Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The use of newcomers' experiences in the urban planning process Gallins, Myra Berk 1971

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

Item Metadata

Download

Media
831-UBC_1971_A8 G34.pdf [ 6.22MB ]
Metadata
JSON: 831-1.0102201.json
JSON-LD: 831-1.0102201-ld.json
RDF/XML (Pretty): 831-1.0102201-rdf.xml
RDF/JSON: 831-1.0102201-rdf.json
Turtle: 831-1.0102201-turtle.txt
N-Triples: 831-1.0102201-rdf-ntriples.txt
Original Record: 831-1.0102201-source.json
Full Text
831-1.0102201-fulltext.txt
Citation
831-1.0102201.ris

Full Text

THE USE OF NEWCOMERS' EXPERIENCES IN THE URBAN PLANNING PROCESS by MYRA BERK GALLINS B.A., Univers i ty of Wisconsin, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the School of Community and Regional Planning We accept th is thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA A p r i l , 1971 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 i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my 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. . . ; School of Community and Regional Planning The University of British Columbia Vancouver 8, Canada Date flffciL. 3 k \<\l\ ABSTRACT The purpose of th is study was to determine whether newcomers to a large urban area, could be of special assistance to the planner. In order to discover whether newcomers could be of special use in the p lan-ning process, i t was necessary to determine i f newcomers' opinions of the c i t y d i f fered from those of longterm residents. For i f they d id ,^ then the planner might be able to derive from the newcomers fresh i n -sight into the planning problems posed by his c i t y . t Hence three hypotheses were formulated: I. Newer res idents ' opinions regarding the qua l i ty of the i r neigh-i bourhood's services and f a c i l i t i e s vary s i g n i f i c a n t l y from those of longterm residents. I I . Newer res idents ' community par t i c ipa t ion varies s i g n i f i c a n t l y from that of longterm residents. I I I . Newer res idents ' views on the qua l i ty of neighbourhood services and f a c i l i t i e s and the extent of the i r community par t i c ipa t ion vary s i g n i f i c a n t l y according to the s ize and locat ion of the i r former residence. . These hypotheses were tested by means of a questionnaire o r a l l y administered to a random sample of 108 residents of the K i t s i l ano area of Vancouver. On the bas i s o f the answers o b t a i n e d , each of the hypoth -eses was submit ted to the s t a t i s t i c a l s c r u t i n y o f m u l t i p l e d i s c r i m i n a n t a n a l y s i s , m u l t i p l e r eg ress i on a n a l y s i s , and percentage compar isons. Newer r e s i den t s were found to have d i f f e r e n t op in ions from long-term r e s iden t s on the q u a l i t y o f t h e i r neighbourhood's s e r v i c e s and f a c i l i t i e s and to p a r t i c i p a t e l e s s ; i n community a c t i v i t i e s . W i th in the newcomer ranks former l o c a t i o n was,an important d i s t i n g u i s h i n g v a r i a b l e . Thus a l l three hypotheses^ were subs t an t i a t ed by the r e s u l t s o f t h i s study as we l l as by some e a r l i e r research f i n d i n g s . Hence i t was concluded tha t res idence length d i f f e r e n c e s a l o n e , were s i g n i f i c a n t enough to dev ise some way o f c o n s u l t i n g newcomersj as i one a id to the urban p lann ing p rocess . The p lanner can p o t e n t i a l l y make use of the newcomers' prev ious exper ience and unencumbered percept ions to ga in f r e sh i n s i g h t i n t o the p lann ing problems posed by h i s c i t y : Page LIST OF TABLES v i i Chapter I. INTRODUCTION . 1 Hypotheses 2 L i terature Review . 4 Summary 10 I I . PROCEDURAL METHODS USED IN THIS STUDY 13 Operat ional izing the Hypothesis .13 Choice of Test Area 14 Boundaries of Test Area 15 Questionnaire Design . ; 16 Sample Selection 19 Administration of the Study . 20 Summary . 24 I I I . CHARACTERISTICS OF PEOPLE INTERVIEWED 26 Personal Character is t ics of the 108 People Interviewed , 26 Personal Character is t ics of Newcomers , 37 Factor Analysis 47 C h a p t e r p a g e Summary 56 IV. RESULTS: HYPOTHESIS I 60 D e s c r i p t i o n o f R e g r e s s i o n A n a l y s i s and D i s c r i m i n a n t A n a l y s i s -60 T e s t s o f t h e F i r s t H y p o t h e s i s by D i s c r i m i n a n t A n a l y s i s 64 T e s t s o f the F i r s t H y p o t h e s i s by R e g r e s s i o n A n a l y s i s 66 T e s t s o f the F i r s t H y p o t h e s i s by Pe r c e n t a g e Comparisons ' 69 Summary 74 V. RESULTS: HYPOTHESIS II . 77 T e s t s o f the Second H y p o t h e s i s by D i s c r i m i n a n t A n a l y s i s 77 T e s t s o f the Second H y p o t h e s i s by: R e g r e s s i o n A n a l y s i s 79 T e s t s o f the Second H y p o t h e s i s by P e r c e n t a g e Comparisons 79 Summary 82 i VI. RESULTS: HYPOTHESIS I I I . 86 T e s t s o f the T h i r d H y p o t h e s i s by D i s c r i m i n a n t A n a l y s i s 86 T e s t s o f the T h i r d H y p o t h e s i s by R e g r e s s i o n A n a l y s i s 89 T e s t s o f the T h i r d H y p o t h e s i s by P e r c e n t a g e and Average Comparisons 91 Summary. 99 Chapter Page VII. SUMMARY AND CONCLUSIONS 103 Summary 103 Suggestions for Further Research 108 Implications for Planning 109 BIBLIOGRAPHY . . . . . . 1 113 1 APPENDICES . . . . . 115 A. Questionnaire 117 B. Letter of No t i f i ca t i on Concerning Interview 126 C. Census Comparison to Determine the Representa-tiveness of th is Study's Sample 128 D. Regression Analysis Results 131 E. Membership Type Breakdown by Residence Length 136 F. Lessons Learned in the Process of Interviewing 138 LIST OF TABLES Table Page I I . 1. Par t ic ipant Response 23 III."I. Educational Attainment 27 2. Occupation 28 3. Occupation (Condensed) 29 4. Income 30 5. Age 31 6. Mari ta l Status 32 7. Stages in Family L i f e Cycle . 33 8. Dwelling Location 35 9. Dwelling Type 35 10. Homeownership 36 11. Education Breakdown by Residence Length 38 12. Occupation Breakdown by Residence Length 38 13. Income Breakdown by Residence Length 39 14. Age Breakdown by Residence Length 39 15. Mari ta l Status Breakdown by Residence Length 40 16. Children Breakdown by Residence Length 40 17. L i f e Cycle Breakdown by Residence Length 42 18. Dwelling Location Breakdown by Residence Length 43 111.19. Dwelling Type Breakdown by Residence Length 44 20. Homeownership Breakdown by Residence Length 44 21. Newcomers' Former Place of Residence ,-46 22. Zero-Order Corre la t ion Matrix for 11 , Variables to be Factor Analyzed 49 23. Summary of Factors 50 24. Loadings for Factor One 51 25. Individuals with High Factor Scores on Factor One .51 26. Loadings for Factor Two 52 27. Individuals with High Factor Scores on Factor Two 53 28. Loadings for Factor Three 54 29. Individuals with High Factor Scores on Factor Three . . . . . . . 55 I V . 1 . F Ratio for Individual Variables 65 2. Number of Cases C l a s s i f i e d into Group 66 . 3. Per cent of Residence Group Not Providing Answers for Rank 71 4. Quality Breakdown by Residence Length 72 V . l . Number of Cases C la s s i f i ed into Group 78 VI . 1. F Ratio for Individual Variables 87 2. Number of Cases C l a s s i f i e d into Group 88 3. Regression Analysis Summary 90 ACKNOWLEDGEMENT I wish to express my special thanks to Dr. Nirmala devi Cherukupalle whose ad-vice was invaluable in preparing this thes is . Thanks are also due to Dr. Robert C o l l i e r for his helpful suggestions. And f i n a l l y , I am grateful to my husband Glenn for his patience, sympathy, and active encouragement. THE USE OF NEWCOMERS' EXPERIENCES IN THE URBAN PLANNING PROCESS INTRODUCTION Purpose The purpose of this study was to determine whether newcomers to a large urban area could be of special assistance to the planner. Spec-i f i c a l l y , could the experience of newcomers in other environments be drawn upon by the planner in formulating planning pol icy? In order to discover whether newcomers could be of special use in the planning process i t was necessary to determine i f newcomers' opinions of the c i t y differed from those of longterm residents . For i f they d i d , then the planner might be able to derive from the newcomers fresh ins ight into planning problems posed by his c i t y . Impetus for this Study Two factors led to the undertaking of th i s study. The f i r s t was the author's conviction that c i t i z e n pa r t i c ipa t ion should play an 1* important part in the planning process. This convict ion dictated that the subject matter of this study be one which attempted to discover a mode by which planners could use c i t i zens in the planning process. But which c i t i zens should be used? This was answered by the second factor Footnotes follow end of chapter. the fact that population trends revealed that in the future Canada would 2 be predominantly an urban society concentrated in a few large c i t i e s . One obvious resul t of th is increased urban concentration w i l l be that there would be many newcomers to the c i t i e s of Canada. Hence the author resolved to investigate whether newcomers as a special resource group; could be used for planning. In order to do th is three hypotheses were formulated. A. Hypotheses The three hypotheses that tested the usefulness of newcomers as an aid to urban planning were that: I . Newer residents 1 opinions regarding the qua l i ty of t he i r neighbourhood's services and f a c i l i t i e s vary s i g n i f i c a n t l y from those of longterm residents . I I . Newer residents ' community pa r t i c ipa t ion varies s i g n i f i c a n t l y from that of longterm residents . , I I I . Newer residents ' views on the qua l i ty of neighbourhood ser-vices and f a c i l i t i e s and the extent of t he i r community par-t i c i p a t i o n vary s i g n i f i c a n t l y according to the s ize and( f-locat ion of the i r former residence. To understand these hypotheses the term newer resident must be defined. This term has two aspects. The f i r s t aspect i s o r i g i n . For the purpose of th is study newer residents were defined as those people moving into the c i t y from places outside the c i t y . The s ize of t he i r former place of residence (or ig in) was broken down into c i t i e s of over 3 and under 100,000 in population. The locat ion of t he i r former place of residence was categorized as: (1) the same province as the test c i ( t y , : (2) a l l other places in Canada, and (3) outside of Canada. The se.cond aspect of the term newer resident i s length of residence. In this- study i t was f e l t that any arb i t rary number of years to define a newer resident would be meaningless. As w i l l be seen, the term was operat ionalized by placing length of residence of the, people used to test the hypotheses,; on a continuum. F i f t y per cent (those with the shortest residence length) were designated newer residents , and the remainder were considered, long-term residents. , <• I t should also be observed that Hypotheses I and III deal with opinions about the qua l i ty of neighbourhood as opposed to ci ty-wide ser-4 vices and f a c i l i t i e s . The reasonj that opinions about neighbourhoods^ were emphasized was because i t has been found in other studies that people can more readi ly a r t i cu la te the i r neighbourhood needs. For ex-ample, in a study conducted in theisouthern United States i t was found that people had a more c r i t i c a l a t t i tude toward t h e i r neighbourhood than 5 toward the c i t y as a whole. Before proceeding with the method by which these hypotheses were tested, i t may be useful to review what has been discovered about the ( effect of differences in residence length on the views of the neighbour-hood by other researchers. B. Li terature Review There were many possible avenues of discussion related to the hypotheses l a i d out above. However, at a l l times the author had to re-member that th is was to be a thesis in Community and Regional Planning, not in Psychology, Sociology, or Geography. Thus a l l discussion of 6 7 migration, perception, and at t i tude formation and change have been considered to be outside the scope of th is thes i s . There was also, tocbe no elaboration on the theory-behind or the costs and benefits of c i t i z e n pa r t i c i pa t i on . For the purposes of this study an assumption was made that some form of c i t i z e n pa r t i c ipa t ion in the planning process i s , v a l -uable, and that obtaining opinions from c i t i zens i s one form of c i t i z e n pa r t i c i pa t i on . . , -t Hence, the l i t e r a tu re to be referred to , though l imi ted in nature, was a l l concerned with topics d i r e c t l y related to the formal hypotheses of th is study. In format f i r s t two authors' reasons for obtaining the opinions of newcomers w i l l be set out, then l i t e r a tu r e pertaining to each of the three hypotheses in turn w i l l be mentioned. i ^ '•  i s Reasons for Obtaining Newcomers' Views D. W. Blackburn from his work with newcomers in Conway, Arkansas was able to state s ix different reasons for using the views of newcomers Q as an aid to community planning: 1. The newcomer sees faul ts that the native w i l l not see for the reason that the native has\ become immune to conditions andjprob-lems. 2. The newcomer will be more objective, less prejudiced, sentimental, and emotional in sizing up community conditions and problems than will the native. 3. The newcomer has standards with which to compare since he has lived in other places, observed other ways of doing things. He is in a position to know what has worked and failed in other places. 4. The consideration of the views of newcomer (sic) provides a way for introducing new ideas from progressive communities elsewhere. The newcomer brings a fresh, new outlook. 5. The consideration of the views of the people, such as the new-comer segment makes for better relations between the administra-tion of the community and the people. This is especially true for the newcomer-since i t makes him feel welcome, wanted, and accepted. 6. The newcomer is less influenced than is the native by other indi-viduals and groups, is not afraid to speak his views. Blackburn also pointed out two disadvantages, (1) that the newj-comer may not have observed and experienced the community long enough to be in position to make suggestions, and (2) that the natives may resent the giving of extra attention to views of the newcomers. However, he concluded that "the disadvantages in using the views of newcomers are. important but the advantages in using them as one aid in community plan-g ning more than offset the disadvantages." j In passing i t might be noted that Joseph Sonnenfeld suggests another reason for learning the views of nonnatives, i.e., to keep them in the community. He believes that "what attracts the nonnative to an area may, i f i t is realized, act to hold the nonnative to that area." 1 0 Hypothesis I Several people have conducted studies, which, in part, related^ satisfaction with the community to* length of residence. The results R. J . Crothers tested the sa t i s fac t ion of 3,500 households by asking "Do you l i k e your community?" He found that length of residence in the community was not markedly or consis tent ly correlated with s a t i s -fact ion with the community. Respondents who had l i v e d in the i r commun-i t i e s for more than 17 years or less than 5 years tended to have high indices of s a t i s f ac t i on . On the other hand, those communities with r e l a t i v e l y high average per capita incomes showed only a small increase in the i r Indices of Sat isfactor iness with increasing length of res idence .^ i Caplow's study of residents in Puerto Rico showed that famil ies who had l i v e d more than f ive years at t h e i r present addresses expressed much less intent ion to move than those who had l i v e d in one place for a shorter time. Caplow f e l t that i t was "en t i re ly conceivable that i f neighborhood d i s sa t i s f ac t ion were the r u l e , the patience of the average family might be gradually eroded in such a way as to produce a steady^ 12 increase of moving intentions with increasing tenure." However,,'he found that this was not the case with the people he s tudied. Instead neighbourhood evaluation rose with increasing tenure. i Gul i ck ' s work with residents in Durham and Greensboro, North Carolina also showed that the percentage of informants expressing high sa t i s fac t ion increased with lengthiof residence. Newcomers tended to be less s a t i s f i e d , generally and s p e c i f i c a l l y , than oldt imers. And the , 13 spec i f i c d i ssa t i s fac t ions recurred;in consistent ways. Robert Sommer's research attempted to investigate more of the i spec i f i c d i s sa t i s fac t ion differences between residence length groups, but in quite a different s e t t i ng , that of a mental hosp i t a l . Breaking down the patients by length of stay he found that long-stay patients (10 years or more) made fewer complaints than medium- or short-stay pat ients . The spec i f i c differences, e . g . , concerning crowding, have l i t t l e a p p l i c a b i l i t y elsewhere except for the fact of t h e i r existence.; Sommer f e l t that " i f meaningful information can be secured from people who have extreme d i f f i c u l t y in communicating and re la t ing to others, the results should be even more useful when one interviews factory, workers, l i b r a r y patrons and a i r l i n e passengers"^ (or residents of the same area of a c i t y ) . ; To conclude, the results of the studies discussed above were somewhat contradictory. However, the majority indicated that d i f f e r -ences in residence length led to marked differences in s a t i s f a c t i o n . , Hypothesis II A l l three studies that attempted to deal with aspects of the second hypothesis about residence length differences in community p a r t i -c ipat ion reached s i m i l a r conclusions. A l l three found that residence length and community involvement were d i r e c t l y re la ted . Terence Lee used Nh.Q., a measure of indiv idual involvement-.in the urban soc ia l /phys ica l mi l i eu to test the re la t ionsh ip . He found that a pos i t ive re la t ionship was shown when the scale was plot ted against years of residence, but only for the period above f ive years. He con-cluded that newcomers to a l o c a l i t y become involved, up to a given.low l e v e l , quite qu ick ly . Thereafter they remain more or less s t a t i c for about f ive years, after which the i r involvement begins to increase 15 s t ead i ly . Gu l i ck ' s work on newcomer enculturation examined both formal association membership and community involvement and found d i rec t r e l a -tionships with length of residence. In Durham and Greensboro those with no formal associations were most frequent among the most recent new-comers. And while membership in three or more associations occurred among people of a l l ages and lengths of residence, i t was most frequent among the migrants who had l i v e d in Greensboro 6 or more years , and j 1 C. second most frequent among the nat ives. The Gulick study also found that the "doers" about public concerns were very heavily concentrated among the people over 35 years who had l i v e d in Greensboro 6 or more i years. "In general, the natives in both age categories [were] somewhat less par t ic ipant than the newcomers of more than 6 years ' residence, ,: though de f in i t e ly more par t ic ipant than the more recent newcomers."^ The th i rd work dealing with residence length differences in,com-munity pa r t i c ipa t ion was Bas i l G. Zimmer's "Par t ic ipa t ion of Migran ts 3 in Urban Structures." He used the married male occupants of a random sample of dwelling units in a midwestern community with a population of 20,000. In tes t ing whether migrants to an urban centre become p a r t i c i -pants to the same extent, he control led for age, sex, education, and occupation to demonstrate the s ignif icance of migrant status i t s e l f . Zimmer found that membership in formal organizations tended to increase d i r e c t l y with length of time in the community wi th in age, occupational and educational categories. High status f a c i l i t a t e d p a r t i c i p a t i o n . How-ever, regardless of s tatus , i t took migrants 10 years or more to become 19 integrated in th is aspect of the organized structure of the community. Thus the three previous studies a l l indicate that there should be differences found between residence length groups, with newcomers p a r t i -cipat ing less than longterm residents. , > Hypothesis III Regarding the t h i rd hypothesis of differences within newcomer groups on the basis of the s ize and locat ion of t he i r former residences, one set of wri ters on urban Canada f e l t that s ize differences are disappear-ing . However, two recent studies conducted in the United States i n d i -cated that such differences s t i l l e x i s t . , 5 James and Robert Simmons in Urban Canada f e l t that rural-urban differences were rapidly disappearing as urban areas expanded, that many of the values and att i tudes of:the. young col lege- t ra ined farmer were indis t inguishable from those of his urban cousin. They stated unequi-vocal ly that now "the urban l i f e - s t y l e and urban mores reach and rule 20 a l l parts of the country." I f the Simmons are correc t , the th i rd r-hypothesis should be disproved by th is study. However, Zimmer in the a r t i c l e mentioned previously above found great differences to s t i l l be in existence. He found that no matter how long farm migrants had been in, thei community, they had a lower p a r t i c i -pation rate than did other types of migrants who had l i v e d in the com-21 munity the same length of time. . A l s o , urban migrants tended to enter Blackburn, too, found s ign i f i can t differences among newcomers from different areas. He had asked a group of 159 people l i v i n g in Conway, Arkansas between one and two years to rate 25 di f ferent se rv ices , programs, and a c t i v i t i e s on a scale from zero to 100. He discovered that ratings followed a def in i te pattern. Those newcomers from Arkansas-rural areas gave the highest ra t ings . Those from Arkansas-urban areas gave the intermediate ra t ings . And those who came from out-of-state ; 23 gave the lowest ra t ings . While the Simmons f e l t that urban-rural differences were d i sap r pearing, these l a s t mentioned studies seem to support the t h i rd hypothesis, i . e . , there should be differences showing up with regard to qua l i ty of services and community pa r t i c ipa t ion depending on the s ize and locat ion of the person's former place of residence. s , C. Summary Having set out the purpose for th is study and the three hypotheses to be tested, a review of the l i t e r a tu r e was then set fo r th . Chapter II w i l l go on to relate the methods decided upon to . t e s t the hypotheses propounded in th is chapter. FOOTNOTES Pros and cons of th is can be found in the bibliography compiled by the author in December, 1970. This bibliography included a complete analysis of c i t i z e n pa r t i c ipa t ion in everyday planning. 2 For example, Economic Council of Canada, Fourth Annual Review: The Canadian Economy From the 1960's to the 1970's (Ottawa: Queen's P r in t e r , 1967), p. 184. 3 100,000 is quite a common d iv id ing point between c i t y s i z e s , = e . g . , I b id . ! [ 4 This i s despite the fact that the hypotheses dealt with newer residents of the c i t y . The difference was of l i t t l e consequence to th is study. 5 Robert L . Wilson, " L i v a b i l i t y of the C i t y : At t i tude and Urban Development," in Urban Growth Dynamics in a_ Regional Cluster of C i t i e s , ed. by F. Stuart Chapin, J r . and Shi r ley F. Weiss (New York: John Wiley and Sons, 1962), pp. 371, 381-382. Migration is usually discussed in terms of rural-to-urban i n -stead of interurban and i s usual ly on the large scale instead of con-sider ing individual decis ions. ^Most perception discussions relate pr imar i ly to c i t y design : anyway. o "Using the Views of Newcomers as an Aid to Community Planning" (paper presented at the Annual Meeting of the Southwestern Socio logica l Assoc ia t ion , New Orleans, Louis iana , A p r i l 7-9, 1966), p. 134. 9 I b i d . Joseph Sonnenfeld, "Variable Values in Space Landscape: An Inquiry into the Nature of Environmental Necessi ty," Journal of Social  Issues, XXII (October, 1966), 78-80. ^ R . J . Crothers, "Factors Related to the Community Index of Sat-i s fac tor iness , " E k i s t i e s , XXX (August, 1970), p. 109. Theodore Caplow, Sheldon Stryker , and Samuel E. Wallace, The  Urban Ambiance (Totowa, N. J . : Bedminster, 1964), p. 198. 13 John Gu l i ck , Charles E. Bowerman, and Kurt W. Black, "Newcomer Enculturation in the C i t y : Att i tudes and P a r t i c i p a t i o n , " in Urban Growth Dynamics, pp. 324, 327. ^Robert Sommer, Personal Space: The Behavioral Basis of Design (Englewood C l i f f s , N. J . : P r e n t i c e - H a l l , I nc . , 1969), pp. 89-90. , 15 Terence Lee, "Urban Neighborhood as a Socio-Spat ial Schema," Human Rela t ions , XXI (August, 1968), p. 259. , _; 16 ' "Newcomer Encul tura t ion ," pp. 342-343. 1 7 I b i d . , p. 347. " - : } 18 In C i t i e s and Socie ty , ed. by Paul K. Hatt and Alber t J . Reiss , J r . (revised e d . ; New York: The Free Press, 1957), pp. 730-738. 1 9 1 b i d . , p. 733. O f ) (The Copp Clark Publishing Company, 1969), p. 6. 21 "Par t i c ipa t ion of Migrants in Urban Structures," in C i t i e s and  Society , p. 733. , • ; 22 23, I b i d . , p. 738. Using the Views of Newcomers," p. 132. PROCEDURAL METHODS USED IN THIS STUDY The purpose of th is chapter i s to discuss the methods used to design and carry out th is study. The solutions to problems w i l l be d i s -cussed in the order in which the problems were met. The problem areas included (A) how the hypotheses were operat ional ized, (B) how the area to be sampled from was se lected, (C) how the boundaries of the test area were defined, (D) how the questionnaire was designed, (E) how. the sample was selected, and (F) how the questionnaire was administered. , A. Operat ional izing the Hypotheses c I t was decided that the best way to test the hypotheses of th i s thesis was to carry out a p i l o t study. This study was to be based on some form of questionnaire. However, to y i e l d a higher percentage of responses, i t was necessary to administer the questionnaire o r a l l y . To d i r e c t l y test the hypotheses i t was decided that a cross section of a l l residents of one area of Metropolitan Vancouver be sampled. I t was f e l t that this would be preferable: to merely studying a cross section of newcomers throughout the c i t y . 1 B. Choice of Test Area In order to aid i n the se lect ion of the test area, a l i s t of f ive c r i t e r i a was compiled. The f i r s t c r i t e r i o n was that the area selected be part of the Ci ty of Vancouver, rather than one of the surrounding mun ic ipa l i t i e s . This i s because i t was postulated that newcomers to a metropolitan area w i l l s e t t l e f i r s t in the c i t y proper. The second c r i t e r i o n was that the area chosen have a high propor-t ion of i t s avai lable housing in the form of suites or apartments.: This is because i t was postulated that newcomers w i l l take up temporary r e s i -dence in a sui te before deciding to purchase a house. The next c r i t e r i o n was that the area to be selected be as hetero-geneous as poss ible . This would allow con t ro l l ing for such factors as d i f fe r ing age, income, and ethnic groups. The fourth c r i t e r i o n in choosing a test area was that the area have some overa l l s t a b i l i t y rather than a high turnover of the popula-t i o n . This constraint was imposedifor two reasons. The f i r s t was.;t05 insure that persons interviewed would have some degree of f a m i l i a r i t y with the area in which they were l i v i n g . The second reason for imposing this c r i t e r i o n was for ease in contacting dwelling units included in . the f i na l sample. Letters were sent out in advance to not i fy people of the impending interview. But in an area of high turnover too many of these l e t t e r s might have been received by people who w i l l have moved out by the time the interviewer v i s i t e d . , And the f i n a l c r i t e r i o n was that the area chosen have an estab-l ished sense of community i den t i t y . One ind ica t ion of th i s would be i f people already l i v i n g there knew the name of the i r community. Thereby an established community could be used to determine whether newcomers have iden t i f i ed with the area at least to the extent of learning i t s name.1 , ; The only area in Metropolitan Vancouver that seemed to f i t a,ll 2 the above c r i t e r i a for the test area was K i t s i l a n o . K i t s i l a n o i s part of the Ci ty of Vancouver with a large number of suites of a l l types included in i t s dwell ings. K i t s i l a n o ' s population is heterogeneous, and there is f a i r l y low population turnover. In add i t ion , there i s an established sense of community i den t i t y . ; C. Boundaries of Test Area There were three already established ways of defining the boun-. daries of K i t s i l a n o . The f i r s t was to use the boundaries determined by the Dominion Bureau of S t a t i s t i c s for the Census. The second was to use the e lec tora l d i s t r i c t boundaries. And f i n a l l y , the boundaries de-fined by the United Community Services of Vancouver could have been, used. However, the three sets of boundaries defined by these outside agencies were rejected because preliminary invest igat ion indicated that the < people in the community iden t i f i ed with a different de f in i t i on of t h e i r area. Hence, the following boundaries were substi tuted for those which might have been adopted: English Bay on the North, Sixteenth Avenue on. the South, Burrard Street on the East , and MacDonald Street on the West. D. Questionnaire Design Two aspects of questionnaire design w i l l be dealt with in th i s sec t ion. The f i r s t deals with what questions should be asked to y i e l d the required information.. The second deals with ways of most effec-t i v e l y asking these questions. Questions to be Asked i To keep the questionnaire as short as possible and yet be able to obtain enough information to use in test ing the hypotheses the questionnaire was divided into three sect ions . The f i r s t section was designed to obtain opinions on the level of qua l i ty of services and f a c i l i t i e s in the immediate v i c i n i t y of the home of the person interviewed., To obtain th is information a l i s t was compiled of eighteen types of services and f a c i l i t i e s to be ranked. The second section was designed to obtain information about^ pa r t i c ipa t ion in community a c t i v i t i e s . This involved awareness of the popular name of K i t s i l a n o , awareness of and pa r t i c ipa t ion in neighbour-hood organizat ions, attendance at government meetings, and involvement in organizations in general. The f ina l section of the questionnaire was designed to e l i c i t e ; information about the personal background of the people interviewed. ^ Along with the more usual questions used to derive socioeconomic status and stage in family l i f e cyc le , i t was necessary to obtain information about length of residence in Vancouver. In add i t ion , newcomers to Vancouver were required to answer questions revealing t h e i r former place of residence and the i r motives for moving to Vancouver. (The com-plete questionnaire i s Appendix A ) . ; , i Administering the Questions During the administration of the questionnaire, respondents;were required to ac t ive ly par t ic ipa te two times. I t was hoped by having the respondents par t ic ipa te that they would be more interested in the i n t e r -view. With regard to the ranking of services and f a c i l i t i e s , people . interviewed were asked to pick from note cards. I t was f e l t that th is would be better than merely reading off 18 names or giving them a typed l i s t of 18 items from which to choose. A sample of the 18 note cards used i s on the next page. The 18 cards were placed, in alphabetical order, in columns of s ix e i ther on a nearby table or on the f l o o r . ; The respondent was asked to choose the . f ive things he f e l t were the biggest problem in his area and the f ive things he f e l t were the least problem, and place them in two separate p i l e s . In this way those services and f a c i l i t i e s he was merely neutral about were el iminated. The person was then forced to place the items selected in t he i r r e l a t ive pos i t ions . ;. The positions were best to worst in the category of least problems-and worst to best in the category of biggest problems. I t was assumed that F I R E PROTECTION P A R K S POLICE TmWmDWfflM features to be evaluated, and would also be more in te res t ing . The other time that people interviewed were required to ac t i ve ly par t ic ipa te during the administration of the questionnaire was in the section dealing with personal cha rac te r i s t i c s . People interviewed;were handed a card showing different income groups and were asked to choose the l e t t e r of the group the i r family belonged to on the basis of t h e i r average annual income. This device of t e l l i n g income by l e t t e r rather than by do l l a r value was also used to avoid what might have been seen; 3 as a personal threat by the respondent. The remainder of the questions required the respondents to p a r t i -cipate to a lesser degree. However, their , task s t i l l required some -effor t on the i r part in that they were ca l led upon to answer open-ended questions. For example, instead of asking "Are you aware of the e x i s -tence of one of the three major neighbourhood organizations: Kitsijlano Ratepayers, K i t s i l a n o Area Resources Counc i l , or K i t s i l a n o Community , Centre?" the respondents were merely asked "Is there any neighbourhood organization?" ! . E. Sample Select ion In order to pick a sample of respondents, a l i s t i n g was made of the one hundred and f i f t y blocks wi th in the area defined previously as K i t s i l a n o . Since a sample made up of newcomers and longer term residents was desired for comparative purposes, i t was decided that areas other than merely those zoned for apartments should be included. For that reason the sample of 17 blocks consisted of 50 per cent blocks randomly chosen from those zoned for mult iple family use, 20 per cent blocks randomly chosen from those zoned for two family use, and 30 per cent blocks randomly chosen from those zoned for s ingle family use only . These blocks are shown on the map on the next page. i The next step was to make a complete l i s t i n g of a l l dwelling units on each of these chosen blocks. For th i s purpose the 1970 } Vancouver Ci ty Directory was used. ; Although i t s usefulness may be ques-tioned when a sample of names is des i red, the Ci ty Directory is extremely useful when accuracy is l imi ted to a correct l i s t i n g of a l l addresses and unit numbers only. Hence, i t was r e l a t i v e l y simple to choose a random sample of 10 household units from each of the 17 blocks. This , gave a to ta l possible sample s ize of 170 households. . F. Administration of the Study Most of the interviews were iconducted.between 19 December 1970 and 18 January 1971 by three different interviewers, of which the author completed 82.4 per cent. In what follows a number of the d i f f i c u l t i e s met in administering the questionnaire w i l l be presented. Among these d i f f i c u l t i e s were what consti tuted an i n e l i g i b l e un i t , what to do i f people to be in t e r -viewed were not ava i l ab le , and what caused people to refuse to be interviewed. uu. uu O i l £i]zzzi nrzi LZZZZJ c 5rfaaaczlJ rzzzzj g|aBrlE3( E Z X I ^4 ZJ LZ ZH Z3 i : . ::iinr.:;_:. toizzzj czi ixczzj rzzi — —<r.-r 1—11 L Z Z Z Z L Z J L Z B R O A 0 w A Y zzz LZZZZJ c z i orzzzi CZZZZJ c :irz,c-g-ZJL" r z i z z i n z ^ c z i i e g r z z z z j c ! : LZZZ):. Z J C ZD r ZJ _ 1 c 3C z i i z z z i L z z z 3 z z z z a . L Z z z z j x IT [ZJzzFTzzzz Z Z Z Z J c z z r c 1 J 1 II I I1! L 1 i • ! ! i 1 1 i ^ C Z 3 T -r z cz :ZJ c z z z i LZZZZJ LZZZZJ LZZZZJ nrzzz] LZ z i i z z z z h i z z z z LZZZZJ" ZZIZZZJJZZZZZD LZ HZZZZDiCZZZ^rz ZZJH xzzza, rf : : j T z z z z n z z z z : c ZTLZ ZZJ,C;Z:J rzzi r zzfizzrzzizzzz I-ZZ • L ZZTZ z Tr:z:zrz:'zir • L Z Z Z Z J : ZJLZZ T7T ZJ-C • LZZZLJ i) r ZJ ZeC 3 C it 3 * ZJ 'i.r SIXTEENTH AVENUE BLOCKS INCLUDED IN SAMPLE i n n n a ZJ 1 inch = 1000 feet D Single Family ED 2-Family H I Multiple Family Occasionally a dwelling unit l i s t e d in the sample had to be con-sidered i n e l i g i b l e . There were several possible reasons for t h i s . Two of the most important included units vacant at the time an interview was attempted and units that did not consti tute a residence. In cases where there was no:one at home at a dwelling unit i n - s eluded in the sample an attempt was made to return at a di f ferent time of day or , i f necessary, on a weekend afternoon. Whenever a potential respondent expressed in teres t but could not see the interviewer at that time, an appointment was made for another time that would be more con^ venient for the respondent. However, the number of return v i s i t s was • l imi ted to two per dwelling un i t . A pretest was carr ied out the weekend of the 21st of November 1970, with no advance warning given to potential respondents. Since a number of negative responses were received, the bulk of the interviews were then conducted only after a l e t t e r of n o t i f i c a t i o n had been sent (see Appendix B) . The p r io r warning seemed to make potential respon-dents more l i k e to cooperate w i l l i n g l y . Some people did refuse, how-, ever. Reasons included not wanting to be bothered, a feel ing that they had nothing to contr ibute, and d i s t rus t of the interviewer, despite receipt of the not i fying l e t t e r . For a complete breakdown of par t ic ipant response the fol lowing table should be consulted. TABLE II.1 PARTICIPANT RESPONSE OF 170 DWELLING UNITS CONTACTED: 108 were interviewed 63.5% 26 refused to be interviewed 15.3% 18 women 7 men 1 couple i 21 not at home after two callbacks 12.4% 15 i n e l i g i b l e units 8.8% 5 vacant 4 unfamiliar with the c i t y 3 out of town for a prolonged period 2 not residences 1 no one English speaking 100.0% G. Summary In th is chapter i t was decided that a p i l o t study be car r ied out to tes t the hypotheses of th is thes i s . The c r i t e r i a used to select the test area were then set fo r th . This was followed by a descr ipt ion of the boundaries of K i t s i l a n o , the test area selected. Next two aspects of questionnaire design were presented. These were what questions; should be asked to y i e l d the required information and how these questions could most e f fec t ive ly be asked. Having described how the questionnaire was designed, the method by which the sample was selected was set f o r t h . And f i n a l l y , some of the p rac t i ca l : aspects of the administration of this study were reported. '• s i Henry Sanoff i n "Social Perception of the Ecological Neighbor-hood" in E k i s t i c s , XXX (August, 1970), p. 132 suggested something s i m i -l a r . In his study of Raleigh, N.C. "When asked about the Ltname3 of t he i r neighborhood, the respondents tended to give the name of t he i r street whereas few tended to locate the i r section geographically with respect to the c i t y . In„neighborhoods that have previously been iden-t i f i e d by a name, this strategy would serve as a measure of the subjec-t ive perceptions of indiv iduals as they do or do not correspond to the public image of t he i r neighborhood." \ 2 The other areas of Vancouver that met most of the c r i t e r i a were Marpole, Ker r i sda le , and the West End. However, these did not f i t the c r i t e r i a for the following reasons: Kerrisdale was not as hetero-geneous at that moment; the West End appeared to have an above average rate of turnover; and Marpole did not seem to r e a l l y have an estab-l ished community i d e n t i t y . 3 Charles Herbert Backstrom and Gerald D. Hursh, Survey Research (Chicago: Northwestern Univers i ty Press, 1963), p. 106. CHARACTERISTICS OF PEOPLE INTERVIEWED The purpose of th is chapter i s to present, from three different perspectives, a p ro f i l e of the charac ter i s t ics of the 108 people i n t e r -viewed for this study. In the f i r s t section of th is chapter an over-view of the personal charac te r i s t ics of the people interviewed w i l l be presented. Having discussed the charac ter i s t ics of a l l the people! i n ; the sample, the second section w i l l focus on the personal charac te r i s t i cs of the newer residents only. In the f i n a l sec t ion , the factor analysis carr ied out on the personal charac te r i s t ics of the entire sample of 108 people w i l l be described. A. Personal Character is t ics of the 108 People Interviewed The personal charac ter i s t ics of the people interviewed can be des-cribed under three basic categories, socioeconomic s ta tus , stage in family l i f e cyc le , and dwelling unit cha rac t e r i s t i c s . These w i l l be ex-amined separately below. Socioeconomic Status Socioeconomic status is commonly based upon three in ter re la ted: var iab les , education, occupation, and income. With regard to education, the sample as a whole was f a i r l y well educated (see Table 1) . The mean and median level of education was high school graduation. Hence, above the median, almost 30 per cent of the sample had attended un ive r s i t y , while 10 per cent had graduated from un ive r s i ty , and another 10 per icent had completed some graduate or pro-fessional t r a in ing beyond univers i ty graduation. Below the median, only 5 people out of 108 ( i . e . , 4.6 per cent) had never reached high school , while 25 people ( i . e . , 23.2 per cent) attended high school but did-not graduate. -TABLE II1.1 EDUCATIONAL ATTAINMENT Education Level Number Per cent Some Elementary (0-7 years) 1 0.93 Elementary Graduation (8 years) 4 3.70 Some High School (9-11 years) 25 23.15 High School Graduation (12 years) 26 24.07 Some Univers i ty (13-15 years) 30 27.78 Univers i ty Graduation (16 years) 11 10.19 Graduate or Professional Training (16+ years) 11 10.19 108 100.00 The second component of socioeconomic status is occupation. From the results of th is sample, as reported in Table 2, i t i s obvious that Ki t s i l ano i s not a "working class" neighbourhood. Rather, 20.4 per cent of the respondents f e l l into the highest occupation category, that of professional and technical workers, while 12.0 per cent of the sample were managers, o f f i c i a l s , and propr ie tors . Thirteen point zero per cent were c l e r i c a l workers. Sales workers comprised 7.4 per cent of the people interviewed, while 11.1 per cent were craftsmen and foremen. Four point s ix per cent were operatives, and 7.4 per cent were service workers. Merely 1.8 per cent were laborers , while 11.1 per cent were students and 6.5 per cent were r e t i r e d . TABLE I I I . 2 OCCUPATION Occupation Category Number Per cent Professional and Technical 22 20.37 Managers, O f f i c i a l s , and Proprietors 13 12.04 C l e r i c a l Workers 14 12.96 Sales Workers 8 7.41 Craftsmen and Foremen 12 11.11 Operatives 5 4.63 Service Workers 8 7.41 Laborers 2 1 .85 Students 12 11.11 Retired 7 6.48 Other 5 4.63 108 100.00 Based on these percentages i t i s c lear that the white co l l a r -b lue c o l l a r dichotomy does not apply to th is data. Therefore a breakdown of occupa-tions was made wi th in the white c o l l a r category. High status white c o l l a r occupations included professional and technical workers and man-agers, o f f i c i a l s , and propr ie tors ; while low status white c o l l a r occupa-tions included sa les , c l e r i c a l a n d service workers. The following condensed table (Table 3) w i l l make clearer the d i s t r i bu t i on of occupa-tions described above. TABLE 111.3 OCCUPATION (Condensed) Occupation Category Number Per cent High Status White Co l l a r 35 33.97 Low Status White Co l l a r 30 29.13 Blue Co l l a r 19 18.45 Students 12 11.65 Reti red 7 6.80 103 100.00 The th i rd component of socioeconomic status is income. Within the sample the most frequently chosen response category was that of people earning $7,000 to $9,999 a year ( i . e . , 27 per cent of those res-2 ponding ) . However, the average yearly income for a household was $5,000 to $6,999. And at the extremes, 16 per cent of the people who re-vealed the i r incomes claimed to have earned less than $3,0Q0 a year, while 10 per cent claimed incomes of more than $15,000 a year. For f u l l de ta i ls of the responses to th is question reference should be made to Table 4. TABLE 111.4 INCOME Yearly Income Number Per cent Refused 6 5.56 Under $3,000 17 15.74 $3,000 to $4,999 17 15.74 $5,000 to $6,999 17 15.74 $7,000 to $9,999 28 25.93 $10,000 to $14,999 13 12.04 $15,000 and Above 10 |9.26 108 100.00 Having examined the three components of socioeconomic s ta tus , ed-ucat ion, occupation, and income, i t can be concluded on the basis of th is sample that the socioeconomic status of K i t s i l a n o i s somewhat heter-ogeneous within the middle class l e v e l s , with some lower middle class and some upper middle class households in evidence. Family L i f e Cycle The concept of family l i f e cycle i s one which is useful in reveal-ing the personal charac ter i s t ics of the people interviewed in this study. To determine an i n d i v i d u a l ' s stage in family l i f e cycle four components are used. These include age, marital s ta tus , presence of chi ldren in the household, and the age of the youngest of these ch i ld ren . Each of these w i l l be examined separately, and then the composite stage in family l i f e cycle w i l l be determined for each person interviewed for : this study. With regard to age, Table 5,reveals that younger age groups were overrepresented. In fac t , exactly 50 per cent of a l l respondents were under 30 years of age. The middle age groups are almost equally repre-sented. That i s , approximately 14 per cent of the sample were 31-40, 15 per cent were 41-50, and 10 per cent were 51-60. Those over 60 years of age comprised approximately 11 per cent of the sample. The mean age of the ent i re sample was 31-40. TABLE I I I .5 AGE Age Level Number Per cent 18-24 31 28.70 25-30 23 21.30 31-40 15 13.89 41-50 16 14.81 51-60 11 10.19 61-65 4 3.70 65+ 8 •7.41 108 100.00 The second component of family l i f e cycle i s mari tal s tatus . As was expected, the majority of cases f e l l into the two main categories. Thi r ty-e ight per cent were s i ng l e , and 47 per cent were married. The remaining 15 per cent were c l a s s i f i e d between married and s i n g l e , i . e . , as being married at one time but not considering themselves married at present. The precise breakdown can be seen in Table 6. TABLE 111.6 MARITAL STATUS Mari ta l Status Number Per cent Single 41 37.96 Widowed 7 6.48 Divorced 4 3.70 Separated 5 4.63 Married 51 47.22 108 100.00 The next component of family l i f e cycle i s presence of ch i ld ren . Within the sample 29 households reported that they had children l i v i n g at home. Of these the majority had only one c h i l d ( i . e . , 48 per cent) . Twenty-eight per cent had 2 ch i ld ren , 14 per cent had 3 ch i ld ren , and 10 per cent had more than 3. Thus the average number of chi ldren per household for the ent i re sample was 0.52. The f i na l component necessary to derive the composite stage in family l i f e cycle i s the age of the youngest c h i l d l i v i n g at home. Of the 29 households with ch i ld ren , the youngest c h i l d in 7 of those house-holds was under the age of 6. For 12 households the youngest c h i l d was between 6 and 14. And in the remaining 10 households the youngest c h i l d was between 14 and 18 years of age. Having presented the data on age, marital s ta tus , number o f - c h i l -dren, and age of ch i ld ren , the stages in family l i f e cycle for each per-3 son interviewed can now be set forth in Table 7. - TABLE I I I .7 STAGES IN FAMILY LIFE CYCLE Li fe Cycle Category Young s ingle Young married, no children Young married, youngest ch i l d under 6 years of age Young married, youngest c h i l d 6 or older Older married, with chi ldren Older married, without children l i v i n g at home Older s ingle Others (1 parent only, plus children under the age of 18) Number 38 18 5 4 12 12 12 Per cent 35.19 16.67 4.63 3.70 11.11 11.11 11.11 6.43 108 100.00 This table indicates that the largest s ingle category of people by stage in family l i f e cycle was composed of those who are young and single ( i . e . , 35 per cent of the sample). The next most frequent cate-gory was young marrieds without chi ldren ( i . e . , 17 per cent) . Following the young marrieds without ch i ld ren , three categories contained the same number of i nd iv idua l s . These three categories included older s ingle people and older married people, both with and without ch i ld ren . The. least commonly occurring household compositions are young married couple with children of any age and the more unusual combinations with only one parent. Dwelling Character is t ics The f ina l factor which w i l l be used to reveal personal character-i s t i c s of the sample is that of the locat ion and type of dwelling in which the persons interviewed l i v e d . With regard to l oca t ion , 75 per cent of the people interviewed l i ved West of Arbutus Street . On a North-South axis, of the four possibl loca t ions , most of the homes of persons interviewed were in the area located between Cornwall and Fourth Avenue (43.5 per cent) . Complete de 4 t a i l s of dwelling locat ion can be seen in Table 8. > Now exactly what types of dwellings did the people interviewed l i v e in? Forty-three of the 108 people interviewed ( i . e . , 39.8 per cent) l i v e d in low-rise apartment blocks. The second most frequent type of dwelling in which respondents l i ved was the s ingle family de-tached house ( i . e . , 32.4 per cent) . The complete breakdown of dwelling type i s presented in Table 9. The f ina l personal charac te r i s t i c of people interviewed that has DWELLING LOCATION East-West Axis East of Arbutus Street West of Arbutus Street North-South Axis North of Cornwall Between Cornwall and Fourth Avenue Between Fourth Avenue and Broadway Between Broadway and Sixteenth Avenue Number 27 81 108 Number 6 47 20 35 108 Per cent 25.00 75.00 100.00 Per cent 5.56 43.52 18.52 32.41 100.00 TABLE I I I .9 DWELLING TYPE Type of Dwelling High-rise apartment Low-rise apartment Suite in a house Duplex Single family detached house Number 4 43 17 9 35 Per cent 3.70 39.81 15.74 8.33 32.41 to do with dwellings i s the extent of homeownership. Seventy-four per cent of the sample were tenants, while the other 26 per cent of the people interviewed l i v e d in dwellings that they owned. Breaking down the sample by type of dwel l ing , 70 per cent of the people interviewed l i v i n g in s ingle family detached houses owned the i r own home. Eighteen per cent of people l i v i n g in a s ingle family house divided into several suites owned the ent i re house. Only 11 per cent of duplex residents owned the i r dwel l ing. And 5 per cent of the people interviewed l i v i n g in low-rise apartments owned the apartment blocks in which they were . l i v i n g . Table 10 summarizes this information on owneroccupancy. , TABLE III .10 HOMEOWNERSHIP Number of Per cent Type of Dwelling Tenants Owners Owner-occupied High-rise apartment 4 0 0.00 Low-rise apartment 41 2 4.65 Suite in a house 14 3 17.65 Duplex 8 1 11.11 Single family house 10 23 69.69 77 29 26.85 (Average) This concludes the discussion of the personal charac ter i s t ics of the 108 people interviewed. The question which must be answered now i s : what are the personal charac ter i s t ics of an important segment of the sample? That i s , what are the charac ter i s t ics of newcomers to Vancouver? B. Personal Character is t ics of Newcomers This section w i l l consist of two parts . The f i r s t w i l l be a b r i e f summary of the information of Section A of th is chapter broken down by different lengths, of residence. The other part of th i s section w i l l deal exc lus ive ly with the two newer groups of residents , and w i l l be a comparison and contrast in terms of the i r former places of residence and the i r reasons for moving to Vancouver. Personal Character is t ics I t was decided that any arb i t rary time span to define "newcomers" would be meaningless. Instead the sample has been divided into approxi-mate qua r t i l e s , so that 27 respondents have l i ved in Vancouver for 0-2 years, 27 for 2.5-6 years , 29 for 7-21 years , and 25 for 22-60 years. The most important personal charac ter i s t ics of newcomers, as de-rived from Tables 11-20, are as fol lows: Tables 11-13 indicate that on the basis of socioeconomic va r iab les , newer residents tend to be better educated, in white c o l l a r occupations and have medium to high incomes. Table 14 shows that newer residents c lus ter in the younger age groups. And while the majority have no ch i ld ren , in Table 16 i t can be seen that newer residents are the ones who tend toward large famil ies of 3 or more ch i ld ren . EDUCATION BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver School Level 0--2 2.5-6 7--21 22--60 Total Number Some Elementary - 100 .00 1 Elementary Grad * 25.00 50 .00 25 .00 4 Some High School 24, .00 12.00 16 .00 48 .00 25 High School Grad ' 30, .77 30.77 26 .92 11 .54 26 Some Univers i ty 26, .67 26.67 30 .00 16 .67 30 Univers i ty Grad 9 .09 36.36 27 .27 27 .27 11 Graduate or Professional Training 36, .36 27.27 36 .36 11 108 i TABLE II I .12 OCCUPATION BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver Occupation Category 0--2 2.5-6 7--21 22--60 Total Number High White Co l l a r 18 .36 31.99 19 .75 29 .90 35 Low White Co l l a r 43 .45 13.69 30, .36 12 .50 30 Blue Co l l a r 22 .23 15.00 28, .33 34 .44 19 Student 33 .33 41. .67 16 .67 8 .33 12 Retired 14 .29 28 .57 57 .14 7 INCOME BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver Total Yearly Income 0-2 2.5-6 7-21 22-60 Number Refused - - 66.67 33.33 6 Under $3,000 11.76 35.29 29.41 23.53 17 $3,000 to $4,999 52.94 23.53 11.76 11.76 17 $5,000 to $6,999 ' 17-65 35.29 29.41 17.65 17 $7,000 to $9,999 25.00 21.43 17.86 35.71 28 $10,000 to $15,000 23.08 30.77 23.08 23.08 13 $15,000 + 30.00 10.00 50.00 10.00 10 108 Table III .14 AGE BREAKDOWN BY RESIDENCE LENGTH j Years of Residence in Vancouver Total Age Level 0-2 2.5-6 7-21 22-60 Number 18-24 38.71 32.26 19.35 9.68 31 25-30 34.78 34.78 26.09 4.35 23 31-40 13.33 40.00 40.00 6.67 15 41-50 18.75 12.50 25.00 43.75 16 51-60 9.09 - 27.27 63.64 11 61-65 25.00 - - 75.00 4 65 + - 12.50 50.00 37.50 8 MARITAL STATUS BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver Mari tal Status 0-2 2.5-6 7-21 22-60 Total Number Single 23.08 35.90 25.64 15.38 39 Wi dowed 14.29 14.29 28.. 57 42.86 7 Divorced 25.00 - 75.00 - 4 Separated '20.00 40.00 20.00 20.00 • 5 Married 25.49 TABLE 19.61 III .16 25.49 29.41 51 108 CHILDREN BREAKDOWN BY RESIDENCE LENGTH Number of Children Years '< 0-2 , of Residence in Vancouver 2.5-6 7-21 22-60 Total Number None 24.05 26.58 29.11 20.25 79 One (1) 28.57 7.14 14.29 50.00 14 Two (2) 37.50 25.00 25.00 12.50 8 Three or More (3+) 14.28 42.86 28.57 14.28 1 108 TABLE I I I . 17 LIFE CYCLE BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver Total L i f e Cycle Category 0--2 | 2.E 3-6 7-21 22-60 Numb< Young Single 34, .21 36, .84 21.05 7.89 38 Young Married, No Children 27 .78 27, .78 38.89 5.56 18 Young Married, Children Under 6 40, .00 40, .00 - 20.00 5 Young Marr ied, Children Over 6 * 50 .00 25 .00 25.00 . - •4 Older Married, With ChiIdren 16 .67 16 .67 16.67 50.00 12 Older Married, No Children 16 .67 25.00 58.33 12 Older Single 16 .67 41.67 41.67 12 Other 14, .29 14 .29 42.86 28.57 7 108 In terms of dwelling unit va r iab les , there is very l i t t l e d i f f e r -ence between residence lengths on an East-West a x i s , while a difference does ex i s t on a North-South ax i s . Table 18 indicates that newer r e s i -dents tend to be located between Broadway and Cornwall , while older residents predominate between Sixteenth Avenue and Broadway. Table 19 c l ea r ly shows that the newer residents mostly l i v e in apartment blocks and duplexes, while more of the longer-term residents are found in suites in houses and in s ingle family detached dwell ings. And therefore i t follows that the newer residents are mostly tenants, while the older residents dominate the homeowners category (Table 20). Having summarized the personal charac ter i s t ics of the people i n -terviewed in terms of different residence lengths, the next part of th i s section w i l l set forth the or igins of newcomers and the i r reasons for s e t t l i ng in Vancouver. Former Places of Residence and Reasons for Moving A l l people interviewed who have l i v e d in Vancouver for less than f ive years were asked where they l ived immediately p r io r to moving to Vancouver. Forty-nine people answered th is question. Of these, 30 (or 61.2 per cent) came from c i t i e s with population over 100,000, while the remaining 19 were from smaller towns. The former place of residence was also c l a s s i f i e d according to i t s loca t ion . On this basis 20 respondents ( i . e . , 40.8 per cent) were from other provinces in Canada, 17 (34.7 per cent) were from places in B r i t i s h Columbia, and the remaining 12 i n d i -5 viduals (24.5 per cent) were from outside of Canada. TABLE III .18 DWELLING LOCATION BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver 0-2 2.5-6 7-21 22-60 East-West Axis i East of Arbutus 29.63 25.93 18.52 25.93 West of Arbutus 23.46 24.69 29.63 22.22 North-South Axis North of Cornwall 33, .33 33 .33 33 .33 Fourth Ave. to Cornwall 23 .40 31 .91 31, .91 12 .77 Broadway to Fourth Ave. 20 .00 35 .00 25 .00 20 .00 16th Ave. to Broadway 28 .57 8 .57 25 .71 37 .14 Total Number 27 81 108 6 47 20 35 108 TABLE I I I . 19 DWELLING TYPE BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver Type of Dwelling 0--2 2.5-6 7-21 22--60 Total Number High-rise Apartment 25 .00 50.00 25.00 4 Low-rise Apartment 30 .23 32.56 20.93 16 .28 43 Suite in a House 29 .41: 11 .76 47.06 11 .76 17 Duplex -22 .22 44.44 22.22 11 .11 • 9 Single Family House 17 .14 14.29 25.71 42 .86 35 108 TABLE 111.20 HOMEOWNERSHIP BREAKDOWN BY RESIDENCE LENGTH Years of Residence in Vancouver Total Occupant's Class 0-2 2.5-6 7-21 22-60 Number Tenant 32.47 29.87 24.68 12.99 77 Owner 6.90 10.34 31.03 51.72 29 106 The next thing to examine in connection with former place of residence i s the difference between the two classes of newcomers--those who have l i ved in Vancouver for less than two years and those who have l i v e d in Vancouver between 2.5 and 6 years. For convenience the former class of newcomers w i l l be referred to as the "newest" group, while those i n Vancouver 2.5-6 years w i l l be c l a s s i f i e d as the "newer" group of residents . I t can be seen from Table 21 that , on the basis of the sample of this study, the older wave of immigrants to Vancouver was mostly made up of people from smaller towns within the province of B r i t i s h Columbia., The newest wave of immigrants, on the other hand, was la rge ly from big c i t i e s across Canada, excluding B r i t i s h Columbia. Or, to phrase th i s ; d i f f e r en t l y , i t would seem that between 3 and 6 years ago Vancouver's growth was a t t r ibutable to a ruralrto-urban migration wi th in the pro-vince. More recent ly , however, migrants to Vancouver are of the i n t e r -urban or , to be more exact, intermetropolitan type. i One reservation should be set forth as to the veraci ty of th is conclusion. The general izat ion just made was based so le ly on new r e s i -dents s t i l l l i v i n g in the K i t s i l ano area. I t i s quite possible that a large group moved to Vancouver from metropolitan Canada between 2 and 6 years ago and e i ther did not se t t l e in K i t s i l a n o , or se t t led there and moved out again. To be able to make a de f in i t i ve statement based on the trends of this sample is impossible. Rather, a pa ra l l e l sample must be drawn from a cross-sect ion of newcomers dwelling throughout Metropolitan Vancouver. TABLE III.21 NEWCOMERS' FORMER PLACES OF RESIDENCE A. Newest Group (0-2 years in Vancouver) Size Location Under 100,000 Over 100,000 Total B r i t i s h Columbia 5 1 6 Canada (except B.C.) 1 14 15 Outside Canada 1 5 6 TOTALS ' 1 20 27 B. Newer Group (2.5-6 years in Vancouver) Size Location Under 100,000 Over 100,000 Total B r i t i s h Columbia 8 3 11 Canada (except B.C. ) 1 4 5 Outside Canada 2 3 5 TOTALS 1 11 10 21 Now the f i n a l question to ask with regard to newcomers i s why did they come to Vancouver? I t would appear that there are three p o s s i b i l -i t i e s . Ei ther there was a push or a pul l or a combination of push and p u l l . To i l l u s t r a t e , "loss of job in Winnipeg" would be a push, while Ft "having a job offer in Vancouver" would be a p u l l . i Seven of 45 people ( i . e . , 15 per cent) gave answers coded as push factors . Six (or 13 per cent) gave answers that had to be coded as a combination of push and p u l l , while the vast majori ty , 32 of 45 ( i . e . , 71 per cent) , named purely pul l factors . This would seem to indicate that Vancouver has a great a t t rac t ion for outs iders . C. Factor Analysis In the previous two sections of th is chapter the personal charac-t e r i s t i c s of the ent i re sample and of the sample broken down by r e s i -dence length were described. Essen t i a l ly 11 key variables were used in those descr ip t ions . These included education, occupation, income, age, marital s ta tus , ch i ld ren , age of youngest c h i l d , locat ion of dwelling on an East-West a x i s , locat ion of dwelling on a North-South a x i s , dwell ing type, and homeownership. In the fol lowing discussion the age of the youngest c h i l d w i l l not be employed.'' However, a variable based on zon-ing w i l l be used. I t was assumed that a factor analysis could be carr ied out to . break the 11 background variables into a few basic dimensions. This was thought to be useful for the purposes of l a t e r analysis since so many of these variables are commonly known to be in te r re la t ed ; e . g . , occupation i s usually related to education and dwelling type is usual ly correlated with homeownership. I t was further supposed o r i g i n a l l y that three factors could be derived to account for most of the var ia t ion in the 11 o r ig ina l var iab les . The three predicted factors corresponded to the topics in the previous sect ions , that i s , socioeconomic s ta tus , stage in family l i f e cyc le , and charac ter i s t ics of dwelling units and the i r loca-t i o n . , \ , The factor analysis was based on the zero-order cor re la t ion matrix reproduced on the next page (Table 22). Surpr is ingly very few of the variables exhibited high in te rcor re la t ions . However, a few variables, were re la ted . Zoning was highly correlated with dwelling type and with North-South a x i s , while homeownership was highly correlated with both: age and dwelling type. I n i t i a l l y i t would appear that there was very l i t t l e cor re la t ion among the three areas socioeconomic s ta tus , l i f e c y c l e , and dwelling cha rac t e r i s t i c s , except for the re la t ionship between agei (a l i f e cycle variable) and homeownership (a dwelling c h a r a c t e r i s t i c ) . The discussion of the factor analysis w i l l show whether th i s was in fact the case. ; g The results of the factor analysis are summarized in Table 23. The names given to the three factors were based, not necessari ly on the variables most highly correlated with them, but rather on the variables that seemed to account the most for high factor scores for pa r t i cu l a r individuals on that fac tor . However, for two of the three factors th is name and that of the most highly correlated var iable were i d e n t i c a l . ZERO-ORDER CORRELATION MATRIX FOR 11 VARIABLES TO BE FACTOR ANALYZED 1 Zoning 2 Mari ta l Status 3 Age 4 Education 5 Occupation 6 Children 7 Home-ownership 8 Income q Dwel1ing type 1 n East-West l u Axis North-South 1 1 Axis 10 11 East-West Axis North-South Axis 1 Z ° n 1 " 9 2 ftllls 3 ** . Educa- r Occu- r Chi1• t ion pation dren 7 H o m e ° w " 8 Income 9 D f 1 1 i n g nership Type * 1.00000 0.18231 * 1.00000 0.37579 0.18139 * 1.00000 -0.14552 -0.00274 -0.48239*1.00000 0.02435 -0.17033 0.07022 -0.00393 * 1.00000 0.24098 ,0.36268 0.14897 -0.03470 -0.19174*1.00000 0.54365 0.29411*0.67066 -0.35906 -0.03374 0.15357*1.00000 0.03754 0.51954 0.12047 0.07053 0.30042 0.10149 * 1.00000 -0.50809 * 0.81336 0.24919 0.45728 -0.25755 0.01849 0.36246*0.62767 0.06616*1.00000 -0.10017 0.04844 -0.06635 0.11370 0.02539 -0.08082 0.00553 0.05509 0.03137 0.05012 0.28636 -0.15553 -0.07976 0.13090 0.39706 -0.02567 0.52212 * 0.73881 10 East-West Axis * 1.00000 -0.19934 11 North-South Axis 1.00000 * Indicates an absolute value qreater than or equal to 0.60000. SUMMARY OF FACTORS Factor Number Main Descript ive Variable Number of Variables Described One North-South Axis 4 Two Income 4 Three Age and Education 3 11 I t appears that the three factors which explain most of the var-i a t ion in this set of data do indeed break down by the three basic dimensions hypothesized e a r l i e r . I t seems that North-South axis repre-sents a dwelling dimension; income represents a socioeconomic dimension; and age and education represent a l i f e cycle dimension. However, as each factor i s now analyzed separately in d e t a i l , i t can be seen that the three seemingly autonomous dimensions do overlap to some degree. ( Factor One Factor One iden t i f i ed the North-South contrasts in zoning, dwel-l i n g type, and homeownership. Table 24 gives the factor loadings, or cor re la t ions , for these four variables most c lose ly associated with Factor One. Since a l l variables had pos i t ive loadings th is i s an i n d i -cation that individuals that had higher rank scores on zoning, i . e . , that were from single family zoned areas, also had higher rank scores on the other three var iab les . This means that they owned the i r own single LOADINGS FOR FACTOR ONE Loading Variable Number Variable Name Pos i t ive Negative 1 Zoning 0.93474 9 , Dwelling Type 0.82426 11 North-South Axis 0.81410 7 Homeownership 0.51496 family detached house in the most southerly area, that area between Broadway and Sixteenth Avenue. Table 25 summarizes for individuals with highest factor scores on Factor One (greater than +1.2 and less than -1.2) the mean rankings of these individuals on a l l four variables associated with this fac tor . TABLE II I .25 INDIVIDUALS WITH HIGH FACTOR SCORES ON FACTOR ONE Average Rankings on Var iab les : Zoning Dwelling Type North-South Homeownership Pos i t ive Scores Range 1.236 to 1.978 19 Subjects 3.000 4.736 4.000 1.722 Negative Scores Range -1.201 to -1.594 7 Subjects 1.000 1.857 1.286 1.000 Total Sample 1.741 3.259 2.778 1.274 The individuals with high posi t ive factor scores on Factor One can be characterized as was stated in the immediately preceding para-graph. At the opposite extreme, people with high negative scores on this factor can be described as individuals renting suites in low-rise apartment blocks in that part of K i t s i l a n o zoned for mul t ip le family use that i s located between Cornwall and English Bay, the most northerly 9 loca t ion . Factor Two i Factor Two i s pr imar i ly a socioeconomic status fac tor . A d i f f e r -ent set of four variables correlated highly with th i s fac tor . Income was the s ingle most highly correlated va r i ab le . (See Table 26) The other three variables included marital s ta tus , occupation, and ch i ld ren . Variable Number 8 2 5 6 TABLE 111.26 LOADINGS FOR FACTOR TWO Variable Name Income Mari ta l Status Occupation Children Loading Pos i t ive Negati ve 0.86689 0.69684 0.68641 0.56312 The negative loading for occupation is deceptive. This occurred because occupation was coded in reverse order from other var iab les . Instead of i giving the higher status occupations the highest number ranks, the pro-fessional category was given a rank of 1. The mean rankings on the four d is t inguishing variables for i n d i -viduals having high scores on Factor Two are given in Table 27. TABLE III .27 INDIVIDUALS WITH HIGH FACTOR SCORES ON FACTOR TWO Average Rankings on Variables : Income Mari ta l Status Occupation Children Pos i t ive Scores Range 1.276 to 2.711 10 Subjects 5.300 5.800 1.500 2.400 Negative Scores Range -1.204 to -1.791 15 Subjects 1.000 1.133 8.867 0.000 Total Sample 3.324 3.713 4.670 0.518 It can be seen that respondents having high pos i t ive factor scores on this factor were mostly in the highest income group with the highest status occupations. Most were married and had ch i ld ren . By contrast high negative scores indicate s ingle i n d i v i d u a l s , and thus no chi ldren in the household. These people were also in the lowest income bracket (mostly earning less than $3,000 year ly) and had lowest status occupations. This group includes 8 of the 12 people in the sample who were presently univers i ty students along with 2 of the 7 re t i red i nd iv idua l s . From the variables highly correlated with this fac tor , two were included in the e a r l i e r analysis in the category socioeconomic status and two in the category stage in family l i f e cyc l e . This can indicate e i ther that the factor analysis did not break down into the three dimen-sions hypothesized or that the variables were wrongly c l a s s i f i e d into the three dimensions e a r l i e r . Judgment w i l l be reserved u n t i l af ter the analysis of the f i n a l factor . Factor Three Factor Three was highly correlated with three of the o r ig ina l 11 var iables . Education was. very highly correlated in a pos i t ive d i r e c t i o n , while age was almost as highly correlated in a negative d i r e c t i o n . Home-ownership was the th i rd variable correlated with th is fac tor . (See Table 28) \ TABLE II I .28 LOADINGS FOR FACTOR THREE Variable Number 4 3 7 Variable Name Education Age Homeownership Loading Pos i t ive Negative 0.82708 0.82221 0.65520 Table 29 of high pos i t ive and negative factor scores indicates that high posi t ive scores went along with highly educated young people who were tenants, while high negative scores were attained by e lder ly people (average age of nearly 60) who had very l i t t l e education compared to the rest of the respondents, yet who owned the i r own dwel l ing . INDIVIDUALS WITH HIGH FACTOR SCORES ON FACTOR THREE Average Rankings on Var iables : Education Age Homeownership Pos i t ive Scores Range 1.203 to 1.818 11 Subjects 6.636 1.364 1.000 Negative Scores Range -1.203 to -2.645 14 Subjects 3.000 5.929 1.929 Total Sample 4.454 2.972 1.274 While the variable homeownership loaded highly on Factor One and Factor Three, only 5 of 26 respondents who had high scores e i ther pos i -t ive or negative on Factor One also had high factor scores on Factor Three. Because of th is education and age can be considered the main var iables . Homeownership made a much smaller contr ibut ion to Factor -Three. Factor Analysis Conclusions Three factors have been derived which account for a great deal of the variance in the 11 o r ig ina l variables (approximately 60 per cent of the va r i a t i on ) . While a l l these variables were o r i g i n a l l y c l a s s i f i e d into socioeconomic s ta tus , stage in family l i f e c y c l e , and dwelling charac ter is t ics categories, the three kinds of categories did not determine the factors . This i s because 1 of the 3 factors included var-iables from 2 of the 3 o r ig ina l se ts , and another of the factors i n -cluded variables from each of the o r ig ina l sets of var iab les . At one stage during the in terpreta t ion of th i s factor analysis i t was surmised that a l l the variables put together determined factors that would so le ly characterize different stages in the family l i f e cyc le . At that point i t was assumed that Factor One described older homeowners with older ch i l d r en , Factor Two higher income younger famil ies who were tenants, and Factor Three unmarried highly educated people with lower status occupations. However, when the complete in terpre ta t ion jus t des-cribed was carr ied out, th i s assumption was found to be erroneous.; Rather, i t was necessary to know a person's factor score on at least 2 and often a l l of the three factors to be able to derive a complete des-c r i p t i on of his cha rac t e r i s t i c s . And th is seemed to be espec ia l ly ; t rue for those individuals who had only moderate or low factor scores. For admittedly, high factor scores represent the extreme cases, and only rarely would people ranking near the. means be i n c l u d e d . ^ D. Summary Chapter III was designed to give a complete background descr ipt ion of the subjects making up the sample. To do th is the f i r s t section tabu-lated the responses to the background questions under three headings, socioeconomic s ta tus , stage in family l i f e c y c l e , and dwelling character-i s t i c s . Next Section B was designed to break down the previously analyzed personal charac te r i s t ics by four residence length categories , 0-2 years , 2.5-6 years , 7-21 years, and 22-60 years. Each of the pre-viously described table was summarized on th is bas is . Then the two newer groups were examined in more de ta i l to discover where they had come from and why. The conclusion was that most people moved to Vancouver because they were pulled by some (as yet undefined) a t t rac t ion in Van-couver. And on the basis of K i t s i l a n o alone i t would seem that the a t t rac t ive power drew mostly rural residents of B r i t i s h Columbia 3-6 years ago, but i s now drawing mostly people from c i t i e s of over 100,000 in other parts of Canada. And in Section C a factor analysis was carr ied out on the data of Section A to see i f the three categories would be maintained. In f ac t , the factor analysis did come up with three basic dimensions in the 'data. But these were en t i t l ed locat ion of dwelling on a North-South a x i s , income, and age and education. The question on the questionnaire dealing with occupation of the main breadwinner in the family was not coded u n t i l a l l interviews were complete. At that time i t was discovered that the information avai lable from respondents would make c l a s s i f i c a t i o n on the basis of Dominion Bur-eau of S t a t i s t i c s categories nearly impossible. For that reason, the categories of the U.S . Census were used. However, two addit ional cate-gories were added. Since K i t s i l a n o i s usual ly assumed to consis t of a high proportion of univers i ty students and of older pensioners, these two categories were added instead of c l a s s i fy ing these groups as "unem-ployed." , 2 ! Six ind iv idua ls interviewed refused to divulge the i r income. . 3 The c l a s s i f i c a t i o n comes from John B. Lansing and L e s l i e K i s h , "Family L i f e Cycle as an Independent Var i ab l e , " American Sociologica l  Review, XXII (October, 1957), 512-519. Since the authors of that a r t i c l e never stated an age to use as a cu t -of f poin t , the age of 35 was used to d is t inguish between young and o l d . ^The sample was o r i g i n a l l y drawn so that 90 dwelling units would be contacted in mul t ip le family zoned areas of K i t s i l a n o , 30 units in 2-family areas, and 50 units in s ingle family areas (p.20). I t would now be useful to see the p ro f i l e of respondents in this regard. Of the 90 possible respondents in mult iple areas, 57 or 63.3 per cent were interviewed; of the 50 possible respondents in s ingle family areas 29 or 58 per cent were interviewed; and of the 30 possible respondents in 2-family areas 22 or 73.3 per cent were interviewed. Thus the f i na l sam-ple consisted of units drawn from zoning areas in almost the exact proportions desired at the outset: 52.78 per cent from mult ip le family areas, 20.37 per cent from 2-family areas, and 26.85 per cent from single family zoned areas of K i t s i l a n o . 5 I t seems that th is information can be used to predict where the newer residents are l i k e l y to se t t le wi th in one area of a c i t y . New-comers from B r i t i s h Columbia can be found almost exc lus ive ly in the mult iple family area of K i t s i l a n o . Not one newcomer from B.C. who was interviewed l i v e d wi th in the s ingle family area. This i s not the case with newer residents coming to Vancouver from other parts of Canada. . One th i rd of these newcomers are in s ingle family zoning, 57 per cent in the mul t ip le areas, and 10 per cent in the 2-family zoned areas. When i t comes to locat ing newcomers from outside of Canada the 2-family area seems to be the most popular. Five of the 12 people from outside Canada dwell in 2-family zoned areas, while the remainder are evenly s p l i t be-tween s ingle family and mul t ip le areas. ^In order to keep the interview as short as poss ib le , the res-ponse to reason for moving to Vancouver was taken at face value, with no probing for hidden or subjective reasons. -, 7 This var iable had too few observations to be of use ( i t only . applied to 22 of 108 respondents). ' Before th i s variable was omitted one factor analysis was run, and th is variable loaded nearly equally on a l l the factors . Thus i t could l o g i c a l l y be deleted with l i t t l e d i f f i c u l t y . 8 ' The computer program-used was UBC BMDX72, implemented from the UCLA BMD Package by Jason Helm, December, 1970. The output of the fac-tor analysis included mean, standard dev ia t ion , maximum, minimum, var-iance, and sum for each va r i ab le ; a complete cor re la t ion matrix for a l l var iab les ; the unrotated and varimax (maximized variance) rotated factor loadings matrices; factor scores rat ing the individual respondents.on, each of the dimensions; and mean, standard devia t ion , and cor re la t ion ; matrix of the factors . One source that was of special help in interpre-t ing the resul ts of th i s factor analysis was "The Socio-Economic Dimen-sions and Spat ial Structure of Canadian C i t i e s " by D. Michael Ray, e_t a l . , a Report on the F i r s t Phase of a CMHC Projec t , Univers i ty of Waterloo, Summer 1968 (mimeographed). g Many of the respondents could have had high negative factor scores on Factor One because they f i t the descr ipt ion of 3 out of the 4 var iables . However, they did not l i v e in the most northerly area. Thus the North-South axis var iable was the determining one, and deserves to name th is dimension. ^ F a c t o r analysis i s based on the zero-order cor re la t ion matr ix. However, very few of the 11 variables from th is data that were factor-analyzed were highly in te rcor re la ted . Thus i t i s not surpr is ing that the three factors only accounted for 60 per cent of the var ia t ion in the set of data. Six ty per cent i s not very high by comparison with other s tudies . For th is reason, and also because of the low degree of i n t e r -co r r e l a t i on , the factor analysis w i l l not be used extensively in test ing the hypotheses. Rather, the raw data w i l l be appl ied. However, regres-sion was carr ied out with the factor scores. Where th is has been done and the results were different than with the raw data, the factor scores' results w i l l be reported in the footnotes. RESULTS: HYPOTHESIS I In this chapter and the two- that fo l low, the tests carr ied out on the three hypotheses set out in Chapter I w i l l be presented. In tes t -ing each of these hypotheses three s t a t i s t i c a l methods were used. These included percentage and average comparisons, regression ana lys i s , and multiple discriminant ana lys is . Percentage comparisons are so widely understood that no descr ipt ion pf that technique w i l l be needed. How-ever, in Section A of th is chapter the two more sophist icated techniques, regression analysis and discriminant ana lys i s , w i l l be described. Following the explanation of these techniques, f i r s t d i sc r imin-ant ana lys i s , then regression analysis and percentage comparisons w i l l be used to test the hypothesis that newer residents ' opinions regarding the qua l i ty of the i r neighbourhood's services and f a c i l i t i e s vary s i g -n i f i c a n t l y from those of longterm residents . A. Description of Regression Analysis and Discriminant Analysis Regression Analysis The f i r s t s t a t i s t i c a l technique that needs to be described is re-gression analys is . The descript ion w i l l include f i r s t what the technique i s and how i t works, second the assumptions underlying the technique, and l a s t l y what the computer output i s when th is technique i s used. This type of analysis can t e l l whether two properties are related in a l inear manner such that the value of one (the independent variable or "predictor") can be used to predict the value of the other (the de-pendent or " c r i t e r i o n " va r i ab l e ) . ••Both simple and mul t ip le regression have been used. Simple regression has one dependent and one independent var iab le . Mul t ip le regression, while only dealing with one dependent] variable at a time, considers several possible independent va r i ab le s . . The computer program used 1 selects independent variables one at a time and adds them to the l i nea r equation such that the variable chosen makes the maximum explanatory contr ibut ion to the regression explanation and 2 eliminates redundancy or in te rcor re la t ion as much as poss ible . The basic assumptions underlying the regression model that should be remembered are 1. that y (the dependent variable) i s a random variable and i t s values are normally d is t r ibu ted around the regression equation; 2. that x 's (the independent variables) are not random variates but are constants; 3. that observations are independent; 4. that the re la t ionship is l i n e a r ; 3 5. that the predictor variables are independent of each other. The computer output includes the level of s ignif icance of the F r a t i o , used to test the s ignif icance of the regression coe f f i c i en t , for o i a b l e w i t h F probabi l i ty less than 0.05. The printout also gives an r figure and the standard error of estimate of the dependent var iab le . 2 r i s the proportion of to ta l var ia t ion that has been explained by the regression l i n e . The standard error i s a measure of the accuracy of . p red ic t ion . I f the standard error i s large ( re la t ive to the mean of the va r i ab l e ) , the conclusion must be that the regression model i s unsatis-factory. Discriminant Analysis The second s t a t i s t i c a l technique that needs to be described i s mult iple discriminant ana lys is . The descript ion w i l l f i r s t include what the technique does and how i t works and then w i l l out l ine the assumptions underlying the technique. Discriminant analysis s tar ts with a predetermined c l a s s i f i c a t i o n of groups ( in this study on the basis of length of residence in Van-couver), while in other multidimensional techniques group membership de-termination follows from the results of the ana lys i s . This technique attempts to maximize the discriminations between soc ia l groups that are perceived as natural ( e . g . , sex) or that have already been defined, 5 t heo re t i c a l l y . In maximizing the*discriminations between the groups the analysis answers the following four questions: 1) Is the a p r i o r i c l a s s i f i c a t i o n under consideration v a l i d in terms of the variables that are being suggested as relevant to the c l a s s i f i c a t i o n ? 2) What "dimen-sions" val idate th is c l a s s i f i c a t i o n , i f i t was found in fact to be, val id? 3) Where do the groups l i e in terms of these dimensions, i . e . , what i s the composite score for each group on the dimensions? and 4) Given the score of a new subject in terms of the o r ig ina l va r iab les , which group does i t f i t in best? In essence the mult iple discriminant analysis technique compares the degree of homogeneity within the groups to the degree of hetero- } geneity between the groups. And the level of s ignif icance of the F ra t io ( in this computer program,'7 always less than 0.05 to be counted s ign i f i can t ) represents the p robabi l i ty that th i s c l a s s i f i c a t i o n might have been produced by chance. There are some basic assumptions underlying this technique that should be mentioned. F i r s t , i t i s assumed that the groups exhaust;the universe with which one i s deal ing. This i s espec ia l ly necessary in order to answer the fourth question mentioned above. A l s o , the l inea r discriminant functions or "dimensions" are assumed to be independent, add i t ive , and orthogonal (which means uncorrec ted with one another). : Summary Regression analysis and discriminant analysis can be summed up in nons ta t i s t i ca l terms. E s s e n t i a l l y , regression analysis indicates to what degree knowing a person's "score" on one variable ( e . g . , residence g length) w i l l predict his score on another var iab le . Discriminant anal-y s i s , on the other hand, indicates what variables discriminate between different groups. In terms of th is study, discriminant analysis picks out those variables on which the different residence length groups have s i g n i f i c a n t l y dif ferent "scores." : B. Tests of the F i r s t Hypothesis by Discriminant Analysis A discriminant analysis was run on 28 qua l i ty variables at one time. These variables were derived from a restructuring of the o r ig ina l information obtained from the questionnaire. This restructuring was accomplished by using two dif ferent procedures. The f i r s t procedure yie lded 18 variables that were derived as fo l lows. The 18 o r ig ina l services and f a c i l i t i e s were given a number corresponding to the ranks to which they were assigned by the people interviewed. That i s , i f a service or f a c i l i t y was placed in the category "best" i t was given a rank of 1; "2nd best" resulted in a rank of 2; and so on down to "worst" with a rank of 10. The second set of variables was obtained as fo l lows . The answers given to each of the 10 possible rankings of services and f a c i l i t i e s (from 5th best to best and from 5th worst to worst) were made into f re-quency scores. The most popular answer in each ranking category was given a score of 10, while answers only rarely mentioned in that category resulted in a score of 1. For example, in the rank category of "best" f a c i l i t y or se rv ice , 30 people mentioned locat ion of major shopping. Therefore th is answer resulted in a score of 10. On the other hand, only 2 people f e l t that a i r po l lu t ion was "best," so th is answer resulted in a score of only 1. Having thus derived 28 qua l i ty va r iab les , these were used in a discriminant analysis of four categories of residence length, 0-2 years , 2.5-6 years, 7-21 years, and 22-60 years. The analysis turned out to be s i g n i f i c a n t . However, in examining those variables that val idated this c l a s s i f i c a t i o n , i t was surpr is ing at f i r s t to notice that both of the s ign i f i can t variables came from the 10 ranking variables and none at a l l from the 18 o r ig ina l service and f a c i l i t y var iab les . The two var-iables that val idated the c l a s s i f i c a t i o n of residence length were the best rank and the 5th best rank. TABLE I V . l F RATIO FOR INDIVIDUAL VARIABLES* Variable F Ratio Degrees of Freedom Level of Signif icance Best 4.0566 3 and 104 0.0091 5th Best 2.8133 3 and 103 . 0.0423 Overall 3.4106 6 and 206 0.0032 S ign i f i can t Only I t would seem from the level of s ignif icance of the F rat ios that there i s l i t t l e p robabi l i ty that this c l a s s i f i c a t i o n might have been pro-duced by chance. The c l a s s i f i c a t i o n has been val ida ted . Next the members of each group were compared to the composite scores for the groups on the va l ida t ing dimensions to see which group they f i t in best. When this step was accomplished the c l a s s i f i c a t i o n seemed to be less useful than was indicated by i t s F p robab i l i t y . When a c l a s s i f i c a t i o n i s a good one most cases w i l l f a l l on the diagonal. But as can be seen in Table 2 th is was not the case with this c l a s s i f i c a t i o n . None of the newest cases, 44.4 per cent of the newer, 51.7 per cent of the o lder , and 36.0 per cent of the oldest f e l l on the TABLE IV.2 NUMBER OF CASES CLASSIFIED INTO GROUP: Newest Newer Older Oldest Newest 0 12 8 7 Newer 0 12 8 7 Older 0 9 15 5 Oldest 0 7 9 9 0 40 40 28 diagonal. There must be some other underlying c l a s s i f i c a t i o n that could help explain th i s unsatisfactory poster ior c l a s s i f i c a t i o n . The prel im-inary conclusion must be that on the basis of opinions on the qua l i ty of neighbourhood services and f a c i l i t i e s the c l a s s i f i c a t i o n of residence lengths is v a l i d but not the only c l a s s i f i c a t i o n that explains the d i f -ferent responses of different i nd iv idua l s . C. Tests of the F i r s t Hypothesis by Regression Analysis Regression analysis was applied to the f i r s t hypothesis using the same 28 qual i ty var iables . The 0.05 s ignif icance level for the F ra t io was used as the c r i t e r i o n for accepting or reject ing the independent var-i ab les . F i r s t the resul ts on the 10 rankings w i l l be presented, then the results on the 18 o r ig ina l services and f a c i l i t i e s . A complete table of a l l mult iple regression results can be found in Appendix D. Residence length proved s ign i f i can t in simple regression equations for only the ranking 5th best. This would seem l i k e l y since 5th best was one of the two s ign i f i can t discr iminat ing var iables . However, when the effects of a l l of the other background variables were held constant in stepwise mult iple regression, length was no longer a s i gn i f i c an t inde-pendent variable for 5th best. Rather, the effects of education, home-ownership, and locat ion of dwelling on a North-South axis were much greater. Residence length became s ign i f i can t in mult iple regression for 4th best and for worst, when i t was not s ign i f i can t on these rankings, in simple regression. Residence length was the only independent var iable that was s ign i f i can t on 4th best. As such i t explained 1.4 per cent of the var ia t ion in the dependent variable (r =0.014). This i s a very small 2 r and indicates that the cor re la t ion between the two variables was only 0.118. This too is not a very high f igure . I t i s p o s i t i v e , however, and indicates that newer residents gave less popular responses on 4th best. Another way of test ing the goodness of the predict ion of 4th best from residence length i s by noting the s ize of the standard e r ro r . The stan-dard error for 4th best was approximately 50 per cent of the value of i t s mean. This too indicates that there i s a great deal of error or var-i a t ion unexplained by residence length alone. In the mult iple regression for the ranking worst, residence length was not the only independent var iable entered as s i g n i f i c a n t . Rather, age was selected f i r s t and accounted for 5.1 per cent of the var ia t ion in worst by i t s e l f . Residence length and age together accoun-ted for 9.2 per cent 'of the var ia t ion in worst. The cor re la t ion between residence length and worst was 0.0752 (a f a i r l y low figure but since i t was posi t ive th is would indicate that newer residents chose less popular responses on worst) . When th is figure i s squared i t would indicate that residence length alone should be explaining 5.6 per cent of the var ia t ion in worst. The difference, between 5.6 per cent and 4.1 per cent (9.2 per cent-5.1 per cent) must be accounted for by the in terac t ion between age and residence length. Even 4.1 per cent i s much higher than the 1.3 per cent explained var ia t ion for 4th best. However, the standard error for worst was also too high, - approximately 50 per cent of i t s mean, g again causing one to question the accuracy of the p red ic t ion . Service and F a c i l i t y Variables Residence length was a s i gn i f i can t independent var iable in simple 2 regression for hospitals only , with an F p robab i l i ty of 0.0127 and an r of 0.0566 (or 5.66 per cent of the var ia t ion explained). In mult iple regression, residence length was s ign i f i can t and i n -cluded again for hospitals only. As was the case for worst, residence length was not the only independent va r iab le . Location on a North-South axis was entered f i r s t and explained 9.4 per cent of the v a r i a t i o n . North-South axis had an F probabi l i ty of ,0.0033, while residence length showed an F probabi l i ty of 0.0317. Together they accounted for 13.2 per cent of the va r i a t i on . This indicates that the contr ibut ion of residence . length alone was 3.8 per cent. Here, too, there was a difference between 2 this value and that of the r in the simple regression that i s accounted for by in terac t ion between North-South axis and residence length. Resi -dence length and hospitals had a corre la t ion of -0.2379. This negative corre la t ion indicates that newer residents rated hospital services lower than older residents. A caveat should be mentioned at th i s time. While the standard error of residence length was only 12 per cent of the value of i t s mean, that for hospitals was greater than 100 per cent of the value of i t s mean. This i s a very high standard error which makes the results rather ques t ionab le . 1 0 2 In summary, although the standard errors were high and the r s low, residence length was s ign i f i can t in predict ing the ranking variables 4th best and worst and the service and f a c i l i t y variable hosp i ta l s . Newer residents on the whole chose less popular answers on these two ranking variables and f e l t that the qua l i ty of hospitals was much worst than did longterm residents . D. Tests of the F i r s t Hypothesis by Percentage Comparisons The l a s t method which was used to test the f i r s t hypothesis was simple percentage comparisons. These comparisons could not be tested with a Chi-square technique because the number of responses in each cate-gory was too few. However, th i s examination can indicate tendencies and thereby add to the evidence of the two more s t a t i s t i c a l techniques. Many respondents were not able to pick out f ive biggest and f ive least problems. Even th is group is meaningful in terms of residence length. Table 3 displays the residence length for people who did not choose any answer for the pa r t i cu la r ranking. Generally speaking, i t would seem that residents of K i t s i l a n o could f ind more good things about the i r area than problems. This comes out from the number of respondents not answering each question. - Once a person did not provide an answer for a rank, a l l ranks lower than that one also were answerless. So that to say that 45 people could not provide answers to the best ranks while 69 people could not f ind enough bad things in the area i s not quite accurate. However, i t does indicate that people on the whole stopped providing answers to the worst rankings e a r l i e r than to the best rank-ings. That i s , they could f ind more things good with the area than bad. The residence length differences are very c lear cut in those not providing answers to the best rankings, but not so c lear in the worst categories. Remembering that the tota ls are not t ru ly accurate but pre-sent trends or impl ica t ions , the newer group (2.5-6 years in Vancouver) only had 5 of i t s members not providing answers to some of* the best questions, while 9 people from the newest and oldest groups did l ikewise . However, 22 people from the older residence length group (7-21 years in Vancouver) could not provide answers for a l l f ive least problem ranks. I t would seem that people in Vancouver for 7-21 years can f ind the least amount of things to praise in t he i r area, and people in Vancouver for 0-2 years can f ind the most to pra ise . PER CENT OF RESIDENCE GROUP NOT PROVIDING ANSWERS FOR RANK: Ranks Newest Newer Older Oldest Total Number •5th Best 18.2 18.2 36.4 27.2 100.0 22 4th Best 18.2 9.1 54.5 18.2 100.0 11 3rd Best 33.3 - 50.0 16.7 100.0 6 2nd Best 25 .0 . 75.0 - 100.0 4 Best - - 100.0 - 100.0 2 Worst 25.0 25.0 50.0 - 100.0 i 4 2nd Worst 16.7 16.7 33.3 33.3 100.0 6 3rd Worst 18.1 27.3 27.3 27.3 100.0 11 4th Worst 25.0 25.0 20.0 30.0 100.0 20 5th Worst 25.0 17.9 25.0 32.1 100.0 28 Quality Rank Variables Third Best Rank. Disregarding the two lowest ranks in the best and worst sections for s i gn i f i can t lack of responses, the f i r s t rank to examine i s that of 3rd best. Here f i r e pro tec t ion , hosp i t a l s , and hous-ing conditions seem dif ferent for dif ferent residence length groups. No one from the newest group mentioned the f i r s t of these, but i t was mainly newest residents mentioning the other two. (See Table 4) ; ? Second Best Rank. In the 2nd best rank there were differences on the mention of shopping, parking, and parks. The newest group mentioned QUALITY RANK BREAKDOWN BY RESIDENCE LENGTH Newest Newer Older Oldest Total Number 3rd Best Fire Protection 30 .0 30 .0 40 .0 100 .0 10 Hospitals 62 .5 12 .5 12 .5 12 .5 100 .0 8 Housing Conditions 100 .0 100 .0 2 2nd Best -Shopping 9 .1 27 .3 36 .3 27 .3 100 .0 11 Parking 60 .0 20 .0 20 .0 100 .0 5 Parks 37 .5 25 .0 12 .5 25 .0 100 .0 16 Best Hospitals 50 .0 25 .0 25 .0 100 .0 4 Shopping 30 .0 33 .3 10 .0 26 .7 100 .0 30 Worst Po l lu t ion 17 .6 23 .5 44 .1 14 .7 99 .9 34 Back Lanes 16 .7 16 .7 33 .3 33 .3 100 .0 6 Fire Protection 100 .0 100 .0 2 Housing Conditions 30 .0 50 .0 10 .0 10 .0 100 .0 10 Public Schools 50 .0 25 .0 25 .0 100 .0 4 Lighting 100 .0 100 .0 2 Maintenance 20 .0 40 .0 40 .0 100 .0 5 2nd Worst • Pol lu t ion 35 .7 21 .4 21 .4 21 .4 99 .9 14 Hospitals 83 .3 16 .7 100 .0 6 Maintenance 33 .3 33 .3 16 .7 16 .7 100 .0 6 Tra f f i c Control 12 .5 12 .5 37 .5 37 .5 100 .0 8 3rd Worst Po l lu t ion 38 .5 23 .1 15 .3 23 .1 100 .0 13 Light ing 33 .3 33 .3 16 .7 16 .7 100 .0 6 Tra f f i c Control 50 .0 25 .0 25 .0 100 .0 4 shopping l eas t , but mentioned parking and parks more than any of the other groups. Best Rank. When the answers in the best category are examined, one can find residence length differences on the mention of hospitals and shopping. Again i t was mostly the newest group that mentioned hos-p i t a l s . I t was also the two newer groups that mentioned shopping most frequently. Worst Rank. There seem to be seven answers d i f ferent for the d i f -ferent residence length groups on the ranking of worst. These were a i r p o l l u t i o n , back lanes, f i r e protec t ion , housing condi t ions , public schools, s t reet l i g h t i n g , and s treet maintenance. People in the newest category gave answers of f i r e pro tec t ion , publ ic schools, and s t reet l i gh t i ng more frequently than a l l the other groups, and answered less frequently on a l l the other 4 services and f a c i l i t i e s . Second Worst Rank. In the category of 2nd worst the newest group answered more frequently a i r p o l l u t i o n , hosp i t a l s , and s treet mainten-ance and less frequently than other groups mentioned t r a f f i c con t ro l . Third Worst Rank. In the 3rd worst rank, the newest residents more frequently mentioned p o l l u t i o n , s t reet l i g h t i n g , and t r a f f i c con-t ro l than longerterm residents . Summary In summary, percentage comparisons demonstrated that people, who had l i v e d in Vancouver for less than 2 years could f ind the most features to praise in the i r area, while people l i v i n g in Vancouver 7-21 years could f ind the leas t things to pra ise . And in each of the ranks from 3rd best to best and from worst to 3rd worst anywhere from 2 to 7 of the service and f a c i l i t y variables were mentioned with seemingly s i g -n i f i c a n t l y different frequencies by the various residence length groups. Newer residents usual ly named items not un iversa l ly popular for the par-t i c u l a r category. E. Summary In this chapter the hypothesis that newer residents ' opinions re-garding the qua l i ty of t he i r neighbourhood's services and f a c i l i t i e s vary s i g n i f i c a n t l y from those of longterm residents was submitted to three types of s t a t i s t i c a l sc ru t iny . Discriminant analysis found:that differences between the 4 d i f f e r -ent residence length groups were s i g n i f i c a n t l y greater than the d i f f e r -ences wi th in each group on the two rank variables best and 5th best. The regression analysis showed that residence length alone was s ign i f i can t in predict ing the ranks 4th best and worst , and in each case i t was demonstrated that newer residents chose less popular answers. . Residence length was also s ign i f i can t in predict ing the rank assigned to hosp i ta l s , with newer residents fee l ing hospital qua l i ty was worse, than longterm residents seemed to f e e l . ; • And f i n a l l y , percentage comparisons showed what seemed to be many s ign i f i can t differences between answers to the dif ferent rank categories between the different residence length groups, with newer residents mostly choosing less un iversa l ly popular responses to each category. UBC TRIP (Triangular Regression Package), by James H. Bjerr ing and Paul Seagraves, November, 1970. 2 Nirmala devi Cherukupalle, "Relationship Between At t r ibu te s : Simple and Mul t ip le Regression and Cor re la t ion , " Mimeographed, p. 19. 3 I b i d . , p. 23. ' • 4 Solomon Re t t i g , "Mult ip le Discriminant A n a l y s i s , " American Socio- l og i ca l Review, XXIX (June, 1954), p. 399. 5 I b i d . , p. 399. Nirmala devi Cherukupalle, " C l a s s i f i c a t i o n Techniques in Planning Ana lys i s , " Socio-Economic Planning Science, IV (September, 1970), pp. 395-396. 7UBC BMD07M, Implemented from the UCLA BMD Package by Paul Sea-graves, October, 1970. Using th is program, i f any variables are entered by de f in i t i on the c l a s s i f i c a t i o n is s i g n i f i c a n t . However, with th is program there is then some problem in knowing exactly what the c l a s s i f i -cation indica tes . p One d i f f i c u l t y with the data of this study should be noted. A l l questions were coded on an ordinal bas i s , not on an in terva l sca le . Thus the l i n e a r i t y of the re la t ionship i s l i m i t e d . g When 4th best and worst were run in regression analysis with the factor scores from Chapter III and standardized scores for residence length, none of the independent variables was s ign i f i can t at the 0.05 l e v e l . 10 1 When hospitals was run as the dependent variable in regression analysis with the three factor scores and with residence length in stan-dardized (z) scores, none of the factors was s i gn i f i c an t at the 0.05 ^ l e v e l . Thus only length was entered in the mul t ip le regression, with r equal to that of the simple regression reported in the text . RESULTS: HYPOTHESIS II The second hypothesis was that newer residents ' community p a r t i -c ipat ion varies s i g n i f i c a n t l y from that of longterm residents . Four variables were tested in this sec t ion . They included awareness of the community's popular name, awareness of neighbourhood organizat ions, attendance at government meetings, and number of voluntary associations to which the indiv idual belonged.^ In Section A the discriminant analysis car r ied out to test the hypothesis w i l l be described. Section B includes the regression analysis on these pa r t i c ipa t ion var iab les . And in Section C, percentage compari-sons w i l l be set fo r th . A. Tests of the Second Hypothesis by Discriminant Analysis A discriminant analysis was conducted on these four variables using the same four residence length categories , newest = 0-2 years , newer = 2.5-6 years , older = 7-21 years , and oldest = 22-60 years. The analysis turned out to be s i g n i f i c a n t . That i s , since one var iable had a high enough F ra t io to be entered, there was an ind ica t ion that differences wi th in the four groups were smaller than between the four groups. The one variable which was entered, that val idated the c l a s s i f i -ca t ion , was neighbourhood organizational awareness. This var iable had an F ra t io of 8.7241 with 3 and 104 degrees of freedom. Thus i t s level of s ignif icance was 0.0000. There was almost no p robab i l i ty that th i s c l a s s i f i c a t i o n might have been produced by chance. The members of each group were next compared to the composite . scores for these groups on th is va l ida t ing dimension to see which group they f i t in best. When th is step was accomplished the c l a s s i f i c a t i o n seemed to be less useful than was indicated by i t s level of s ign i f i cance . While 92.6 per cent of the newest cases f e l l on the diagonal , only 14.8 per cent of the newer, 55.2 per cent of the o lder , and none of the oldest did so. In th is c l a s s i f i c a t i o n i t would seem that there i s almost a dichotomy. There must be some other underlying c l a s s i f i c a t i o n that would help to explain the c l a s s i f i c a t i o n of Table 1. TABLE V . l NUMBER OF CASES CLASSIFIED INTO GROUP: Older Oldest 2 0 11 0 16 0 15 . 0 The preliminary conclusion must be that on the basis of the four p a r t i -c ipat ion variables the c l a s s i f i c a t i o n of residence lengths was v a l i d Newest Newer Newest 25 0 Newer 12 4 Older 9 4 Oldest 10 0 but not the only c l a s s i f i c a t i o n that explains the dif ferent responses of different i nd iv idua l s . B. Tests of the Second Hypothesis by Regression Analysis Regression analysis revealed that at the 0.05 leve l of s i g n i f i -cance residence length was only entered as a s i gn i f i can t independent variable in predict ing awareness of neighbourhood organizat ions, both 2 in simple and mult iple regression. See Appendix D for a complete l i s t -ing of the mult iple regression r e su l t s . Residence length was the only independent var iable of 12 possible variables that was entered as predict ing neighbourhood organization 3 awareness. Residence length had an F p robab i l i ty of 0.0001 and ex-plained 14.75 per cent of the var ia t ion in organization awareness. This 2 4 was the highest r encountered thus far in the data of th is study. However, the standard error of y , an ind ica t ion of the accuracy of. the p red ic t ion , was approximately 47.9 per cent of the value of i t s mean, causing one to question the sa t is factor iness of th is regression model. C. Tests of the Second Hypothesis by Percentage Comparisons  Organization Awareness Variable Since neighbourhood organization awareness showed up in both of the p r io r types of s t a t i s t i c a l ana lys i s , this, var iable w i l l be examined f i r s t in terms of percentage comparisons. The differences among residence length groups were immediately obvious. In the newest group only 2 of 27 people interviewed mentioned any organizat ion, compared to 15 of 27 in the newer group, 20 of 29 in the o lder , and 16 of 25 in the oldest . To phrase this another way, of those people not aware of any organization at a l l , 45.3 per cent were from the newest residence length group, 22.6 per cent from the newer, 17.0 per cent from the o lder , and 15.1 per cent from the oldest . And the opposite holds true for those-people aware of the major organizations. In th i s aware group only 4.4 per cent have l i ved in Vancouver less than 2 years , 24.4 per cent 2 .5-6 years, 35.6 per cent 7-21 years, and another 35.6 per cent have l i v e d in Vancouver 22-60 years. This d i rec t re la t ionship ( i . e . , the longer the residence the more aware of organization) holds even when con t ro l l ing for the zoning of the area. Government Meeting Attendance Variable The figures concerning government meeting attendance are also re-veal ing. Two people from each of the categories newest, newer, and oldest attended, while s ix people attended from the older residence length category. Those from the newest and oldest categories attended only p o l i t i c a l meetings. The newer category was evenly divided between medical meetings and meetings that were planning issue oriented. The older category, on the other hand, had some of i t s 6 people attending each of the three c l a s s i f i e d types of meetings. The older group was the most involved in meetings about planning oriented issues ( e . g . , board of variance, rapid t r ans i t discussions, e t c . ) . The results of the regression analysis on government meeting attendance are borne out by examining the frequency comparisons. Educa-t ion and age (Factor 3) were quite determining fac tors , as can be seen from the fact that 9 of 12 people who attended government meetings had some univers i ty education, and 10 of the 12 were under the age of 30. Now not ic ing that 13 of 27 people in the newest and 15 of 27 people in the newer group f e l l into the highest education categories, while 20 of 27 in the newest and 18 of 27 in the newer group are under 30 years of age, i t would seem that more of these people residing in Vancouver less than 7 years had the background necessary to give them a propensity to attend. Since they did not attend government meetings, on the whole, i t would seem that length of residence has a res t ra in ing ef fec t . Membership Variable The next pa r t i c ipa t ion variable to be examined in de ta i l i s member-ship in voluntary associa t ions . The two residence length categories with least pa r t i c ipa t ion were the newest (0-2 years) and older (7-21 years) with 33 per cent and 38 per cent belonging respect ive ly . The other two categories, newer and oldes t , each had 56 per cent belonging. However, the results are quite different when looking at only those indiv iduals who belong to more than one organizat ion. In this case there was a d i rec t r e l a t ionsh ip , i . e . , increasing residence length went along ,with an increased number of organizations to which members of that residence length category belong. Eight per cent in the newest category belonged to more than 1 organizat ion, 16 per cent in the newer, 21 per cent in the o lder , and 28 per cent in the oldest . On the basis of number of organizations belonged to by the people interviewed i t i s c lear that newer residents do not j o i n voluntary associations immediately upon the i r a r r i v a l in a new c i t y , but only after some time. K i t s i l a n o Awareness Variable The f ina l pa r t i c ipa t ion variable i s awareness of the area's popu-l a r name of K i t s i l a n o . ^ Jhose not knowing that the i r area had the name • K i t s i l a n o were 19 per cent of the newest group, 8 per cent of the newer, 7.7 per cent of the o lder , and 14.3 per cent of the oldest group. Thus the newest a r r i va l s in Vancouver were least l i k e l y to be aware of the name of the area they were l i v i n g i n . The great surprise was the oldest g group, which was the next largest group unaware of the name K i t s i l a n o . D. Summary To summarize, three s t a t i s t i c a l techniques were used to test the second hypothesis that newer residents ' community pa r t i c ipa t ion varies s i g n i f i c a n t l y from that of longterm residents. A l l three techniques found neighbourhood organization awareness to be the variable that showed the most differences between the various residence length groups. I t val idated the c l a s s i f i c a t i o n in discriminant ana lys is . That i s , the differences wi th in each residence length group on awareness of neighbour-hood organizations were s i g n i f i c a n t l y less than the differences between the groups. In mult iple regression analysis residence length alone explained 15 per cent of the var ia t ion in awareness of neighbourhood organizations. The percentage comparisons also reinforced these resul ts by showing a d i rec t r e la t ionsh ip . The longer the period of residence in Vancouver the more aware people were of the existence of neighbourhood organizations. While none of the three other pa r t i c ipa t ion variables proved s i g -n i f i can t in discriminant or regression ana lys i s , there were c lear i n d i -cations of trends demonstrated by percentage comparisons. Those people l i v i n g in Vancouver between 7 and 21 years attended government meetings, and espec ia l ly planning issue oriented ones, more than any of the other residence length groups. This was despite the fact that the newest r e s i -dents had the greatest propensity to attend from the i r age and level of education. A d i rec t re la t ionship was demonstrated between residence length and membership in more than one voluntary associa t ion. Newcomers did not seem to j o i n immediately upon the i r a r r i v a l . And f i n a l l y , awareness of the area's name (and iden t i ty as) K i t s i l a n o was least for newest a r r i v a l s . FOOTNOTES Appendix A contains the complete de ta i l s of the to ta l sample's rankings on these var iab les . In summary, most people were aware of the area's name, but less than hal f knew of the existence of any neighbour-hood organizations. Only 11 per cent of the sample had attended any government meetings in the l a s t year. And regarding voluntary associa-t ion membership, while 55. per cent of the sample belonged to no organ-i z a t i o n s , 18 per cent belonged to two or more associa t ions . The most frequently mentioned types were the hobby/sport type and re l ig ious organizations. i 2 Awareness of the name K i t s i l a n o seemed to be predicted mostly by area of residence, government meeting attendance by level of education at tained, and number of organizations belonged to by income. * 3 This was true for raw scores only. For standardized residence length and factor scores, Factor 2 (mostly income) was also s i g n i f i c a n t . And together 20.5 per cent of the var ia t ion in awareness of neighbourhood organizations was explained. 4 i However, r e l a t ive to other studies th i s does not have very;great explanatory power. 5 And the re la t ionship became even more confused when income was control led for (income was the one s ign i f i can t variable in regression . ana lys i s ) . Slhile the responses of the dif ferent residence length group mem-bers on type of membership are quite in t e res t ing , they are of l i t t l e use in test ing the hypothesis. The interested reader can f ind a table of membership type broken down by length of residence in Vancouver in Appendix E. 7There was an unexpected d i f f i c u l t y that arose in the course of conducting the interviews. At that time i t was discovered that 2 of the blocks out of the 17 in the sample were almost un iversa l ly considered by the i r residents not to be part of K i t s i l a n o . These were the blocks South of Broadway but East of Arbutus. A few of the oldest residents recal led when the i r area had the name Talten Place and thus f e l t that they had never been part of K i t s i l a n o . For th is reason, the 15 people interviewed from this area have been excluded from the analysis of th var iab le . g There is some p o s s i b i l i t y that people l i v i n g near Sixteenth Avenue and MacDonald Street lack i d e n t i f i c a t i o n with K i t s i l a n o . RESULTS: HYPOTHESIS III In this chapter the same three s t a t i s t i c a l techniques w i l l be applied to the th i rd hypothesis, which may be stated as fo l lows: Newer residents ' views on the qua l i ty of neighbourhood services and fac-i l i t i e s and the extent of t h e i r community pa r t i c ipa t ion vary s i g n i f i c a n t l y according to the s ize and locat ion of t he i r former residence. While reading i t may be helpful to keep in mind that rural indicates c i t i e s with population less than 100,000. A. Tests of the Third Hypothesis by Discriminant Analysis  Size and Location Separately Separate discriminant analyses were run on the c l a s s i f i c a t i o n by s ize and by loca t ion , each with the same 28 qua l i ty variables and 4 par-t i c i p a t i o n variables used in Chapters IV and V. In both analyses the c l a s s i f i c a t i o n s were val idated by three s i g -n i f i can t var iab les . Neighbourhood organization awareness val idated both c l a s s i f i c a t i ons and was the only pa r t i c ipa t ion variable included. The other two variables in each case were d i f fe ren t . The s ize analysis included 3rd best (a qua l i ty rank) and pol ice pro tec t ion , while the locat ion analysis included parking and parks (both o r ig ina l service and f ac i . l i t y va r i ab les ) . Table 1 summarizes the de ta i l s j u s t i f y ing the i n -clusion of these var iab les . • . TABLE VI.1 ' F RATIO FOR INDIVIDUAL VARIABLES . Signif icance C l a s s i f i c a t i o n Variable - F Ratio Degrees of Freedom Level Size 3rd Best 10.0331 1 and 44 0.0029 Orgaware* 6.2345 . 1 and 44 0.0156 Pol ice 4.2560 1 and 44 0.0427 Overall 6.7853 3 and 44 0.0008 Location Parks 5.0813 2 and 43 0.0104 Parking 3.8236 2 and 43 0.029D Orgaware 3.6462 2 and 43 0.0337 Overall 4.2246 6 and 86 0.0010 * Neighbourhood organization awareness From the level of s ignif icance of the F rat ios there seems to be l i t t l e p robabi l i ty that these c l a s s i f i c a t i ons might have been produced by chance. However, the goodness of the c l a s s i f i c a t i o n s can only be de-termined by comparing the members of each group to the composite score for the i r group to see which group they f i t best i n . Table 2 shows most cases f a l l i n g on the diagonals (for the s ize c l a s s i f i c a t i o n 72.2 per cent of the rural and 76.7 per cent of the urban; and for the locat ion c l a s s i f i c a t i o n 70.6 per cent of B . C . , 55.0 per cent of Canada, and 81.8 per cent of Other). This means that the two c l a s s i f i c a t i o n s on the basis of s ize and locat ion of former residence can account for most of the responses of dif ferent indiv iduals without a l ternat ive c l a s s i f i c a t i o n s , and of the 32 variables these 5 were the only s ign i f i can t discr iminat ing ones. TABLE VI.2 NUMBER OF CASES CLASSIFIED INTO: Rural Urban  Rural 13 5 Urban 7 23 B.C. Canada Other B.C. 12 2 3 Canada 4 11 5 Other 0 2 9 Size and Location Combined Another discriminant analysis was run on a c l a s s i f i c a t i o n that was a combination of former s ize and former loca t ion . Since there were too few cases to work with in 4 of the 6 categories, the groups used were only ru ra l -B .C . and urban-Canada. The c l a s s i f i c a t i o n was va l ida ted , but by only the one pa r t i c ipa t ion var iable awareness of neighbourhood organizations. This variable (and the ent i re analysis) had an F ra t io of 8.702 with 1 and 29 degrees of freedom, which gave an F p robab i l i ty of 0.006. When the members of each group were compared to the compo-s i t e score for t he i r group, the c l a s s i f i c a t i o n turned out to be f a i r l y good. Ninety-four per cent of the urban-Canada group members were; placed in the urban-Canada category, while 53.8 per cent of the r u r a l -B.C. group f e l l in the r u r a l - B . C . category. This seems to indicate that a l l urban-Canada group members (but 1) were a l i k e in t he i r knowledge of neighbourhood organizat ions, while r u r a l - B . C . group members were almost evenly s p l i t in the i r awareness. However, there was an underlying; dimension wi th in th i s group that can account for this s p l i t that w i l l show up in the percentage comparisons. B. Tests of the Third Hypothesis by Regression Analysis A regression analysis was run with the same 28 qua l i ty and 4 par-t i c i p a t i o n variables with s ize and locat ion of former residence as independent v a r i a b l e s . 1 The analysis showed that former s ize and former locat ion were s ign i f i can t in predict ing 2 and 3 dependent va r iab les , respect ively . However, there was no overlap in the variables predicted. Size of former residence was related to the pa r t i c ipa t ion variable awareness of neighbourhood organizations and the qual i ty rank 3rd best, while former locat ion was only related to the o r ig ina l service and f a c i l i t y variables parking, senior c i t i zens programs, and street mainten-ance. As each of the regression equations had quite high standard errors for the independent va r i ab le , the analysis may be questioned. Table 3 summarizes for each variable predicted the levels of s ignif icance 2 and r . From this table i t can be seen that on the whole s ize of former residence predicted more of the var ia t ion in the var iables for which i t 3 was s ign i f i can t than did locat ion of former residence. < Former Size Orgawareness 3rd Best TABLE VI.3 REGRESSION ANALYSIS SUMMARY Coeff ic ient F Ratio F Probabi1ity -0.5222 -2.3778 5.7814 7.1657 0.0193 0.0100 0.1117 0.1348 Former Location Parking -1.6697 5.7889 Senior Ci t izens Program -0.8394 4.0009 Street Maintenance -1.3578 4.7903 0.0193 0.0488 0.0320 0.1118 0.0800 0.0943 To summarize the implicat ions or indicat ions of these regression equations, new residents from rural areas were more aware of the e x i s -tence of neighbourhood organizations and chose the more popular answers 4 for the t h i rd best rank than newcomers from urban areas. Newer r e s i -dents from B r i t i s h Columbia seemed to feel that parking, senior c i t i zens programs, and s treet maintenance were much bigger problems than was f e l t by newcomers from other provinces of Canada and from outside 5 Canada. C. Tests of the Third Hypothesis by Percentage and Average Comparisons  The f ive different variables entered as s i gn i f i can t in the d i s -criminant analysis w i l l be discussed f i r s t , and then other in teres t ing relat ionships w i l l be explored. Organization Awareness Variable Since awareness of neighbourhood organizations was s ign i f i can t in the discriminant analyses of both s ize and locat ion of former residence, this var iable w i l l be examined f i r s t . S i ze . Only 57.9 per cent (11 of 19) of the rural newcomers were unaware of any neighbourhood organizat ion, compared to 89.6 per cent (26 of 29) urban newcomers who were unaware. Of those who were aware, 100 per cent of the urban group knew the three major organizations ( i . e . , Ratepayers, Resource Counc i l , and Community Centre), while 37.4 per cent of the rural group mentioned other organizations not considered as main ones in the neighbourhood. Locat ion. When locat ion of former residence was examined separ-ately for both newer groups combined, only 53.0 per cent (9 of 17) of the people coming from B r i t i s h Columbia were unaware of any neighbourhood organizat ion. This figure can be compared to 83.3 per cent (10 of 12) unaware from places outside Canada and 94.7 per cent (18 of 19) unaware from places in Canada other than B . C . ! New residents coming from B r i t i s h Columbia were the only ones mentioning organizations other than the three major ones. Contro l l ing for Length of Residence. Since e a r l i e r explanations showed that awareness increased with increasing length of residence, next s ize and locat ion have been combined, but wi th in each residence length category. Those who did not know of any organization were in proportion to the trends of former locat ion for the two newest residence length groups that were described in Section B of Chapter I I I . What i s more important to investigate i s the background of those 8 people who were aware of the major organizations. Two who knew of the major organizations were newest residents , ( in Vancouver less than 2 years) . One was from a small town in B r i t i s h Columbia, and the other from a large c i t y somewhere in Canada outside of B .C . The s ix others knowing major organizations have a l l l i v e d in Vancouver between 2-1/2 and 6 years. They were mostly from small towns in B .C. o r i g i n a l l y (50 per cent of the 6) . The other three people who knew were evenly divided among urban areas of B . C . , and places outside of Canada, both rural and urban. I t i s s ign i f i can t to note that not a s ingle person who knew of the major organizations and who l i v e d in Vancouver between 2-1/2 and 6 years was from Canada outside of B r i t i s h Columbia! Size Discriminating Variables The next variables to be examined are those that proved s i g n i f i -cant in discr iminat ing between former s ize groups, i . e . , 3rd best rank and pol ice protect ion. Third Best Rank. The difference being looked for in the 3rd best rank variable i s a difference in the frequency of the response mentioned. That i s , one of the groups w i l l have chosen the most popular answers and the other the least popular. Looking at the average rank for each group gives a much clearer demonstration than looking at the per cent answer-ing in each category. The small town group had a mean rank of 7 .1 , while the large c i t y group had a mean rank of 4 .2 . When these figures were compared to the overal l mean of 5.6, the rural group was 1.5 points above the mean, and the urban group was 1.4 points below the overal l mean. People from the bigger c i t i e s almost never mentioned the most popular answer. Pol ice Protec t ion . The other var iable s i gn i f i can t in d iscr imina-i t ing between the c i t y s ize of former places of residence was pol ice protect ion. I t was postulated that reactions toward th is variable would be quite different for different zoning areas. Thus the sample of newer residents was broken down according to mult iple family area and the areas other than mult iple family. A l l rural people l i v i n g in mult iple family zones who ranked pol ice protection f e l t that i t was r e l a t i v e l y good. People from urban areas, on the other hand, placed pol ice protection in the worst problem category 27 per cent of the time. .Average rank can also be used to demonstrate the differences between the groups on th is va r iab le . The average ranks of pol ice protection in the mult iple family area were 2.0 for rural people interviewed and 4.6 for former urban residents . The results i n the 2-family and s ingle family areas were s l i g h t l y d i f ferent . Seventy-five per cent of the former rural and 67 per cent of the former urban residents were neutral about pol ice protection (com-pared to only 36 per cent and 28 per cent in mult iple areas, respective-l y ) . Contrary to the resul ts in the mult iple area, none of the small : town people ranked pol ice protection as good, while only 8 per cent of the big c i t y people ranked th is service as bad! The average ranks, were 7.5 for former rural residents and 4.75 for former urban residents. Thus i t can be seen that former urban residents , no matter what zoning area they l i v e i n , in K i t s i l a n o , rank pol ice protection around 4 .7 . Since there were only 2 ranks making up the average for small town r e s i -dents in nonmultiple areas, the overal l rank average was construed as being more meaningful. This overa l l mean turned out to be 3 .2 , lower than the urban average, and yet above the ent i re sample's mean of 2 .1 . That i s , rural newcomers found pol ice protection to be better than did newcomers from urban areas. Location Discriminating Variables Both of the remaining variables that discriminated among the former locat ion c l a s s i f i c a t i o n s were of the type that pol ice protection was, i . e . , an o r ig ina l service and f a c i l i t y va r iab le . These two var-i ab le s , parking and parks, have been separated by zoning area a l so , be-cause the i r qua l i ty may vary within K i t s i l a n o . Parking. In the mult iple zone 85 per cent of those from B.C. placed parking in the bad categories, compared to only 54 per cent of those from the rest of Canada and 50 per cent of those from outside Canada. The average ranks follow the same order: 8.5 for B . C . , 6.7 for Canada, and 6.0 for other places, compared to the overal l sample mean of 5.1. Results were s l i g h t l y different in the other zoning areas. S i x t y -seven per cent of B . C i t e s , 11 per cent of those from Canada, and 50 per cent of those from outside Canada did not rank th is variable (compared to 7 per cent, 18 per cent, and 0 per cent in mult iple areas). The num-ber of people placing parking in the bad categories went down in each case (33 per cent B . C . , 44 per cent Canada, and 26 per cent other) . The average ranks have, concomitantly, been reduced to 6.0 for B . C . , 5.75 for Canada, and 5.25 for other. A l l these figures are s t i l l higher than the overal l mean of 5.1. They also seem to indicate that parking i s more of a problem in mult iple family areas. This i s probably a true ; ind ica t ion because of the small percentage of onsite parking and the re-cently introduced regulation permitting parking only on one side of the s t reet . However, people from B.C. who have recently moved to Vancouver g s t i l l rate parking much worse than newcomers from other areas. Parks. The f ina l d iscr iminat ing var iable to discuss i s parks. There were not enough people ranking th is variable in the nonmultiple areas to use this d i s t i n c t i o n except for people coming from other pro-vinces of Canada. So, o v e r a l l , 35 per cent of the people from B . C . , 15 per cent of those from elsewhere in Canada, and 50 per cent of those from outside Canada were neutral about the qua l i ty of parks. The aver-age rankings were 1.90 B . C . , 3.83 Canada, and 1.68 other. Breaking down the Canada group, the rank was 2.8 in mult iple areas (close to K i t s i l a n o Beach) and 5.3 in other areas. This seems to be a s i gn i f i can t d i f f e r -ence. I t might be inferred that other areas of Canada have better (or more) parks in s ingle family areas than is true in K i t s i l a n o and in many places outside of Canada. Thus people moving to Vancouver from other parts of Canada have shown different expectations with regard to the amount of parkland there should be in the community. Pa r t i c ipa t ion Variables Next a b r i e f examination w i l l be made of s i gn i f i can t differences between former c i t y sizes and former locations on the remaining p a r t i c i -pations var iab les , K i t s i l a n o awareness, membership, and government meeting attendance. K i t s i l a n o Awareness Var iable . In terms of awareness of the area's name of K i t s i l a n o , those not knowing the name were 11.8 per cent of those from rural areas but 16.7 per cent of those from urban loca-t ions . I t was mostly the urban group that mentioned a l ternat ive names for the area. Looking at the same question from the standpoint of former l oca t ion , 6.7 per cent of the newcomers from B.C. and 12.5 per cent of the newcomers from elsewhere in Canada did not know the name K i t s i l a n o , compared to 30.0 per cent of those from outside Canada. Although the numbers in each case were too small to give s ign i f i can t re-gression r e su l t s , these percentages do seem to indicate meaningful d i f -ferences, espec ia l ly in terms of locat ion of former residence. Membership Var iable : S i ze . When examining the number of voluntary associations belonged to in terms of former s i z e , of those not belonging to any 67 per cent were from urban areas (33 per cent r u r a l ) . Of those belonging to 1 organization 54 per cent were from rural areas and 46 per cent from urban. And of those belonging to more than 1 assoc ia t ion , 67 per cent were from urban and 33 per cent from rural areas. This i s an in teres t ing reversa l . Former urban people belong to e i ther no organiza-tions or more than 1, while people from rural backgrounds dominate the category of people belonging to 1 only . Membership Var iable : Locat ion. Regarding locat ion of former r e s i -dence, of those not belonging to any voluntary assoc ia t ion , 45 per cent were from Canada, 33 per cent from outside Canada, and 22 per cent from B.C. Of those belonging to only one organization 60 per cent were from B . C . , 27 per cent from elsewhere in Canada, and 13 per cent from outside Canada. And f i n a l l y , of those belonging to more than 1 assoc ia t ion , 50 per cent were from Canada, 33 per cent from B . C . , and 17 per cent from other places. Th i s , too, shows some reversa l , with Canadians leading among those who do not belong at a l l and among those who belong to more than 1 organizat ion, while former B .C . residents far outnumber both g other former locations among those who belong to exactly 1 organizat ion. Government Meeting Attendance Var iab le . The f i n a l pa r t i c ipa t ion variable is attendance at government meetings. Although the attendance for newcomers was very low, 3 out of the 4 people who did attend were from rural areas. Looking at these same four people from the former locat ion perspective, ha l f were from B r i t i s h Columbia and ha l f from Canada. 1 0 Quali ty Rank Differences There were a few other qua l i ty rankings which showed seemingly s ign i f i can t differences among the former s ize and locat ion groups, although few of these were as dramatical ly different as the variables discussed e a r l i e r . S i ze . There was a difference of 0.8 between the means of the rural and urban groups on two ranks, worst and 2nd worst. In each case the overal l sample mean f e l l between the means for rural and urban. And in each case the rural group had the higher mean—indicating a tendency to choose more popular answers. The urban group, on the other hand, for these two problem categories, tended to choose things that were not universa l ly regarded as a big problem. Location. In terms of the 3 former locat ion groups' mean rank-ings there were three var iab les , 3rd best, 2nd best, and worst, that exhibited a spread of more than 1.1 between the highest and the lowest. In a l l of these the B .C . group had the highest rank for i t s mean (most popular answers) and the outside of Canada group had the lowest mean. Thus, i t has been shown that there was a trend for rural people and for B r i t i s h Columbia people (who were often the same) to pick more popular responses, while the i r counterparts picked out things not un iversa l ly thought of for the pa r t i cu la r rank. D. Summary In summary, th is chapter applied discriminant ana lys i s , regression analysis and percentage and average comparisons to the data obtained from the questionnaire in order to test the hypothesis that newer r e s i -dents' views on the qual i ty of neighbourhood services and f a c i l i t i e s and the extent of t he i r community pa r t i c ipa t ion vary s i g n i f i c a n t l y according to the s ize and locat ion of t he i r former residence. Size Dealing f i r s t with s i z e , the discriminant analysis proved s i g n i f i -cant and the c l a s s i f i c a t i o n of former residence s ize used was val idated by three var iab les , 3rd best (a qua l i ty rank), pol ice protection (an p r ig ina l service and f a c i l i t y va r i ab l e ) , and awareness of neighbourhood organizations (a pa r t i c ipa t ion va r i ab l e ) . The posterior c l a s s i f i c a t i o n was exce l len t , with over 70 per cent of the cases f a l l i n g on the diagonal. Former s ize was s ign i f i can t in regression analysis in predict ing organ-iza t ion awareness and 3rd best, explaining over 10 per cent of the*var-i a t ion in each. The d i rec t ion of these relat ionships was explored by means of percentage and average comparisons. I t was shown that people from big c i t i e s of over 100,000 knew of the existence of neighbourhood organiza-tions much less often than small town newcomers. The people from big c i t i e s also mentioned less popular answers for 3rd best and found pol ice protection to be much worse in qua l i ty than did newcomers from more rural areas. Two other pa r t i c ipa t ion variables and two other qua l i ty variables showed marked differences, between rural and urban newcomers. Big c i t y newcomers belonged to e i ther no voluntary associations or to more than one, while rural newcomers were most l i k e l y to belong to exactly 1. A l so , 3 of the 4 newcomers who attended government meetings were from small towns. The two qua l i ty rank variables that showed differences in each case had the small town people choosing the more popular answers compared to big c i t y people. Location When differences between the 3 possible locations were examined, the discriminant analysis proved s i g n i f i c a n t , with the c l a s s i f i c a t i o n being val idated by neighbourhood organization awareness and 2 of the o r ig ina l 18 service and f a c i l i t y var iab les , parks and parking. Except, for the group of newcomers from Canada both of the other groups had over 70 per cent of the i r members c l a s s i f i e d pos te r ior ly into the i r o r ig ina l category. In regression ana lys i s , locat ion predicted only o r ig ina l service and f a c i l i t y va r iab les , i . e . , parking, senior c i t i zens programs, and street maintenance. This indicated that s ize was more determining than locat ion in regard to neighbourhood organization awareness. However, the d i rec t ion of the re la t ionships that were s ign i f i can t were a l l the-same. In each case people from B r i t i s h Columbia f e l t that the service or f a c i l i t y was a much bigger problem than was f e l t by newcomers from other provinces of Canada and from outside of Canada. Percentage and average comparisons indicated some trends concern-ing the former locat ion of newcomers. Regarding awareness of neighbour-hood organizat ions, people from B r i t i s h Columbia were much more aware than people from ei ther of the other loca t ions . B . C i t e s also knew the name K i t s i l a n o much more frequently than other newcomers and belonged to a s ingle voluntary associat ion more. However, newcomers from other parts of Canada dominated those who did not belong to any associations or belonged to more than one. The two dis t inguishing service and f a c i l i t y variables were not consistent in the i r d i r e c t i o n . B . C i t e s found parking to be much worse than any of the other groups, while people from other parts of Canada ranked parks much worse than the other two groups. Three of the qua l i ty ranks showed marked differences between the means of the groups. In each case people from B r i t i s h Columbia chose the most popular responses and those from outside Canada chose the most unusual items for each rank. FOOTNOTES Using only data from newcomers for whom information on these two variables was ava i l ab le . 2 Greater than 50 per cent of the value of i t s mean. 3 2 Even so, the r was always less than 15 per cent, so neither were accounting for a major proportion of the variance in any of the var iables . 4 This seems to come about because people have much stronger opin-ions about the best and 2nd best ranks. Thus only on 3rd best can real perceptive differences a r i se . 5 This was a conservative estimate because neutral answers were coded as blank and interpreted by the computer as zero. They thus weighted answers toward the pos i t ive s ide . However, th is ranking has a s l i g h t l y different meaning. A rank of less than 5 meant that the item was considered one of the better fea-tures of the area, while a rank closer to 10 indicates that the feature was considered a big problem. People who did not mention the pa r t i cu la r item at a l l were not included in the rank computation, but were consid-ered by the computer in deriving the overal l sample's mean. Again these neutral answers weight toward the pos i t ive s ide , because they should r ea l ly have been given a value of 5, between the best and worst cate-gories . 7 I t i s quite d i f f i c u l t to in terpret the implicat ions of these , rankings because individual perceptions of pol ice services d i f f e r so con-siderably between different age groups these days. Q Only 1 person ranked parking in nonmultiple zones. When he is averaged with' those from mult iple areas, the average is s t i l l nearly 8.3. g There was no s ign i f i can t difference between the former c i t y sizes in the types of organizations to which people belonged. But more B . C i t e s belonged to the hobby/sport type than any other group. 1 0These were -12 per cent of the B.C. group, and 10 per cent of the Canada group. SUMMARY AND CONCLUSIONS A. Summary The present invest igat ion was designed to test hypotheses r e l a t -ing to differences between newer residents in a c i t y and older residents and to differences within the newcomer group on the basis of s ize and locat ion of the i r former places of residence, to see i f th is information could be used for community planning. These hypotheses were tested by means of a questionnaire o r a l l y administered to a random sample of 108 residents of the K i t s i l a n o area of Vancouver. The respondents had-l i ved in Vancouver for varying amounts of time. The person interviewed who l i ved in Vancouver for the shortest period was in the c i t y for less than a year , while the longest resident had l i ved in the c i t y for 57 years. On the basis of the answers obtained each of the hypotheses was submitted to s t a t i s t i c a l sc ru t iny . In tes t ing each hypothesis, f i r s t discriminant analysis was used to see whether the basic c l a s s i f i c a t i o n was val idated. Secondly, regression analysis was applied to see what variables residence length (or s ize or locat ion of former residence) , could predic t . And f i n a l l y , percentage and average comparisons were made to investigate the actual nature and d i rec t ion of differences. The results of tes t ing each hypothesis w i l l be summarized i n d i v i d u a l l y . Hypothesis I The f i r s t hypothesis was that newer residents ' opinions regarding the qual i ty of t he i r neighbourhood's services and f a c i l i t i e s vary s i g n i f -i can t ly from those of longterm residents. On the basis of the results discussed in Chapter IV, the conclusion must be that th is hypothesis has been supported by the evidence. On the basis of a s i gn i f i can t d iscr im-inant ana lys i s , s i gn i f i can t mult iple regression, and percentage compari-sons, newer residents chose less popular answers for 4 of the 10 rank variables (Best, 4th Best, 5th Best, and Worst) and f e l t that hospital qua l i ty was worse than longterm residents seemed to f e e l . Thus the f i r s t hypothesis i s accurate. That i s , i t seems that there are s ign i f i can t differences between ind iv idua l s ' opinions of neighbourhood services and f a c i l i t i e s depending upon the i r length of residence in the c i t y , despite some s t a t i s t i c a l shortcomings l i k e inexact poster ior c l a s s i f i c a t i o n and f a i r l y high standard errors and low p r e d i c t a b i l i t y in the regression analys is . This hypothesis was not tes t ing for sa t i s fac t ion with the community but rather for d i f fe r ing reactions to the qua l i ty of neighbourhood ser-vices and f a c i l i t i e s . However* the results can be compared with the e a r l i e r studies presented in Chapter I in one respect. A proxy for s a t i s -faction can be derived from people, being interviewed who could not provide 5 best and 5 worst features from a l i s t of 18. In th i s regard, while there was l i t t l e difference in the worst categories, people who had l i ved in Vancouver for less than 2 years could f ind the most features to praise in the i r area. Those in Vancouver 7-21 years could find the least things to pra ise . This substantiates Crothers ' f ind ings , in that residents for less than 7 years and more than 21 years (less than 5 and more than 17 in his case) seemed most s a t i s f i e d , i f more good features mentioned are any i nd i ca t i on . Concomitantly, th is proxy for ; 2 3 4 sa t i s fac t ion contradicts the work of Caplow, Gu l i ck , and Sommer.. However, more relevant to the purpose of th is study, the hypoth-esis of opinion differences has been substantiated. Hypothesis II The second hypothesis was that newer residents ' community p a r t i -cipat ion varies s i g n i f i c a n t l y from that of longterm residents . The re-sul ts in Chapter V demonstrate overwhelming support for th is hypothesis. And the d i rec t ion of the difference was extremely consistent . On the basis of a s ign i f i can t discriminant ana lys i s , a s i gn i f i can t mult iple , regression, and percentage comparisons, newer residents were much less aware of the existence of neighbourhood organizations than longterm , residents . Percentage comparisons also indicated that newest residents were least l i k e l y to be aware of the i r area's popular name and least l i k e l y to belong to any voluntary associat ions. The only residence length category that attended government meetings, and espec ia l ly plan-ning issue oriented ones, to any noticeable extent was that of people in Vancouver between 7 and 21 years. Here again there were some d i f f i c u l t i e s with the more s o p h i s t i -cated s t a t i s t i c a l techniques, e . g . , high standard errors and inexact poster ior c l a s s i f i c a t i o n . Despite these shortcomings, the indicat ions were obvious and consistant with e a r l i e r findings in th is regard (Lee, 6 7 Gul ick , and Zimmer ) . The conclusion must be that newer residents are less l i k e l y to par t ic ipa te in community a c t i v i t i e s than longterm r e s i -dents. The var ia t ion between the residence length groups was s ign i f i can t enough to support the contention of th is second hypothesis. Hypothesis III The th i rd and las t hypothesis was that newer residents ' views on the qua l i ty of neighbourhood services and f a c i l i t i e s and the extent of the i r community pa r t i c ipa t ion vary s i g n i f i c a n t l y according to the s ize and locat ion of t he i r former residence. There seemed to'be enough e v i -dence from the results presented in Chapter VI to support th i s contention. Looking f i r s t at former s i z e , the d i s t i n c t i o n between people coming from c i t i e s over and under 100,000 in population seems j u s t i f i e d . Rural newcomers were much more l i k e l y to par t ic ipate in community a c t i v i t i e s than big c i t y newcomers and. were also more l i k e l y to go along with the majority and pick the most popular answers for the qua l i ty ranks. Among the newcomers, i t was those from urban areas over 100,000 who saw things d i f f e r en t l y . In terms of former l oca t ion , newcomers from B r i t i s h Columbia f e l t that 3 of the service and f a c i l i t y variables were much bigger problems than was f e l t by other newcomers. Former B.C. residents were the most l i k e l y to par t ic ipa te in community a c t i v i t i e s , and they also tended to choose the most popular responses for qua l i ty ranks. I t was the group from outside Canada that saw things d i f fe ren t ly as evidenced by the i r choice of the least frequently mentioned services and f a c i l i t i e s for , each rank. In summary, th is evidence from Chapter VI seems to indica te . tha t there are s ign i f i can t differences within the newcomers ranks on the basis of s ize and locat ion of t he i r former place of residence. People formerly from B r i t i s h Columbia and those from towns under 100,000 had the greatest degree of community p a r t i c i p a t i o n . However, in terms of qua l i ty var iab les , i t seems that s ize differences were greater in d i s -t inguishing the populari ty of the response given to f i l l a pa r t i cu la r rank, while locat ion differences were greater concerning the ranks to which the o r ig ina l service and f a c i l i t y variables were assigned. On the basis of th is analysis there is enough evidence to refute Q the Simmons' b e l i e f in the disappearance of urban-rural differences. 9 • However, the evidence also contradicts Zimmer's f inding that urban migrants tend to enter the a c t i v i t i e s of the community more rap id ly . But the c o n f l i c t i n g results in th is respect may be because th is study; did not separate farm migrants from migrants coming from small towns. Then too, Zimmer seemed to go much below 100,000 in population to con-sider a person an urban migrant. While the results are not s t r i c t l y comparable with B l a c k b u r n ' s 1 0 work e i the r , there seems to be evidence of agreement. That i s , the farther away people had come from the more c r i t i c a l they were concerning the qua l i ty of the i r new neighbourhood's services and f a c i l i t i e s , as evidenced by the i r choice of least popular answers for each rank. Conclusion I t seems from the evidence summarized above that a l l three hypo-theses have been substantiated by the results of th is study as well as by some e a r l i e r research f indings . The second hypothesis was more c l ea r ly demonstrated than, the other two, but in any case indicat ions are that a l l three were correct i n . t h e i r assert ions. B. Suggestions for Further Research Having completed the p i l o t test of the three hypotheses propounded in Chapter I , some suggestions for future research can be given. In the author's opinion, i t would be useful to -direct research in two ways. The f i r s t would be to expand upon the study presented herein using more subjects. This would allow more invest igat ion into differences between s ize and locat ion of newcomers' former places of residence. But more; importantly, i t would allow the researcher to control for the effects of other variables while tes t ing residence length. For example, since i n -come was the main determining var iable with regard to membership in voluntary associa t ions , the ideal would be to separate respondents into the four residence length groups wi th in income categories before running the discriminant ana lys is . I f that were done, the effect of residence length might be g rea t e r . 1 1 : The second d i rec t ion which further research might take would be to obtain responses, not from a l l people in one area, but from newer residents only. With a sample of newcomers one could f ind out more about migration trends—where newcomers are coming from, why, and where they se t t l e in the c i t y . Knowing where they l i v e wi th in one c i t y would also help l a t e r on in applying the r e su l t s , i . e . , in obtaining the opinions and advice of newcomers as part of the planning process. C. Implications for Planning Essen t i a l ly three implicat ions for planning can be derived from this study. The most important i s that newcomers can indeed be of assistance to planners in the planning process. Because th is study re-vealed that newcomers differed s i g n i f i c a n t l y in t he i r views of the c i t y from longterm residents , i t may be>posited that the newcomer sees prob-lems in his new c i t y to which the longterm resident has become immune. I f th i s i s the case, the planner can use the newcomers' previous exper-ience and unencumbered perceptions to gain fresh ins ight into the plan-ning problems posed by his c i t y . I f newcomers can be used, the question arises which newcomers would be of greatest assistance to the planner? The answer to th is ques-t ion leads to the second impl ica t ion for planning to be derived from this study. That i s , the resul ts indicate that the most c r i t i c a l new-comers, and hence perhaps the most useful newcomers for the planner to contact, would be those newcomers who came from c i t i e s of over 100,000, especia l ly those c i t i e s of over 100,000 located outside of Canada. This is because in Hypothesis III i t was shown that the people who picked the least popular services and f a c i l i t i e s for each rank were those from c i t i e s over 100,000 and those from outside of Canada. Assuming newcomers w i l l be used in the planning process, the th i rd impl icat ion for planning of th is study is derived from the author's experience in attempting to e l i c i t the opinions of newcomers. I t i s be-l ieved that in order to best obtain the opinions of newcomers an a l t e r -nate procedure should be substi tuted for the one used in th i s study. Some form of d i rec t contact must be maintained. However, a more open-ended interview s i tua t ion should be substituted for the highly structured 12 questionnaire administered in th is study. The benefits for the planner of an open-ended method would be that the people would respond more f ree ly . And i t i s from seemingly extraneous remarks that the plan-ner may derive his best information. For example, in th i s study the ; author observed that newcomers were unable to mention t h e i r greatest concerns because they were prohibited from doing so by the structure of the questionnaire. In summary, three implicat ions for planning have been presented. F i r s t , newcomers can be of assistance in the planning process. Second, of a l l newcomers those of greatest potential assistance would be new a r r iva l s from c i t i e s larger than 100,000 in population, and from outside of Canada. And f i n a l l y , an a l ternat ive form of e l i c i t i n g the information of newcomers should be used. I l l FOOTNOTES R. J . Crothers, "Factors Related to the Community Index of S a t i s -factor iness ," E k i s t i e s , XXX (August, 1970), p. 109. 2 Theodore Caplow, Sheldon Stryker , and Samuel E. Wallace, The  Urban Ambiance (Totowa, N. J . : Bedminster, 1964), p. 198. 3 John Gul ick , Char les 'E . Bowerman, and Kurt W. Black , "Newcomer Enculturation in the C i t y : Att i tudes and P a r t i c i p a t i o n , " in Urban Growth  Dynamics in a Regional Cluster of C i t i e s , ed. by F. Stuart Chapin, J n and Shi r ley F. Weiss (New York: John Wiley and Sons, 1962), pp. 324, 327. 4 Robert Sommer, Personal Space:, The Behavioral Basis of Design (Englewood C l i f f s , N. J . : P r e n t i c e - H a l l , I nc . , 1969), pp. 89-90. 5 Terence Lee, "Urban Neighborhood as a Socio-Spat ial Schema," Human Rela t ions , XXI (August, 1968), p. 259. c "Newcomer Encul tura t ion ," pp. 342-347. ^Bas i l G. Zimmer, "Par t i c ipa t ion of Migrants in Urban Structures ," in C i t i e s and Society, ed. by Paul K. Hatt and Alber t J . Reiss , J r . (Revised ed . ; New York: The Free Press, 1957), pp. 730-738. 8 James and Robert Simmons, Urban Canada (The Copp Clark Publishing Company, 1969), p. 6. Q "Par t i c ipa t ion of Migrants," pp. 733-738. 1 0 D . W. Blackburn, "Using the Views of Newcomers as an Aid to Com-munity Planning," (paper presented at the Annual Meeting of the South-western Socio logica l Assoc ia t ion , New Orleans, Louis iana, A p r i l 7-9, 1966), p. 132. ^ B a s i l G. Zimmer's work is a good example of t h i s . See " P a r t i c i -pation of Migrants," pp. 730-738. 12 Of course the highly structured questionnaire was necessary in this study to es tabl ish that differences ex is ted . BIBLIOGRAPHY BIBLIOGRAPHY Axel rod, Morr i s . "Urban Structure and Social P a r t i c i p a t i o n . " C i t i e s  and Society. Edited by Paul K. Hatt and Albert J . Reiss , J r . Rev-ised ed. New York: The Free Press, 1957. Backstrom, Charles Herbert and Gerald D. Hursh. Survey Research. Chicago: Northwestern Univers i ty Press, 1963. Blackburn, D. W. "Using the Views of Newcomers as an Aid to Community Planning." Paper presented at the Annual Meeting of the Southwestern Sociologica l Assoc ia t ion , New Orleans, Louis iana , A p r i l 7-9, 1966, Caplow, Theodore; Sheldon Stryker; and Samuel E. Wallace. The Urban  Ambiance. Totowa, N. J . : Bedminster, 1964. Cherukupalle, Nirmala devi . "C la s s i f i c a t i on Techniques in Planning Ana lys i s . " Socio-Economic Planning Science, IV (September, 1970), 395-405. 1 . "Relationship Between At t r ibu tes : Simple and Mul t ip le Regres-sion and Cor re la t ion . " (Mimeographed) Crothers, R. J . "Factors Related to the Community Index of S a t i s f a c t o r i -ness." E k i s t i c s , XXX (August, 1970), 107-109. Economic Council of Canada. Fourth Annual Review: The Canadian Economy  From the 1960's to the 1970's. Ottawa: Queen's P r i n t e r , 1967. Gul ick , John; Charles E. Bowerman; and Kurt W. Black. "Newcomer Encul-turat ion in the C i t y : Att i tudes and P a r t i c i p a t i o n . " Urban Growth  Dynamics in a Regional Cluster of C i t i e s . Edited by F. Stuart Chapin, J r . and Shi r ley F. Weiss. New York: Wiley, 1962. K e l l e r , Suzanne. The Urban Neighborhood. New York: Random House, 1968. Lamanna, Richard A. "Value Consensus Among Urban Residents," Journal of the American Ins t i tu te of Planners, XXX (November, 1964), 317-323. Lansing, John B. and Les l i e K i s h . "Family L i f e Cycle as an Independent Var iab le . " American Sociological Review, XXII (October, 1957) 512-519. Lansing, John B. and Robert W. Marans. "Evaluation of Neighborhood Qua l i ty . " Journal of the American Ins t i tu te of Planners, XXXV (May, 1969), 195-199. Lee, Terence. "Urban Neighborhood as a Socio-Spat ial Schema." Human  Relat ions , XXI (August, 1968), 241-267. Lowenthal, David. "Geography, Experience, and Imagination; Towards a Geographical Epistemology." Annals of the Associat ion of American  Geographers, LI (September, 1961), 241-260. Milgram, Stanley. "The Experience of L iv ing in C i t i e s . " E k i s t i c s , XXX (August, 1970), 145-150. Re t t ig , Solomon. "Mult iple Discriminant Ana lys i s . " American Sociolog- i c a l Review, XXIX (June,-1964), 398-402. Ross i , Peter H. Why People Move. Glencoe, I l l i n o i s : The Free Press, 1955. Sanoff, Henry. "Social Perception of the Ecological Neighborhood." E k i s t i c s , XXX (August, 1970), 130-132. Saroff, Jerome R. and Alberta Z. Levi tan. Survey Manual For Comprehen-sive Urban Planning; the Use of Opinion Surveys and Sampling Tech- niques in the Planning Process. Alaska College Ins t i tu te of S o c i a l , Economic, and Government Research, SEG Report Number 19, 1969. Simmons, James and Robert Simmons. Urban Canada. The Copp Clark Publ i sh-ing Company, 1969. Sommer, Robert. Personal Space: The Behavioral Basis of Design. Englewood C l i f f s , N. J . : Prentice-Hal 1, Inc . , 1969. Sonnenfeld, Joseph. "Variable Values in Space Landscape: An Inquiry into the Nature of Environmental Necessi ty." Journal of Social Issues, XXII (October, 1966), 71-82. Univers i ty of L o u i s v i l l e Urban Studies Center. Report to FalIs of the  Ohio Metropolitan Counci1 of Governments on Urban Decisions Project , May, 1968. Wilson, Robert L . " L i v a b i l i t y of the C i t y : At t i tude and Urban Develop-ment." Urban Growth Dynamics in a Regional Cluster of C i t i e s . Edited by F. Stuart Chapin, J r . and Shi r ley F. Weiss. New York: John Wiley and Sons, 1962. Zimmer, Bas i l G. "Par t i c ipa t ion of Migrants in Urban Structures ." C i t i e s and Society. Edited by Paul K. Hatt and Alber t J . Reiss , J r . Revised ed. New York: The Free Press, 1957. APPENDICES APPENDIX A APPENDIX A QUESTIONNAIRE ' About how long have you l i v e d in Vancouver? {For those here less than two years:j What was the main reason you had for moving here? 63 No answer 7 Push factors 6 Combination of both 32 Pu l l factors 108 Could you please t e l l me where you l i ved in the two years before you came to Vancouver? 59 No answer 19 Population under 100,000 30 Population over 100,000 108 59 No answer 17 B r i t i s h Columbia 20 Other province in Canada 12 Other country 108 Have you thought ser ious ly of moving out of Vancouver during the l a s t year? No answer 29 Yes 79 No 108 * Not analyzed in this thesis What do you l i k e most about l i v i n g in Vancouver? 5 Nothing 3 Fami l i a r i t y 4 Economics 5 Other Miscellaneous 9 People 15 A c t i v i t ies 26 Climate 41 Natural Beauty 108 Now, out of this l i s t of 18 items could you please pick out the 5_ things that are the least problem in your area and the 5 things that you see as the biggest problem in your area. In other words, which things are you most s a t i s f i e d with and least s a t i s f i e d with? Now would you go through your p i l e of the 5 things that present the least problem and order them, so that the card on top is the thing you are happiest wi th . Now would you go through your p i l e of the 5 things that present the b ig -gest problem and order them, so that the card on top i s the thing you are most d i s s a t i s f i e d wi th . 1 A i r Po l lu t ion A Parking 2 Back Lanes B Parks 3 Day Care C Pol ice Protection 4 Fire Protection D Public Schools 5 Hospitals E Recreation F a c i l i t i e s 6 Housing Conditions F Senior Ci t izens Programs 7 Landscaping of Gardens G Street Light ing 8 Library Services H Street Maintenance 9 Location of Shopping I T ra f f i c Control What i s the least problem in your area: that thing that you are happiest with? 2 Nothing 2 A i r Po l lu t ion 6 Fire Protection 4 Hospitals 4 Landscaping of Private Gardens 10 Library Services 30 Location of Major Shopping 3 Parking 28 Parks 9 Pol ice Protection 2 Public Schools 5 Recreation F a c i l i t i e s 3 Street Light ing 108 What i s the biggest problem in your area: that thing that you are most d i s s a t i s f i e d with? 5 Nothing 34 A i r Po l lu t ion 6 Back Lanes 2 Fire Protection 4 Hospitals 10 Housing Conditions 1 Landscaping of Private Gardens 1 Library Services 1 Location of Major Shopping 19 Parking 1 Pol ice Protection 4 Public Schools 4 Recreation F a c i l i t i e s 2 Senior Ci t izens Programs 2 Street Light ing 5 Street Maintenance 7 Tra f f i c Control 108 Now I ' d l i k e to ask you some questions about your neighbourhood and your a c t i v i t i e s . Would you please give me the boundaries of what you consider your neigh-bourhood? I t would help i f you would draw an X where your home i s . Does i t have a popular name? — No answer 85 Yes (Ki t s ) 14 No 9 Yes (but incorrect name) 2 No Answer 53 No 8 Miscellaneous 45 Major Ones 108 Major Ones Mentioned: Ratepayers * 89 No 19 Yes 108 Resource Counci1 106 No 2 Yes 108 Community Centre 83 No 25 Yes 108 j l f yes:i Have you ever attended one of the i r meetings? 65 No answer 8 Yes 35 No 108 Did you attend any o f f i c i a l meetings or open hearings of government bodies th is year? No answer 12 Yes 96 No I f yes: Please t e l l me the i r names & the number of times you attended? Type 2 Special-Medical 4 Special-Planning 6 Gene ra l -Po l i t i ca l 12 Number of Times Attended 3 1 5 2 2 3 2 4 12 Did you take part in the discussion during the meetings or hearings? 96 No answer 4 Yes 8 No 108 * Do you believe that once you have elected a person to represent you in local government that he w i l l represent your interests and should not be bothered by people t ry ing to t e l l him how to act? 8 No answer 38 Yes 62 No 108 Do you belong to any organizations—that i s , groups l i k e the PTA or c i v i c groups, c lubs , lodges, church groups, veterans organizat ions, and t'ie l i ke? Any other? 59 None 8 Fra ternal /Socia l 8 Professional 8 Public Af fa i r s 3 Public Service 2 Business 12 Church/Religious 2 Veteran/Pat r io t ic 3 Cul tura l /Aes thet ic 20 Hobby/Sports 125 Number Belonged To 59 None 30 1 14 2 3 3 1 4 1 5 108 I have jus t a few more questions to ask about you and your family. What i s your current marital status? No answer Single Widowed Divorced Separated Engaged to be married Married What i s your age? ( I f no answer, estimate) No answer 19-24 25-30 31-40 41-50 51-60 61-65 65+ What i s the name of the l a s t school you attended? What was the l a s t grade you completed in school? No answer 0-7 (Some elementary) 8 9-11 12 (Elementary graduation) (Some high school) (High school graduation) 13-15 (Some univers i ty) 16 (Universi ty graduation) 16+ (Graduate or professional t ra in ing) Could you please t e l l me what the main bread-winner in your family does for a l i v i n g ? Do you have any chi ldren l i v i n g at home? No—0 1 2 e tc . I f yes:[ What schools do they attend? No answer No answer Private Elementary Public High school Do you own your own home or are you renting i t ? Here i s a card showing different income groups. Just give me the l e t t e r of the group your family i s i n . (Your average annual salary) No answer A Under $3000 B $3000 to $4,999 C $5000 to $6,999 D $7000 to $9,999 E $10,000 to $14,999 F $15,000 and above No answer Univers i ty or Vocational No answer Renting Own (Not to be asked o r a l l y ) Sex 57 Male 51 Female High-r ise apartment Low-rise apartment Suite in a house Duplex Single family detached house I 'd l i k e to thank you for your assistance to our survey, and espec ia l ly for sparing me your time during th is busy season. Before I leave, do you have any other comments for me about anything? Are there any questions I can answer? Thank you again. COMMENTS Area type l=Multiple 2=2-Family 3=Single Family Interviewer 1- Myra Gall ins 2- Peter Fisher 3- Elizabeth Darragh Interview Number APPENDIX B APPENDIX C APPENDIX C CENSUS COMPARISON TO DETERMINE THE REPRESENTATIVENESS OF THIS STUDY'S SAMPLE Age Under 24 25-64 65+ Mari ta l Status Single Married Widowed Number of Children None 1-2 3-4 5+ This Sample 28.7% 63.9 7.4 100.0% 41.7% 51.8 6.5 100.0% 73.2% 20.4 5.5 0.9 (Tracts 15,16,22) K i t s i l a n o 22.2% 58.2 19.5 99.9% 33.2% 53.5 13.3 100.0% 55.9% 35.8 6.8 1 .5 100.0% 100.0% Homeownership Owner 26.8% 20.0% Tenant 73.2 80.0 100.0% 100.0% Dwelling Unit Type Single Family, Attached 8.3% 4.8% Single Family, Detached - m 32.4 16.4 Apartments & Flats 5.9.3 78.8 100.0% 100.0% In this appendix the sample of 108 respondents was compared with data from Census Tracts 15, 16, and 22 from the 1966 Census of Canada on f ive population cha rac t e r i s t i c s . Many assumptions were made at the s ta r t to overcome some of the d i f f i c u l t i e s in making th is comparison. For example, i t was assumed that the comparison would be f a i r l y accurate despite noncoterminous boundaries, and despite the evolution of the K i t s i l ano area during the four years since the 1966 Census. Given these assumptions, most of the differences between th is study's sample and the population of K i t s i l a n o can be accounted for by the way that the sample was drawn. For example, since only private dwellings were included, by de f in i t i on rest homes, where many older people l i v e , were excluded. Hence, i t can be concluded that there are no exceptional ly large differences (unaccounted for by sample select ion) that would make th is sample unrepre-sentative or unre l iab le . APPENDIX D APPENDIX D REGRESSION ANALYSIS RESULTS Dependent Variable Independent Variable Coeff. F 1 Ratio F Prob 2 r Quali ty Variables * Pol lu t ion Zoning -1.5222 11 .7793 0.0010 0 .1000 Back Lanes Ownership 1.6392 4 .4681 0.0349 0 .0404' Day Care Ownership Dwelling type -1.9200 0.5364 10 6 .2202 .6030 0.0020 0.0112 0 .0977 Fire Protect . Education -0.4182 6 .9214 0.0095 0 .0613 Hospitals * Res. Length North-South -0.6233 -0.9979 4 9 .6376 .1385 0.0317 0.0033 0 .1321 Housing Cond. Occupation Dwelling Type -0.2115 -0.7730 4 8 .2658 .1132 0.0391 0.0053 0 .1131 Gardens — Library Age Dwelling Type -0.4499 0.4787 8 4 .1522 .9956 0.0052 0.0261 0 .0809 Shopping — Parking Children -0.9890 7 .9969 0.0056 0 .0701 Parks — Pol ice --Public Schools : Education 0.4814 4 .8422 0.0283 0 .0437 Glossary of Terms page 133. Dependent Independent Variable Variable Coeff. F Ratio F Prob r_ Quali ty Variables (Cont 'd.) Ownership -1.4235 5 .9609 0.0156 Recreation 0 .0909 EastrWest -1.3053 4 .1851 0.0409 Sr . Ci t izens North-South -0.7086 5 .5791 0.0190 0 .0500 Street L igh t . --Street Maint. — Traf f i c Cont. Ownership • -1.8247 6 .0283 0.0150 0 .0538 Par t i c ipa t ion Variables : (Raw Scores) East-West -0.3189 6 .0163 0.0151 KitsAwareness 0 . 1480 North-South 0.1713 8 .6237 0.0042 Orgawareness Res. Length 0.3380 18 .3342 0.0001 0 .1475 Govt. Attend Education -0.0576 7 .0489 0.0086 0 .0624 Membership Income 0.1350 6 .2496 0.0134 0 .0557 Par t i c ipa t ion Variables : (Factor Scores) KitsAwareness Factor 1 0.2346 6 .1628 0.014Q 0 .0549 Factor 2 0.2460 7 .6559 0.0066 Orgawareness 0. .2054 Res. Length 0.3836 19 .5145 0.0000 Govt. Attend Factor 3 -0.2035 4 .5524 0,0333 0 .0412 Membership Factor 2 0.1853 6 . 1368 0.0142 0 .0547 Dependent Variable Po l lu t ion Back Lanes Day Care Fire Protect . Hospitals Housing Cond. Gardens Library Shopping Parking Parks Pol ice Public Schools Recreation Sr . Ci t izens Street L igh t . Street Maint. Tra f f i c Cont. APPENDIX D GLOSSARY OF TERMS = A i r Po l lu t ion = Back Lanes = Day Care Centres = Fire Protection = Hospitals = Housing Conditions = Landscaping of Priyate Gardeps = Library Services = Location of Major Shopping = Parking = Parks r- Pol ice Protection ? Publ ic Schools s= Recreation F a c i l i t i e s = Senior Ci t izens Programs - Street Light ing = Street Maintenance = Tra f f i c Control Dependent Variable (Cont 'd.) KitsAwareness Orgawareness Govt. Attend Membership K i t s i l a n o Awareness Neighbourhood Organization Aware-ness Government Attendance Membership Number Independent Variable Res. Length North-South East-West Residence Length North-South Axis East-West Axis APPENDIX E APPENDIX E MEMBERSHIP BREAKDOWN BY RESIDENCE LENGTH Type of Organization 0-2 2.5-6 7-21 22-60 None 17 12 18 11 Fraternal /Socia l 2 0 3 3 Professional 1 4 1 2 Public Affa i r s 1 3 . 3 1 Publ ic Service 1 1 1 0 Business 0 0 0 2 Church/Religious 1 1 3 7 Veteran/Patri o t i c 0 0 0 2 Cul tura l /Aes the t ic 0 0 2 1 Hobby/Sports 4 8 4 4 APPENDIX F LESSONS LEARNED IN THE PROCESS OF INTERVIEWING Many lessons can be learned from this attempt at cooperation from people in the i r homes. Some of these problems cannot be overcome, but having an awareness of the p o s s i b i l i t y of t he i r occurrence may a l l ev i a t e some frustrat ions in others about to attempt something s i m i l a r . Some of the phobias of potential respondents probably w i l l never be overcome and may be encountered at any time. These refer to people who demonstrate feelings against the un ive r s i t y , young people, evening v i s i t o r s , and ca l l e r s of any kind. People who have the l a s t two types of phobias would probably par t ic ipa te by mail or over the telephone but w i l l not l e t anyone into the i r homes. Another f rus t ra t ing experience i s when a potential respondent says that he cannot see you at the moment but w i l l d e f i n i t e l y see you i f you w i l l jus t come back Friday night or Tuesday at 1 P . M . , and then he is not at home when you make a special t r i p to meet the time he_ set as being convenient for him. Occasionally a person w i l l make a def in i te appointment and seem l i k e l y to cooperate i f you jus t return at his con-venience. Then when you do return he absolutely refuses to see you at a l l and obviously never had any intent ion of cooperating. Apparently this type of person does not believe that you r ea l l y w i l l return and feels that this i s a method of refusing without having to say "no." Something that seemed to be a problem at f i r s t turned out to be an advantage during the course of th i s study. The interviewers at f i r s t had d i f f i c u l t i e s obtaining entry into apartments that had locked doors and a buzzer system. However, l a t e r i t was found that th is allowed the person inside to hear who was c a l l i n g and the nature of the interview; The usual opening l ine became "He l lo , I am here about that l e t t e r you re-ceived from the un ivers i ty l a s t Thursday." I t seemed in the long run-that admittance was gained much more readi ly th is way than by merely knocking on the person's door. The only problem came when people claimed that they never received the l e t t e r (true for about 10 per cent of the units contacted). An addit ional problem encountered in the course of conducting th is study was that the Christmas and New Year's Holiday Season is not a good time to t ry t o . f i n d people at home or to have them give you the i r time. Considering the time of year when most of the interviews were carr ied out, the response rate was higher than would have been expected. Time became one of the biggest d i f f i c u l t i e s in conducting th is study. There were four types of people who took much longer than average in answering the questions. Some people took extra time because they were being very del iberate and thinking quite careful ly about the cor-rectness of each answer. The second type of slow respondent was the per-son who was putting on a show to entertain others present during the . interview. To overcome th is problem one should t ry to interview an individual alone in order to get his undivided a t tent ion . The th i rd type of slow person was one who was lonely and jus t glad to have someone present with whom to t a l k . In th is case, without being rude the in t e r -viewer should t ry to be patient but to leave after a reasonable amount of time. The f ina l type of slow respondent encountered was a person who was jus t t o t a l l y inconsiderate. For example, in one s i tua t ion the interviewer was to ld that the interview would have to wait u n t i l the person had tea, and the interview ended up taking over an hour because the tea had to be completely prepared and eaten before the respondent would answer any questions. This occurrence seems quite rare and also d i f f i c u l t to pred ic t , but should be avoided at a l l costs . The f ina l lesson to mention, and one that has already been dis^ cussed by many researchers before, i s the importance of carefu l ly trained and interested interviewers. In t ra in ing one cannot assume anything but must mention everything e x p l i c i t l y and more than once. One should also observe the t ra in ing interviewer during more than one dry-run administra-t ion of the questions to be asked. In observing note how he codes pre-coded questions, whether he probes,when answers are unclear ( e . g . , occupation--Alberta Wheat Pool ) , whether his answers are readable ( e . g . , are his l ' s confusable with his 7 ' s ? ) , whether he uses the same i n f l e c -t ion on a question at each reading, whether he adds any comments between questions that might "di rect" answers, e tc . The main consideration in h i r ing an interviewer should be his i n -terest in the project and wi l l ingness to make the necessary time commit-ment. Desire for the salary being paid is not enough, espec ia l ly when time is short and you are counting on your interviewer to be working with his best e f fo r t . R e l i a b i l i t y of the interviewer becomes c r u c i a l , when le t te r s of n o t i f i c a t i o n include the name of the interviewer. This i s because potential respondents might not understand the need for a subs t i tu t ion , so be wary of the entire project and thereby lose c o n f i -dence in the sponsoring organizat ion. In summary, people contemplating a s i m i l a r interview s i tua t ion should be aware of (1) phobias and quirks of respondents that may cause refusals to be interviewed, (2) the best ways pf gaining admittance, (3) times of the year that are d i f f i c u l t for potential respondents, (4) reasons why respondents may take too long in answering, and (5) the importance of careful ly trained and interested interviewers . 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0102201/manifest

Comment

Related Items