 6 . 5 % 11 .5% 13 19.9 20 .6% 6 .5% 6 5.7 9.5% 10.5% 24 28 .8 23 . 1% 13.7% 38 32 .9 36 .5% 19 .0% 28 32 .9 26 .9% 14 .0% 14 9.4 13.5% 24 .6% 17 11.4 4 1 . 5 % 9 .7% 13 13.0 3 1 . 7 % 6 .5% 10 13.0 24 .4% 5.0% 1 3.7 2.4% 1.8% 26 19.4 37. 1% 14 .9% 24 22 .2 34. 3% 12.0% 12 22 .2 17 .1% 6 .0% 8 6 .3 11.4% 14.0% 175 200 200 57 2 7 . 7 % 31 .6% 31 .6% 9 .0% ROW TOTAL 354 56 .0% 63 10 .0% 104 16 .5% 41 6 .5% 70 11 .1% 632 100.0% CHI-SQUARE D .F . 2 9 . 3 8 4 3 9 12 STAT IST IC SIGNIFICANCE 0 .0035 SYMMETRIC MIN E . F . 3.698 CELLS WITH E . F . < 5 1 OF WITH GROUPING DEPENDENT 20 ( 5.0%) WITH S ITE DEPENDENT LAMBDA SOMERS' D 0 .05775 -0.09696 0 .00000 -0.09139 0.09491 -0.10325 STAT I ST IC VALUE SIGNIFICANCE KENDALL ' S TAU B KENDALL ' S TAU C GAMMA 0 .09714 0.08711 0 .14292 0 .0022 0 .0022 Table 4.23: Crosstabulation of Grouping by Site, for the weekend data. 141 PAIRS WEEKDAY WEEKEND OF SITE CHI-SQ SIG LAMBDA CHI-SQ SIG LAMBDA SKYWAY VS MALL(R) : 8.61505 0.0715 0.08361 3.40447 0.4926 0.03429 SKYWAY VS MALL(F) : 17.17437 0.0018 0.11037 18.74010 0.0009 0.16571 SKYWAY VS MARQ.AVE.: 22.94636 0.0001 0.09187 6.70286 0.1524 0.0000 (NOTE : The underlined s i g n i f i c a n c e values i n the t a b l e demonstrates evidence of s i g n i f i c a n t d i f f e r e n c e at 95% confidence l e v e l ) . Table 4.24: Pairwise comparison of facilities for the proportional distribution of Grouping types. Data Source: Appendix II; Tables 23 to 28. KRUSKAL-WALLIS 1-WAY ANOVA SITE F a c i l i t y Observed BY GROUPING MEAN RANK CASES 611.41 891 GROUPING = 1 A lone 563.92 124 GROUPING = 2 Two, Same Sex 603.57 75 GROUPING = 3 Two, Man & Woman 442 . 14 44 GROUPING = 4 3 or More A d u l t s 4 11.18 48 GROUPING = 5 Fami1y, Chi l d r e n 1182 TOTAL CORRECTED FOR TIES CASES CHI-SOUARE SIGNIFICANCE CHI-SQUARE SIGNIFICANCE 1182 25.7532 0.0000 27.4733 0.0000 Table 4.25: Mean ranking of Grouping categories on the facilities, during weekdays. 142 - - - - KRUSKAL-WALL IS 1-WAY ANOVA SITE F a c i l i t y Observed BY GROUPING MEAN RANK CASES 332.80 354 GROUPING = 1 A lone 285.56 63 GROUPING = 2 Two, Same Sex 330.30 104 GROUPING = 3 Two, Man & Woman 254.55 4 1 GROUPING = 4 3 or More A d u l t s 277 .69 70 GROUPING = 5 F a m i l y , C h i l d r e n 632 TOTAL CASES 632 CHI-SQUARE 13.1093 SIGNIFICANCE 0.0108 CORRECTED FOR TIES CHI-SQUARE SIGNIFICANCE 14.3325 0.0063 Table 4.26: Mean ranking of Grouping types on the facilities, during weekend. COUNT ROW PCT COL PCT GROUPING A lone ROLE V i s i t o r T o u r i s t S e c u r i t y Other Wo Not Dete ROW Guard r k e r r mined TOTAL 1.00| 2.00| 3.00| 4.00| 5.00| 1 .00 Two, Two , 4 .00 3 or More A d u l t s Fam i 1 y , 2 .00 Same Sex 3 . 00 Man & Woman 5.00 Chi 1dren COLUMN TOTAL 6 100.0 7 .6 79 27 . 9 15 5. 3 58 25 . 4 73.4 1 1 4 . 8 73.3 1 . 4 100.0 149 65 . 4 84.2 9 3.9 81.8 8 29.6 10. 1 1 3.7 6 . 7 16 59 . 3 9.0 2 7.4 18.2 6 33 . 3 7.6 3 16.7 20.0 9 50.0 5. 1 1 25.0 1 . 3 3 75.0 1 . 7 177 62 . 5 1 1 3.9 228 80.6 27 9.5 18 6 . 4 4 1 .4 6 2. 1 283 100 .0 Table 4.27: Crosstabulation of Grouping by Role for Marquette Avenue, for the weekday data. 143 COUNT ROW PCT COL PCT PURPOSE Ac c e s s , Shopping Promenad To Work, R e c r e a t i S e a r c h i n W a l k i n g ing Working ng g 1.00 | 2.00| 3.00| 4.00| 5 . 00| 6 . 00 | GROUPING A 1 one 1 .00 Two , Two , F a m i l y , 2.00 Same Sex 3 . 00 Man & Woman 4 .00 3 or More A d u l t s 5.00 Ch i l d r e n COLUMN TOTAL 446 74.6 77 12.9 100.0 2.4 41 6.9 27 4.5 4 . 7 381 77.3 85.4 60 12.2 77.9 21 4.3 51 .2 24 4.9 88.9 4 .8 100.0 3 .6 100.0 39 62.9 8.7 9 14.5 11.7 13 21.0 31.7 1 1 .6 3.7 21 75.0 4 . 7 3 10. 7 3.9 4 14 . 3 9.8 5 35.7 1 . 1 5 35. 7 6.5 2 14.3 4.9 2 14.3 7 . 4 ROW TOTAL 493 82.4 62 10 . 4 28 4 . 7 14 2.3 598 100.0 Table 4.28: Crosstabulation of Grouping by Purpose for the weekday data, for all workers observed on the facilities. COUNT ROW PCT COL PCT SITE Skyway 1 GROUPING A lone N i c o l l e t N i c o l l e t Marquett Mai 1 R Ma 11 F e Avenue 00| 2.00| 3.00| 4.00| 1 .00 2.00 Two, Same Sex Two . 3.00 Man & Woman 4 .00 3 or More A d u l t s COLUMN TOTAL 19 31.7 67.9 16 26. 7 88.9 14 23. 3 82.4 1 1 18.3 78.6 4 44 . 4 14.3 2 22.2 1 1.8 3 33.3 21.4 2 66 . 7 7 . 1 1 33. 3 5.6 3 60.0 10. 7 1 20.0 5.6 1 20.0 5.9 28 18 17 14 36.4 23.4 22.1 18.2 ROW TOTAL 60 77.9 1 1 5 6.5 77 100.0 Table 4.29: Crosstabulation of Grouping by Site, for all Workers observed Shopping during weekdays. 144 unsafe (perceived safety) from the skyway to the street, and with decreasing level of retail facility and activities, Kendall's Tau B and C showed significant results throughout the week. The association, although not impressive (Somers' D -0.14507 for weekdays and -0.12113 for weekend) was moderately strong and negative. This implies that as the perceived safety level and retail activities decreased (from the Skyways to the Street), the size of the user groups decreased. The phenomena was stronger during weekdays than weekend. The weaker weekend association can be explained by the fact that with substantial decrease in automobile traffic on the roads, visitors and tourists tend to spread out of the core area into other areas inside downtown. Also, free parking facility during weekend, probably attracted more number of visitors in larger groups to the downtown. 4.3.0 Activities on the Facilities 4.3.1 PURPOSE The facilities differed significantly in the proportional distribution of mobile activity types. The Chi-square test gave sufficient evidence in support of such a belief (sig. 0.0000 for weekdays: Table 4.32, and 0.0000 for weekend with reduced categories to meet test assumptions: Table 29, appendix II). The association was moderately strong, as can be observed in the values of lambda (0.1258 for weekdays and 0.118 for weekend) (Table 4.32, Table 4.33). A pair-wise comparison of the site categories (with reclassification to meet Chi-square test assumptions) made the differences clearer (Table 4.34). The skyway differed significantly from all the other three sites during both, weekdays and weekend, in the proportional distribution of purpose categories. Difference existed, however, in the 145 strength of association between each pair of variable. During weekdays, the variability between skyway and Marquette Avenue (Lambda = 0.28269) was considerably stronger than between skyway and Mall(F) (Lambda = 0.21070). The least variability existed between the skyway and Mall(R), with a Lambda value of only 0.12040. Variations in the strength of association for the weekend data was not clear due to the 0.0000 value of Lambda between the skyway and Marquette Avenue. Beside the larger difference of the activity type between the street and the Skyway System, the data exhibit a lot more intricate differences. During the weekdays as well as the weekends, searching as an activity type occurred mostly on the skyway (Table 4.32 & 4.33). This finding is consistent with notions forwarded by various authors that the Skyway System are disorienting. The most visible evidence of disorientation was the large number of users (not in the sample) who spent considerable time studying maps at the few locations where such maps have been provided. However, disorientation does not seem to have had any impact on the route choice decision on the theoretically potential affected group; the elderly. This may be due to the fact that the elderly users tend to have a constant habituated route pattern. In fact, people disoriented on the skyway were more likely to be adults (73.7% of all users searching; Table 4.35) and visitors (68.4% of all people searching; Table 4.36). The adult visitors who searched their way were probably occasional visitors. However, disorientation did not seem to have any major impact on the route choice of the visitors as a whole, since the proportions of visitors on the Skyway and the Mall was not considerably different (Table 4.16, Table 4.17). A part of the explanation could be the then ongoing renovation work on the mall, which made using the mall relatively difficult, and may have had resulted in a shifting of some of the visitors on to the skyway network. On the whole, however, the proportion of users searching was negligible (0.9% during weekdays and 1.3% during weekend, of all users surveyed), leading to sufficient grounds to believe that the skyways may not be disorienting to the majority of users. This inference is consistent with Robertson's finding that most of the surveyed users, in all the five cities, did not 146 experience disorientation, neither did disorientation seem to have had any association with the level of signage inside the facility (Robertson, 1988). During weekdays, shopping, promenading, and hanging out were more likely to be observed on the skyway and mall(R), than Mall(F) and the street (Appendix III, Fig.9). Especially important is the fact that shopping as an activity was observed more on the skyway than Mall(R), which is the principal retail street in downtown Minneapolis. A part of the phenomena can be explained by the then ongoing construction work on Nicollet Mall. However, the physical environment probably constituted another important factor. Details of the variations and shortcomings in the physical environment is discussed in the next chapter. There was negligible variation in the proportion of promenaders on the Skyway, Mall(R) and Mall(F). However, Marquette Ave. entertained a considerably lesser proportion of promenaders. Several factors, including the lesser degree of complexity in the street environment, more automobile traffic, a lesser perceived safety level, lesser density and number of people on the sidewalks, and lesser activity level, may explain such a phenomena. The distribution of recreating activities provided expected results. Most of the people recreating were observed on the mall, as compared to the street or the skyway. The negligible degree of recreation on the skyway was not unexpected, since the facility design is not conducive to such form of activities (jogging, bicycling, roller skating, playing, etc.). The facility management strategy of the skyway also strictly prohibits any recreational activity. On the other hand, the streets, with all the motorized traffic are not the safest place for such user groups, which leaves the mall for people to recreate. The lesser degree of recreational activity observed on Mall(R), again, can be attributed to the then ongoing construction work. Clearly, the skyway offered the least opportunities for such activity types. 147 The street (Marquette Ave.) was host to one predominant activity type; access/walking. Where as walking constituted to only 40.1% to 49% of all activity types on the other facilities, 68.9% of all activity types on the street belonged to this category, during weekdays. A similar trend can be observed during weekend, when 42.1% of all users on the street were observed walking as against 22% to 26.9% for the other facilities. This, again, supports the previous data, that most of the users of the streets were workers, who used the sidewalks for traversing between their place of work and other downtown destinations. Even during weekend, when most of the people using the streets were visitors (50.9%; Table 4.37) and not workers, access/walking remained the major activity type (42% of all users were engaged in this activity type), which again reflects on the lesser compatibility of the streets to any form of activity other than access. Some of the activities changed during weekend in their average distribution on the facilities (Appendix III, Fig. 10). More people were observed hanging out on the mall than on the skyway or the street. A possible explanation could be that the skyway with reduced operating hours, negligible number of workers (constituting one of the largest category of skyway users), and lesser people, probably did not seem as attractive as the mall for hanging out. Good, sunny weather could have been another factor. The mall on the other hand was always open, and was host to a larger variety of activity types, provided a more festive environment during weekend, as compared to the skyway. However, for the majority of users, the skyway seemingly remained as attractive as Mall(R), and to a lesser extent Mall(F), to engage in various activity types. Contrary to the primary management strategy, lesser proportion of people used the skyway corridors for access, during weekdays, as compared to the other facilities. With a comparatively larger proportion of people using it for shopping, promenading, or even hanging out, its performance as a public place was quite admirable considering the private ownership and strict control of the facility segments. 148 SI2E COUNT EXP VAL ROW PCT COL PCT SITE Skyway N i c o l l e t N i c o l l e t Marquett ROW M a l l R M a l l F e Avenue TOTAL 1.00| 2.001 3.00| 4. 00 | 1 .00 2.00 3.50 COLUMN TOTAL 204 225.4 22.9% 68.2% 224 226. 1 25. 1% 74 . 7% 235 226 . 1 26 . 4% 78. 3% 228 213.3 25.6% 80.6% 891 75.4% 52 50. 3 26 . 1% 17.4% 53 50. 5 26.6% 17.7% 49 50. 5 24.6% 16.3% 45 47 .6 22.6% 15.9% 199 16.8% 43 23.3 46.7% 14.4% 23 23.4 25.0% 7 . 7% 16 23.4 17.4% 5.3% 10 22.0 10.9% 3.5% 92 7.8% 299 25. 3% 300 25. 4% 300 25.4% 283 1182 23.9% 100.0% CHI-SQUARE O.F. SIGNIFICANCE MIN E.F. CELLS WITH E.F.< 5 29,38436 0.0001 22.027 NONE STATISTIC SYMMETRIC WITH SIZE DEPENDENT WITH SITE DEPENDENT LAMBDA SOMERS' D 0.02643 -0.10050 0.00000 0.07688 0.03515 -0.14507 STATISTIC VALUE SIGNIFICANCE KENDALL'S TAU B KENDALL'S TAU C GAMMA -0.10561 -0.08647 -0.19298 0.0000 0.0000 Table 4.30: Crosstabulation of Size categories by Site for the weekday data. SIZE COUNT EXP VAL ROW PCT COL PCT SITE Skyway 1 N i c o l l e t N i c o l l e t Marquett M a l l R Ma 11 F e Avenue 00| 2.00| 3.00| 4.00| 1 .00 87 98.0 24.6% 49 . 7% 102 112.0 28. 8% 51.0% 137 112.0 38. 7% 68.5% 28 31.9 7.9% 49 . 1% 2.00 45 46.2 26.9% 25.7% 61 52.8 36.5% 30.5% 41 52.8 24.6% 20.5% 20 15.1 12.0% 35. 1% 3.50 43 30. 7 38.7% 24.6% 37 35. 1 33.3% 18.5% 22 35. 1 19.8% 11.0% 9 10.0 8. 1% 15.8% COLUMN 175 TOTAL 27.7% 200 31.6% 200 31.6% 57 9.0% ROW TOTAL 354 56 .0% 167 26.4% 1 1 1 17.6% 632 100.0% CHI-SQUARE O F . SIGNIFICANCE MIN E.F. CELLS WITH E.F.< 5 23.75466 0.0006 10.011 NONE STATISTIC SYMMETRIC WITH SIZE DEPENDENT WITH SITE DEPENDENT LAMBDA SOMERS' D 0.05775 -0. 10909 0.00000 -0.09922 0.09491 -0.12113 STATISTIC VALUE SIGNIFICANCE KENDALL'S TAU B KENDALL'S TAU C GAMMA -0.10963 -0.10640 -0.16749 0.0009 0.0009 Table 4.31: Crosstabulation of Size categories by Site for the weekend data. 150 COUNT EXP VAL ROW PCT COL PCT PURPOSE 1 .00 A c c e s s , W a l k i n g S ITE Skyway 1 .001 N l c o l l e t M a l l R 2.001 N i c o l l e t M a l l F 3 . 0 0 | 120 151.0 20 . 1% 40. 1% 135 151 .5 22. 6% 4 5 . 0 % 147 151 .5 24. 6% 4 9 . 0 % Marque t t e Avenue 4 . 0 0 | + 195 142.9 32 .7% 6 8 . 9 % Shopp1ng 2 .00 86 57 .2 38. 1% 2 8 . 8 % 72 57 .4 3 1 . 9 % 2 4 . 0 % 40 57 .4 17 .7% 13 .3% 28 54 . 1 12.4% 9 .9% 3.00 P romenad ing 4 .00 To Work. Work ing 62 60 . 7 2 5 . 8 % 2 0 . 7 % 70 6 0 . 9 2 9 . 2 % 2 3 . 3 % 70 6 0 . 9 2 9 . 2 % 2 3 . 3 % 38 57 .5 15 .8% 13.4% 12 9.4 32 .4% 4 . 0 % 2 9.4 5.4% . 7% 11 9.4 29 . 7% 3.7% 12 8 .9 32 .4% 4 .2% ROW TOTAL 597 5 0 . 5 % 226 19. 1% 240 20. 3% 37 3.1% 5 00 0 14 29 5 48 R e c r e a t 1 n g 12. 1 12 .2 12.2 11 5 4. 1% . 0 % 29 .2% 60 . 4% 10 4% . 0 % 4 . 7% 9 .7% 1 8% 6 00 10 0 0 1 11 S e a r c h I n g 2 .8 2 .8 2.8 2 6 . 9 % 9 0 . 9 % .0% . 0 % 9 1% 3.3% .0% .0% 4% 7 00 9 7 3 4 23 H a n g i n g Out 5.8 5 .8 5.8 5 5 1.9% 39. 1% 30 .4% 13 .0% 17 4% 3 .0% 2 . 3% 1.0% 1 4% COLUMN 299 300 300 283 1182 TOTAL 25 . 3% 25 .4% 2 5 . 4 % 23 9% 100.0% CHI - SQUARE D . F . S IGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 149.05793 18 0 .0000 2.634 4 OF 28 ( 14.3%) STATIST IC LAMBDA SOMERS' D SYMMETRIC 0 .07566 -0.12454 WITH PURPOSE DEPENDENT 0 .00000 -0 . 1 1741 WITH SITE DEPENDENT 0 .12585 -0.13259 STATIST IC KENDALL 'S TAU B KENDALL 'S TAU C VALUE •0.12477 -0.11739 SIGNIFICANCE 0 .0000 0 .0000 Table 4.32: Crosstabulation of Purpose types by Site for the weekday data. 151 COUNT EXP VAL ROW PCT COL PCT PURPOSE 1 .00 A c c e s s , W a l k i n g SITE Skyway N i c o l l e t M a l l R 1.00| 2 . 0 0 | N i c o l l e t Marque t t M a l l F e Avenue 3 .00 | 4 . 0 0 | 2 .00 S h o p p i n g 3 .00 R e c r e a t i ng 5.00 S e a r c h i n g 6 .00 H a n g i n g Out 7 .00 P r o m e n a d i n g 4 .00 To Work. Work ing COLUMN TOTAL 47 44 .0 29 .6% 2 6 . 9 % 44 50 .3 2 7 . 7 % 2 2 . 0 % 44 50. 3 2 7 . 7 % 2 2 . 0 % 24 14.3 15. 1% 42 . 1% 59 47 .6 34 .3% 33 .7% 68 54 .4 39 .5% 34 .0% 35 54 .4 2 0 . 3 % 17.5% 10 15.5 5.8% 17.5% 53 59 .8 24 .5% 30. 3% 62 68 .4 28. 7% 31 .0% 84 68 .4 38 .9% 4 2 . 0 % 17 19.5 7.9% 2 9 . 8 % 5 2.2 6 2 . 5 % 2 .9% 3 2.5 37 .5% 1.5% 0 2.5 .0% .0% 0 . 7 .0% .0% 0 14. 1 .0% .0% 16 16. 1 31 .4% 8.0% 32 16. 1 6 2 . 7 % 16.0% 3 4 .6 5.9% 5.3% 8 2.2 100.0% 4 .6% 0 2.5 .0% .0% 0 2.5 .0% .0% 0 . 7 .0% .0% 3 5.0 16.7% 1.7% 7 5.7 38 .9% 3.5% 5 5.7 27 .8% 2 .5% 3 1.6 16 .7% 5.3% 175 200 200 57 27 .7% 31 .6% 31 .6% 9 .0% ROW TOTAL 159 25 .2% 172 27 .2% 216 34 .2% 8 1.3% 51 8 .1% 8 1.3% 18 2 .8% 632 100.0% CHI-SOUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 8 8 . 8 9 0 3 3 18 0 .0000 0 .722 11 OF 28 ( 39.3%) LAMBDA SOMERS 1 STAT IST IC STAT IST IC KENDALL ' S TAU 6 KENDALL ' S TAU C SYMMETRIC 0 .08255 0 .06565 VALUE 0.06566 0 .06360 WITH PURPOSE DEPENDENT 0.04567 0.06672 SIGNIFICANCE 0.0253 0 .0253 WITH S ITE DEPENDENT 0 .11806 0 .06462 Table 4.33: Crosstabulation of Purpose types by Site for the weekend data. PAIRS OF SITE CHI-SQ WEEKDAY SIG LAMBDA CHI-SQ WEEKEND SIG LAMBDA SKYWAY VS MALL(R) : 33.99899 0.0000 0.12040 16.9069 0.0020 0.02857 SKYWAY VS MALL(F) : 62.05080 0.0000 0.21070 41.46021 0.0000 0.18286 SKYWAY VS MARQ.AVE.: 67.02343 0.0000 0.28269 7.8897 0.0483 0.0000 (NOTE : The underlined s i g n i f i c a n c e values i n the t a b l e demonstrates evidence of s i g n i f i c a n t d i f f e r e n c e at 95% confidence l e v e l ) . Table 4.34: Pairwise comparison of facilities for the proportional distribution of Purpose Types. Data Source: Appendix II; Tables 30 to 35. COUNT ROW PCT COL PCT PURPOSE 1 .00 A c c e s s , Wa1k ing AGE C h i l d Teen A d u l t E l d e r l y ROW <13 13 t o 20 to >64 TOTAL 1.00| 2.00| 3.00| 4.00| 2.00 Shopp i ng 3 .00 Promenading 4 .00 To Work, Working 5.00 R e c r e a t i n g 6.00 Se a r c h i n g 7 .00 Hanging Out COLUMN TOTAL 34 1 .9 56 3. 1 3 . 4 8.8 8 1 . 1 14.3 717 94.8 46 . 3 28 3.7 15.9 10 2.5 29 . 4 16 4.0 28 . 6 325 81.7 21.0 47 11.8 26. 7 17 3.7 50.0 14 3. 1 25.0 333 73.0 21.5 92 20. 2 52. 3 3 6 . 7 5.4 42 93.3 2.7 7 7 . 1 12.5 91 91.9 5.9 1 1.0 .6 3 15.8 8 . 8 1 5.3 1 .8 14 73.7 .9 1 5.3 .6 1 2 . 4 2.9 7 17.1 12.5 26 63 . 4 1 . 7 7 17.1 4.0 1548 176 85.3 9.7 756 41.7 398 21.9 456 25. 1 45 2.5 99 5.5 19 1 .0 4 1 2 . 3 1814 100.0 Table 4.35: Crosstabulation of Purpose by Age for all facilities, ignoring day. COUNT ROW PCT COL PCT ROLE V i s i t o r T o u r i s t 1 .00 | PURPOSE • 1 .00 Access , W a l k i n g S e c u r i t y Other Wo Not Dete Guard r k e r rmined 2. 00 | 3.00| 4.00| 5.00| 2.00 Shopping 3.00 Promenad ing 4 .00 To Work, Work i n g 5.00 Recre a t i n g 6 .00 Search i ng 7 .00 Hanging Out 200 26 . 5 24. 1 44 5.8 20. 1 1 . 1 7 . 1 477 63. 1 74.5 34 4.5 30.9 272 68. 3 32. 7 42 10.6 19 . 2 1 . 3 7: 1 79 19.8 12.3 4 1 .0 3.6 269 59 .0 32.4 126 27.6 57 . 5 1 .2 7 . 1 46 10. 1 7.2 14 3. 1 12.7 1 2.2 . 1 1 2.2 . 5 1 1 24 .4 78.6 30 66.7 4 . 7 2 4.4 1 .8 74 74 . 7 8.9 2 2.0 . 9 4 4.0 .6 19 19.2 17.3 13 68 . 4 1 .6 3 15.8 1 . 4 3 15.8 .5 2 4 . 9 .2 1 2.4 .5 1 2.4 .2 37 90.2 33.6 COLUMN TOTAL 831 45.8 219 12.1 14 .8 640 35. 3 110 ROW TOTAL 756 41.7 398 21.9 456 25. 1 45 2.5 99 5.5 19 1 .0 41 2.3 1814 Table 4.36: Crosstabulation of Purpose by Role for all facilities, ROLE V i s i t o r COUNT ROW PCT COL PCT PURPOSE T o u r i s t Not Dete rmined 1 . 00 | 2.001 5.001 1 .00 A c c e s s , W a l k i n g 17 70.8 58.6 3 12.5 15.8 4 16 . 7 44 . 4 2 .00 Shopping 2 20.0 6.9 8 80.0 42. 1 3.00 Promenad i ng 9 52.9 31.0 8 47 . 1 42. 1 5.00 R e c r e a t i n g 1 33. 3 3.4 2 66 . 7 22 2 7 .00 Hanging Out 3 100.0 33. 3 COLUMN TOTAL 29 50.9 19 33. 3 9 15.8 6.1 100.0 ignoring day. ROW TOTAL 24 42. 1 10 17.5 1 7 29.8 3 5.3 3 5.3 57 100.0 Table 4.37: Crosstabulation of Purpose by Role for Marquette Avenue, for weekday data. 154 4.3.2 STATIONARY ACTIVITIES Stationary activities in and around the core areas did not occur to the same degree in the four facilities. This was concluded from the Chi-square test for weekday and weekend activities, which provided sufficient evidence to believe that differences existed (sig. 0.0030 for weekdays and 0.0035 for weekends, after reclassifying the variables to meet the test assumptions; Table 36 & 37, appendix II). The associations were moderately strong as shown by the lambda (0.1385 for weekdays and 0.1724 for weekends; Table 4.38 & 4.39). With marked deviation from the trend in the preceding analysis, a pair-wise comparison of the facilities demonstrated no evidence at the 95% significance level in support of any difference between the Skyway System and Marquette Avenue (Table 4.40), where as the differences between skyway and Mall(R), and skyway and Mall(F) were not only significant but were also strong which can be observed in the corresponding Lambda values. The phenomena implies the presence of very limited types of stationary activities on the skyway network, very similar to the usage pattern on the street. It is consistent with the primary objectives of the management strategies, wherein the skyway corridors were designed primarily as access routes. Since the degree of activities are partly dependant on the degree of control, considering the facilities as ranging from larger control in the skyway system to lesser control on Marquette Avenue, the variable site could be considered as ordinal categorical. The Kruskal-Wallis one-way A N O V A test demonstrates significant difference among the average distribution of activity types on the four facilities (Table 4.41, Table 4.42; Appendix III, Fig. 11 & 12). Although the skyway was host to lesser variety of activity types, the skyway segments during weekdays was more likely to witness people meet, than any other facility. Nearly 56.3% of all people meeting on all facilities, occurred on the skyways. This may not have been true, however, for the entire system. Since most of the survey sites were located inside the core, and overlooking atria spaces, the degree of meeting may have been overrepresented in the data. Furthermore, 155 most of the people observed meeting on the skyway were either workers (44.4% of the users engaged in that category) or belonged to the 'not determined' user categories (55.6%) (Table 4.43). Workers being the regular, users of the system, it was very likely for them to meet other workers and acquaintances spontaneously. The 'not determined' users, similarly, constituted the majority of people hanging out on the skyways (Table 4.36), and were very likely to meet their peers. The degree of meeting on the other three sites was not considerably different, although, the portion of Nicollet Mall under renovation would probably experience a lot more of that activity type after the completion of the construction work. The furnished portion of the mall, probably due to its remoteness from the core, did not experience this activity type to any large degree. Moreover, the presence of a large plaza on its side (Peavey Plaza), possibly results in scheduled meeting occurring inside the plaza rather than on the streets. People eat to a considerably larger degree on Mall(F) than any other facility. This may simply be due to the larger availability of eating places on the mall. Although the skyway segments have a lot of eating places, the most common users of those establishments were the workers. During good weather even workers could be observed treading down to the mall, for food, instead of the skyway. The virtual absence of eating activity type on Mall(R) may be largely attributed to the lack of outdoor eating facilities (sitting as well) during the renovation process. People watch/groom and talk more on Mall(F) than on the skyway. The poor performance of Mall(R), again, can be explained by the then ongoing renovation work. The comparatively poor performance of the skyway system in the stationary activity sector, may be ascribed to the severe lack of furnishing elements on the system. The skyways did not seem to be an attractive place for people wanting to engage in stationary activities, as compared to the mall. It can be observed from the data that most of the people engaged in 156 the watching/grooming, talking, and eating on the four facilities, were sitting (watching grooming 44.8%, talking 54.7%, and eating 73.5%, ignoring the day of the week; Table 4.44). This finding is also consistent with earlier findings by Whyte in his study of outdoor, clustered public spaces (Whyte, 1980), that the provisions of seats among other furnishings is a major determinant of the success of public spaces. A similar phenomena can be observed in the studing/reading category, where 93.1% of all observed were in the sitting posture. That explains why most of the people engaged in this activity type were observed in Mall(F). Tighter control and regulation, and management policies geared against stationary activities, also act against the use of skyway space for such activities. Lastly, performing as an activity type was purely restricted to Nicollet Mall. The effects of lack of such activities on the skyway have been discussed later. The only major stationary activity more likely to happen on the street was waiting (73.3%). However, this waiting occurred chiefly during the office closing hours, when workers (81.8% of all people observed waiting) (Table 4.45) could be observed waiting for a hike in their partner's/spouse's car. People waiting near bus stops have been totally excluded from the sample. With prohibition of vehicular traffic on the mall, such type of activity was totally absent. Ignoring the waiting category, non of the other stationary activity types were more likely to occur on Marquette Ave, as compared to other facilities, and with severely restricted types of mobile activities occurring on the streets, the areas outside the mall and the core areas represented the duller side of downtown Minneapolis, for most part of the day. During weekends, an increased proportion of visitors and tourists, in groups and with families, on the skyway (Table 4.46 as compared to Table 4.47) led to substantially larger proportion of the users watching/grooming (60.4% as compared to 31.9% during weekdays). Due to the increased traffic of visitors and tourists, a substantially larger 157 proportion of users waiting also occurred (Table 4.48 as compared to Table 4.43). However, this waiting differs from the one occurring on the street since, on the skyways the users.were.mainly composed of family, members, waiting for their relatives. In the rest of the activity types, the trend was similar to that of the weekdays. None of the issues including control & regulation, perceived safety, ease of accessibility, or orientation, seem to have had as much bearing on the degree of stationary activity on each facility as the user facilities and amenities provided by the physical environment. Many other factors not considered in this analysis, probably had some degree of bearing on the stationary activity level. Proximity to the core, degree of retail activities, density of user population, the external weather condition, and the variety of other activities occurring in the overall environment are just a few worth considering. 4.3.3 POSTURE TYPES From the previous section, it is fairly apparent that the degree and type of stationary activity is highly associated with posture, since both the variables are associated with the degree and quality of furnishing offered by each facility's physical environment, and the quality of physical environment as a whole. Consequently, some of the discussions regarding posture types have been covered in the previous section. Statistical analysis of the data on posture types on the facilities showed that a significant (Chi-square sig. 0.0000) difference between all the pairs of facilities existed during both, weekdays and weekend (for reclassified variables to meet Chi-square test assumptions; Table 43 & 44 appendix II). The corresponding lambda measures at 0.26154 for weekdays, and 0.24138 for weekend makes it amply clear that the difference was not only significant but also considerably strong (Table 4.49 & 4.50). The variations became clearer 158 COUNT EXP VAL ROW PCT COL PCT ACTIVITY 1 .00 Wa t c h ing/Groom i n SITE Skyway 1 N i c o l l e t N i c o l l e t M a r q u e t t M a l l R M a l l F e Avenue 00| 2.00| 3.00| 4. 00 | Talk i n g Eat i n g 2.00 3.00 4 .00 S t u d i n g / R e a d i n g 5.00 P e r f o r m i n g Wa i t i n g Meet i n g 6 .00 7 .00 22 17.3 35.5% 31.9% 3 5.0 4.8% 15.0% 28 29 . 4 45.2% 23.9% 9 10.3 14.5% 22.0% 23 31.0 20.7% 33. 3% 13 9.0 11.7% 65.0% 60 52.6 54 1% 51.3% 15 18.4 13.5% 36.6% 5 3.9 35. 7% 7.2% 0 1 . 1 .0% .0% 8 6.6 57 . 1% 6.8% 1 2 . 3 7. 1% 2.4% 6 7.0 24 .0% 8.7% 0 2.0 .0% .0% 16 11.8 64.0% 13.7% 3 4 . 1 12.0% 7.3% 0 1 . 1 .0% .0% 1 . 3 25.0% 5.0% 3 1.9 75.0% 2.6% 0 . 7 .0% .0% 4 4 . 2 26. 7% 5.8% 0 1 .2 .0% .0% 0 7 . 1 .0% .0% 11 2.5 73.3% 26.8% 9 4 . 5 56 . 3% 13.0% 3 1 . 3 18.8% 15.0% 2 7.6 12.5% 1 . 7% 2 2.7 12.5% 4.9% COLUMN 69 TOTAL 27.9% 20 8. 1% 117 47 . 4% 41 16.6% ROW TOTAL 62 25. 1% 1 1 1 44.9% 14 5.7% 25 10. 1% 4 1.6% 15 6. 1% 16 6 . 5% 247 100.0% CHI-SQUARE O.F. 66.61720 18 SIGNIFICANCE 0.0000 MIN E.F. CELLS WITH E.F.< 5 0.324 15 OF 28 ( 53.6%) STATISTIC LAMBDA SYMMETRIC 0.06767 WITH ACTIVITY DEPENDENT 0.00000 WITH SITE DEPENDENT 0. 13846 Table 4.38: Crosstabulation of Activity types by Site, for weekday data. 159 COUNT EXP VAL ROW PCT COL PCT ACTIVITY 1 .00 Watch ing/Groomin SITE Sk yway 1 N i c o l l e t N i c o l l e t Marquett Mal1 R Ma 11 F e Avenue 001 2.00| 3.00| 4.001 TaIk i n g Eat i n g 2.00 3.00 4 .00 S t u d i n g / R e a d i n g 5.00 P e r f o r m i n g Wa i t i n g Meet i n g 6.00 7 .00 COLUMN TOTAL ' 37.9% 32 10 10 2 20.4 15 .0 16 . 2 2 3 59. 3% 18 .5% 18 .5% 3 7% 60.4% 25 .6% 23 .8% 33 3% 14 15 16 3 18.2 13 . 4 14 . 4 2 1 29 . 2% 31 . 3% 33 . 3% 6 3% 26.4% 38 . 5% 38 . 1% 50 0% 3 9 8 0 7.6 5 .6 6 .0 9 15.0% 45 .0% 40 . 0% 0% 5.7% 23 . 1% 19 .0% 0% 0 1 3 0 1.5 1 . 1 1 . 2 2 .0% 25 .0% 75 .0% 0% .0% 2 .6% 7 . 1% 0% 0 1 1 0 .8 .6 .6 1 .0% 50 .0% 50 .0% 0% .0% 2 .6% 2 .4% 0% 4 3 1 1 3.4 2 .5 2 . 7 4 44 . 4% 33 . 3% 1 1 . 1% 11 1% 7.5% 7 . 7% 2 . 4% 16 7% 0 0 3 0 1 . 1 .8 .9 1 .0% .0% 100 .0% 0% .0% .0% 7 . 1% 0% 53 39 42 6 27 .9% 30.0% 4 . 3% ROW TOTAL 54 38.6% 48 34 . 3% 20 14.3% 4 2.9% 2 1.4% 9 .4% 3 . 1% 140 100.0% CHI - SQUARE 33.90711 D F . 18 SIGNIFICANCE 0.0129 MIN E.F. CELLS WITH E.F.< 5 0.086 19 OF 28 ( 67.9%) STATISTIC LAMBDA SYMMETRIC 0. 15607 WITH ACTIVITY DEPENDENT 0.13953 WITH SITE DEPENDENT 0. 17241 Table 4.39: Crosstabulation of Activity types by Site, for weekend data. PAIRS OF SITE CHI-SQ WEEKDAY SIG LAMBDA CHI-SQ WEEKEND SIG LAMBDA SKYWAY VS MALL(R) : ... 7.8783 0.0486 0.0000 12.0399 0.0024 0.20513 SKYWAY VS MALL(F) : 13.61683 0.0035 0.10145 14.0941 0.0009 0.26190 SKYWAY VS MARQ.AVE.: 4.7723 0.1892 0.0000 (NOTE : The underlined significance values in the table demonstrates evidence of significant difference at 95% confidence level; No recategorization was possible for the pairwise crosstabulation of Skyway System vs Marquette Avenue, during weekend,in order to satisfy the Chi-square test assumptions) Table 4.40: Pairwise comparison of facilities for the proportional distribution of Activity types. Data Source: Appendix II; Tables 38 to 42. - - - - KRUSKAL- WALL IS 1-WAY AN0VA SITE F a c i l i t y Observed BY ACTIVITY MEAN RANK CASES 116.06 62 ACTIVITY = 1 Watch ing/Grooming 127.24 1 1 1 ACTIVITY = 2 Ta lk ing 113.29 14 ACTIVITY = 3 Eat ing 130.36 25 ACTIVITY = 4 S t u d i n g / R e a d i n g 130.88 4 ACTIVITY = 5 P e r f o r m i n g 175.80 15 ACTIVITY = 6 Wa i t ing 81 .47 16 ACTIVITY = 7 Meet ing 247 TOTAL CORRECTED FOR TIES CASES CHI-SQUARE SIGNIFICANCE CHI - SQUARE SIGNIFICANCE 247 15.0989 0.0195 17.4186 0.0079 Table 4.41: Mean ranking of Activity types on the facilities, during weekdays. KRUSKAL- WALL IS 1 - WAY ANOVA SITE F a c i l i t y Observed BY ACTIVITY MEAN RANK CASES 55 6 3 54 ACTIVITY-= 1 Watch ing/Grooming 77 1 1 48 ACTIVITY = 2 Tal k i n g 82 30 20 ACTIVITY = 3 E a t i n g 103 38 4 ACTIVITY = 4 S t u d i n g / R e a d i n g 93 25 2 ACTIVITY = 5 Per form i n g 64 22 9 ACTIVITY = 6 Wa i t i n g 1 13 50 3 ACTIVITY = 7 Meet i n g 140 TOTAL CASES 140 CHI-SQUARE SIGNIFICANCE 17.0734 0.0090 CORRECTED FOR TIES CHI-SQUARE SIGNIFICANCE 19.0318 0.0041 Table 4.42: Mean ranking of Activity types on the facilities, during weekend. ACTIVITY 1 .00 Watch ing/Groom i n ROLE V i s i t o r T o u r i s t S e c u r i t y Other Wo Not Dete Guard r k e r r m ined 1.00| 2.00| 3.00| 4.00| 5.00| COUNT ROW PCT COL PCT Talk i n g Eat i n g 2.00 3.00 4 .00 St uding/Read i n g Wa i t i n g Meet i n g 6 .00 7 .00 COLUMN TOTAL 15 r -2 r ¥ -3 2 68 .2 9 1 13 .6 9 1 41 . 7 28 6 20 .0 22 2 16 2 2 3 69 .6 8 7 8 7 13 .0 44 .4 28 6 100 0 20 .0 - + - - + - + - - + -1 3 1 20 .0 60 0 20 0 2 .8 42 9 1 1 1 - + - . + . - + - - + -3 3 50 .0 50 .0 8 . 3 20 .0 - + - . + _ - + -1 2 1 25 .0 50 .0 25 0 2 . 8 13 . 3 1 1 1 . . . . - + - . + - + - - + -4 5 44 . 4 55 6 26 . 7 55 6 36 7 2 15 9 52 . 2 10 1 2 9 21 . 7 13 0 ROW TOTAL 22 31.9 23 33.3 5 7.2 6 8.7 4 5.8 9 13.0 69 100.0 Table 4.43: Crosstabulation of Activity by Role for the Skyway System, for weekday data. ACTIVITY 1 .00 Watch ing/Groom i n POSTURE COUNT ROW PCT S i t t i n g S t a n d i n g L y i n g L e a n i n g K n e e l i n g COL PCT / S q u a t t i 1.00| 2.00| 3.00| 4.00| 5.00| Ta1k ing Eat i n g 2.00 3.00 4 .00 S t u d i n g / R e a d i n g 5 .00 P e r f o r m i n g Wa i t i n g Meet i n g 6 .00 7 .00 COLUMN TOTAL 52 • 44 . 8 26.0 42 36 . 2 28.4 3 2.6 42.9 17 14.7 58.6 2 1 . 7 66. 7 87 54 . 7 43.5 63 39.6 42.6 3 1 .9 42.9 6 3.8 20.7 25 73.5 12.5 9 26.5 6. 1 27 93. 1 13.5 1 3.4 . 7 1 3.4 3.4 1 16 . 7 .5 4 66 . 7 2.7 1 16.7 33. 3 6 25.0 3.0 15 62.5 10. 1 3 12.5 10.3 2 10.5 1 .0 14 73.7 9.5 1 5.3 14 . 3 2 10.5 6.9 200 51.7 148 38 . 2 7 1 . 8 29 7 . 5 ROW TOTAL 1 16 30.0 159 41.1 34 8.8 29 7.5 6 1.6 24 6.2 19 4.9 387 100 .0 Table 4.44: Crosstabulation of Activity by Posture on the facilities, ignoring day. 163 ACTIVITY ROLE V i s i t o r T o u r i s t Other Wo Not Dete r k e r rmined 1.00| 2.00 | 4.001 5.00| COUNT ROW PCT COL PCT 1 .00 Watching/Groomin Talk i n g Eat ing 2.00 3.00 4 .00 S t u d i n g / R e a d i n g Wa i t i n g Mee t i ng 6 .00 7 .00 COLUMN TOTAL 1 11.1 16.7 2 13.3 33. 3 1 100.0 16.7 1 9 . 1 16 . 7 1 50.0 16 . 7 6 14.6 1 33 . 3 50.0 1 9. 1 50.0 2 4.9 6 66 . 7 20.0 13 86. 7 43. 3 33. 3 3.3 9 81.8 30.0 1 50.0 3.3 30 73.2 2 22.2 66 . 7 33. 3 33. 3 ROW TOTAL 9 22.0 15 36.6 1 2.4 3 7 . 3 1 1 26 .8 2 4.9 3 41 7.3 100.0 Table 4.45: Crosstabulation of Activity by Role for Marquette Avenue, for weekday data. 164 COUNT ROW PCT COL PCT ROLE V i s i t o r Tour i s t 1 .00 2.00 GROUPING Alone Two, Sam Two, Man 3 or Mor ROW e Sex & Woman e A d u l t s TOTAL 1.00| 2.00| 3.00| 4.00| 3.00 Secur i t y Guard 4 .00 Other Worker 5.00 Not Determined COLUMN TOTAL 17 47 . 2 54. B 1 14.3 3.2 50.0 3.2 8 53. 3 25.8 4 44 . 4 12.9 31 44 .9 3 8.3 50.0 2 28.6 33. 3 1 6.7 16.7 12 33. 3 63.2 4 57 . 1 21.1 1 50.0 5.3 2 13.3 10.5 4 11.1 30.8 4 26.7 30.8 6 8.7 19 27 .5 5 55.6 38.5 13 18.8 36 52.2 7 10. 1 2 2.9 15 21.7 13 69 100.n Table 4.46: Crosstabulation of Role by Grouping for the Skyway System, for weekday data. ROLE COUNT ROW PCT COL PCT 1 .00 V i s i t o r Tour i s t 2.00 3.00 S e c u r i t y Guard Not GROUPING Alone Two, Sam Two, Man 3 or Mor F a m i l y , e Sex & Woman e A d u l t s C h i l d r e n 1.001 2.00| 3.001 4.001 5.00| 5.00 De term i n e d COLUMN TOTAL 12 29 . 3 66 . 7 1 33.3 5.6 5 100 .0 27 .8 18 34 .0 4 9.8 50.0 2 50.0 25.0 2 66 . 7 25.0 8 15.1 10 24.4 83. 3 12 29 . 3 100.0 2 50.0 16 . 7 12 22.6 12 22.6 3 7 . 3 100.0 ROW TOTAL 41 77 . 4 4 7.5 3 5.7 5 9 . 4 3 53 5.7 100.0 Table 4.47: Crosstabulation of Role by Grouping for Skyway, for weekend data. 165 from the pair-wise comparison of sites (Table 4.51). Very similar in pattern to the variations in stationary activities, the skyway and the Marquette Avenue demonstrated insufficient evidence of any significant difference. Once again, Mall(R)and Mall(F) proved to be significantly dissimilar from the skyway segments, in the proportional distribution of posture types. Furthermore, the maximum variability was demonstrated in the comparison between skyway and Mall(F), which was not unexpected considering the marked difference in facility furnishing on the two systems. COUNT ROW PCT COL PCT ACTIVITY 1 .00 Watch ing/Groomin ROLE V i s i t o r T o u r i s t S e c u r i t y Not Dete ROW Guard rmined TOTAL 1.00| 2.00| 3.00| 5.00| 22 68.8 53. 7 2 6.3 50 .0 3 9.4 100.0 5 15.6 100.0 Ta 1 k i n g Eat i n g Wa i t i n g 2.00 3.00 6 .00 COLUMN TOTAL 12 85. 7 29 . 3 2 14.3 50.0 3 100.0 7 . 3 4 100.0 9.8 41 77.4 4 7.5 3 5.7 5 9.4 32 60.4 14 26.4 3 5.7 4 7 . 5 53 100.0 Table 4.48: Crosstabulation of Activity by Role on Skyway, during weekend. An isolated comparison of each posture type was in confirmation with the above finding (Appendix III, Fig. 13 & 14). Most of the people engaged in leaning posture types were observed on the skywalks (80% of all users observed). On the skyway network they were more likely to be found near the skywalk segments overlooking atria spaces. Leaning, however, on atrium skyway segments had got no relations with the shortage of seating spaces on skyway corridors. Users could be observed leaning on skyway parapets watching the activities occurring in the atria spaces below. Such segments offered 166 marvellous viewing points for the users, and many users made use of it. But leaning, by itself, did not constitute any major posture type on the skyway. Standing was the most common posture observed .on the skyway segments, which can be attributed to the shortage of furnishing elements on the facility. As can be observed from the data, only 30.4% of all the users engaged in stationary activities on the skyway were sitting, as opposed to 84.6% on Mall(F). In contrast with the skyway (58%), only 12% of the users were observed standing in the furnished portion of Nicollet Mall. The large proportion of users observed standing on Mall(R) (90% of all posture types), can be explained by the then ongoing construction work on the mall which left no area for the users to sit or lean on. Users observed sitting on Marquette Avenue, which constituted a large proportion, deserves further analysis. A majority of the people sitting on Marquette Ave. belonged to the workers category, sitting outside on the window sills of the adjacent buildings, smoking and conversing with co-workers. Table 4.52 shows the proportional distribution. Workers constituted 72.2% of all the people sitting on Marquette Ave., and 84.6% of them were observed engaged in talking (while smoking). The smokers were a continuous presence on the sidewalks outside office entrances, and the window sills provided comfortable seating to large number of such people, outside the mall. Standing, on the whole, was the second most frequently observed posture types, after sitting. People standing (58% of all observed), were more likely to be observed on the skyway due to the low level of furnishing, than on Mall(F) or Marquette Avenue. Very few people lied down. And most of them could be observed in the furnished part of Nicollet Mall, again explained by the higher degree of furnishing. People observed sitting, hence, were also more likely to be on Mall(F) than on any other facility (71.7% of all people observed sitting). 167 COUNT EXP VAL ROW PCT COL PCT POSTURE S i t t i n g Stand i n g Ly i n g Lean i n g 1 .00 2.00 3.00 4 .00 COLUMN TOTAL SITE Sk yway 1 .00 I N i c o l l e t N i c o l l e t Marquett ROW M a l l R M a l l F e Avenue TOTAL 2.00| 3.00| 4.00| 21 38.6 15.2% 30.4% 40 26 .0 43.0% 58.0% 0 1 . 7 .0% .0% 8 2.8 80.0% 11.6% 1 1 69 27 .9% 0 99 18 . 2 65.4 22.9 .0% 71.7% 13.0% .0% 84 .6% 43.9% 18 14 21 .5 44 . 1 15.4 .4% 15. 1% 22.6% .0% 12.0% 51.2% 2 4 0 .5 2.8 1.0 . 3% 66 . 7% .0% .0% 3.4% .0% 0 0 2 .8 4.7 1 . 7 .0% .0% 20.0% .0% .0% 4.9% 20 117 41 . 1% 47 .4% 16.6% 138 55.9% 93 37 . 7% 6 . 4% 10 4.0% 247 100.0% CHI-SQUARE D.F. 105.33006 STATISTIC SIGNIFICANCE 0.0000 MIN E.F. CELLS WITH E.F.< 5 0.486 8 OF 16 ( 50.0%) SYMMETRIC WITH POSTURE DEPENDENT WITH SITE DEPENDENT LAMBDA 0. 30962 0. 36697 0.26154 Table 4.49: Crosstabulation of Posture types by Site, for weekday data. 168 COUNT EXP VAL ROW PCT COL PCT SITE Sk yway N i c o l l e t N i c o l l e t Marquett POSTURE S i 11 i n g S t a n d i n g Ly i n g Lean i ng 1 .00 2 .00 3.00 4.00 5.00 Knee 1 i ng/Squat t i COLUMN TOTAL CHI-SQUARE D.F. 47.30272 12 STATISTIC Mal 1 R Mal 1 F e Avenue 1 .00 2.00 3.00 4 .00 21 9 32 0 23.5 17.3 18 .6 2.7 33.9% 14.5% 51 . 6% .0% 39.6% 23. 1% 76 . 2% .0% 17 25 9 4 20.8 15.3 16 . 5 2.4 30.9% 45.5% 16 .4% 7 . 3% 32 . 1% 64 . 1% 21 .4% 66.7% 1 0 0 0 . 4 . 3 . 3 .0 100.0% .0% .0% .0% 1 .9% .0% .0% .0% 14 3 0 2 7.2 5.3 5 . 7 .8 73.7% 15.8% .0% 10.5% 26.4% 7.7% .0% 33.3% 0 2 1 0 1 . 1 .8 .9 . 1 .0% 66.7% 33 .3% .0% .0% 5. 1% 2 . 4% .0% 53 39 42 6 37.9% 27.9% 30 .0% 4 . 3% •LAMBDA SIGNIFICANCE MIN E.F 0.0000 SYMMETRIC 0.24848 0.043 ROW TOTAL 62 44 . 3% 55 39. 3% 1 .7% 19 13.6% 3 2.1% 140 100.0% CELLS WITH E.F.< 5 11 OF 20 ( 55.0%) WITH POSTURE DEPENDENT 0.25641 WITH SITE DEPENDENT 0.24138 Table 4.50: Crosstabulation of Posture types by Site, for weekend data. PAIRS OF SITE CHI-SQ WEEKDAY SIG LAMBDA CHI-SQ WEEKEND SIG LAMBDA SKYWAY VS MALL(R) : 5.66755 0.0173 0.0000 9.41132 0.0090 0.20513 SKYWAY VS MALL(F) : 55.88664 0.0000 0.43478 15.93452 0.0003 0.26190 SKYWAY VS MARQ.AVE.: 2.803 0.2462 0.0000 1.50665 0.2197 0.0000 (NOTE : The underlined s i g n i f i c a n c e values i n the table demonstrates evidence of s i g n i f i c a n t d i f f e r e n c e at 95% confidence l e v e l ) . Table 4.51: Pairwise comparison of facilities for the proportional distribution of Posture types. Data Source: Appendix II; Tables 45 to 50. COUNT ROW PCT COL PCT ROLE V i s i t o r Touris t Other Wo Not Dete ROW rker rmined TOTAL 1.00| 2.00| 4.00| 5.00| ACTIVITY 1 .00 Watch ing/Groom in T a 1k i ng Eat ing 2 .00 3.00 100.0 100.0 4 .00 Studing/Reading COLUMN TOTAL 1 5.6 50.0 100 .0 1 5.6 2 50.0 15.4 1 1 100.0 84.6 13 72.2 2 50.0 66 . 7 50.0 33 . 3 3 16 . 7 4 22. 2 1 1 61.1 1 5.6 2 11.1 18 100.0 Table 4.52: Crosstabulation of Activity types by Role for Marquette Avenue, for weekday data, for all people observed sitting. 170 The weekdays and weekend observations were not significantly different. A few people were observed to be leaning/squatting, and all were invariably on Nicollet Mall. 4.4.0 Differing Usage Different spatial configurations, physical settings, urban policies, and management strategies lead to variations in the usage pattern in the four facilities. However, one facility, the Marquette Avenue, proved to be significantly different from the Skyway System in almost all the areas of comparison. In contrast to the differences between the Skyway System and Marquette Avenue, Mall(R) and the skyways showed little evidence of dissimilarity in the proportional distribution of user age, gender, role, and grouping during, both weekdays and weekend. However, an area deserving attention is the pair-wise comparison of the weekend data of Skyway System and Marquette Avenue. In spite of the wide variations established between the systems based on the weekday data, the weekend analysis did not provide, in the comparisons in age, gender, and grouping, enough evidence for any significant differences. Exact cause of such a marked deviation in skyway performance in unknown, but the partial closure of some of the skyway segments and retail establishments during weekend may partly explain the phenomena. Other possibilities include the reduction in automobile traffic on city street, which attracts more visitors and tourists to the areas outside the core, and thus in a way reducing the variability in usage pattern between the street and the skyway. In spite of the lack of evidence for dissimilarity, one major difference remains. The volume of users on the street during weekend was considerablj' less than on the Skyway System (this aspect has not been considered in this survey), which is amply evident from the number of users who fell within the survey sample (57 on Marquette Avenue as against 175 on the Skyway 171 System). If the volume of users on the facilities would be considered as an indicator of system performance, the street would prove to be far inferior as an urban public space than the skyway corridors. Thus where as Mall(R) showed little evidence of dissimilaritj'. from the skyway in facility usage, Marquette Avenue was dissimilar either in proportional distribution or absolute degree of usage in all four variables. In the pair wise comparison of Purpose, all pairs provided sufficient evidence of difference. But, here again, Marquette Avenue had larger dissimilarity, as shown by the degree of association, from the Skyway System, as compared to both the portions on Nicollet Mall. The onry variables, where evidence of dissimilarity between the skyway segments and the street was not available were Stationary Activity, and Posture. The lack of dissimilarity was once again nullified by the almost negligible degree of stationary activities occurring on the streets, as compared to the skyways or the mall, in absolute degree of usage. In fact, as compared to the skyway segments and the mall, the street was host to a more restrictive user type and usage category. The street, which has been a part of the urban setting since the beginning of urban settlements, did not prove to be the preferential location for any activity other than access, for the pedestrians. Most of the users on the streets were downtown workers, and during weekends the streets outside core area remained deserted. Particularly important is the fact that Marquette Ave. too, has a variety of retail establishments on either side of the street, which makes it comparable to certain degree with Nicollet Mall, as far as retail facilities are concerned. Survey points on Marquette Avenue were also not remotely located. Shoppers, however, were not more likely to be found on Marquette Avenue, as compared to the Skyway or Mall(R). The street lacked considerably in stationary activities too. Clearly, Mall(R) and the Skyway had fewer dissimilarities, and served a wider variety of users, and usage, as compared to the street outside the core. It would not be wrong to assume, that for the people of Minneapolis, the core area of the central city constitutes the downtown, leaving the rest of the downtown streets for use as access routes by the workers. 172 Unlike Mall(R), Mall(F) also varied considerably from the Skyway System in the facility usage pattern. However, the phenomena may be largely attributed to the extremely-remote location of the survey .points as compared to the survey.locations on the other three facilities. Completion of the renovation work should see an improvement in the performance of Mall(F). Nevertheless this portion of Nicollet Mall demonstrated some remarkable performance in the areas of stationary activity and posture, unparalleled by the other facilities. Its performance as a public space is beyond any doubts, but it serves a different range of functions as compared to the indoor skyway segments. Interestingly enough, although stationary activities on the skyway segments were limited to a few categories, the degree of activity was comparable to both the portions of Nicollet Mall. 27.9% and 37.9% of all people observed engaged in stationary activities in downtown Minneapolis , during weekdays and weekend respectively, were seen on the Skyway System (Table 4.38 & 4.39). In terms of degree of activity, it was a comparatively remarkable performance. It highlights the necessity of effecting certain amount of change to the facility management strategy by allowing some category of stationary activity to occur on the network, and to provide facilities and amenities for such kind of usage. It also demonstrates the potential of the Skyway System in catering to the social needs of people in urban public spaces, which has not yet been realized. More interesting is the fact that the Skyway and the portion of Nicollet Mall under renovation, demonstrated the least variability as urban public spaces. The then ongoing renovation work on Mall(R) should have theoretically altered, more specifically reduced the facility's performance as a public space. The good performance of the facility, in spite of the construction work deserves further research efforts. The skyway and Mall(R) demonstrated little dissimilarity in user population, and had similar usage pattern. The major area where the skyway system differed significantly from the mall, and to a large degree, was in the variety of stationary activities, and posture types. But the lack in 173 variety was more or less compensated by the high degree of stationary activities observed on the system corridors. In most of the other aspects, the Skyway System experienced a usage, pattern very similar to-the Nicollet. Mall,-either, in .proportional distribution, or. in degree of usage, which has been hailed by academicians as a success in the designing of urban public spaces (chapter two). They together served a wide variety of users, engaged in a wider variety of activity types, and represented a considerable improvement in design of public spaces from the conventional city street and sidewalk. The differences between the two portions of the mall (under renovation, and furnished), needs a few more variables to be explained. Chief among those, as explained earlier, includes, proximity to the core, degree of retail facility, and density of user population, and the then ongoing renovation work. Any variability in the proportional distribution of disabled users on the four facilities could not be analyzed due to the negligible number of disabled users who fell within the sampled population (appendix II, Table 51 & 52). The possibility of very few disabled people visiting the core area is worth exploring, moreover since, of all the people surveyed only 0.615% constituted the disabled users. Such a low figure may hint at possible deficiencies in core area accessibility rather than system accessibility. But not much can be concluded from the collected data. Among the issues raised by authors from time to time, the perceived safety level seem to have had a predictable association with the users and usage pattern. Ease of accessibility to the system did not prove to be a determining factor, though, ease of accessibility to downtown from the suburban areas might be having some association on user's trip making decisions, specifically since to a large number of users, availability and affordability of parking spaces downtown did matter (gathered from personal interviews). Control and regulation had an association only in assisting in increasing the perceived 174 safet3' level. And orientation did not prove to be any major problem for the system users. Control and regulation from the viewpoint of private ownership of public places did not seem to have-had any ..impact on. facility .performance .of the Skyway. System. Although privately controlled, the presence of commercially unattractive users including retired elderly and teenagers hanging out on the corridors, the almost uniform distribution of 'not determined' users on all facilities, and the absence of any dissimilarity in the distribution of ethnic minorities does not support the contention that skyways are less public than the mall or the sidewalks. Moreover, users were using the skyways less for access and more for visiting, shopping, or promenading, as compared to the other facilities. That speaks in favour of the performance of the Skyway System as a public place in comparison to the mall. The only issue (beside perceived safety level), which seems to have had a considerable degree of association with the users and usage pattern was the quality of the physical environment, and the facility and amenity offered by the physical setting of each facility. The impact of furnishing on user activities have already been partly probed in section 4.3.0. A few areas have not been considered. The next chapter probes those areas of the physical environment of the facilities, and endeavors to explain some of the phenomena observed in the usage pattern in this chapter. 175 C H A P T E R F I V E CITY C O R E A N D T H E P E R C E P T U A L S C E N A R I O 5.1.0 Starting Out The Skyway System and Mall(R) demonstrated striking similarity in user population, and lesser dissimilarity in usage pattern as compared to Marquette Avenue. The analysis in Chapter Four proved this fact. However, many other aspects have not been accorded due attention. This particularly draws importance from the fact that physically and spatially the two facilities, with similar usage pattern, have vastly different physical design attributes, and that raises the first array of questions in the areas of potential interaction between the physical environment, the usage pattern, and the user's choice of facility. The second area of importance originates from factors, external to the collected data from the survey conducted in Downtown Minneapolis. This area addresses several possible limitations to the present findings, considering, among other areas, the vast differences in the physical features of individual Skyway segments. Many other areas of interest befit consideration. Primary among those, is the role of these urban landforms (Skyway and the Mall), which transcends the boundaries of the human physical world. Such considerations raise different questions, hitherto not considered. The most suitable example, to illustrate the above point, could be the meaning of the Skyway System to the regular elderly visitors (gathered through personal interviews). For the elderly, going downtown is more out of long term habit than out of interest or attraction to the downtown establishments or even the newly incorporated improvements to the physical landscape. 176 Most of the users in this age group clearly remember the days when city center vitality was at its peak, and the images of a healthy, vibrant, lively downtown are strongly engrained in their .cognitive, mental maps. They feel- the gradual deterioration of the downtown environment during the 50's and 60's as unfortunate, at the least. They despise the indoor, suburban shopping/business establishments, and they despise the newly created indoor environment in the downtown to the same degree. But they still continue to visit the downtown, in spite of severe transportation related problems. Their major concern arises from the perceptual vision of a dormant downtown where festivities are a forgotten phenomena, and any physical improvement disregarding this activity area proves meaningless to them. Their daily appearance was a testimony to their strong desire to bring back the vitality. In contrast to such indifference from regular users, many rare visitors of the facilities expressed pride in the high-tech development. Thus, public perception of the environment could widely differ, ranging form a positive, optimistic attitudes, to total disregard. The study warrants a broader understanding of the perceptions of the users, their needs, and the role the physical design play in the overall context of modification of the lifestyle of the urban inhabitants. In this chapter an attempt will be made to address most of the above mentioned areas of interest. However, the subsequent sections would include two primary emphasis areas. The first will be directed towards investigating any theoretical relationship, wherever possible, between the physical character of the environment and the facility usage pattern. It would include a comparison of the physical environment of the Skyway and the Mall, and to search for possible areas where some aspects of the usage phenomenon described in Chapter Four could be explained. The second emphasis would concentrate on studying possible shortcomings of the conclusions drawn from the collected data, including the problems involved with generalizing the survey findings (the explanations included in this chapter regarding the physical environment is based purely on phenomenological observations by the author, and does not enjoy a statistical foundation). 177 5.2.0 The Urban Forms and Social Activities 5.2.1 AREAS OVERLOOKED The findings of the present study, that the Skyway and the Mall experienced similar usage pattern, may encounter certain difficulties in universal applicability. The difficulty originates from two areas of potential impact on the system's general usage pattern, which has been overlooked in the data analysis, due to the limited scope of the research work. The first major significant difference could arise out of the limited range of Skyway segment types which were surveyed in this study. The physical design of the various segments in the network vary considerably. As an initial step towards classification of the segments on the basis of physical design, the entire system can be broadly categorized into nine different (not mutually exclusive) types : (1) Segments within the primary (major) network. (2) Segments within the secondary (feeder) network. (3) Segments along/overlooking atrium spaces (Fig.5.1). (4) Segments (connecting corridors) deep inside buildings (Fig.5.2). (5) Segments at the periphery of buildings (Fig.5.3). (6) Free standing segments supported independently (Fig.5.4). (7) Segments with retailing/businesses on one or both edges (Fig.5.5). (8) Segments without any retail/business establishments (Fig.5.2). (9) Segments through major departmental stores (Fig.5.6). The surveyed segments represent only a section of the above categories, and the inclusion of all the different types of segments in the survey could have resulted in some variations in the collected data, since the segments surveyed represent some of the better used 178 segments in the network. Another important aspect with potential impacts is the relative scale of the two systems surveyed. The Skyway System covers a wide area extending over a number of blocks in the Central Business District (32 blocks and connecting almost all downtown activity types; Downtown Council of Minneapolis, 1988), unlike the Nicollet Mall which is restricted to a solitary street (8 blocks long; Shostack, 1978) in downtown Minneapolis. Such a difference in scale exposes the Skyway System to more number and variety of users as compared to the mall. Any comparison of the systems ignoring the above two points would introduce certain degree of error to the assessment process. During the process of comparative analysis of the physical environments of the two systems, in the subsequent sections, the possible impacts of these two factors on the conclusions drawn in Chapter Four, will be discussed. Moreover, all references to the Skyway System in this chapter would mean the entire system and not only the segments where survey points were located, unless specific segment names are stated. Fig.5.1: Skyway segment inside the IDS Center atrium lobby. Fig.5.4: Free standing skyway corridors in downtown Minneapolis. Fig.5.5: Skyway segments with retail establishments defining the edge 181 Fig.5.7: Skyway corridor inside the North Star Building; photograph taken during weekend. 182 5.2.2 CONFIGURATION OF SPACES In spite of the differences in physical design, all the Skyway segments had one striking similarity; all the segments have been designed as pedestrian access corridors. Differences between the official development standards of the skyway network and the grade level sidewalks are negligible (Downtown Council of Minneapolis, 1988; Minneapolis Skyway System Design Standards); both the systems have been designed and managed as continuous systems, free of impediments. Such a developmental strategy has manifested itself in a distinct directional rendering of the skyway spaces. As a result, where as the Mall itself serves as a place to visit, the skyways, even with the potential of being the same, resemble the sidewalks at the grade level in its usage, where people use the corridors and passages mainly to traverse from one point to another. The purely linear form of the skyways restrict any form of stationary or quasi-stationary activities. This has already been demonstrated in sections 4.3.2 and 4.3.3, where the system was shown to lack in stationary activities. The similarity of the skyways in other aspects of usage pattern, with Nicollet Mall, established in the data analysis, could be misleading since all of the segments observed were adjacent to the Mall in the core area, many connecting major retail anchors, and most of the segments overlooking atrium spaces, the positive impacts of which would be explained later in this chapter. At the least, the atria spaces, wherever available, catered to a large variety of user needs including resting, eating or smoking, reading and watching, unavailable on other segments, and attracting in the process a large number of users. The attractiveness of the atria segments to stationary activity types can be demonstrated by the comparison of the degree of stationary activities on the North Star segment (Fig.5.7) with those of the other two. The primary difference between the North Star segment and the other two segments surveyed, namely the IDS Center (Fig.5.8) and City Center skyway (Fig.5.9), lies in the spatial configuration. The other two overlook atria spaces, 183 184 have visible vertical link (escalators) to the lower level, and, with large glazed areas in the atria roofs and lateral building street interface, are relatively more transparent to the external environment. In contrast, the North Star segment is essentially a corridor with retail establishments lining one or both sides, little transparency, and no visible vertical connection. The total number of users on the North Star segment observed to be engaged in stationary activities (observations recorded between 8:00 a.m. and 12:00 noon), during weekdays, constituted only 17.4% of all users, engaged in stationary activities, surveyed on the entire Skyway System (Table 5.1, Fig.5.10, Fig.5.11) as compared to 63.8% observed on the City Center segment. The IDS segment, however, even with an atria space, did not play host to any considerable proportion of users. This may be explained by the acute lack of furnishing elements on the IDS segment, and also by the difference in the time period of observation. Nevertheless, the presence of the atria space partly explains the difference. Most of the surveyed segments represented some of the best used in the entire system, and as compared to the surveyed portions, other segments on the system had spatial configurations with relatively lesser positive impacts on system usage. Barring the few segments along atria spaces, the skyway bridges, and the portion of the system along building periphery, major portions of other routes are positioned deep inside buildings, where dependence on signs and orientation maps for navigation becomes a necessity (Fig.5.7 as compared to Fig.5.12). A survey of those segments could have provided a very different result from the one obtained in the present study, especially in the number of visitors and elderly on the system, and the number of users disoriented (searching). The most vulnerable sections of the corridors were the areas where two or more skyway corridors met. Arbitrary changes in the directions (angles) of these inside corridors to suit the building's structural needs further aggravated the problem of disorientation at intersections. The dilemma, on the other end, was the problem of monotony of straight long corridors. The most used and the most lively segments in the System formed a part of a closed network along and around 185, COUNT ROW PCT COL PCT TIME 1 .00 8A.M. t o 9 A.M. 9 to 10 10 to 11 2 .00 3.00 4 .00 11 t o 12 Noon 5.00 Noon to 1 P.M. 1 to 2 2 to 3 3 t o 4 4 t o 5 5 t o 6 ACTIVITY Watching T a l k i n g E a t i n g S t u d i n g / W a i t i n g M e e t i n g ROW /Groomin Reading TOTAL 1.00| 2.00| 3.00| 4.00| 6.00| 7.00| 6.00 7 .00 8.00 9.00 10.00 COLUMN TOTAL 1 50.0 4 . 5 2 100.0 9. 1 2 33. 3 9 . 1 3 21.4 13.6 1 12.5 4.5 4 40.0 18.2 1 8 . 3 i . 5 4 57 . 1 18.2 1 50.0 16.7 2 100 .0 8.7 2 33 3 8 . 7 1 16.7 20.0 5 35.7 21.7 2 14.3 33. 3 4 50.0 17.4 2 25.0 33.3 4 40.0 17.4 2 20.0 40.0 2 16 . 7 8.7 1 8.3 20.0 1 8.3 16. 7 2 28.6 8.7 1 14.3 20.0 1 16 . 7 25.0 1 12.5 25.0 2 16 . 7 50.0 4 28.6 44 . 4 5 4 1.7 55.6 2 2.9 2 2.9 2 2.9 6 8.7 14 20. 3 8 11.6 10 14.5 12 17.4 7 10. 1 4 2 6 66. 7 33. 3 8.7 18.2 8.7 22 23 5 6 4 9 69 31.9 33. 3 7 . 2 8.7 5.8 13.0 100.0 Table 5.1: Temporal distribution of users by activity on the Skyway System, during weekdays. 186 STATIONARY ACTIVITIES MINNEAPOLIS SKYWAY SYSTEM TOTAL NUMBER Of USERS OBSERVED 1 1 North 8Ur City Center 108 8KYVMY 8E0MENT3 E 2 Talk E H Eat ESS toad LU3 w*it CD M M ESS TOTAL Fig.5.10: Comparison between skyway segments for the degree of Stationary Activity types. STATIONARY ACTIVITIES MINNEAPOLIS SKYWAY SYSTEM City Center 84% DISTRIBUTION ON SEGMENT TYPES Fig.5.11: Proportional distribution of Stationary Activities on the three Skyway Segments surveyed (percentage of all users observed). 187 the Nicollet Mall, inside the core area, especially those along atrium spaces, and with business/retail establishments defining the edges. Segments connecting and within the major anchor stores including the Daytons', Carson Pirie Scott, and others made up livelier spaces than the rest. In contrast, corridor spaces inside buildings without retail establishments or atria spaces, even inside the closed network, were primarily used for access, and devoid of any other form of activity. When such corridors fell outside any closed network (connected to the network at only one end), the decrease in proportional representation of users other than downtown workers could be easily noticed. The best example to illustrate the above point was the usage pattern on the skyway inside North Star Building. The skyway and Mall(R) showed no statistically significant difference in the proportional representation of workers (Table 4.16). With 5.5% difference in the proportions, and a 95% confidence interval ranging from -2.45% to +13.45%, there was not enough evidence to support a true difference in the population, in general. However, 70.0% of all users surveyed on the North Star skyway, during weekdays, between 8:00 A . M . and Noon when observations were being recorded on the North Star corridor, were workers. In contrast, 56.0% of all users surveyed on the Mall(R) during the same time period were workers; a difference of 14.0% (Table 5.2, 5.3). A 95% confidence interval for this difference in proportional representation ranges from +0.77% to +27.23%, suggesting enough evidence to believe that a true difference existed. This considerable overrepresentation of workers on the North Star corridor can be seen in Fig.5.13, between 8:00 A . M . and Noon, as compared to the same period in Fig.5.14. Workers on the North Star segment were overrepresented even in comparison to the other two segments; the City Center and the IDS Center. In contrast to the North Star Building, workers represented only 25% and 47.25% of all users, in City Center and IDS routes respectively (Table 5.2, Fig.5.15). Evidences supporting the difference is sufficient with the corresponding 95% confidence intervals being +32.64% to +57.36%, and +9.47% to + 36.03%. The proximity of the North Star segment to the mall and the core, hints at 188 TIME 1 .00 8A.M. to 9 A.M. ROLE COUNT ROW PCT ^ V i s i t o r T o u r i s t S e c u r i t y Other Wo Not Dete Guard r k e r r m ined 1.00| 2.00| 3.00| 4 00| 5.00| 9 t o 10 10 t o 11 2.00 3.00 4 .00 11 to 12 Noon 5.00 Noon t o 1 P.M. 1 t o 2 2 t o 3 3 t o 4 4 t o 5 5 t o 6 6 t o 7 7 t o 8 6 .00 7 .00 8.00 9 .00 10.00 1 1 .00 12.00 COLUMN TOTAL ROW TOTAL 4 16.0 I 8 2 I .0 I 19 76 .0 1 25 8 . 4 4 16.0 8 2 1 .0 | 8 2 0 15 60.0 2 8.0 | 25 8.4 4 16.0 6 2 I .0 | 19 I 76 .0 1 25 8.4 5 20.0 8 2 I •0 I 4 J 1 7 1 68.0 1 25 8.4 9 36.0 I 12 3 I .0 | 4 J 12 I 48.0 | 1 25 8.4 17 68.0 I 12 .2 I 20.0 1 25 8.4 16 64 .0 I 12 3 1 •0 I 16.0 2 8.0 1 25 8.4 14 56.0 I 16 o I 16.0 3 12.0 1 25 8.4 4 16.0 12 1 1 68.0 | 1 4.0 1 25 8.4 8 32.0 8 1 1 5210 2 8.0 1 25 8.4 8 32.0 I 12 I 1 10 | 40.0 | 4 16.0 1 25 8.4 10 41.7 I 12 1 1 ' I 29.2 | 4 16 . 7 1 24 8.0 103 34 . 4 10 32 7 1 4 3 142 47.5 18 6.0 299 100 .0 Table 5.2: Temporal distribution of users by Role on the Skyway System, during weekdays. TIME 1 .00 8 A.M. to 9 A.M. ROLE COUNT ROW PCT V i s i t o r Tourist Securi ty Other Wo Not Dete ROW Guard rker rmined TOTAL 1.00| 2 .001 3.001 4.00| 5.001 9 to 10 10 to 11 2.00 3.00 4 .00 11 to 12 Noon 5.00 Noon to 1 P.M. 1 to 2 2 to 3 3 to 4 4 to 5 5 to 6 6 to 7 7 to 8 6.00 7 .00 8.00 9.00 10.00 11 .00 12.00 COLUMN TOTAL 6 1 1 1 7 I 1 1 25 24 .0 1 4.0 68.0 I 4 .0 1 8.3 8 2 1 14 I 1 j 25 32.0 8 o 1 56.0 I 4 .0 1 8.3 11 4 1 10 I 1 25 44.0 16 •0 1 40.0 I 1 8.3 8 2 1 15 1 1 25 32.0 8 •0 1 60.0 1 1 8.3 8 3 1 13 1 1 1 25 32.0 12 o 1 52.0 1 4 .0 1 8.3 9 3 1 12 1 1 I 25 36 .0 12 •0 1 • 48.0 | 4 .0 1 8.3 13 2 1 6 1 4 1 25 52.0 8 •0 1 24.0 1 16 .0 1 8.3 14 2 1 8 1 1 1 25 56.0 8 •0 1 32.0 1 4 .0 1 8.3 12 2 1 10 1 1 I 25 48 .0 8 o 1 40.0 1 4 .0 1 8.3 12 3 1 7 1 3 1 25 48.0 12 o 1 28.0 1 12 .0 1 8.3 11 5 1 8 1 1 1 25 44.0 20 o 1 32.0 1 4 .0 1 8.3 14 4 1 6 1 1 1 25 56 .0 16 o 1 24 .0 1 4 .0 1 8.3 126 32 1 126 15 300 42.0 10 7 . 3 42.0 5 .0 100.0 Table 5.3: Temporal distributing of users by Role on Nicollet Mall, during weekdays. 190 DISTRIBUTION OF USERS BY ROLE TIME 8EWE8 DATA P E R C E N T A G E D I S T R I B U T I O N -99 - 1010 - 1 « 1 • B B - l l - 2 2 - 3 3 - 4 4 - 6 B - 6 6 - 7 7 -HOURUT INTERVALS 9M VISITOR E2 TOUR IST d] 6 E C . 3 U A R O S3 W O R K E R S Eia N O T D E T . Fig.5.13: Tempore! distribution of users by Role on the Skyway System, during weekdays. DISTRIBUTION OF USERS BY ROLE TIME SERIES DATA P E R C E N T A G E D I S T R I B U T I O N •89- 1 0 1 0 - i n t - 1 2 1 2 - 1 1 - 2 2 - a a - 4 4 - 6 6 - 6 6 - 7 7 -H O U R L Y I N T E R M M J B • i V I S I T O R E2 T 0 U P J 8 T EZD 8 E C . Q U A R D cSS W O R K E R S E9 N O T D E T . Fig.5.14: Temporal distribution of users by Role on Nicollet Mall, during weekdays. North Star City Center IDS Visitor Em Tourist CZZ1 Soc.Quard Worker ESI Not Det. Fig.5.15: Workers on the Skyway System: Distribution of users by Role on the three segments surveyed. 192 the possibility of existence of a larger significant difference for corridors located at some distance from the core area, and among those, the ones with dead walls defining their edges could constitute the least used segments in the system, specifically as urban public spaces. Considering such drawbacks and differences in the physical environment of the Skyway System, it would not be erroneous in expecting sharp deviations in the usage pattern of various segments, in a comparative usage analysis of all the categories of segments within the system. On the whole, it may not be erroneous to believe that most of the segments on the skyway network, except a few, are frequented more by workers than any other user category. In contrast to the spatial organization of the skyway segments, the spatial configuration of the mall is far from linear. Although the line of buildings on the edges represent a linear form, the curvilinear road network and the varying widths of the sidewalks give the entire space a clustered configuration (Fig.5.16, Fig.5.22). The continuous curve of the road, along with the plantings and furnishings, visually divides the entire length of the mall into a series of adjacent clustered spaces, reducing effectively the entire system down to human scale. Not only the spatial organization but also the vertical organization of the walls and facades, throughout the length of the Nicollet Mall, helps in rendering the use message of the mall to one of recreational activity rather than access. Such an environmental use message is consistent with the observed findings, where the furnished portion of Nicollet Mall demonstrated the maximum degree of stationary activities inside the core area (sections 4.3.2, 4.3.3, Fig.5.17). The clustering effects on the mall are further strengthened by the Skyway System. Some portions of the skyway segments, the skyway bridges between buildings, act as space dividing elements for the grade level public areas (Fig.5.18). The maximum benefit of this potential is realized in the mall areas where skyway bridges connect buildings on all four sides. This potential clustering factor can be effectively utilized at other places, to enhance the attractiveness, 193 and support the use message of all downtown grade-level public pedestrian facilities. However, simultaneous care is warranted in not affecting the aesthetics of downtown environment through such development, specifically in the issue related to compatibility of architectural styles, and more importantly the blocking of landmark views (although not included in this study, the issue of architectural compatibility and blocking of landmark views and vistas has been addressed as an important area by many authors including Robertson, 1988 and Whyte, 1988. Its impact on system performance is not known, and would have required a different research approach to discover any statistically valid relationship. No significant impact on system usage was, however, visible). "If Ul Fig.5.16: The clustering effects of furnishing elements and the curvilinear pavement on Nicollet Mall. The relative proportions of the mall's defining surfaces also prove supportive to the recreational use message. With taller buildings set back from the mall facade, the mall's cross-sectional organization define an effective proportion of 1:1 approximately (Fig.5.19). On the other hand, the skyway routes resemble tunnels with distinct directional attributes, leading to the use of the space more as linear access routes than public miscellaneous 194 activity spaces. The segments surveyed, however, did not serve as mere access routes, due to a number of factors including their proximity to the core, and the attractiveness of the ..physical- setting, especially .the atria, spaces.. Nevertheless, minor .variation in usage pattern could be observed on the North Star segment where the proportional representation of people using the system for access was more than any of the other segments surveyed, as well as the adjacent Nicollet Mall (Fig.5.20, Table 5.4). Statistically, a surveyed difference of 37% showed a significant difference between the skyway segment in North Star building and the City Center, with a 95% confidence interval of +24.77% to +49.23%. The same was not true for the segment in the IDS center, where evidence for a significant difference is unavailable. The possibility that other corridor type segments in the system are largely used for access, is evident in this comparison. In such a scenario, not much is possible in the modification of the skyway's physical environment from a directional vocabulary to clustered configuration, since, any form of clustered organization on the skyway segments contribute substantially to the problem of disorientation. However, some reduction to the scale could be effected through appropriate design and incorporation of space dividing elements, and modification to the edge descriptions. Such possibilities will be explored in the succeeding sections. 5.2.3 T H E EDGES AND DEFINING SURFACES The edges constituted some of the major differences between the two systems, although not always in the same direction. In terms of visual complexity, skyways fall far behind the grade level Malls. Even on segments with retailing activities at edges skyways tend to be monotonous in comparison to the Mall, with every retail/business units, on most of the segments, bearing similar external physical character. This may be partly due to rigid design of first floor level spaces by the facility designers. These corridors were 195 140 TOTAL NUMBER OF USERS OBSERVED SKYWAY SYSTEM NIC. MALL(R) NIC. MALL(F) SYSTEMS SURVEYED MARQUETTE AVE. • i Watch ^ Talk EMI Eat H§§ read cnm wait I I Meet M Perform b=d TOTAL Fig.5.17: Stationary activity in the core during weekdays: a comparison between the four facilities surveyed. 196 Fig.5.19: Cross-sectional organization of Nicollet Mall: View from IDS Center skyway bridge. TOTAL NUMBER OF USERS IN SAMPLE 60 I North Star City Center IDS SKYWAY SEGMENTS H i Access Shop 1 1 Promenade H§3 To Work Iiiiiii Search 1 1 Hang out Fig.5.20: Comparison of skyway segments by the trip Purpose of users surveyed. 197 TIME 1 .00 8 A.M. to 9 A.M. PURPOSE COUNT ROW PCT Access. Shopping Promenad To Work, Searchin Hanging Walking ing Working g Out 1.00 | 2. 00| 3.001 4.001 6 .00| 7 . 001 ROW TOTAL 9 to 10 10 to 11 2.00 3.00 4.00 11 to 12 Noon 5.00 Noon to 1 P.M. 1 to 2 2 to 3 3 to 4 4 to 5 5 to 6 6 to 7 7 to 8 6.00 7 .00 8.00 9.00 10.00 11 .00 12.00 COLUMN TOTAL 15 60.0 28 7 .0 4 1 1 o 1 1 4.0 I 1 I 4.0 | 1 25 8.4 1 1 44.0 16 4 .0 16 4 1 o 1 2 8.0 | 2 I 8.0 I 8 2 0 1 25 8.4 14 56.0 24 6 .0 16 4 1 o 1 | 1 1 4.0 1 | 25 8.4 14 56.0 20 5 .0 16 'o 1 2 8.0 I | 1 25 8.4 7 28.0 32 8 .0 32 .! 1 1 4 .0 I 1 I 4.0 | 1 25 8.4 4 16.0 28 7 .0 40 10 1 o I 1 4.0 | 3 I 12.0 | 1 25 8.4 < 16.0 3 6 9 .0 32 .! I I 2 1 8.0 | 8 2 0 1 25 8.4 2 8.0 48 'I 28 1 1 2 8.0 | I 8 2 0 1 25 8.4 16 64 .0 16 'o 16 4 1 o 1 1 4.0 | I | 25 8.4 13 52.0 28 7 .0 16 4 1 o 1 | I 4 1 0 | 25 8 . 4 10 40.0 36 9 .0 16 4 1 o 1 2 8.0 I | | 25 8.4 10 41.7 33 . 5 16 4 1 • 7 1 | I 8 2 3 24 8.0 120 40. 1 28 B6 .8 20 52 . 7 12 4 .0 10 3.3 3 9 0 299 100.0 Table 5.4: Temporal distribution of users by trip Purpose on the Skyway System, during weekdays. 198 undoubtedly better than the ones with opaque edges (Fig.5.5), but some more flexibility in the design could lead to significant improvement (this feeling by the author coincides with the feelings of many of the users, mainly.visitors, who.were interviewed during the data collection process). If the proportional representation of visitors is any indication of their feelings, the collected data confirms it. Although sufficient statistical evidence is not available to show any difference, the proportional representation of visitors on the Mall was 7.6% larger than the corresponding representation on the skyway (Table 4.16). The effect is more evident on the North Star segment where the proportional representation of visitors among user population was the minimum, at 16.6% as compared to 54.3% in City Center, and 29.1% in IDS Center (Table 5.2, Fig.5.15). Statistically, the differences are significant, with the corresponding 95% confidence intervals being +25.5% to +49.8%, and +0.95% to +24.04%, for the differences between North Star and City Center, and North Star and IDS Center respectively. Although, the time period of the respective observations varied, questioning the validity of the results, it gives some ground to believe a real difference may exist due, partly, to the differences in complexity of the physical environment; a monotonous environment in the North Star segment as against the more versatile physical environment in the City Center and IDS routes. The impact of the monotonous environment of the North Star segment is also reflected in the range of activity types. In the two major activity types other than access, namely shopping and promenading, the North Star segment fell far behind the City Center skyway in performance. Out of all the people surveyed on the skyways, the proportion of users observed shopping on the North Star segment was 14% lesser than the corresponding proportion on the City Center segment. Similarly, the proportion of promenaders in North Star building was 20% lesser than the proportion in City Center (Table 5.4, Fig.5.20). With 95% confidence intervals of +1.57% to +26.43%, and +8.67% to +31.33% respectively, there is enough statistical evidence to support the observed difference. The observations on the IDS segment, however, does not provide sufficient evidence. In 199 addition, the time period of observation for the three segments differed, leading to further erosion of validity. Nevertheless, some grounds supporting the difference is available from the collected data.- Access/ walking constituted-the major activity on this. segment (section 5.2.2), and thus, as a public urban space, the North Star segment fall far behind the other two, which may be partly explained by the level of visual complexity of the physical environment. Since the retail/business establishments constituted one of the major varieties of the edge structures, the planning and spatial allocation of such establishments deserve some attention, to achieve proper degree of usage. The skyway usage showed similarity to the grade level from the retail planning viewpoint. The heaviest usage in the system occurred between major retail anchor points including the Daytons', IDS, and Carson Pirie Scott, most of which, coincidentally, include atrium spaces inside the buildings. Due to the City's primary intention of developing pedestrian access routes through the Skyway System, the shops, advertisement windows and displays on the segments form the secondary space organizers. The varieties of material, color and texture used on the segments are numerous, but the major directional vocabulary of the routes tend to render the network as a pedestrian freeway. Users are prohibited from sitting anywhere on the network (seats are provided at only a few places), standing, loitering, and in the case of the City Center skyway, from even photographing without prior permission (the question of public areas in private domain has been a long standing controversy. Krushelnickie, 1985; Brown, Maclean, and Sijpkes, 1985; Robertson, 1985; Warner, 1985, have, among others, raised this issue not only in relation to skyway usage but also in regard to other similar grade level, indoor, quasi-public environment. The analysis in chapter four could not establish any direct relationship between restrictions and facility usage pattern. As has been seen, the Skyway System and the relatively less restrictive Nicollet Mall showed similar usage pattern. However, such restrictions does support the primary objective of 200 developing the system as a pedestrian access route). This type of physical design goes hand in hand with the facility management strategy too (Downtown Council of Minneapolis, .1988;. Minneapolis _ sky way- System- Design--Standards), and hence the Skyway System cannot be termed as inefficient, from the viewpoint of the sj'stem's primary management objective. The second major area of impact on system usage was the openings, which are mainly restricted to the skyway bridges between buildings, and the atrium roofs. Beside the aesthetical considerations, the openings on the bridges and the view through the transparent atrium roofs serve a multiple function of being a major navigational aid, a major contributor of daylight, and the primary linkage of the system to the external environment ("..., to what extent will skyways alter our ability and desire to sense the downtown environment and decipher its differences from other environments?"', Robertson, 1988 raised this pertinent issue during a comprehensive analysis of the Skyway System. The openings on the bridges, and more importantly in the atrium walls and roofs, offer a potential solution to the problem. It was not unusual thus to observe the maximum usage in the areas with larger transparency. To establish a relationship, however, a different research approach is warranted). The differences observed between the North Star corridor and the other two segments, which has already been described in the preceding paragraphs, can be partly attributed to the larger degree of transparency of the City Center and IDS routes to the external environment. It is difficult, with the present level of data, to assess the impact of the degree of transparency, to system usage, but it may not be wrong to assume the presence of a significant relationship, considering the larger variety of users and activities observed on the IDS and City Center networks. No less important was the design of the entrances to the system's areas and establishments. At several locations, the doors leading into the buildings, seemingly, 201 served as major deciding element of the route choice. First, the skyway segments have, on the upper level pathways, lesser number of doors as compared to the building entrances at -...grade.level. Along many.-sky.way routes, doors.ar.e.provided.on.only one side of the skyway bridges, where as at grade level, pedestrians encounter two sets of doors while travelling from one building to another, one set on each building at the street-building interface. Secondly, the design of the doors differ too. Many of the grade level entrances to the buildings have been provided with revolving doors, where as majority of the building entrances at the skyway level have been provided with automatic or free swinging doors. The revolving doors at grade level, undoubtedly, are more energy efficient. Nevertheless, they cut down the pedestrian traffic flow substantially as compared to the free swinging ones at the upper level. Moreover, the negative impacts of the revolving doors on the attractiveness of the system to special user groups including the handicapped, elderly, and parents with babies, is an area deserving further investigation. In terms of pure traffic planning, a single or double layer of free swinging door with cut off lobbies, or the automatic doors allow faster and more volume of pedestrian traffic than a single layer of revolving door (Fruin, 1971, p.53). This probably explains, to some extent, the then increase in attractiveness of the Skyway System as compared to grade level, after the completion of the first major closed network (Podolske and Heglund, 1976). The major impact of the entrance design, however, seem to be more on the volume of user traffic (which has not been included in this study) than the type of user traffic. The design and location of the shop windows constituted another significant difference between the skyway and the Mall. During the last few years, where as the opportunity for window shopping increased on the skyway routes, it decreased at the grade level, in the mall (from personal interviews of the shoppers). The most appropriate example is the Daytons' departmental store. During the time when the skyway routes were not popular, the external facades of the Daytons' building on Nicollet Mall were lined 202 with display windows. As the popularity of the skyway network grew, the display windows on the mall were gradually replaced with blank walls, and the display windows were shifted.to.the upper_ level. The_.changes.were.probably, done. to..allure. more customers.from. the skyways to the upper level of the departmental store. For stores like Daytons', with two levels of shopping areas, the effect on the business at mall level were probably negligible. But such measures poses a serious negative impact on the mall level environment. The possible impacts of such measure are not evident in the collected data, where, the proportional representation of visitors, tourists, and users shopping and promenading on the skyways and the mall were not significantly different (Table 4.16, 4.17, 4.32, & 4.33). In spite of the erosion of window shopping facility on the mall, and the difficulties arising out of the renovation work, absence of any difference may reflect, to some extent, the relative edge of the mall over the skyways, as a downtown attraction. However, keeping the objective of central city revitalization in mind, any competitiveness between the two systems is probably irrelevant. Coordinated planning strategies to reduce the competitiveness of the two systems, and increase the supplementarity is warranted. Practical solutions to the problem of eroding window shopping facility could have been addressed through Public Decrees, and appropriate Zoning Bylaws, in a coordinated effort to increase the vitality and versatility of both the systems. Another impact of the shop window design on the mall environment, lies in the scale. In areas where single departmental stores cover entire blocks, like Carson Pirie Scott in the City Center building, Gaviidae Center in the opposite block, and the Daytons' (Fig 3.4), all shop windows have identical physical characteristics, effectively reducing the complexity of the entire block (Fig.5.21). This also affects the usage since the expensive stores on either side of the block on Nicollet Mall between 6th and 7th Streets seemed to be mostly patronized by upper/upper middle class users, leaving the rest of the users, from lesser status group, to use that block on the malls for access/promenading only. Data regarding the socioeconomic status of the users is impossible to collect through observation, and hence does not form a 203 part of this stud}'. But, the problem is typical to the mall and is not prominent on the skyway since, the skyway segments mostly have small business establishments on its edges, or run directly through departmental stores. Fig.5.21: Facade of City Center Building on Nicollet Mall; deficient complexity. One of the major benefits of the mall's physical configuration was contributed by its walls and facades. In comparison to the skyways, buildings on the mall's edges have a larger variety of architectural styles, materials and detailing, and more importantly, sufficient width of the spaces to see and receive the visual information. Owing to the lesser width, the upper level environment does not enjoy the same degree of opportunity to deliver visual stimulus to its users. However, a large portion of the advantages derived from the diversity is negated by the entrance design to the mall establishments (discussed in the preceding paragraph on entrance door design). The skyway establishments, due to the better access facility from the skyway segments, make shopping easier. The collected data does not show, however, any significant difference between the proportional representations of shoppers on the two facilities (Table 4.32, 4.33). Reasons for such a 204 phenomena could be numerous, and probably extraneous to the scope of this study. The question regarding the design of entrances brings into focus another important impact - -— area.-The-.mall. may. .have.a-larger .variety, and_more..diverse.physical feature .of.its walls . and facades, but the positive impact of such a diverse physical environment is contrasted by the comparative]}' lesser degree of interrelationship between the public and private domains. The interrelationship of the public and private/quasi public domains on the skyway segments is stronger than on the mall. The first factor is the openings. The upper level establishments with larger openings, shop windows, and open doors combine the areas for access on the segments with the adjoining establishments cohesively. Walking into the establishments on the skyway network is easier than walking into their counterpart establishments on the mall. The climate acts as a determining factor. Where as the skyway establishments have the opportunity to leave the doors open all the time, the establishments on the mall are compelled to keep their doors closed due to the weather conditions as well as environmental pollution factors. As a result, they lose the bond between the public and the private domain. In addition, the openings and windows on the mall facade are proportionately smaller than the one on the skyways. Another important factor was the brightness of the private domain. The interior brightness of the upper level establishments as compared to the skyway pedestrian thoroughfare remains significantly higher, thus attracting considerable attention. In contrast, the interior brightness of the mall establishments remains considerably lesser than the external sunlit public spaces. As a result, where as the establishments on the skyways act as focal points, the establishments on the mall, visually, only define the mall edges, and the mall becomes the focus of attention at grade level. Once again, the general activity pattern on the two systems did not expose any significant differences to reflect the differences in the physical environment. 205 As observed in the preceding sections, the mall suffers from some serious disadvantages in comparison to the skyways, in some of the numerous aspects of the _ -.physical environment...Probably,. such a.combination._of.positive and negative, impacts,.on both the systems, result in an equilibrium of the attracting forces. Equally likely is the possibility that there are many other areas of macro/micro level influences totally ignored in this study, which either could have explained the absence or exposed the presence of any significant differences. But before addressing those issues, some other areas of significant impact is worth exploring; the furnishing, and the activities. 5.2.4 F U R N I S H I N G O N T H E F A C I L I T I E S One of the most crucial determining factor of activities on the two systems seemed to be the level of furnishing on each. The problems arising with the acute lack of furnishing elements on the skyways have already been discussed in sections 4.3.2 and 4.3.3. Beside the problem in facility design, the facility management strategies also discourages any type of activity on the system other than access. The potential of the Skyway System to serve other activity types, nevertheless, should not be underestimated. Most of the skyway routes, as a result, seemingly serve more as access routes for inter-downtown trips, than a public place, as compared to the mall where access is only one type of the wide range of activities taking place. Although this characteristic is not evident in the data, a survey of the entire system, as against a few segments inside the core, could show the difference, and an example of such a usage pattern has already been demonstrated as regard to the North Star segment (section 5.2.2). The problem is not one of ambiguous allocation of activity areas. In both the facilities, activity areas have been more or less clearly demarcated through the use of appropriate material and furnishing elements. The few 206 furnishing elements on the skyway like the garbage bins and the sand boxes are located in such a fashion as to eliminate any obstruction to the free flow of pedestrian traffic. The question regarding the frequent change in ground material, color, and texture on the skyway segments, and its effect on use message have been addressed by some authors (Robertson, 1988). The primarj' argument had been regarding the question of public environment, where the authors argued that the frequent change of material on the segments render the segments as private areas in Minneapolis, and the system in St.Paul has been hailed as the ideal due to the uniformity of material and ground treatment on the entire system. However, the analysis in chapter four does not support such a contention. People, in Minneapolis, use the skyways, a private area, as much as they do the mall, a public place, in most of the activity types. It has also been commonly acknowledged that the high quality of materials used on the skyway segments, coupled with the frequent changes in material and detailing, render the area as spaces for the upper/upper middle class people. However, lower class users have not been actively discouraged from using the skyways as can be observed from the data (data on the socioeconomic status of the users, which is a combination of the level and type of education, income, and occupation of the people, is impossible to obtain through simple observation. The remarks made here are based on the data collected regarding the users in the 'others' Role category, most of whom seemingly belonged to the lower class in social status). The distribution of the types of users on the skyway and the Mall did not show any significant difference (section 4.2.4). There were no more lower class users (the 'Others' category) on the mall as on the skyways. As discussed in the same section the entire core area probably does not fit into the perceptual vision of those users as ideal spaces, which is evident from the fact that most of the users from that category use the adjacent Hennepin Avenue as their activity spaces. If evaluation is a major step before action, there is ho evidence that the skyways were evaluated differently than the mall, as far as the users social status is concerned. 207 The comparative evaluation from the activity viewpoint, however, probably accounts for some of the differences in usage pattern between the two systems. The skyways, in .general, were more limiting in,usage types than the mall (description of the North.Star section in section 5.2.2 and 5.2.3; section 4.3.2 and 4.3.3 on stationary activities). As a result, people in forms of activities other than shopping and access tended to be on the mall, and a few segments on the skyway network with high levels, and larger variety of activities. Majority of the users, other than workers, who use the skyways probably restrict themselves to locations where retail establishments act as powerful focuses, or where the segments cross atria spaces. One of the main reasons for a possible variation in evaluation of the two systems from the activity viewpoint is the contrast in the level and type of system furnishing. Among the furnishing elements, the seating facility is most crucial to user needs. This has been discussed in chapter four. People sit for various reasons; to rest (extremely important to the elderly and handicapped users), to eat, to read, to meet, or to watch/groom (mainly by the retired elderly users). The skyways severely lack in the provision of seating facility, although such facilities can be accommodated by implementing proper changes to the management strategies. Basic furnishing is so crucial that even in the portion of Nicollet Mall under renovation and the Marquette Avenue, users could be frequently observed sitting on unfinished pavements and fountain parapets, and in some instances eating lunch while standing, primarily due to the lack of provision of seating facilities. If the skyways are to be made more attractive, seats are the first furnishing elements which should be provided on each route, at strategic locations, and substantially. The larger degree and variety of activities on the City Center segment as compared to the North Star segment may be partially attributed to the provision of seating facility on the City Center route (section 5.2.2, 5.2.3). Further, the degree of stationary activities on different segments of the skyway network reflect the impact. 63.8% of all stationary activities observed on the 208 skyway network was inside City Center. Around 43.18% of them were observed sitting. In contrast, only 17.4% and 18.8% of all stationary activities was observed in the North Star .... — . .-..and IDS segments respectively (Table.5.1, Fig.5.10,.Pig.5.11) , The skyways also lack in the provision of focal elements except for the few atria spaces. On the grade level mall, and even in the portion of the mall under renovation, a large variety of physical elements and activities act as focal elements. Among those are fountains, trees, sculptures, and several activities including the street side entertainer/artists. A unique focal element was the construction machinery and equipments, and to a large extent, the construction work in progress attracted a lot of attention from the users, which even included the downtown workers, who, while using the mall for access, tended to slow down and promenade, glancing at the ongoing construction. The mall under renovation acted as a live theater, although a different type than conventional ones, constantly providing the onlookers a host of new information. Almost every user seemed to love watching the people work on the renovation project, and they loved watching it proceed day by day; each day a day nearer to completion. They watched their city change in slow motion, and that provided a unique kind of visual stimulation. Cyclists could be observed challenging the circuitous routes laid out by the construction team, and the promenaders never forgot to stand by and watch for a few minutes. The mall, though physically demolished, was not dead. All of these stimulants were absent on most of the segments of the skyway network, barring the few atria and the walkway bridges. The bridges are amongst the prime attractions on the skyway, chiefly due to the vantage viewing point they provide on the city streets. However, during peak traffic, people stopping on the bridges constitute potential hindrances to the basic management strategy. It also is illegal to stop for a long period, and the design of the bridges does not provide for this user need. As a result, most of the active segments on the skyways tend to be in the vicinity of focal points like the atria. The impact of the absence of focal points can 209 be observed partly in the usage pattern of North Star segment (section 5.2.2, 5.2.3) as against the other two segments. Beside acting as focal elements, the trees, benches, canopies and the skyway bridges, among other features, help in reducing the scale of the mall to a human level. The clustering effects of the skyway bridges and the canopies on the Nicollet Mall between 6th and 7th Street are the only elements which successfully negate the dulling effect of the block long, monotonous, building facade and shop windows of the City Center and Gaviidae buildings. Larger windows, brighter shop/display windows, and better entrance doors, as well as more number & varieties of canopies on the Nicollet Mall and other grade level streets including the Marquette Avenue could substantially enhance the use message of the Mall, and induce a stronger interrelationship between the public and the private domains. The skyways lack in such space dividing elements, which could go a long way in reducing the strong directional attribute of the designed spaces. A lot can be learnt from several features of mall and even the skyways, as regards to the impacts of space dividing elements. The foremost example on the skyway are the routes through departmental stores. The display areas in the store effectively reduce the scale as well as the directional attribute of corridors. Such segments, in spite of being one of the most disorienting routes, never lose complexity. Since many of the users of the upper level establishments are regular/quasi-regular ones, the problems of orientation as regards to cognitive mapping does not arise, and the sensory perceptions continually encounter new levels and types of information due to the frequent major and minor alterations made in the department store's displays. The same effect can be witnessed on the mall near the sidewalk vendors, at some locations, displaying and selling their merchandise freely. The possible outcome of such urban design strategies can also be seen in the weekday Farmer's Market held on each Thursday on Peavey Plaza, located on Nicollet Mall(F). They not only attract more users to the mall, but also very effectively reduce the scale of the mall to a more human 210 level, simultaneously introducing a healthy complexity level. Certain level of spatial division would not be detrimental to the interest of skyway establishments, if properly coordinated with the .primary objective_..of_..providing. access thoroughfares. Due to the distinct edge of the climate controlled skyways over the mall in harsh weather cities like Minneapolis, and many other North American urban agglomerations, several such minor improvements to the physical environment may bring about substantial improvement to the system's attractiveness as a public place, even outside the core. Implementing such ideas would not prove to be difficult in privately owned systems. Owners could be encouraged in privately owned systems, through tools including Bonus Zoning (which has been used in the past by many cities concerning public facility improvement projects; Fruin, 1973), and other incentives, to follow guidelines laid down by the city planners, with the ultimate objective of providing good quality urban spaces to the walking public. Since most of the publicly owned systems get financed through public taxes, government bonds, and public grants it may prove relatively difficult to obtain finances for such ventures. However, pilot projects at small scale could be implemented, by including such measures within the overall strategies for downtown revitalization, to see their success rates. Success of pilot projects could make obtaining funds an easier task for the city. The strategy would provide an alternative to the users, rather than shifting users from one system to another. Whyte's assessment that food constitutes one of the strongest attractions in the plazas of New York (Whyte, 1980; regarding clustered public domain) holds true for linear public domain like the mall, the skyways, and other grade level streets as well. The best evidences were the seats provided on the sidewalks, including in front of restaurants on the mall. The seats in front of the restaurants were the most used seats on the mall, and the other fixed benches fell far behind in attracting users. An evidence of such a phenomena can be seen in Fig. 5.22 and 5.23, where the distribution of users engaged in stationary 211 activities on Mall(F) have been graphically plotted. During both the time periods, the area in front of the restaurant (shown inside the dashed circle) remained the busiest (due to practical difficulties only one side of.the facility .could be. observed at a time.. Hence .data on stationary activities on the north pavement of Mall(F) between 1:00 p.m. and 4:00 p.m. is not available). Beside the factor of social comfort due to the free standing seats provided by the restaurants, the factor of food availability probably exercised equal strength. In many areas even outside the mall, at grade level, the areas outside restaurants constituted the busiest areas in the block. A good example of it can be observed on Marquette Avenue between 6th and 7th Street (Fig.5.24). The only restaurant within three consecutive blocks (shown within dashed circle) attracted the largest concentration of users on the sidewalk, during the time period of observation. The activities near the food vendors on the mall is another evidence. During the hours between eleven in the morning and two in the afternoon every day, during which the food vendors opened their movable shops, stationary activities increased substantially (Fig.5.25), chiefly near and around the food vendors (Fig.5.26). Beside eating, people engaged in many other forms of stationary activities near the areas where food was available; more than any other area on the block. Although food is available on the skyway segments, the restaurants are not strategically located to generate or attract other forms of activity. The lack of furnishings and improper location of restaurants on the skyway explains to some extent the lesser degree of stationary activities on the skyways as compared to Mall(F). The skyways lacked another important asset; the sun. However, in Minneapolis, for a substantial part of the day sun does not reach the ground on the mall. This is primarily due to two factors. First, the North-East South-West orientation of the mall. And second, the shadows cast by high rising structures in the core area. The attractiveness of the mall, due to the Sun, in comparison to the skyway did not seem to be an important impact factor. 212 Ld Ld LY I— (/) I • • •• pj i t - 7 RESTAUR* SHDP SHOP SHDP L-r-i r YMCA BUILDING u u LY I— </) JZ I— OJ NDTEi EACH DDT REPRESENTS ONE USER DN THE FACILITY Fig.5.22: Stationary user distribution on Mall(F) North sidewalk, between 9:00 a.m. and 12:00 noon, during weekdays. Ld Ld LY r -</) C7 rue m i « VENDORS ^ ^ KXTMJRAHT SCATS \ cm KICKS NICOLLET MALL i pins Rjwra BOXCI O M O I O B B O Q 1 — / Z) SHDP SHDP SHDP REST/ L-n r YMCA BUILDING ' Ld Ld LY I— C/) IE I— OU NOTE. EACH DDT REPRESENTS ONE USER DN THE FACILITY Fig.5.23: Stationary? user distribution on Mall(F) North sidewalk, between 4:00 p.m. and 8:00 p.m., during weekdays. 213 NDRTHWEST CENTER L d L d LY I— (/) H \— MARQUETTE AVENUE SKYWAY BRIDGE Q. • IE \ / • RESTAURANT Q _ • X CO • X IO C L • X to L d L d LY f— X I— NDRTHSTAR BUILDING NDTEi EACH DOT REPRESENTS DNE USER DN THE FACILITY RESTAURANT OUTDOOR SEATS NOT SHDVN DN THE DRAWING Fig.5.24: Distribution of Stationary- users on Marquette Avenue South sidewalk, between 1:00 p.m. and 4:00 p.m., during weekdays. 8 - 9 9 - 1010 - 1111 - 12 12 -1 1 - 2 2 - 3 3 - 4 4 - 5 5 - 6 6 - 7 7 - 8 — WATCH - t - TALK - * - EAT READ -x- PERFORM WAIT MEET TOTAL Fig.5.25: Temporal distribution of users by stationary Activity on Nicollet Mall(F), during weekdays. 214 Ld Ld QL h-CO I P E A V E Y P L A Z A V E N D O R S G d. \ =& • •• V \ OUTDOOR RESTAURANT SCATS c m IOCHCS NICOLLET MALL PLANTER KIXES ~£ • ~t—. i O I O m O y Oi QL ZD < h-CO Ld SHDP SHDP SHOP "Lr-. r YMCA BUILDING Ld Ld QL t— C/0 •c I— cu NQTEi EACH DDT REPRESENTS DNE USER DN THE FACILITY Fig.5.26: Stationary User distribution on Mall(F), South pavement between 1:00 p.m. and 4:00 p.m., on weekdays. 5.2.5 ACTIVITIES AS DIFFERENTIATING FACTORS People tend to get attracted to spaces with activities rather than isolated, unused spaces (Whyte, 1980). This phenomenon was clearly evident in the core area of downtown Minneapolis. It partly explains the concentration of comparatively large number of skyway users near and around the city core, leaving vast tracks of the system's segments virtually unused, and the major concentration of most stationary activities on the mall in the vicinity of the restaurants and food vendors. In a larger context, it also explains the larger degree and variety of stationary activities occurring on the mall as compared to the 215 skyway. From such a perspective viewpoint, the mall are more successful as public places, and are conducive to a larger variety of activities, including recreational activities like -jogging, walking,, roller, skating, .cycling,, and other such sports, which was totally absent on the skyway, and can never be accommodated due to design constraints. But activities like standing and watching also are rare on the skyway segments, except for the skyway bridges and atrium galleries, which has already been demonstrated as regard to the North Star segment. Street characters including artists, musicians, and entertainers, are allowed on the mall and atrium spaces, but never on the skyways. Festivals and exhibition rarely get to the upper level, but are frequent on the mall. As public spaces from the viewpoint of social activity, the skyways, in general, lie far behind the grade level malls, but not necessarily due to limited design potential. Since the skyways along atrium spaces seemed to be the most active segments, they deserve some further explanation. The only important welcome break in the environment on the skyways were the atriums/hubs. More users on the system could be observed circulating around the atrium spaces as opposed to upper level shopping corridors (the remark is based on unscientific observation). The atria act as the nerve centers of the Skyway System. Most of the activities are concentrated around the atrium spaces on the skyway routes, making these multilevel interior spaces the hub of downtown activities. Most notable hubs are the IDS Crystal Court, and the City Center (Carpenter, 1975; Robertson, 1988), and many more are on the line for recognition. On the skyways, the presence of the atrium and the skyway bridges served a more crucial role than the furnitures, and even in the absence of furnishing elements, activities around atrium spaces remained at its peak. An analogy can be drawn between the role of the atria in the interior spaces and the plazas in the exterior spaces. The areas on the mall with the most varying and largest degree of activity types were the areas adjoining the plazas. Although the activity level inside the plaza was comparatively higher, it influenced the activity levels on 216 the adjoining mall and the sidewalks. The potential of closely integrating the mall with the plazas, and the skyway with the atria into a single integral pedestrian system is worth exploring. The presence of. the. plazas not only offers a diversion to the mall's linear spatial configuration, but also offers a substantial change to the activity levels and types on the mall, enhancing the overall complexity of the core area environment. Smaller plazas on the mall also experience heavy usage. People frequently use the spaces for eating, watching, meeting, talking, or simply resting. Similar activities could be observed in the atriums, but all were restricted to the lower level. Even without direct encouragement of activities on the upper level segments, mere transparence of the segments to areas with high degree of activities, outdoor or indoor, influence the skyway activity level positively. The importance of lateral transparence can be partly seen in the atrium space inside the T C F Tower, where a laterally enclosed atrium did not attract activity level any where near the level of the IDS Crystal Court (no data has been collected to prove this). The atria spaces, in addition, have a definite advantage over the plazas. The skyway segments more or less surround the atria spaces, where as the plazas remain confined to the edges on grade level. This centrality of location of the atria spaces in the skyway network makes the atria more effective affective spaces in comparison to their counterpart at grade level. 5.2.6 SUMMING Up Probably, the presence of the atrium spaces in two of the three skyway Segment surveyed explains the similarity in usage pattern between the skyways and Nicollet Mall. Another distinct possibility is the balancing impact of the hosts of positive and negative factors in each facility, thus making the facilities equally attractive. To briefly consider the various positive and negative aspects, the general configuration of the spaces on the skyways is less attractive as an urban public place in comparison to the spatial 217 composition of the mall, except for the few skyway corridors running through atrium spaces/hubs and departmental stores. The skyway segments are also far inferior in the level of .visual, complexity, than, the mall, severely lacking, in. visual ...stimulation in large portions of the system. It lacks any considerable degree of system furnishing, especially seating facilities, as well as in the availability of focal elements to act as coordinators of spaces and activities. In absence of any space dividing elements on the corridors, not only is the distinct directional attribute of the linear spaces overemphasized, but the overall physical environment also suffers from monotony. Although there is no dearth in food availability on the system, the locations of such facilities are not strategic enough to attract or generate other activity types on the system, as compared to similar establishments on the mall. The system, in terms of the degree of prevailing stationary activities, stands far behind the mall in comparison. On the other hand, the skyways have a better physical environment in several key areas affecting the facility's performance. The entrance design on the corridors of the system as well as to the system establishments on edges are far superior to their counterpart in the mall. Shopping is seemingly easier on the upper level, and the provision of vantage viewing points on the system, providing attractive views of the streets, buildings, as well as the atrium spaces, act as extremely important attractors. One area in which the upper level areas have a cutting edge over the grade-level mall is in the provision of user facilities including public washrooms, and access to climate controlled rest areas. Since neither of the two systems have a physical environment with relatively net positive impact, the possibility of the forces of attraction/repulsion reaching equilibrium is an area warranting further investigation. As addressed earlier, inclusion in the survey of all the different types of skyway segments discussed at the beginning of this chapter, including segments outside the core area, could possibly have resulted in a very different explanation of system usage pattern. Such an action would also have had taken into 218 consideration the impacts of the relative scale of the two systems (section 5.2.1) on system usage. The impact of the renovation work of the mall on system usage has not been ... considered.in the.analysis too,.which would..have.necessitated some.difference in approach to the method adopted in this study. 5.3.0 A Better Environment In spite of all the areas incomprehensible with the present level of collected data, one conclusion can be clearly drawn from the data analysis. The skyways in the core, in spite of several handicaps in design flexibility, and a less suitable physical environment as a public place in comparison to the mall, have managed to attract hosts of users into its fold every day. A part of the attraction may be arising from some relatively less positive attributes of the mall as described in the previous sections, but by itself, the skyways constitute a major attraction to the city dwellers, in the modern urban landscape. The system, has developed into a closely integrated pedestrian access network, and several major segments get utilized to their optimum level during most hours of the day. From a transportation viewpoint, the system cannot be termed as unsuccessful. Although many of the segments remain under utilized for substantial period of time (Whyte, 1988), any cost-beneFit analysis (section 2.8.8) should take into account the fact that in every transportation system, several networks are developed through cross subsidization, to enhance the system's overall benefit (most of the assessments of the Skyway System in the past have been concentrated within the realms of transportation planning. Majority of the authors, including Ponte, 1986; Heglund, 1982; Podolske and Heglund, 1976, have judged the system to be successful from the viewpoints of traffic congestion, pedestrian safety, reduction in traffic accident rate, vehicle delay savings, and vehicle fuel savings. Hence it may be safe to assume that the Skyway System provides some innovative 219 solution to transportation related problems, and would rate high in any cost-benefit analysis). A better environment can be created not by looking at the system from the narrow perspective of transportation planning. The key, to certain extent, lies in encouraging design initiatives aimed at integrating the Skyway System and the Mall into a single integrated pedestrian facility. As measures towards making the two systems supplementary, certain variations should be introduced in the intended usage of the two systems, to reduce the competition between lower and upper level facilities, and to cater to, to some degree, mutually exclusive needs of the users. The positive and negative attributes of both the systems should be thoroughly considered before designing a comprehensive downtown pedestrian system. These measures, nevertheless, would not totally address the problem of downtown attractiveness vis-a-vis the suburban establishments. Macro level planning policies with wider influence are necessary for such measures. But an integrated pedestrian system could increase the attractiveness at the micro level substantially. Without any change to the prevailing management strategies, the Skyway System could possibly be made more attractive by the incorporation of some minor changes to the physical design of the facility (the following recommendations are mere suggestions and not well researched ideas for implementation. Further research is necessary to assess the impacts of the suggested changes). Building corridors linking skyway bridges should ideally have some portions of it aligned on the external wall of the buildings, thus alleviating marginally the problem of disorientation, mainly to irregular users. Such a measure could introduce the possibility of direct connection of the skyway segments with the side walks at grade level (Robertson, in his 1988 survey cited the example of the Cincinnati Skyway System, where similar measures have resulted in considerable improvement to system accessibility as well as visibility), thus increasing the attractiveness of the system. The 220 possibility of building skyway bridges at street intersections rather than at mid-block locations should be explored, which could increase user accessibility to the system (in a comprehensive...evaluation of all possible., locations of skyway ..segments, Podolske and Heglund, 1976, however, suggested mid-block location as the best of the alternatives. The topic constitutes an ideal area for further research). Small changes in angles of the pathways and confusing intersections should be avoided as far as practicable. To make matters easier, every new development in the central city should be asked to keep provisions for skyway corridors in the original design, irrespective of whether the system actually gets built simultaneously or not. Corridors with dead walls on edges should be made wider, as compared to other segments, to avoid the feeling of congestion on the system. Although the issue of private ownership, and restriction of activities on the segments could not prove to be a factor effecting any dissimilarit}' in usage pattern between the Skyway System and the less restrictive, and more public Nicollet Mall, a more uniform code of rule and regulation on the entire network, and some degree of control in the hand of the public authorities would probably prove beneficial. A little more liberty to the owners of the skyway establishments in designing their shop fronts and display patterns could substantially reduce the monotony of the corridor environment. The edges of the system, especially on the segments with retail/business facilities could be given varying physical characteristics by manipulating the frontage, displays and signs, to introduce some uniqueness to each segment, in activities as well as visual stimulation. Transparency of the system to the external environment should be given more importance. The success of the IDS Crystal Court and City Center atria is not only due to its central location in the core area, but also probably due to the larger transparency to the external public domain, and the mall activities. Although atrium spaces are one of the most successful areas on the system, it would not be practical to include such spaces in all downtown buildings. Moreover, the incorporation of more 221 atriums should follow any considerable increase in user traffic to support such development. Along with the transparence, the inclusion of vantage viewing points is essential.. First, any desired modification to the design of skyway bridges, should be made in order to cater to the users who find the view from above the street interesting to watch. The transparency of lateral walls on skyway bridges, atrium walls and roofs go a long way in helping the users get the feeling of the context. Some more transparency of the corridors could only have positive impacts, not negative. Another vantage point is the atrium gallery. Minor cantilevered projections on the skyway routes overlooking the atrium spaces attract a lot of users. Instances of such occurrence can be frequently observed in the City Center route, at the first and second gallery levels. Certain coordination of the vantage viewing points with users rest area could turn out to be extremely attractive physical features of the system. Creation of such points of attraction on the system, along with the availability of food for the users would further appreciate the attractiveness. Food availability, by itself, could be a major attraction, and proper planning to include some speciality restaurants at strategic locations could turn out to me major destinations for all types of users (in his survey during 1988, Robertson discovered the virtual absence of any expensive restaurants on any of the five systems, suggesting that the upper level market is geared towards the lunch hour traffic. Hence, experiments may be made by including speciality restaurants to see if any changes occur to the overall usage pattern, and system attractiveness). Such moves could possibly act as activity generators. Minor modifications to the management strategies could allow for some degree of activities, other than access, in those specially designed focuses, and such special focal attractions at intermittent locations could also break the prevailing monotonous character of system corridors. Further provision of user amenities and facilities in the focuses, and location of visible mechanical vertical transportation facility in it vicinity, preferably linking the system with at-grade pedestrian areas, could considerable enhance the attractiveness of the system. However, any of the suggested measures may not boost the systems attractiveness as 222 much as the provision of furnishings. Seats at these focal points and other locations on the corridors could cater to large number of users. One important area where seats are urgently required is.near .the location of directional maps. For users, especially the tourists in groups, families, and visitors with children and shopping bags, some provision to park the belongings and children could be a lot of help. The need of seats for the elderly and the handicapped is self explanatory, and need not be explained. Along with the modifications to the facility's physical environment, some changes to the management strategies, as will be explained next, are essential to make the entire system a much better public place. Keeping access as the primary role of the facility, minor changes to the management policies could prove to be the most positive contribution to the system's attractiveness. 5.4.0 A Public Place ? But, in spite of the comparably good performance, are the skyway spaces truly public? The answer is no, and hence, before concluding the discussion it is essential to address a few crucial issues. In a broad definition, public places can be described as spaces where restrictions on activities and users are non existent. A guarantee on absolute freedom of activity is absolutely essential to bring out the flavor of publicness in any outdoor or indoor urban area. For that matter, even a perception, of restriction on activity or usage, in the minds of the users could generate negative feelings about the space, inside them. In a true democratic sense, politics, culture, sports, festivities, congregations, and discussions, at the least, define the various facets of the otherwise complex term. From such a holistic viewpoint, the skyways lie far behind the mall as urban public places. It may be erroneous to infer from the analysis in this thesis that the skyways are truly public, considering the similarity in users and degree of usage pattern between the mall 223 and its segments. Whether activities like placarding, striking, congregating, or campaigning should be allowed on the skyway segments could be an ideal area for wide ranging, debate. It would .probably, rest. on. the. users,.in each urban area, to decide on such critical issues. The term public place was used in this thesis from a much narrower viewpoint of similarity in user population and usage, and did not include an assessment of the degree of actual or perceived freedom the users enjoy on each system to engage in activities of their choice. Clearly, many activities, including campaigning, loitering, sitting, or congregating are banned on the skyway network. It is only in the other sectors that the skyway performance come close to that of the mall, a truly public place. However, looking at the data one finds a larger degree and variety of activity on the skyway and the mall as compared to the streets. The factor explaining a large portion of such a trend is perceived safety. People perceive the streets and sidewalks in the central city as unsafe places to engage in any activity type. As a result, increasing the safety of users, perceived or real, on the city streets should be addressed simultaneously with the issues concerning the public usage of skyway system. As discussed earlier, activities like sports, politics, or campaigning on the skyways are debatable. However, without raising any controversy, two main areas of public activities could be introduced, and should be introduced to increase the attractiveness of the system, as well as to make it more public. Both the areas are directly related to the management policies. The first area pertains to the degree of facility furnishing. Substantial changes to the management policies is essential to include and allow most types of stationary activities on the segments, at interval locations. The second area addresses an issue of more public nature including allowing activities like festivals, exhibitions, sales, and street performers at a few locations. Both the features could enhance the system performance as a public place to a large degree. Allowing such activities goes hand in hand with the suggestion in this thesis regarding creating intermittent focuses at the upper level (section 5.3.0). Of course, such changes in management policies need a simultaneous restructuring of the skyway spatial 224 configuration. A move in that direction, would probably render the Skyway System a better tool for revitalization, since the increase in publicness could lead to an increase in its attractiveness to large number of.people outside.the central.city, and.the suburbs. .. 5.5.0 Going Downtown In light of the intended development of the system as a city revitalization strategy, the modification of the physical features of the urban landscape may not be the only factor to exercise any significant bearing on the suburban dweller's decision to visit the central city. A visit around the city raises many questions. After three decades of development, the city core, probably, is still not the first place people plan to visit during their free time. The large expanse of lake shores and the Mississippi river front continue to serve the recreational needs of the major bulk of urban dwellers. Many areas inside the downtown are still being considered as dangerous by the common public. And suburban retail establishments seem to be having a clear edge over the downtown in retail sales. The development of the Skyway System, therefore, should be considered from a wider perspective of macro level development policies. Some of the important areas of policy development, among others, include an effort towards a coordinated development of the Central City and suburban public recreational areas, more coordination in the areas of automobile traffic, downtown access, parking provision and pricing policies, downtown residential population, year round downtown activity planning, and a careful assessment and planning of downtown vs suburban attractions. The Skyway System not only provide areas where people find protection from harsh external weather and automobile traffic. This is amply evident form the finding that 225 even during summer months and weekends the skyways attract a user population comparable to the mall below. Cold winters could be increasing the attractiveness to _. several degrees. Designed .judiciously.,—and .integrated, properly..with . the grade level networks, the system could be an effective solution to numerous downtowns where compact development is a necessity for future growth, and where downtown streets no longer copes with the increase in automobile traffic. However, the survival, to some extent, depends on the user population's acceptance and support of such form of indoor oriented development of quasi-public environments. The people of Minneapolis seem to be supportive, and to a large extent, even take pride in the ultra modern touch such developments provide to their central city. The System in Minneapolis has an added advantage. The entire downtown has a single retail street, which gives the multi-block retail areas of the skyways tremendous advantage. In cities where retailing covers more than a couple of blocks in the downtown, the feasibility of an upper level retail development could be questioned. A good example could be the city of Vancouver where large multi-block street level retail bases have been established, and the climate remains mild throughout the year. The potential of system development in such a case is questionable. Moreover, the streets of downtown Vancouver are not perceived as unsafe by the city population. The possibility that the people of Minneapolis restrict themselves to the skyway routes due to their negative impression about the streets is not unfounded. Probably, had the sidewalks of Minneapolis been safer, the impact of private control of the skyways, on its usage pattern, could have been more tangible. Besides, the downtown residential population makes a significant difference. The streets of Minneapolis remains virtually deserted from six in the evening till eight in the morning every day, in spite of. the presence of two large scale pedestrian attractions. However, the same is not true for a city like Vancouver of approximately the same size, or larger cities like Chicago, where the central cities remain active for most part of the day and the evening. Without going into specific details, one of the major differences seem to be the size of residential population of 226 the downtown areas. Without any substantial residential population, underutilization of facilities during off hours could probably be expected, as in the case of downtown Minneapolis. Another .aspect .deserving attention is. the. retail facility. Retailing does not seem to be the only ideal factor for increasing downtown attraction, since the suburban facilities have an edge in many important areas; primarily parking. Parking cost and availability, trip cost, and traffic congestion exercise considerable bearing on trip making decision, and it seems easier for people to shop in the suburban facilities than the central city. To increase downtown's attractiveness, and the System's attractiveness, attention should be diverted to many other special provisions and activities/environments not available in the suburbs. To be successful as an agent of central city revitalization, the Skyway System should generate, and be conducive to special activities, beside being a pedestrian thoroughfare. Wilcox, 1980, explained the significance of attraction management to the success of downtown revitalization strategies. He stressed the importance of integrated coordinated regular annual promotional, entertainment and visitor generation programs, to increase the attraction of downtown as a whole. Many downtowns have long lost their former advantage as the prime retail area, and "..the business of the new downtown is the capture of customers and visitors, and the stimulation of as many return visits as possible" (Wilcox, 1980). From such a viewpoint, stress on the coordinated development of Nicollet Mall, and the Skyway System as a single pedestrian strategy, along with the planning of generation of special year round activities on the facilities is essential. A research strategy with much larger scope is warranted for such a comprehensive study. The present study has provided one important piece of information. Inside the core, with the inclusion of special facilities like atrium spaces, and segments linking major retail facilities, the skyways probably enjoy the same degree of attractiveness as their grade level counterpart, the Mall, and demonstrate lesser dissimilarity in usage pattern as 227 compared to the streets. In addition, the two systems, together, represent a significant improvement to urban environment from the conventional grade level streets and sidewalks. .Considering, .the... private., .control, of ..the skyway .corridors, it is a.remarkable performance. However, the performance cannot be generalized for the entire system. In addition, the positive impact of perceived unsafety on city streets, on the performance of the skyways is an area worthy of analysis. The next step towards the assessment of the Skyway System from an architectural viewpoint, should include a comparative analysis of the large varieties of skyway segments, a broad classification of which has been attempted in the beginning of this chapter. The survey location in the next step should cover the entire central city area, rather than core areas. Such an analysis could provide the architects with the knowledge regarding the better and lesser used segments, their interaction with the physical environment, and the various shortcomings in facility design. A substantial improvement to the present finding can be achieved through the employment of other alternative research methods (triangulation of measurement process), to supplement the findings of the survey through observation, as well as the survey of the facilities during all the other seasons; Autumn, Winter, and Spring. Lastly, any future endeavor in this direction should include the survey of the system in every other North American city to increase the universality of the research application. ******************* BIBLIOGRAPHY Ambrose, David M . , "Longitudinal Assessment of Businesses in a Central Business District", Mid-South Quarterly^ Business Review, Vol.l8,_ No.3, October 1980, pp.3-5 & 1 2 . " ~ ~ " Argan, Giulio C , The Renaissance City, Translated by Susan Edna, George Braziller, New York, 1969. Barnett, Jonathan, "Helping Downtown Compete with the Suburbs", Architectural Record, Vol.151, January 1974, pp.127-135. Bentham, C.G. and Haynes, R .M. , "The Effect of Vehicular Traffic Restriction on Pedestrian Number", Traffic Engineering and Control, Vol.26, No.4, April 1985, p.210. Benz, Gregory P., and Lutin, Jerome M . , "Midtown Underground Pedestrian Connection Study", Journal of Transportation Engineering, Vol.109, No.4, July 1983, pp.487-498. Brown, David and Sijpkes, Pieter, "Critiquing the Underground City", Planning, Vol.51, No.3, March 1985, pp. 16-17. Brown, David, MacLean, Michael, & Sijpkes,Pieter, "The Indoor City - Can it be successfully integrated into the urban fabric?", City Magazine, Fall 1985, pp. 16-21. Brown, David, Sijpkes, Pieter, & Maclean, Michael, "The Community Role of Public Indoor Space", Journal of Architectural and Planning Research, Vol.3, No.2, May 1986, pp. 161-172. Carlson, David and Carlson, Mary R.S., "The pedestrian Mall: Its Role in Revitalization of Downtown Areas", Urban Land, Vol.33, No.5, May 1974, pp.3-9. Carpenter, Edward K. "Making Minneapolis Work", Design and Environment, Vol.6, No.2, Summer 1975, pp.33-45 Ching, Francis D. K. , Architecture: Form, Space and Order, Van Nostrand Reinhold Company Inc., New York 1979. 229 Clark, Roger H . , & Pause, Michael, Precedents in Architecture, Van Nostrand Reinhold Company Inc., New York 1985. Cottle, R.W., "The experience of cities in the improvement of the pedestrian environment". Greater-London Council Intelligence Unit Quarterly Bulletin, ' \ o . 2 l " , December 1972, pp.2 1 - 3 4 . " " Cranz, Galen, "The Sociology of Public Space", Design Quarterly, Vol.129, 1985, pp.22-24. Curran, Raymond J . , Architecture and the Urban Experience, Van Nostrand Reinhold Company Inc., New York 1983. Deane, Philip, Constantinos Doxiadis; Master Builder for Free Man, Dpbbs Ferry, Oceana Publications Inc., New York, 1965. DeVito, Mathias J . , "Retailing Plays Key Role in Downtown Renaissance", Journal of Housing, Vol.37, No.4, April 1980, pp. 197-200. Dillon, David, "Dallas: A Case Study in Skyway Economics", Design Quarterly, Vol.129, 1985, pp. 25-28. Downtown Council of Minneapolis, and City of Minneapolis Planing Department, Minneapolis Metro Center Metro 2000 Plan, January 1988. Dueker, Kenneth J . , Pendleton, James H . , and Rao, Rishinath L . , "Evaluation of Portland Transit Mall", Transportation Research Record, No.931, 1983, pp. 1-6. Etherington, A.Bruce, Streets in the Sky, Human Settlements Division, Asian Institute of Technology, Bangkok, Thailand 1983. Evenson, Norma, Le Corbusier : The Machine and the Grand Design, George Braziller Inc., New York, 1969. Ferriss, Hugh, The Metropolis of Tommorrow, I.Washburn, New York, 1962. Fishman, Robert, Urban Utopias in the Twentieth Century, New York, Basic Books, 1977. Francis, Mark, "Mapping Downtown Activity", Journal of Architectural and Planning Research, Vol.1, No.1, June 1984, pp.21-35. 230 Fraser, Jack, "Sidewalks in the Sky", Journal of the American Institute of Architects, Vol.54, Dec. 1970, pp.55-56. Fruin, John J . , "Pedway systems in urban centers", Civil Engineering, Vol.43, No.9, Sept 1973, pp.63-66. Fruin, John J . , Pedestrian Planning and Design, Metropolitan Association of Urban Designers and Environmental Planners, Inc., New York 1971. Galantay, E . Y . , Constandse, A . K . , and Obha, T., New Towns World Wide, The Hague, The Netherlands: International Federation of Housing and Planning, 1985. Gibbs-smith, Charles, and Rees, Gareth, The inventions of Leonardo Da Vinci, Phaidon Press, Oxford 1978. Goldfield, David R., "Historic Planning and Redevelopment in Minneapolis", Journal of American Institute of Planners, Vol.42, No . l , January 1976, pp.76-86. Gomez-Ibanez, Jose A. & Fauth, Gary R., "Downtown Auto Restraint Policies", Journal of Transportation Economics and Policy, Vol.14, No.2, May 1980, pp. 133-153. Grafton, John, New York in the Nineteenth Century, Dover Publications Inc., New York, 1977. Greater London Council, The planning of a New Town, 1965. Hall, Wayne Robert, Spatial Behavior in Victory Square: The Social Geography of an  Inner-city Park, A thesis submitted in the partial fulfilment of the requirements for the degree of Master of Arts in the Department of Geography, University of British Columbia, April 1974 (not published). Heglund, C. Todd, "Des Moines Skywalk System a", Transport Engineering Journal, Vol.108, Jan.1982, pp.1-13. Henderson, Arthur, "Pedestrianisation: Dos and Donts", Built Environment, Vol.3, No.7, July 1974, pp.345-347. Hiorns, Frederick R., and George, G. , Town Building in History, Harrap & Co. Ltd, London, 1956. 231 Irvin, Lawrence M . , and Groy, Jeffrey B., The Minneapolis Skyway System: What it is  and Why it Works, City of Minneapolis City Planning Department, January 1982. Jacob, Bernard, "Architectural Form in the Urban Setting", Design Quarterly, Vol.129, 1985, pp.29-31. Jones, Kay Bea, "Genus and Genius: An Architectural Morphology of urban Genoa", Journal of Architectural Education, Vol.43, No.4, Summer 1990, pp. 16-26. Karash, Karla, "Downtown Retail in Boston : Social Trends and Impacts of a Pedestrian Mall", Transportation Research Record, No.931, 1983, pp.7-11. Klein, Norman & Arensberg, Walter, "Auto-Free Zones: Giving Cities Back to People", City, Vol.6, No.2, March/April 1972, pp.45-52. Koehler, Cortus T. , "A Comparative Study of Three Suburban Malls: The Influence of Physical Environment on Pedestrian Behavior", Journal of Environmental Systems, Vol.3, No.1, Spring 1973, pp. 17-25. Krushelnicki, Bruce, "Shopping Centers: Public Agora or Privileged Place?", City Magazine, Fall 1985, pp.22-25. Larson, Daniel E . , and Solie, Robyn L . , "Minneapolis Skyway Retail Study", Towle Report, Second Quarter 1988, prepared by Towle Real Estate, Minneapolis. Levine, Marvin, Marchon, Iris, & Hanley, Gerard, "The Placement and Misplacement of You-Are-Here Maps", Environment and Behavior, Vol.16, No.2, March 1984, pp.139-157. Levinson, Herbert S., "Streets for People and Transit", Transportation Quarterly, Vol.40, No.4, Oct. 1986, pp.503-520. Llewellyn-Smith, Michael, "Towards a Model of Pedestrian Movement for the city of Sydney", Royal Australian Planning Institute Journal, Vol.10, No.4, October 1972, pp.149-152. Loukissas, Philippos J . , "Automobile Restricted Zones in Downtowns : Lessons from U M T A ' s Demonstration Program", Transportation Research Record, No.991, 1984, pp.22-31. Loukissas, Philoppos J . & Mann, Stuart H . , "Implementation of Downtown Automobile-Use Management Projects", Transportation Research Record, No. 1046, 1985, pp.76-84. Lynch, Kevin, A Theory of Good City Form, Cambridge, Massachusetts : MIT Press, '""'1981. May, Peter, "Feet and Pedals - few rights of way", Built Environment Quarterly, Vol.1, No.2, Sept.1975, pp.117-118. Minneapolis Skyway Advisory Committee, "Minneapolis Skyway System Design Standards", November 21st, 1985. Moreno, Elena Marcheso, "The Many Uses of Postoccupancy Evaluation", Architecture, Vol.78, No.4, April 1989, pp.119-121. Morphew, Carol "The Ins and Outs of Skyways", Planning, Vol.50, March 1984, p.23-26. Morphew, Carol, "The Ins and Outs of Skyways", Planning, Vol.50, March 1984, pp.23-26. Onibokun, Adepoju, "A Comprehensive Evaluation of Pedestrian Malls in the United States", The AppraisalJournal, Vol.43, No.2, April 1975, pp.202-218. P/A Inquiry, "Remaking Malls", Progressive Architecture, Vol.69, No. 12, November 1988, pp.96-101. Pastier, John, "To Live and Drive in L . A . " , Planning, Vol.51, No.2, Feb. 1986, p.21-25. Patricios, N . N . , "Human Aspects of Planning Shopping Centers", Environment and Behavior, Vol.11, No.4, December 1979, pp.511.538. Paumier, Ceril B., "Design Elements For a Downtown Marketplace", Journal of Housing, Vol.39, No.3, May/June 1982, pp.76-79. Planning, "Downtown Seven Good Bets", Planning, Vol.47, No.5, May 1981, pp.12-18. Podolske, Richard C. and Heglund C.Todd, "Skyways in Minneapolis/ St.Paul", Urban Land, Vol.35, No.8, September 1976, pp.3-12. 233 Ponte, Vincent, "Reflection on the Pedestrian System", Urban Design International, Fall 1986, pp. 28-33. Preiser, Wolfgang F . E . , "The Use of Ethological Methods In Environmental Analysis: A Case Study", Environmental .Design: Research and Practice. One. Proceedings of the EDRA3/AR8 Conference, University of California at Los Angeles, January 1972, Edited by William J . Mitchell, pp.3-2-1 to 3-2-10. Rapoport, Amos, Human Aspects of Urban Form, Pergamon Press, Printed in Great Britain by A.Wheaton & Co. Exeter, Published by Robert Maxwell 1977, M.C. . Reis, Donald, "Public-Private Cooperation in Developing an Underground Pedestrian System", Underground Space, Vol.6, No.6, May 1982, pp.337-341. Robertson, Jaquelin "Private Space in the Public Realm", Design Quaterly, Vol 129, 1985, pp.5-7. Robertson, Kent A . , "Designing Downtown Redevelopment Policy : The Problem of Knowledge", Journal of Architectural and Planning Research, Vol.2, June 1985, pp.129-139. Robertson, Kent A. , "Pedestrian Skywalk Systems: Downtown's Great Hope or Pathways to Ruin?", Transportation Quarterly, Vol.42, No.3, July 1988, pp.457-484. Robertson, Kent A . , "Pedestrian Skywalks in Calgary, Canada", Cities, August 1987, pp.207-214. Robertson, Kent A. , "Pedestrian Skywalks: The Esthetic Impact on Downtown", Landscape, Vol.29, Spring 1987, pp.43-47. Rowe, Colin, "I Stood in Venice on the Bridge of Sigh", Design Quarterly, Vol.129, 1985, pp.8-15. Rudofsky, Bernard, Streets for People; a Primer for Americans, Doubleday and Company, Inc, Garden City, New York, 1969. Sands, Peter, & Ciolek, T. Matthew, "Computer Simulation and the Study of Spatial Behavior of Pedestrians: An Exploratory Example", Man-Environment Systems, Vol.9, No.6, 1979, pp.281-292. 234 Saxon, Richard, Atrium Buildings : Development and Design, London : The Architectural Press, 1983. Schlivek, Louis B., "Four Places Where Urban Design and Planning are Paying Dividents: Minneapolis", AlA Journal, Vol.64, No.2, August 1975, pp.32-40. Seneviratne, Prianka, and Fraser, Philip, "Issues Related to Planning for Pedestrian Needs in Central Business Districts", Transportation Research Record, Vol.1141, 1987, pp.7-14. Shastock, Hannah, City Center Pedestrian Systems: A State-of-the-art Review, a Working Paper prepared for the Urban Transport Research Branch of Canadian Surface Transport Administration, March 1978. Short, Dale, "Birmingham reclaims its downtown", Place, Vol.3, No.5, May 1983, pp.4-6. Smith, James, "Pedestrianisation - Shopping Streets in Scotland", Planner, Vol.71, No.5, May 1985, pp. 12-16. U.S. News & World Report, "How Shopping Malls are Changing Life in U.S.", June 18, 1973, pp.43-46. Uhlig, Klaus, Pedestrian Areas From Malls to Complete Networks, Academy Editions London , Verlag Gerd Hatje, Stuttgart, Germany 1979. Vernez-Moudon, Anne, and Lanconte, Pierre, Streets as Public Property: Opportunities  for Public/Private Interaction in Planning and Design, Proceedings of an International Institute held at the College of Architecture and Urban Planning, University of Washington, Seattle, Washington 1983, Moudon and Lanconte 1983. Villecco, Marguerite, "Urban Renewal Goes Underground", Architecture Plus, Vol.1, No.5, June 1973, pp.20-29. Warner, Sam Bass Jr. , "The Liberal City", Design Quarterly Vol.129, 1985, pp. 16-21. Webb, Eugene J . , Campbell, Donald T. , Schwartz, Richard D., and Sechrest, Lee, Unobtrusive Measures: Non-reactive Research in the Social Sciences, Rand McNally College Publishing Company, Chicago, U.S.A. 1966. 235 Weisbrod, Glen, "Business and Travel Impacts of Boston's Downtown Crossing Automobile-Restricted Zone", Transportation Research Record, No.882, 1982, pp.25-32. Whyte, William H . , City: Rediscovering the Center, New York, London: Doubleday, 1988. Whyte, William H . , The Social Life of Small L^rban Spaces, The Conservation Foundation, Washington D.C. 1980. Wiedenhoeft, Ronald, "Minneapolis: a Closer Look", Urban Land, Vol.34, No.9, Oct. 1975, pp. 8-17. Wilcox, David A. , "Attractions Management is Key to a Revitalized Downtown's Success", Journal of Housing, Vol.37, No.7, July 1980, pp.377-382. Wolff, Anthony, "Main Street Vs. the Suburban Mall", E.P.O., Vol.1, No.2, March/April 1979,.pp.l2-17. Zucker, Paul, Town and Square From Agora to the Village Green, New York, Columbia university Press, New York, 1959. ************** 236 APPENDIX I  B E H A V I O R A L D A T A RECORDING SHEETS HA.S.K THESE BEHAVIORAL OBSERVATION DATA RECORDING SHEET STANDARD DATA: " Date : Location : Time Period : Weather : UsER DATA: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Jtae - Child (<13. indep.1 Teenaeer (13-19) Adult (20-641 ----Elderly (pwr 65.1 = ™ = Sex - Male Female = = ---- - ~ = = Ethnicity - White Black Others — User Type - Workers Viators Tourists Security Staff M/P with Baby iXhera (vgr., ffla., etc) Physically Cane/Crutch Impared - Wheel Chair s = Grouping Alone 30 pattern - Sameser Couple MvP CoucJe Groups 13 cr more) Family t = Class - Lower Upper 21 26 ACTTvTrv' DATA : (MOBK ACTIVITY TYPED Grid location : 1 8 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 22 23 24 25 27 28 29 Access, Walloiut shoppies Prpmenadiiifl Tb Work. Working Recreating Searching fffinging out _ _ COMMENTS : i i . .. -Fig.l: The 'Behavioral Data Recording Sheet' used for recording details of Mobile Activity types. 237 M.AS.A THESIS BfEAVTBRAL OBSERVWN DATA RBCORDtNG SHEET STANDARD DATA: Date : Location: Time Period: Weather : USER DATA: 1 8 3 4 5 6 r 8 9 10 11 12 13 14 15 16 17 18 19 SO 21 2g g3 24 25 26 27 28 29 Ase - Child (<13, indep.) Teenaser (13-19) Adult (20-M) Elderly (over 65) Sex - Male Female = = = — — Ethnicity - White Black Others = User Type - Workers Waters Tourists Vendor* Street Entertainer Security Staff M/F with Baby Others regr., Hka, etc) Phyaicallv Cane/Crutch tapered - Wheel Chair = Grouping Alone pattern - Samesex Couple M/P Gourde Groups (3 or more) Family = = Class - Lower Upper 28 ACTIVITY DATA: (STATIONARY AmVTTY TYPE) Grid location : 1 g 3 4 5 6 7 8 9 10 11 18 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 29 Watchiia/Gironiiris Talking Eating Studing /Reading Performing Waiting Meeting Body Posture Sitting Lying rYnpplirw/f'irnTBtting . Fig.2: The 'Behavioral Data Recording Sheet' used for recording details of Stationary Activity types. 238 APPENDIX II C H A P T E R F O U R T A B L E S COUNT EXP VAL ROW PCT COL PCT SITE Skyway 1 AGE 1 N i c o l l e t N i c o l l e t Marquett Mall R Mall F e Avenue 00| 2.00| 3.00| 4.00| ROW TOTAL Ch i l d & 1 .00 Teen 14 11.4 34 . 1% 8.0% 14 13.0 34 . 1% 7.0% 9 13.0 22.0% 4.5% 4 3.7 9.8% 7.0% 41 6.5% Adul ts 2.00 133 145 . 1 25. 4% 76 .0% 174 165.8 33. 2% 87.0% 166 165.8 31.7% 8 3.0% 51 47.3 9 . 7% 89.5% 524 82.9% Elder ly 3 .00 28 18.6 4 1.8% 16.0% 12 21.2 17.9% 6.0% 25 21.2 37 . 3% 12.5% 2 6.0 3.0% 3.5% 67 10.6% COLUMN TOTAL 175 27 . 7% 200 31.6% 200 31.6% 57 9.0% 632 100.0% CHI-SOUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 15.83860 6 0.0146 3.698 1 OF 12 ( 8.3%) WITH AGE 1 WITH SITE STATISTIC SYMMETRIC DEPENDENT DEPENDENT LAMBDA SOMERS' 0.02963 -0.02203 0.00000 -0.01559 0.03704 -0.03751 STATISTIC VALUE SIGNIFICANCE KENDALL'S TAU B -0.02418 0.2512 KENDALL'S TAU C -0.01672 0.2512 GAMMA -0.05239 Table 1: Crosstabulation of Age Groups by Site, for the weekend data (for a re-categorized Age variable). 239 AGE Chi ld COUNT EXP VAL ROW PCT COL PCT SI TE Sk yway 1 N i c o l l e t Ma 1 1 R 2 ROW TOTAL 1 < 13 7 6.5 53.8% 2. 3% 6 6.5 46.2% 2.0% 13-2.2% Teen 2 13 to 13 11.5 56 . 5% 4 . 3% 10 11.5 4 3.5% 3.3% 23 3.8% Adult 3 20 to 246 249 .1 49 . 3% 82 . 3% 253 249.9 50. 7% 84 . 3% 499 83. 3% Elder ly 4 >64 33 31 .9 51.6% 11.0% 31 32. 1 48.4% 10.3% 64 10. 7% COLUMN TOTAL 299 49.9% 300 599 50.1% 100.0% CHI-SOUARE D F . SIGNIFICANCE MIN E F . CELLS WITH E . F . < 5 0.62726 3 0.8902 6.489 NONE WITH AGE WITH SITE STATISTIC SYMMETRIC DEPENDENT DEPENDENT LAMBDA 0.01504 0.00000 0.02007 Table 2: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Age groups, during weekday. 240 AGE C h i l d COUNT EXP VAL ROW PCT COL PCT SITE Skyway N i c o l l e t ROW M a l l R TOTAL •I 2| 1 <13 7 5.6 58. 3% 4.0% 5 6.4 4 1.7% 2.5% 12 3.2% Teen 2 13 to 7 7 . 5 4 3.8% 4.0% 9 8.5 56 . 3% 4.5% 16 4 . 3% A d u l t 3 20 to 133 143.3 43.3% 76 .0% 174 163.7 56. 7% 87.0% 307 81.9% E l d e r l y 4 >64 28 18.7 70.0% 16.0% 12 21.3 30.0% 6.0% 40 10. 7% COLUMN TOTAL 175 46. 7% 200 53.3% 375 100.0% CHI-SOUARE D F . 10.84042 STATISTIC SIGNIFICANCE 0.0126 MIN E.F. CELLS WITH E.F.< 5 SYMMETRIC 5.600 WITH AGE DEPENDENT NONE WITH SITE DEPENDENT LAMBDA 0.07407 0.00000 0. 10286 Table 3: Pairwise comparison of Skyway System and Mall(R) for the proportional representation of Age groups, during weekend. 241 AGE Ch i l d Teen Adul t COUNT EXP VAL ROW PCT COL PCT SITE 1 Skyway N i c o 1 l e t M a l l F 1| 2| 1 <13 2 13 to 3 20 to E l d e r l y >64 COLUMN TOTAL CHI - SQUARE O F . 6.96684 3 STATISTIC 7 4 . 5 7 7.8% 2.3% 2 4 . 5 22.2% . 7% 13 9.5 68 . 4% 4 . 3% 6 9.5 31.6% 2.0% 246 255.6 48.0% 82. 3% 266 256. 4 52.0% 88 . 7% 33 29 . 5 55 . 9% 11.0% 26 29 . 5 44 . 1% 8.7% 299 300 49.9% 50.1% ROW TOTAL 9 1 .5% SIGNIFICANCE 0.0730 • 19 3 . 2% 512 85.5% 59 9 . 8% 599 100.0% MIN E.F. CELLS WITH E.F.< 5 SYMMETRIC 4 . 492 WITH AGE DEPENDENT 2 OF 8 ( 25.0%) WITH SITE 1 DEPENDENT LAMBDA 0.04922 0.00000 0.06355 Table 4: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Age Groups, during weekdays. 242 AGE Chi Id COUNT EXP VAL ROW PCT COL PCT SI TE 1 Sk yway N i c o l l e t ROW M a l l F TOTAL 1| 2| 1 <13 7 4.2 77.6% 4.0% 2 4.8 22 . 2% 1 .0% 9 2.4% Teen 2 13 to 7 6.5 50.0% 4.0% 7 7.5 50 .0% 3.5% 14 3.7% Adul t 3 20 to 133 1 39 . 5 44 . 5% 76.0% 166 159 . 5 55 . 5% 8 3.0% 299 79 . 7% E l d e r l y 4 >64 28 24 . 7 52.8% 16.0% 25 28. 3 47.2% 12.5% 53 14.1% COLUMN TOTAL 175 46 . 7% 200 375 53.3% 100.0% CHI-SQUARE D.F. 4.94504 STATISTIC SIGNIFICANCE 0.1759 MIN E.F. SYMMETRIC 4.200 WITH AGE DEPENDENT CELLS WITH E.F.< 5 2 OF 8 ( 25.0%) WITH SITE 1 DEPENDENT LAMBDA 0.03187 0.00000 0.04571 Table 5: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Age groups, during weekend. 243 AGE Ch i l d Teen A d u l t COUNT EXP VAL ROW PCT COL PCT SITE2 Sk yway 1| Marquet t e avenue 21 ROW TOTAL t <13 2 13 t o 3 20 to E l d e r l y >64 COLUMN TOTAL 7 5.7 63.6% 2.3% 4 5.3 36 . 4% 1 . 4% 1 1 1 . 9% 13 7.2 92.9% 4 . 3% 1 6.8 7 . 1% . 4% 14 2.4% 246 259 . 4 48. 7% 82. 3% 259 245.6 51.3% 91.5% 505 86.8% 33 26 . 7 6 3.5% 11.0% 19 25. 3 36 . 5% 6.7% 52 8.9% 299 51.4% 283 48.6% 582 100.0% CHI-SQUARE O F . 14 . 77907 STATISTIC SIGNIFICANCE 0.0020 MIN E.F. CELLS WITH E .F.< 5 SYMMETRIC 5. 349 WITH AGE DEPENDENT NONE WITH SITE2 DEPENDENT LAMBDA 0 . 0.16 1 1 0.00000 0.04594 Table 6: Pairwise comparison of Skyway System anal Marquette Avenue for the proportional distribution of Age groups, during weekdays. 244 SITE2 COUNT EXP VAL Sk yway Marquet t ROW ROW PCT e avenue TOTAL COL PCT 1 AGE 1 7 1 8 Ch i Id <13 6 . 87 . 4 . 0 5% or. 2.0 12.5% 1 .8% 3.4% 2 7 3 10 Teen 13 to 7 . 70. 5 0% 2.5 30.0% 4.3% 4 0% 5.3% 3 133 51 184 A d u l t 20 to 138. 72. 76 . 8 3% 0% 45. 2 27 . 7% 89.5% 79. 3% 4 28 2 30 E l d e r l y >64 22 6 7.4 12.9% 93 3% 6 . 7% 16 0% 3.5% COLUMN 175 57 232 TOTAL 75 4% 24.6% 100.0% CHI-SOUARE D.F. 6.960 12 STATISTIC LAMBDA SIGNIFICANCE 0.0732 MIN E.F. SYMMETRIC 0.00000 1 . 966 WITH AGE DEPENDENT CELLS WITH E.F.< 5 2 OF 8 ( 25.0%) WITH SITE2 DEPENOENT 0.00000 0.00000 Table 7: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Age Groups, during weekend. 245 GENDER Ma le Female COUNT EXP VAL now per COL PCT SITE Skyway N i co 1 let Ma 1 1 R II 2| ROW TOTAL 1 COLUMN TOTAL 136 157 293 146 . 3 146 . 7 48 . 9% 46.4% 53.6% 45 . 5% 52. 3% 163 143 306 152. 7 153. 3 51.1% 53. 3% 46. 7% 54 . 5% 47.7% 299 300 599 49.9% 50 . 1% 100.0% CHI-SQUARE D.F . SIGNIFICANCE MIN E . F . CELLS WITH E .F .< 5 2.54326 2.81065 STATISTIC LAMBDA 0.1108 0.0936 SYMMETRIC 0.05743 146.255 NONE ( BEFORE YATES CORRECTION ) WITH GENDER DEPENDENT 0.04778 WITH SITE DEPENDENT 0.06689 Table 8: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Gender groups, during weekdays. GENDER COUNT EXP VAL ROW PCT COL PCT SITE Sk yway Nico l let Mal 1 R 1| 2| 1 84 104 Ma le 87 . 7 100. 3 44 . 7% 55. 3% 48.0% 52.0% 2 91 96 Fema1e 87 . 3 99. 7 48. 7% 51.3% 52.0% 48.0% + - + COLUMN 175 200 TOTAL 46 . 7% 53.3% ROW TOTAL 188 50. 1% 187 49.9% 375 100.0% CHI-SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 0 . 44805 0.59734 STATISTIC 0.5033 0 .4396 SYMMETRIC 87.267 NONE ( BEFORE YATES CORRECTION ) WITH GENDER DEPENDENT WITH SITE DEPENDENT LAMBDA 0.01934 0.03743 0.00000 Table 9: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Gender groups, during weekend. 246 GENDER Ma l e Fema1e COUNT EXP VAL S H E 1 Skyway N i c o l l e t ROW ROW PCT Ma 1 1 F TOTAL COL PCT 1 21 1 136 200 336 167 . 7 168 . 3 56 . 1% ..... .. . 40 . 5%... - -59.-5% . 45.5% 66 . 7% 2 163 100 263 131.3 131.7 4 3.9% 62.0% 38.0% 54.5% 33. 3% v + COLUMN 299 300 599 TOTAL 49.9% 50. 1% 100.0% CHI - SQUARE D.F . SIGNIFICANCE MIN E.F. CELLS WITH E.F.< 5 26.42687 27.28014 STATISTIC 0.0000 0.0000 SYMMETRIC 131.280 NONE ( BEFORE YATES CORRECTION WITH GENDER DEPENDENT WITH SITE1 DEPENDENT LAMBDA 0. 16014 0.10266 0.21070 Table 10: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Gender groups, during weekdays. GENDER Male Fema1e SI TE 1 COUNT EXP VAL Skyway N i c o l l e t ROW ROW PCT Ma 1 1 F TOTAL COL PCT 1 2| 1 84 122 206 96. 1 109.9 54.9% 40.8% 59. 2% 48.0% 6 1.0% 2 91 78 169 78.9 90. 1 45 . 1% 53.8% 46 . 2% 52.0% 39 .0% + + - ..._ + COLUMN 175 200 375 TOTAL 46 . 7% 53. 3% 100.0% CHI-SQUARE D.F. SIGNIFICANCE MIN E F . CELLS WITH E.F.< 5 5 . 85707 6.37136 STATISTIC 0.0155 0.01 16 SYMMETRIC 78.867 NONE ( BEFORE YATES CORRECTION WITH GENDER DEPENDENT WITH SITE 1 DEPENDENT LAMBDA 0.05814 0.04 142 0.07429 Table 11: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Gender groups, during weekend. 247 GENDER Ma le Female SITE2 COUNT EXP VAL Sk yway Marquet t ROW ROW PCT e avenue TOTAL COL PCT 1 2| 1 136 165 301 154.6 146.4 51.7% 4 5 2% 54 8% 45 . 5% 58 . 3% 2 163 118 281 144.4 136.6 48 . 3% 58 .0% 4 2.0% 54 . 5% 4 1.7% COLUMN 299 283 582 TOTAL 51.4% 48.6% 100.0% CHI - SQUARE D.F SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 9.06 132 1 9.56779 1 STATISTIC 0.0026 0 .0020 136.637 NONE BEFORE YATES CORRECTION ) SYMMETRIC WITH GENDER DEPENDENT WITH SITE2 DEPENDENT LAMBDA 0.09929 0.09609 0. 10247 Table 12: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Gender groups, during weekdays. SITE2 COUNT EXP VAL Skyway Marquett ROW ROW PCT e avenue TOTAL COL PCT 1| 2| GENDER 1 84 32 Ma le 87.5 28.5 72.4% 27 .6% 48 .0% 56. 1% 2 91 25 Female 87 .5 28.5 78 . 4% 21.6% 52 .0% 4 3.9% COLUMN 175 57 TOTAL 75 . 4% 24 . 6% 116 50.0% 1 16 50.0% 232 100.0% CHI - SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 0.83729 1 . 13965 STATISTIC LAMBDA 0.3602 0.2857 28.500 NONE ( BEFORE YATES CORRECTION ) SYMMETRIC 0.04046 WITH GENDER DEPENDENT 0.06034 WITH SITE2 DEPENDENT 0.00000 Table. 13: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Gender groups, during weekend. 248 COUNT EXP VAL SITE Sk yway N i c o l l e t N i c o l l e t Marquett ROW ETHNIC Wh i te ROW PCT COL PCT 1 .00 Ma 1 1 R Ma 1 1 F 2.00| 3.00 e Avenue TOTAL 4 .001 1 .00 Ma jor i t 254 263. 3 24 . 4% 84 .9% 266 264 . 2 25.6% 88. 7% 266 264 . 2 25.6% 88. 7% 255 249 . 2 24 . 5% 90. 1% 104 1 88. 1% Black 2.00 32 25.0 32. 3% 10. 7% 19 25. 1 19.2% 6.3% 25 . 25. 1 25.3% 8.3% 23 23.7 23.2% 8. 1% 99 8.4% Other 3.00 Minor i t 13 10.6 31.0% 4 . 3% 15 10.7 35.7% 5.0% 9 10.7 21.4% 3.0% 5 10. 1 11.9% 1 . 8% 42 3.6% COLUMN 299 300 300 283 1182 CHI-SQUARE 9.03467 TOTAL D.F . 25.3% 25.4% 25.4% 23.9% 100.0% SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 0. 1716 10.056 NONE STATISTIC LAMBDA SOMERS' D SYMMETRIC 0.01271 -0.04136 STATISTIC VALUE WITH ETHNIC DEPENDENT 0.00000 -0.02664 SIGNIFICANCE WITH SITE DEPENDENT 0.01474 -0.09245 KENDALL'S TAU B KENDALL'S TAU C GAMMA -0.04963 -0.02996 -0 . 12322 0.0293 0.0293 Table 14: Crosstabulation of Ethnic groups by Site, for the weekday data. 249 ETHNIC COUNT EXP VAL ROW PCT COL PCT SITE Sk yway N i c o l l e t N i c o l l e t Marquett ROW Mall R Mall F e Avenue TOTAL 1.001 2.00 | 3. 00 | 4. 001 White 1 .00 Ma jor i t 148 150. 1 27 . 3% 84 . 6% 168 17 1.5 31.0% ' 84.0% 177 171.5 32. 7% 88.5% 49 48.9 9.0% 86 .0% Black 2 .00 17 15.0 31.5% 9 . 7% 20 17.1 37 .0% 10.0% 12 17.1 22.2% 6.0% 5 4.9 9 . 3% 8 . 8% Other 3 .00 Minor i t 10 10.0 27.8% 5.7% 12 11.4 33 . 3% 6.0% 1 1 11.4 30 . 6% 5.5% 3 3.2 8.3% 5.3% COLUMN 175 TOTAL 27.7% 200 31.6% 200 31.6% 57 9.0% 542 85.8% 54 8 . 5% 36 5.7% 632 100.0% CHI - SQUARE 2 . 63632 D . F . STATISTIC SIGNIFICANCE 0.8529 SYMMETRIC MIN E . F . CELLS WITH E . F . < 5 3.247 2 OF 12 ( 16.7%) WITH ETHNIC DEPENDENT WITH SITE DEPENDENT LAMBDA SOMERS' D STATISTIC 0.01724 -0.03018 VALUE 0 .00000 -0.02045 SIGNIFICANCE 0.02083 -0.05757 KENDALL'S TAU B KENDALL'S TAU C GAMMA -0.03431 -0.02193 -0.08037 0.1713 0.1713 Table 15: Crosstabulation of Ethnic groups by Site, for the weekend data. 250 ROLE 1 V i s i tor Tour i •? t Workers Others COUNT EXP VAL ROW PCT COL PCT SITE Sk yway N i c o l l e t N i c o l l e t Marquett Ma 11 R Ma 11 F e Avenue 1.001 2.00| 3.001 4.001 1 .00 2.00 3.00 4 .00 COLUMN TOTAL 131 119.1 30 .57. 74 .9% 140 136 . 1 32 .6% 70 .0% 130 136 . 1 30.2% 65.0% 29 38.8 6 . 7% 50; 9% 21 31.8 18.3% 12.0% 37 36 . 4 32 . 2% 18.5% 38 36 . 4 33.0% 19.0% 19 10.4 16 . 5% 33. 3% 15 12.5 33. 3% 8.6% 13 14.2 28 . 9% 6 . 5% 17 14.2 37 .8% 8.5% 0 4 . 1 .0% .0% 8 11.6 19.0% 4.6% 10 13.3 23.8% 5.0% 15 13.3 35. 7% 7 . 5% 9 3.8 21.4% 15.8% 175 200 200 57 27.7% 31.6% 31.6% 9.0% ROW TOTAL 4 30 68.0% 1 15 18.2% 45 7 . 1% 42 6.6% 632 100.0% CHI - SQUARE 29.55674 D.F . STATISTIC LAMBDA SOMERS' D STATISTIC KENDALL'S TAU B KENDALL'S TAU C GAMMA SIGNIFICANCE 0.0005 MIN E . F . SYMMETRIC 0.01577 0 . 10780 VALUE 0. 10964 0.08692 0.18220 3. 788 WITH ROLE 1 DEPENDENT CELLS WITH E . F . < 5 2 OF 16 ( 12.5%) WITH SITE DEPENDENT 0.00000 0.09119 SIGNIFICANCE 0 .0009 0.0009 0.02315 0. 13183 Table 16: Crosstabulation of Role types by Site for the weekend data (for a re-categorized Role variable). 251 COUNT EXP VAL ROW PCT COL PCT SITE Sk yway 1 ROLE V i s i tor Tour i s t Secur i t y Guard i Other Worker Not Determined COLUMN TOTAL N i c o l l e t Ma 11 R I 2| 103 114.3 4 5.0% 34 .4% 126 114.7 55.0% 42.0% 32 31.9 50.0% 10. 7% 4 2.5 80 .0% 1 . 3% 32 32. 1 50.0% 10. 7% 1 2.5 20.0% . 3% 142 133.8 53.0% 47.5% 126 134.2 47.0% 42.0% 18 16 . 5 54 . 5% 6.0% 15 16.5 45.5% 5.0% 299 49 .9% 300 50. 1% ROW TOTAL 229 38.2% 64 10. 7% 5 .8% 268 44 . 7% 33 5.5% 599 100.0% CHI•SQUARE D . F . 5.33634 STATISTIC SIGNIFICANCE 0.2545 MIN E . F . CELLS WITH E . F . < 5 SYMMETRIC 2. 496 WITH ROLE DEPENDENT 2 OF 10 ( 20.0%) WITH SITE DEPENDENT LAMBDA 0.03492 0.00000 0.07358 Table 17: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Role types , during weekdays. 252 COUNT EXP VAL ROW PCT COL PCT SITE Sk yway 1 ROLE V i s i t o r Tour i s t Secur i ty Guard Other Worker c Not Determined COLUMN TOTAL N i c o l l e t Ma 1 1 R 1| 2| CHI - SQUARE 6.3364 1 D . F . 131 126.5 48. 37. 74 .97. 140 144.5 51 . 77. 70.0% 21 27 . 1 36 . 2% 12.07. 37 30.9 63.8% 18.5% 3 1 . 4 100.07. 1 . 77. 0 1.6 .07. .0% 12 11.7 48 .07. 6.97. 13 13.3 52.0% 6.5% 8 8.4 44 . 47. 4.67. 10 9.6 55.6% 5.0% 175 46 . 7% 200 53. 3% SIGNIFICANCE 0.1754 ROW TOTAL 27 1 7 2.3% 58 15.5% 3 .8% 25 6.7% 18 4.8% 375 100.0% MIN E . F . STATISTIC SYMMETRIC 1 .400 WITH ROLE DEPENDENT CELLS WITH E . F . < 5 2 OF 10 ( 20.0%) WITH SITE DEPENDENT LAMBDA 0.01075 0.00000 0.01714 Table 18: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Role types, during weekend. 253 COUNT EXP VAL ROW PCT COL PCT ROLE V i s i tor Tour i s t SI TE 1 Sk yway Nico l let Ma 11 F 1| 2| 1 Secur i t y Guard Other Worker Not Determined COLUMN TOTAL 103 97 . 8 52.6% 34 . 4% 93 98. 2 4 7.4% 31.0% 32 28.5 56 . 1% 10. 7% 25 28.5 43.9% 8.3% 4 4.5 44 . 4% 1 . 3% 5 4.5 55 . 6% 1 . 7% 142 147.3 48 . 1% 4 7.5% 153 147 . 7 51.9% 51.0% 18 21.0 42.9% 6.0% 24 21.0 57 . 1% 8.0% 299 49 .9% 300 50. 1% ROW TOTAL 196 32 . 7% 57 9.5% 9 1 .5% 295 49 . 2% 42 7.0% 599 100.0% CHI-SOUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 2.74662 0.601 1 4 . 492 2 OF 10 ( 20.0%) STATISTIC SYMMETRIC WITH ROLE DEPENDENT WITH SITE 1 DEPENDENT LAMBDA 0.02819 0.00000 0.05686 Table 19: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Role types, during weekday. 254 COUNT EXP VAL ROW PCT COL PCT 1 ROLE V i s i tor Tour i s t Secur i t y Guard i Other Worker c Not Determined COLUMN TOTAL SI TE 1 Skyway N i co1 let Ma 1 1 F 1| 2| 131 121.8 50.2% 74 . 9% 21 27 . 5 35.6% 12.0% 3 1 . 4 100.0% 1 . 7% 12 13.5 4 1.4% 6.9% 8 10.7 34 . 8% 4.6% 130 139.2 49 . 8% 65.0% 38 31.5 64 . 4% 19.0% 0 1 .6 .0% .0% 17 15.5 58 .6% 8.5% 15 12. 3 65. 2% 7 . 5% 175 46 . 7% 200 53. 3% ROW TOTAL 261 69 .6% 59 15.7% 3 . 8% 29 7 . 7% 23 6 . 1% 375 100.0% CHI-SOUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 9.26917 STATISTIC 0.0547 SYMMETRIC 1 .400 WITH ROLE DEPENDENT 2 OF 10 ( 20.0%) WITH SITE1 DEPENDENT LAMBDA 0.01384 0 .00000 0.02286 Table 20: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Role types, during weekend. 255 SITE2 COUNT EXP VAL ROW PCT COL PCT ROLE 1 Vis 1 tor 2 Tour i s t 3 Secur i t y Guard 4 Other Worker 5 Not Determined CHI - SQUARE 16.21594 D.F . COLUMN TOTAL Sk yway Marquet t ROW e avenue TOTAL 1 2 f 103 79 182 93.5 88.5 31.3% 56 .6% 43.4% 34 . 4% 27.9% 32 15 47 24. 1 22.9 8 . 1% 68. 1% 31.9% 10. 7% 5.3% 4 1 5 2.6 2.4 .9% 80 .0% 20.0% 1 . 3% .4% 142 177 319 163.9 155. 1 54 . 8% 44 . 5% 55.5% 47.5% 62.5% 18 1 1 29 14.9 14.1 5.0% 62 . 1% 37 .9% 6.0% 3.9% 299 283 582 51.4% 48.6% 100.0% SIGNIFICANCE MIN E F . 0.0027 STATISTIC SYMMETRIC 2.431 WITH ROLE DEPENDENT CELLS WITH E . F . < 5 2 OF 10 ( 20.0%) WITH SITE2 DEPENDENT LAMBDA 0.06410 0.00000 0. 12367 Table 21: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Role types, during weekdays. 256 COUNT EXP VAL ROW PCT COL PCT ROLE 1 V is i tor Tour i s t Workers Other s SITE 1 Sk yway Marquet t e Avenue 1| 2| 1 131 29 120. 7 39 . 3 81.9% 18.1% 74.9% 50. 9% 2 21 19 30. 2 9.8 52.5% 47.5% 12.0% 33. 3% 3 15 0 11.3 3.7 100.0% .0% 8.6% .0% 4 8 9 12.8 4 . 2 47.1% 52.9% 4.6% 15.8% COLUMN 175 57 TOTAL 75.4% 24.6% CHI-SQUARE D . F . 27.20415 3 STATISTIC ROW TOTAL 160 69.0% 40 17.2% 15 6.5% 17 7.3% 232 100.0% SIGNIFICANCE 0.0000 SYMMETRIC MIN E . F . 3.685 CELLS WITH E . F . < 5 2 OF LAMBDA SOMERS' D SI A I 1STIC KENDALL'S TAU B KENDALL'S TAU C GAMMA 0.007 7 5 0. 19792 VALUE 0. 19971 0 . 16937 0. 39947 WITH ROLE 1 DEPENDENT 0.00000 0.22847 SIGNIFICANCE 0.0008 0.0008 8 ( 25.0%) WITH SITE1 DEPENDENT 0.01754 0. 17457 Table 22: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Role types, during weekend (for a re-categorized Role variable). 257 COUNT EXP VAL ROW PCT COL PCT SITE Skyway 1 GROUPING A lone 2 Two. Same Sex 3 Two, Man & Woman 4 3 or More Adults 5 Fam 1 1 y . Chi ldren COLUMN TOTAL 1| N i c o l l e t ROW Mall R TOTAL 2| CHI-SQUARE 8 . 6 1505 D . F . STATISTIC 204 213.6 47.7% 68.2% 224 214.4 52. 3% 74 . 7% 36 33.4 53.7% 12.0% 31 33.6 46 . 3% 10.3% 16 19.0 42. 1% 5.4% 22 19.0 57.9% 7 . 3% 20 14.5 69 .0% 6.7% 9 14.5 31.0% 3.0% 23 18.5 62.2% 7 . 7% 14 18.5 37 .8% 4. 7% 299 300 49.9% 50.1% SIGNIFICANCE 0.07 15 428 71.5% 67 11.2% 38 6 . 3% 29 4.8% 37 6.2% 599 100.0% MIN E . F . 14.476 CELLS WITH E . F . < 5 NONE LAMBDA SYMMETRIC 0.05319 WITH GROUPING DEPENDENT 0.00000 WITH SITE DEPENDENT 0.08361 Table 23: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Grouping types, during weekdays. 258 COUNT EXP VAL ROW PCT COL PCT SITE Sk yway N ico1 let Ma 1 1 R H 21 1 GROUPING A lone 2 Two, Same Sex 3 Two, Man & Woman 4 3 or More Adults 5 Fam i 1 y . Chi ldren COLUMN TOTAL 87 88.2 46 .0% 49 . 7% 102 100.8 54 .0% 51.0% 21 20.5 47.7% 12.0% 23 23.5 52. 3% 11.5% 24 28.9 38. 7% 13.7% 38 33. 1 61.3% 19.0% 1 7 14.0 56 . 7% 9 . 7% 13 16.0 4 3.3% 6 . 5% 26 23.3 52.0% 14.9% 24 26. 7 48 .0% 12.0% 175 200 46.7% 53.3% ROW TOTAL 189 50. 4% 44 11.7% 62 16.5% 30 8.0% 50 13.3% 375 100.0% CHI-SQUARE D . F . 3.40447 STATISTIC SIGNIFICANCE 0.4926 MIN E . F . 14.000 CELLS WITH E . F . < 5 NONE SYMMETRIC WITH GROUPING DEPENDENT WITH SITE DEPENDENT LAMBDA 0.01662 0.00000 0.03429 Table 24: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Grouping types, during weekend. 259 COUNT EXP VAL ROW PCT COL PCT 1 GROUPING A lone 2 Two, Same Sex 3 Two, Man & Woman 4 3 or More Adults 5 Fam i1y, C h i l d r e n COLUMN TOTAL SITE! Sk yway N ico 1 let Ma 1 1 F 1| 2| CHI - SQUARE 17.17437 D.F . 204 219.1 46 . 5% 68.2% 235 219.9 53.5% 78.3% 36 32.9 54.5% 12.0% 30 33. 1 45.5% 10.0% 16 17.5 45 . 7% 5 . 4% 19 17.5 54 . 3% 6 . 3% 20 15.5 64 .5% 6.7% 1 1 15.5 35.5% 3.7% 23 14.0 82. 1% 7.7% 5 14.0 17.9% 1.7% 299 300 49.9% 50.1% SIGNIFICANCE ROW TOTAL 439 7 3.3% 66 11.0% 35 5.8% 31 5.2% 28 4.7% 599 100.0% 0.0018 MIN E . F . 13.977 CELLS WITH E . F . < 5 NONE STATISTIC SYMMETRIC WITH GROUPING OEPENDENT WITH SITE 1 OEPENDENT LAMBDA 0 .07 190 0.00000 0.11037 Table 25: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Grouping types, during weekdays. 260 COUNT EXP VAL ROW PCT COL PCT SI TE 1 Sk yway 1 GROUPING A lone Two, Same Sex Two, Man & Woman 3 or More Adu11 s Fami 1 y, Chi ldren COLUMN TOTAL N i co 1 let Ma 1 1 F 1| 2| 87 104 . 5 38.8% 49 . 7% 137 119.5 61.2% 68 . 5% 21 15.9 61.8% 12.0% 13 18. 1 38 . 2% 6 . 5% 24 24 . 3 46 . 2% 13.7% 28 27 . 7 53.8% 14.0% 17 12.6 6 3.0% 9 . 7% 10 14.4 37.0% 5.0% 26 17.7 68 . 4% 14.9% 12 20. 3 31.6% 6.0% 175 46 . 7% 200 53. 3% ROW TOTAL 224 59 . 7% 34 9. 1% 52 13.9% 27 7 . 2% 38 10 . 1% 375 100.0% CHI-SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 18 . 74010 STATISTIC LAMBDA 0.0009 12.600 NONE SYMMETRIC 0.08896 WITH GROUPING DEPENDENT 0.00000 WITH SITE 1 DEPENDENT 0. 16571 Table 26: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Grouping types, during weekend. 261 COUNT EXP VAL ROW PCT COL PCT SITE2 Sk yway 1 GROUPING A lone 2 Two, Same Sex 3 Two, Man & Woman 4 3 or More Adults 5 Fam i 1 y , Chi ldren COLUMN TOTAL Marquet t e avenue 1| 2| CHI SQUARE 22.94636 D.F . 204 228 221 .9 210. 1 4 7.2?. 52.8% 68 . 2% 80.6% 36 27 32.4 30 . 6 57 . 1% 42 . 9% 12.0% 9 . 5% 16 18 17.5 16 . 5 47.1% 52.9% 5.4% 6 . 4% 20 4 12.3 11.7 83. 3% 16.7% 6 . 7% 1 . 4% 23 6 14.9 14.1 79. 3% 20. 7% 7 . 7% 2. 1% + + 299 283 51.4% 48.6% SIGNIFICANCE 0.0001 ROW TOTAL 432 74.2% 63 10.8% 34 5.8% 24 4. 1% 29 5.0% 582 100.0% MIN E F . 11.670 CELLS WITH E . F . < 5 NONE STATISTIC SYMMETRIC WITH GROUPING DEPENDENT WITH SITES DEPENDENT LAMBDA 0.06005 0.00000 0.09187 Table 27: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Grouping types, during weekdays. 262 COUNT EXP VAL ROW PCT COL PCT SITE? Skyway Marquett e avenue l l 21 1 GROUPING A lone 2 Two. Same Sex 3 Two, Man & Woman 4 3 or More Adu 11s 5 Fami 1 y. Chi ldren COLUMN TOTAL 87 86 . 7 75. 7% 49. 7% 28 28. 3 24 . 3% 49. 1% 21 20. 4 77.8% 12.0% 6 6.6 22 . 2% 10.5% 24 28. 7 63.2% 13.7% 14 9 . 3 36 . 8% 24 . 6% 1 7 13.6 94 . 4% 9. 7% 1 4 . 4 5.6% 1 . 8% 26 25.6 76 . 5% 14.9% 8 8.4 23.5% 14.0% 175 57 75.4% 24.6% ROW TOTAL 1 15 49 .6% 27 11.6% 38 16 . 4% 18 7.8% 34 14.7% 232 100.0% CHI-SQUARE D . F . 6.70286 STATISTIC SIGNIFICANCE 0.1524 MIN E . F . 4. 422 CELLS WITH E . F . < 5 SYMMETRIC 1 OF WITH GROUPING DEPENDENT 10 ( 10.0%) WITH SITE2 DEPENDENT L AMBDA 0.00000 0.00000 0.00000 Table 28: Pairwise comparison of Skyway System and Marquette proportional distribution of Grouping types, during weekend. Avenue for the 263 SITE COUNT EXP VAL ROW PCT COL PCT PURPOSE 1 1 .00 Access/Work ing Sk yway Nico l let Mai 1 R 1.00| 2.00| N i c o l l e t Marquett Ma 1 1 F e Avenue 3.00| 4.00| Shopp ing 2 .00 3.00 Promenad ing/Sear 4 .00 Recrea t i ng 5 .00 Hanging Out COLUMN TOTAL 52 46 . 2 31.1% 29 . 7% 47 52.8 28. 1% 23.5% 44 52.8 26 . 3% 22.0% 24 15. 1 14.4% 42. 1% 59 47 .6 34 . 3% 33. 7% 68 54.4 39.5% 34 .0% 35 54 .4 20. 3% 17.5% 10 15.5 5.8% 17.5% 61 62.0 27.2% 34.9% 62 70.9 27 . 7% 31.0% 84 70.9 37 . 5% 42.0% 17 20. 2 7.6% 29.8% 0 14.1 .0% .0% 16 16 . 1 31.4% 8.0% 32 16. 1 62.7% 16.0% 3 4.6 5.9% 5.3% 3 5.0 16.7% 1 . 7% 7 5.7 38.9% 3.5% 5 5.7 27 . 8% 2.5% 3 1 .6 16.7% 5.3% 175 27 . 7% 200 31.6% 200 31.6% 57 9.0% ROW TOTAL 167 26 . 4% 172 27.2% 224 35.4% 51 8. 1% 18 2.8H 632 100.0% CHI - SQUARE 59.81581 D . F . 12 STATISTIC LAMBDA SOMERS' D SIGNIFICANCE 0.0000 SYMMETRIC 0.06667 0.10608 MIN E . F . CELLS WITH E . F . < 5 1.623 3 OF 20 ( 15.0%) WITH PURPOSE 1 DEPENDENT 0.03186 0.10669 WITH SITE DEPENDENT 0.09954 0. 10547 STATISTIC KENDALL'S TAU B KENDALL'S TAU C GAMMA VALUE 0. 10608 0. 10170 0.14639 SIGNIFICANCE 0.0008 0.0008 Table 29: Crosstabulation of trip Purpose by Site, for weekend data (for a re-categorized Purpose variable). 264 COUNT EXP VAL ROW PCT COL PCT SITE Skyway PURPOSE N i c o l l e t Mai 1 R 1| 2| 1 Access, Walk ing Shopp ing Promenad ing To Work, Work ing Recreat ing Search ing Hanging Out 120 127 . 3 47 . 1% 40. 1% 135 127 . 7 52 . 9% 45.0% 86 78.9 54 . 4% 28.8% 72 79. 1 45.6% 24 .0% 62 65.9 4 7.0% 20. 7% 70 66 . 1 53.0% 23. 3% 12 7.0 85. 7% 4.0% 2 7.0 14.3% . 7% 0 7.0 .0% .0% 14 7.0 100.0% 4 . 7% 10 5.0 100.0% 3.3% 0 5.0 .0% .0% 9 8.0 56 . 3% 3.0% 7 8.0 4 3.8% 2.3% COLUMN TOTAL 299 49 . 9% 300 50. 1% ROW TOTAL 255 42.6% 158 26 . 4% 132 22.0% 14 2.3% 14 2.3% 10 1 . 7% 16 2.7% 599 100.0% CHI-SQUARE 33.99899 D.F . STATISTIC LAM8DA SIGNIFICANCE 0.0000 SYMMETRIC 0.05599 MIN E F . CELLS WITH E . F . < 5 4 .992 1 OF 14 ( 7.1%) WITH PURPOSE DEPENDENT 0.00000 WITH SITE DEPENDENT 0.12040 Table 30: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Purpose types, during weekday. 265 COUNT EXP VAL ROW PCT COL PCT PURPOSE 1 SITE Sk yway Nico1 let Mal 1 R 1| 2| 1 Access/Work ing Shopp ing Promenad i ng/Sear Recreat ing Hanging Out 52 46 . 2 52. 5% 29. 7% 47 52 . 8 4 7.5% 23.5% 59 59 . 3 46.5% 33 . 7% 68 67 . 7 53.5% 34 .0% 6 t 57 . 4 49 . 6% 34.9% 62 65.6 50 . 4% 31.0% 0 7.5 .0% .0% 16 8.5 100.0% 8.0% 3 4 . 7 30.0% 1 . 7% 7 5.3 70 .0% 3.5% COLUMN TOTAL 175 46. 7% 200 53. 3% ROW TOTAL 99 26.4% 127 33.9% 123 32.8% 16 4 . 3% 10 2.7% 375 100.0% CHI - SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 16.90C92 STATISTIC 0 .0020 SYMMETRIC 4 .667 1 OF WITH PURPOSE 1 DEPENDENT 10 ( 10.0%) WITH SITE DEPENDENT LAMBDA 0.01655 0.00806 0.02857 Table 31: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Purpose types, during weekend (for a re-categorized Purpose variable). 266 COUNT EXP VAL ROW PCT COL PCT SITE 1 Sk yway PURPOSE Nico l let Ma 1 1 F 1| 2| - + + 1 Access , Wa1k ing Shopp ing Promenad i ng To Work. Work ing Recrea t ing Search ing Hanging Out 120 133. 3 44 .9% 40. 1% 147 133. 7 55. 1% 49 .0% 86 62.9 68. 3% 28.8% 40 63. 1 31.7% 13.3% 62 65.9 4 7.0% 20 . 7% 70 66 . 1 53.0% 23. 3% 12 11.5 52.2% 4.0% 1 1 11.5 47 .8% 3.7% 0 14.5 .0% .0% 29 14.5 100.0% 9 . 7% 10 5.0 100.0% 3.3% 0 5.0 .0% .0% 9 6.0 75.0% 3.0% 3 6.0 25.0% 1 .0% COLUMN TOTAL 299 49.9% 300 50. 1% ROW TOTAL 267 44 .6% 126 21.0% 132 22.0% 23 3.8% 29 4.8% 10 1 . 7% 12 2.0% 599 100.0% CHI-SQUARE 62.05080 D . F . STATISTIC LAMBDA SIGNIFICANCE 0.0000 SYMMETRIC 0.09984 MIN E . F . CELLS WITH E . F . < 5 4.992 1 OF 14 ( 7.1%) WITH PURPOSE DEPENDENT 0.00000 WITH SITE 1 DEPENDENT 0.21070 Table 32: Pairwise comparison of Skyway System and Mall(F) for distribution of Purpose types, during weekdays. the proportional 267 COUNT EXP VAL ROW PCT COL PCT PURPOSE 1 SI TE 1 Skyway 1| Nico 1 let Mal 1 F 21 1 Access/Work ing Shopp i ng Promenad ing/Sear Recreat ing Hanging Out 52 44.8 54 . 2% 29. 7% 44 51.2 45.8% 22.0% 59 43.9 62. 87. 33 . 7% 35 50. 1 37.2% 17.5% 61 67 . 7 4 2.1% 34 . 9% 84 77.3 57 .9% 42.0% 0 14.9 .0% .0% 32 17.1 100.0% 16.0% 3 3.7 37.5% 1 . 7% 5 4.3 62.5% 2.5% COLUMN TOTAL 175 46 . 7% 200 53. 3% ROW TOTAL 96 25 . 6% 94 25. 1% 145 38. 7% 32 8.5% 8 2. 1% 375 100.0% CHI - S Q U A R E 41.46021 D . F . STATISTIC LAMBDA S I G N I F I C A N C E 0.0000 MIN E . F . CELLS WITH E . F . < 5 3.733 2 OF 10 ( 20.0%) SYMMETRIC 0.07901 WITH PURPOSE 1 DEPENDENT ) 0.00000 WITH SITE1 DEPENDENT 0.18286 Table 33: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Purpose types, during weekend (for a re-categorized Purpose variable). 268 SITE2 COUNT EXP VAL ROW PCT COL PCT 1 Wa Ik ing PURPOSE Access, Shopp i ng 3 Promenad ing 4 To Work. Work ing 5 Recreat ing Search ing Hanging Out Skyway Marquet t ROW e avenue TOTAL 1 2| 120 195 315 161 .8 153. 2 54. 1% 38. 1% 61.9% 40. 1% 68 . 9% 86 28 114 58.6 55. 4 19.6% 75.4% 24.6% 28.8% 9.9% + + 62 38 100 51.4 48.6 17.2% 62.0% 38.0% 20. 7% 13.4% + + 12 12 24 12.3 11.7 4. 1% 50.0% 50.0% 4.0% 4 . 2% 0 2.6 .0% .0% 5 2.4 100.0% 1 .8% 10 5.7 90. 9% 3.3% 1 3 1% 4% 7 9 4 O 6 . 7 6.3 69.2% 30. 8% 3.0% 1 . 4% COLUMN 299 283 TOTAL 51.4% 48.6% 5 .9% 1 1 1 .9% 13 2.2% 582 100.0% CHI SQUARE 67.02343 D . F . STATISTIC LAMBDA SIGNIFICANCE 0.0000 SYMMETRIC 0 . 14545 MIN E F . CELLS WITH E . F . < 5 2.431 2 OF 14 ( 14.3%) WITH PURPOSE DEPENDENT 0.00000 WITH SITE2 DEPENDENT 0. 28269 Table 34: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Purpose types, during weekdays. 269 COUNT EXP VAL ROW PCT COL PCT SITE2 Skyway PURPOSE 1 Marque11 e avenue 1| 2| 1 52 24 Access/Working 57 . 3 18.7 68 . 4% 31.6% 29 . 7% 42. 1% 2 59 10 Shopp ing 52.0 17.0 85.5% 14.5% 33. 7% 17.5% 3 61 20 M i see 1 1aneous 61.1 19.9 75. 3% 24 . 7% 34 .9% 35. 1% Hanging Out 4 3 3 O 4 . 5 1 . 5 50 .0% 50.0% 1 . 7% 5.3% COLUMN 175 57 TOTAL 75 . 4% 24 .6% ROW TOTAL 76 32.8% 69 29 . 7% 81 34.9% 6 2.6% 232 100.0% CHI - SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 7.88978 STATISTIC 0.0483 SYMMETRIC 1.474 2 OF WITH PURPOSE 1 DEPENDENT 8 ( 25.0%) WITH SITE2 DEPENDENT LAMBDA 0.01923 0.02649 0.00000 Table 35: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Purpose types, during weekend (for a re-categorized Purpose variable). 270 COUNT EXP VAL ROW PCT COL PCT ACT IVE 1 1 .00 Watch ing/Groom in SITE Skyway 1 .001 N i c o l l e t N i c o l l e t Marquett Ma 11 R Ma 11 F e Avenue 2.00| 3.00| 2.00 Talk ing 3.00 M i see 1 1aneous M e e t i n g 4 .00 COLUMN TOTAL 69 27 .9% 20 8. 1% 117 47 . 4% 4.00 | 22 17 . 3 35. 5% 31.9% 3 5.0 4.8% 15.0% 28 29.4 45.2% 23.9% 9 10. 3 14.5% 22.0% 23 31.0 20. 7% 33 . 3% 13 9.0 11.7% 65 .0% 60 52.6 54 . 1% 51.3% 15 18 . 4 13.5% 36.6% 15 16 . 2 25.9% 21.7% 1 4 . 7 1 . 7% 5.0% 27 27.5 46 .6% 23. 1% 15 9.6 25.9% 36 .6% 9 4 . 5 56 . 3% 13.0% 3 1 . 3 18.8% 15.0% 2 7.6 12.5% 1 . 7% 2 2.7 12.5% 4.9% 41 16.6% ROW TOTAL 62 25. 1% 1 1 1 44.9% 58 23. 5% 16 6.5% 247 100.0% CHI - SQUARE D . F . 24 .95450 S T A T I S T I C SIGNIFICANCE 0.0030 MIN E . F . CELLS WITH E . F . < 5 1.296 4 OF 16 ( 25.0%) SYMMETRIC WITH ACTIVE 1 DEPENDENT WITH SITE DEPENDENT L A M B D A 0.02632 0.00000 0.05385 Table 36: Crosstabulation of Activity by Site, for weekday data (for a re-categorized Activity variable). 271 SITE COUNT EXP VAL Skyway N i c o l l e t N i c o l l e t Marquett ROW ROW PCT Mall R Ma 11 F e Avenue TOTAL COL PCT 1.00| 2.00| 3.00| 4.00| ACT I VE 1 + + + + + 1 .00 32 10 10 2 54 Watch ing/Groom in 20. 4 15.0 16.2 2 3 38.6% 59. 3% 18.5% 18.5% 3 7% 60.4% 25.6% 23.8% 33 3% 2.00 14 15 16 3 48 Talk ing 18.2 13.4 14.4 2 1 34.3% 29.2% 31.3% 33.3% 6 3% 26 . 4% 38.5% • 38. 1% 50 0% 3 .00 7 14 16 1 38 Misce l laneous 14.4 10.6 11.4 1 6 27. 1% 18.4% 36.8% 42. 1% 2 6% 13.2% 35.9% 38. 1% 16 7% COLUMN 53 39 42 6 140 TOTAL 37 .9% 27 .9% 30.0% 4 3% 100.0% CHI-SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 19.39746 6 0.0035 1.629 3 OF 12 ( 25.0%) WITH ACTIVE1 WITH SITE STATISTIC SYMMETRIC DEPENDENT DEPENDENT LAMBDA 0.13295 0.13953 0.12644 Table 37: Crosstabulation of Activity by Site, for weekend data (for a re-categorized Activity variable). 272 COUNT EXP VAL ROW PCT COL PCT ACT IVE 1 SITE Skyway N i c o l l e t ROW Mall R TOTAL 1| 2| 1 Watch i ng/Groom in Talk ing M i see 11aneous Meet ing COLUMN TOTAL 22 19.4 88.0% 31.9% 23 27.9 6 3.9% 33. 3% 15 12.4 93.8% 21.7% 9 9 . 3 75.0% 13.0% 69 7 7.5% 3 5.6 12.0% 15.0% 13 8. 1 36 . 1% 65.0% 1 3.6 6 . 3% 5.0% 3 2.7 25.0% 15.0% 25 28. 1% 36 40. 4% 16 18.0% 12 13.5% 20 89 22.5% 100.0% CHI-SOUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 7.87831 STATISTIC 0.0486 2.697 2 OF SYMMETRIC WITH ACTIVE1 DEPENDENT 8 ( 25.0%) WITH SITE DEPENDENT LAMBDA 0.00000 0.00000 0.00000 Table 38: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Activity types, during weekdays (for a re-categorized Activity variable). 273 COUNT EXP VAL ROW PCT COL PCT ACT I VE 1 SITE Skyway N i c o l l e t ROW Mal l R TOTAL 1| 2| 1 32 10 Watch ing/Groom in 24 . 2 17.8 76.2% 23.8% 60.4% 25.6% 2 14 15 Ta lk ing 16 . 7 12.3 48 . 3% 51.7% 26 . 4% 38.5% 3 7 14 Others 12. 1 8.9 33. 3% 66 . 7% 13.2% 35.9% COLUMN 53 39 TOTAL 57.6% 42.4% 42 45. 7% 29 31.5% 21 22.8% 92 100.0% CHI-SQUARE 12.03999 D . F . STATISTIC SIGNIFICANCE 0.0024 MIN E . F . 8.902 CELLS WITH E . F . < 5 NONE SYMMETRIC WITH ACTIVE 1 OEPENDENT WITH SITE OEPENDENT LAMBDA 0.14607 0.10000 0.20513 Table 39: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Activity types, during weekend (for a re-categorized Activity variable). 274 COUNT EXP VAL ROW PCT COL PCT ACTIVE 1 SI TE 1 Skyway N i c o l l e t ROW Mal1 F TOTAL 1| 2| 1 Watch ing/Groom in Talk ing M i see 11aneous Meet ing COLUMN TOTAL CHI-SQUARE D.F 13.61683 22 18.5 44 .0% 31.9% 28 31.5 56.0% 23.9% 23 30.8 27 . 7% 33 . 3% 60 52.2 72. 3% 51.3% 15 15.6 35. 7% 21.7% 27 26 . 4 64 . 3% 23. 1% 9 4 . 1 81.8% 13.0% 2 6.9 18.2% 1 . 7% 69 117 50 26.9% 83 44.6% 42 22 . 6% 1 1 5.9% 37 . 1% SIGNIFICANCE 0.0035 186 62.9% 100.0% MIN E . F . 4 .081 CELLS WITH E . F . < 5 1 OF 8 ( 12.5%) STATISTIC SYMMETRIC WITH ACTIVE 1 DEPENDENT WITH SITE 1 DEPENDENT LAMBDA 0.04070 0.00000 0. 10145 Table 40: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Activity types, during weekdays (for a re-categorized Activity variable). 275 COUNT EXP VAL ROW PCT COL PCT ACTIVE 1 SI TE 1 Skyway N i c o l l e t ROW Mall F TOTAL 1| 2| 1 Watch ing/Groom in 32 23.4 76.2% 60 . 4% 10 18.6 23.8% 23.8% 2 Talk ing 14 16.7 46 . 7% 26.4% 16 13.3 53. 3% 38. 1% 3 Others 7 12.8 30. 4% 13.2% 16 10. 2 69.6% 38. 1% COLUMN TOTAL 53 55.8% 42 44 . 2% 42 44 . 2% 30 31.6% 23 24.2% 95 CHI-SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 14.09417 STATISTIC LAMBDA 0.0009 SYMMETRIC 0. 17895 10.168 NONE WITH ACT IVE 1 DEPENDENT 0. 11321 WITH SITE 1 DEPENDENT 0.26190 Table 41: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Activity types, during weekend (for a re-categorized Activity variable). 276 COUNT EXP VAL ROW PCT COL PCT SITE2 Sk yway ACTIVE 1 Marquet t e avenue I 2| 1 Watch ing/Groom in Talk ing Mi see 1 1 aneous Meet ing 22 19.4 7 1.0% 31.9% 9 11.6 29 .0% 22.0% 23 23.8 60 . 5% 33. 3% 15 14.2 39.5% 36.6% 15 18.8 50.0% 21.7% 15 11.2 50.0% 36 . 6% ROW TOTAL 31 28.2% 38 34 .5% 30 27 . 3% 4 9 2 11 6.9 4 . 1 10.0% 81.8% 18.2% 13.0% 4.9% COLUMN 69 41 110 TOTAL 62.7% 37 . 3% 100.0% CHI - SQUARE 4.77231 D.F . STATISTIC SIGNIFICANCE 0. 1892 MIN E . F . CELLS WITH E . F . < 5 4.100 1 OF 8 ( 12.5%) SYMMETRIC WITH ACTIVE 1 DEPENDENT WITH SITE2 DEPENDENT LAMBDA 0.00000 0.00000 0.00000 Table 42: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Activity types , during weekdays (for a re-categorized Activity variable). 277 SITE COUNT POSTURE 1 S i 11 ing EXP ROW COL VAL PCT PCT Skyway N i c o l l e t N i c o l l e t Marquett ROW Mall R Mall F e Avenue TOTAL 1.00| 2.00| 3.00| 4.00| 1 00 21 38.6 15.2% 30. 4% 0 11.2 .0% .0% 99 65.4 71.7% 84 .6% 18 22.9 13.0% 43.9% 138 55.9% Stand ing 2 00 40 26 .0 4 3.0% 58.0% 18 7 . 5 19.4% 90.0% 14 44 . 1 15. 1% 12.0% 21 15.4 22.6% 51.2% 93 37. 7% Other 3 00 8 4.5 50.0% 11.6% 2 1 . 3 12.5% 10.0% 4 7.6 25 .0% 3.4% 2 2.7 12.5% 4.9% 16 6 . 5% COLUMN TOTAL 69 27.9% 20 8 . 1% 117 47.4% 41 16 . 6% 247 100.0% CHI - SQUARE D.F . SIGNIFICANCE MIN E . F . CELLS 87 . 1 7076 6 0.0000 1 .296 3 OF 12 ( 25.0%) STATISTIC SYMMETRIC WITH POSTURE 1 DEPENDENT WITH SITE DEPENDENT LAMBDA 0.29289 0. 36697 0.23077 Table 43: Crosstabulation of Posture types by Site, for weekday data (for a re-categorized Posture variable). 278 SITE COUNT POSTURE 1 S i t t ing EXP ROW COL VAL PCT PCT Skyway N i c o l l e t N i c o l l e t Marquett Ma 11 R Ma 11 F e Avenue 1.OO| 2.OO| 3.OO| 4.OO 1 OO 21 23.5 33.9% 39 . 6% 9 17.3 14.5% 23. 1% 32 18.6 51.6% 76 . 2% 0 2.7 .0% .0% Standing 2 OO 17 20.8 30.9% 32. 1% 25 15.3 45.5% 64 . 1% 9 16 . 5 16.4% 2 1.4% 4 2.4 7 . 3% 66 . 7% Other 3 OO 15 8.7 65.2% 28. 3% 5 6.4 21.7% 12.8% 1 6.9 4 . 3% 2.4% 2 1.0 8.7% 33. 3% COLUMN TOTAL 53 37.9% 39 27.9% 42 30.0% 6 4 . 3% ROW TOTAL 62 44 . 3% 55 39 . 3% 23 16.4% 140 100.0% CHI-SQUARE 38.84748 D . F . STATISTIC SIGNIFICANCE 0.0000 MIN E . F . CELLS WITH E . F . < 5 0.986 3 OF 12 ( 25.0%) SYMMETRIC WITH POSTURE 1 DEPENDENT WITH SITE DEPENDENT LAMBDA 0.23636 0.25641 0. 21839 Table 44: Crosstabulation of Posture types by Site, for weekend data (for a re-categorized Posture variable). 279 COUNT EXP VAL ROW PCT COL PCT POSTURE 1 St and ing other SITE Skyway N i c o l l e t ROW Mall R TOTAL 1| 2| + 1 40 18 45.0 13.0 69 .0% 31.0% 58 . or. 90.0% 2 29 2 24.0 7.0 93.5% 6.5% 42.0% 10.0% COLUMN 69 20 TOTAL 7 7.5% 22.5% 58 65.2% 31 34 . 8% 89 100.0% CHI - SQUARE O F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 5.66755 7.00754 STATISTIC 0.0173 0.0081 6.966 NONE ( BEFORE YATES CORRECTION ) SYMMETRIC WITH POSTURE 1 DEPENDENT WITH SITE DEPENDENT LAMBDA 0.00000 0.00000 0.00000 Table 45: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Posture types, during weekdays (for a re-categorized Posture variable). 280 COUNT EXP VAL ROW PCT COL FC T SITE Sk yway POSTURE 1 N i c o l l e t ROW Mall R TOTAL H Z | S i 11 ing 1 21 17.3 70.0% 39 . 6% 9 12.7 30.0% 23. 1% Stand ing 2 17 24.2 40.5% 32. 1% 25 17.8 59.5% 64. 1% Other 3 15 11.5 75.0% 28 . 3% 5 8.5 25.0% 12.8% COLUMN TOTAL 53 57.6% 39 42.4% 30 32.6% 42 45. 7% 20 21.7% 92 100.0% CHI - SQUARE D . F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 9.41132 STATISTIC LAMBDA 0.0090 SYMMETRIC 0. 13483 8.478 NONE WITH POSTURE 1 DEPENDENT 0.08000 WITH SITE DEPENDENT 0.20513 Table 46: Pairwise comparison of Skyway System and Mall(R) for the proportional distribution of Posture types, during weekend (for a re-categorized Posture variable). 281 SI IE1 COUNT POSTURE 1 S i 11 ing EXP ROW COL VAL PCT PCT Skyway N i c o l l e t Mall F 1| 2 ROW TOTAL 1 21 44.5 17.5*. 30.4% 99 75. 5 82. 5% 84 .6% 120 64 . 5% Stand ing 2 40 20.0 74 . 1% 58.0% 14 34 .0 25.9% 12.0% 54 29 .0% Other 3 8 4.5 66 . 7% 11.6% 4 7.5 33. 3% 3.4% 12 6.5% COLUMN TOTAL 69 37 . 1% 117 186 62.9% 100.0% CHI-SQUARE D.F . 55.88664 2 STATISTIC SIGNIFICANCE 0.0000 SYMMETRIC MIN E F . CELLS WITH E . F . < 5 4.452 1 OF 6 ( 16.7%) WITH POSTURE 1 DEPENDENT WITH SITE 1 DEPENDENT LAMBDA 0.36296 0.28788 0.43478 Table 47: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Posture types, during weekdays (for a rejcategorized Posture variable). 282 COUNT EXP VAL ROW PCT COL PCT SI TE 1 Sk yway POSTURE 1 S i 11 ing Stand ing Other N i c o l l e t Mal 1 F 1| 2| ROW TOTAL 1 21 32 53 29 .6 23 . 4 55 . 8% 39.6% 60 . 4% 39 . 6% 76 . 2% 2 1 7 9 26 14.5 1 1 . 5 27 . 4% 65.4% 34 .6% 32. 1% 21 . 4% 3 15 1 16 8.9 7 . 1 16.8% 93.8% 6 . 3% 28. 3% 2 . 4% + -COLUMN 53 42 95 TOTAL 55.8% 44 .2% 100.0% CHI-SQUARE D . F . 15.93452 STATISTIC SIGNIFICANCE 0.0003 MIN E . F . 7 .074 CELLS WITH E . F . < 5 SYMMETRIC NONE WITH POSTURE 1 DEPENDENT WITH SITE1 DEPENDENT LAMBDA 0.13095 0.00000 0.26190 Table 48: Pairwise comparison of Skyway System and Mall(F) for the proportional distribution of Posture types, during weekend (for a re-categorized Posture variable). 283 COUNT EXP VAL ROW PCT COL PCT POSTURE 1 S i 11 ing Stand ing Other SITE2 Skyway Marquett ROW e avenue TOTAL 1| 2| 1 COLUMN TOTAL CHI - SQUARE 2.80315 D . F . STATISTIC 21 18 24 . 5 14.5 53.8% 46.2% 30 . 4% 43.9% 40 21 38 . 3 22. 7 65 . 6% 34 . 4% 58 .0% 51.2% 8 2 6 . 3 3.7 80 .0% 20 .0% 11.6% 4 . 9% 69 4 1 62 . 7% 37 . 3% 39 35.5% 61 55. 5% 10 9. 1% LAMBDA 1 10 100.0% SIGNIFICANCE MIN E . F 0.2462 SYMMETRIC 0.00000 3. 727 CELLS WITH E . F . < 5 1 OF 6 ( 16.7%) WITH POSTURE 1 DEPENDENT 0.00000 WITH SITE2 DEPENDENT 0.00000 Table 49: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Posture types, during weekdays (for a re-categorized Posture variable). 284 COUNT EXP VAL ROW PCT COL PCT POSTURE 1 S t and ing other SITE? Sk yway 1| Marquet t e avenue 1 COLUMN TOTAL 1 7 18.9 81.0% 32. 1% 36 34 . 1 94 . 7% 6 7.9% 53 89.8% 4 2. 1 19.0% 66 . 7% 2 3.9 5.3% 33. 3% 6 10.2% ROW TOTAL 21 35 6% 38 64 . 4% 59 100.0% CHI - SQUARE D.F . SIGNIFICANCE MIN E . F . CELLS WITH E . F . < 5 1 .50665 2.81324 0.2197 0.0935 2.136 2 OF 4 ( 50.0%) ( BEFORE YATES CORRECTION ) STATISTIC SYMMETRIC WITH POSTURE 1 DEPENDENT WITH SITE2 DEPENDENT LAMBDA 0.07407 0.09524 0.00000 Table 50: Pairwise comparison of Skyway System and Marquette Avenue for the proportional distribution of Posture types, during weekend (for a re-categorized Posture variable). 285 COUNT EXP VAL ROW PCT COL PCT SITE Sk yway DISABLED N i c o l l e t N i c o l l e t Marquett Mall R Mall F e Avenue 1.001 2. 001 3. 00 | 4.001 1 .00 1 2 0 3 Cane. Crutches 7 1 3 1 3 2 7 16 7% 33 3% 0% 50 or. 100 0% 100 0% 0% 75 or. 2.00 0 0 2 1 Whee lcha ir 3 7 7 1 3 0% 0% 66 7% 33 3% 0% 0% 100 0% 25 or. COLUMN TOTAL 11. 1 % 2 22 . 2% 2 22. 2% 4 44 . 4? ROW TOTAL 66 6 7% 3 33 . 3% 9 100.0% CHI - SQUARE D . F . 5.62500 STATISTIC LAMBDA SOMERS' STATISTIC SIGNIFICANCE 0.1314 SYMMETRIC 0. 37500 0. 1304 3 VALUE MIN E . F . 0. 333 CELLS WITH E . F . < 5 8 OF WITH DISABLED DEPENOENT 0.66667 0. 10714 SIGNIFICANCE 8 (100.0%) WITH SITE DEPENOENT 0.20000 0.16667 KENDALL ' S KENDALL'S GAMMA TAU B TAU C 0. 13363 0. 14815 0 . 20000 Table 51: Crosstabulation of Disabled categories by Site, for weekday data. 286 COUNT EXP VAL ROW PCT COL PCT DISABLED 1 .00 Cane. Crutches SITE Sk yway Nico l let Mai 1 F 1.00| 3.00| 2.00 Whee lcha ir COLUMN TOTAL 2 1 .0 100.07. 100.0% 0 1 .0 .0% .0% 2 50.0% 0 1.0 .0% .0% 2 1.0 100.0% 100.0% 2 50 .0% ROW TOTAL 2 50 .0% 2 50.07. 4 100.0% STATISTIC FISHER'S EXACT TEST STATISTIC LAMBDA SOMERS' D STATISTIC ONE TAIL 0. 16667 SYMMETRIC 1.00000 1.00000 VALUE TWO TAIL 0.33333 WITH DISABLED DEPENDENT 1.00000 1.00000 SIGNIFICANCE WITH SITE DEPENDENT 1.00000 1.00000 KENDALL'S TAU B KENDALL'S TAU C GAMMA 1 .00000 1.00000 1.00000 Table 52: Crosstabulation of Disabled categories by Site, for weekend data. 287 APPENDIX III CHAPTER FOUR BAR CHARTS 100 80 60 40 20 PERCENTAGE OF ALL USERS d i i SKYWAY S Y S T E M N I C O L L E T MAL L(R)N I C O L L E T MALL(F) M A R Q U E T T E AVE. SYSTEMS SURVEYED CHILD TEEN E H ADULT ELDERLY WEEKDAY DATA Fig.l: Proportional Distribution of Users by Age on the Surveyed Facilities, during Weekdays. 100 80 60 40 20 PERCENTAGE OF ALL USERS SKYWAY S Y S T E M N I C O L L E T M A L L ( R ) N I C O L L E T MALL(F) M A R O U E T T E AVE. SYSTEMS SURVEYED CHILD TEEN ADULT ELDERLY WEEKEND DATA Fig.2: Proportional Distribution of Users by Age on the Surveyed Facilities, during Weekend. 288 PERCENTAGE OF ALL USERS SKYWAY SYSTEM N ICOLLET MALL(R) NICOLLET MALL(F) M A R Q U E T T E AVE. SYSTEMS SURVEYED MALE t22a FEMALE WEEKDAY DATA Fig.3: Proportional Distribution of Users by Gender on the Surveyed Facilities, during Weekdays. PERCENTAGE OF ALL USERS SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) M A R Q U E T T E AVE. SYSTEMS SURVEYED MALE ^ FEMALE WEEKEND DATA Fig.4: Proportional Distribution of Users by Gender on the Surveyed Facilities, during Weekend. PERCENTAGE OF ALL USERS SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) MARQUETTE AVE. SYSTEMS SURVEYED • i VISITOR HH TOURIST EMI SEC.QUARD M WORKER DUi NOT DET. WEEKDAY DATA Fig.5: Proportional Distribution of Users by Role on the Surveyed Facilities, during Weekdays. PERCENTAGE OF ALL USERS SKYWAY SYSTEM NICOLLET MALL(R) N ICOLLET MALL(F) M A R Q U E T T E AVE. SYSTEMS SURVEYED • i VISITOR EUl TOURIST LMO SEC.QUARD £M1 WORKER BHD) NOT DET. WEEKEND DATA Fig.6: Proportional Distribution of Users by Role on the Surveyed Facilities, during Weekend. 290 PERCENTAGE OF ALL USERS 100 | SKYWAY SYSTEM NICOLLET MALL(R)N I COLLET MALL(F) MARQUETTE AVE. SYSTEMS SURVEYED ALONE [§§1 2 S.S. EM 2 O.S ^ § 3 /MORE HULO FAMILY WEEKDAY DATA Fig.7: Proportional Distribution of Users by Grouping on the Surveyed Facilities, during Weekdays. PERCENTAGE OF ALL USERS SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) MARQUETTE AVE. SYSTEMS SURVEYED ALONE tH2 2 S.S. EM 2 O.S E§H 3 /MORE 003 FAMILY WEEKEND DATA Fig.8: Proportional Distribution of Users by Grouping on the Surveyed Facilities, during Weekend. 291 PERCENTAGE OF ALL USERS SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) M A R Q U E T T E AVE. SYSTEMS S U R V E Y E D WM ACCESS HI! SHOP • PROM. HI WORK SHiD RECRE. • SEARCH W HANQ WEEKDAY DATA Fig.9: Proportional Distribution of Users by Purpose on the Surveyed Facilities, during Weekdays. PERCENTAGE OF A L L USERS 50 i SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) M A R Q U E T T E AVE. SYSTEMS S U R V E Y E D • 1 ACCESS m SHOP E Z J PROM. H I WORK LHHO RECRE. • SEARCH W HANQ W E E K E N D DATA Fig.10: Proportional Distribution of Users by Purpose on the Surveyed Facilities, during Weekend. 292 PERCENTAGE OF ALL USERS 70 I SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) MARQUETTE AVE. SYSTEMS SURVEYED • i WATCH TALK L U EAT READ LUOI] PERF. • WAIT W MEET WEEKDAY DATA Fig.11: Proportional Distribution of Users by Activity on the Surveyed Facilities, during Weekdays. PERCENTAGE OF ALL USERS SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) MARQUETTE AVE. SYSTEMS SURVEYED •1 WATCH gSi TALK • EAT •HI READ \nnn PERF. • WAIT W MEET WEEKEND DATA Fig.12: Proportional Distribution of Users by Activity on the Surveyed Facilities, during Weekend. 293 100 80 60 40 20 P E R C E N T A G E O F A L L U S E R S 1 1 jl JBII SKYWAY SYSTEM NICOLLET MALL(R)NICOLLET MALL(F) MARQUETTE AVE. S Y S T E M S S U R V E Y E D SIT STAND E l l LIE L E A N W E E K D A Y DATA Fig.13: Proportional Distribution of Users by Posture on the Surveyed Facilities, during Weekdays. P E R C E N T A G E O F A L L U S E R S 1 1 1 , Mil I F7771 SKYWAY SYSTEM NICOLLET MALL(R) NICOLLET MALL(F) MARQUETTE AVE. S Y S T E M S S U R V E Y E D SIT §H! STAND ME LIE M L E A N H I K N E E L W E E K E N D DATA Fig.14: Proportional Distribution of Users by Posture on the Surveyed Facilities, during Weekend. 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098624/manifest

Comment

Related Items