UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The utilization of space in an isotropic environment : a predictive model of beach user behaviour Evans, Laurence Kenneth 1977

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1977_A1 E83.pdf [ 7.31MB ]
JSON: 1.0076856.json
JSON-LD: 1.0076856+ld.json
RDF/XML (Pretty): 1.0076856.xml
RDF/JSON: 1.0076856+rdf.json
Turtle: 1.0076856+rdf-turtle.txt
N-Triples: 1.0076856+rdf-ntriples.txt
Original Record: 1.0076856 +original-record.json
Full Text

Full Text

THE UTILIZATION OF SPACE IN AN ISOTROPIC ENVIRONMENT: A PREDICTIVE MODEL OF BEACH USER BEHAVIOUR. LAURENCE KENNETH EVANS M.S., University of Utah, 1969 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY THE FACULTY OF GRADUATE STUDIES Interdisciplinary Studies: School of Community & Regional Planning Institute of Animal Resource Ecology Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November, 1977 by in Laurence Kenneth Evans, 1977 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 representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Community Regional Planning The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date ?A October 1977 i ' ABSTRACT A public beach served as the site for a study of the impact of in creasing density on human spatial behaviour. This setting provided a ; unique environment where the range of observed densities was wide, user behaviour could be monitored unobtrusively, and where effects due to the social and physical environment were not confounded. The specific goals of the research were: 1) to test the hypothesis that beach users require a minimum amount of intergroup space ard that such distances will be related to proper social functioning (cf. Edward T. Hall's proxemic zones),, 2) demonstrate a relationship between the overall spatial pattern of beach users and density, and 3) relate indi vidual personality dispositions, mood states and socio/demographic differences to observed respondent spatial behaviour. Aerial photography was used to gather data concerning the spatial distribution of 1791 groups located on three public beaches or sunning areas. Coincidental psycho/demographic data were obtained by means of a paper and pencil survey for a subsample of 266 subjects located on the beaches during the 27 photographic sampling runs completed. A Monte Carlo simulation technique coupled with a 'distance to nearest neighbour' model were used to analyse the spatial pattern of beach users over the range of densities observed. Results indicate that at densities less than 110 groups/hectare the observed spatial pattern does not differ significantly from random. At higher densities however, users tend to maximize the distance to near neighbours which results in aypattern statistically described as uniform. The average distance separating groups at densities greater than 110 groups/hectare approached a constant at 2.7 meters. This latter observation plus Hall's claim that such distances may be utilized toreffectively screen or insulate persons from unwanted social inter action suggests that beach users adapt to increasing density by obtaining ii just enough space to maintain the social integrity of the group. Survey results using groups produced few significant correlations and stepwise regression analysis indicated characteristically lowpredic-tability of target spatial variables. Analysis of response patterns of lone individuals however, produced a substantial increase/.in the ability of selected independent variables to account for variance in dependent variables. For example, respondent nearest neighbour distance was 2 predicted moderately well by six independent variables (R = .47). Similarly, eight variables accounted for 57% of the variance in the dependent variable which measured the amount of space demarcated by a respondent's personal possessions. These results suggest that at lower densities beach users may choose sites in relation to other "users which reflect individual preferences and since preferences are varied a random spatial pattern is observed. However, as space becomes limiting at higher densities such needs anddesires may remain unfulfilled. Finally, based on the above results maximum 'psychological carrying capacity' estimates were calculated and the implications for the planning and design professions discussed. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ACKNOWLEDGEMENT CHAPTER I. Spatial Behaviour and the Regulation of Social Interaction. - Goals and objectives. - The beach as an isotropic environment. - Origin of concepts. - Individual distance. - Personal Space. - Crowding. - Space as a limiting factor: The first hypothesi - Density and spatial pattern. II. Individual Differences and Spatial Behaviour: The Role of Personality and Socio-economic Characteristics. - Introduction - The role of personality. - Personality correlates. - The survey instruments. - The Environmental Response Inventory. - The Leisure Activities Blank. - Mood Adjective Checklist. - Socio-economic and demographic characteristics. III. An Approach to the Study of Human Spatial Behaviour Methods and Applications. - The study areas. - Aerial photography. - Pattern analysis. - Space-time study. - Survey dissemination. - Psychological and demographic information. - Environmental Response Inventory. - Leisure Activities Blank. - Socio-economic variables. - Mood adjective checklist. iv Pago IV. Spacing Behaviour on Public Beaches: Analysis of Results. - User distribution and external environmental features. - Density and spatial pattern. T Density and distance to nearest neighbor. V. Group Characteristics VI. Beach Carrying Capacities. - Introduction. - Carrying capacities and the response to density. VII. Predictors of Spatial Behaviour: Analysis of Survey Results. - Predictors of spatial behaviour - all groups., - Lone individuals - a second look at the data0 - predictors of respondents' distance to nearest neighbor. - Predictors of 'group area1. - Marked group area and density, - Density and group size. VIII. Discussion and Summary of Results. 75 - Summary of results. - Methodology applications. - Implications for planning and design. - Toward further research. LITERATURE CITED 83 APPENDICES A B C D 89 100 108 110 V Iii ST OF FIGURES FIGURE PAGE 1. Map depicting Kitsilano and English Bay study areas ' 30 2. Map depicting Skaha study area 31 3. The distance between points is not maximized by a hexagonal 37 pattern 4. A pattern of equilateral triangles maximizes the inter-point 37 distance 5. With more points in the same area, a pattern of squares 37 maximizes the distance 6. Spatial characteristics of beach users over time 47 7. Spatial characteristics of beach users over time 47 8. Spatial characteristics of beach users over time 47 9. Spatial characteristics of beach users over time 48 10. Spatial characteristics of beach users over time 48 11. Spatial characteristics of beach users over time 48 12. Relationship between density and pattern 50 13. Average distance between first nearest neighbours (NND) plotted against density for 27 runs 52 14. Density and group area 56 15. Density and group size 8 16. Hypothetical representation of two neighboring groups "marked" and " minimum" space boundaries 61 17. Random pattern characteristic of low densities 110 18. Transition from random to regular pattern 1119. Regular pattern characteristic of high densities 111 vi LIST OF TABLES TABLE PAGE 1. Proxemic zones for North Americans 7 2. Environmental Response Inventory Scales 21 3. Seven LEISURE ACTIVITIES BLANK - Past Factors 23 4. Correlations of the Adjectives with Eight Mood Factors 27 5. Group Area Related to the Number of Individuals in a Group 57 6. Maximum beach population estimates 62 2 7. Initial partial correlations, F probabilities and final r values for three dependent variables 66 2 8. r variables for spatial and group dependent variables 68 9. Significant independent variables contributing to the dependent variables, centroid to centroid and nearest approach nearest neighbor distances (cc/nnd and na/nnd). 70 10. Significant independent variables contributing to the depen dent variable 'group area'. 72 11. ERI variables: means and standard deviations for English Bay, Kitsilano, and Skaha Beaches 101 12. LAB variables: means and standard deviations for English Bay, Kitsilano, and Skaha Beaches 1013. McKechnie's (1973) LAB Results compared with those from the present study 102 14. Socio/demographic variables: means and standard deviations for English Bay, Kitsilano and Skaha beaches 104 15. Mood score means and standard deviations for English Bay, Kitsilano, and Skaha beaches 107 16. Comparison of means for the two conditions: surveyed and not surveyed 109 ACKNOWLEDGEMENTS An interdisciplinary dissertation cannot help but represent the efforts of a variety of people in addition to the author. Among these, William Rees my research supervisor has contributed more than his share of time, effort and enthusiasm. To him I would like to express a warm thank you. I wish also to thank the members of the guidance committee for their ideas, critical evaluation and sup port during the course of my research and writing. Of these, John Collins and C.S. 'Buzz' Holling deserve special mention since each contributed to the research far in excess of committee re quirements. For help in the computer analysis including the simulation project, I wish to thank Steve Borden of the TARE computing facility In addition, thanks are in order for Frank Maurer who aided the re search by making the photographic and digitizer equipment so readil available. To the School of Community and Regional Planning and the Ford Foundation who helped fund the research, I wish to extend my sincere gratitude. I wish also to express my appreciation to Edward T. Hall who acted as external examiner. To my mate, Anna Friesen I wish to demonstrate my apprec iation for her unfailing support, encouragement and love during the course of my work. Her help was all that it could and should be. Finally, I would like to express my thanks to my parents for their continued help and understanding. viii CHAPTER I SPATIAL BEHAVIOUR AND THE REGULATION OF SOCIAL INTERACTION - 1 -Coals and Objectives This study focuses on the problem of how users of public beach facilities respond spatially to increasing numbers of individuals in a shared space and how this response Is associated with certain personality, socio-economic and demographic characteristics. More specifically the objectives of the study are: to develop a methodology for obtaining unobtrusive measures of beach user spatial and group behaviour. to describe the pattern or spatial distribution of users as a function of density. - to determine the extent to which aspects of the physical environment surrounding the beach afiect user spatial distribution. to delineate how users adapt to decreasing avail able space as density increases. to examine the extent to which personality and socio-economic characteristics could be used as predictors of spatial and group behaviour. The above goals formed the basis for three general hypotheses and one corrollary which I wished to test. These will be discussed more fully in later sections, however for the present they may briefly be stated as: 1) As space becomes limiting at higher densities the distance between nearest neighbours will approach a minimum value the limit of which will be determined by cultural norms relating to the regulation of social interaction. a) Corrollary: The spatial pattern exhibited by beach users will be related to changes in user density. 2) The distnnces maintained between nearest neighbours will be some function of certain internal psychol ogical dispositions. 3) Nearest neighbor distance will be related to user socio-economic and demographic characteristics. - 2 -Although tho results of otlior analyses nro nlsn ro por tod, those three hypotheses reflect the major objectives of the study. These objectives relate to n broader theoretical perspective involving the ways in which space may be used as a regulating mechanism of social interaction. There are many ways in which social animals Including man attempt to regulate the kind and intensity of social inter action, however as Kummer (1971) points out, the manipulation of space is the safest technique available, T/his statement is underlined since close physical proximity in many specie's carries an implied threat of aggression and such distances bring interactants within the range of whatever weapons are available. Many signals have evolved to counteract "aggressive responses between members of a species since activities related to reproduction and social bonding would not be possible were there no mechanism available to counter act such aggressive behaviours. Tinbergen's (1952) now well known des cription of the zig-zag dance of the male stickleback is an often cited example of* signalling devices which serve to inhibit agonistic responses while stimulating reproductive behaviours. In manr space may still be viewed as an important component re lating to regulation of social interaction, however, modern methods of com munication complicate the issue. The telephone, newspaper, television, radio, etc., all serve to increase perceived social interaction. Distance, at least for these examples becomes meaningless. The present research sought to examine some of the ways humans use space where the potential for social interaction increases dramatically. A public beach was chosen as a potentially useful environment in which to conduct the research since densities varied over a wide range and since perceived and actual density could be assumed to be similar if not the same. The rationale for the choice of a beach as a research setting plus issues and concepts relating to distance, crowding and space control will be discussed as they pertain to the study goals. - 3 -The beach as an isotropic environment. Since most studies of spacing behaviour in humans have been con ducted under laboratory conditions where the setting is often contrived, is of short duration, or has taken place in complex settings where effects due to the physical environment are not knownt I chose a public beach as a site least likely to suffer from shortcomings such as these. A public beach represents a setting which is generally uniform and relatively free from 'artifactual constraints' . Based on the assumption that in such an environment effects due to the physical setting would be minimized I pro posed that observed changes in behavioural patterns must result largely from social or interpersonal factors. Such 'free field' settings are referred to as 'isotropic'. An isotropic environment such as a beach thus has the advantage of allowing the investigator to examine behaviour relating to site selection, spacing, and group phenomena in isolation from physical setting effects. By controlling for these effects and by ob serving behaviour unobtrusively, I hoped to provide 'baseline' results concerning the ways in which people use space under conditions of varying density. An example may illustrate the point I wish to make. In an un-crowded theater a patron could choose a seat which both satisfied his desired location with respect to the stage as well as his spatial prefer ence vis a' vis other patrons. In contrast when densities are high, few alternatives remain and a choice of seats would require sitting within inches of another patron. Since the seats have been placed as they have by an 'authority', the arrangement acts as a sanction which allows the new arrival to, in effect, invade the personal space1 of the person already in place. A beach without such constraints should thus provide a way of determining spatial preferences based solely on the internal needs of the individual. It is in this sense that the term 'baseline' is used. - A -Finnlly, two other advantages of using a beach as a source for the study of human spatial behaviour are important. First, the actual density of the beach as expressed as individuals or groups/unit area may be assumed to be the same as, or at least close to the density which users perceive. Rapoport (1975) emphasizes the importance of differentia ting between perceived and actual density when one looks for effects due to density considerations. He argues that for many environments these two values may be quite different. Of course, methodologically it is far easier to accurately measure the actual number of people in an area than to pin down how many individuals a person believes are there. A beach setting avoids this problem since an individual can visually identify the number of people in the vicinity. The final aspect of a beach which is important to a study of density effects is that an ethological approach may be used. Using this technique the behaviour of individuals and groups may be observed unob trusively with little or no bias introduced by the investigator's presence. In this way, observed behaviour may be considered 'natural' for the setting involved and conclusions more easily translateable to everyday (real world) events. Origin of concepts Wilson (1975) reviews issues and concepts relating to spacing behaviour as it applies to social organisms and discusses six components of social spacing: l)total range, 2) home range, 3) core area, 4) terri tory, 5) individual distance, and 6) dominance. These categories may be briefly summarized in the following manner: Total range: the area traversed by an individual animal over its entire life cycle (Goin & Goin, 1962). Home range: the area over which an animal habitually travels (Seton, 1909; Burt, 1962). Core area: that area of heaviest useage within the home range (Kaufmann, 1962). - 5 -Territory: an area occupied more or less exclusively through means of overt defense or advertisement (Noble, 1939; Brown, 1964). Individual distance: the minimum distance routinely kept between individuals of a species (Hediger, 1941, 1955; Conder, 1949). Dominance: the assertion of one member of a group over another which gains the dominant individual increased access to resources such as food, water, sleeping sites, space, etc. With respect to these concepts Wilson (1975) stresses that the behaviour of species in general show a continuously graded series with the boundaries of each of the above classifications becoming blurred and indistinct. The primary point is that variations occur within, as well as between species with each behavioural manifestation serving a distinct biological and/or social function. Thus(a bird may maintain a home range with respect to its feeding activities, a territory surrounding the nest site and individual distance and dominance characteristics within a flock. Other examples of this continuum relate to changes in breeding fluctuations and varying life cycle conditions each requiring different spatial strategies. In recent years the concept of territoriality has been invoked to explain a varietv of human behaviours, most notably Lorenz (1966) and Ardrey <"1966) . At least one study of beach user behaviour has utilized the territory concept for theoretical constructs (Edney and Jordan -Edney, 1974). Based upon the most commonly accepted definitions of what constitutes a territory this approach seems unwarranted. The point was made by Becker and Mayo (1971) that personal space and individual distance concepts were more appropriate than territorv in defining the spacing behaviour of cafeteria patrons since in their study, subjects were not willing to defend the area denoted by their possessions. These authors concluded that individual distance concepts were more parsimonious in that they made fewer assumptions about the relative value of a space. For the - 6 -present study the concepts relating to individual distance seemed most closely to reflect the goals of the research as well as the observed behaviour of users. Individual distance. In studying the spatial behaviour of beach users, I was pre dominantly interested in the distances separating individuals and groups of individuals. Studies of spacing mechanisms of social animals have referred to two measures of intraspecific spacing: personal or individual distance and social distance (Hediger, 1941,1955; Conder, 1949) Personal distance refers to a minimum distance which individual animals routinely keep between themselves and others, whereas social distance relates to the distance beyond which an animal apparently experiences a strong attraction to return to its social group. These two concepts underline the dynamic quality of spacing in social animals,. In his re view of sociobiology, Wilson (1975) highlighted this dynamic character istic by defining personal distance as "the compromise struck by animals that are both attracted to other members of their own species and repelled by them at short distances." (p. 257). Much of the literature on social animals is concerned with interindividual spacing behaviour. However, a few studies have indica ted that groups may act collectively in maintaining intergroup distances. For example, Blank and Ash (1956) showed that coveys of partridge (P. perdix), although exhibiting overlapping home ranges, normally remain separated by a certain minimum distance. Similarly, both individuals and groups of sandhill cranes (Grus canadensis) space themselves evenly with groups primarily composed of family units (Miller and Stephen, 1966). Other species exhibiting similar patterns are baboons (Hall and Devore, 1965), white-crowned sparrows (Zonotrichia leucophrys) (Blanchard and Erickson, 1949; Morton, 1967) and wintering flocks of Juncos (J. hyemalis and J. oregamus) (Sabine, 1956). - 7 -Hall (1966) has studied the ways in which humans space them selves in a variety of social situations across several cultures. Hall suggests that for humans, personal distance is but one of four zones. Based on observations of behaviour patterns at various interpersonal dis tances Hall has termed the other three, * intimate','social', and 'public'. Table 1 outlines these zones for North Americans and demonstrates the behaviours characteristic of each. Since these zones seem to have func-^ tional properties related TO me type and form of social interaction permissable for North Americans, I will return to these ideas when discussing the testing of relevant hypotheses. Table 1. Proxemic zones for.North Americans.. (from Hall, 1966) TYPE OF ZONE Intimate Personal DISTANCES (meters) 0-.46 .46-1.22 Social 1.22-3.66 • Public 3.66-7.62 CHARACTERISTIC BEHAVIOURS bodily contact possible, eg. lovemaking, comforting, protecting, wrestling, etc. bodily contact possible at close phase; most encounters between friends and close associates occur in this range. social gatherings, imper sonal business dealings, formal business and social events conducted at the far phase. behaviours which do not require interpersonal in volvement occur within this range; public speakers and important people are often observed in this range. - 8 -Personal space. To this point I have referred to the spatial characteristics of animals as a linear distance. Excluding territoriality, evidence exists which indicates that members of social species maintain an area or volume around themselves which is relatively exclusive of others. McBride (1964) has studied this phenomenon in various species and terms it a 'social force field'. His research indicates that these areas are not circular and tend to 'axtend further in the direction in which the animal is facing. McBride's social force field has been extensively studied in humans and has been labeled 'personal space' by Sommer (1966; 1969). As a testimony to the*interest in the field of personal space, Altman (1975) has documented over 200 studies conducted since the early 1960's. Most of this work is tangential to the present study. However, researcli concerning intrusions into personal space boundaries as well as the personality correlates of personal space behaviour is relevant. The former will be discussed below, however the latter will be reviewed later. Research concerning intrusion of personal space has confirmed the notion that unwarranted crossing of personal space boundaries is a powerful event. Two early studies by Felipe & Sommer (1966) underscore this statement. The first study examined the flight reactions of mental patients when a confederate sat beside a patient at a distance of approx imately six inches. In a park setting about one-third of the patients left within two minutes, about one-half left within nine minutes and over two-thirds left within 20 minutes. Many overt signs of discomfort such as fidgeting, mumbling and nervous rubbing of body parts were commonly observed. Similar results were also found in the second study which examined spacing behaviour of students in a library setting. Patterson, Mullens and Romano (1971) added strength to these findings through observ ations of facial expressions and body orientation of patrons. In this .library study, the incidence of such behaviour as blocking themselves off, leaning away, and glaring increased the closer a confederate was to a subject - 9 -In a study of the extent to which level of arousal (as measured by the Galvanic Skin Response-GSR) was correlated with interpersonal dis tance, McBride, King and James (1965) found that subjects who were app roached at distances of 1, 3 and 9 feet showed lower GSR readings as distance'increased. Another finding indicated that approaches from the side produced lower readings than those from the front. Tn another study of intrusion Efrnn and Clioyno (1973) observed shopping mall patrons to determine willingness to pass between two con federates standing at varying distances. They found patrons seldom passed between the confederates when they were closer than four feet apart. It is significant that this distance is at the edge of Hall's (1966) personal and social space zones referenced earlier. The above study thus tends to reinforce the functional validity of Hall's zones. In a later study, Efran and Cheyne (1974) forced subjects to pass between two closely interacting confederates. Results showed that subjects displayed more agonistic facial gestures and later reported less positive mood ratings than did controls. Predicted heart rate changes however, were not obtained. Argyle and Dean (1965) obtained similar results when subjects approached a photograph of a person or an actual person. Eye contact typically decreased as distance decreased and other signs of tension were also reported when subjects were approx imately two feet from the target person. The studies cited above relate to two basic themes. The first is that in general ,North Americans maintain similar conventions about the amount of space appropriate for different social and interpersonal occasions. The second is that personal space boundaries do exist and to violate them evokes feelings and bodily symptoms indicative of emotional distress both for the intruder and the person being intruded upon. With respect to the present study, these findings suggest that users of a public beach facility should choose sites which allow for at least a minimum amount of space which is consistent with social norms and personal needs. Hypotheses were generated to test this and other related constructs and will be dealt with in more detail in a later section. - 10 -Crowding. The goals of the project did not include a direct attempt to assess the possible impact of crowding on beach users. This decision was based on the author's belief that although models exist to define various aspects of the crowding experience (Stokols, 1976), the 'state of the art' is such that an accurate assessment of such characteristics remains problematical. Because of the difficulties in defining when crowding may be said to occur, I preferred to concentrate on describing the actual behaviour of users which could be attributed to increasing numbers of others within the beach environment. Such a research strategy, I believe, should serve to produce a baseline indicator of how people actually respond to the amount pf space available and precludes the ne cessity of determining how individuals feel about the experience in question. In addition, a primary hypothesis tested by the study pre-* dieted a minimum intergroup distance within which newly arriving users would not situate. Given the confirmation of such a hypothesis, one might conclude that crowding had not occurred since users were able to accrue some minimum amount of space necessary for maintaining a satisfactory experience. In other words, one would not expect respondents to express more than mild dissatisfaction as long as their basic needs for space were being met. Of course, to definitively answer questions relating to crowding it would be necessary to interview potential users who did not participate because of extreme densities. This strategy was not employed since such persons were not easily indentifiable. Although as I have pointed out, crowding was not a central issue for the present study, two other research efforts directed towards beach users have made attempts in this direction. The first, by Brougham (1968), attempted to assess levels of perceived crowding through a variety of questions relating to the perceived quality of the beach experience. The results of his study indicated that with eighteen independent variables relating to crowding and socio/demographic characteristics of users, only 12% of the variance of the dependent variable (nearest neighbor distance) could be accounted for. Further, the crowding index used by Brougham was significant only at the 0.90 probability level and contrary to expectations - 11 -the groups perceiving the beach as overcrowded were found at greater distances than those not objecting to the number of others present. The second study which attempted to measure perceived crowding in a beach setting was conducted by Edne y and Jordan-Edney (1974). The method they used involved asking beach users two questions; 1) How many people did they (the users) think the beach could hold before it became overcrowded? and 2) How did they see the beach at the time - crowded, average, or underpopulated? Responses to these questions were then compared to a measure of nearest neighbor distance. An analysis of responses to the second question which attempted to measure crowding directly, produced no significant differences between the crowding indices and nearest neighbor distance values. The data relating to estimates of the beach capacity was significantly correlated with distance, but only when disaggregated by group size and composition by sex. Although the authors attempt to explain these findings using two alternate hypotheses relating to 'focus of attention' versus 'sense of control1 these were not tested. This section has been included to demonstrate some of the difficulties associated with assessing the crowding experience directly. Because of such problems I chose not to include direct measures of crowding. However, the questionaire designed for the study did include several personality variables which could be used as indirect indices of tolerance to crowding. These will be discussed in a later, section. Space as a limiting factor: The first hypothesis. In a previous section I briefly referred to Hall's (1966) work in which he classifies observed social behaviours according to the distance separating interactants. Hall's scheme (see Table 1) suggests that certain classes of behaviour characteristic of North Americans can be grouped conveniently into four distance zones. These zones reflect increasing interpersonal and sensory involvement as distance decreases. These ob servations imply that distance, relative to potential or actual interact ants, carries meaning and this meaning relates to the type and quality of - 12 -social interaction sought. Similarly, as'Rapaport (1975) points out in a recent article, the definition of space by agreed upon rules serves as an organizing.element and thus decreases the amount of information needing to be processed at any given time. Thus by dividing classes of behaviours along a space continuum, the need to communicate behavioural intentions, other than by the use of distance, is diminished. Borrowing from Hall's and Rapaport's assessment of the functional qualities of space use I predicted that people on a public beach maintain basic spatial needs and that these needs are related to the regulation of social Interaction. Further, I hypothesized that these needs and prefer ences are influenced in part by various personality and cultural norms. In an attempt to examine the efficacy of these propositions, I chose Hall's (1966) schema of four distance zones as an initial source for theoretical constructs. Referring to Table 1, Hall's zone, 'public' is characterized by behaviours not requiring interpersonal involvement. Since most people were observed to maintain their group identity and since at low densities space was not thought to be limiting, I predicted that under these density conditions most users would locate at: nearest neighbor distances greater than 3.7 meters (12 ft.). Further, as pressures due to increasing density were realized, I also predicted that observed distances would compress to some point within Hall's (1966) 'social distance zone' (far phase). This predic tion is based on the argument that users could be expected to adopt various adaptational strategies which would allow somewhat closer inter group distances. The far phase of the social distance zone (2.1-3.7 meters) is the most likely lower limit of this compressibility since Hall's characterization includes the statement that social distance (far phase), "can be used to insulate or screen people from each other." In contrast, the close phase (1.2 - 2.1 meters) is typified by behaviours which are more casual than the far phase and contains more elements in dicative of social involvement, although of an impersonal nature. - 13 -Based on the above arguments the.following is a statement of the first hypothesis to be tested: - At low to moderate density, distances to nearest neighbor will be greater than 3.7 meters, however as space becomes limiting at higher densities, distances will approach a minimum value between 2.1 and 3.7 meters. Such extreme densities would create a situation in which newly arriving groups would be forced to violate spatial norms, seek out another less crowded beach or return home. The testing of the preceeding hypothesis was carried out with the aid of aerial photography covering three beaches over a broad range of densities for each. This technique allowed for an instantaneous record to be made of the position in space of users of an entire public beach. Density and spatial pattern. The dispersion of objects in space and time are studied through pattern analysis. Such analyses are widespread in such fields as ecology (Pielou, 1969) Grieg-Smith, 1964) and social geography (Dacey, 1964; Getis, 1964). The analysis of pattern is contingent upon three types of spatial distribution, of which two, regular and aggregated represent the opposite ends of a continuum. The third type, random, refers to the special case where the placement of a point or individual is uninfluenced by any other point. Aggregated or clumped patterns are exhibited when there is a higher probability that two or more points will be found in close proximity. A perfectly regular or uniform pattern is character ized by a set of points where all distances between points are maximized. In this extreme case, the pattern is expressed as a hexagonal lattice, since this type of distribution is the most efficient way to pack a space or volume. Of course, a distribution may be classified as either regular, random, or aggregated in the statistical sense without totally satisfying the conditions above. - 14 -The dispersion of animals in space results from interactions with the physical environment as well as the presence or absence of other individuals (Brown & Orians, 1970), Since for a public beach I have assumed that the environment is structurally uniform (isotropic) then any changes in the spatial pattern as exhibited by beach users can be expected to result from largely social as opposed to environmental sources. If users of a public beach are viewed as having spatial needs which are manifest as culturally or biologically appropriate inter personal distance, then one might expect that above a certain overall density the spatial pattern exhibited by the population would bo uniform as people strive to maintain the minimum amount of space which they re quire. The process by which this might occur is easy to visualize since each individual or group arriving at the beach would strive to gain at least the minimum amount of space which was required. Contrarily, at low densities each group could obtain much more than this minimum amount and thus there would be no psychological 'pressure' from other groups that would influence the positioning of new arrivals. Thus at these densities differences related to individual personality characteristics could be manifest. Since the expression of these charateristics may be visualized as many and varied, a pattern approaching random should be observed. _ In a study cited previously, Brougham (1968) proposed a similar argument to the one above. Through the use of oblique aerial photographs over a one day period, Brougham sought to examine the effects of density on spatial pattern and perceived crowding at Pinery Provincial Park beach in Ontario. His results seem to indicate that beach users did attempt to max imize the space available to them as evidenced by an 'R' value (a measure of the extent to which the distribution of objects in space conform to one of three patterns, random, regular or aggregated) significant in the direction of a regular pattern. Inexplicably, the 'R' values (although significantly different from a random pattern) tended to decrease as - 15 -density increased. These results are not conclusive however, since the use of oblique photographs may not have produced reliable measurements, and since the photographs were taken for a single day only, the sample size ,as well as the range of observed densities were relatively small. Since densities were expressed in relative as opposed to absolute units direct comparisons between his study and the present research are not possible. To test these arguments, an analysis of the spatial distribution of beach users examined the following corollary of the first hypothesis mentioned previously: - The distribution ofa groups over the beach surface will approach a random pattern at low densities and as den sity increases the distribution of groups will exhibit an increasingly regular pattern. In summary, I predicted that since other studies have shown that humans and other social animals maintain certain spatial requirements which are related to proper social functioning, there should exist a den sity range over which people on a public beach would maximize the space between themselves and neighboring groups thus resulting in a uniform spatial pattern. - 15a -CHAPTER II INDIVIDUAL DIFFERENCES AND SPATIAL BEHAVIOUR: THE ROLE OF PERSONALITY AND SOCIO-ECONOMIC CHARACTERISTICS. - 16 -Introduction. The hypotheses listed in the preyious section were generated to aid in determining how entire populations of beach users behave. How ever, describing aggregate behaviour does little to explain hw differ ences between individuals influence observed patterns of behaviour.. ' In order to determine some of the factors influencing the spatial behaviour of individuals, I chose to administer a survey to randomly selected samples of beach users at a variety of different densities. The choice of survey instruments included tests designed to assess a broad range of environ mental dispositions, mood variables and socio-economic and demographic charac teristics. Besides interpersonal distance measures, I also chose to inves tigate how the personal attributes referred to above relate to the area circumscribed by an individiial's or group's personal possessions (marked group area). These objectives were oriented toward gaining an understand ing of psychological and socio-demographic characteristics as related to various aspects of beach user spacing and group behaviour. The role of personality. The literature relating to personality correlates of spacing behaviour is substantial. In general, however, the studies show little coherence, with lack of theoretical underpinnings being the most likely cause (Altman, 1975). Other than Hall's (1966) qualitative observations and cross-cultural comparisons few models exist which attempt to explain the role of personality in personal space preferences. An early model by Argyle and Dean (1965) proposed an equili brium hypothesis which suggested that behavioural shifts occur to main tain desired levels of intimacy and social interaction. Thus such behav iours as eye contact, body orientation, facinl expressions, etc., operate to create desired interpersonal distances which they suggested are commensurate with the type of social interaction involved. - 17 -An additional attempt to develop a theoretical approach to spacing behaviour was proposed by Duke and Norwicki (1972). They sug gested that appropriate distancing behaviours are related to social-learning models and that reinforcements act as the driving force for learning culturally defined spacing norms. Altman (1975) suggests that spacing behaviour is, "one of a series of self/other boundary mechanisms that function in the service of desired levels of interaction." Central to Altman's hypothesis is the concept of privacy as a boundary control process. According to Altman, one of the ways in which people achieve desired levels of privacy is through the use of space. Personality correlates. Of the specific studies relating personality and spatial behaviour, only two areas maintain any degree of consistency. The first concerns the effect of anxiety on interpersonal distance. In general, measures indicating high levels of anxiety correlate with increased per sonal distance (Smith, 1953, 1954; Luft, 1966; Weinstein, 1968; Patterson, 1973; Karabenich and Meisels, 1972; and Bailey, Hartnett, and Gibson, 1972). The second area of research where personality attributes have been related to spacing behaviour comes from studies of the introversion/ extroversion complex. For the most part, subjects scoring highly on measures of extroversion are observed to maintain gloser individual distances than those with elevated introversion profiles (Williams, 1971; Cook, 1970; Patterson and Holmes, 19G6). In another study which related scores on "exhibitionism" and "impulsivity" scales, Sewoll (1973) reported a significant negative correlation between tin? personality measures nnd distance. Contrary results, however, were obtained by Meisels and Canter (1970). - IR -Finally, a study dealing with attitudes and perceptions of crowding on a public beach underscores the potential importance of person ality characteristics of users of a recreational resource. This study, conducted by Meyer and Bryan (1974) at Long Beach, Vancouver Island, British Columbia attempted to correlate user responses relating perceived crowding to site density. They found that most respondents felt the num ber of people at their site was "about right". Since Long Beach is ex tensive with many sites available, Meyer and Bryan concluded that users may have selected the site which was consistent with personal crowding preferences. Although not specifically tested by Meyer and Bryan, this ex planation is central to questions relating to the present research. Implicit in the prediction that people select the density which is comen-surate with spacing needs and preferences is the notion that site selec tion is mediated through various personality processes within the indiv idual. In this way, a user by having previous knowledge of when a beach was less or more crowded could choose the time of day or day of the week most likely to fulfill basic internal needs. Similarly, once at the beach the user may choose a section which is more or less crowded and situate at a comfortable distance from neighbors. The key question regarding the discussion above is: "What are the most likely characteristics of the user which influence such decisions?" Since the literature contained no guiding theory and few studies leading to such theory I chose to use scales derived principally from the field of environmental psychology. This decision was based on the argument that the ways in which people perceive and respond to the physical environment may offer valid insights into other behaviours relating to interpersonal functioning. For example, with reference to the present study, a person who manifests a positive orientation toward the high density urban envir onment should be more likely to be observed at the beach during high den sity periods or at shorter distances than someone who prefers the quiet and solitude of a more rural environment. Similarly, a person who affirms a preference for highly stimulating environments or activities should be observed in closer proximity to others and at higher densities than a respondent who is typically fearful of such environments. 'The main thrust of this component of the study was to determine whether a variety of personality and socio-economic characteristics of beach users are related to spacing preferences. Since I predicted earlier that due to the pressures associated with high density conditions, nearest neighbor distance would be constant at these times, the above preferences could only be effectively manifest at lower densities when choices are not inhibited by crowding influences. Thus the prediction is that at lower densities users may select*a site based on preferences mediated by personality characteristics, whereas at high densities site selection is a function of adaptational processes directly related to gaining the min imum amount of space required for controlling social interation between neighboring groups. The survey instruments. The survey instruments which I chose were designed to measure a respondent'sorientat ion toward various aspects of: 1) the physical and to a lesser extent social environment, 2) the recreational environ ment, and 3) his own internal "moods" within the setting. An additional section elicited background information designed to tap important socio economic and demographic characteristics of the user. Two of these tests, the Environmental Response Inventory (ERI) and the Leisure Activities Blank (LAB) were developed by George McKechnie (1974). The mood scale was designed by Lorr, Daston and Smith (1967) and the background section was developed for the study by the author. A facsimile of the survey occurs in Appendix A. - 20 -The Environmental Response Inventory. The ERI was specifically designed to assess what Craik (1966, 1969, 1970a, 1970b) has termed "environmental dispositions." Environmental dispositions are defined as relatively enduring psychological dimensions which are, used by the individual to describe and evaluate various aspects of the physical environment. The ERI consists of 184 statements which tap a diversity of environmental themes, most of which relate to the non-human environment. The remainder relate to various aspects of the human social environment. The inventory yields scores on eight scales plus one test re liability scale (termed communality) designed as a validity check for response bias. The nine scales and McKechnie's (1974) description of each are listed in Table 2. To facilitate the process of enumerating the hypotheses derived for the study I have also included in Table 2 a sign indicating the predicted direction of the correlation between each scale and the distance from a respondent and his or her nearest neighbor. In this way for example, a respondent scoring highly on the Urbanism scale is thus predicted to be found nearer than average to the closest neighboring group. This is based on the argument that people reporting a positive orientation to high density urban environments should tolerate or even seek out settings where crowds are likely to occur. Similar arguments can be generated for the other eight scales as well. A more specific discussion of the scales and response format occurs in the section on methods. The Leisure Activities Blank. The second part of the survey (Leisure Activities Blank, LAB) consisted of a comprehensive range of leisure and recreational activities which subjects responded to on the basis of past participation for each item. Through factor analysis McKechnie (1974) developed seven scales which he broadly classified as: Mechanics, Crafts, Intellectual, Slow Living, Neighborhood Sports, Glamour Sports, and Fast Living. Table 3 lists the activities and their factor loadings for each of the seven scales. The choice of the LAB for the present study was based on the argument that the activities a person voluntarily chooses to participate Table 2: Environmental Response Inventory Scales (adapted from McKechnie, 1973) Scale and Major Themes: High Scorers often Described as: Low Scorers often Described as: PASTORALISM. (+) Opposition to land development: concern about population growth: preservation of natural forces as shapers of human life: sensitivity to pure environmental experiences: self-sufficiency in the natural en vironment. URBANISM. (-) Enjoyment of high density living; appre ciation of unusual and varied stimulus patterns of the city; Interest in cultural life; en joyment of interpersonal rich ness and diversity. ENVIRONMENTAL ADAPTATION. (+) Modification of the environ ment to satisfy needs and desires, and to provide com fort and leisure: opposition to government control over private land use: preference for highly designed or adapted environments: use of tech-nolofry to solve environmental problems: preference for stylized environmental details. STIMULUS SEEKING. (-) Interest in travel and exploration of unusual places: enjoyment of complex and intense physical sensations; breadth of interests. Aesthetic, affectionate, compli cated, distractible, outspoken, progressive, rebellious, uncon ventional, unpredictable, selfish. Critical, skeptical, responsive to urban aesthetics, high-brow, concerned with philosophical problems in life, valuing Intel-tectual activity, managerial in terests. Autocratic, condescending, con servative, efficient, inter-prising, extraverted, hard-headed, mannerly, methodical, power and money oriented, judgmental, aesthetically unresponsive. Advonturous, disorderly, dis tractible, dreamy, easy-going, immature, impulsive, progressive, unconventional, ^independable. Apathetic, conscientious, con servative, conventional, delib erate, dependable, friendly, honest, practical, self-con trolled. Conscientious, conventional, friendly, generous, non verbal, opportunistic, robust, simple, unselfish. Artistic, awkward, compas sionate, curious, distractible, idealistic, introspective, moody, non-conforming, sensitive, sensuous, worrying, forthright. Conscientious, conservative, fastidious, practical, res ponsible, rigid, severe, stingy. Table 2 (continued) Scale and Major Themes: ENVIRONMENTAL TRUST. (-) General environmental openness, respon siveness, and trust, competence in finding one's way about the environment . vs Fear of poten tially dangerous environments: security of home: fear of being alone and unprotected. ANTIQUARIANISM. (-) Enjoyment of antiques and historical places: preference for tradi tional vs modern design: aesthetic sensitivity to man-made environments and to land scape; appreciation of cul tural artifacts of earlier eras; tendency to collect ob jects for their emotional significance. NEED FOR PRIVACY. ( + ) Need for physical isolation from stimuli: enjoyment of soli tude; dislike of neighboring; need for freedom from dis traction. MECHANICAL ORIENTATION. . (+) Interest in mechanics in its various forms: enjoyment in working with one's hands, interest in technological processes and basic prin ciples of science: app reciation of the functional properties of objects. COMMUNALITY. ( + ) A Validity scale, tapping honest, attentive, and care ful test-taking attitude; response to items in statis tically modal manner. High Scorers often Described as: Low Scorers often Described as: Capable, competent, diligent, efficient, helpful, ingenious, resourceful, stable, thorough, tolerant, well-adjusted. Bitter, cold, coarse, dis satisfied, distrustful, intol erant, moody, prejudiced, spendthrift, unkind. Affectionate, artistic, change able, dependent, dreamy, emotional, forgiving, ideal istic, introspective, aes thetically reactive, warm. Coarse, cool, conservative, deliberate, mischievous, moralistic, practical, sky, stolid, unemotional. Aloof, arrogant, autocratic, bitter, cold, formal, hard hearted, sulky, polished, resentful, stubborn. Appreciative, cooperative, easy-going, friendly, seeking reassurance, warm, seeks ac ceptance, lacks confidence, introverted. Arrogant, conceited, ego tistical, hard-hearted, mas culine, self-seeking, in flexible, sociable, mani pulative . Affectionate, feminine, generous, sincere, under standing, submissive, sym pathetic, warm. Calm, civilized, initiative, mannerly, patient, tactful, trusting, rule-following. Hard-headed, flirtatious, good looking, immature, opportun istic, versatile, witty, inde pendent-minded, psychologically complex. Table 3: Seven LEISURE ACTIVITIES BLANK - Past Factors (adapted from McKechnie, 1974) Factor 1: Mechanics Factor 2: Crafts Loading* # Item Loading * •# Item .327 2 A inn tour radio .440 22 Ceramics or pottery .353 6 Auto racing .289 27 Collecting things .722 7 Auto repairing .521 29 Cooking and baking .455 13 Billiards or pool .301 30 Crossword puzzles . 516 18 Boxing .446 31 Dancing (ballet, mod) .311 19 Camping . 506 34 Designing clothes .683 21 Carpentry . 476 43 Flower arranging .488 37 Electroni cs .354 45 Folk dancing .405 41 Fishing (saltwater) . 415 57 Home decorating .484 42 Fishing (fresh) .455 63 Jewelry making .246 44 Flying (or gliding) .412 64 Jig-saw puzzles . 423 60 Horseshoes . 539 69 Knitting-crocheting . 575 61 Hunting . 351 70 Leatherwork .682 72 Marksmanship .603 79 Needlework . 829 73 Mechanics .435 80 Painting and drawing .709 74 Metalwork .351 91 Sculpture . 469 75 Model building .641 92 Sewing .336 81 Playing poker .493 115 Weaving . 524 111 Vol. fire fighting .410 116 Weight lifting . 523 118 Wrest 1ing .622 121 Woodworking Factor 4: Slow Living .413 32 Social dancing Factor 3: Intellectual . 422 35 Dining out . 413 36 Driving .329 1 Acting (dramatics) .323 39 Exercising .631 4 Attending concerts .430 49 Gardening .344 8 Backpacking . 497 50 Going to movies .357 24 Chess .380 71 Listening to radio . 442 26 Civic Organizations .324 " 83 Playing records .328 33 Darkroom work .331 87 Reading: light .705 51 Going to plays .329 94 Sightseeing .424 56 Hiking or walking .476 98 Social drinking . 246 84 Musical instruments .473 100 Sunbathing . 541 85 Political activities ,288 104 Taking pictures .426 86 Reading: serious .450 105 Talking on telephone .2 51 95 Singing .472 109 V i s1t i ng f ri end s .353 107 Travel abroad .340 112 Watch team sports . 548 108 Vi s i t ing museums .453 113 Watch TV shows . 4 28 119 Writing poetry, etc. .438 117 ' Window shopping .494 28 Conservat ion-ecology .347 120 Writing letters *A11 factor loadings reported here are positive. - '/A -Table 3: continued Factor 5: Neighborhood Sports Factor 6: Glamour Sports Loading * ff Item Loading * # I tern .407 9 Badminton .356 3 Archery .628 10 Baseball .430 15 Boating (rowing) .644 11 Basketball .439 20 Canoeing .355 12 Bicycling .339 59 Horseback riding .324 17 Bowling . 275 62 Ice Skating .370 23 Checkers or Go . 498 , 76 Motor Boating . 506 46 Football .372 77 Motorcycling . 402 65 Jogging .376 78 Mountain climbing .338 68 Kite Flying . 551 90 Sailing .226 93 Shuffleboard i 550 96 Ski ing . 436 99 Squash or Handball .350 97 Skindiving .389 103 Ping pong a .455 101 Surfboard ing .540 110 Volleyball . 410 102 Swimming .381 106 Tennis Factor 7: Fast Living . 583 114 Water skiing .284 47 Fraternal organizations .419 48 Gambling (casino) .442 52 Going to Horseraces .354 53 Going to Nightclubs - 25 -in during periods of leisure time may be related to other personal psy chological characteristics. Evidence relating to this argument stems from McKechnie's (1974) study where he was able to correlate each LAB factor with various Environmental Response Inventory scales, socio-economic and environmental attitude variables. Based on these results he types high scores on each of the LAB factors in the following manner: Mechanics: ". . .a rugged, mechanically-minded male, who enjoys the outdoors, working with his hands and getting away from home for periods of time." Crafts: "... a woman who enjoys doing things at home: decorating the house, making clothing for the family, or engaging in other activities to make the home a cozy and emotionally satisfying place." Intellectual: "A high scorer. . . is from an educationally and econ omically priviledged sector of society, enjoys the natural environment and desires to preserve it, and devotes his leisure time to pursuing this and other worthwhile community goals." Slow Living: ". . . a person for whom the home is a refuge from com muting to and from a white collar job, who might relax by sett ling down on the patio and passively enjoying his periods of lei sure." Neighborhood Sports: ". . . a young, well educated male who enjoys the outdoors so long as some sort of playing field is nearby and a game is on." Glamor Sports: "The person scoring high on Glamor Sports seems to like getting out in the environment and enjoying the intense stim ulation that such activities as motorcycling, waterskiing, and sailing can afford. The high scorer on the factor is typically a young, well educated male; he is pro-conservation and enjoys sports equipment as a means of stimulating environmental exper ience." Fast Living: * Not typed by McKechnie since the factor had but four item definers. Although McKechnie does not relate the LAB factors to personality traits per se the typology which he derives does show how people differ according to their individual leisure activity patterns. A study which did relate leisure activity patterns to person ality traits was conducted by Lamphear (1970). He noted that subjects with "normal" MMPI (Minnesota Multiphasic Personality Inventory) scores maintain recreation patterns which are significantly different from those with elevated profiles. For the present study, predictions of spacing behaviour as it corresponds to scores on the LAB were made based on the extent to which activities within a factor were predominantly oriented toward solitary or individual, pastimes versus group activities of a more gregarious nature. This criterion led to the following predicted correlations between LAB factor scores and respondent nearest neighbor distance: Mechanics, Crafts, Slow Living, and Glamour Sports - larger distance to nearest neighbor; Neighborhood Sports and Fast Living - smaller distance to nearest neighbor. A prediction of spacing and the Intellectual factor was not made since the activities seemed to be evenly split with respect to the selection criterion Mood Adjective Checklist. The mood scale consists of sixty adjectives which, when sub jected to factor analysis by Lorr, Daston & Smith (1967) produced eight identifiable mood factors (Table 4). These they termed: Cheerful, Energetic, Anger-hostility, Tense-anxious, Depressed, Inert-fatigued, Thoughtful, and Relaxed-composed. Since the other portions of the survey were included to assess more enduring psychological dimensions which might relate to spatial behaviour, the mood adjective checklist was in serted as a way of measuring more momentary and transient:aspects of a subject's psychological profile. In this way I hoped to determine the extent to which a respondent's mood was influenced by the level of spatial press due to the proximity of others. One might expect, for example that a respondent who had chosen a site well away from more crowded por tions of the beach and who had been subsequently intruded upon by another group would show elevated scores on the more "negative" mood variables. Although, it was not possible to know when such a scenario occurred the correlation between the mood variables and distance measures would indicate, in a relative way, the extent that this and similar situations prevailed. A final aspect of the survey related to the decision to make the mood checklist an optional feature of the questional re package. Since the time required to complete the survey was lengthy (about 35 - 40 min utes), I reasoned that if a respondent was not enjoying the beach exper ience prior to filling out the survey then he would be less likely to Table 4: Correlations of the Adjectives with Eight Mood Factors (Adapted from Lorr et a.l, 1967). ' Factor 1: Cheerful Factor 2: Energetic Loading ff Item Loading ff I tern .70 •Elated .62 1 Active .69 35 On top of the world . 56 42 Energetic .60 6 Excited .54 38 Full of pep .56 39 Light-hearted .53 50 Alert .56 49 Carefree .51 24 Vigorous ,.54 12 Gay .50 55 Lively . 52 2 Cheerful .44 47 Enthusiastic .51 34 Happy-go-lucky .34 9 Pretty good .33 58 Optimistic „ Factor 3 : Anger-Hostility Factor 4: Tense-Anxious .68 27 Furious .59 10 Nervous .67 13 Annoyed .53 51 Anxious .65 5 Angry .39 53 Shaky .45 54 Spiteful .36 59 Worried .45 15 Resentful .36 3 Jittery . 44 48 Ready to fight .31 26 Tense .41 7 Bad-tempered .30 9 On edge .33 52 Grouchy Factor 5 : Thoughtful Factor 6 : Depressed .62 30 Introspective .61 36 Hopeless . 58 33 Thoughtful .59 16 Helpless . 56 22 Contemplative . 57 19 Worthless . 55 11 Pensive .36 46 Unhappy .40 14 Earnest .32 44 Lonely .35 25 Serious .29 56 Blue .32 32 Preoccupied .28 20 Frightened .26 8 Apathetic Factor 7 : Inert-Fatigued Factor 8: Relaxed-Composed .66 37 Weary .59 21 Calm .66 40 Tired .52 45 At ease .43 17 Sluggish .44 43 Composed .38 60 Lethargic .44 41 Relaxed .38 31 Lazy .34 18 Serene .25 57 Listless .29 23 Nonchalant .28 28 Languid •'Elated' deleted from present survey as it was mis-typed 'hated'. - 28 -complete the final segment if given a choice. Thus to the extent that crowding influences are related to a decrement in user satisfaction, I predicted that a respondent who chooses not to fill out the mood survey would be found at high densities and thus small nearest neighbor distances. Socio-economic and demographic characteristics. Questions relating to a respondent's socio-economic background were included to determine the extent to which such variables as age, sex, marital status, income, etc. were related to spatial and group behaviour as well as a way of describing the sample. - 28a -CHAPTER III An Approach to the Study of Human Spatial Behaviour: METHODS AND APPLICATIONS The study areas. Three beaches were chosen ns suitable environments for the pur poses of the study. The sites chosen were relatively distinctive thus ensuring as comprehensive a data base as possible. Two of these areas, English Bay beach and a grassy, sunning area near Kitsilano beach are located near the center of Vancouver, British Columbia (49° 17' No. Lat., 123° 8' Long.). The third site is located on Skaha Lake near the city of Pentiction, British Columbia (49° 27' No. Lat., 119° 36' Long.). Maps depicting- the three areas are shown in figures 1 and 2. English Bay is a gently curving, sand beach with the ocean along the western edge. The beach is characterized by two distinctive areas, one of which contains logs placed in parallel rows by the local parks board. Beach users utilize these logs as back supports and as a result,distributions of users in this area tends to be linear. This area runs parallel to the shoreline and is situated on the landward half of the beach. The other area nearer the ocean, has no logs and is thus free from such environmental influences. Since the research required a uniform envir onment it was this latter site which was chosen as the study area. The dimensions of this section of the beach are approximately 470 meters by 12 to 43 meters depending on the level of the tide. The average area as calculated from the aerial photographs was 1.20 hectares. The site also contained a centrally located beach house/refresh ment stand. Access to the beach was varied with some on street parking and a parking lot located near the south-east end. The second study site, Kitsilano, is a complex area of ocean fronted beach containing "backrest" logs and two adjacent rectangular sunning areas covered with grass rather than sand. The southern-most sunning area was chosen because of its uniformity, wide range of density (over time) and basic rectangular shape. The area is 146 x 31 meters (.453 hectares)and is virtually free from physical obstructions such as back rest logs which might act to influence spacing behaviour. A small exception is a pathway slanting diagonally across the east end. Entrance is open except on the north and south where a seawall and a fence respec tively limit direct access. A parking lot exists near the east end and a refreshment stand is situated on the west. - 30 -Figure 1. Map depicting Kitsilano and English Bay study areas. - 31 -Figure 2. Map depicting Skaha study area. The study site at Penticton (Skaha beach) is the middle of three beaches on the north shore of Skaha Lnke. The study area dimensions are 260 x 25 meters (.54 hectares). A refreshment stand is located on the landward side near the mid-point and access is not limited. Skaha was chosen because of its uniform character and because most users are vacationers and originate outside the immediate area. In fact the survey revealed that over 95% of all respondents did not reside in or near the city of Penticton. In summary, the three sites were chosen for their essentially within beach uniform character, although certain aspects such as substrate type and refreshment stand location differed between beaches. Aerial Photography. Data concerning spatial behaviour and group phenomena were col lected via aerial photography. This technique although complicated and prone to mechanical problems was chosen over others because of its ability to produce large quantities of date quickly as well as the relative ease with which the information can be digitized for.computer analysis. Another important aspect was that data could be collected unobtrusively so that all behaviour is observed as 'natural' and thus uninfluenced by the observer. Before outlining the methodology, three terms are operationally defined for purposes of clarity: 1) group - any set of interacting in dividuals situated in close proximity to one another so as to form an eas ily recognizable unit (for most purposes a lone individual is also labelled a group except where it is important to distinguish between single persons and larger numbers of people), 2) run - any aerial photographic pass over the study area which resulted in a complete record of the beach and Its users, and 3) boundary - the geographical limits of the beach or sunning area except in the case of English Bay where only that portion of the beach between backrest logs and the water was used. Although aerial photography is often a complex and rather costly undertaking, a method was devised which satisfied the requirements of the research and was at the same time only moderately expensive. Of the 27 runs finally used with the analysis, 23 were completed without the use of commercial aerial photographic techniques or equipment. The method entailed the use of a light aircraft (Cessna 150) - 33 -converted for aerial photographic purposes by removing both doors. This practice provided the required visibility for both pilot and photograph er where the key to a successful run was the maintenance of the proper flight path directly above the study area. The camera, a 35 mm, motorized Nikon, equipped with a 135 mm lens was hand held and set to maintain a film transfer rate of 2.5 frames per second. At a flight altitude of 305 meters (1,000 feet), this equip ment provided a good image of people and most of their beach articles. With an airspeed of 113-145 Km per hour (70-90 miles per hour) there was sufficient interframe overlap to ensure a complete record of the beach on any given run. In order to scale the photographs for any given run three ob jects such as sidewalks, diving platforms, slides, etc, were chosen as environmental features easily visible within the photograph and permanent enough not to be moved during the duration of the study. Two of the ob jects were located at the respective ends of any given beach and a third was situated in the middle. A conversion factor wa s obtained for each of the three objects by dividing the known ground dimension of the ob ject by its corresponding image dimension. In order to minimize errors due to altitude fluctuations the three conversion factors were averaged to obtain the final conversion value. These errors were minimal as the difference between the two most different values for any given run, was most often two to three percent and only once approached ten percent. All film analyses were completed using a Vanguard motion anal yzer and DEC 11/45 computer. The motion analyzer is designed so that a single photographic frame is projected onto an opaque glass screen. Two thin wires running at right angles to each other act as cross-hairs and their movement is controlled by the rotation of two knobs on the console. By moving the cross-hairs over the screen to tip desired point and by activating a switch, the X, Y co-ordinates of the point are trans ferred onto computer compatible paper punch tape. By digitizing around the perimeter of a group or individual as indicated by their physical belongings I was able to construct a representation of each group's spatial boundaries. This process was carried out for every group on the beach. The program designed for the project connects the points for any given group in such a way as to construct an irregular polygon. The area subtended by each polygon was defined as the 'marked group area' for any given group. Since the motion analyzer was not infallible, all polygons (group areas) were plotted using a standard calcomplotter and as a result of this process large errors due to digitizer malfunctions could be detected by visual inspection of the completed maps. Since it takes many frames to compose one beach image, it was necessary to subtract the overlap from each pair of frames. This was done by digitizing six recognizable points (objects on the beach surface) within each frame, three at the 'top' and three at the 'bottom'. Each set of three points was picked such that one was at the extreme left of the frame, one in the middle and one on the extreme right. After the analysis of that frame the 'lower' three points were found on the 'top' of the next frame and their positions placed on the tape. At this time the points were chosen and punched for the next frame at the 'bottom' of the frame currently being analyzed. The three distances between corres ponding points on adjacent frames were then averaged. This average dist ance between corresponding points on adjacent frames was thus the amount of overlap for each frame. Averaging the distances of the widely separate points was done to minimize errors due to lens abberation and possible deviations resulting from the aircraft not maintaining its position dir ectly above the study site. These errors were thought to be samll since a high quality lens was used and care was taken to maintain a flight path directly above the beach. In addition to the digitizing process the number of people in each group was entered into the record. For this study it was generally easy to determine what did and what did not define a group, although one can visualize an area so crowded that group definitions by purely photo graphic means becomes difficult. Such densities were not observed and in most circumstances the placement of beach articles was sufficient to dem arcate one group from another. In order to determine the overall beach density during a run the area of each beach was required. This was obtained by digitizing around the perimeter of that portion of the study plot represented on one frame. These points were then taken as the vertices of a polygon and the area computed. - 35 -This process was completed for each frame with care taken to superimpose adjacent sides of each pair of contiguous polygons. Again these values are plotted to validate data transcription and the result ing polygons fitted together to represent the outline of the study area. This process was necessary once only for Skaha and Kitsilano . However, since English Bay is located directly on the ocean, a. separate area compu tation for each run was required due to/changes in the tide level. Pattern Analysis. The analysis of pattern originated with and has been developed primarily through work of ecologists and biometricians. Gleason (1920) was the first to develop a method describing pattern type using sample quadrats and the Poisson series. Criticism of techniques utilizing quadrat methods center around the influence of quadrat size on frequency data (Curtis and Mcintosh, 1950; Skellam, 1952) and because of these criticisms, newer techniques were used for this study. Another technique widely used by ecologists is the distance to nearest neighbor technique. This method of pattern analysis was origin ated by Dice (1952) and subsequently elaborated upon by Skellam (1952), Clark and Evans (1954), Morisita (1954) and Thompson (1956). This tech nique was used for the present study since it is the most accepted and widely used method of pattern analysis and since nearest neighbor dis tances were easily calculated from the aerial photographs. The method as described by Clark and Evans (1954) consists of measuring the distance between an individual and his nearest neighbor, where individuals are chosen by some random process. An alternate method involves calculating the distance between individuals and their nearest neighbors for all members of the population. Of course, this variation only applies when the population is discrete and small enough that such measurement is feasible. Such was the case for this study, so that near est neighbor distances for all groups on a beach for any given run were measured. Here a "run" refers to a photographic sequence of the entire length of a beach. The pattern satistic *'R' is defined by Clark and Evans (1954) from the ratio of the observed to expected mean nearest neighbor distance 36 -such that R= r /v , where r is equal to the mean observed nearest nelgh-o e o bor distance and r is equal to the mean expected nearest neighbor distance. e -The mean expected nearest neighbor distance (r ) is the mean distance e ' which would be expected if the population in question were distributed at random. Clark and Evans show that r is equal to 1/2 \/d~7 where d equals e the density in individuals per unit area. The value of R is shown by Clark and Evans to exhibit a limited range with a lower limit of zero and an upper limit of 2.1491. Thus perfectly random, aggregated or regu lar patterns are described respectively, by R values of 1, 0, and 2.1491. Maximum aggregation occurs when all members of a population fall on the same locus and thus the mean distance to nearest neighbor is zero. Per fect uniformity exists when inter-individual distance is maximized. Un der these conditions a hexagonal pattern is formed and each member of the population (except those at the periphery) will be equidistant from six other individuals. A test of the significant departure of from r is assesed by letting Z equal the standard variate of the normal curve such that Z=(r - r)/rjr, where O r equals the standard error of the mean dis-o e e e tance to nearest neighbor in a randomly distributed population of the same density as the observed. The standard error (O r )' as derived by Clark and Evans (1954) is expressed as: . a r ' = 0.26136/ /nd , e where n equals the number of measurements and d is the density. Another study of spacing behaviour of beach users has shown that the upper limit of R is influenced by linear environments such as beaches when densities are low (Brougham, 1968). This is based on the fact that under such conditions the primary assumptions associated with the nearest neighbor distance statistic: are violated. The derivation of the formulae for r and also the upper limit of the R statistic, 2.1491, are based e upon the assumptions of an infinite number of points and an unbounded surface. These assumptions are rarely, if ever met in practice. However, when the violation is extreme such as in Figures 3, 4, and 5a hexagonal distribution does not maximize the spacing between points. r 37 -Figure 3. The distance between points is not maximized by a hexagonal pattern. (Adapted from Brougham, 1968) • • • Figure 4. A pattern of equilateral triangles maximizes the inter-point distance. • • • • • Figure 5. With more points in the same area, a pattern of squares maximizes the distance. -38 -In a test of the effect of a linear, bounded surface with few points (81 points in an area 40" x 0.866")' Brougham found that with a distribution similar to the one in Fig. 3, an R statistic of 3.0590 was obtained, a considerable deviation from the theoretical maximum value of 2.1491. Since English Bay, Kitsilano and Skaha beaches reflect vary ing degrees of linearity and since much of the work depended upon an accurate assessment of pattern, the following procedure was developed: Because re must be an unbiased estimate of the mean distance between points scattered by some random process, a Monte Carlo simulation routine was used to place points across a rectangular representation of each beach. Although English Bay beach was not exactly rectangular, it was so considered for the purpose of the simulation. Errors introduced by this simplification seems small since the beach is a long, gently curving beach and is probably perceived as rectangular by users. For each run, the length and average width (with the same area as the actual beach in question was entered with the observed number of groups. For example, English Bay run number ten was observed to maintain an area of approxim ately 12,690 square meters (a length of 470 meters and an average width of 27 meters). For purposes of the simulation, the values used were the untransformed screen dimensions, i.e. the dimensions as taken from the projected image on the motion analyzer. The above values were: length = 152 cm., average width = 9 cm. and area = 1,368 square cm. The number of groups for run number ten as observed from the film was 76. Given the area dimensions and the number of groups the model was programmed to scatter these groups as points (centroids of polygons) in a random fash ion over the available area. This process was iterated 100 times and for each iteration the average observed one to four nearest neighbor distances was calculated. This process was completed for all 27 runs and as a result of the iterative procedure each simulation produced a sampling distribution of means and therefore a reliable estimate of the true pop ulation mean nearest neighbor distance for each of the four orders. In addition, the process allows one to calculate the standard error of sample means ( a r ) to be used in place of the theoretical value as derived from s the Clark and Evans method. The simulated mean nearest neighbor distance - 39 -for each run then becomes the expected value, r , for that same run. e The Clark and Evans model is therefore: R = r /r OS where r equals the mean of the simulated sampling distribution of the mean. The results indicated'a difference between the simulated and theoretical values especially at low densities and thus the method seems of worth for studies with similar physical constraints. Since a large number of R values were calculated, an effort was made to simplify comparisons between runs by normalizing R according to the formula: Z„ = (r - r )/ n r R o s G s Using this formula any Z value between - 1,96 indicates a random spatial pattern at the 0.05 probability level. Any value greater than 1.96 was considered a reflection of a regular pattern and any value less than -1.96 was considered clumped or aggregated. Two strategies were employed to calculate the distance to near est neighbor: (1) the centroid of each polygon was calculated and used as a point source for the 'R' statistic, and (2) 'nearest approach distance' calculated as the distance between the edges of any two nearest neighbor polygons. I predicted that the latter distance would be mofe responsive to any possible interactive effects due to psychological variables than the dis tance between group area centroids. This prediction was based upon reas oning which supposed that an individual or group probably decided where to 'settle' on the basis of distances between edges of groups rather than on center to center distances. The nearest approach distance was used as a dependent variable in that portion of the study designed to determine possible spatial group correlates of the various psychological indices; how ever, its use in the analysis of pattern was precluded for theoretical rea sons. For example, in a recent paper Mohn and Stavem (1974) showed that for a Monte Carlo simulation of randomly sp aced discs (red blood cells) in a haemocytometer, the poisson, binominal and hypergeometric models provided poor fits to the data. Although two of their models fitted the empirical results reasonably well, the fact that the sizes of the discs were relatively - 40 -uniform made their use in the present study difficult since the group areas were thought to be too variable to be applicable to their models. Space-time study. Since external features of the beaches such as bath houses, parking lots and refreshment stands seemed to exert a certain amount of influence over the general distribution patterns of beach users, an effort was made to determine the relative effects of these aspects of the beach environment. To maximize the range of observed densities the study took place at the Kitsilano site on a Saturday when crowds were expected to be large. As mentioned previously, the Kitsilano study area extends in an east west direction with a parking lot near the east end and refresh ment stand adjacent to the west end. Since these two features were at opposite ends of the beach, it was possible to determine their relative effects. To determine the spatial distribution of is ers as the day pro gressed, hourly maps were made. To facilitate the mapping, the study site was divided into 20 quadrats 15.2 meters on a side with the outside corners marked with engineer tape. Maps were drawn to scale and at the appointed time the position and number of people in each group were placed on the sheet. The first census commenced at 1030 hours and a final count was made at 1530 hours, making a total of six separate enumerations. Plots of these data thus provided a time-series of the placement of each group over the study area. In this way effects due to environmental fea tures such as the refreshment stand, parking lot and swimming pool could be ascertained. Survey dissemination. Two main factors influenced the choice of sampling methods for this phase of the research. First, since one of the primary goals of the study depended upon the coincident gathering of both overt and behavioural data relating to the distribution of beach users as well as subjective responses concerning various psychological variables, it was necessary to coordinate the aerial photography with the dispersal of the survey - 41 -booklets. This was accomplished by having the person distributing the surveys (surveyor), telephone the airport when the appropriate density was observed. At this time, the surveyor began distributing the booklets. When this was completed, a large brightly colored marker panel was sit uated in a predetermined spot thus signalling the aerial photographer to begin the photographic run over the study area. This method was found to work well except for low density situations. On most: days the length of time which passed between low to medium density conditions was so short that a surveyor might begin handing out surveys at a low density, but by the time the aerial photographs were taken, enough new beach users had arrived as to make the density fall within the medium range. This problem was largely overcome by sampling on weekdays when the beach did not fill as fast. The second aspect of the study areas which determined sampling procedures was the extreme length to width ratio of two of the three study sites. Because of this problem English Bay and Skaha were sampled in a slightly different manner than Kitsilano. The method developed for Kitsilano consisted of dividing the sunning area into six equally spaced 'lanes' which ran the length of the study site. Prior to distributing' the booklets, the surveyor picked a number between one and six from a hat and used the corresponding imaginary line as the sampling transect. The surveyor then proceeded to distribute the booklets by pac ing a prescribed number of paces. The person chosen was the closest individual in.the closest group within the forward 180° vision of the surveyor. If the person declined to complete the survey, the next closest group was chosen , and so on. After a subject had agreed to complete the questionaire, the surveyor returned to the transect and again paced the re quired number of steps before approaching the next potential respondent. This process was continued until ten surveys had been given out or, as happened occasionally at low densities, everyone had been asked. The number of paces between stops was determined'by the length of the study area such that in most cases the complete length of the beach was surveyed. - 42 -This process was modified for English Bay and Skaha since they were so narrow that choosing 'lanes' was impractical. For these beaches the surveyor picked a midline path for the sampling transect and as with Kitsilano used the pacing technique to distribute the surveys evenly over, the length of the beach. The most important aspect of the dissemination of surveys was the need to assure that the respondents would be visually identifiable in the aerial photograph. This was accomplished with the aid of two foot square black and white marker panels, each with a distinctive pat tern. After a potential respondent had been told of the purpose of the research, and had accepted the invitation to fill out the survey, the per son distributing the surveys staked down the distinctive panel beside the individual or group .and placed the corresponding marker panel symbol on the cover of the survey. Since the panel was visible in the aerial photo graph, each subject's responses on the survey could be correlated with his or her spatial and group characteristics. Virtually no one questioned the significance of the panel, apparently believing it was necessary to guide the surveyor back to the spot when collecting the surveys. The cover of the survey booklet contained a title 'Recreational Attitude Survey' as well as the purpose of the research and a short list of instructions. The planning aspects of the survey was implied by the label 'School of Community and Regional Planning, University of British Columbia.' The front cover also contained blanks for information which the surveyor obtained directly from the respondent. The light intensity on the beaches was characteristically high. To reduce eyestrain, the booklet was printed on blue paper. Psychological and demographic information. Briefly again, the survey of personal characteristics consis ted of four parts: the Environmental Response Inventory (ERI), the Lei sure Activities Blank (LAB),and a socio/demographic section and a mood adjective checklist (MACL). The ERI, LAB and socio/demographic profile were included to tap the more stable psychological dimensions, whereas the adjective checklist measured more transient mood states. -43 -The Environmental Response Inventory. The ERI consists of 184 statements or items which pertain to various aspects of designed and natural environments. To respond to these statements a subject circles the extent to which they agree or disagree with the item according to the following pattern: 1) SA = strongly agree, 2) A = agrees, 3) N = neutral, 4) D = disagree, and 5) SD = strongly disagree. On the basis of each subject's response pattern, a numerical score is obtained for the nine separate factors or dispositions. Documented on pages 21 and ,22.. Leisure Activities Blank. McKechnie's (1973) stated goal in producing the IAB was to present "a summary picture of a respondent's self-reported past recrea tion and leisure behaviours." This he accomplished by developing a list of 121 leisure activities which he felt comprehensively surveyed the cur rently popular recreation pastimes in the United States. The response format used in the present study is similar to the one developed by McKechnie except where he used a four point response scale, I used five. The response format which differed from McKechnie's was, " you occasion ally participate in the activity at this time." The addition of this statement allowed for five response types and thus conformed with the ERI in this regard. The response format was developed as follows: Below is a list of leisure and recreational activities. For each activity indicate the extent of your participation using the following system: N - You have never engaged in the activity. T - You tried it once or a few times. U - You used to do it regularly, but not no longer do it regularly. 0 - You occasionally participate in the activity at this time. . R - You currently participate regularly in the activity. Check the appropriate blank to indicate your partic ipation in each o,f the following activities: In addition to the 121 leisure activities space was provided for additional activities which the respondent participated in but which was not included in the list. In McKechnie's sample of 288 subjects and for the present study only a few additional activities were mentioned. Socio-economic Variables. In order to ascertain the degree of association between demogra phic and socio-economic variables with beach behaviour, a series of ten questions concerning the following categories were asked: 1. age 2. sex 3. marital status 4. number of children 5. number of siblings 6. education 7. number of years residing in six different sizes of urban centres (6 variables) 8. number of automobiles 9. occupation 10. household income An additional question tapped the amount of leisure time spent in the urban environment, and in rural environments. This variable was included to test whether persons who spend a majority of their time in non-urban activities have a higher need for space than those participa ting in predominantly urban activities. Mood Adjective Checklist. The mood adjective checklist was developed by Lorr et al (1967). This survey consists of 60 adjectives which act as descriptors of various mood conditions. When these adjectives were factor analyzed by Lorr and his associates, eight factors emerged. Individual factors are listed on page 27. The survey used for the present research used only 59 adjectives rather than 60, since the adjective 'elated' was mis-typed as 'hated' and as a consequence was dropped from the analysis. A re-factoring of the adjectives as used for tlie present study produced essentially the same factors as Lorr et al (1967) and as n result their factors were used to calculate respondent scores for the mood scale. The checklist was made as an optional part of the survey, and each respondent chose whether he wished to complete this section. This final section was titled an 'Optional Word List Survey' and the response format was as follows: If you feel you have any extra time there is an optional survey below which consists of 60 words which describe how you may feel at this time. The survey takes about five minutes, and is designed to measure your personal feel ings at this time. If you wish to complete the survey, for each word merely circle the number which best indicates how you feel at this moment according to the following scheme: 1. Not at all 2. A little 3. Moderately 4. Strongly 5. Extremely Work quickly — first impressions are usually the most accurate. Each adjective was followed by numbers from one to five and a respondent merely circled the number most nearly approximating his immed iate feelings. -. 45a -CHAPTER IV SPACING BEHAVIOUR ON PUBLIC BEACHES: ANALYSIS OF RESULTS. - 46 -User distribution and external environmental features. A primary assumption of this study was that a beach represents an isotropic environment free from artifactual constraints. However, initial observations suggested that users did respond to certain aspects of the beach environment, namely, bath houses, parking lots, and refresh ment stands, especially at lower densities. In an attempt to quantify these observations I carried out a space-time study of the distributional pattern of users for the Kitsilano site for one complete day. (See methods) The results of the census are represented by six separate plots (Figs. 6 -11), each the result of a single hourly census. A group's pos ition is indicated by a dot and the diameter of the dot is proportional to the number of persons in the group. Only those groups which were act ually present during any one census are represented by each figure. Referring to the figures in succession, the first people to arrive tend to locate near the west end, close to the refreshment stand and densities continue to be higher in this area throughout the day. It also seems that larger sized groups tend to concentrate in this area as well. Although these results tend to suggest that for Kitsilano there are behavioural effects due to certain physical structures surround ing the beach, later results indicate that the effect is slight since at no time did the pattern statistic (R) suggest an aggregated pattern was present for any of the three beaches studied. The most probable explan ation is that environmental features such as refreshment stands exert a small attractive force but the repellent force of situating near other groups quickly becomes the dominant determinant in the site selection process of newly arriving groups. Density and spatial pattern. The testing of the first hypothesis and its corollary are the subject of this and the following section. To re-orient the reader they are restated below: - 47 -Figures G - 11. Spatial chni-acteri s I ics of bench users over time Figure 6. Time - 1030 hours, Density - 11 grOups/hectare Figure 7. Time - 1130 hours, Density - 42 groups/hectare ••• Figure 8. Time - 1230 hours, Density - 86 groups/hectare - 48 -Figure 9. Time - 1330 hours, Density - 139. groups/hootn re • • • • i i. Figure 10. Time - 1430 hours,.Density - 188 .groups/hectare » • • • • • • • • • • • • • • * • • • • • • • * » * • • • • • • « • • • * • • • • • • • • • * • * • • Figure 11. Time -.1530 hours, Density - 168 groups/hectare • • 9 9 * • • • • • • * » • • o • • • • • • * 9 • • • • • • • i • • • t • • • » » • • • • • • • • t... 4 * -.49 -Hypothesis: At low to moderate densities, distances to nearest neighbor will be greater than 3.7 meters, however, as space becomes limiting at higher densities, distances will approach a minimum value between 2.1 • and 3.7 meters. Corollary: The distribution of groups over the beach will approach a random pattern at low densities and as density increases the distribution of groups will exhibit an increasingly regular pattern. The results relating to the corollary will be discussed first. To test the relationship between density and pattern, the norm alized R statistic (ZR) was calculated from nearest neighbor distance data covering 1791 groups for 27 separate photographic runs. Inspection of Fig. 12 reveals a basically linear relationship between density and Z^ (points fitted by least squares regression). These data further ind icate a continuous trend from a random pattern toward a regular distri bution, that departs significantly from random (p< .05) at 110 groups/ hectare. Although basically confirming the corollary relating density to pattern, I must point out that except for one case (Skaha), only Kit silano regularly reached densities sufficiently high to exhibit a Z value greater than 1.96. This factor does not seriously detract from the val idity of the results since the overall trend is basically linear and it seems Justified to expect that if higher densities could be sampled from these other sites they would show the same trend as Kitsilano. (See App endix D for photographs of typical spatial patterns of users as a result of different density conditions) These results suggest several implications relating to site sel ection and the relative influence of other users on this behaviour. The results lend support to the argument that within a density range of 20 -110 groups/hectare, beach users are able to select a site based on inter nal needs and preferences without reference to other groups. In addition, since an aggregated pattern was not observed, major characteristics of the physical environment do not exert a significant effect. Thus if one views available space on the beach as a resource, then the first group or individual to arrive at the beach has unlimited freedom to exploit this, resource. As each new group arrives and chooses a site, the "degrees - 50 -Figure 12. Relationship between density and pattern. -r 27.0 54.0 —1 1 6J.0 100.D 135.0 1E2.D DENSITY (GROUPS/HECTRRE) 1(39.0 —I 216.0 —1 1 243.0 270 Pattern is measured by the normalized 'IT statistic (7,R). The dashed line represents the .05 probability level for a regular pattern, r = 0. Z =-1.88 + 0.03 (d), where d equals the density in groups/hectare. KR= Kitsilano, S = Skaha, and E = English Bay. -51 -of freedom" for a subsequently arriving group have diminished, i.e. the physical space taken up by a group, plus as I will show in a later sec tion, a certain amount of space surrounding the group, is not available for use by any other group. However, since the pattern is basically random for densities less than 110 groups/hectare, it seems evident that groups arriving within this density range have enough vdegrees of freedom" or options available to them to select a site according to personal pref erences, relatively uninfluenced by the social and psychological factors relating to space needs and preferences. Further, as densities exceed 110 groups/hectare, there is a decreasing probability that any newly arriving group can locate solely on the basis.of personal preferences without regard for their own spatial needs, i.e. other groups are the major influence with respect to site selection. The degree to which these spatial needs become primary is reflected in the extremely high ZR values associated with higher densities. Since these values indicate an extremely low probability that the pattern reg ularity is due to chance,users arriving at•the beach must, choose to locate with reference to in situ groups such that distances between the chosen spot and near neighbors are maximized. This inter-neighbor distance and how it is influenced by density forms the basis of the following section. Density and distance to nearest neighbor. An analysis of the average distance to nearest neighbor (nnd) was carried out to test the hypothesis that spacing behaviour of beach users is related to density and that average nearest neighbor distance at a given density is related to cultural norms and proper social func tioning. To make this test it was necessary to examine the relationship between the average nearest neighbor distance for all runs on all beaches and the density.nt which each run was completed. Figure 13 is a graphical representation of these data where the two curves shown are: 1) the average distances between polygon centrolds of nearest noighbor groups, and 2) the average distances between the edges of nearest neighbor poly gons. These measures are termed average centroid to centroid nearest neighbor distance (cc/nnd) and nearest approach nearest neighbor distance (na/nnd) respectively. (See page 39 for details of these distance measures). - 52 -Figure 13. Average distance between first nearest neighbors (NNU) plotted againtit density for 27 runs. X "1 _ , , , , T -| Q>0 27.0 54.0 B1.0 1C3.0 135.0 152.0 • . DENSITY (GR0UP5/HECTRRE) ID9.C 2)6.0 I 2-13.0 270.0 The symbols E, K and S refer to the three study sites, English Bay, Kit silano, and Skaha for the centroid to centroid NNO. The points denoted by X refer to the measure nearest approach NND, whereas the letters E, K and S refer to centroid to centroid NND. Distances are in meters and the sample is based on 1791 groups. The first aspect of Figure 13 to note is the existence of a constant relationship between the two measures of nnd such that the av erage difference between corresponding values of cc/nnd and na/nnd is 2.2 meters (S.D. = 0.23). The relatively small standard deviation about the mean differ ence between.cc/nnd and na/nnd is explained by two other results relating to group size and space characteristics. First, over 83% of all per sons observed on the three sites were found in either one or two person groups and second, the average group area for one and two person groups as delineated by a group's possessions does not change appreciably over the observed density range. Most important, Figure 13 shows that nnd becomes asymptotic to the x-axis (density) at about 110-120 grps/ha., i.e. within an observed density range of 110 - 264 grps/ha the average cc/nnd remains constant at 5.0 meters (S.D. = 0.36) and the average na/nnd is 2.7 meters (S.D. = 0.45). Thus within the 110-264 grps/ha density range, most groups main tain an edge to edge distance of about 2.7 meters which remains invari ant despite density. Since the distribution, even though statistically regular, is patchy, newly arriving groups can "fit" into the remaining spaces or holes. Figure 12 shows that the pattern of spacing becomes statistic ally regular (ZR < 1.96) at about the same density as nnd becomes asymp totic i.e.,110 groups/hectare. This fact is important since it is possible to conceive of beach users exhibiting a regular spatial pattern while con tinuing to decrease the distance between nearest neighbors until the point is reached where groups' spatial boundaries touch their neighbors and average na/nnd is zero. However, this did not occur. At approxim ately 110 groups/hectare beach users have adapted to the influence of crowded conditions on the beach, and have done so by maximizing the dis tance between their near neighbors thus producing a regular distribution. The distance between neighbors remains constant above this density sug gesting that there is a limit to the compressibility of any group's spatial preferences. •r 54 -The limit to the compressibility of space preferences as dem onstrated above was 2.7 meters. This value (na/nnd) is the average dis tance between the edges of two neighboring groups' marked space at den sities greater than 110 grps/ha. and lies near the mid-point between the extremes of Hall's (1966) social distance zone (far phase). These results are thus consistent with the first hypothesis which stated that at higher densities distance to nearest neighbor would approach a minimum value and that this distance would fall within Hall's social distance zone (far phase). These results suggest that individuals and groups adapt to in creasing density by maintaining a minimum 'bubble' of space around them selves. This minimum distance between neighbors is likely related to the control of unwanted social interaction and may thus be a privacy reg ulation mechanism. This proposition is consistent with Hall's (1966) claim that, for North Americans, the social distance zone (far phase) is often used to screen or insulate one person or group from another. In summary, the results relating to the effects of density on spatial pattern and distance to nearest neighbor indicates that for the three beaches studied, users respond to increasing numbers of others in characteristic ways. First, at low densities the observed spatial pattern is random and thus it is proposed that effects due to other groups (soc ial effects) are minimal. Further, since the choice of a site seems not to be affected by other groups, users maintain more degrees of freedom in the process of selecting a site. Second, as density increases space be comes a limiting factor with respect to site choice. This is reflected in the existence of a statistically regular pattern as well as a constant average distance to nearest neighbor at densities, greater than 110 grps/ha. Finally, a mechanism is proposed to explain these findings which relates inter-group distances to the control of social interaction. - 54a -CHAPTER V GROUP CHARACTERISTICS Marked group area and density. Although not tested as specific hypotheses, several aspects of the group size and group area of beach users are presented here since they are relevant to calculations in the following chapter on maximum beach population size. A group's area was defined previously as the area circumscribed by the sides of a polygon, the vertices of which were beach paraphernalia, possessions owned or shared by a group, or in many cases the bodies of users themselves. Possessions used in this way have been termed spatial or "territorial" markers (Sommer and Becker, 1966; Becker, 1973; and for the present study, the area included within these objects has been oper ationally defined as a group's marked space. Are the marked areas of groups influenced by density? Figure 14 indicates that for group sizes of from one to four persons, density had lit tle or no effect on the marked area for a given size group. These results sug gest that groups do not decrease the size of the marked space as a way of adapting to increasing density. These and other previous results sug gest that tactical space-saving maneuvers may not be necessary on the part of in situ groups since new arrivals seem reluctant to situate within the 2.7 meter zone referred to earlier. Another result expressed by Figure 14 is that the mean group area grows in linear proportion to the number of people for groups of one to three persons. Table 5 shows that for these group sizes (1 -3), the space utilized increases by approximately two square meters for each group size. The next four group size classes (4 - 7) increase by amounts ranging from 2.6 to 3.4 square meters. These latter values must be viewed with caution however, since sample sizes are small and standard deviations substantial. These data may be explained in two ways. First, each per son may bring to the beach a certain requirement for space which remains uninfluenced by the proximity of other individuals and groups. Such spa tial needs if real would then be functionally related to the personal space construct of Hall (1966) and Sommer (1969). Second, since beach equipment such as towels and blankets are relatively uniform in size and shape, these articles may determine the spacing of individuals within a group. These may not necessarily be competing explanations since users - 56 -Figure 14. Density and group area. tn u X X X X X X X X 0 X X J,? I * X A* L*5 X xo^ J ^ *x o * K - x X >< -£ -X X ^ "X X X X X X X v * * X X X + -—± ' + X •+r + -t-X — —75 ~w ST* o. n i^Tn n ir,?.0 inoTo ?]"ii7o 243.0 270.0 Vo 27 0 510 fli.o 100.0 i.is.o ir,?.o mo.o ZIB.O °-° 2 '° • DENSITY tGRQUPS/HEClRNL) Means for one to four person groups represented by solid lines. Lone individuals (+); Two person groups (X); Three person groups (O) ; Four person groups (*). -'57 -may choose towels, blankets, etc. which reflect personal space needs and preferences. In a test of a possible relationship between marked space and the distance between neighboring groups (nearest approach nnd), analysis by simple correlation produced an r value of -0.07. This result suggest there is no clear relationship between the size of the marked space anu the distance between the edges of neighboring groups' marked areas. Table 5: Group Area Related to the Number of Individuals in a Group Group Size Class 1 2 3 4 5 6 7 7 Total 2 Mean (meters ) 1.9 3.9 5.7 9.0 12.1 14.7 18.1 11.1 3.8 S. D. 1.0 2.0 3.3 7.7 11.4 10.7 26.6 4.1 3.4 N 828 664 155 102 22 10 8 2 1791 Density and group size. A final characteristic of beach user groups worthy of mention, concerns the relationship between density and the number of individuals in a group (group size). . Analysis by simple linear regression indicates 2 a moderate increase in group size as density rises (r = 0.29, p _^ 0.01). Figure 15 shows this result graphically. This result leads one to specu late that individuals or groups come to the beach at low densities be cause of a high need for privacy or for the solitude which these times afford. Evidence from the survey results however, does not support this hypothesis. For example, the ERI scale 'need for privacy' only correla ted -0.06 with nearest neighbor distance measures. Other possible causes of this relationship may be related to the temporal and structural dynamics of group formation, however this relationship was not tested and there fore remains hypothetical. - 58a -CHAPTER VI CARRYING CAPACITIES AND THE BEACH EXPERIENCE - 59 -Introduction. Previous results indicated most people on a public beach re spond to increasing density by choosing sites with at least 2.7 meters between their own and neighboring groups' marked areas. From other re search it seems likely that this distance is functionally related to the regulation of social interaction. Thus it should be possible to calculate the maximum population size which each beach could sustain and still allow users the benefit of this minimum space requirement. Resource managers often refer to the carrying capacity of a site or geographical region. As the term applies to recreation, it generally refers to the number of persons (level of use) which a resource can sup port without a loss in user satisfaction and without a decrement in the quality of the physical environment. (For a review of the carrying ca pacity concept see Verburg & Rees, 1975) For the purpose of calculating maximum tolerable use rates for the present study, I have disregarded effects due to the user on the physical environment, since other than litter, the environment seems resilient to high intensity use rates. Of course for beaches with vegetated dunes, effects due to overuse could be severe. Before proceeding with the results of these calculations several terms require clarification: * Group size - the average number of individuals per group for all observations at each of the study sites. * Marked group area - the average observed space subtended by (lying within) the personal possessions of user groups. * Minimum group space - the calculated group space requirement (in addition to marked space) based on the average nearest neighbor distance values for densities ?" 110 groups/hectare. *- Maximum carrying capacity estimate - the number of average size groups which each beach could sustain, based upon observed spatial behaviour. * Load factor - a value representing the extent to which each site reached its maximum carrying capacity. - 60 -Carrying capacity and the response to density. The calculation of the carrying capacities for each of the three sites requires knowledge of the average centroid to centroid near est neighbor distance for densities exhibiting a constant nearest neigh bor distance, i.e. densities 2l 110 groups/hectare. Since capacity estimates required a constant representation of the average group areas, I chose a regular hexagon as a suitable geometric shape for this purpose. This shape approximates a circle (a study by Edney and Jordan-Edney, 1974 suggested the areas claimed by beach users approximated a circle), how ever in contrast to circular areas, hexagons leave no space unaccounted for. For purposes of calculating the carrying capacity estimates, the distance between any two nearest neighbor group's marked space boundaries was considered to be*shared evenly, i.e. each group maintained jurisdiction over one-half the intergroup space. In reality, this space is most likely perceived by users as common property with each group utilizing the space jointly. In any event, the equal space assumption above is merely utilized for calculation purposes and is not meant to convey the existence of such behaviour. Figure 16 is a conceptual representation of the spatial config uration of any two average groups at or above 110 groups/hectare. The smaller of the two hexagons simulates the 'marked area', whereas the larger is a representation of the minimum spatial requirement of the group. To calculate the minimum spatial requirement of a group (large hexagon), one need only calculate the area of the triangle ADE and multiply by six. Referring to figure 16, the altitude- of the triangle ADE is equivalent to one-half the centroid to centroid distance AA'. Knowledge of the altitude AC, allows one to calculate the area of the triangle ADE by the formula Area = h2 / where h equals the altitude AC. The maximum carry ing capacity (# of groups) is therefore the area of the beach divided by the minimum space reauirement (large hexagon). Table 6 contains the results of these calculations as well as the maximum densities at each site which were observed during the study. By comparing the ma-ximum theoretical population density for each beach with the highest observed densitv (load factor), it is readily apparent that at no time did any of the sites reach these limits. Kitsilano maintained the highest recorded density (264 grps/ha). During this - 61 -Figure 16. Hypothetical representation of two neighboring groups' "marked" and "minimum" space boundaries. - 62 -Table 6. Maximum beach population estimates. E. BAY KITSILANO SKAHA 2 Beach Area (meters ) 11967 4522 6500 Group Size 1.6 1. 8 2.1 Marked Group Area (meters^) 3.0 3.9 4.5 Minimum Group Space* 21.7 21.7 21.7 Maximum Carrying Capacity of Beach (number of groups) 551 208 299 Maximum Carrying Capacity (groups/hectare) 461 461 461 Maximum Observed Density 151 264 175 Load Factor (observed/maximum) 33% 57% 38% * Based.on average nearest neighbor distance values across beaches for densities > 110 groups/hectare. - 63 -period the beach was 57% of the estimated maximum carrying capacity. English Bay and Skaha beaches maintained densities which were only 33% and 38% respectively, of their estimates. These results suggest that even at the highest recorded densities newcomers were still able to locate in a spot which would allow them the minimum intergroup distance (nearest approach nearest neighbor distance) of 2.7 meters. This is an important point since it suggests that a density was never reached where all open space was utilized, thus forcing newcomers to locate within the spatial boundaries of others. Note that if beach densities exceeded carrying capacity estimates derived above, the distances between neighboring groups would be reduced sharply. For example, if an 'average group' were to locate midway be tween neighboring groups at the maximum carrying capacity density, there would remain but 0.3 meters between the edges of any two of the three groups in question.. Such distances fall within what Hall (1966) classifies as the 'intimate distance - far phase'. He characterizes this zone by stating that the use of such distances in public is not considered proper by most adult North Americans. He goes on to state however, that in many situations such as crowded elevators, trains, buses, etc., other tactics are used which serve to decrease visual and body contact. It seems probable that as such densities are approached, most users of a public beach facility would search for another site or return home, thus foregoing the experience rather than subjecting themselves to such close inter personal distances. Of course beaches do exist where such high density conditions occassionally occur. It is likely that for these sites, users are highly motivated to participate either because no other sites are available or for reasons relating to costs involved in reaching the beach. Beaches where such levels of crowding occur would provide ideal environments for studying the range of tactics used by persons and groups in order to cope with personal space violations which occur at these extreme densities. The above results are important since they demonstrate the effectiveness of using actual participant behaviour to arrive at carrying capacity estimates. What is apparent is that for the areas studied, users rarely select sites which violate prevailing social norms with respect to appropriate spacing behaviour. Since we assume that this behaviour serves - 64 -some basic social function, the manager or planner of such facilities may make policy decisions based on the needs and preferences of users themselves. In this way the decision maker can be reasonably well assured of adeauately serving the interests of the greatest number of people without detracting seriously from the quality of the experience. Using this criterion, we may speak of optimal solutions to design and manage ment problems. Of course, at such densities individual satisfaction may not be maximimal, however a satisfactory experience can be expected to be provided for the largest number of people. Such behaviourally based guidelines would surely serve as an improvement over more arbi trary techniques commonly used in the past. - 64a -CHAPTER VII PREDICTORS OF SPATIAL BEHAVIOUR: ANALYSIS OF SURVEY RESPONSES - 65 -Predictors of spatial behaviour - all groups. Results th a preceeding section indicated the existence of a minimum intergroup distance which was associated with high density condi tions. This distance was in turn postulated as a mechanism by which peo ple control social interaction. The purpose of the following section is to present evidence which demonstrates that certain psychological dis positions, mood states and socio-economic characteristics are related to spading and group behaviour. A total of 329 surveys were distributed (English Bay - 105, Kitsilano - 127, Skaha - 97) of which 23 were unuseable and 46 deleted due to camera malfunction or because the marker flag was not visible in the photograph. Of the 266 surveys remaining,84 were from English Bay, and 99 from Kitsilano and 83 from Skaha. The data were analyzed by a stepwise multiple regression pro gram with an F probability to accept and reject potential independent variables of .05000 and .050001 respectively. Forty-eight independent variables (ERI - 8, LAB - 7, Mood - 9, Socio-economic - 24) were used as potential predictors of three dependent variables (two measures of near est neighbor distance, and the area marked by a group). Table 7 shows the partial correlations at the first step in the regression analysis as well as the final values for the dependent var iables. Inspection shows few significant correlations and characteristi cally low predictability of target variables. Although discouraging, these results suggested an alternate approach which focussed attention on res ponse patterns of solitary individuals not part of larger groups. Lone Individuals - A second look at the data. The analysis of survey data for lone individuals was based on the argument that solitary persons are probably more in control of where they locate, since in groups the decision may be made by someone other than the respondent, or may be a collective decision which does not ex actly reflect the desires of the person completing the survey. Similarly, if only one individual in the group makes the decision as to site loca tion, then as group size increases there exists a decreasing probability Table 7. Initial partial correlations, F probabi variables. (all group sizes considered 1 VARIABLES GROUP AREA ERI Partial Corr. F Prob. PASTORALISM URBANISM ENVIRCNMENTAL ADAPTATION STIMULUS SEEKING ENVIRONMENTAL TRUST ANTIQUARIAN!. SM NEED FOR PRIVACY MECHANICAL ORIENTATION COMMUNALITY LAB MECHANICS CRAFTS INTELLECTUAL SLOW LIVING NEIGHBORHOOD SPORTS GLAMOUR SPORTS FAST LIVING MOOD CHEERFUL ENERGETIC ANGRY TENSE-ANXIOUS THOUGHTFUL DEPRESSED FATIGUED RELAXED MOOD 0. 322 0. 060 0.065 0. 001 0. 077 0.097 0.107 0.001 0.036 0.050 0.091 0.088 0.016 0.073 0.031 0.089 0.093 0.038 0.050 0.088 0.096 0.148 0.092 0.029 0.012 0.001 0.396 0. 353 0.937 0.270 0.161 0.122 0.934 0.616 0.479 0.189 0.203 0.805 0.294 0.661 0.198 0.179 0. 594 0.485 0.204 0.164 0.032 0.185 0.683 0.845 es and final values for three dependent pproximate degrees of freedom = 207) CENTROID TO CENTROID NEAREST APPROACH NND NND Partial Corr. F Prob. Partial Corr. F Prob. 0.022 0. 0,59 0.017 0.128 0.001 0.047 0.036 0.059 0.013 0.748 0.406 0.796 0.062 0.936 0. 510 0.609 0.408 0.834 0. 008 0.078 0.018 0.115 0.002 0.034 0.047 0.062 0.011 0. 875 0.265 0.786 0.096 0.923 0.629 0. 512 0.381 0.852 0.133 0.027 0.041 0.139 0.217 0.096 0.182 0.053 0.703 0.569 0.043 0.002 0.166 0.009 0.153 0.008 0.000 0.140 0.197 0.058 0.171 0.026 0.874 0.948 0.042 0.005 0.416 0.013 0.063 0.062 0.064 0.066 0.103 0.026 0.042 0.011 0.120 0.375 0.376 0.364 0.351 0.135 0.714 0. 557 0.845 0.082 0.036 0.048 0.124 0.014 0.102 0.080 0.020 0.021 0.142 0.616 0.497 0.071 0. 827 0.140 0.252 0.765 0.760 0.039 Table 7. (continued) VARIABLES GROUP AREA SOCIO/DEMOGRAPHIC Partial Corr. F Prob. AGE 0.029 SEX 0.032 MARITAL STATUS 0.091 NO. OF CHILDREN 0.105 NO. OF SIBLINGS 0.092 EDUCATION 0.069 NO. OF YEARS LIVED IN CITIES V.'ITH POPULATIONS OF : . OVER ONE MILLION 0.098 100,000 - ONE MILLION 0.068 50,000 - 100,000 0.111 10,000 - 50,000 0.110 5,000 - 10,000 0.133 BELOW 5,000 0.067 NO. OF AUTOMOBILES 0.T35 JOB CATEGORY 0.027 INCOME 0.07% TIME RECREATING OUTSIDE URBAN ENVIRONMENT 0.015 DAY OF THE WEEK 0.117 TIME OF DAY 0.049 NO. OF CHILDREN IN GROUP 0.119 % OF GROUP WHO WERE MALES 0.071 % OF GROUP WHO WERE FEMALES 0.016 % OF GROUP WHO WERE CHILDREN 0.094 DISTANCE TO HOME CITY 0.065 POPULATION OF HOME CITY 0.026 R2 0.18 0.684 0.651 0.190 0.130 0.184 0.324 0.155 0.332 0.106 0.111 0.053 0.342 0.049 0.696 0.286 0.814 0.089 0.495 0. 084 0.312 0. 804 0.174 0.353 0.709 CENTROID TO CENTROID NND Partial Corr. F Prob. NEAREST APPROACH NND Partial Corr. F Prob. 0.080 0.042 0.099 0.135 0.050 0.019 0.005 0.085 0.071 0.072 0.088 0.174 0.140 0.015 0.089 0.127 0.044 0.212 0.047 0.025 0.019 0.081 0.010 0.096 0 . 22 0. 253 0. 558 0.153 0.050 0.479 0.779 0. 898 0.221 0.308 0.301 0.205 0.012 0.042 0. 810 0.198 0.066 0. 538 0.003 0. 509 0.716 0.781 0.245 0. 862 0.167 0.093 0.065 0.149 0.171 0.046 0. 040 0.019 0.114 0.035 0.084 0.096 0.177 0.181 0.007 0.106 0.134 0.024 0. 215 0. 007 0. 029 0. 014 0.040 0. 082 0. 219 0.19 0.180 0.359 0.030 0.013 0. 517 0. 579 0.779 0.099 0.618 0.225 0.164 0.010 0.009 0.882 0.123 0.051 0. 726 0.002 0.888 0.679 0.824 0. 573 0.240 0.002 - 68 -that the survey had been given to the person making the decision. In addition, I argued that with respect to the-area which any size group marks as theirs, lone individuals will have more control over the size of the area than a single individual in a larger group. Based on these arguments, the dependent variables cc/nnd, na/nnd, and group area, were analyzed for lone individuals only. Because of a sampling discrepancey associated with the nearest neighbor distance values (cc/nnd and na/nnd) for English Bay, these data were omitted from the analysis as well. (See Appendix C for a discussion of this problem) To reiterate, on the basis of the above considerations, the dependent variables cc/nnd and na/nnd were analyzed for lone individuals at Kitsilano and Skaha beaches only, whereas the variable group area was analyzed for lone individuals at all beaches. Since the former two var iables (cc/nnd and na/nnd) were based on response patterns and spatial data for lone individuals at two of the three sites the number of observa tions declined sharply (n = 64 with approximately 54 degrees of freedom). The results of this analysis must therefore be viewed with a certain de gree of caution since the number of observations approach the number of independent variables used in the analysis. 2 Table 8 presents the R values associated with the various analyses conducted for all beaches and group sizes, plus those completed for lone individuals only. These results indicate a comparatively large increase in the ability of the selected variables to account for the var iance in the dependent variables when English Bay was dropped from the analysis and when groups containing two or more individuals were ommited. 2 Table 8: R Values For Spatial And Group Dependent Variables. Group Area cc/nnd na/nnd EKS (all/Si) .18/.57 .22/.10 .19/.08 KS (all/Si) .26/.66 .20/.46 .21/.47 E = English Bay; K = Kitsilano; S = Skaha; all = all group sizes; Si = single individuals. - 69 -Predictors of respondents' distance to nearest neighbor. The two distance measures, nearest approach nnd and centroid to centroid nnd were predicted by identical independent variables and main tained very similar R2 values for lone individuals. The only differences were the extent to which each independent variable contributed 2 to the overall R. . The Independent variables: accounting for a significant amount of the variance in the two distance measures are listed in Table 9. These data indicate that people from small towns, those with high scores on 'pastoralism* , 'environmental adaptation' and the mood var iable 'relaxed' are all found at a greater than average distance from their nearest neighbor. Thus users who have spent much of their lives in rural environments, or who haye a 'pastoral' disposition tend to choose a site with more intergroup space. The 'pastoralism' variable is defined by McKechnie with such phrases as, "concern about population growth and preservation of natural resources, including open space." These results would suggest that insofar as .the 'pastoralism' scale measures a respon dent's attitude and needs for space, the spatial behaviour of beach users tends to validate the scale. The second best predictor of intergroup distance is the 'man over nature' variable (environmental adaptation). A person scoring highly on 'environmental adaptation' may be characterized as one who seeks to modify "the environment to satisfy needs and desires, and to provide com fort and leisure." This person is also "opposed to governmental control over private land use" and shows a "preference for highly designed or adapted environments." Adjective descriptors include, "autocratic, con descending, conservative, efficient, judgemental" etc. It seems likely that high scorers on 'environmental adaptation' have a basic need to control their environment and in an isotropic setting such as a beach dominating space is the most available way of maintaining this control. The inclusion of the mood variable 'relaxed' is important since (ills indicates that respondents at larger distances tend to bo less tense than those individuals situated nearer other groups. - 70 -Table 9: Significant independent variables contributing to the dependent variables, centroid to centroid and nearest approach nearest neighbor distances (cc/nnd and na/nnd). VARIABLE F PROBABILITY CC/NND NA/NND NORMALIZED REGRESSION COEFFICIENT CC/NND NA/NND 1. Years lived in cities with populations between 5,000 -10,000 people. .0003 . 0003 .4506 . 4608 2. 'Urbanism' .0010 .0021 -.4494 -.4128 3. 'Environmental Adaptation' .0124 .0193 .3072 .2836 4. 'Pastoralism' .0188 .0117 .2919 .3136 5. Number of children .0104 .0050 -.2910 -.3217 6. 'Relaxed' .0354 .0368 . 2505 .2466 (Results based upon lone individuals at Kitsilano and Skaha beaches only) - 71 -The two variables negatively correlated with distance to near est neighbors were the ERI scale 'urbanism' and'the number of children a respondent claimed.' . The scale 'urbanism' describes subjects who are or iented to high density urban environments and who maintain an interest in the unusual arid varied aspects of city life. High scorers on 'urbanism' are thus those people who either enjoy the crowds a city affords or are those who are capable of adapting to such high density environments,, Since our urban public beaches often reflect such environments, it is not surprising that this scale was a strong predictor of user spacing behaviours„ The correspondence of the second variable ' number of children' with smaller intergroup distances may be explained by the fact that a person who chooses to have a large family is probably more gregarious and enjoys being around larger groups of people. This latter statement, of course, is conjectural and awaits further testing. Surprisingly, the Environmental Response Inventory scale, 'Need for Privacy' did not correlate significantly with the distance to nearest neighbor measures. In fact, the correlation was close to zero (r = -.05) for both nearest neighbor distance measures. One explanation for these results may be that the beaches chosen for the study are primarily urban public beaches which may only attract individuals with diminished privacy needs. These settings may thus convey the image of a crowded, high stimulus environment, even though the beaches exhibit lower densities for a portion of each day. Predictors of 'group area*. -The R value in Table 8 relating to the dependent variable 'group area' for single individuals at all sites was 0.57. The variables contributing to the regression equation are summarized in Table 10 . (Eng lish Bay included). These data suggest that lone individuals who attain larger amounts of marked space may be characterized as having more auto mobiles than average, spend more recreational time away from the urban environment, have more children, have spent more of their lives in cities with populations between 50,000 to 100,000 and are more often at the beach on weekends when densities are highest. Similarly, individuals with - 72 -Table 10. Significant independent variables contributing to the dep endent variable 'group area'. VARIABLE' 1. Number of automobiles 2. % recreation time spent outside the city 3. 'Environmental Trust' 4. Number of children 5. Years lived in cities with fewer than 5,000 people 6. Years lived in cities with population sizes between 50,000 - 100,000 people 7. Years lived in cities with population sizes between 10,000 - 50,000 people 8. Day of the week F PROBABILITY < .0001 .0005 NORMALIZED REGRESSION COEFFICIENT .4602 .3306 .0006 .0002 .0029 -.2966 .2797 -.2696 ,0179 ,2207 .0209 -.1996 ,0447 1720 Values based upon lone individuals at. English Bay, Kitsilano and Skaha Beaches. (n = 85) - 73 -smaller than average marked spaces tend to have a higher trust of poten tially threatening environments (Environmental Trust) and have spent more of their lives in small towns and cities with populations between 10,000 and 50,000 people, , Five variables correlated positively with an individuals' marked space. Two of these, the number of children claimed by the respon dent and the number of automobiles in the household may actually be arti facts attributable to the data collection technique utilized. For example, even though persons were visible in the photographs as lone individuals, a few may actually have been part of a larger family group. Children, for example, may have been playing elsewhere pr other members of the,fam ily may have been strolling nearby. Since families could be expected to have a higher probability of owning more than one car and since such a family group would maintain larger marked areas on the beach, these var iables would be selected as significant predictors of the dependent var iable, group area. The data do not permit a test of this hypothesis and thus such an explanation remains conjectural. The inclusion of 'percent time spent recreating outside the city' can be justified since it could be argued that each variable, large group area and high percentage of recreation time away from the city, are related to an increased need for open space. It is interesting to note that people who come to the beach on weekends also have larger marked spaces. This may reflect the need to buffer oneself from others by the use of space, since weekends offer the user the highest density conditions in which to recreate. The latter relationship is conjectural since earlier results on the effect of density on group area indicated little or no effect. It is also of interest that people who spend many years in-small towns seem to maintain smaller marked areas. This result is in contrast to an earlier finding which suggested that persons from smaller cities tended to be at greater distances from their nearest neighbor. One explanation for these results may be that most people who came from small cities and towns were vacationers at Skaha beach and thus in order to save space on route may have brought fewer beach articles with them and thus had fewer materials to spread around. This argument may also - 74 -explain the finding that people living much of their lives in moderately large cities obtained larger amounts of marked space. Beach users who scored highly on the ERI variable "Environmental Trust" tended to maintain smaller amounts of space which was marked. McKech-liie's definition of this variable sheds light on this correlation: Environmental Trust: General environmental opennesg, responsiveness, and trust; competence in finding one's way about the environment. vs Fear of intentionally dangerour environments, security of house; fear of being alone and unprotected. * McKechnie also describes high scorers on this variable as: Capable, competent, diligent, efficient, helpful, ingen ious, resourceful, stable, thorough, well adjusted. and conversely low scorers as: Bitter, cold, coarse, disatisfied, distrustful, intol erant, moody, prejudiced, spendthrift, unkind. The picture which emerges is that respondents categorized as capable, competent, well adjusted etc., have a greater ability to cope with smaller amounts of space whereas low scorers require larger indiv idual marked areas as a buffer against a perceived, inhospitable environ ment . In summary, the three dependent variables,, group area and the two nearest neighbor distance variables were moderately well predicted with percentage variance accounted for ranging from 46 to 57 percent. In most cases the independent variable selected by the regres sion analysis were those that could be easily explained on the basis of their individual content and meaning. The independent variables of most interest were those relating to respondent's scores on various ERI scales and those relating to the number of years a subject had spent in high density urban centres versus those who had lived predomintly in smaller towns and villages. Tn general those respondents who maintain pastoral attitudes and who have spent a large proportion of their lives in small towns are more often found at greater distances from their closest neigh bors than average and those who derive satisfaction from high density environments are most likely observed in close proximity to their near neighbors. -. 74a -CHAPTER VIII DISCUSSION AND SUMMARY OF RESULTS - 75 -Summary of results. The two most important objectives of the studv as set forth in the introduction were first, to determine if behavioural shifts occur in response to increasing density and second, to examine the ex tent to .which individual personality characteristics are related to spatial behaviour differences. With respect to the first objective, evidence was presented which indicate shifts in behaviour did occur (analysis of the spatial pattern of users showed a gradual change from random at low densities to regular at high densities) and coinciding with these events users began to choose sites which were on the average 2.7 meters from their nearest neighbor. This distance fell within Hall's (1966) 'social distance zone' (far phase) which he claims is used by North Americans to effectively insulate themselves from unwanted social interaction. The results of the present study thus add strong empirical support for Hall's claim. The second major objective was achieved by analyzing the spatial behaviour of beach users who chose to complete a questionaire designed for the study. The survey, composed of items dealing with environmentally based dispositions, participation in leisure activities, mood states and socio-demographic characteristics was analyzed by a stepwise multiple regression technique.. The dependent variable in this analysis was the observed subject to nearest neighbor distance as obtained from the aerial photo graphs. The results indicated only a limited ability to predict the depen dent variable when data for all group sizes were used. Based on the ar gument that lone individuals are more in control of the site selection process than groups of two or more, the data were reanalyzed for solitary respondents only. These results showed a substantial increase in the a-mount of variance accounted for by the selected independent variables. The most salient variables selected as significant predictors of distance to nearest neighbor measures were the Environmental Response Inventory variables: 'Urbanism', 'Pastoralism', and 'Environmental Adaptation'. In addition direct experiential measures of the number of years a respon dent had lived in towns and cities of various sizes proved to be of impor tance. Other significant variables included the number of children a respondent claimed and the mood variable 'relaxed'. - 76 -In addition to the problems associated with sampling for groups containing two or more individuals, one other factor may explain why the selected independent variables were not capable of explaining a greater percentage of the total variance. Because of the dynamic quality of the beach environment, a user may have chosen a site under different con ditions from those obtaining when he was selected as a respondent and subsequently photographed. This would decrease the predictive power of the independent variables since a user's spatial environment would have changed as more people arrived at the beach and filled in the area around him. A future research strategy might be devised which delineated the spatial characteristics of a beach user, immediately upon his selecting a site. Of couurse, methods other than aerial photography would have to be used for obtaining data on the spacing behaviour of users in such a study. Finally, based on the observed spatial behaviour of users men tioned above, the carrying capacity of each of the three study sites were calculated. Based on these calculations it was shown that at no time during the study did densities at the three sites surpass the estimated upper limits. These results indicated that on the average, conditions were never so crowded that new arrivals were forced to select a site within the 2.7 meter zone referred to above. The extent to which this situation prevailed as a result of new arrivals choosing not to participate in the beach experience is not known. Methodology applications. Webb et al. (1966) , among others,; have effectively demonstrated the advantages of using multi-method approaches to problem solving in the social sciences. In particular, Webb et al. argue persuasively for the expanded use of nonreactive techniques to assess human behaviour, The present study sought to utilize each of these research strategies. Firstly, the spacing and group behaviour of beach users were studied in a completely unobtrusive way through the use of aerial photography. This technique circumvented obvious sources of bias where the objective of the research is known or suspected by the subject . Such a procedure maximized the - 77 -probability that observed behaviour was typical and uninfluenced by the presence of the investigator or his equipment. Secondly, the study was undertaken in a 'natural' as opposed to a laboratory setting and as a re sult, conclusions reached offer ''.real world' validity with little or no fear that results represent artifacts of experimental conditions. Finally, questionaire response patterns were correlated with each subject's spatial and group behaviour. Thus, both survey data and extant behaviour were jointly utilized to broadly describe how people in an isotropic environment respond to fluctuations in density conditions. This procedure offered some insight into individual personality differences and the extent to which they relate to beach user"behaviour. Implications for planning and design. Although some caution must be applied grhen placing the results of the present study in a planning or design context, I should point out that most space standards have been based primarily pn arbitrary decisions. For example, the California Outdoor Recreation Committee Report (1960) set the optimum space allotment for beaches at 100 square 2 feet (9.3 meters ) per person. This value was based on seasonal attendance records excluding the three most crowded days which fell on holidays. This report further stated that if attendance was higher than 70% of the density on the sixth most crowded day, then the area was considered over-used. The present study represents a distinct improvement over such guidelines in that the behavioural characteristics of users have been used to arrive at social carrying capacity estimates. It is interesting 2 to compare the California standard of 100 ft. /person, to results based on the present research. This comparison may be roughly made by dividing the average group size estimate (1.8 persons/group) into the minimum group 2 2 space standard (21.7 meters )/ This value (12.1 meters ) represents over a 20% increase over the 9.3 meter2 estimate from the above report. We may conclude that, to the extent the two populations (California and western Canada) share similar spatial needs and preferences, the California beaches would be considered 'over-crowded' by the criterion suggested by the present study, at population densities considered 'optimal' by the C.O.R.C. report. - 78 -An important aspect of the results relevant to the planning and design professions concerns the carrying capacity estimates referred to above. To place these results in perspective it is necessary to examine the question of optimality which the carrying capacity concept implies. Depending upon ones frame of reference, the population estimates derived for the three beaches may reflect 'maxima' instead of 'optima'. I stress this point for two related reasons. The first point concerns the extent to which people adapt to high density conditions and how such adaptation relates to user preferences and satisfactions. One might expect that users present during high density .periods might possess skills which allow them to cope successfully with con ditions relating to crowded environments. However, coping successfully does not necessarily imply maximum satisfaction. For example, users may tolerate such conditions at less than optimum satisfaction and participate in the activity even though they might prefer to experience the beach at lower density levels. Thus we might predict that a certain segment of society maintains skills which allow them to cope successfully with conditions relating to crowded environments and do sp, even though their preferences might dictate otherwise. Since I made no direct attempt to asses user prefer ences and satisfaction based on such factors as perceived crowding , It is not possible to know the extent to which this problem applies to the question of optimum vs maximum carrying capacities. The second reason for emphasizing the optimality question is that the present results do not allow one to know the extent to which the sample of beach users is representative of the overall source population. Only those people actually at the beach were sampled and thus no data exist for those individuals who do not participate. For example, certain people may forego a trip to the beach because they per ceive the area to be over crowded, too far away, or facilities not con sistent with expectations. Such people are thus 'filtered out' and there fore not represented in any sampling procedure utilizing on site inter views or observations. Similarly, as the results have shown, personal ities of users differ and certain differences seem to relate to spacing preferences. These differences, of course, only relate to the sample, - 79 -and given the sample was representative*, of users at the three sites in general. Since it is difficult to know whether various selection pro cesses mitigate against certain segments of the regional population it is doubtful that the personality indices are reflective of people in gen eral. Without an understanding of the relative proportion of key person ality variables of the source population it is also doubtful whether optimal carrying capacity estimates can be calculated. For example, in the present study respondents with elevated profiles on the pastor alism scale were observed to require more open space than those scoring highly on the urbanism scale. It seems probable that others not found in the sample would score more highly on the pastoralism scale and have even higher needs for space. Such people would rarely visit sites such as the three areas in the present study since their need for open space could not easily be satisfied in such environments. The arguments above would suggest that to equitably manage rec reational resources such as beaches, the manager should sample the source population to determine the relative proportion of individuals maintaining relevant personality characteristics. Armed with such data the designer or manager should be in a position to build or maintain facilities con sistent with the needs of both actual and potential users. Such a strat egy would eliminate the possibility of selecting against various segments of society. Granting, that, in many cases, the manager may not have suffic ient resources to complete the requirements of such a study, an alternate strategy might consist of ensuring the presence of a range of facilities, each satisfying one segment of the range of user preferences. Such a tactic would provide valuable data on use rates for each facility and the manager could then infer the relative need for each class of facility. As mentioned in an earlier section, if such guidelines or procedures are not forthcoming, then the estimates derived from the present study would probabably serve to ensure adequate satisfaction for the largest number of people. At least using these estimates one can be relatively well assured of not seriously detracting from the spatial"hee~ds~of users. - 80 -The discussion to this point has centered on space guidelines for public beaches; however, since the results were consistent with other more general work (Hall, 1966 for example), the findings may apply to other settings. The results seem especially applicable to environments where the type of social interaction (or lack of it) is consistent with behaviours characteristic of Hall's 'social and public distance zones'. For example, in settings where the maintenance of the integrity of the social group (including solitary indivduals) is important, the present research suggests designers should allow for at least 2.7 meters between the boundaries of any two design elements. A typical example of the type of design setting where these results could be applied, are airport, train and bus waiting areas. The present research suggests that for these areas, seating clusters should not be placed much closer than the 2.7 meter zone above. Other areas where these results might apply are, plazas, parks, restaurants, etc. Of course, in these settings the use of plants and other suitable perceptual barrier systems might be used to effectively decrease this space requirement The most important point of the above discussion is that space itself communicates the need to be separate from others. By struc turing space in this way, other more costly, behaviourally oriented space control mechanisms need not be called upon by the individual to maintain the social identity of the group. In an age where high density environ ments are often common features of our daily lives, the use of space stan dards based on behavioural criteria of actual and potential users seem crucial. Without such standards, it seems likely that many environments will continue to compromise user needs and as a result exacerbate the stress such high density settings undoubtedly offer. Toward further research. Many questions relating to the ways in which people structure and use space, expecially with respect to increasing density remain unanswered. The present research served as a source for many such questions, the most important of which are listed below: - 81 -1) What arc the effects of various visual and auditory barriers on perceived density and spatial needs? 2) Does the experience of living in large urban envir onments provide people with coping strategies not available to rural inhabitants? 3) If these coping strategies exist what are their form and how do they function under varying density conditions? 4) What are the cultural traditions and behaviour patterns that increase or decrease the ability of indiv iduals and groups to cope with close spatial proximity and high density conditions? 5) What are the forces of selection in determining who uses a particular environment and how do these forces relate to the observed population, i.e. who are the people who decided not to participate? Answers to these questions might be best acquired by studying a population of beach users from a large metropolitan area such as New York, Los Angeles, or Hong Kong where beach densities reach levels in excess of the maximum derived from the present study. Of special inter est would be the plot of average nearest neighbor distance versus density. For example one might predict a threshold effect such that as in the pre sent study a similar asymptote would be observed until the maximum derived density was reached whereupon a new lower asymptote would appear. This would indicate a need for a basic amount of space but would differ in that users at the highest densities would be satisfied to obtain much less space in order to participate in the beach experience. This process if observed might be linked to prior expectations of how crowded the beach would be or to the perceived costs and benefits of the recreational exper ience. Such high density beaches would also offer the opportunity to examine ways of limiting social interaction other than the regulation of space. Such mechanisms might include behaviours associated with minimizing - 82 -eye contact, such as controlling body orientation and gaze, lying face down, falling asleep, reading, etc. Similar behaviours have been shown to serve as powerful regulators of social interaction nnd sensory input (Chance 1962; Argyle and Dean, 1965; Grant, 1969; Goldberg. Kiesler and Collins, 1969; McGrew, 1972; Efran and Cheyne, 1974). Such a study would provide important data on how and when these behaviours are used and whe ther their occurrence and frequency are related to density considerations in a natural setting. The answers to these questions and others would provide valuable information pertaining to issues and concepts relating to crowding, stim ulus overload and the effects of "selection in determining the composition of any given referent group. A public beach may be one of the most useful settings in which to conduct such research. - 83 -LITERATURE CITED Altman, J. 1975. The Env i ronmcnt and Social Behaviour: Privacy, Personal Space, Territory, Crowding. Monterey: Brooks/Cole Publishing Co. Ardrey, R. 1966. The Territorial Imperative. Atheneum, New York. Argylo, M. & Dean, J. 1965. Eye contact, distance & affiliation. Sociometry 28:289-304. Bailey, K. G., Hartnett, J. J. & Gibson, S. W. 1972. Tmplied threat and the territorial factor in personal space. Psychological Reports 30:263-270 Barker, R„ I960. Ecology and motivation. Nebraska Symposium on Motivation 8:1-50. Barker, R„ & Gump, P. 1964. Big School, Small School. Stanford Univ. Press. Becker, F. D. & Mayo, C. 1971. Delineating personal distance and terri tory, Environment and Behaviour 3:375-382. Becker, F.D. 1973. Study of spatial markers. J. of Personality and Social Psychology 26:439. Blanchard, D. B., & Erickson, M. M. 1949. The cycle of the Gambel sparrow. University of California Publications in Zoology 47:255-318 Blank, T. H. & Ash, J. S. 1956. The concept of territory in the partridge Perdix p. perdix. Ibis 98:379-389. Blishen, B. 1958. The construction and use of an occupational class scale. Canadian Journal of Economics and Political Science 24:519-53lT ' Brougham, J. 1968. An evaluation of crowding upon the quality of the recreational experience. M.A. Thesis. Department of Geography University of Western Ontario. Brown, J.L. 1964. The evolution of diversity in avian territorial systems. The Wilson Bulletin 76:160-169. Brown, J. L. & Orians, G. H. 1970. Spacing patterns in mobile animals. Annual Review of Ecology and Systematics. 1:239-262. Burt, W.H. 1943. Territoriality and home range concepts as applied to mammals. Journal of Mammalogy 24(3):346-52. California Public Outdoor Recreation Plan, Part II. California Public Outdoor Recreation Committee. 1960. p. 48. Chance, M. R. A. 1962. An interpretation of some agonistic postures; the role of "cut-off" acts and postures. Symposia of the Geological Society of London 8:71-89. - 84 -Clarke, P. J. & Evans, F. C. 1954. Distance to nearest neighbour as a measure of spatial relationships in populations. Ecology 35:445-453. Conder, P. J. 1949. Individual distance. Ibis 91:649-655. Cook, M. 1970. Experiments on orientation and proxemics. Human Relations 23:61-76. Craik, K. II. 1966. The prospects for an environmental psychology. I PAR Research Bulletin. Berkeley: University of California. Craik, K. H. 1969. Assessing environmental dispositions. Paper read at the Annual Meetings of the American Psychological Association, Washington, D.C. Craik, K. H. 1970 (a). The environmental dispositions of environmental decision-makers. Annals of the American Academy of Political and  Social Science 389:87-94. Craik, K. H. 1970 (b). Environmental Psychology. In K. II. Craik, et al New Directions in Psychology IV. New York: Holt, Rinehart and Winston. Curtis, J. T. & Mcintosh, R. P. 1950. The inter-relationship of certain analytic phyto-sociological characters. Ecology 31:434-455. Dacey, M. F. 1964. Two dimensional random point patterns: A Review and an interpretation. Papers and Proceedings of the Regional Science Association 13:41-58. Davis, Go & Ayres, V„ 1973, Photographic methods on behavior in the human milieu: New developments and critique. In Environmental Design Research Assoc. Vol. 2,. pp. 280-285. ' '. ~~ Dice, L. R. 1952. Measure of spacing between individuals within a pop ulation. Contributions of the Laboratory of Verebrate Biology. University of Michigan 55:1-23. Duke, M. P. & Nowicki, S. Jr. 1972. A new measure and social learning model for interpersonal distance. Journal of Experimental Research in Personality 6:119-132. Edney, J. J. & Jordon-Edhey, N.L. 1974. Territorial spacing on a beach. Sociometry 37 (1): 92-104. Efran,M. G. & Cheyne, J.A. 1973. Shared space: Cooperative control of spatial areas by two interacting individuals. Canadian Journal of Behavioural Science 5:201. Efran, M. G. & Cheyne, J.A. 1974. Affective concomitance of the inva sion of shared space: Behavioural, physiological and verbal indica tors. Journal of Personality and Social Psychology 29:219-226. - 85 -Felipe, N. J. & Sommer, R. 1966. Invasions of personal space. Social Problems 14 (2):206-214 -Getis, A. 1964. Temporal land-use pattern with the use of nearest neigh bour and quadi-at methods. Annals of the Association of American Geographers 54:391-399. Gleason, H. A. 1920. Some applications of the quadrat method. Bulletin of the Torey Botanical Club 47:21-33. Goldberg, G.N., Keisler, C. A. & Collins, 13. E. 1969. Visual behaviour and face-to-face distance during interaction. Sociometry 32:43-53. Goin, Olive B. & Goin, C.J. 1962. Introduction to Herpetology. W.H. Freeman, San Francisco. Grant, E. C. 1969. Human faciad expression. Man 4:525-536. Greig-Smith, P. 1964. Quantitative Plant Ecology. Second edition. London: Butterworth and Co. Limited. Hall, E.T. 1966. The Hidden Dimension. New York: Doubleday. Hall, E. R. L. & Devore, I. 1965. Baboon social behaviour. In Primate Behaviour: Field Studies of Monkeys and Apes. (I. Devore, Ed.). 53-110. New York: Holt, Rinehart and Winston. Hardwick, W. G. & Collins, J. B. 1973. Vancouver Urban Futures Project: Interim Report Series. University of British Columbia, Departments of Geography and Adult Education. Hediger, H. 1941. Biologische Gesetzmassigkeiten im Verhalten von Wirbeltieren. Mitteilungen der Naturforschenden Gesellshaft Bern, 1940, pp.37-55" " ~ ' Hediger, H. 1955. Studies of the psychology and behaviour of captive animals in zoos and circuses, trans, by G. Sircom. Criterion Books, New York. Reprinted as The psychology and behaviour of animals in zoos and circuses, Dover, New York, 1968. Karabenick, S. & Meisels, M. 1972. Effects of performance evaluation on interpersonal distance. Journal of Personality 40:275-286. Kaufman, J.H. 1962. Ecology and social behavior of the coati, Nasua narica, on Barro Colorado Island, Panama. University of California  Publications in Zoology 60:95-222. Kummer, H. 1971. Spacing mechanisms in social behaviour. In Man and Beast: Comparative Social Behaviour. (J.F. Eisenberg & W. S. Dillon, Eds.) Smithsonian Institution Press. - 86 -Lamphear, S. C. 1970. Personality and recreation: Study of particpant behaviour in selected outdoor recreation activities. Dissertation  Abstract International 30, 12-B.5314. Lorenz, K.Z. 1966. On Aggression. Harcourt, Brace & Jovanovich, New York. Lorr, M., Daston, P. & Smith, I. R. 1967. An analysis of mood states. Education and Psychological Measurement 27:89-96. Luft, J. 1966. On nonverbal interaction.. Journal of Psychology 63:261-26* McBride, G. 1964. A general theory of social organization and behaviour. University of Queensland Papers. Faculty of Veterinary Science Vol. 1 (2). The University of Queensland Press. St. Lucia. McBride, G., King, M.G. & James, J.W. 1965. Social proximity effects on galvanic skin responses in adult humans. Journal of Psychology 61:153-157. McGrew, W. C. 1972. An ethoiogical study of children's behaviour. New York: Academic Press. McKechnic, G. E. 1973. Manual for the Environmental Response Inventory (ERI). Research Edition, Consulting Psychologists Press, Inc., Palo Alto, California. McKechnie, G. E. 1974. The psychological structure of leisure: past behaviour. Journal of Leisure Research 6:27-45. Meisels, M. & Canter, F. M. 1970. Personal space and personality charac teristics: A non-confirmation. Psychological Reports 27:287-290. Meyer, P. A. & Bryan, R. C. 1974. Recreational crowding and the prospects for fish-related recreation - Pacific Rim National Park. Fisheries and Marine Service, Southern Operations Branch, Pacific Region; Technical Report series No. PAC/T-74-22. Miller, R. S. & Stephen, W. J. D. 1966. Spatial relationships in flocks of sandhill cranes. (Grus canadensis). Ecology 47:323-327. Mohn, E. & Stavem, P. 1974. On the distribution of randomly placed discs. Biometrics 30:137-156. Morisita, M. 1954. Estimation of population density by spacing method. Mem. Faculty of Science, Kyushu University. Series E 1:187-197. Morton, M. L. 1967. Diurnal feeding patterns in white-crowned sparrows, Zonotrichia leucophrys gambelii. Condor 69:45-64. Noble, G.K. 1939. The role of dominance in the social life of birds. Auk 56:263-73. - 87 -Patterson, M. L. 1973. Stability of non-verbal immediacy behaviours. Journal of Experimental Social Psychology. 9:97-109. Patterson, M. L. & Holmes, D. S. 1966. Social interaction correlates of MMPI extroversion-introversion scale. American Psychologist 21:724-725. Patterson, M. L., Mullens, S. & Romano, J. 1971. Compensatory reactions to spatial intrusion. Sociometry 34:114-121. Pielou, E. C. 1969. An Introduction to Mathematical Ecology. New York: John Wiley and Sons, Inc. Rapoport, A. 1975„ Toward a redefinition of density,, Environment & Behavior 7.(2) :133-1580 " Sabine, W. S. 1956. Integrating mechanisms of the junco winter flock. Condor 58:338-341. Seton, E.T. 1909. Life-histories of northern animals: An account of the mammals of Manitoba, 2 vols. Charles Scribner's Sons, New York. Sewell, A. F. & Heisler, J.T. 1973, Personality-correlates of proximity preferences. Journal of Psychology 85:151. Skellam, J. G. 1952. Studies in statistical ecology. I. Spatial pattern, Biometrika 39:346-362. Smith, G. H. 1953. Size-distance judgements of human faces (projected images). Journal of General Psychology 49:45-64. Smith, G. H. 1954. Personality scores and personal distance effect. Journal of Social Psychology 39:57-62. Sommer, R. 1966. Man's proximate environment. Journal of Social Issues 22 (4) :59-70. Sommer, R. 1969. Personal space: The behavioural basis of design. Englewood Cliffs, N.J. : Prentice-Hall, Inc. Sommer, R. & Becker, F. D. 1966. Territorial defense and the good neigh bor. Journal, of Personality and Social Psychology 11:85-92. Stokols, D..1976. The experience of crowding in primary and secondary environments. Environment & Behavior 8(1) :49-86„ Thompson, H. R. 1956. Distribution of distance to Nth neighbour in a popu lation of randomly distributed individuals. Ecology 37 (2):391-394. - 88 -Tinbergen, N. 1952. The curious behavior of the Stickleback. Scientific  American 182 (6):22-26, Verburg, K. & Rees, W. E. 1975. The carrying capacity of recreational lands: A review. Occasional Paper No. 1, Parks Canada, Prairie Region. Winnipeg, Manitoba. Webb, E.J., Campbell, D.T., Schwartz, R.D. & Sechrest, L. 1966. Unobtrusive Measures: Nonreactive Research in the Social Sciences. Rand McNally, Chicago, 111. Weinstein, L. 1968. The mother-child schema, anxiety, and academic achieve ment in elementary school boys. Child Development 39:257-264. Wicker,.A. 1973. Undermanning theory and research; Implications for the study of psychological and behavioral effects of excess populations. Representation Research in Social Psychology 4:185-206, Wicker, A,, McGrath, J, and Armstrong, G. 1972. Organization size and be havior setting capacity as determinants of member participation. Behavioral Science 17:499-513. Willems, E.P. & Campbell, D.E. 1976. One path through the cafeteria. Environment & Behavior 8(1):125-140. Williams, J. L. 1971. Personal space and its relation to extroversion-introversion. Canadian Journal of Behavioural Science 3:156-160. Wilson, E. 1975. Sociobiology: The New Synthesis. Cambridge, Mass.: Harvard University Press. 88a -APPENDIX A THE SURVEY - 89 -School of Community & University of British Time Group M F Map No. Regional Planning Area_ Columbia Date RECREATIONAL ATTITUDE SURVEY As part of a study to determine how people view particular recreational environments, we have devised three surveys which are contained in this booklet. Your co-operation will provide a better understanding of how people perceive and behave with respect to each other as well as toward certain aspects of recreational.settings. The first section contains 184 statements concerning various aspects of the environment and your own attitudes. The second is a checklist of recreational and leisure activities which you may have participated in at one time or another. The third is a short list of questions concerning your own background which will help in understanding why people differ. No name or other identification is required and your anonymity is ensured. Work quickly -- first impressions are usually the most accurate. Most people finish in 20 - 25 minutes. Each section is self-explanatory and contains its own set of instructions. Thank you for your co-operation and time! The first section begins on the next page. - 90 -- 2 -ENVIRONMENTAL RESPONSE INVENTORY Please read each statement and decide quickly whether you personally agree or disagree with it. To respond, simply circle the answer to the left of each statement according to these categories: SA ~ Strongly Agree A — Agree N — Neutral D — Disagree SD — Strongly Disagree Again, work quickly. Do not be concerned if some items seem similar to ones you have seen earlier. SA A N D SD 1. I like amusement parks. SA A N D SD 2. I would enjoy the work ' of an architect SA A N D SD 3. SA A N D SD 4. SA AND SD SA A N D SD 6. SA AND SD 7. SA A N D SD 8. SA AND SD 9. Machines increase man's freedom. I prefer to live in an area where neighbours keep to themselves. I would enjoy driving a racing car. The idea of walking into the forest and "living off the land" for a week appeals to me. Life in the city is more interesting than life on a farm. I would enjoy building a radio. Travelling isn't really worth the effort. SA A N D SD 10. I have my best thoughts when I am alone. SA A N D SD 11. I enjoy browsing in bookstores. SA A N D SD 12. It would be fun to move around and live in different parts of the country. SA A N D SD 13. It is boring to spend all day working with . your hands. SA A N D SD 14. SA A N D SD 15. SA A N D SD 16. SA AND SD 17. SA AND SD 18. SA A N D SD 19. It is exciting to go shopping in a large city. There should be a law against skyscrapers. I like to be by myself much of the time. I enjoy browsing in antique shops. I sometimes daydream of being stranded on a tropical island. I like places that have the feeling of being old. SA A N D SD 20. I shudder at the thought of finding a spider in my bed. SA A N D SD 21. SA A N D SD 22. SA A N D SD 23. SA A N D SD 24. SA A N D SD 25. SA AND SD 26. SA A N D SD 27. SA A N D SD 28. I would enjoy traveling around the world on a sailing ship. Alleys are interesting places to explore. I prefer a stick-shift car to one with an automatic transmission. I like crystal chandeliers. I like homes with stone floors. I like the variety of stimulation one finds in the city. I usually save spare nuts and bolt3. I get annoyed when my neighbours are noisy. SR A N D SD 29. SA A N D SD 30. SA A N D SD 31. SA A N D SD 32. SA AN D SD 33. SA A N D SD 34. SA AND SD 35. SA A N D SD 36. SA A N D SD 37. SA A N D SD 38. SA A N D SD 39. SA AND SD 40. SA A N D SD 41. SA AN D SD 42. SA AND SD 43. SA AND SD 44 . SA AND SD 45. SA AND SD 4G. SA A N D SD 47. When buying clothes, I usually look more for comfort than for style. _ • I am quite skillful with my hands. It's annoying to have to share an office or work space with someone. I like to visit historic places. Suburbs should replace the city as the centre of cultural life. I would prefer working with precision power tools. I have difficulty concentrating when things are noisy. I would rather remodel an old house than build a new one. Wo must move ahead and not worry about past failures. Cities are too noisy and crowded for me. I often feel uneasy in a large crowd of people. I can repair just about anything around the house. I often have trouble getting the privacy I want. There should be a law against anyone owning more than a thousand acres of land. I feel most secure when I am working around the house. It is hopeless to try to save our cities. It would be fun to own some old-fashioned costumes. Motorcycles should be kept out of recreation areas. I like modern furniture better than the more traditional styles. SA A N D SD 48. SA A N D SD 49. SA A N D SD 50. SA A N D. SD 51. SA A N D SD 52. SA A N D SD 53. SA A N D SD 54. SA A N D SD 55. SA A N D SD 56. SA A N D SD 57. SA AND SD 58. SA AND SD 59. SA A N D SD 60. SA A N D SD 61. SA A N D SD 62. I would like a job that involved a lot of traveling. It is important for me to own top quality equipment. As a child, I often watched when someone repaired things around the house. I like the sounds of a city street. Old sections of the city are more interest ing than the new areas. I often feel lonely when I am by myself. As a child, I was taught respect for all living things. It is good for man to submit to the forces of nature. I prefer friends who are reliable and even-tempered. I often think of settling down on a farm some day. I don't like being completely alone. I would like to live in a modern, planned community. Zoning laws and other building controls are necessary to protect the rights of the public. I like things that have precision moving parts. I would enjoy enter taining famous people. - 4 -SA A N D SD 63. SA A N D SD 64. SA A N D SD 65. SA AND SD 66. SA A N D SD 67. SA AND SD 68. SA A N D SD 69. SA A N D SD 70. SA A N D SD 71. SA AND SD 72. SA A N D SD 73. SA A N D SD 74. SA AN D SD 75. SA A N D SD 76. SA AND SD 77. SA A N D SD 78. SA A N D SD 79. I often feel that I am a part of the space around me. I can identify many of the local flowers and trees. I would like to work with computers. I have vivid memories of where I lived as a child. Our national forests should be preserved in their natural state, with roads and buildings prohibited. Plying in a small airplane would make me nervous. As a child, I was afraid of being outside by myself. It is better if people live out their lives in one place. I would enjoy owning a fancy watch. ' I would enjoy riding a motorcycle. Making rain by artificially "seeded" clouds is a great technological advance. I enjoy staying up all night. I am happiest when I am alone. No child should have to grow up in a rural area. I get annoyed when people drop by my house without warning. A fireplace adds a special feeling of cor.incss to a room. It's interesting to learn about the history of the Place where you live. SA A N D SD 80. SA A N D SD 81. SA A N D SD 82. SA A N D SD 83. SA A N D SD 84. SA A N D SD 85. SA A N D SD 86. SA A N D SD 87. SA A N D SD 88. SA A N D SD 89. SA A N D SD 90. SA A N D SD 91. SA A N D SD 92. SA A N D SD 93. SA A N D SD 94. SA A N D SD 95. SA A N D SD 96. It is fun to make scale models of things. I would enjoy living the rest of my life in a large city. Electricity fascinates me. I like social gatherings where I can enjoy myself without worrying about other people. I don't think that I would ever want to be hypnotized Small town life is too boring for me. Fertilizers improve the quality of food. I often get the feeling that I just must be alone. A person has a right to modify the environment to suit his needs. Sometimes I'm afraid of too much stimulation - from sounds, colours, odors, etc I understand the architect ural idea that form follows function. I enjoy working in a flower garden. I enjoy owning a good piece of equipment, even if I don't get to use it much. I pride myself on having a home which is always open to friends. Fences make good neighbours I'd rather live in the suburbs than in the city. A complex technological society cannot tolerate individuality. - 5 -SA AND SD 97. SA AND SD 98. SA AN D SD 99. SA A N D SD 100. SA A N D SD 101. SA A N D SD 102. SA A N D SD 103. SA A N D SD 104. SA A N D SD 105. SA AND SD 106. SA A N D SD 107. SA A N D SD 108. SA AND SD 109. SA A N D SD 110. SA A N D SD 111. SA A N D SD 112. SA A N D SD 113. I enjoy a change in the weather even when it .turns bad. It is unsafe to ride on buses these days. Country people are more honest than city people. Hiking is boring. I'd be afraid to live in a place where there were no people nearby. I find street noise very distracting. I have always been some what of a daredevil. I would enjoy riding in a crowded subway. I am quite sensitive to the "character" of a building. I like to ride on roller coasters. I enjoy tinkering with mechanical things. I do not like to loan things to neighbours. I would enjoy living in a historic house. Sometimes I wish I had power over the forces of nature. I have no interest in ballet. I like to read about the history of places. DirtH control practices should be accepted by everyone. SA A N D SD 114. SA A N D SD 115. SA A N D SD 116. SA A N D SD 117. SA A N D SD 118. SA A N D SD 119. SA A N D SD 120. SA A N D SD 121. SA A N D SD 122. SA A N D SD 123. SA A HD SD 124. SA A N D SD 125. SA A N D SD 126. SA A N D SD 127. SA A N D SD 128. SA A N D SD 129. Jet air travel is one of the great advances of our society. I have vivid memories of the neighbourhood where I grew up. I would enjoy going to the opera. Today people are too isolated from the forces of nature. It is easy for me to work undistracted in most situations. I like to dress in the latest fashions. I seldom pay attention to what I eat. It is dangerous to work around heavy machinery. The wilderness is cruel and harsh. Modern buildings are seldom as attractive as older ones. I like experimental art. I often wish for the seclusion of a weekend retreat. I would like to own an expensive camera. Building projects which disrupt the ecology should be abandoned and the land returned to its natural state. The problems of the cities will never be solved. I am easily distracted by people moving about. -94 -SA A N D SD 130. SA A N D SD 131. SA A N D SD 132. SA A N D SD 133. SA A N D SD 134. SA A N D SD 135. SA A N D SD 136. SA AND SD 137. SA A N D SD 138. SA A N D SD 139. SA A N D SD 140. SA AND SD 141. SA A N D SD 142. SA A N D SD 143. SA A N D SD 144. SA A N D SD 145. I often have trouble finding my way around a new area. In spite of all talk about pollution, the earth is still a safe place to live. I need more variety in my life than other people seem to need. I usually avoid public rest rooms. I often have trouble figuring out how to use household appliances. I usually enjoy having lots of people around. I would enjoy watching movies made 15 or 20 years ago. Natural resources must be preserved even if people must do without. I like to get up early to see the sun rise. I am afraid of driving in the city. Trespassing laws should be more carefully enforced. I am an adventurous person. I often have strong emotional reactions to buildings. There is too little emphasis on privacy in our society. It is dangerous nowadays to live in a large city. I seldom vary the route I take to everyday destinations. SA A N D SD 146. SA A N D SD 147. SA A N D SD 148. SA A N D SD 149. SA A N D SD 150. SA A N D SD 151. SA A N D SD 152. SA A N D SD 153. SA A N D SD 154. SA A N D SD 155. SA A N D SD 156. SA A N D SD 157. SA A N D SD 158. SA AND SD 159. SA A N D SD 160. SA A N D SD 161. SA A N D SD 162. SA A N D SD 163. It is important for me to feel that I am in harmony with the forces of nature. When it comes to fixing things, I am hopeless. Modern communities are plastic and ugly. Science does as much harm as good. I get upset if I must do too many things at once. I would feel safer on the highway if speed limits were reduced. I would like to take flying lessons. Most jewellry is a waste of money. I like to say hello to my neighbours. I enjoy collecting things that most people would consider junk. There are often times when I need complete silence. I worry a lot about the rising crime rate. The cultural life of a big city is very important to me I like to go to shopping centres where everything is in one place. I ara fond of oriental rugs. I am afraid of heights. People who try to repair appliances themselves usually end up breaking them I would like to live in a palace or a castle. - 95' -- 7 -SA A N D SD 164. SA A N D SD 165. SA A N D SD 166. SA A N D SD 167. SA A N D SD 168. SA A N D SD 169. SA A N D SD 170. SA AND SD 171. SA A N D SD 172. SA A N D SD 173. SA A N D SD 174. Sight-seeing is tedious and boring. The cities contain the best aspects of modern life. It's nice to buy a new car every year or so. Bathtubs have become obsolete. Places often play an important role in my dreams. I would like to build a cabin in the woods. I enjoy being in dangerous places. Everyone should have the opportunity to live in a great city. It's fun to walk in the rain even if you get wet. Old buildings are usually depressing. I would enjoy living on a houseboat. SA A N D SD 175. SA A N D SD 176. SA A N D SD 177. SA A N D SD 178. SA A N D SD 179. SA A N D SD 180. SA A N D SD 181. SA AND SD 182. SA A N D SD 183. SA A N D SD 184. Computers may someday take over the world. I like to be on the move, not tied down to any one place. Mental problems are more common in the city than in the country. Odors often bring back distant memories. I like to care for animals. A man should spend his leisure time at home with his family. If I had the money, I would enjoy owning an expensive stereo set. I feel a great attraction to the sea. I would rather sleep on the open ground, than in a tent. Given enough time, science will solve most human problems. -96- -LEISURE ACTIVITIES BLANK Balow is a list of leisure and recreational activities. For each activity indicate the extent of your participation using the following system: N - You have never engaged in the activity. T You tried it once or a few times. 0 - You used to to it regularly, but now no longer do it regularly. 0 - You occasionally participate in the . activity at this time. R - You currently participate regularly in the activity. Check the appropriate blank to indicate your participation in each of the following activities: >. iH rH iO >. •P C •H 0 0 U 4J •H <0 u •o 0) rH UJ -0 id 3 > •H 0) u tr> 01 U U) o 01 z E-i D o oz, N T u 0 R 10 >. 4J c rH •H 0 0 IH 4J -rl 10 u •d Ul rH di 01 •a 10 3 > •H o CP Cl U 01 u <1) H o a; N T u 0 R 1 Acting (dramatics 2 Amateur radio 3 Archery 4 Attending concerts 5 Attending auctions 6 Auto racing 7 Auto repairing 8 Back packing 9 Badminton 10 Baseball or Softball 11 Basketball 12 Bicycling .13 Billiards or pool 14 Bird watching 15 Boating (rowing) 16 Bookbinding 17 Bowling 18 Boxing 19 Camping 20 Canoeing 21. Carpentry 22 Ceramics or pottery 23 Checkers or go 24 Clu-sr. 25 Child-related activities (e.g. , scouts, PTA) 26 Civic organizations 27 Collecting (antiques, coins, etc.) 28 Conservation or ecology organizations 29 Cooking and baking 30 Crossword puzzles 31 Dancing ballet or modern 32 Dancing (social) 33 Darkroom work (photography) 34 Designing clothes 35 Dining out 36 Driving (motoring) 37 Electronics 30 Encounter groups 39 Exercising 40 Fencing 41. Fishing (deep-sea) 42 Fishing (fresh water) 43 Flower arranging 44 Flying (or gliding) 45 Folkdancing 46 Footbal1 47 Fraternal organizations 48 Gambling (casino) 49 Gardening 50 Going to movies - 97 -. •• LEISURE ACTIVITIES BLANK, p. 2. >, rH" Hi >. r. fH •ri 0 o - H JJ •a u T) tn ai •o 10 3 > •H ai o C u Ul u 01 sr. E-i D O OS N T U 0 R 4J c •H o o U .u •<-t D TI 111 0) •a 3 > •H o • al U in o a) Z E-i g a; N T U (3 R 51 Going to plays or lectures 91 Sculpture 52 Going to horseraces 92 Sewing 53 Going to nightclubs 3 Shuffleboard 54 Golf 94 Sightseeing 55 Gymnastics 5 Singing 56 Hiking or walking ; 96 Skiing 57 Home decorating 97 Skin diving 58 Homeowner organizations . 98 Social drinking 59 Horseback riding ; 9 Squash or handball 60 Horseshoes 100 Sunbathing 61 Hunting 1 Surfboarding 62 Ice skating 102 Swimming 63 Jewelry making 3 Table tennis (ping 64 Jig-saw puzzles pong) 65 Jogging 104 Taking pictures 66 Judo or karate ^ (photography) 67 Keeping pets 105 Talking on telephone 68 Kite flying ; ; 106 Tennis 69 Knitting or crocheting 107 Travelling abroad 70 Leatherworking 108 Visiting Museums 109 Visiting friends 71 Listening to the radio ' 110 Volleyball 72 Marksmanship \ 73 Mechanics 111 Volunteer fire fighting 74 Mctalworking 11.2 Watching team sports 75 Model building 113 Watching TV shows 76 Motorboating ; 114 Waterskiing 77 Motorcycling . 5 Weaving 78 Mountain climbing 116 Wei.ghtlifting 79 Needlework 117 Windowshopping 80 Painting and drawing 118 Wrestling 119 Writing poetry or 81 Playing poker stories 82 Playing bridge . 120 Writing letters 83 Playing records (music) 84 Playing a musical 121 Woodworking and instrument related crafts 85 Political activities . 86 Reading (books, plays, Others not listed (specify): poetry) 87 Reading (newspapers, ; . magazines) . ; •88 Religious organizations 89 Roller skating 90 Sailing . BACKGROUND INFORMATION Age: Sex: F M Marital Status: a. Single d. Divorced b. Married e. Separated c. Widowed f. Co-habiting If you have any children, how many? How many brothers and sisters do you have? Check the highest level of education which you attained? Elementary Some High School High School graduate Some University or College University Degree Some graduate work M.A. or equivalent Ph.D., M.D., L.L.B., E.D.D., etc. How many years have you lived in each of these urban centres: over 1 million -10,000 - 50,000 100,000 - 1 million 5,000 - 10,000 50,000 - 100,000 below 5,000 How many automobiles are at your disposal in your household? What is your occupation? Please be specific. What was your household income before taxes during the last tax year? Of your total time spent in recreational and leisure activities, what percentage is spent away from the city as opposed to in the city? Away from the city »( + in the city % = 100% OPTIONAL WORD LIST SURVEY If you feci you have any extra time there is an optional survey below which consists of 60 words which describe how you may feel at this time. The survey takes about 5 minutes and is designed to measure your personal feelings at this time. If you wish to complete the survey, for each word merely circle the number which best indicates how you feel at this moment according to the following scheme: 1. Not at all 2. A little 3. Moderately 4. Strongly 5. Extremely Work, quickly — first impressions are usually the most accurate. rH rH rH >1 rt a m >. rH •p rH 4-1 •p <d e id •P ki c OJ •rJ a> 0 u 4J rH •0 u p. 0 0 4J X Z < Ul w 1 2 3 4 5 rH >. rH rH >. id III HI >. rH rH •P rH 0) P •P rd Oi G id •P c (1) •H 01 o U JJ rH XI n *J 0 0 •p X 2 «: s in w 1 2 3 4 5 1. Active 1 2 3 4 5 2. Cheerful 1 2 3 4 5 3. Jittery 1 2 3 4 5 4. Pretty good 1 2 3 4 5 5. Angry 1 2 3 4 5 6. Excited 1 2 3 4 5 7. Bad-tempered 1 2 3 4 5 8. Apathetic 1 2 3 4 5 9. On edge 1 o 3 4 5 10. Nervous 1 .2 3 4 5 11. Pensive 1 2 3 4 ' 5 .12. Gay 1 2 3 4 5 13. Annoyed 1 2 3 4 5 14. Earnest 1 2 3 .4 5 15. Resentful 1 2 3 4 5 16. Helpless 1 2 3' 4 5 17. Sluggish 1 2 3 4 5 18. Serene 1 2 3 4 5 19. Worthless 1 2 3 4 5 20. Frightened 1 2 3 4 5 21. Calm 1 2 3 4 5 22. Contemplative 1 2 3 4 5 23. Nonchalant 1 2 3 4 5 24. Vigorous . 1 2 3 4 5 25. Serious 1 2 3 4 5 26. Tense 1 2 3 4 5 27. Furious 1 2 3 4 5 28. Languid 1 2 3 4 •5 29. Hated 1 2 3 4 5 30. Introspective 1 2 3 •  4 5 31. Lazy 1 2 3 4 5 32. Treoccupied 1 2 3 4 5 33. Thoughtful 1 2 3 4 5 34. Happy-go-lucky 1 2 3 4 ' 5 35. Top of the world 1 2 3 4 5 36. Hopeless 1 2 3 4 5 37. Weary 1 2 3 4 5 38. Full of pep 1 2 3 4 5 39. Light-hearted 1 2 3 4 5 40. Tired 1 2 '3 4 5 41. Relaxed 1 2 3 4 5 42. . Energetic 1 2 3 4 5 43. Composed 1 2 3 4 5 44. Lonely 1 2 3 4 5 45. At ease 1 2 3 4 5 46. Unhappy 1 2 3 4 5 47. Enthusiastic 1 2 3 4 5 48. Ready to fight 1 2 3 4 5 49. Carefree 1 2 3 4 5 50. Alert 1 2 3 4 5 51. Anxious 1 2 3 4 5 52. Grouchy 1 2 3 4 5 53. Shaky 1 2 3 4 5 54. Spiteful 1 2 3 4 5 55. Lively 1 2 3 . 4 5 56. Blue 1 2 3 4 5 57. Listless 1 2 3 4 5 58. Optimistic 1 2 3 4 5 59. Worried 1 2 3 4 5 60. Lethargic 1 2 3 4 5 - 99a -APPENDIX B BEACH USER PROFILE: SURVEY DIMENSIONS - 100 -Environmental Response Inventory. The means and standard deviations (by beach) for the first section of the survey (Environmental Response Inventory, ERI) are listed in Table 11 . Comparison of these results with those of Hardwick & Collins (1973) (Vancouver Urban Futures Project) and McKechnie (1973) indicate overall congruence with samples from Vancouver, British Columbia, Marin County, California and a cross section of students from U. S. colleges and universities. Leisure Activities Blank. The second por. tion of the survey concerned the participation by respondents in 121 leisure activities (Leisure Activities Blank, LAB). The means and standard deviations are listed in Table 12. These results are not directly comparable to McKechnie's means since his survey used a four point response format, whereas I used five. By multiplying the means for the present study by 0.8, a rough comparison is possible. Table 13 lists the transformed means for the beach study, McKechnie's means. The results oft tests (correlated means) indicated that the trans formed means for the beach results differed from McKechnie's for the fol lowing scales: Mechanics, Slow Living, and Neighborhood Sports (p< .01). Of the three scales which differ, only "slow living" is easily explained. On McKechnie's LAB the activity "sunbathing" loads highly on the factor "slow living" and thus, persons on a beach engaged in this activity could be expected to maintain a higher score than those persons sampled from the population at large who were not engaged in similar activities at the time the survey was administered. The differences between the scales "mechanics" and "neighborhood sports" may be due to differing recreational preferences of Canadians and Americans, since McKechnie took his sample from Marin County, California. - 101 -Table 11. ERI variables: means and standard deviations for English Bay, Kitsilano, and Skaha Beaches, (n. = 266) VARIABLE ENGLISH BAY KITSILANO SKAHA GRAND MEAN Mean S.D. Mean S. D. Mean S.D. PASTORALISM 77.3 9.6 77. 5 15. 8 78. 1 11. 2 77. 1 URBANISM 59.9 9.4 58.8 9. 6 56. 1 10. 0 57. 5 ENVIRONMENTAL ADAPTATION 68.9 9.4 67.1 9. 2 69. 0 10. 7 69. 1 STIMULUS SEEKING 67.6 11.0 70.1 10. 4 71. 0 12. 4 68. 9 ENVIRONMENTAL TRUST 61.9 9.3 63.0 8. 1 63. 4 8. 5 62. 6 ANTIQUARIANISM 67.2 10.2 69.7 9. 7 66. 9 9. 8 66. 9 NEED FOR PRIVACY 54.6 7.6 54.2 7. 0 54. 1 8. 2 54. 6 MECHANICAL ORIENTATION 63.6 9.3 62.0 8. 9 64. 5 9. 1 63. 5 COMMUNAL!TY 80.7 6.3 81.7 6. 4 80. 5 10. 0 80. 8 Table 12. LAB variables: means and standard deviations for English Bay, Kitsilano, and Skaha Beaches. VARIABLE ENGLISH BAY KITSILANO SKAHA GRAND MEAN Mean: S.D, Mean S.D. Mean: S.D. MECHANICS 38 .6 10.7 39. 7 10. 9 46.3 12. 7 42. 3 INTELLECTUAL 41 .4 8.9 42. 8 7. 5 40.6 9. 3 40. 7 CRAFTS 41 .6 12.1 40. 8 10. 1 37. 8 10. 3 39. 5 SLOW LIVING 74 .8 10.4 77. 8 6. 0 76.9 8. 2 76. 9 NEIGHBORHOOD SPORTS 33 .9 8.7 37. 2 8. 0 37.9 7. 6 36. 6 GLAMOUR SPORTS 33 .6 7.5 36. 1 8. 7 36 .6 8. 6 35. 2 FAST LIVING 9 .0 2.9 8. 7 2. 8 9.5 2. 7 9. 0 - 102 -Table 13. McKechnie's (1973) LAB Results Compared With Those From The Present Study. Scales McKechnie S.D. Present Study S.O. Mechanics 38.4 •11,3 33. 8 12.8 Crafts 33.0 8.8 31. 6 11.4 Intellectual 34.2 8.4 32. 6 9.3 Slow Living 59.7 9.1 61. 5 8.5 Neighbourhood Sports 26.0 5.6 29. 3 8.4 Glamour Sports 27.0 6.9 28. 2 8.9 (Note: McKechnie did not use the scale "fast living" in further analyses). - 103 -Socio-economic and demographic characteristics. The results of the questions requesting background information (socio/demographic data) of beach respondents is included in Table 14, • . Before categorizing users for these dimensions, several items require clarification. The first question (age of respondent) samples only those users 18 and over. Although the decision to exclude persons younger than 18 from the sample was to a certain extent arbirtary, I felt that the sur vey was more applicable to adults whose attitudes and opinions are prob ably more stable and less subject to change. The average age of the res pondent (30.1) is thus biased upwards compared to that of the beach pop ulation for this study. . • To ease mathematical computation, variable three (marital sta tus) was reduced from six response possibilities to two. Thus, unmarried (1) included "single", widowed", "divorced", and "separated" whereas married (2) included the category "co-habiting". The results thus indicate that slightly over one half (58%) of all respondents were mar ried. Question six regarding education, was divided into eight res ponse blanks where a respondent checked his level of education. Low numbers correspond to low education level attained and vice versa. Question seven asked respondents the number of years they had lived in six different size urban centers. These data thus represent six independent variables and were asked in the following order: 1. Over one million 2. 100,000 - one million 3. 50,000 - 100,000 4. 10,000 - 50,000 o 5. 5,000 - 10,000 6. Below 5,000 Responses to question nine, "What is your occupation?" were categorized according to an occupation class scale (Blishen, 1958) which is a scheme whereby a respondent's Job is ranked according to its rela tive prestige. In this case low numbers are associated with high status. There are a total of seven classes with the following four classifications Table 14. Socio/demographic variables: means and standard deviations for English Bay, Kitsilano and Skaha beaches. (n=266) VARIABLE ENGLISH KITSILANO SKAHA BAY GRAND Mean S.D. Mean S.D. Mean S.D. MEAN AGE 31.1 8.6 29.4 9.9 30.2 9.9 30.7 SEX- 1.4 0.4 1.4 0.5 1.6 0.5 1.4 MARITAL STATUS 1.5 0.5 1.4 0.5 1.6 0.5 1.6 NO. OF CHILDREN 1.1 1.4 0.7 1.2 1.3 1.6 1.2 NO. OF SIBLINGS 3.6 2.4 2.9 3.1 2.8 2.3 3.1 EDUCATION 3.6 1.2 3.8 1.4 3.6 1.5 3.5 NO". OF YEARS LIVED IN CITIES WITH POPULATIONS OF: OVER ONE MILLION 8.6 9.8 6.6 10.0 5.2 8.3 5.6 100,000 - ONE MILLION 10.5 10.4 8.0 10.3 8.5 9.9 8.6 50,000 - 100,000 2.9 7.2 2.8 7.2 2.6 6.0 3.1 10,000 - 50,000 3.2 5.4 3.8 6.4 3.0 6.6 3.6 5,000 - 10,000 0.9 1.9 2.3 5.2 2.1 5.1 1.9 BELOW 5,000 1.6 3.2 4.6 6.6 7.2 10.3 5.8 NO. OF AUTOMOBILES 1.0 0.7 0.9 0.7 1.5 0.9 1.3 JOB CATEGORY 3.8 1.3 3.1 1.2 3.5 1.2 3.5 INCOME 8191 3534 10677 9301 12848 6893 11324 % TIME RECREATING OUTSIDE URBAN ENVIRONMENT 34 24 27 20 40 30 35 DISTANCE TO HOME 498 1136 221 835 331 664 360 POPULATION OF HOME CITY 1052500 552660 1084600 285680 676740 492880 824385 - 105 -excluded from the analysis: 1) housewife, 2) retired, 3) unemployed, and 4) student. Income (question ten) is expressed as total household income before taxes for the previous tax year. A final question provided a description of the percentage of time a user spent in recreational and leisure activities outside the urban environment. Additional information collected from the respondent by the surveyor included the following: 1) Total number of people in the group from which the respon dent was selected. 2) Group composition according to sex. 3) The number of children in the group. 4) The straight line distance in miles from the city where the beach was located, to the city where the respondent resided. Note: This variable only applied to non-residents. 5) Population of the city of origin of the respondent. Note: This variable was collected for every respondent whether resident or non resident. The variables, day of the week and time of day were encoded so that a low number corresponded to low density conditions. For example, densities follow a generally increasing trend from Monday through the weedend, and from morning to afternoon. As a result, Monday was labelled as one, and time was encoded on a 24 hour clock basis, and morning hours were thus numerically smaller than afternoon times. The average beach, user who responded to the survey may be cat egorized by the socio/demographic variables in the following manner: 1) Approximately 31 years old (biased upwards because no users less than 17 were asked to complete the;', survey) . 2) 45% were males. 3) 58% were married. 4) Number of children - 1.2. 5) Number of siblings - 3.1. 6) Level of education reached,equal to a point between high school graduate and some college. - 106 -7) Spent the most years in urban centers which were very large (50,000 - over 1 Million) and very small (less than 5,000). 8) Had at their disposal 1.3 automobiles. 9) Maintained occupations which were exactly midway between extremes on a prestige scale. 10) Had a household income before taxes of $11,324. 11) Spent a little over one-third of their recreational time away from the city. 12) Non-residents were, on the average, 360 straight line miles from home. 13) The average population of the city where the respondent res ided was approximately 825,000 ppople. Mood Adjective Checklist. The analysis of the mood scale resulted in a profile much as one would expect given the context of the beach environment. Table 15 indicates that respondents scored most highly on the factors "cheerful", "energetic", "thoughtful", and "relaxed". As mentioned previously, these results are probably biased toward these more"positive" mood states, since the checklist was an optional feature of the survey. Of the 266 surveys finally included in the analysis, 68% elected to complete the mood, checklist. - 107 -Table 15. Mood score means and sta Kitsilano, and Skaha bea ENGLISH VARIABLE BAY Mean.' S.D. CHEERFUL 29.0 5.2 ENERGETIC 20.1 4.3 ANGER - HOSTILITY 10.7 2.9 TENSE - ANXIOUS 10.3 3.0 DEPRESSED 10.4 2.7 INERT - FATIGUED 10.7 2.7 THOUGHTFUL 18.1 3.8 RELAXED - COMPOSED 18.1 3.1 MOOD 1.6 0.6 dard deviations for English Bay, lies. GRAND KITSILANO SKAHA MEAN Mean S.D. Mean S. D. 28.1 6.8 28. ,7 6. 3 29.0 19.8 5.9 19. ,2 5. 9 19.7 11.7 4.1 9. . 5 2. 2 10.6 10.7 4.0 9. , 2 2. 6 10.0 12.0 4.6 9, . 8 2. 1 10.7 11.8 3.2 11. .9 3. 3 11.7 18.2 4.6 17. ,6 4. 2 17.6 17.9 4.2 18. . 8 3. 2 18.4 1.8 0.4 1. ,7 0. 4 1.7 - 107a -APPENDIX C SOURCES OF SAMPLING ERROR - 108 -Interviewer effects. To test for possible interviewer effects a 3 X 3 (3 interview ers and 3 beaches) analysis of variance was conducted using two dependent variables: the percentage of respondents refusing to complete a survey and the percentage of partially completed surveys for any given inter viewer (percentages were normalized by an arc-sine transformation). The use of these variables was based on the argument that any negative effects due to interviewer-respondent interactions would be reflected in the proportion of respondents who were unwilling to fill out a survey or not complete it once they had accepted the proposal. Based upon these two variables, effects due to interviewer were not significant (survey unfinished, F probability > .27; survey pro posal rejected, F probability >.32). Although this test may not have covered all sources of interviewer bias, it was considered sufficient for the purposes of the present research. Sampling procedure. The second test of sample bias utilized discriminant analysis and was concerned with determining .whether the sample taken from the beach population was typical. For this test four spatial and group var iables were compared for those individuals receiving surveys (n = 266) versus the total population sampled whether surveyed or not (n =1791). The variables employed were: 1) total number of people in the group, 2) the area of a group as evidenced by personal markers, 3) centroid to centroid nearest neighbor distance, and 4) nearest approach nearest neighbor distance. The results of the analysis for the two conditions, 'surveyed' and 'not surveyed*, indicated a significant difference between means (p < .01). Comparison of the means for surveyed and unsurveyed users for the three beaches indicated that surveyed subjects sampled on English Bay and Skaha were from larger groups with larger areas and res pondents at English Bay tended to be on the average more distant from their nearest neighbors. The mean values for these variables and condi tions are listed in Table 16. - 109 -Table 16. Comparison of means for the three sites for the two conditions: surveyed by questionnaire (s), and surveyed +< not surveyed (ns) . E = English Bay, K = Kitsilano, Sk = Skaha. Areas are in meters and distances are in meters. E/ns E/S K/ns K/s Sk/ns Sk/s Group size 1.63 2.48 1.81 1.92 2.08 2.33 Group area 2.95 3.64 3.81 4.10 4.43 4.89 cc/nnd 7.00 10.04 5.63 6.52 5.42 5.56 na/nnd 4. 97 7.92 3.40 4.36 3.05 3.11 The results which indicate larger mean group sizes and areas for English Bay and Skaha survey respondents can be explained by the fact that group members at these two sites were often observed to be swimming, strolling, going to the refreshment stand, etc. and were thus not visible in the aerial photographs. Since a respondent indicated on the survey booklet the number of individuals in the group whether immediately pre sent or not, these data would obviously be different from those obtained from the photographs. Kitsilano did not show this discrepancy as much as the other two sites since it was a much more compact area and most people tended to stay at their site. This may have been due to the fact that the area was grass covered and was not as hot as the other beaches which were sandy. The second difference between surveyed and unsurveyed data men tioned previously was that surveyed respondents at English Bay tended to have greater nearest neighbor distances than the average. Thus respondents with greater than average nnd's were over-represented in the sample. These results are not explained easily since three different interviewers distribtited surveys on that beach and thus it seems unlikely that inter viewer bias was the source. Since no explanation was apparent the data for nearest neighbor distance for English Bay was not used in the regres sion analyses and thus the results for these two distance measures are based on data from Skaha and Kitsilano only. - 109a -APPENDIX D REPRESENTATIVE DENSITY AND PATTERN CONFIGURATIONS Figure 17. (Kitsilano -- 110 -Random pattern characterictic of density = 66 groups/hectare). low densities. - Ill -Figure 19. Regular pattern characteristic of high densities (Kitsilano - density - 264 groups/hectare). 


Citation Scheme:


Usage Statistics

Country Views Downloads
France 9 0
United States 4 0
Japan 3 0
China 2 22
Germany 1 17
City Views Downloads
Unknown 9 17
Tokyo 3 0
Ashburn 2 0
Beijing 2 0
Martinsried 1 0
Buffalo 1 0
Seattle 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items