|\ FOLLOW-UP STUDY IN A TEMPORARY SHELTER FOR THE HOMELESS: A LOOK AT QUALITY OF LIFE AND SOCIAL SUPPORTS by Andrea Lynn SoTarz Department of PsychoTogy Michigan State University A DISSERTATION Submitted to Michigan State University in partiaT fulfiTTment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychoiogy 1986 ."t TV:\‘«);:.3: Copyright by he _ ANDREA LYNN SOLARZ 1 1986 E I 49:?» Lyye ”9” .‘1' ABSTRACT FOLLOW—UP STUDY IN A TEMPORARY SHELTER FOR THE HOMELESS: A LOOK AT QUALITY OF LIFE AND SOCIAL SUPPORTS by Andrea Lynn Solarz The current study examined one group of the homeless; guests of a temporary shelter agency. The study had two major aims. The first aim was to gather descriptive information about the homeless population. Through interviews, information was gathered on study participants' background and demographic characteristics, psychiatric morbidity, social support systems, and perceptions about their quality of life. From these data, a correlational analysis using Tryon and Bailey's method (BCTRY) identified four defining clusters: psychiatric history, transiency, criminal behavior, and criminal victimization. These produced a total of seven meaningful O-Types. The second aim of this study was to examine whether rates of return for scheduled follow-up appointments could be increased by the use of different research procedures. Using a 2 x 2 factorial design, comparisons were made between the use of two types of payment (cash or material goods) and between the use of two types of appointment cards (permanent or regular). Participants were randomly assigned to one of the four conditions (n = 30 per cell). A total of 23.3% of the participants returned for their scheduled follow-up interview. A Chi- square analysis with return for follow—up as the dependent variable and type of payment and type of appointment card as the two independent variables failed to detect any statistically significant differences in return rates. ':.:j r I" 3" "'J 9'“ . .' ‘1‘! ma? ~. , 5H 7‘ 7" ._i2W(_.-; A total of 125 shelter guests were interviewed as part of this study. Simple random sampling methods were used. Participation was voluntary. It was anticipated that this group of homeless persons would have poor quality of life overall. To a certain extent, these expectations were substantiated. Participants were frequent victims of crime, suffered extreme financial hardship, were unemployed, and had frequent contacts with physicians. Many also had a history of residential instability. The O-Typing analysis in this study clearly revealed a number of distinct subgroups. These results tended to refute many of the current stereotypes of the homeless. This group was clearly not primarily a deinstitutionalized population. Few, if any, were homeless by choice, and the majority were not particularly transient, but had generally lived in the same city for a number of years. L's!) e H n9 3 Nut 9*?" 'n‘ 9,): 3??! To (“Two TOOQ x 3 n OJ .0 ‘- . m D J in .0 1 OJ 1: .n. :. riL j'i Vgn‘oz-yaq Jnoiarv-i ,I I.“ . - ". .- will"? ‘flfllflaltrv; ,. .._.r .I .-5 flowers e: ._ . C. . T “MID I an“: disem- . with I. Dedicated to .1D‘ Tamara Adriana O'Connell - whose birth coincided with the completion of this work ayj;ff " 2; wt?“ ‘. «L its swabs . the ir em“! ‘9‘ . 5 ,.1 ACKNOWLEDGMENTS This large undertaking would not have been possible without the help of others.» While I cannot mention all of them here, there are a few who I would especially like to thank now. My Dissertation Committee deserves special mention. Dr. Robin Redner, the Chair, has provided me with continuous support, both academic and emotional, since the incep- tion of this project. Her feedback has been invaluable, and I am grate— ful to her for pushing me as hard as she did. Dr. Carol Mowbray has helped me in two roles — as a member of my Committee and as my boss at the Michigan Department of Mental Health. This project simply would not have taken place without the opportunities she provided me to conduct research on the homeless. The original idea to conduct this study emerged from conversations with her about this important issue. In addition, she has supported me in obtaining dissertation funding from the Department of Mental Health. I clearly owe her a lot. Dr. Anne Bogat provided me with invaluable information on measurement issues in social support. Her interest in the project and always incisive feed- back are very much appreciated. Dr. Neal Schmitt, as always, was avail- able to answer any and all questions about statistical analysis. I would also like to express my appreciation to the dedicated staff of COTS Shelter. Their cooperation and interest in the project were imperative for its success. Their acceptance of me and my interviewers made the job a lot easier, and we always looked forward to spending time with them. In particular, I mention Ms. Jarrie Tent, the Executive vi yaw? «Win ;,"-5)hnu camel smT Tilv- 9'1.‘ .' IQ RIM .‘l'n ng "M ?‘.‘I') .AT.H .rvs'nai'ufisqoa: wiam‘ .‘-~F.‘I9fl2 2T0?) To "file‘mflm ‘93 - ~31; - .. r J‘ . o...,.‘.. Director of the shelter. Her belief that this kind of work is important made it possible for the research to be conducted at COTS. The data collection would not have gone as smoothly without the help of my student interviewers. Special thanks go to Sue Dupuis for her good companionship and insight. My many friends provided me with support that kept me going even through the most difficult stages of this project. First and foremost I thank my Disseration Support Group: Leah Gensheimer, Julie Parisian, Mary Sullivan, and Sara Woodkraft. Our weekly meetings kept me going, kept me realistic, and gave me a needed opportunity to talk about my work. If I leave the group now, as I must, I leave it only in body, and not in spirit. I also thank Joe Bornstein for continuing support, his many phone calls, and emergency use of his computer. Thanks to Brian Mavis for leading me through the intricacies of BCTRY, and to do Weth for being such a great weekend study buddy. As always, the love of my family has been a continuous source of support. I'm sure they are just as glad as I am that this work is finished! This research was supported through dissertation grants from the Michigan Department of Mental Health and the Urban Affairs Program at Michigan State University. This funding was an immense help, as the project was quite costly. Appreciation also goes to Dr. Bonnie Morrison of the Urban Affairs Program for her continuing interest and support. Finally, I thank with love the homeless men and women who shared their lives with me during this study. They showed me a side of the world that I had not known before. The lessons that I learned from them will be with me always. ‘ u. . . , .’.__ 3%} v‘n.‘ .... ,I'i Rik? - 1' " '.' . ‘ I . - ' f r .,‘ . “' emanate... *‘. - iirin‘ '9‘ ‘ . .. - _ - * ""rw vx'l‘ m..- u. :mo’n: man no cloo‘w new)!!!” ‘0‘ aways-ems”: m up hr" -,,-.-. -Iifl?xl‘IflWlT0‘l boon ion cm X" If” fui'émufl-5fl3 ave; n3iu innd ‘; 7% :.' ‘tiis ,. l4 TABLE OF CONTENTS LIST OF TABLES .......................... LIST OF FIGURES .......................... CHAPTER I. INTRODUCTION ....................... Part I - Identifying the Homeless ............ Framework 1 - Demographic Characteristics ....... Mental Illness and the Homeless .......... Women Among the Homeless .............. Age of the Homeless ................ General Criticisms of Previous Research ...... Framework 2 - Residential Patterns of Homelessness Framework 3 - Causes of Homelessness ......... Global Reasons for Homelessness .......... Specific Reasons for Homelessness ......... Framework 4 — Subjective Assessments ......... of Homelessness Quality of Life of the Homeless .......... Social Support Systems and the Homeless ...... Summary - Part I ................... Part II — Follow—Up with the Homeless .......... Review of Follow-Up Research ............. Research Objectives ................... METHOD .......................... Setting . ........................ Subjects ......................... Non-Participants ................... Research Design and Procedures .............. Data Collection .................... ' viii 45 48 ‘ 351”, , . _ men? 10 rzu "IL? ICU . ‘ i -i..‘l {magi-i {9.3 III. MEASUREMENT INSTRUMENTS . . . . . . . . . . . . . . . . . 53 Measure Development ................... 53 Quality of Life ..................... 57 Subjective Quality of Life Ratings .......... 57 Objective Quality of Life . . ............. 62 Psychiatric Morbidity .................. 64 SCL-lO ........................ 64 Psychiatric History .................. 67 Social Support ...................... 68 Subjective Social Support ............... 68 Objective Social Support ............... 70 Return for Follow-Up Measures .............. 72 Experiences at COTS Shelter .............. 72 Satisfaction with COTS Shelter ............ 72 Satisfaction with the Interviewer ........... 74 Reliability of Measures ................. 74 DESCRIPTIVE RESULTS ................... 77 Housing Quality of Life ................. 77 Objective Housing Quality of Life ........... 77 Shelter Use History ................ 77 Recent Residential History ............. 79 Subjective Housing Quality of Life .......... 83 Finances and Employment Quality of Life ......... 86 Objective Finances and Employment ........... 86 Quality of Life Income ....................... 86 Employment ..................... 90 Subjective Finances and Employment Quality ...... 90 of Life Safety Quality of Life .................. 92 Objective Safety Quality of Life ........... 92 Subjective Safety Quality of Life .......... 94 Self Quality of Life ...... » ............. 94 Physical Health .................... 94 General Health ................... 94 Alcohol Use .................... 101 Drug Use ...................... 101 Mental Health ..................... 103 Objective Mental Health Variables ......... 103 Subjective Mental Health Variables ......... 106 Subjective Self Quality of Life ............ 106 ix .idaihobm W.“ . 9M3 30 ”3'50 '. | . fvgbuu 9V71”m . i: TL; (’4'f ’ _ (”‘0 99'3”” - Mia physician“ 0! - ll" Social Relations Quality of Life ............. 109 Objective Measures of Social Support ......... 109 Social Support Networks .............. 109 Family Relationships ................ 118 Community Involvement ............... 120 Subjective Measures of Social Support ......... 120 Subjective Characteristics of Network Members . . . 120 Subjective Social Relations Quality of Life . . . . 121 Global Quality of Life .................. 126 Typological Analysis ................... 130 V. FOLLOW-UP RESULTS .................... 157 Analysis of Hypotheses One and Two ............ 157 Analysis of Hypothesis Three ............... 158 Satisfaction with COTS Shelter ............ 158 Satisfaction with the Interview . . . . . . . . . . . . 160 Behavioral Infractions at COTS Shelter ........ 160 Initial versus Follow-Up Comparisons ........... 163 Experiences After Leaving the Shelter .......... 163 VI. DISCUSSION ........................ 171 Quality of Life ...................... 171 Objective Quality of Life ............... 171 Subjective Quality of Life ............... 174 Social Support ...................... 175 Existence of Social Support ............... 175 Structure of Social Support .............. 176 Functional Content of Social Support .......... 178 Mental Health Status ................... 180 Typological Analysis ................... 182 Follow-Up Interview Results. ...... . ......... 184 Improving Follow-Up Through the InterView Process . . . 186 Improving Follow-Up Through the Choice of a Research Site .............. 187 Improving Follow-Up Through Tracking Procedures . . . . 188 Generalizability of Results ................ 191 Conclusions ........................ 192 FOOTNOTES ............................. 195 x ”9.21.576." ”0F "H b .ztmimono i’l '_ :néamvfom': {'50 , . .' 7.; acumen-SAT: 2:17; wn‘ ev'ms’cd . " .-.'.-_.. ‘5', ’ iihgfwa .1 ;_3 '~Y".“ .LJ, ”-1. - ?’ 1929.830 we” gas , ,ioerngtheHomeless.......'....201 ,ci'pa'u‘t Agreement} w -'Haterial Goods Checklist -. ' Interview Measurement Instrunents U" . n .- .‘ ‘ WI . Einzzaesponse 7 Cards :' Fiz. -COTS' Record Data Collection Form Xenon am you». wvuflwm mmfi‘ox Much” is; 0' .Erdrfl “haul. Table I—l CON LIST OF TABLES Characteristics of Non-Participants versus Participants . . Reasons for Non-Participation ............... Summary of Measures .................... Life Domains Quality of Life Scales ........ .. . . . SCL-IO Subscores ..................... Social Support Quality of Life Scale ........... Shelter Satisfaction Scale ................ Attitude Toward Interview Scale .............. Shelter Use History .................... Residential Mobility ................... Number of Times Homeless in the Past ........... Recent Residential History ................ Reasons for Leaving Last Place Stayed ........... Correlations Between Housing QOL Scale and Objective Housing Variables ............... Sources of Income in the Past Six Months .......... Largest Source of Income During the Previous Month ..... Amount of Income During Past Year ............. Work History ....................... Correlations Between Finances and Employment QOL Scale and Objective Work and Finances Variables xii 44 46 54 59 65 66 69 73 75 78 80 81 82 84 85 87 88 89 91 93 'iL-O ;: 1 - .x, ' -- ...mi':’:_¢§‘i 8'3; Table l 21- Robbery/Mugging Victimizations During Past Six Months . . . 95 22. Assault/Beating Victimizations During Past Six Months . . . 96 23. Burglary/Theft Victimizations During Past Six Months . . . 97 24. Correlations Between Safety QOL Scale and Victimization Variables ................ 98 25. General Health Status ................... 99 26. Number of Physician Visits During Past Year ........ 100 27. Main Reason for not Following Prescription ........ 101 28. Frequency of Alcohol Use During Past Month ........ 102 29. Frequency of Marijuana Use During Past Month ....... 102 30. Psychiatric Hospitalization History ............ 104 31. Type of Prescribed Psychotropic Medicine ......... 107 32. SCL-IO Scores ....................... 108 33. Correlations Between Self QOL Scale and Physical Health Variables ............... 110 34. Correlations Between Self QOL Scale and Mental Health Variables ................ 111 35. Number of People Providing Each Type of Social Support . . 112 36. Negative Social Support .................. 114 37. Percentages of Specialist Social Supporters ........ 115 38. Characteristics of Social Supporters ........... 116 39. Family Composition .................... 119 40. Reciprocity of Support .................. 122 41. Social Support Quality of Life Scale Scores ........ 122 42. Correlations Between Social Support QOL and Selected Social Relations Variables .......... 123 43. Correlations Between Family QOL Scale and Selected Social Relations Variables .......... 124 xiii ' I '. . a)... - z . —:‘ ”$3. 7 a": - . ‘3 I) : .37: it? 7:69 pnin‘apWM-afl ,, - .0, ‘ ”37$“ If? IFE‘FI P37 TOG evo‘3h37m33 TV ”'."V'mu is. f- .m- m! adqifisei‘mquimgwlmul .2! . 74-73 fidk‘wulfl v‘flOIJSI‘VV‘VO) .83 .a in‘. immzinrxi‘l - - .fs' ‘ . .' e. '0 I .0. Z 5333': *1 3r '3 ". 323,‘ dQ’.‘fiH [51'9”‘qa . ' ‘50; (I ‘l ' “27””: 5mg ”5‘.” '0 VJ'lfuv Jimmie Isaac: . 3'- -‘- ,_ “‘$- Table l 44. Correlations Between Independence and Leisure QOL and Selected Variables ................. 125 45. Summary of Quality of Life Scale Means .......... 127 46. Intercorrelations Between Quality of Life Scales ..... 128 47. Prediction of Global Quality of Life Using Life Domain QOL Scores .............. 129 48. Variables Entered Into V-Analysis ............. 131 49. Final Clusters Derived from Pre-Set Analysis ....... 133 50. Correlations Between Oblique Cluster Domains ....... 136 51. Number of Members And Homogeneity of O-Types Derived from the Typological Analysis ......... 145 52. Means, Standard Deviations, and Homgeneity of the Eight Derived O-Types ................. 146 53. Demographic and Miscellaneous Information by O-Type . . . 147 54. Housing Variables by O-Type .............. '. . 148 55. Work and Finances by O-Type ................ 149 56. Safety by O-Type ..................... 151 57. Self Variables by O—Type ................. 152 58. Social Support Variables by O-Type ............ 154 59. Criminal Behavior by O-Type ................ 156 60. Incident Reports While at Shelter ............. 161 61. Reasons for Termination .................. 162 62. Reasons for Being Yellow-Tagged .............. 162 63. Comparisons Between Initial and Follow—Up QOL Scale Scores ............... 164 64. Comparisons Between Initial and Follow-Up SCL-IO Scores . . 165 65. Residence Six Weeks After Initial Interview ........ 167 66. Sources of Income During Six Weeks Prior to Second Interview ............... 168 xiv V- .p .3“: .‘gi‘ _):I W... WW m‘ mm, , aroma M ed ‘ 1"? an ,0}! 631-78 3‘“ 10 (39751.10 M N 9;? 110313101 3. ‘ n: .mifaion‘l .Tb A 3" :ii in 1. "Maria?" 4.15.. IiLq. . 35307:. a. ‘21.:2 .200 w- «a (.03 o 7 1 \_ .. uh, y..- - . 9 v ‘6 o M ii €30.41” 7’? ,- ' I 0 3| a! N. M N f n. o: in u n. .. m u ...c ‘J H. n m new Figure 1. 10. Experimental Design: Appointment Card for Follow-Up Interview Cluster Scores Cluster Scores Cluster Scores Cluster Scores Cluster Scores Psychiatric Cluster Scores Cluster Scores Transiency Group ..................... LIST OF FIGURES for O-Type for O-Type for O-Type for O-Type for O—Type Group .................... for O-Type for O-Type 1: 2: Type of Payment by Type of Appointment Card .................... Lower Deviancy Group High Victimization Group High Transiency Group . . . High Psychiatric Group High Transiency - High High Criminality Group High Criminality - High Percent Follow-Up Return by Type of Payment and Type of Appointment Card uuuuuuuuuuuuuuuuu 47 49 138 139 140 141 142 143 144 .1 0-1 oz ’2 hm ”Owen? .~- -' OWN/U swam: in ”won! 0 n 9 o , '- n m m. t n n 9 M . 0. H. mm _ V. n. - LIST OF FIGURES Figure 1. Experimental Design: Type of Payment by Type of Appointment Card .................... 47 2. Appointment Card for Follow-Up Interview ......... 49 3. Cluster Scores for O-Type 1: Lower Deviancy Group . . . . 138 4. Cluster Scores for O-Type 2: High Victimization Group . . 139 5. Cluster Scores for O—Type 3: High Transiency Group . . . . 140 6. Cluster Scores for O-Type 4: High Psychiatric Group . . . . 141 7. Cluster Scores for O-Type 5: High Transiency - High Psychiatric Group .................... 142 8. Cluster Scores for O’Type 6: High Criminality Group . . . . 143 9. Cluster Scores for O-Type 7: High Criminality - High Transiency Group ..................... 144 10. Percent Follow-Up Return by Type of Payment and Type of Appointment Card ................. 159 .p e i. E CHAPTER I INTRODUCTION Homelessness in America is not a new problem. However, the often romanticized hobos and boxcar adventurers of yesteryear have been replaced by a different picture today; that of homeless "new poor," deinstitutionalized mental patients, and "street people." The Great Depression of the 1930‘s spawned large numbers of homeless. However, post World War II national affluence, along with the natural decline of urban skid rows (Miller, 1982; Bahr, 1973), reduced the problem to a level which removed it from the national consciousness. In the 1980's, homelessness has reemerged as a problem. Estimates of the numbers of homeless in this country range from as few as 250,000 to as many as three million (Bassuk, 1984; Holden, 1986). The problem is particularly acute in the nation‘s urban centers. There are as many as 36,000 homeless in New York City (Baxter & Hopper, 1981), 15,000 in Los Angeles, and from 13,000 to 27,000 in Detroit (Smith, 1984; United Community Services of Detroit, 1985). National spokes- persons and advocates for the disenfranchised poor, such as National Coalition for Creative Nonviolence leader Mitch Snyder (Hombs & Snyder, 1982; Katz, 1984; Pichirallo, 1986; Schwartz, 1984), have voiced moral outrage at the plight of the homeless, and at the callousness of a national leader who remarked that many of the homeless were that way “by choice" (Hopper, 1984; Thomas, 1985). In April 1985, a Congressional 1 J it . D . _ 7.5.. _TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTI:ZIIIIIIIIIIIIIIIIQPIIII investigating committee declared homelessness to be a national emergency and called upon the President to mobilize a special Federal effort to combat the problem (Congress, 1985). In recent months, the issue of homelessness has become a "popular“ one for columnists, feature writers, and news commentators (e.g. Grove, 1984; Krauthammer, 1985; McKay, 1986; Rabinowitz, 1985; Raspberry, 1986; Reid, 1984; Roberts, 1985). Recently, the issue was popularized through the national appeals of "Hands Across America" and "Comic Relief" (Christoff, 1986; "Comic Relief," 1986; "5 Million Join,“ 1986; "From Sea to Sea," 1986). Whether this attention actually mobilizes an effec— tive national response to the problem remains to be seen. Besides the immediate concern of lack of housing, the homeless often suffer from a wide range of additional problems. For example, the I homeless are often victims of crime, harassment, and sexual abuse (Baxter & Hopper, 1981; Blumberg et al., 1978). In southern California, citizens calling themselves “troll-busters" prey on the homeless, and a Fort Lauderdale official has suggested topping local garbage with rat poison in order to deter people from salvaging discarded food (Leo, 1985). Shelters, supposedly a place of refuge for the homeless, are often more dangerous than life on the streets. As Mowbray (1985) asserts, the reality of life on the street and in the shelter environ- ment is: “a miserable, often inhumane existence that robs people of self-respect, shames them, exacerbates their health and mental health problems, and provides little help to escape. With few personal resources remaining, most people are not clever or persistent enough to fight their way out of this cycle of poverty“ (p.5). 370ml m! JGMMW “ .l‘ulm 0: new: M In amt“ an- n) 1!!!st m a F' .1: , whom mean nl math.“ ‘30? gar us") .48er :3 It. )nqe I. .ft'L'L “(VIN .0'9” .3“ BMW‘“ -' ”7" n 5’; m Mar-m ‘ Gad item _ EMLI a‘L’: The right to housing is not Constitutionally guaranteed (Collin, 1984). In some jurisdictions, however, the right to housing has been mandated through other mechanisms. For example, some states such as New York have state constitutions which provide that "the aid, care and support of the needy are public concerns and shall be provided by the State" (Collin, 1985, p. 326). In the recent case of gallaflgn vs; Egrgy (1981), this was the basis for the ruling that the city of New York must provide adequate shelter to those homeless men who apply for it, and that the shelter must meet certain specified standards (Hopper et al., 1982). More recently, in November of 1984, District of Columbia voters overwhelmingly approved an initiative requiring that the city guarantee residents "adequate overnight shelter“ (Pianin, 1984). The new law was promptly challenged by the city which charged that the initiative violated the city charter by forcing the District to appropriate funds. After a series of court decisions, the referendum was eventually upheld in May of 1986 (Walsh, 1986). Thus, a basis for remedying the problem of homelessness is begin- ning to be established through the courts and legislation. However, at best, these mandated measures are band-aid solutions to the problem. While they help to guarantee a right to shelter, they do nothing to prevent the problem of homelessness from occurring, or to address the precursors of homelessness such as poverty, unemployment, and the lack of affordable housing. Other strategies must also be used to address the problem of homelessness. The current study examined one group of the homeless; guests of a temporary shelter agency. The study had two major aims. The first aim was to gather descriptive information about the homeless population. Q I; “ Jflfonomdhuwkfi 31:3,“ ‘. with)" o! 1 m. 77.11.19v9md "muggy-m! “fl" 1 I )131‘z s-.3513 ; -3 nmemmnsm .an9 AW . .' Pry-1 53'"; ”"1 " ‘ mam a1: 10 110m 17'HJ) '93532 ,‘(BDI) ma 9 fi . ”‘iJ;E;U: :0 ; ‘41u151q 151303 .umemm oldab'mfis 1o ”huffing!” ;:- Through interviews, information was gathered on study participants' background and demographic characteristics, psychiatric morbidity, social support systems, and perceptions about their quality.of life. Because homeless populations are by nature quite transient, it is difficult for researchers to conduct longitudinal research with them, or for social service agencies to provide needed follow-up services. One method of tracking individuals, either for research purposes or for service provision, is to make appointments with them for a subsequent meeting. Therefore, the second aim of this study was to examine whether rates of return for scheduled follow-up appointments could be increased through the use of different research procedures. In the following sections, reviews and critiques of the background literature pertinent to the two research aims of the study will be presented. In Part 1, information relevant to the first research aim will be presented. Information relevant to the second research aim will be presented in Part II. This will be followed by a discussion of the research hypotheses. Part I — Identifying the Homeless The homeless are not a new topic for research. Sociologists in particular have been studying skid row residents for decades. Early discussions of the homeless include that by Solenberger (1914) in her study of 95g Thousand Homeless yen, and that by Sutherland and Locke (1936) in their impressive study of unemployed men in Chicago shelters during the Depression. Later, urban renewal programs of the 1950's and 1960‘s produced a new wave of skid row studies. A particularly . . ". ‘. > i : Q‘ g ' no be'rsvlJiy a" “TM“! W " rug hm 3': ,«:‘\-.lz.i:9336'rsfl3 ”normal. ”I 'v « . . i- ‘19 grifwg. 3 ar.~.~4.,;w-yq but ,amsiav: Jioqquz 70"., - ~ ~ -~_:':-\'.n names .! '19:.” , , 4n w I? - » mutant 9 9'7. rI. . n; in '10, mm _ ‘5‘: f-m’ Mi} '. -n._.rg;,u3.gib i}- - J 9&6 fizuonl 5110 lo {bani - -' . M '. 1m: nt- ‘ ‘33 c.4»._‘»->/-.-r:-i . ‘ESI' 1+.— comprehensive study was that by Bahr and Caplow which compared residents of New York City's Bowery with populations of settled men (Bahr, 1973; Bahr & Caplow, 1974). Further descriptions of early skid row studies can be found in Miller (1982), Bahr (1973), and Bahr and Caplow (1974). With the emergence in the early 1980's of homelessness as a signi- ficant public problem (Stern, 1984), researchers have again begun to focus resources on the study of this issue. This has been fueled in part by an increased availability of federal and state funds for study in this area (Bachrach, 1984b), including a number of studies funded by the National Institute of Mental Health (e.g., Fischer et al., 1986; Human Services Research Institute, 1985; Robertson & Cousineau, 1986; Roth et al., 1985; Solarz & Mowbray, 1985a, 1985b). However, the tradi- tional stereotype of the homeless person - that of the unmarried, middle-aged or elderly, skid row alcoholic male - no longer describes a "typical“ homeless person. During the last ten to fifteen years, the characteristics of the homeless have been changing. A broad research base which reflects the contemporary population of homeless has not yet been developed (Bachrach, 1984a, 1984b; Milburn & Watts, 1984). For the most part, research on the homeless has been descriptive or epidemio- logical in nature, and the majority of resources have been focused on defining the population, performing needs assessments, and examining the incidence of mental illness. Many studies have measured the incidence of certain characteristics in the homeless population, such as rates of psychiatric morbidity, and study samples are generally well described in terms of their average age, racial background, and other such demogra- phic variables. However, more methodologically sophisticated research '_"-'.(57~7r‘r 24M -£1'107 ' .y “ V: 593 -O.".r 00 ' _ :noiésiudnq rm. mm run M- " ,me: ,wi'fl I ' :jr-‘L'f. 15,0111}? 20:; “NM n' have?) “I, '~ Via 3"! MTV v.7 In)" 3.‘ ' c0)?" .qéq ' l A] .1. ..- , -,MD-]3;qug "‘0 "fl ' involving interventions with this population are, for the most part, nonexistent. Because the characteristics of the homeless have changed, new strategies for dealing with the problems of homelessness which are more appropriate for addressing the needs of the new population must also be developed. There are several levels on which the problem of homeless- ness may be addressed. These include primary, secondary, and tertiary approaches to prevention (Heller & Monahan, 1977). A radical, or pri- mary preventive approach to addressing this problem focuses on inter— vening to "solve" one or all of the causes of homelessness at a societal level. These are attempts to prevent problems from ever occurring. For example, ensuring that adequate housing is accessible to all citizens could be considered primary prevention of homelessness. A secondary preventive approach might involve addressing the problems of those "precipitating events" which lead to an individual becoming homeless. For example, programs to aid individuals in their transition from insti- tutions to the community, or shelters for the temporarily homeless might be considered secondary prevention. These are attempts to identify problems early on and to reduce their length and severity. Finally, tertiary prevention involves treating the impairment which results from problems. For example, psychological counseling for the homeless to help them deal with the problems they are experiencing would be an approach at this level. In order to address the problem of homelessness on any of these levels, it is necessary that the population be adequately described, and that the needs of the homeless be understood (Kroll et al., 1986). There are a number of different frameworks which may be useful for "Ar 5": ‘ v" . . - - ~ , ”ks-.97“ nnnnfwoq-am'm 0" --.,3 ‘:‘Kt'.'. 9M sealed. '45'1‘ at“ 29199171..» 1 .3. --.1u.‘ruoaqqt 'i-uzwfovob "-‘l 7-H)" .‘.)\‘,li -. . ._'l . ‘H. 18%,: 5'3 “@9795“? am» a: ”Me-.9}: describing and understanding this population. For example, the homeless may be examined in terms of their demographic characteristics, pattern of homelessness or residential history, individual causes of homeless- ness, and/or subjective assessments of homelessness. Each of these methods of describing and understanding the population will be discussed in the following sections. Framework 1 — Demographic Characteristics Research indicates that the homeless population now contains greater numbers of women and mentally ill persons than previously, and is younger. Each of these changes will be discussed below, followed by a discussion of some of the limitations of current research. For more comprehensive reviews of this literature, refer to Bachrach (1984c) or to the several bibliographies which have been compiled on homelessness (Garoogian, 1984; Kenton, 1983; Sexton, 1984). Mental Illness and the Homeless Perhaps the most important change in the homeless population is the increased numbers of mentally ill persons who find themselves a part of this group. Estimates of the numbers of the homeless who are mentally ill range from less than 25 percent (e.g. Segal et al., 1977) to as high as 91 percent (Bassuk et al., 1984). In their study of admissions to an Illinois state hospital, Appleby and Desai (1985) found that the rate of homelessness among psychiatric admissions has increased substantially over the past decade, and suggest that the incidence of homelessness among the mentally ill is at least three to four times that of the general population. Even minimum estimates indicate that a substantial number of the homeless may be in need of mental health services. The . ~- . .- ~ ‘ - .. i , .- ,. Jana-”1913515113 grumnpmm nah: i.) am” in gum“ 4-y- . “affine”: mi noncluqoq am: unmitm u‘ I _.u- .£ .. . , . . "o5 «animon '-: _:4 - _ , an: urn...“ :w 10 anneal“ .r . , 4 0 L 33.9 f'Ju? iokbna .- 1 .r‘ vet. ‘r. abort}. ‘ 9:13 a! «31,; 2r3nrlfl a uezasvinl 96d erase to :l’xts‘rnateq 900m azsnzzsl'saoti _ - 9mm; 5 3.38111: 11.: E‘¥ lgk \ y movement during the 1960's to deinstitutionalize mental patients to less restrictive community settings has generally been blamed for "dumping" the mentally ill onto the streets without follow-up or community support. More properly stated, humane treatment policies were inade- quately implemented (Bachrach, 1984d; Bassuk, 1984; Fustero, 1984; Hope & Young, 1984: Jones, 1983; Kaufman, 1984; Lipton & Sabatini, 1984; Shadish, 1984; Talbott & Lamb, 1984). Additionally, as federal and state funding priorities changed, necessary fiscal support for properly initiating the policy was withdrawn, or never allocated. Whatever the cause, it is generally agreed that large numbers of the homeless have serious mental health problems (Bachrach, 1984c). W In the past, the presence of women on skid rows was considered to be rare (Bahr, 1973; Blumberg, 1978). Bahr (1973) notes that studies published in the late 1950's and early 1960's reported that fewer than five percent of skid row samples were women. He cautions, however, that many homeless women may have been overlooked in previous studies, as they would rarely be present in the places where social scientists studied the homeless. While there was likely some error in past esti— mates of the proportion of the homeless who were women, homeless women were clearly greatly outnumbered by men. Today, homeless men still outnumber homeless women. However, greater numbers of women are joining the ranks of the homeless, and they may comprise as much as twenty-five percent of the homeless population (Arce et al., 1983; Arce & Vergare, 1984; Bachrach, 1984d; Crystal, 1984; Lipton et al., 1983; Stoner, 1983). It is difficult, however, to . “. 'u.q raJnam asilznnriuzircnisb :3 4'0391 an! qukatlnl O at? L'--.=~m.lv.i . ' as" 3mm 37'.” 21mm Nffi' I I £3.11." in." . . . ' . ”:21; 911‘ ‘JMI? "V umm" v‘m Iwoq : 7m: 1.: (snuf- .,“"'..'OY I 5‘; ‘ 'srecfla as . ' i‘ r. I ‘u.., ) 53“?“ \ qsvtswo‘i .33: u cueian‘ 4, indomimu “'73.. n...‘ maimed (abnT 1135339391311» 9&1 retaliated! mm to 2190mm . 3‘ _.,, _" . I... '1‘ .' --‘.‘_~'.’ 3' ”r . ‘ '. .7 '~ L1 ”749,5. -J~ “5 'V'fi’ 1mm " V ~ . . L l“ &‘.n..' estimate the exact percentages of women in the homeless population over- all from the proportions of women utilizing shelters, as these numbers are determined in great part by the lesser availability of shelter beds for women. Most research to date has focused on homeless men, in part because of the greater access which researchers have t0'men through shelters. Nonetheless, some researchers have included women in their samples (e.g. Lipton et al., 1983; Arce et al., 1983; Ball & Havassy, 1984; Bassuk et al., 1984; Crystal, 1984; Roth et al., 1985; Solarz & Mowbray, 1985a, 1985b), or have focused exclusively on women (Depp & Ackiss, 1983). There is evidence that while homeless men and women share many characteristics, women differ on a number of variables. For example, Crystal (1984) found that women were more likely to be married, were more likely to have had histories of psychiatric treatment, and were less likely to have been involved in the correctional system. The specific reasons for the increase in the numbers of homeless women are unknown. Some research results suggest that women under treatment by the mental health system were disproportionally affected by deinstitutionalization policies (Crystal, 1984; Bachrach, 1984d). Alternatively, Stoner (1983) suggests that the burden of poverty falls disproportionately on families headed by women. This "feminization of poverty" has led today to greater numbers of women being forced to the streets in the absence of societal supports. Before the phenomenon of increasing numbers of women among the homeless can be fully understood, more research must be conducted which focuses specifically on the precipitating circumstances by which women become homeless. , _._' 3891”)“ 9112 mi «Wu ”mm .6925»: as mum: ~ Wm, ,,.__.,,,,_°m‘m ”"7 f’ ' W T‘ -“’.1‘1, 133:; a. but"... «can 1.. “M" 9.30" 141 ‘0 5111900. *4 1l ‘ i I q ,‘4 1‘ l' ,. l I --l .' ‘7 T3” (ITSVW sic-lee: ..::.'znqsua_ "rsm we to molds am of 21' ‘2'" “49:39“ mmkum " yr!“- Age of the Homeless Another recent change in the homeless population is that the mean age has decreased. In 1936, Sutherland and Locke reported an average age of 45 years among the homeless in Chicago during the Depression. In two later studies, Levinson (1957) found a mean age of 48.5 years in his study in 1955 of white homeless men in New York City, and Bahr (1973) reported a median age between 50 and 54 years for studies of homeless men conducted during the 1960's. Recent studies, however, have gen- erally reported mean ages in the mid-thirties for the homeless (Lipton et al., 1983; Ropers & Robertson, 1984; Arce et al., 1983; Ball 8 Havassy, 1984; Bassuk et al., 1984; Crystal, 1984; Depp & Ackiss, 1983; Fischer, 1984; Kroll et al., 1986; Lamb & Grant, 1983, Solarz & Mowbray, 1985a, 1985b). Several reasons for this phenomenon may be suggested. One reason relates to the deinstitutionalization policies described above. Today, young people who would have been institutionalized 15 to 20 years ago are given only brief and episodic care (Bassuk, 1984). Consequently, these individuals are placed in the community and, if they have no viable housing options or the skills necessary to maintain themselves on their own, may find themselves homeless. This problem is particularly acute now as post World War II “baby boomers" enter their 20’s and 30's, the ages when schizophrenia and other chronic mental illnesses often develop (Smith, 1984). Another contributor to the lowering of the mean age may be the recent economic recession and the concomitant high levels of unemploy- ment. Unemployment may affect younger workers more than it does older workers in several ways. Younger workers may have less seniority (and i. .L r A" 5m- ' .995 15'1 r. Cu} .33»: I .. 273‘le 03:0: bin ' . . _ f . V. _ .—' ’U‘ E 9.3 ‘ art-3'" (A . . i £333” ”MN" ?°‘€'9'T?f‘_9=__ 1941995 _.£E§1;. 11 thus be the first to be laid off) and fewer skills (and thus are less likely to be hired) than older workers. In addition, they may have less savings and fewer benefits, and therefore be less able to weather periods of unemployment. Thus, younger workers may be more likely to find themselves without a job and "on the street." General Criticisms of Previous Research There are a number of methodological problems in much of the research done to date on the homeless. These include the lack of a uniform definition of the population and problems in labeling the mentally ill. Defining the homeless. As stated previously, estimates of the numbers of homeless Americans are widely disparate. Government agencies generally give the lowest estimates. For example, the Federal Depart- ment of Housing and Urban Development reported in 1984 that between 250,000 and 350,000 persons may be homeless daily (cited in Bassuk, 1984). Advocates for the homeless, however, claim that as many as 2.5 million Americans may be homeless, and that these numbers are continual- ly growing (Bachrach, 1984c; Hombs & Snyder, 1982; Arce & Vergare, 1984; Bassuk, 1984; Hopper, 1984). Some of the reasons for such different estimates may be political. In addition, however, there are serious methodological difficulties in calculating accurate numbers of a fluctu- ating and transient population which, to a great extent, remains hidden from view (Arce & Vergare, 1984; Bachrach, 1984c). Of primary impor— tance, however, is the fact that there is no consistent definition of homelessness as this makes it difficult to interpret much of the '5?" mm 1.71.?) .;" ‘. we" ~m I” b 5' 3" ‘5’ 3"“ V 5 ‘ ‘ ‘- i . n ' ' If?! ave" Yam (5." ‘N‘ a rim-um . r i .bsnm 9‘ U z. . ‘ ‘Il'n'l 'v'lw9' b" a“; graham :,-..-,. ,, m ans 9mm .3zsez .dfih’5*n§' .,. _ .;15391 .31 3939‘! s 931A.) we" 90.53:: 9-.flili 1.‘ 'llllii 12 research or to determine the comparability of results (Bachrach, 1984b, 1984c; Milburn & Watts, 1984). This problem of inconsistent definintions is reflected in the manner in which study samples are selected. A common method for obtain- ing participants in studies on the homeless is to draw samples from emergency shelter clients (e g. Ropers & Robertson, 1984; Arce et al., 1983; Bassuk et al. 1984; Fischer, 1984; Laufer, 1981; Levinson, I957; Solarz & Mowbray, 1985a, 1985b). While these studies are often pre- sented as research on the "homeless,” there is very likely a broad range of this very diverse population who never use these facilities, and who remain unstudied. While it is still important to study the subgroup of homeless "shelter users," it is also important to recognize that this group does not encompass the entire population of homeless. Some efforts have been made to formulate definitions of the home- less. In their efforts to examine the issue of homelessness in Michigan, a state Task Force on the Homeless developed the following working definition of the homeless: “Homeless individuals are those who lack a permanent residence (a place of one's own where one can both sleep and receive mail) because of inadequate resources, inadequate access to resources, inadequate management of resources, or because they are unable or unwilling to accept a traditional residential setting for other reasons." (Solarz et al., 1986, p. 4) This definition includes a wide range of individuals who might be considered homeless, including domestic assault victims and runaway/ throwaway youth, who are sometimes excluded from other definitions of the homeless. "5%- . ‘ "3'53 " F. . 'V. a .'0 K. .‘. t ' . : .‘ m) 21:11:91 '10 vhf-imam any-mom Q~ ' .(‘3393 ”fife” I . f .0 911.3 m bwifmrrgc ~' 3 -' ' '1‘ "w“ ,“:_n:nl V; 93!an If“: 0 -a)5305 up“ >rn , : -_ ». «-. ; x«r- q:rnu of V 'ronrvil16q - ‘Q :‘3 vmsq‘im '. 28 .58?! A .vr..'08 . x 2)au‘arv‘rrm lt' 99mg! 3..., (, ’ o est-ulsat sum“? no 2an 13 Identifying the mentally ill. These problems of definition extend to the concept of the homeless mentally ill (Arce & Vergare, 1984; Bachrach, 1984a, 1984c). For example, researchers have used such varied indices of mental illness as prior psychiatric hospitalization, self- report inventories, and assessments by professional staff. This problem is compounded by the fact that most studies are "one-shot" assessments of psychiatric status, which do not allow for longitudinal observations. It also appears that the probability of reporting high estimates of the incidence of psychopathology may be dependent, in part, on the profes- sional orientation of the researchers (Bassuk, 1984; Talbott & Lamb, 1984). Different disciplines tend to use different criteria for label— ing individuals as mentally ill. For example, some of the highest esti- mates have been made by Bassuk (Bassuk et al., 1984) from the Harvard Medical School Department of Psychiatry, while Ropers and Robertson (1984) from the UCLA School of Public Health suggest that the majority of the homeless are not deinstitutionalized. Another caution should be taken when evaluating the reported pre- ponderance of mental illness among the homeless. As Baxter and Hopper (1982) maintain, researchers and clinicians generally evaluate homeless people when they are experiencing highly stressful conditions. Those who did not suffer from mental illness before they became homeless may become exhausted and disoriented as a consequence of the daily stresses involved in surviving on the streets. Baxter and Hopper further remark that "were the same individuals to receive several nights of sleep, a nutritional diet and warm social contact, some of their symptoms might subside" (p. 400). While this is not intended to minimize the real problems of the homeless who are mentally ill, it does indicate a Eff-".7 15921 in zaafdom scam "I ll v; p _ . . . ‘ H :1 235$! ,e‘visp’va‘ .- avvv‘ ' vl'arnvr Hume» 9K) in moods) "th3! I‘l“L.‘ '~ .J) {i(... ' 0. V' , .' .1136?! , .-‘ Nam 10 as) va'i' ’V ' , ' “.""-""J) 2' ' mg ‘10 H muquii t-‘t- «:3 a. v. . if . .1“? . '51 image 3‘53,” 3.439113%“? on: an: mcfi‘ ‘9...” v‘- _.. ‘ _. . >8”? .- - .. ‘ ,‘ I 'g'A: ‘ agar, " e E; :V-‘ different point of intervention for addressing needs. That is, in some cases, primary importance should be placed on obtaining housing rather than on "counseling" or psychological assessment. Finally, as Mowbray (1985) suggests, placing the blame for homelessness on deinstitutionali- zation may simply distract attention from more direct causes such as poverty, unemployment, and the lack of affordable housing. There is substantial evidence that the homeless are not a homo- geneous group. Consequently, it is not appropriate for researchers to assume that their sample is representative of the homeless in general. Homelessness is an issue which can elicit highly emotional responses from people, and there are a preponderance of stereotypes about the homeless. Therefore, it is particularly important that data be gathered on the characteristics of today's homeless. In this study, background and demographic information, including information on psychiatric mOrbi- dity, was gathered in order to describe one sub—population of the home- less, i.e., those who use shelters. While this study was not prepared to address all of the many limitations of the research in this area to date, the information obtained will be useful for describing this parti- cular segment of the homeless. A framework for understanding the home- less based on residential patterns is presented in the next section. Framework 2 - Residential Patterns of Homelessness It is clear that it is not appropriate to speak of the homeless as a homogeneous population with similar kinds of problems. In fact, they are a very diverse group. While it appears difficult to categorize the homeless in terms of their demographic characteristics, some researchers . T | .‘,.-_.‘”' I ‘ "1:5; m‘ V." do "'7'“ V" ”'N’ ' I‘M'VQ . ..‘.wv'a“ no I\".V.'~'"*1 ‘ .1} ‘ :.’£.'*.til"‘qub :en a. 3: no: "6)"3 2': :I Qiiililjlgtiaiflfiwimzoiuqoq z 4 >5 . -\’- I A- o,‘: .— .‘_.'.4‘ 15 have attempted to develop frameworks for understanding the homeless based on residential history. Arce et al. (1983) grouped the homeless into three different classes. First are the “street people." These individuals regularly live on the street, and have lived there for more than a month. They are generally over the age of 40, have a history of substance abuse and/or have been diagnosed as schizophrenic. They also have a variety of health problems and a history of state mental hospitalization. The second group is the "episodic homeless." These people alternate between being domiciled and being on the street. When undomiciled, it is generally for periods of less than a month. They are generally younger than the street people, and tend to be diagnosed as having a personality disorder, an affective disorder, or a problem with substance abuse. They are likely not to have a history of prior hospitalization for mental illness, but have had sporadic contact with a number of different human service agencies. The third group is the "situationally home- less." These are people who are undergoing an acute personal crisis and have a temporary need for shelter. For example, individuals evicted from their residence or forced to leave because of a broken heating system might be included in this group, as would those temporarily stranded in a city because of bad road conditions. In their study, Arce et al. determined that 43 percent of the individuals in their shelter sample were street people, 32 percent were episodic homeless, and 13 percent were situationally homeless. (They did not have enough informa- tion to classify the remaining 12 percent of their sample.) Ropers and Robertson (1984) provide another taxonomy based on resi- dential history. They group the homeless into groups of “long term," mil "ibn'. ".‘v? FLlefimgq9 :01' 5'” w ”a " * «um-Luna I in I ' "7. iv 3‘9" -,.- i * 1:5"- - ' will. ~ ’ - “in v“ 'l'l - ’3‘5 .-v .' v 7,. :9,“ ‘ , --.' ”Emile ‘ ‘, 4‘3. _'.-. 15319.4: .. 1‘ .\ J ’ ‘tl ‘ ' - , ' '1 ‘” . ‘v‘vl'hij‘JL; .is J, .. 3 , f‘ f '_ ; ;- 5~ ““8 .z-e: six-.1 vim .. . . ,. l). . 4 ' , - -~ ‘4) .71. , “.‘rLU q. ..,. _ 7 . ,, 93:4 ’5311. Swan aim . :39 9:594: 599,41P15M) , .aeaiswd filament-1dr: new: ~ ' _ 1‘: u... ‘1” ‘ ‘ 3" I . “"TZ. .- g _- “. r .. . ~ : '9'? . . - .7 . V 1‘ I": ‘-' 9 l - . ~H .,. :I ' ‘ ’ 6' _ ._ ...- .7... 16 "episodic," and "transitional" homeless. The long term homeless have no present residence and have been homeless for longer than twelve months. The episodic homeless have no present residence, have been homeless for less than a year, and have at least one prior episode of homelessness. Finally, the transitional homeless have no present stable residence, have been homeless for less than twelve months, but have no previous history of being homeless. While these taxonomies may present useful ways of categorizing the homeless, they provide little information about the etiology of home- lessness. Furthermore, they do not indicate how meaningful interven- tions for addressing the problem of homelessness might be developed. Framework 3 - Causes of Homelessness Another framework for describing the homeless is in terms of the types of events which led to their becoming homeless. The causes of homelessness are myriad. However, they may be grouped into categories of global or society based causes, and specific or individual causes. Global Reasons for Homelessness Baxter and Hopper (1981) cite three major social and economic developments leading to homelessness today. First, inflation and unem- ployment, coupled with reductions in funding of social programs, have resulted in more and more people falling outside of the "safety net," and onto the streets. Baxter and Hopper note that unemployment, in addition to its obvious economic impact on families and individuals, also causes a great deal of stress and disrupts personal support net- works. Consequently, individuals who are without a source of income, who are unable to find a job, and who lack social supports to fall back 1‘- H11 O '.. 5- .0... . u 1 O 2': . .izz‘fiaoen «we: on»! w” emission “immflmmr m ‘ O V '\ i- ., ' 3W9" “vi/"N ‘2 ‘--‘ 7 4' v .-‘ “‘va we"! a.” ma ”MM“ 1’. J“ 1 vi}! ~l ‘ ‘3 V '.'u :E‘ .au“ 3‘».’“ uyv . nod!’ m7 .vlfsn‘ *va oval ' ' 'Hfiflm e :-. ".39” V59 ‘- 17 on in times of stress, may also find themselves homeless when they are unable to pay their rent. On top of these problems, gentrification of inner cities has dis- placed thousands of individuals, with no provision for replacement housing. This has resulted in a severe reduction in the number of single room occupancy hotels (SRO's), the traditional residences of the poor (Bassuk, 1984; Fustero, 1984; Kasinitz, 1984; Lipton et al., 1983). For example, it has been estimated that the number of SRO rooms in New York City decreased from 50,454 to 18,853 between 1975 and mid—1981 (Kasinitz, 1984). Finally, the well-meaning, but poorly implemented, deinstitution- alization policies of the 1960's resulted in large numbers of the mentally ill being released into communities. Without provision for adequate after-care and follow—up, many ex-patients ended up "on the streets," unable to cope effectively on their own in the community. Joining their numbers are the many mentally ill individuals who are now refused admission into hospitals under stricter entry criteria. Specific Reasons for Homelessness These global antecedents to homelessness are relevant for under- standing the general climate which has led to an increase in the numbers of homeless. However, they do not necessarily help us to understand the immediate events which precipitate homelessness for individuals. The reasons why individuals find themselves homeless are varied. A common reason is eviction from prior residence by landlord or relatives. In addition, poor conditions of affordable residences, or catastrophic events such as broken heating pipes, sometimes force individuals onto . :yic- ,4 f t-m.‘ v.>.'n v30: 13¢!!! ‘02 “~11 Mo ‘_ 49'.) ‘90 001 «5" '. 'uv‘vzauodJ -'- ‘ "1 .oniz - mi «(9012 woo a? , . .— . .‘7 “‘7...“ .6 -. ,.“,‘ v '1 1:41. on 2» .::rz5v9 5339119“? [32:21an my} aim: if , ‘ " 9929139)} swaeaa'iwon aév ivi’aoi 1d» , ....;g"k ' n1: ' 18 the street to fend for themselves (Simpson, 1984). Once housing is lost, however, it may be difficult if not impossible to replace because of a lack of affordable residences. Individuals with limited economic resources may also exhaust their familial and other social resources. Abandoned by family and friends after histories of mental illness or involvement in the criminal justice system, these people can no longer stay with those on whom they have relied for support. Unable to support themselves on their own, they end up on the street or going from shelter to shelter. Included in this group are individuals whose families have been disrupted by divorce, death, or abuse. Another segment of the homeless has been released from institutions (either mental or penal) without adequate follow-up and after—care. Although they may technically have been released to a residence (al- though some may simply be released to shelters), these residences may in fact be substandard, temporary, or simply manufactured by the client in an effort to meet the requirements for discharge. Consequently, without adequate post-release monitoring, these individuals may be unable to maintain themselves in a residence. In addition, as noted above, they may not have the necessary social support systems to facilitate their transition back into the community. Unable to cope effectively in the community, they end up among the homeless. In their sample of guests of a Los Angeles rescue mission, Roper and Robertson (1984) reported that the most frequently cited reason for homelessness was unemployment (34 percent), followed by the lack of money (21 percent). Both of these reasons are clearly linked to eco- nomic need. Thus, the most important precipitating event leading to m .1080: mnmermtmmma~ : B‘fl vn'ur1i‘.j,1‘-? 5.4: V.“ ,P .‘96.- ’unL-Jng V0 I ~'&.mivibnl ? . I ‘f .nn .aria » ' . .71.. ‘9’?. V . ~sz L .‘. ‘ - " “' ’; 3:” “,3'320'» 911-1.: 3. .ru > ’ -‘ i . ..A a b .— .3; 393‘”? mm mam? V-k-m' .57.} .‘ 74.." ,_ homelessness is often simply that a source of income has run out. Indi- viduals may lose income in a variety of ways. For some, a job may be lost or unemployment benefits have run out. Others may lose their source of economic support through divorce or estrangement from family members. In some cases, support checks have been stolen or lost, or support payments are not substantial enough to pay for living costs for an entire month. Problems arising from the lack of low-income housing, lack of familial supports, and failure of aftercare provision can often be alleviated if adequate funds are available to pay for housing. Thus, the causes of homelessness are many, both on the societal and individual levels. Because individuals may become homeless as a result of the combined influence of many different factors, it is difficult to categorize the homeless into groups based on the cause of their particular case of homelessness. Nonetheless, information about the causes of homelessness remains critical for understanding how, and at what points, the problem may be addressed. To date, most of the information available on the causes of home- lessness consists of theoretical musings by "experts" on the etiology of the problem. As is somewhat self-evident, a common base of many of the causes of homelessness is an economic one. However, this information does little to indicate feasible solutions to the problem of homeless— ness, short of advocating a redistribution of the country‘s wealth. Few researchers have adequately documented the precipitating events which lead to homelessness. . ,7 -.v,. '601 can smvnm : 931110: a ind: If“: min at , .- ,mu 3 . wu ’= T‘s-Ian (- m any.) 930! m 3 \- evsrt 7" . 'n-moimu 1’ ‘ ammo )9 )0 A -L -filmri 3'.) ."i“1:‘jv'-: ,3, , r , . .v v jg m 2"T_Jflu03 £3.13 1rvnug’v".‘;1:_.;u,- ‘0": “.7 -. 20 Framework 4 - Subjective Assessments of Homelessness Another method for describing the homeless is in tenms of specific characteristics of their lives, and in terms of the subjective assess- ments which the homeless make of those characteristics. In other words, information may be gathered from the homeless themselves on how they perceive different aspects of their lives. Research to date has gener- ally been limited to gathering "hard" or factual, rather than perceptual data. Thus, little information is available on the attitudes which homeless people have about their situation. The addition of subjective information adds a very important per- spective to research with the homeless. This population is signifi- cantly disenfranchised from the centers of power, from economic re- sources, and from housing. Much of what is published about the home— less, both in the professional and lay literature, is written by "experts“ who describe the lives of the homeless based on culturally biased appraisals of objective characteristics of their lifestyles. Few have the homeless speak for themselves. Research which includes subjective data is important for a number of reasons. For example, subjective assessments of need by the target population may be the most appropriate for determining the most pressing areas of need. In addition, comparisons between subjective assessments by domiciled individuals with those made by the homeless may be useful for understanding the etiology of homelessness, the effects of homeless- ness on individuals, important similarities between the populations, and the like. These kinds of assessments may also be useful for indicating areas for possible intervention research and for developing programs which will be accepted by the population. -~,.1n , fl ""'-0 o ,* ‘« Mt? ni .a‘r :HY-amn‘ ad: pnidtuzfi 10‘ W2- 1- -9 '.' " r» . 5» *3!!! svnt-aiciu-s v“? ~ I“ .— :vnr‘v .“w ? ‘féfi’ ’2‘ 2313?? ‘ '.-“.-. 335‘“ ”5530 "- .‘ ' “ am. in"? 5.13 21.1161! .- ‘9’. . . . l-r‘ h" '75," "OI3W' .. - ‘li h avrn‘iqv . 'r F . ‘ ~ ~ - . m. ' -. T .535!) ". I :‘-‘ 3:4 ‘ x .{UQ . ;;..'-§ns~:nee92m “‘u-r . TU £5975 , . jé £933- .91 96 {em ,v ,. .. . . 1,; . , ‘ . . , 21 An important topic of research where this approach has generally been lacking is that of the quality of life of the homeless. In the following section, research examining objective and subjective aspects of the quality of life of the homeless will be discussed. This will be followed by a discussion of research on a specific element of quality of life - that of social support. Quality of Life of the Homeless Little is known about how the homeless perceive the quality of their lives. A popular conception is that the homeless are that way because they are independent (or crazy) individuals who choose to be homeless, because they are lazy, or because they refuse to accept the help that is offered them. This perception is sometimes cultivated by the media who spotlight individuals who appear to have chosen a homeless lifestyle (e.g. Grove, 1984). These assertions act to minimize the magnitude of the problem and to rationalize the withdrawal of resources for this population. Clearly there is a need to gather information on how the homeless perceive their quality of life. Quality of life refers to the "sense of well-being and satisfaction experienced by people under their current life conditions" (Lehman, 1983b, p. 143). It may be assessed globally with respect to life "in general," or with respect to specific life domains. Campbell (1981) describes twelve areas of life which concern almost all people, and which are largely responsible for satisfaction with life in general. These include the domains of marriage, family life, friendships, stan- dard of living, work, neighborhood, city or town of residence, the nation, housing, education, health, and the self. According to om?- ‘ 3‘5" .i‘vni.<,q:. i'-‘. awn-i nJ‘noea'fi M “.91qu 30., .1. ,5»; a! 9n! name: 90‘ ’ v.“v'bun .- newt". :v‘! . 51" l r was! ~ . 22 Campbell, the domains which are most highly related to general life satisfaction are, in descending order of importance, the self, standard of living, family life, marriage, friends, and work. Additional dis- cussions of the concept of quality of life can be found in Andrews and Withey, 1976; Campbell (1981); Campbell et al. (1976); Murrell et al. (1983); Murrell and Norris (1983); Andrews and McKennell (1980); Bubolz et al. (1980); Flanagan (1978); McKennell and Andrews (1983); and Widgery (1982). Quality of life can be measured using objective or subjective indices. Objective indicators include such things as income, marital status, work status, quality of housing, physical health, criminal victimization, and frequency of social relations. However, psychologi- cal measures are also needed in order to gain an understanding of how individuals assess the intrinsic value and quality of their lifestyles (Zautra, 1983). If only objective measures of quality of life are considered, it is clear that the homeless have a very poor quality of life. The homeless are less likely than the general population to be married (Bassuk et al., 1984; Fischer, 1984; Ropers & Robertson, 1984; Roth et al., 1985) are isolated from their families (Bassuk et al., 1984; Fischer, 1984) and have diminished social support systems (Cohen & Sokolovsky, 1983; Fischer, 1984). The homeless also have few material possessions, are generally unemployed (Ball & Havassy, 1984; Fischer, 1984; Ropers & Robertson, 1984) have no permanent residence in a community, and are in poor physical health (Baxter & Hopper, 1981; Darnton-Hill, 1984; Fischer, 1984; Ropers & Robertson, 1984; Solarz & Mowbray, 1985a, g I- A ‘b‘. .a I . o . 5 ‘ 3° ' 'l-‘ . ' l‘ . " 1'. 3". D J. '1‘, .u hamisa v. “(on-’15:. mime: «if. “I. ' L115; «'3 ' viv' , 3., 233.3 rvn'bnv vase n? ,m .r‘ 3 ma - ‘1 ‘. -=.!,i ‘0 t - " 2ft: Eon-nu . “ ._ «Hm? . .vsm'f, . :lLBQI ; . ~ .' V ‘ fi’hfi “53st? -.am;gn?z1'i Jemima, even (#891 —)- _v 23 1985b). In addition, the homeless have particularly deficient housing situations. Studies which address subjective quality of life of the homeless are scarce. However, studies which examine the quality of life of the mentally ill or of the poor may provide some information about the homeless with similar characteristics. Several of these studies are described below. Ball and Havassy (1984) conducted a survey of the problems and needs of homeless consumers of psychiatric services. The problems most frequently mentioned by the respondents were having no place to live indoors (94.2%), having no money (88.3%), and not having a job (47.6%). In addition, the most often expressed need was for affordable housing (86.0%), followed by the need for financial entitlements (73.7%), and for employment (40.4%). Respondents also expressed concern about their privacy and personal and physical protection. This study suffered from a number of methodological problems in that the sample was non—random and self-selected, and the interviews were not standardized. However, the results indicate that this group of people perceive their quality of life as unsatisfactory in a number of areas. In their survey of 979 urban and non-urban homeless in Ohio, Roth et al. (1985) reported that homeless respondents appeared to be much less satisfied with their lives than a general sample of Ohio residents. Only a third of the homeless reported that their lives had been "very satisfactory" or “somewhat statisfactory," compared to 86.5% of the general population sample. Conversely, 28.1% of the homeless reported that their lives had been "not at all satisfying" or "not very satis- fying," compared to 2.8% of the Ohio sample. .5 ‘o .3. "£2 ulnaluanisq svfili' 199m ill: JO?!“ . . . | at; and»: mom: 'i '19.‘<1~‘DH 93163! i-: v(if 1".”3‘“ . 3 9319.12“ {2.1“- ‘ 5"." d "6 2’ ." 'v.v r" w- ” 1398-18.11 7v 1 ’ ' b‘lfl’ii 0 11:9,“? {9"} _ .511 -. . '.' ,- ‘6’.“ ”gain '_,j{jioJ.9£viéi;a;e.3anmz" 10 "210338,,w' Ti: I." 24 Cohen and Sokolovsky (1983) measured life satisfaction among home- less men. Results indicated that approximately one-half of men living on the Bowery in New York City felt "things were getting worse“ and that they were “not satisfied with life." Furthennore, Bowery residents were more likely to have lower life satisfaction scale scores than were a similar group of men residing in single room occupancy hotels. While not specifically addressing subjective quality of life of the homeless, Campbell (1981) presents information on perceived quality of life according to the income level of the respondent. Those in the lowest income quartile were least likely to describe themselves as "very happy," and most likely to report themselves as "not too happy." In addition, level of income was positively associated with satisfaction with health. Although Lehman (1982, 1983a, 1983b) did not measure subjective quality of life of the homeless, he did examine quality of life among a population of chronically mentally ill persons in community settings. Structured interviews were used to study the life areas of living situa- tion, family, social relations, leisure, work, safety, finances, and health. Results indicated that over half of the sample felt “mostly satisfied" or better about their lives, except in the areas of work, finances, and personal safety. When compared to the population in gen- eral, however, this group felt less satisfied with their quality of life. Global well-being was most consistently associated with personal safety, social relations, finances, leisure, and health care variables. Ratings of satisfaction with various life domains were more frequently and strongly associated with global well-being than were objective ".4 '9. .“.’.‘ 'noiha‘ieimz sic! bsznm'oélt- 1m, Ma‘d* . .fi, 4111f rise «1 Hart; .v v)", 7;. i'iQ'TS 7m: minim! 23W?” . - “t 1'.-.'; h-Ji W13 "1 (WM -v .3 s r‘a: too“ 519' A -« a (rent n-j’ . a I t‘-‘g3 ‘51?“‘2 .-i_v“‘ ‘Idni nsnri in'swod .1575 ‘ 36am saw poled—(law iadola ;vV~v’ ”g: ”at“ ,z‘s'ananfi .minis-u io'rooa ‘ '— ____________________________________________________________________::::]‘II 25 measures. The objective factors which were most often related to higher ratings of global well-being were not having been a victim of a crime, lower use of health care services, more frequent and intimate social contacts in the home, being employed, and more privacy in the board and care homes. Lehman (1983a) concluded that chronic mental patients (the greatest proportion of whom were diagnosed as schizophrenic) were able to provide statistically reliable responses to the interview. Lehman (1983b) also determined that psychiatric symptoms did not significantly affect the relationships among the quality of life ratings, except in the health domain. In summary, examination of objective aspects of quality of life indicates, as would be expected, that the homeless have a poor quality of life. While it may also be legitimate to assume that the homeless also perceive the quality of their lives to be poor, there has been little study involving these subjective assessments of quality of life by the homeless. In addition to filling a gap in the literature, infor- mation on subjective quality of life can provide a measure of goodness- of-fit between the population and their environment. Furthermore, assessments of the quality of life in various domains may indicate areas which need to be targeted for resource allocation (Murrell & Norris, 1983). In this study, subjective as well as objective measures of quality of life were obtained. As stated previously, quality of life may be assessed in terms of specific life domains. Several of these domains, for example those of marriage, family life, and friendships, are closely related to the concept of social support. In the following section, the concept of social support will be addressed. ’ sis-r flquo 13m raw ‘mfi auras? svnm ”I, in mm" v r we! 1 M u.‘ '- 4‘3" c-‘o’r‘J'u id?" ‘G ‘;I59d 10 gal. .~ , ' 7‘!‘ ‘13 m 21 -;.'.-». :51; b91131: 2A -. ants-no em a! '_- .. .‘zwry- Le";- ”‘82 our 26 W 'The homeless are generally considered to be socially isolated and lacking in social support. As with studies of the quality of life among the homeless, research on social support has generally been restricted to assessments of objective rather than subjective variables. In this section, "social support" will be defined, followed by a discussion of research on social support and the homeless. Although the concept known as "social support“ is a relatively new one (House & Kahn, 1983), it has quickly become an extremely popular subject in social science research. Research has been conducted on the general population as well as on a number of special populations (e 9., Belle, 1983; Cohen & Sokolovsky, 1978; Garrison, 1978; Hammer, 1981; Hammer et al., 1978; Henderson et al., 1978; Lipton et al., 1981; Marsella & Snyder, 1981; Mitchell, 1982; Patterson & Patterson, 1981; Perrucci & Targ, 1982; Thoits, 1982; Tolsdorf, 1976). A primary focus of social support research has been its association with physical and mental health (Berkman & Syme, 1979; Brugha et al., 1984; Cohen & Hoberman, 1983; Davies et al., 1983; Donald & Ware, 1982; Gore, 1978; Hoberman, 1983; House et al., 1982; Lin & Dean, 1984; Lin et al., 1981; Monroe, 1983; Phillips, 1981; Sandler & Barrera, 1984; Sarason et al., 1983; Schaefer et al., 1981), and as mentioned in the previous section, social support is seen as an integral part of quality of life (Flanagan, 1978). However, research has been somewhat hindered by the fact that there are no standard definitions of social support, a deficiency that has inspired several authors to make recent attempts to operationalize the construct (e.g. Bruhn & Philips, 1983). 3L; -L. "1.; f 1.."- ‘r'W? z'll' éflffl' A ..’v< rue who; at 9. , Xvi.” 559‘“ “‘i-‘mfi5233' 3»: ..-ur:sga‘. '. "59991 Ri' »’ 1;. .0: Joe 5‘31‘QDU‘II ,.:;v'T"J:nL.£ ,, , ‘; s-‘l “3“" ”imofi one A imafmmm hobos}: on 5135 ‘ -.‘I , . ‘ -’ _ o ’ I ‘r 27 Social support actually refers to a number of different aspects of social relationships. Evaluation of social supports can include assess- ments of the existence or quantity of relationships, of the structure of relationships, and of the functional content of social relationships (House & Kahn, 1983). Existence or quantity of relationships refers to such things as whether or not an individual is married, the number of friends or associations he Or she has and the frequency of contact with them, organizational membership, and the like. The structure of social relationships refers to such characteristics as the level of reciprocity of support relationships and their durability, density of networks, and the characteristics of the support person. The functional aspects of social support may be described by the subjective quality of support relationships, the sources of support, and the types of support. From their review of the literature, Barrera and Ainlay (1983) identified six categories or types of social support: material aid, behavioral assis- tance, intimate interaction, guidance, feedback, and positive social interaction. Further discussions of social support can be found in a number of review articles and books (Beels, 1981; Bruhn & Philips, 1983; Ell, 1984; Gottlieb, 1981; Heitzmann & Kaplan, 1983; Liem & Liem, 1978; Thoits, 1982). The homeless have long been considered a disaffiliated and socially isolated group. Long before the term "social support" was coined, Sutherland and Locke (1936) discussed the isolation of the homeless from family and other groups. They noted that over half had little 0r no contact with their parental families, most had never married or were isolated from their spouses, and that the homeless developed few close 28 personal relationships. Solenberger (1914) also presented data indica- ting that high proportions of the homeless were unmarried. Modern researchers have continued to examine social support among the homeless. This has consisted, primarily, of gathering data on objective variables such as marital status. Research indicates that most of the homeless are single (Bassuk et al., 1984; Fischer, 1984; Kroll et al., 1986; Ropers & Robertson, 1984; Solarz & Mowbray, 1985a, 1985b), although homeless women may be less likely to be single (Crystal, 1984). Bassuk et al. (1984) found that 74 percent of their sample of shelter guests had no relationships with family members and 73 percent had no friends to provide support. Forty-percent of the respondents reported that they had no relationships with anyone. Of those respondents who had a history of psychiatric hospitalization, 90 percent had no friends or family. Fischer (1984) also concluded that homeless men had impoverished social networks compared to a sample of men in general households. Forty-five percent of the homeless reported no contacts with friends, compared to seven percent of the general sam- ple. Similarly, 31 percent of the homeless claimed no contacts with relatives, compared to four percent of the household men. Finally, two- thirds of the homeless had formed no confiding relationships, and none had more than two confidants. In contrast, only one-third of the house-2 hold sample had no confidants, and one quarter had three or more confi- dants. Solarz and Mowbray (1985a) reported that only about a third of the shelter guests they surveyed felt that they had a lot of friends. In addition, less than half (41.6%) had contact with a friend at least weekly, while just over half (52.9%) had at least weekly contact with a relative. 29 The most comprehensive study of social networks among the homeless has been that of Cohen and Sokolovsky (1983) in their study comparing a sample of homeless Bowery men with men living in single room occupancy hotels (SROS). This study did not, however, examine subjective proper- ties of social support systems, and its generalizability is somewhat limited because it focused only on elderly men. The researchers collected information on network size and configuration, as well as on the frequency, duration, transactional content and directionality of social networks. Cohen and Sokolovsky present a slightly more optimis- tic picture of the social lives of the homeless than have some other researchers. They report that half of their homeless sample had contact with at least one kin member. In addition, although Bowery men had small networks, they had more transactions per contact than did the SRO men. However, comparisons between the two groups indicated that SRO men had more outside non-kin and kin contacts, many more contacts with females, and reported being lonely less often. Thus, to date, research on social support systems of the homeless has been somewhat limited. Research on the quantitative aspects of social support has consistently found that the homeless have impover- ished social support systems. However, little information is available on subjective aspects of social support, or on the types of social support received. In this study, information was gathered on both objective and subjective measures of social support. Summary - Part I In summary, there are a number of frameworks which can be used to provide important information for describing and understanding the 30 homeless. In order to obtain a comprehensive picture of homelessness, it is necessary for research to integrate several approaches to studying this problem. In this study, a wide array of descriptive information was gathered on the homeless. Information was obtained on background and demographic characteristics including psychiatric morbidity, preci- pitating cause of homelessness, and recent residential history. In addition, information was gathered on objective and subjective charac- teristics of the quality of life of the homeless, and of their social support systems. This concludes the discussion of the research pertinent to addressing the first aim of the proposed study. The second aim of this research was to systematically examine the effects of different research procedures on return rates for follow-up appointments by homeless per- sons. In the next section, the second aim of this research will be addressed. Part II - Follow-up with the Hbleless There are a number of occasions when follow—up contact with service clients or research participants might be desirable. For service provi- sion, additional contacts are often required to effectively follow through on service or treatment plans. Both researchers and service providers may wish to observe individuals longitudinally to determine the effectiveness of services or intervention programs, observe differ- ential client outcomes, determine the current status of former clients, obtain information about client satisfaction with services received, assess the need for additional services to previously served clients, 31 and to help identify unserved client populations (Reagles, 1979). Researchers can, to give a few examples, use follow-up contacts as an empirical tool to observe changes in behavior or attitudes over time, examine issues of reliability and validity of measurement instruments, and to study developmental processes. The second aim of this study was to systematically examine the effects of different research procedures on whether or not homeless interview participants returned for a scheduled follow-up appointment. In this section, some of the reasons for conducting follow-up research with the homeless will be discussed, followed by a discussion of methods of obtaining higher follow-up rates and a brief review and critique of the follow-up literature. An important limitation of past research is that it is almost exclusively cross-sectional. Because of the mobility of the homeless population, and the difficulties inherent in tracking people who have no fixed address, researchers have generally avoided doing longitudinal research with this group. Bachrach (1984b) notes that a barrier to research is that it is "often difficult to inaugurate epidemiological inquiries and to track study subjects who have already been identified" (p. 913). Arce et al. (1983) also mention the extreme difficulty they encountered in relocating shelter residents. Most information available on residential patterns, mobility, social support systems, and psychi- atric status of the homeless relies on retrospective interview accounts and/or reviews of archival records. While some more qualitative studies of the homeless indicate longer term interaCtions with studied indivi- duals, these accounts have not systematically assessed changes in 32 individuals or their living situations over time (e g., Hopper & Baxter, 1981). In studies of the homeless, follow-up studies can be an effective strategy for obtaining information about residential patterns and his- tories, helping to identify and describe the various sub-populations among the homeless, measuring involvement in the mental health system, and observing relationships between different aspects of social support systems and such variables as psychiatric status, residential status, or general health status. In addition, such studies may be used to examine the effectiveness of various service programs or interventions on the establishment of permanent housing and reduction of recidivism into psychiatric hospitals. An obvious barrier to obtaining follow-up information on the home- less is their mobility. This includes movement between different areas within a city, and thus to different service or catchment areas, as well as movement from city to city (Bachrach, 1984c; Ball & Havassy, 1984; Ropers & Robertson, 1984). Another important factor is that many of the homeless are mentally ill, and thus may have greater difficulty in remembering appointments and meeting obligations. Furthermore, a lack of economic resources may make it difficult to obtain transportation to follow-up appointments, particularly if the individual has relocated to a different area of the city. Efforts to make follow-up contacts by telephone are likely to be ineffective. A general lack of permanent housing makes it less likely that the homeless client has been able to provide a telephone number for future contact, and the very poor often are unable to afford any telephone service. Ball and Havassy (1984) point out that the struggle to meet basic survival needs in the urban 33 outdoors makes it especially difficult for the homeless to keep clinic appointments. Clearly, it is imperative that methods for increasing follow-up return rates be developed if homeless individuals are to be effectively monitored or studied over time, or if services are to be adequately provided. Dillman (1978), in his excellent book on designing mail and tele- phone surveys, discusses why people respond to interviews. This dis- cussion is also pertinent to thinking about why individuals might agree to participate in a follow-up contact, and to developing methods to maximize the likelihood that they will follow through on that agreement. According to Dillman, the process of obtaining participation can be viewed as a case of "social exchange." According to the theory of social exchange, the actions of individuals are motivated by the return that the individual expects those actions to bring from others. Briefly, behavior is a function of the ratio between the perceived costs of performing an activity and the rewards which the actor expects to receive from the other party at a later time. Response, or participa- tion, may be maximized then by minimizing costs, maximizing rewards, and establishing trust that the rewards will be provided. Dillman suggests a number of ways of maximizing participation by considering these tenets of social exchange theory. Respondents may be rewarded if the interviewer shows positive regard for the participant, gives verbal appreciation, uses a consulting approach, supports the respondent's values, offers tangible rewards, and makes the interview or questionnaire interesting. Costs to the respondent may be reduced by making the task appear brief, reducing the physical and mental effort 34 that is required to complete the task, and attempting to eliminate chances for embarrassment, implication of subordination, and any direct monetary cost. Finally, trust may be established by providing a token of appreciation in advance, and by identifying with a known organization that has legitimacy. It is difficult to determine from published research accounts how each of these issues has been addressed in previous research on the homeless. However, the importance of establishing rapport with homeless research participants has been mentioned by researchers (e.g., Bachrach, 1984b). Researchers such as Baxter and Hopper (1981) who employ field observation methods, have perhaps placed the greatest priority on this issue. Reported participation rates for randomly sampled participants range from 40 percent (Cohen & Sokolovsky, 1983) to 98 percent (Fischer, 1984). Higher rates of participation appear to be dependent, in part, on the offering of tangible rewards. For example, in the study by Cohen and Sokolovsky which had only a moderate response rate, apparently no compensation was offered for participating. In contrast, in Fischer's study which had a very high response rate, respondents were offered a gratuity of five dollars. Review of Follow-up Research Research on follow-up methods has generally focused on ways of obtaining returns of mailed questionnaires (Amour & Bedell, 1978; Futrell & Lamb, 1981; Hinrichs, 1975; Jones, 1979; Miner, 1983; Stafford, 1966), participation in telephone interviews, or both (Davis & Yates, 1983). Comprehensive reviews of these kinds of studies can be found in Dillman (1978), Kanuk and Berenson (1975), and Linsky (1975). 35 Several studies which specifically address the issues of increasing returns for follow-up appointments or which compare different methods for increasing compliance will be briefly reviewed below. Fruensgaard and his colleagues (1983), in a Danish study, followed up 70 unemployed patients consecutively admitted to an emergency psy- chiatric department. While they were in the hospital, participants in the study were informed that they would be called later for two inter- views; in six months, and again one year later. The researchers report that after the appointments were made (6 months and 1 year later), only 54 percent of the participants appeared for their six month follow-up appointment, and only 46 percent attended their one-year follow-up appointment. Most participants were subsequently interviewed in their own homes after more assertive attempts to contact them. The authors suggest that the main reason that the written request for attending a follow-up appointment was not successful was that participants were reluctant to come in contact again with the psychiatric department. This suggests that for these individuals, the costs of participating, i.e., the anxiety produced from returning to the treatment site, exceeded any rewards which might be accrued from participation. In addition, current drug or alcohol abuse prevented some participants from attending the meeting. In a French study, LaHarpe et al. (1983) compared three methods for inducing alcoholics who had previously missed a return appointment to subsequently return to a health center for contact. The three methods were used successively on different groups Of patients at the health center. In the first stage, letters were sent to the patient's treating physician, notifying them that the patient had missed his or her 36 appointment, and leaving the decision to call the patient in for treat- ment to the judgement of the physician. In the second stage, letters were sent to selected patients who were believed to be particularly motivated to return for an appointment; and in the third stage, a letter was sent to all patients. Results indicated that the most effective method for increasing returns for appointments was that of sending letters to all patients (45 percent return), followed by sending a letter to motivated patients (36 percent return), and sending a letter to treating physicians (21 percent return). Without additional informa- tion about specific research procedures, it is difficult to assess this study in terms of Dillman's framework. In addition, a number of methodological issues make this assessment difficult. With respect to the physician contact group, it is possible that physicians did not follow through and contact clients. With the "motivated" group approach, it is possible that the identification strategy was not accu- rate. Overall, the study may suffer from the same problem cited for the Fruensgaard study. That is, the costs of returning to the treatment site may have exceeded any potential rewards of contact. Three client follow-up methods were compared by Warner et al. (1983). A total of 1100 clients who had received treatment from a community mental health center were each randomly assigned to experimen- tal groups; face-to—face interview, telephone interview, or mailed questionnaire. Follow-up assessment was planned to occur 180 days after intake. In the face-to-face interview group, appointments were made by telephone or by making visits to the client's home and then scheduling an interview. For those in the mailed questionnaire group, three mail 37 contacts were made before the client was classified as a nonrespondent. A maximum of five attempts were made during different times of the day to contact each client in the telephone interview group before he or she was considered to be a nonrespondent. Overall, a 34 percent response rate was obtained. The greatest response was found for the face-to-face interview group, with 49 percent of clients in this condition completing interviews. Response rates were less for the mail questionnaire (30 percent) and the telephone interview (25 percent). Overall, a large percentage of the clients could not be contacted. Thirty-eight percent in the face-to-face interview group could not be contacted, 68 percent of the telephone interview group could not be contacted, and 35 percent of the mail questionnaires were returned undelivered. Refusals to par- ticipate were low for those in both the face-to-face and telephone groups (7 to 13 percent); however, 34 percent of the mailed question- naires were never returned. Once again, it is difficult to assess this study in terms of social exchange without additional information about research procedures (e.g. wording of contact, etc.). In addition, the high number of clients who could not be contacted makes it somewhat difficult to interpret the results. The face-to—face interview, how- ever, was clearly superior in obtaining the highest rates of participa- tion. In this condition, personal contacts were made by research staff. This approach may have maximized the appearance that the researchers regarded the respondents positively, an intangible reward. In addition, the personal contact provided more opportunities to give verbal appreciation to the participant. In each of these studies, samples included individuals likely to be similar to persons who may be part of the homeless population; that is, 38 alcoholics and users of psychiatric and mental health services. Results indicate that certain procedures may increase the probability that participants will return for follow-up interviews or appointments. It is apparent that merely scheduling meetings in advance is not sufficient to ensure that those meetings will be attended. In the current study, comparisons were made between using cash or material goods incentives in obtaining returns for follow-up appointments. Thus, participants received tangible rewards for their participation. In order to estab- lish trust, these rewards were provided prior to the interview. Intangible rewards were maximized by using a consulting approach and showing positive regard for the participants. While the interview was lengthy and required that participants provide personal information (and hence was "costly"), efforts were made to reduce costs by making the tasks easy to understand and complete. While no direct monetary costs were required from the participants, they needed to provide their own transportation to the interview site. It was anticipated that the tan- gible rewards which they received would offset any such costs. In addi- tion, the use of a more permanent appointment card, which was less likely to be accidentally destroyed, was examined. Research Objectives A review of research conducted on the homeless indicates a number of gaps in the literature. The majority of studies have a number of methodological limitations, and tend to focus on a limited range of characteristics. 39 As described previously, the current study had two major aims. The first aim was to gather descriptive information about the homeless popu- lation using a sample of guests at a temporary shelter. In order to accomplish this aim, information was collected on: 1. Background and demographic characteristics 2. Psychiatric morbidity 3. Subjective and objective quality of life 4. Social support systems While research indicates that respondents are more likely to parti- cipate in research if they receive a tangible reward, it is not known whether cash or material goods make the most effective incentives. The second goal of the study was to systematically examine the effects of different research procedures on whether or not interview participants returned for scheduled follow-up appointments. In order to address this aim, the following research questions were examined: 1. Are cash incentives or incentives of material goods more effective in obtaining returns for follow-up appointments? It was predicted that respondents in the "cash" group would be more likely to return. Cash may be more rewarding because it indicates that the researcher trusts the participant to make his or her own purchase decision, and thus implies positive regard. 2. Are participants more likely to return if they are given perma- nent-type appointment cards instead of traditional paper appointment cards? It was predicted that more participants would return in the perma- nent-type appointment card group. The permanent-type appointment card used in this study was more difficult to lose or destroy. In addition, the use of this type of card, which may appear to be more valuable than 40 typical paper cards, could enhance the appearance that the role of the participant was important. This would increase the intangible rewards for the participant by emphasizing the positive regard with which the researcher held the participants. It could also help to establish trust that the researcher would follow through with providing the incentive upon return for the follow—up interview. 3. Are participants who are more satisfied with services they have received within the interview setting (i.e. the temporary shelter) more likely to return for a subsequent interview at the same location? It was hypothesized that the costs of returning to the interview setting would be less if the prior experience in that setting was posi- tive. Therefore, it was predicted that those who were more satisfied with their experiences in the shelter setting would be more likely to return. CHAPTER II METHOD Setting Established in 1981, the Coalition on Temporary Shelter (COTS) was organized to address the needs and problems of the homeless in Detroit.1 The coalition consists of a number of social and human service agencies and churches in the Detroit area. In July 1982, COTS opened a temporary shelter with a nightly capacity of approximately 45 guests. In April, 1985, the shelter moved to a larger facility which accommodates as many as 72 individuals each night. COTS shelter serves men, women, and families. This temporary shelter facility provides a number of services to clients, including aiding them in locating more permanent housing. Individuals come to the shelter after being referred by local social service agencies or by self-referrals. After their arrival at COTS, guests meet with a case planner who identifies any service needs of the client and develops a plan to address those needs. For example, a client may need help in obtaining general assistance payments, or in locating available housing. Once the service plan is successfully completed (or sufficient satisfactory progress is made), clients check out of the facility (are discharged ). All clients, however, do not successfully follow the service plan. They may choose not to adhere to the plan, they may 41 42 voluntarily leave the shelter without going through discharge proce- dures, or they may be evicted from the facility because of behavioral problems or rule violations. In these instances, their cases are "ter- minated." Thus, clients may leave the facility under two conditions: discharged or terminated.’ Individuals who are restricted by staff from returning to the shelter are "yellow-tagged." Clients may be yellow- tagged when they are considered to be a behavioral risk (e.g., because of violent behavior), have a total of three shelter stays within a cer- tain period of time, or miss curfew on consecutive shelter stays. Subjects. A total of 125 shelter guests were interviewed as part of this study. Of these, 120 participated in the follow-up experiment.2 The sample consisted of 79 males and 46 females.' They had a mean age of 33.4 years (SD = 10.5; Range: 17 to 72 years). Approximately twenty percent were under the age of 25, while fewer than three percent were over the age of 60. Nearly eighty percent of the participants were Black, with the remaining being White (20.8%) or of another racial back- ground. Just over half of the participants (54.1%) had graduated from high school.3 About a fifth of all participants (22.6%) reported that they had completed some college classes. Nearly a quarter (22.8%) of the men in the study were veterans, representing 14.4% of the sample. No women had been in the armed services.4 Shelter records indicated that these participants stayed at the shelter an average of 16.7 days, with a range of from one to 95 days. 43 Over half (53'6%) of all participants had records of arrest; 26.4% had records of incarceration in jail and 13.6% in prison. Extensive information on criminal history and behavior is contained in Appendix A. A total of 150 individuals were approached to participate in the study. Of these, 88 were men, and 62 were women. Eighty-one (92.0%) of the men, 81 and 47 (75.8%) of the women agreed to participate. The overall participation rate for the study was 85.3%. Of the 128 individuals who agreed to participate, 125 actually completed the interview. The interviews of two men were terminated by the researcher, one because of a language barrier, and one because the individual was apparently mentally retarded and had difficulty under- standing the questions. Another woman withdrew because she became too tired to complete the interview. Thus, a total of 83.3% of those approached for participation completed the interview, or 89.8% 0f the men and 74.2% of the women. Non-Participants The 25 non-participants (including the three individuals who with- drew after beginning the interview) were compared to the participants on a number of characteristics. These comparisons are summarized in Table 1. Most of the non-participants and participants were Black, although the proportion of Whites was higher among the non-participants t(148) = 2.07, p < .05. A greater pr0portion of the non-participants were females t(148) = 2.56, p < .05. There was no statistically signi- ficant difference in the percentages of non-participants and partici- pants with histories of psychiatric hospitalization. Data on the ages of non-participants were not available. However, the mean age of the 44 Table 1 Characteristics of Non-participants versus Participants Race Black White Other Gender Female Male Previously in Michigan State Psychiatric Hospital % of Non-Participants (n=25) 52.0 40.0 8.0 64.0 36.0 20.0 % of Participants (n=125) 78.4 20.8 0.8 36.8 63.2 16.0 45 non—participants did not appear to be significantly different from that of the participants. A number of reasons were given by the 22 individuals who did not wish to participate in the interview. These reasons are presented in Table 2, along with the percentages of non-participants for whom each reasons applied. Research Design and Procedures This research study had two majorcomponents. The first was an interview study in which extensive information was gathered on a number of areas. The informatiOn included data on objective and subjective quality of life, psychiatric morbidity, and social support. The measures used are described fully in the next chapter. The second major component of the study was an experiment. Using a 2 X 2 factorial design, comparisons were made between different methods of eliciting returns for a follow-up appointment. The two factors to be varied were type of payment (cash or material goods) and type of appointment card (regular or permanent) received. Participants were randomly assigned to one of four conditions. In order to control for gender effects between conditions, assignment to each condition was stratified by sex. Each cell had an‘n of 30 (11 women and 19 men). The research design is outlined in Figure 1. At the post interview, participants received either a cash payment of five dollars or a package of material goods with a retail value equal to that of the cash incentive. Participants assigned to the material goods condition selected, from a pool of available goods, the items they would receive upon return for the follow-up interview. Selections were 46 Table 2 Reasons for Non-Participation (n=22) Didn't like idea of interview/not interested - 95.4% (Apparent) Psychotic episode/mental illness - 27.3% Leaving town (can't return for follow-up) - - 9.1% Compensation not high enough --------- 9.1% Ill, not feeling well ------------ 4.5% Note. Percentages total to more than 100 because more than one reason for not partiCipating could be cited for each participant. 47 .ucmu acmEucwoaam mo waxy xa ucmsxmn we on»? ”cmvmmv _wucmewcmaxw .H mczmwm ONH u m cm H m so u m 4wpe_me new; peepceu .cewueczmwwceu x__5mw :e cewpescewcH .xpmwxce eweeca use .cewpmeueEem .cewmmeceee mcwmmemme umwwxeecu eepesxm wmewmewecexme eepw-u;mwm .acw_ we we_.a=c weeewm Le wwecm>e emmmwmmm em—< .xuewem ecu .mcwmze; .wwmm .mecmecmemecw use eczmwmw .meececww use xcez we mcweEee ewww we xpwwmsc paeee mmcwwmew emmmemm< .eue .e:-3e__ew Lew emccsuec uceawuwpcee Lenpezz .ucmewewpcee we meet ecu xmm .3mw>ceucw e:-3ewwew we memo .epm .Lee53c cewpeuwwwpceew .eEec Lezew>cepcw .pceawewucea we meet ecu xem .3mw>cep:w we memo eecweueo :ewpeELewcH meczmeez we xcmsszm m m—eew we ecweccewumeso “Leaazm wewuem ofiueum wwwg we xew_a=o m>weaafin=m ummgm Lm>eu zmw>ceucw-pmem umecm Le>eu 3mw>cepcuueca mezmme 30w>LmHCH ecumeez 55 x 3ww>wmusw s=-zo_.os x 3mw>smusw we_e_se emceemwswse< see: .mEeuw sewpeewmwuem seewesm ueuawesw emw< .eeeesw we meesuem .zseumws em: sepwesm .xseumws _ewuseuwmes .sew>esee weswewsu .sewuerewuuw> weswswse .xsepmws em: mzsu use weseewe .xseumws ewseewsexmg .aseemws sewees ._m>mw sewueesue mswuuwusw .eeeu xsepmws wesemseq m>wmsepxu .peeusee we xeseacesw mswuawusw .mxsezpes op uepeswses msemsea useuseeew ewes messy se sewuessewsw .qwsmsewuewms we museusesew use .awsmsewpewes we xuwuesewems .szesx usew 3e; .seusmm .ewsmsewpewes mswuawesw .mxseZHms “seesaw weweem ea umpeswEes weauw>wusw seem so sewuessewsw .uepes usesesm weweem sew; sewueewmwuem .useeezm weweem m>wuemes se\use .wesewpeee .eeseumwmme weewueese .sewpeELewsw use eew>ue .ewsmsewseeeeu mswuw>esa msemsea we ueuseues meEez .mxsezuws wewuem uzeee sewueELewsw uesweueo sewueELewsH mesumeez we ase552m seesaweeoav m e_eew sewueEsewsH usaesmxeem use uwsqesmeemo we mswessewumwao useeezm weweem mw eswessewummao useeaam wewuem Nu eswessewummzo psoaszm weweem mLzmmmZ 56 3mw>swpsw azuzewwew x zmw>sepsw wewuwsw emceemwcwsee see: .mapepm emsesumwu use .msewpeeswsw _esew>eses .xeem we semsew mswuzwesw .xeum sepwmsm so sewwessewsw .mweuwemes eueum sw msewu -erweHwemes ewspewsexme “men so epeo .mswumee nausewwew sew stapes uwsez yes so smsuesz we sewuewuese ueuawesw .smzew>seusw saw; use useexee sew; sewuuewmwpem we mswues useewewuses ueswepeo sewpeEsewsH mesemeez weixue553m fleesswesouv m e_eew useeem exeese see_esm mwoe museemm sawee: weusmz we useapseawo mmszmemz we>wsus< seweeewe>m 3ew>sepsw ecumeez 57 and the internal consistency of that scale was then examined to see if the new item fit the scale empirically. Items were also moved from one scale to another if they correlated mOre highly with another scale than with its own scale, and they fit rationally in the second scale. Several of the measures contained a priori defined rational scales. These rational scales were analyzed to determine whether they fit the obtained data. In general, the goal was to develop independent scales with high internal consistency. However, the foremost criterion was that scales make "content sense." The measures and the final scales are described in the following sections: subjective and objective quality of life measures, psychi- atric morbidity measures, social support measures, and measures of vari- ables hypothesized to be related to return for follow-up interviews. Copies of the measures used in the initial interview can be found in Appendices O and E. (Quality of Life Both objective and subjective measures of quality of life (QOL) were used in this study. In the following section, the subjective measures of QOL will be described. This will be followed by a discussion of the objective measures of QOL. Subjective_9uality of Life Ratings Assessments were made of global quality of life and of quality of life in various life domains. The subjective quality of life measure used in this study was a modification of the quality of life measures used by Andrews and Withey (1976). Twenty-five items were selected from the Andrews and Withey pool of items for the measure in this study. 58 Several criteria were used to select items. First, items were used that were predictive of general or overall quality of life in the Andrews and Withey studies. Second, items were included to tap a broad range of life concerns. Third, certain items were included because they were of particular interest for the study's population. Finally, two new items were developed to assess areas of satisfaction which had not been covered by Andrews and Withey, and which were specific to the poor or homeless (e.g. contacts with social service or welfare agencies). In general, the wording of items was preserved from the original studies. However, some items were modified to reflect this study's population. For example, housing items were changed to reflect the fact that respondents were currently undomiciled. The original "delighted" — terrible response scale was used. However, the "delighted" option was expanded to read delighted or extremely pleased." This was done because respondents and interviewers did not respond well to the word delighted during the pilot phase. In each of the life domain scales, participants were asked to evaluate their QOL with respect to their current feelings, taking into account what had happened in their lives over the past year. Using the scale development procedures described earlier, scales were developed to reflect perceived quality of life in the life domains of housing, work and finances, family, self, leisure and independence, and safety. The final scales are presented in Table 4 along with the item-total correlations and internal consistency reliabilities. Two items on the subjective QOL measure did not fit well into any ‘scale. These items assessed satisfaction with contact with social 59 Table 4 Life Domains Quality of Life Scales Leisure and Independence Quality of Life Scale Alpha = .78 Corrected . Item-Total Item Correlation How do you feel about: The amount of fun and enjoyment you have .58 The responsibilities you have for , .41 members of your family The things you do and times you have .56 with your friends Your independence and freedom .61 The way you spend your spare time .62 Work and Finances Quality of Life Scale Alpha = .80 Corrected Item—Total Item . Correlation How do you feel about: How secure you are financially .67 Your employment situation .53 The income you have .70 Your standard of living .56 60 Table 4 (continued) Life Domains Quality of Life Scales Safety Quality of Life Scale Alpha = .50 Corrected _ Item-Total Item Correlation How do you feel about: How secure you are from people who might .33 steal or destroy your property About your personal safety .33 Family Quality of Life Alpha = .75 Corrected Item-Total Item Correlation How do you feel about your: Close adult relatives .60 Family life .60 Housing Quality of Life Scale Alpha = .55 Corrected Item-Total Item Correlation How do you feel about the: Place where you stayed before the shelter .37 Places you have lived over the past year .43 Privacy you have .30 61 Table 4 (continued) Life Domains Quality of Life Scales Self_9uality of Life Scale Alpha = .78 Corrected Item-Total Item Correlation How do you feel about: Yourself .60 Your health and physical condition .41 What you are accomplishing in your life .26 Your emotional and psychological well-being .43 The way you handle problems that come up - .33 .34 How much you are accepted and included by others 62 service agencies and the chances of getting a good job if looking for one. In addition to the scales assessing perceived quality of life in various life domains global or "overall measures of quality of life were made. Andrews and Withey favored the use of a two item global well-being scale using the "delighted-terrible" scale. In this measure, the same question about overall quality of life is administered twice; once at the beginning of the interview, and again at the end. The two scores are then averaged to form the scale score. Andrews and Withey found this measure to be one of the most sensitive of a set of alterna— tive measures that was used to obtain ratings of respondents current life-as-a-whole. This measure was also used in this study as a measure of global quality of life.7a Cronbach's alpha for this scale was .73. ijectivegguality of Life In addition to the subjective indices of quality of life described above, objective indices of QOL were assessed in the areas of housing, work and finance, safety, and self (includes physical and mental health). For the most part, items were created for the purposes of this study, or were adapted from the measures used by the Michigan Department of Mental Health in their studies of the chronically mentally ill home- less (Solarz & Mowbray, 1985). Two versions of the objective quality of life measures were developed; one for the pre-interview and another for the post-interview. In general, the post-measure simply updated the information obtained in the pre-interview. 9 Housing information included shelter use history, recent residen- tial history, and homelessness history. Data gathered in the area of work and finances included sources of income, amount of income, and 63 recent work history. Several objective measures of physical health were obtained. These included the number of times the respondent saw a physician during the previous year (and whether or not medicines had been prescribed during the past six months), alcohol use (including information on the frequency of use as well as treatment history) and use of illegal drugs, including marijuana. (Measures of mental health status are described in the next section.) Safety quality of life was assessed by gathering data on criminal victimization during the previous six months. This measure was based on the National Crime Survey which is sponsored by the Bureau of Justice Statistics of the U.S. Department of Justice Statistics and has been administered annually by the U.S. Bureau of the Census since 1973 (U.S. Dept. of Justice, 1983). The survey gathers information on the nature and extent of crime, characteristics of victims, and characteristics of criminal events. Respondents are asked for information about incidents of victimization which occurred during the previous six months. A subset of four criminal offenses was identified for inclusion in this study. The selected offenses (robbery, assault, threats of assault, and property theft) were chosen to represent a range of per; sonal crimes, as well as property offenses. Within each offense cate- gory, additional information was gathered with respect to the nature of the offense. From these victimization data, scales were developed to reflect the amount of criminal victimization experienced by respondents.8 One-item scales measured the number of times participants had been victims of each of the assessed offenses. In addition, a scale was developed to 64 measure the number of times each participants had been a victim of any one of these offenses during the previous six months. This was done by simply adding the numbers of victimization incidents in each offense category to form a total. Psychiatric Morbidity m The SCL-IO is a ten item psychological self-report symptom scale designed to assess psychological symptom status (Nguyen et al., 1983). It is a shortened version of the longer SCL-90-R, a widely used self- report symptom inventory. Respondents were asked to indicate how much they had been distressed by each of ten symptoms or problems over the past seven days. Responses could range from "not at all" to "extremely." The SCL—IO consists of three subscales representing depression (six items), somatization (two items), and phobic anxiety (two items). The item content of each subscale is presented in Table 5. Reliability analysis with the data obtained in this study confirmed the a priori scale structure. Internal consistency scores of .85 were obtained for the Depression Subscale, of .69 for the Somatization Subscale, and of .73 for the Phobic Anxiety Subscale. Interscale correlations are presented in Table 6. ' Nguyen et al. (1983) present normative data on the SCL-IO based on 3,628 clients of community mental health centers, public health centers, and freestanding mental health clinics. This population scored a mean of 14.54 out of a possible maximum score of 40. This indicated that respondents were bothered, overall, between a little bit and 65 Table 5 SCL-IO Subscales Corrected Item—Total Scale Correlation Alppg Depression ------------------------- .85 Lonely .61 No interest in things .51 Blue .70 TenSe or keyed up .66 Feelings of worthlessness .65 Lonely even when with people .67 Somatization ------------------------ .69 Weak in part of body .53 Heavy feelings in arms or legs .53 Phobic Anxiety ....................... (.73 Afraid in open spaces .58 or on the streets Afraid to go out of house alone .58 Scale Depression Somatization Phobic Anxiety 66 Table 6 Intercorrelations Between SCL-IO Subscales Depression Somatization Phobic Anxiety 1.00 .62 1.00 .50 .67 1.00 Notes. Correlations have been corrected for attenuation. All correlations are significant at p < .001 67 ”moderately" by the symptoms described in the measure. A coefficient alpha of .88 was obtained for the SCL-IO, indicating that measure has good internal consistency. Normative data for the SCL-IO are not available on a shelter popu- lation. However, the Michigan State Department of Mental Health used the Brief Symptom Inventory (Derogatis & Melisaratos, 1983), a longer version of the SCL-IO, in studies on the health status of shelter guests, and results indicated that this population was able to use the instrument effectively. Psychiatric History ,E§ychotropjc medications. Participants were asked whether they had been prescribed any medicines during the previous six months and, if so, the names of those medications. Prescribed medicines were coded into categories of psychotropic medications, or as non-psychotropic medicines. ,ngchiatric hospitalization history. Information on psychiatric hospitalization history was obtained from three sources. The first source was self-report information gathered during the interview. In addition, access was gained to official records of psychiatric hospital- izations within the Michigan state system through the Michigan State Department of Mental Health. Finally, shelter intake records, which included information about prior psychiatric hospitalization, were re- viewed. These data were combined to form a measure of prior psychiatric hospitalization. If the participant admitted to prior hospitalization during the research interview or during their shelter intake interview, they were coded as having a history of hospitalization. Similarly, if 68 their name appeared in Department of Mental Health records, they were coded as having been previously hospitalized. Social Support Subjective Social Support The primary measure used to assess subjective quality of life in this area was based on a measure developed by Bogat et al (1983). This measure has been used with a variety of populations, including adults and children. The measure was modified to reflect the needs of this study. Subjective measures of social support were made in the areas of affective/emotional support, instrumental/practical assistance, advice and information, and companionship. Within these areas, data were gathered on satisfaction with the quality and quantity of the support received. In order to ensure comparability, these satisfaction ratings were made using the same "delighted—terrible" scale that was used to make subjective ratings of quality of life. These items were combined to form a Social Support Quality of Life Scale using the rational- empirical methods described previously. This scale contained five sub- scales. First, subscale scores were calculated as measures of satisfac- tion with each of the four assessed types of social support. Next, a one-item subscale assessing overall satisfaction with the quantity and quality of social support was included (See Table 7). Further information was obtained to assess the level of reciprocity in each support relationship. These ratings were made for all relation- ships identified. Respondents indicated whether the exchange of support in the relationship was equal, whether the respondent provided more 69 Table 7 Social Support Quality of Life Scale Alpha = .88 Corrected Item-Total Item Correlation How do you feel about your: Amount of companionship .57 Quality of companionship .53 Amount of advice and information .58 Quality of advice and information .59 Amount of practical support .65 Quality of emotional support .70 Amount of emotional support .69 .60 Quality of emotional support Amount and quality of social support overall .67 70 support, or whether the nominee provided more support. These ratings were then aggregated across all nominees, and the percentages of rela- tionships falling into each reciprocity category were calculated. Finally, respondents reported the importance of their relationship with each network member on a scale that ranged from "extremely important" to "not at all important.” Importance of relationship ratings were aggregated across all nominees, and the mean importance of relationships calculated. Objective Social Support Several measures of objective social support were obtained. Objec- tive measures of social support included marital status, church member- ship and attendance, and participation in voluntary organizations. Data were also gathered on the number of good friends and frequency of con- tact, as well as on the number of relatives in the area and the frequency of contact with them. As noted above, information was gathered on several types of social support. Respondents were asked to indicate who provided them with companionship, advice and information, practical assistance, and emo- tional support. Two questions were used to nominate names to each category of positive support. The number of unique names given within each category of support was calculated to form a measure of network size within each category of support. As many as ten names could be nominated for each of the two questions asked about each type of support. Thus, up to twenty names could be given for each type of support, or a possible maximum of 80 positive supporters if no names were given more than once. The number of unique names of persons named 71 to positive support categories formed an overall measure of positive network size. In addition to gathering information about these areas of "positive support,” participants were asked for names of persons who provided them with "negative support." These were individuals who made their lives difficult. Additional objective information was obtained about each person who had been named as providing some type of support (positive and/or nega- tive). This included the relationship to the respondent, the gender of the supporter, and how long the respondent had known the supporter. Information on the relationships of nominees were grouped into categories of nuclear and non-nuclear family, friends, or others. Data were then aggregated across all nominees, and the percentages of nomi- nees in each network falling into each of these categories were calcu- lated. Gender data were also aggregated across all respondents and the percentages of male and female network members calculated. Another characteristic of network members is whether they are "specialists" or "generalists." Specialists provide only one type of support, while generalists may provide several types of support. In this study, network nominees were coded as Specialists if they provided only one type of support. That is, they were named to only one cate- gory, such as companionship. If they provided more than one type of support, they were coded as generalists. Scales were then constructed to indicate the percentages of supporters in each category who were specialists versus generalists. 72 Return for Follow-Up Measures Experiences at COTS Shelter Information was gathered from shelter records on numbers and types of behavioral infractions incurred during the shelter stay, discharge status, and whether participants had been "yellow-tagged" (see Appendix F). Satisfaction with COTS Temporary Shelter A measure of satisfaction with COTS shelter was developed. This Shelter Satisfaction measure was based on the eight-item Client Satis- faction questionnaire (CSQ-B) developed by Nguyen et al. (1983). The CSQ-8 has been extensively field tested on over 3,500 users of both inpatient and outpatient mental health services. Nguyen et al. reported an internal consistency value .87 for the scale, concluding that it had excellent internal consistency, was well-received by respondents, and was applicable to a wide range of service settings. For the purposes of this study, a subset of three of the eight CSQ- 8 items were used. These items measured satisfaction with the amount of help received, an assessment of whether the respondent would return to COTS if he or she needed these services, and overall satisfaction with services received. Two additional items were added to the scale. These items measured feelings of safety in the shelter setting, and how much the participant liked the staff of COTS. Items were combined into a scale using the general scale develop— ment methods described earlier. A Cronbach's alpha of .74 was obtained for the scale (see Table 8). 73 Table 8 Shelter Satisfaction Scale ' Alpha = .74 Item _— How safe do you feel in this shelter? How satisfied are you with the amount of help you have received? If you stay at a shelter again sometime, what are the chances you will come back to this shelter? How do you feel about people who work at the shelter? ‘ Overall, how satisfied are you with the services you have received at the shelter? Corrected Item-Total Correlation .50 .55 .33 .50 .66 74 Satisfaction with the Interview At the conclusion of the interview, participants were asked how they felt about the interview and the payment they would receive for their participation in the study. The questionnaire consisted of four items: two rated satisfaction with the payment, and two rated satisfac- tion with the interview. The items were combined to form an Interview Satisfaction Scale. The coefficient alpha obtained for the scale was .61 (See Table 9). It was thought that the respondents might be good eassessors of whether they would return for their subsequent (follow-up) appointment. Therefore, at the conclusion of the interview, participants indicated the chances that they would return for the follow-up interview. This was done as part of the Interview Evaluation. Responses could range from definitely would to "definitely would not" return for the follow- up appointment. Reliability of Measures Reliability can be assessed in a number of ways. Information on internal consistency, an indicator of one type of reliability, has been presented with the discussion of measures. Another indication of the reliability of measures is test-retest reliability. In this studV, the interval between test administrations was six weeks. It was expected that mean changes would occur in most measures because of real instabil- ity in the phenomena being measured. Therefore. test-retest reliability was not assessed. A concern sometimes voiced by those providing services to the home- less. as well as by researchers, is whether or not information obtained 75 Table 9 Attitude Toward Interview Scale Alpha = .61 Corrected Item—Total Item Correlation What did you think about the interview? .41 (very interesting to very boring) How satisfied are you with the payment? .46 (very satisfied to very dissatisfied) How useful are the payments? .34 (very useful to not at all useful) What did you think of the interviewer? .39 (like a lot to not like at all) 76 from homeless persons is reliable. This is particularly true when in- formation is being gathered about sensitive areas such as mental health history, use of illegal drugs, or criminal behavior. In order to assess the reliability of self-report information obtained in this study, com- parisons were made between self-report responses and archival data. Data were available to assess the reliability of information about the history of psychiatric hospitalization (Michigan Department of Mental Health computerized records, COTS intake records), and about various aspects of criminal history (Michigan Department of Corrections prison files, Michigan State Police conviction data). Psychiatric and criminal histories are areas where one might expect to obtain less reliable self-report information. Comparisons between self-report and archival records revealed that for both criminal history and psychiatric history information, the respondents provided more ex- tensive information than could generally be found in archival records.9 Thus, results indicated that these respondents were relatively reliable sources of this kind of information. By extension, it can be concluded that they were also reliable sources of information about less sensitive topics. CHAPTER IV DESCRIPTIVE RESULTS In order to simplify the presentation of this large amount of data, results will be grouped in terms of life domains of housing, income and finances, safety, social relations and social support, and self (in- cluding mental and physical health). After the general descriptive information is presented, these data will be used to develop a taxonomy of the shelter users in this study. Unless otherwise indicated, the information presented in tables is based on an N of 125. Housing Quality of Life ijective Housing Quality of Life Shelter Use History For a large minority of the shelter guests (41.9%), the current stay was their first in a shelter or mission. For over half (56.8%) the current stay was their first during the previous 12-month period, with the great majority (92.8%) having stayed at a shelter three or fewer times during that time. The first reported shelter stay was in 1964. Most, however, experienced their first shelter stay in 1983 or later (88.7%). Thus, overall, shelter use was a relatively recent phenomenon for this group. Additional information on shelter use history is presented in Table 10. 77 78 Table 10 Shelter Use Histopy Total number of stays during the past year % of sample One -------------- 56.8 Two -------------- 25.6 Three ------------- 10.4 'Four or more --------- 7.2 M = 1.87 SD = 1.54 Total number shelter stays (a = 124) % of sample One -------------- 41.9 Two -------------- 28.3 Three ------------- 15.3 Four or more --------- 14.5 M = 2.39 (I) U II 1.99 79 Recent Residential History Mobility. As can be seen in Table 11, the majority of the partici- pants had been residents of the Detroit area for a year or longer. However, there was also a transient minority who had been in the area for less than a month. While participants were relatively stable with respect to maintaining residence within one city, they were highly mobile within that area. On the average, respondents had stayed in four places in the six-month period prior to their shelter stay. Just 12.2% had stayed at only one previous address in the past six months. Infor- mation on residential mobility is summarized in Table 11. Homelessness history. The majority of participants considered themselves to be homeless at the time of the interview (70.4%). In addition, 42.0% reported that they had been homeless in the past, with the number of previous periods of homelessness ranging from one to seven (see Table 12). The mean number of previous periods of homelessness for those who had been homeless in the past was 3.0 (SD = 2.3). Although the majority considered themselves to be homeless at the time of the interview, most of the respondents (80.3%) felt that they "probably wouldn't" or "definitely wouldn’t" stay at a shelter again in the future. Prior residences. Participants had lived in a variety of settings during the six months prior to their shelter stay (see Table 13 ). The majority (76.8%) of the participants had mostly been living in a house or apartment durino this period, although a few (8.0%) had been staying mostly in shelters or on the street during the previous six months. While approximately half of the participants had stayed in a house or apartment the night before coming to the shelter, 35.2% had spent the 80 Table 11 Residential Mobility Length of residence in Detroit % of sample Less than one month --------- 12.0 One to less than six months ----- 5.6 Six months to less than 12 months - - 4.0 More than a year ---------- 78.4 ’Number of residences during past six months (N = 123] % of sample One ................. 12.2 Two ----------------- 25.2 Three ------------- 20 3 Four ------------- 13 8 Five ------------- 12 2 Six or more ------------- 15-3 M = 4.03 SD = 3.92 *Time at residence before shelter % of sample Less than one month --------- 75.2 One to less than three months s - - - 10.4 Three to less than twelve months - - 7.2 More than one year --------- 7.2 81 Table 12 Number of Times Homeless in the Past (n = 100) % of sample None --------------- 58.0 One ---------------- 17.0 Two ---------------- 6.0 Three --------------- 6.0 Four or more ----------- 13.0 _ = 1.27 SD = 2.12 82 Table 13 Recent Residential History Primary residence type for the past six months % of sample House or apartment --------------- 76.8 Room or hotel ----------------- 9,5 Shelters, street, abandoned buildings, etc. - - - 8.0 Group living (e.g. drug rehab program) ----- 3.2 Jail or prison ----------------- 2.4 Residence type for night prior to shelter stay % of sample House or apartment --------------- 50.4 Room or hotel ----------------- 11,2 Shelter, street, abandoned building, etc. - - - 31.2 Group living (e 9. drug rehab program) ----- 3.2 Other (includes bus, airport, hallway, church) - 4.0 83 previous night either in another shelter, on the street (including the woods or an abandoned house or vehicle), or in other transient settings (e.g., airport, bus, church, or apartment hallway). Respondents reported a number of reasons for leaving the place where they had stayed the night before coming to the shelter. The most common reason cited for leaving was that this place was only temporarv. Additional reasons for leaving the last place staved are presented in Table 14. These reasons for leaving the last residence refer pply to the place stayed the nioht before coming to the shelter: they are not necessarily the events that precipitated the current incidence of homelessness, or a long history of homelessness. Supjective Housing Quality of Life The Housing Quality of Life (QOL) Scale measured satisfaction with privacv and residences over the past year. Ratings were made using the seven-point delighted-terrible scale. The mean score on the Housing QOL Scale was 3.9 (SD = 1.38), indicating that participants generally felt "mixed" about their recent residences. Using Pearson correlations, relationships were examined between a number of objective variables related to housing and scores on the Housing QOL scale. These correlations are presented in Table 15. Among these variables, Housing Scale scores were most highly related with the total number of times a person had staved in a shelter before. That is, those who had a greater number of shelter stays were generally lfiéé satisfied with their housing situation over the past year. 84 Table 14 Reasons for Leaving Last Place Stayed % of sample Temporary residence only --------------- 26.6 Interpersonal conflict with household members ----- 18.4 Referred to COTS, desired COTS' services - - - e - — - 15.3 On the street, needed shelter ------------- 15.2 Economic reasons (e.g. couldn't pay rent) ------- 12.0 Desire for independence, place of one's own ------ 6.5 Criminal victimization, physical abuse, unsafe conditions ---------------- 5.6 Overcrowded ---------------------- 5.6 Evicted ------------------------ 4.8 Disaster (e 9. fire) ----------------- 3.2 Discharge from program or hospital ---------- 2.4 Exceeded number of allowed days at another shelter - - 2.4 Note. Percentages add up to greater than 100 because participants could provide multiple reasons for leaving the last place stayed. 85 Table 15 Correlations Between Housing QOL Scale and Objective Housing Variables Variable 3 Number of: Times in shelter past year -.22 ** Times ever in shelter -.23 *** Places lived past six months -.18 * Cities lived in past year .04 Consider self to be homeless .15 * Length of time lived in Detroit -.13 * p < .05 ** p < 01 *** E < :005 86 Finances and Employwent Quality of Life ijective Finances and Employment Quality of Life 122-fie Participants reported a wide range of sources of income during the previous six months (see Table 16). Nearly half (49.6%) of the partici- pants reported receipt of public assistance (e.g. welfare, AFDC); this was the most frequently reported source of income. The next most fre— quently reported source of income was work, with 47.2% of the respon- dents indicating that they had received money from working during the past six months. Sixteen percent of the respondents indicated that they had some source of illegal income during the previous six months. While partici- pants were not directly asked the illegal source of income, many of the respondents volunteered this information. Illegal sources of income included selling controlled substances (including their own prescribed medicines), shoplifting for personal needs, stealing items to sell, and leaving restaurants without paying for meals. Drug trafficking was the most commonly mentioned source of illegal income. Participants also reported their largest source of income during the previous mpnpp (see Table 17). The most frequently mentioned source was public assistance, with 29.6% reporting that this had been their main source of income during the past month. This was closely followed by work as a main source of income (23.2% of participants). Additional information was obtained on respondents total amount of income during the previous year (see Table 18) Nearly three-quarters (74.2%) reported that their income for the last year was less than $5,000. A minority of 87 Table 16 Sources of Income in the Past Six Months Sppggg % Receiving Public Assistance ------------- 49.6 Work ------------------- 47,2 Family ------------------ 40.0 Social Security Income ($51, $501) - - - - 16.1 Illegal sources -------------- 16.0 Friends ------------------ 14.5 Savings ------------------ 11.2 Panhandling -------------- .' - 11.2 Unemployment compensation --------- 5.6 Veterans benefits ------------- 4.8 Plasma center --------------- 3.2 Returnable bottles ------------ 3.2 Child support/alimony ----------- 0.0 Other (e.g. selling belongings) ------ 3.2 NOte. Percentages total to over 100 because respondents often indicated that they had more than one source of income. 88 Table 17 Largest Source of Income Duripg the Previous Month Spprpp % Receivjpg Public Assistance ------------- 29.6 Work ------------------- 23.2 Family ------------------ 10.4 Social Security Income (SSI,SSDI) ----- 12.8 Illegal sources -------------- 8.8 Friends ------------------ 3.2 Savings ------------------ 1.6 Panhandling ---------------- 2.4 Unemployment compensation --------- 2.4 Veterans benefits ------------- 0.0 Plasma center --------------- 0.8 Returnable bottles ------------ 0.8 Child support/alimony ----------- 0.0 Other (e.g. selling belongings) ------ 1.6 No income past month ----------- 2.4 89 Table 18 Amount of Income During Past Year Total income in the_past year % of sample Less than $1,999 -------------- 29.2 $2,000 to $2,999 -------------- 22.5 $3,000 to $4,999 -------------- 22.5 $5,000 to $7,999 -------------- 15.0 $8,000 to $9,999 -------------- 3.3 $10,000 to $14,999 ------------- 2.5 $15,000 or more -------------- 5.0 Income earned by working past year % of sample None -------------------- 50.0 Less than $1,999 -------------- 30.0 $2,000 to $2,999 -------------- 7.5 $3,000 to $4,999 -------------- 3.3 $5,000 to $7,999 -------------- 3.3 $8,000 to $9,999 -------------- 1.7 $10,000 to $14,999 ------------- 2.5 $15,000 or more -------------- 1.7 Note. A small minority of participants refused to answer these questions, some apparently because they had extensive illegal sources of income or had earned work income while receiving public assistance benefits. 90 the participants reported that they had income of $15,000 or more during the previous year (5.0%). lo nt Participants were asked to report how long it had been since they had worked at a job that lasted two weeks or longgr (see Table 19). 0f the respondents who had a work historv, one third (34.5%) reported that they had worked within the previous six months, although the maiority of the participants (58.8%) had been unemployed for over a year. Half of the respondents reported that they had earned some money from work during the previous year (not including illegal sources of income, but including odd jobs and sporadic employment). The majority of these individuals (60.0%), however, had earned less than $1,999 from work during that period (see Table 18). Participants reported that thev usually held a variety of jobs when thev worked (see Table 19). The majority of participants usuallv worked at unskilled or skilled blue-collar jobs, although a significant portion had held white—collar jobs. The most frequent iob classification held by the resoondents was general laborer. Subjective Finances and Employmentyguality of Life The Work and Finances Quality of Life (00L) Scale assessed how participants felt about their economic and employment situation. The mean score on the Work and Finances QOL Scale was 2.6 (§Q = 1.38), indi- cating that participants felt between "unhappy" and "mostly dissatis- fied" with their finances and emplovment situation. Nearly three- quarters (73.4%) of the respondents indicated that they felt "mostly 91 Table 19 Work History Last employment held for at least two weeks % of sample Currently working ------------- 3.4 During past month ------------- 10.9 1 to < 6 months ago ------------ 20.2 6 to < 12 months ago ----------- 6.7 1 to 2 years ago ------------- 14.3 2 to < 3 years ago ------------ 7.6 More than 3 years ago ----------- 37.0 Usual type of job % of sample General laborer (e.g. warehouse work, handyman, yardwork, general factory work) --------- 35.5 Clerical, secretarial ---------------- 14.9 Food service, cleaning (e.g. cook, waitress, housekeeper, dishwasher, janitor) -------- 12.4 Skilled blue-collar, craftsperson (e.g. crane operator, truck driver, welder, foreman) ----- 9.9 Personal or protective services (e.g. barber, cab driver, babysitter, security guard) ----- 9.9 Sales, cashier -------------------- 5.8 Health services (e.g. nurse, nurse's aide) ------ 5.8 Professional, technical (e.g. electronic technician, musician, nutritionist) ------------- 5.8 Note. The above percentages are based on the 96.0% of participants who indicated that they had a work history. Four percent of the respondents had never worked for pay at a job lasting two weeks or longer. 92 dissatisfied" or worse about their work and financial situation (i.e. scored 3.0 or lower). Relationships were examined between Work and Finances 00L Scale scores and a number of obiective indicators of work and finances quality of life. These Pearson correlations are presented in Table 20. The variable most highlv correlated with these scales was receipt of public assistance income during the previous six months. That is, individuals who reported receiving this kind of incomes also tended to report being more satisfied with their financial and employment situation. Safety Quality of Life Objective Safety Quality of Life Objective safety quality of life was assessed through information on criminal victimization. Participants were asked whether thev had been a victim of a robberv or mugGing, an assault, a threat of violence, and/or burglary or theft during the previous six months. Over half (54.4%) of the respondents reported being victimized at least once during the previous six months. Of those who had been victimized, 52.9% indicated that they had been a victim of more than one type of crime during this period, with burglary/theft being the mast commonlv reported tvpe of victimization. Overall, 19.2% of the participants reported that someone had threatened to beat them up or harm them (i.e., with a knife, gun, or other weapon) during the previous six months. Those who reported that they had been threatened with violence during the previous six months had been threatened an average of 2.5 times (SD = 1.9), with a range of from one to over seven incidents of victimization. Additional 93 Table 20 Correlations Between Finances and Employment QOL Scale and ijective Work and Finances Variables Variable r Public assistance past 6 months -------- .15 * Income from panhandling and/or illegal sources past 6 months ------ -.3O ** Length of time since last worked ------- -.09 Total income past year ------------ -.07 Work income past year ------------- -.08 * p < .05 ** p < .001 94 information about robbery, assault, and burglary Victimizations is reported in Tables 21 through 23. Subjective Safety Qualigy of Life Participants rated their satisfaction with their safety on the seven-point delighted-terrible scale. The mean score on the Safety QOL Scale was 3.9 (S2 = 1.48), indicating that, overall, participants felt mixed about their safety. The relationships between subjective ratings and various objective variables were examined using Pearson correlations. These results are summarized in Table 24. Self Quality of Life Physical Health General Health Participants were asked to rate their health as compared to other people their age. As shown in Table 25 the maiority of respondents rated their health as being good or excellent, although over a third rated their health as fair or poor. Most of the participants felt that their health was the same or better than it was two years ago, although over a third felt that their health had gotten worse during this period. Respondents had a high rate of contact with physicians during the previous vear. Only 20.8% said that they had ppt been to see a doctor during this time (see Table 26). A large minority of participants (42.4%) reported that they had received prescriptions for medicines during the previous six months. Manv of these individuals (63.6%), however, indicated that thev currentlv were ppt taking their medicines 95 Table 21 Robbery/Mugging Victimizations During Past Six Months Percentage of sample victimized ------- 21.6 Mean number of Victimizations/victima - - - - 1.48 Location of Last Robbery % of victims Own home ------------------ 26.9 Area near home --------------- 15.4 Other person's home ------------ 7.7 On street ----------------- 38.5 Other (e.g. bar, store) ---------- 11.5 Relationship to Robber % of victims Stranger ------------------ 61.5 Known -------------------- 38.5 a A maximum of seven total Victimizations were coded. 96 Table 22 Assault/Beating Victimizations During Past Six Months Percentage of sample victimized ------- 19.2 Mean number of Victimizations/victima - - - - 2.25 Location of Last Attack % of victims Own home ------------------ 45.8 Other person's home ------------- 12.5 On street - - - - - - - - 4 - 3 e ------ 20.8 Other (e.g. motel, bar, vehicle) ------ 20.8 ’Relationship to attacker % of victims Stranger ------------------ 20.8 Known - -'- - - -' ------------- 79.2 Type of attack for most recent assaultb % of victims Rape (% of females only) ---------- 14.3 Shot -------------------- 8.7 Knifed ------------------- 4.3 Hit with object -------------- 50.0 Hit, punched, etc. ------------- 54.2 TA maximum of seven total Victimizations were coded. b Percentages add up to greater than 100% because the incident may have included multiple types of assault. 97 Table 23 Burglary/Theft Victimizations During Past Six Months Percentage of sample victimized ------- 34.6 Mean number of Victimizations/victima - - - - 2.49 Location of Last Burglary % of victims Own home ------------------ 65.1 Shelter ------------------ 7.0 Other person's home ------------ 14.0 On street ----------------- 2.3 Other ------------------- 11 6 a A maximum of seven total Victimizations were coded. 98 V Table 24 Correlations Between the Safety QOL Scale and Victimization Variables Variable E Total number times victimized -.28 *** Total number types of Victimizations -.22 ** Victim of robbery .03 Victim of assault -.20 * Victim of threat -.16 * Victim of burglary/ th9ft -.21 ** * p < .05 ** p < .01 *** p < .001 99 Table 25 General Health Status Self-report health rating % of sample Excellent ------------------ 21-5 Good -------------------- 39.2 Fair -------------------- 27.2 Poor -------------------- 12.0 Currentlhealth status compared to health two years ago % of sample Better now ------------------ 29-6 Worse now --------------- 36-0 100 Table 26 Number of Physican Visits During Past Year % of sample None -------------------- 20.8 One -------------------- 22.4 Two to three ---------------- 23.2 Four to ten ------- ' --------- 18.4 11 to 23 ------------------ 8.8 24 or more ----------------- 6,4 M = 4.9 §D = 6 9 Note. A maximdm of 24 physician contacts were recorded. The actual range of visits reported was from none to more than 97. 101 Table 27 Main Reason for not Followinngrescrjption (fl = 36) % of sample No longer needed -------------- 41.7 Can't afford to refill ----------- 33.3 Don't like side effects ---------- 13.9 No Medicaid card - - — - - - L ------- 5.6 5.6 Ran out .................. 101 according to prescription, often because they could not afford to refill their prescription (see Table 27) The types of medicines for which prescriptions had been received varied. Over half (52.8%) of those who had received prescriptions had been prescribed analgesics, such as Motrine or Tylenol. Nearly a third (30.2%) had received prescriptions for psychotropic medicines. The remaining prescriptions were for a variety of medications including high blood pressure medicine, anti- biotics, insulin, cold medicine, and vitamins. Alcohol Use The majority (89.6%) of the participants admitted to previously drinking alcoholic beverages. Information on the frequency of alcohol use among participants is presented in Table 28. Nearly a quarter (22.3%) of the 112 alcohol users had at some time been in treatment for alcohol problems (including detoxification, in- patient rehabilitation, outpatient programs, and halfway houses). Over half (54.2%) of those who had been in alcohol treatment programs had been so within the past six months, with one participant reporting that he was currently under treatment for alcohol problems. Another 20.8% had been in alcohol treatment programs between six and twelve months previously. A quarter had not been in treatment for a year or longer. In addition, a number of alcohol users (18.8%) had been involved in Alcoholics Anonymous (AA), with one third of those being current members. Drug Use A majority of participants (78.4%) admitted to using marijuana at some time, with 62.2% reporting that they had smoked marijuana during the previous month. A quarter of all participants (25.6%) had used 102 Table 28 Frequency of Alcohol Use During Past Month % of sample Daily -------------------- 8.9 More than once weekly ------------ 24.2 Weekly ------------------- 9.7 2 to 3 times per month ----------- 22.6 Once per month --------------- 8.9 Not at all ----------------- 25.8a a 40.6% of the individuals giving this response reported that they never consumed alcoholic beverages. Table 29 Frequency of Marijuana Use During Past Month (2 = 98) % of marijuana users Daily - - - -. ---------------- 4.1 More than once weekly ------------ 21.4 Weekly ------------------- 7.1 2 to 3 times per month ----------- 15.3 Once per month --------------- 14.3 Not at all ----------------- 37.8 Note. Refers only to those who have previously used marijuana. 103 marijuana at least weekly during the last month (see Table 29) . Many of the participants (42.7%) reported that they had at some time used illegal drugs other than marijuana such as heroin, cocaine, or LSD. However, only 22.6% of those indicated that they had used any of these drugs within the past months. Overall, 79.2% of the respondents reported that they had used illegal drugs at some time.‘ Nearly a quarter (24.2%) of those who admitted to a history of drug use reported that they had been in a drug treatment program at some time. Over half (54.2%) of those who had been treated for drug problems »had been in treatment within the past six months. Another 37.5% had not been in treatment for a year Or longer. Mental Health Objective Mental Health Variables Psychiatric hospitalization history. Nearly a third (32.0%) of the participants had a history of psychiatric hospitalization. The number of self-reported psychiatric hospitalizations ranged from one to one hundred with half of these individuals reporting that they had experienced only one previous hospitalization for emotional problems (see Table 3O).9a Half of those with a history of psychiatric hospitalization had been in the hospital within the last 24 months. A few (12.5%) had their last psychiatric hospitalization at least ten years prior to their shelter stay, with one participant experiencing his last psychiatric hospitalization over forty years previously. The average age of first psychiatric hospitalization was calculated to be 25.3 years of age, with 104 Table.3O Psychiatric Hospjtalization Histgry (r1 = 40) Number of previous_psyppiatric hospitalizationsfi % of sample 1 ...................... 52.5 2 to 3 ---------------- 22 5 4 or more ----------------- 25-0 y = 3.60 SD = 4.5 Time since last psychiatric hospitalization? % of sample Less than 6 months ------------- 27.5 6 to < 24 months -------------- 22.5 2 to < 5 years --------------- 12.5 5 to < 10 years -------------- 20.0 10 years or longer ------------- 17.5 M = 5.7 years SD = 8.1 a Includes only those with a history of previous psychiatric hospitalization. A maximum of nine hospitalizations were recorded. Number of hospitalizations was determined by taking the greater number indicated by self-report data or Department of Mental Health records. b Time since last hospitalization was determined by using the most recent date of hospitalization indicated from either self-report or Department of Mental Health data. 105 Table 30 (continued) Psychiatric Hospjtalization Histgry Age at firsjjppychiatric hospitalization % of sample Under age 21 ---------------- 25.0 21 to less than 25 ------------- 22.5 25 to less than 30 ------------- 30.0 30 or older ----------------- 22.5 M = 25.3 SD .— 7.5 106 the age at first psychiatric hospitalization ranging from age 11 to age 50 (see Table 3O)10. Use of psychotropic medications. Many of the participants 14;.4%) had received prescriptions for medications during the previous six months. Of those, 30.2% (or 12.8% of the total sample) had received prescriptions for psychotropic medicines. The most common prescription for psychotropics was for neuroleptics (e g. Thorazine, Prolixin, Mellaril) (see Table 31). Half of those who had received prescriptions for psvchotropics during the past six months reported that they were not taking their medicines as prescribed. The most common reason reported for not taking prescribed psychotropic medicines was not liking the side-effects (62.5%), followed by not being able to pay to refill the prescription (25.0%). Subjective Mental Health Variables Self-reportedjpsychological symptoms. Participants indicated how much they had been bothered during the past week by a number of psycho- logical svmptoms on a scale ranging from not at all to "extremelv." The 10—item SCL—IO consisted of a 6-item Depression Subscale, a 2-item Somatization Subscale, and a 2-item Phobic Anxiety Subscale. Scale scores on the SCL-IO are presented in Table 32. Subjective Self Quality of Life Scale A Self Oualitv of Life Scale reflected how respondents felt about themselves. This subscale included ratings of mental and phvsical health, satisfaction with problem resolution, satisfaction with accom- plishments, feelings about themselves, and feelings about acceptance by 107 Table 31 Iype of Prescribed Psychotropic Medicine (2 =16) % of sample Neuroleptics (e.g. Thorazine, Prolixin, Mellaril) --------------- 50.0 Antianxiety agents (e.g. Librium, Valium) - - 25.0 Anticonvulsants (e.g. Dilantin) ------- 18.8 Antidepressants (e.g. Elavil, Lithium) - - - 12.5 Sedatives and hypnotics ----------- 6.3 Note. Figures total to over 100 percent because some participants had been prescribed more than one psychotropic medicine. 108 Table 32 SCL—IO Scores SCL-IO (total) ........... Subscales: Depression .......... Somatization ......... Phobic Anxiety ........ Mean 2.42 2.69 2.04 2.00 1.06 1.21 1.24 Note. Scale: not at all a little bit moderately quite a bit extremely U'l-DWNH II II H II II 109 others. As with the other quality of life scales, ratings were made on the seven-point delighted-terrible scale. The mean score for the Self Scale was 4.6 (§Q = 1.21), indicating that on the average respondents felt between mostly satisfied and "mixed" about themselves. Pearson produce moment correlations between the Self OOL Scale and a number of physical health variables are presented in Table 33. Satis- faction with self was most highly correlated with participants' ratings of their health, as well as with whether they felt their health had improved or deteriorated over the past two years. Relationships between Self OOL Scale scores and a number of mental health variables are presented in Table 34. Self OOL Scale scores were most highlv correlated with the total scores received on the SCL-lO, with individuals scoring high on the SCL-IO generally reporting that they were less satisfied with themselves. Social Relations Quality of Life Objective Measures of Social Support Social Support Networks Numbers of positive supporters. Respondents named an average of 6.0 supporters across all types of social support (Range = 0-24) (see Table 35, It should be noted that just over ten percent of the partici- pants indicated np positive supporters. Numbers of negative suppprters. Participants were asked whether there were anv individuals who made their lives difficult. Nearly half of the participants provided names of negative supporters (49.6%; 110 Table 33 Correlations Between Self QOL Scale and Physical Health Variables Variable r Health ratings -------------- .43 *** How current health compares to health two years ago -------- .37 *** No. times seen doctor past year ----- -.06 Received prescription for medicines during past 6 months -------- .13 Frequency of alcohol use during . past month ------------- -.10 Membership in Alcoholics Anonymousa ------------- -.22 * Frequency of marijuana use during past month ------------- .12 Previous use of illegal drugs other than marijuana -------- -.25 ** Use of illegal drugs other than marijuana during past month ----- -.O6 a Participants reported whether they (1) had never been in Alcoholics Anonymous, (2) had previously been in AA, but were no longer a member, or (3) were current members of AA. * p < .01 ** p < .005 *** p < .001 111 Table 34 Correlations Between Self QOL Scores and Mental Health Variables Variable SCL-IO Total Score ............ Depression Subscale --------- Phobic Anxiety Subscale ------- Somatization Subscale -------- Speech Rating Scale ........... Emotional State Ratings Scale ------ Received prescription for psychotropic medicines during past 6 months History of psychiatric hospitalization — - Length of time since most recent psychiatric hospitalization (for those with history only) l-s .78 .77 .47 .64 .83 .60 .29 .13 * p < .01 ** p < .001 ** ** *‘k ** *‘k *‘k ** .38 * 112 Table 35 Number of People Providing Each Type of Social Support ! Companionship ---------------- 2.76 Advice and Information ----------- 2.06 Practical Assistance ------------ 2.26 Emotional Support -------------- 3.38 Total Number Positive Supporters ------ 6.04 §p 2.7 1.5 1.9 2.4 4.2 113 M = 1.8: SQ = 1.1; Range = 1-6). An average of 8.0% of all supporters named provided pnly negative support. Only one participant indicated that those nominated to his or her network provided oply negative support. This compared to the 46.4% of the participants who nominated pply positive supporters to their social support networks. Data on negative network members are presented in Table 36. Specialists versus generalists. In this study, supporters were coded as specialists if they provided only one type of support. If they provided more than one type of support, they were coded as generalists. Overall, networks were made up of half specialists and half generalists. Additional data on support specialists are presented in Table 37. Relationship of supporter. Over half of the supporters named were relatives. Additional information on the relationships of supporters is presented in Table 38 along with data on the gender of persons nominated to social networks. Participants also indicated which of the individuals nominated to their social support network was most important to them (see Table 38). The majority identified a nuclear family member as the person in their network who was most important to them. Overall, a parent was most frequently identified as the most important person (26.7%), followed by a son or daughter (25.8%), friend or romantic partner (23.3%), or sib- ling (13.3%). A spouse was identified as the most important person by only 5.0% of the respondents. The remaining 5.9% of most important' persons included professionals (e.g. therapist, caseworker, lawver), acquaintances, and individuals with other such relationships to the respondent. 114 Table 36 Negative Social Support Number of negative supporters % of sample None ------------------------ 50.4 1 ------------------------- 26.4 2 ------------------------- 13.6 3 ------------------------- 4.8 4 to 6 ----------------------- 4 8 _ = .90 SD = 1.20 Type of support provided by nominees to social networks Only positive support ------------ 85.6 Both positive and negative --------- 5.4 Only negative support ------------ 8.9 115 Table 37 Percentages of Specialist Social Supporters Iype of support Mean % SD n Overall -------------- 50.0 30.7 119 Companionship ----------- 37.3 38.4 102 Advice and Information ------ 21.3 34.9 103 Practical ------------- 24.7 34.9 104 Emotional ------------ '- 29.7 30.5 111 Note. Percentages are based only on those cases where supporters were named; i.e., where the denominator was greater than zero. 116 Table 38 Characteristics of Social Supporters Relationships of social supporters to participants % of Relationship supporters SQ Nuclear family (e.g., parent, sibling, child, spouse) ------- 47.4 31.0 Any relative (including nuclear family) --------------- 57.2 31.6 Friends - - - - - - - - 4 -------- 30.3 29.0 Others (professionals, acquaintances, landlord, etc.) ----------- 12.5 19.5 Relationship of most important social supporter % of Relationship supporters 'Nuclear family (e.g., parent, sibling, child, spouse) ---------- 70.8 Other relatives ----------------- 3.3 Friends ----------------- ~- - - - 23.3 Others (professionals, acquaintances, landlord, etc.) -------------- 5.9 Gender of social supporters % of Gender supporters SQ Female ------------------ 54.6 26.3 Male ------------------- 45.4 26.3 117 Table 38 (continued) Characteristics of Social Supporters Length of relationship with social supporter % of Time known of supporters SD 6 months or less ----------- '- 10.0 17.4 > 6 to 12 months ------------ 3.6 9.9 > 1 to 5 years ------------- 17.4 21.0 More than 5 years ------------ 69.0 27.6 Note. The above information is based on data provided by the 96.0% (p = 120) of respondents who nominated individuals to their social support networks. ‘h-h 118 Lppgth of relationships. Participants indicated how long they had known each of the individuals nominated to their social network (see Table 38. The majority of all individuals named to social networks had been known to the respondent for over five years (69.0%). This was not surprising given the high numbers of relatives identified by partici— pants as providing social support. Nearlv all (95.0%) of the partici- pants who were able to identify someone as providing social support . named at least one person who they had known for at least five years. About 25.0% nominated only persons whom they had known for more than five years to their social network. Few of the participants named individual network members whom they had known for a week or less (11.5%), with a total of 38.3% of the par- ticipants indicating that persons whom they had known for six months or less were a part of their social network. In some cases, these were individuals whom the participant had met through their shelter stay (i.e.; COTS staff, other shelter guests). FamilyiRelationshjps Information was obtained on family composition and on contacts with relatives. Just over a quarter (27.0%) claimed a steady romantic rela- tionship with either a spouse, girlfriend or boyfriend. Data on marital status are described in Table 39. Most of the participants had children (see Table 39) Over a quarter of participants had children aged five years old or younger (26.4%). Nearly half of the participants (48.0%) reported that thev had children aged twelve or younger, and 60.0% indicated that they had children eighteen years old or younger. 119 Table 39 Family Composition Marital status % of sample Single, never married --------------- 51.2 Divorced ----------- ' ---------- 22.4 Widowed ---------------------- 3.2 Separated --------------------- 18.4 Married, living with spouse ------------ 4.8 Number of children % of sample None ----------------------- 35.2 1 ------------------------ 30.4 2 ........................ 16.8 3 or more -------------------- 17 6 M = 1.4 §_ =1.7 NDmber of children with respondent at shelter % of sample None ----------------------- 79,3 1 ------------------------ 12.9 2 or more .................... 7 3 M = .34 SD = .84 120 The majority of participants (81.6%) reported that thev had rela- tives in the area. On the average, respondents indicated that thev had contact with a relative approximately three to four times a month. Over half of the participants (62.4%) reported that they had contact with a relative at least once during the previous month. Community Involvement Many participants (43.2%) indicated that they had voluntarily attended religious services during the past month, with an average attendance of three to four times. Very few respondents (12.8%) claimed to be involved in clubs or groups. On the average, those who were involved in groups indicated that they were "fairly active" in group activities. Subjective Measures of Social Support Subjective Characteristics of Network Members Importance ratings of social network members were made on a seven- point scale from "not at all important" to "extremely important." The average importance rating was 5.6 (SQ = 1.78), indicating that the average importance of the relationships with those named to the network was between "somewhat important" and "verv important." Over ten percent (13.3%) of the participants rated their relationships with 211 network members as "extremely important." Few (6.7%) of the participants reported average importance ratings of 4.0 or less for their networks, with a score of four indicating that the relationship was "equally important and unimportant." Note that ratings of the importance of relationships with individuals who provide negative support were included in these overall ratings. 121 For each person named in their network, respondents indicated whether the other person provided more support, whether the exchange of support was equal, or whether the participant provided more support in the relationship. As shown in Table 40, respondents felt that they were receiving at least as much support as they were giving in the large majority of their relationships. Subjective Social Relations Quality of Life Subjective social support quality of life. The Social Support quality of Life Scale score provided a measure of overall satisfaction with social support networks (M = 4.7, SQ = 1.1). For the most part, ' participants felt "mostly satisfied" to "mixed" about their social support (see Table 41). Relationships between the Social Support QOL Scale scores and a number of social relations variables are presented in Table 42. Subjective family quality of life. The mean scale score for the Family QOL Scale was 4.2 (SQ = 1.7), indicating that, on the average, participants felt "mixed" about their relationships with their families. Pearson correlations indicated that the Family QOL score was most highly related to frequency of contact with relatives in the Detroit area during the past month, with those having more frequent contact being more satisfied with their family relationships (see Table 43). Leisure and independence_guality of life. The mean score on the Leisure and Independence QOL Scale was 4.4 (§Q = 1.3), indicating that respondents generally felt between "mixed" and “mostlv satisfied" with their leisure time. Information is proVided in Table 44 on the 122 Table 40 Reciprocity of Support % of sample Other person provides more support ----------------- 16.1 Equal amounts of support provided ------- 46.5 Respondent provides more support -------- 37.4 Note. These figures are based on the relationships described by the 96.0% (n = 120) participants who nominated persons to their social support networks. Table 41 Social Support Quality of Life Scale Scores Type of social support M 50 Overall ---------------- 4.7 1.1 Companionship ---------- 4.5 1.6 Advice and Information ------ 4.9 1.3 Practical Assistance ------- 4.6 1.5 Emotional Support -------- 4.8 1.5 Note. Scale: terrible unhappy mostly dissatisfied mixed (about equally satisfied and dissatisfied) mostly satisfied pleased delighted or extremely pleased metho-J II II II H II H II 123 Table 42 Correlations Between Social Support QOL and Selected Social Relations Variables Variable 1 Presence of relatives in area ------- -.11 Frequency of contact with relatives in area ------------ .19 * Marital status --------------- .00 No. of children aged 18 or younger ----- -.08 Attended religious services past month ---------------- .19 * No. of close friends ------------ .25 ** No. positive supporter named to social network ------------ .32 *** Frequency of contact with most important person in social network - - - - .22 ** % of social network made up of negative supporters ----------- -.17 * * p < .05 ** p < .01 *** p < .001 Variable Correlations Between Family QOL Scale and 124 Table 43 Selected Social Relations Variables Presence of relatives in the area ------ Frequency of contact with relatives in the area ----------- Marital status ---------------- Have steady romantic relationship ------ Have children ................ I-s .26 ** .08 .16 * I'M- l'Ol'U AA 00 125 Table 44 Correlations Between Independence and Leisure QOL and Selected Variables Variable Have relatives in the area - - - - - - - - 7 Frequency of contact with relatives in area ............ Marital status ............... No. of children aged 18 or younger ----- Attended religious services past month ................ N0. of Close friends ------------ No. of companionship social supporters - - Frequency of contact with most important person in social network - - - - Receive public assistance (i.e., welfare or $51, 5501) .......... How long since last worked --------- I-s .12 .02 .00 .19 .13 .13 .18 .20 .09 .02 p < .05 126 relationships between scores on this scale and a number of other social relations variables. Global Quality of Life The mean score on the Global Quality of Life Scale was 4.3 (§Q = 1.4), indicating that participants felt between "mixed" and "mostly satisfied" about their lives as a whole. The relationships between Global Quality of Life and the various life domains were examined. Table 45 summarizes the scores on each of these scale. Correlations between the Global Quality of Life Scale scores and scores on the life domain scales are presented in Table 46. To further examine these relationships, a step-wise multiple regression analysis was conducted using the HEW REGRESSION procedure of the Statistical Package for the Social Sciences (Hull and Nie, 1981). The program selected variables for inclusion in the analyses when they met a minimum criteria of an 5 value with a p < .05, and a tolerance level greater than .01. All life domain satisfaction scales were included in the regression analyses. Two life domain satisfaction scales, the Self OOL Scale and the Work and Finances QOL Scale, entered the prediction equation for Global Quality of Life, yielding an 32 of .52. Thus, approximately half of the variance in Global Quality of Life scores was explained by these two scales. A list of the predictors and a summary table for the regression analysis are presented in Table 47. 127 Table 45 Summary of Quality of Life Scale Means Scale Mean .59 Global QOL ----------------- 4.3 1.4 Housing QOL ---------------- 3.9 1.4 Finances and Employment QOL -------- 2.6 1.4 Safety QOL ----------------- 3.9 1.5 Self QOL ------------------ 4.6 1.2 Social Support QOL ------------- 4.7 1.1 Family QOL ----------------- 4.2 1.7 Leisure and Independence QOL -------- 4.4 1.3. Note. Scale: 1 = Terrible 2 = Unhappy 3 = Mostly dissatisfied 4 = Mixed 5 = Mostly satisfied 6 = Pleased 7 = Delighted or extremely pleased 128 m se new; ueuaeEee mum; msewuewmsseu woo. v m we useewwwsmwm msewuewescee ww< .mmw eu oHH Eesw mswmses .seweeaseeee sew uepeessee use; msewpewessee mmesw "muez oo.H we. me. He. ow. oe. we. Kw. museuseamusw use msamwee oo.H we. mm. mm. He. em. we. xwwses oo.H we. oe. me. se. me. ecosssm _ewoom oo.H we. mm. mm. mm. wwem oo.H mm. mm. so. xuewem oo.H me. He. usages—saw use meusesww oo.H we. mswmae: oo.H weeewu mesmuseeeusw xwwseu psoaeum wwem xuewem aseexoweEm mswmue: weeewu e eszmwme _ewoom a mmesesws meweum mwwe we xuwwezo seeZHmm msewuewesseesmusw es eweew. 129 .eee eweeeem weweem ese .eee swwsew .eee ewemwee use museuseamusw .eoo mumwem .eoo mswmze: ewes sewuezee esp ousw umwwusm wes mmweewce> .muez seem. mmwe. weeewmeeew we. we. Nw. eee. Ne.Nm eewe. ewem. See meeeeeww eee ewe: .N ee. me. we. eee. Ne.ew meme. weee. eee wwem .w eeemee mm m m weesw m we mm m eswweeee seem we .eewm ew w meweewwe> meweem ewwe we wewweee ewesee ewwe mewme ewwe we wewweee weeewe we seweeweewe we mweew 130 Typological Analysis In order to obtain a better understanding of the participants in this study, a correlational analysis using Tryon and Bailey's (1970) method (BCTRY) was conducted. First, a selection of 37 descriptive variables (see Table 48) was submitted to an empirical V-analysis followed by a pre-set cluster analysis (Tryon and Bailey, 1975). In developing the clusters, variables were generally dropped from further consideration if their loading was less than .40 and their communality was below .20. In this manner, only the most significant variables were retained for further analysis. Through these analyses, six empirical dimensions or clusters were identified: 1. Criminal behavior 2. Psychiatric history 3. Transiency 4. Criminal victimization 5. Work history 6. Social support These dimensions are described in Table 49. Following the identification of the clusters, data were submitted to an O-Type analysis to develop typologies, or types of homeless persons. This procedure grouped subjects into clusters based on scores on the defining variables from the pre-set cluster analysis (i.e., on the six clusters noted above). When all six of the initially defined clusters were included in the O-Type analysis, a total of 21 groups were identified. As the goal of this analysis was to simplify the presentation of this large amount of 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 131 Table 48 Variables Entered into V—Analysis Sex Race Age Education Health rating Psychiatric hospitalization history Alcohol use Marijuana use Number times robbed Number times assaulted Number times threatened with violence Number times burglarized When last worked Number times in shelter during past year Number cities lived in during past year Type of place lived in most of time during past six months Arrest history Number jail terms Number prison terms Income last year Marital status Whether lived with both parents until age 16 Work income Money from family 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. Social Public 132 Table 48 (continued) Variables Entered into V-Analysis Security Income assistance income Panhandling income Money from friends Money from savings Income Number Number Number Number Number Return Number from selling plasma, collecting bottles of companions of advice and information social supporters of practical assistance social supporters of emotional social supporters of negative social supporters for follow-up appointment of children with participant at shelter 133 Table 49 Final Clusters Derived from Pre-Set Analysis Cluster Loading Cluster 1: Criminality reliability = .67 1. Number of jail terms .71 2. Arrest history .60 3. No. of prison terms _ .58 Cluster 2: Psychiatric history reliability = .68 1. Social Security Income .88 2. History of psychiatric hospitalization .48 3. Health rating -.47 Cluster 3: Transiency reliability = .54 1. Number of places lived past six months .68 2. Number of times in shelter past year .49 3. Income from selling plasma, returning bottles .39 Cluster 4: Criminal Victimization reliability = .57 1. Number of times assaulted _ .58 2. Number of times threatened with violence .53 3. Number of times burglarized .39 134 Table 49 (continued) Final Clusters Derived from Pre-Set Analysis Cluster Loading Cluster 5: Work History reliability = .77 1. When last worked .80 2. Income from work last year .77 Cluster 6: Social Support reliability = .82 1. Number of advice and and information supporters .76 2. Number of practical assistance supporters ..72 3. Number of emotional social supporters .71 4. Number of companionship supporters .64 135 information, the profile types were consolidated by reducing the number of defining clusters. It was decided to use the four clusters of psy- chiatric history, transiency, criminal behavior, and criminal victimiza- tion to describe the participants. These clusters represent character- istics that are often believed to be descriptive of the homeless. Intercorrelations between the clusters are presented in Table 50. When data were resubmitted to the O—Type analysis with four pre-set clusters, a total of eight O-Types, which accounted for 121 of the 125 participants, were identified. Based on their patterns of cluster scores, three of the remaining participants were assigned to O-Type groups by the researcher. Thus, a total of 124 of the participants were assigned to O-Types. The numbers of individuals in each O-Type ranged from two to 57. The O-Type with only two individuals was dropped from further analysis. The following O-Types were identified: 1. Lower Devianqy - Members of this group tended to have the lowest scores on all four clusters. 2. High Victimization - This group generally resembled O-Type 1, except that they had experienced high rates of criminal victimization during the previous six months. 3. High Transiency - Individuals in this group were highly transient. They also exhibited low to moderate rates of criminal behavior and had mental illness histories. 4. High Psychiatric - Individuals in this group exhibited high rates of psychiatric problems. 5. High Transiency, High Psychiatric - Members assigned to this group were highly transient and generally had psychiatric histories. 136 Table 50 Correlations Between Oblique Cluster Domains Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 1 -- .02 .24 -.O3 Cluster 2 .02 -- .12 .33 Cluster 3 .24 .12 —- -.13 Cluster 4 -.O3 ,33 -.13 -- 137 6. High Criminality - Members of this group showed high rates of criminal behavior. 7. High Criminality, ngh Transiency - Individuals in this group were similar to those in O-Type 6, except that they were also highly transient. The final seven O-Types are presented graphically in Figures 3 through 9. The numbers of members within each typology, along with assessments of overall homogeneity are presented in Table 51. Means, standard deviations, and homogeneity for each O-Type within each cluster are presented in Table 52. Finally, comparisons of O-Types on selected variables are presented in Table 53 through 59. .QJOLw muco_>wo Lora; H unable Low moLoum Lm+mapu .m 0L:m_m Lewwapu co_+0N.e.#u_> muc¢_uc0Lh U_Lvo_£ummm Poc_§_Lu ..0.0.0.0.0.0.0.0.0.0.. . 0 0 - Sfi mfififid}3&€€dfi3 Inflflfidlxuflfilv ..0.0.00.0. ..0..00000. unnbufidrxxhxc .Ehficdlnhflcilu . O O O O O O O O O O . vODObopOpODOpOpOpOpOpo ”IBRhfififillnnh ..000.000.00000000000 0.. Na ..vé NV . mu mw.w¢ es.m¢ 1 am 1 so 1 an an mLoum Lm+mawu uwN_uLouce+m 139 ca_weu_e_+u.> .QJOLm co_+0N_e_eu_> £v_z N mamhio Low moLoum vamapu Lw+m3wu. Juan—wCOLH u.Lvo_£ume Poc_s_Lu . ..‘.%.0*. . 0'000'0” $€€€€€fifi€ ....0....0.. . .0. 0 0 0...... ...... . ......0.. 0.0.0... 0.00... .........0. ......0..... 0........ . .....00... ........0.. . .....0000. ......0.... .....0...... ......0...0 ............ ......0... .0... .... .... 00. .... 0.. 0.. ......0. .0 . .. 0.0.... .0. .00.... ... .0.... 0... ...... .........0. ......0.... ........00. ......000. .IB???..§8, 3%.. 0. db . . ..mfifii. ......00. .. .32323333V9 .......... .0... . .0 .. ... mifidd$€€a 00....... .0....... adddtddl .00.... 0 0 0 0 0 0 0 ....«aauuuuuuunuu» 0.0.0.0.0..0. "n.00 v .. bfihfikbhhhhfl .......... ...... .0... .0... ..... . . .0. . . 0.0.0... 0. ....0 0.. ....00. .. . . ... cfiéfiflfiéfiéfia ........... 33% . .' l .v 0L3m_m au am so as am mmLoum Lmemapu uwN_uLoucoem 140 .QJOLw mucw_mCOLh $6.: m wamhlo Low mmLoum Lawmawu .m 0L3m_m Lo+mapu co_+ow.e_+0r> June—«cokh U_L#o_£ummm Foc_s_Lu ......0.0...0.0.0.0.0.. - , a a? w......00.n .......0... mQQQQQQQQQWw ébfiébfibfififim . ......0.... 0.......... .....00.... ......0.... ......000.. 0......0... ......0.... ........0.. ......0.... ........... 0 dddddddtdda .03233333)? 8 .9 .u.......... ......... 1 am mm.sm .... .00..0....0.00...00... .......0... .0.......0. ........... 1 an ........... ........0.. .00....0... ......0.... ......0.... ........... .........0. 0000....... ......0..... ’332?333333 000.000.004.000000000. Nm.wh mmLoum Lw+mapu uwN_uLouco+m 142 .QJOLu Lummm £U_: I mucw_mc0Lh 10.: m mamklo Low mmLoum Laymapu .n 0L:m_m Lmvmapu co_+ew_e_eu_>. Juan—«COLh u_Lvo_£uawm Poe—swuu ....................... ...... 0.0.... .... ......0............... nu nfififi$3§§$§h €€€é€d33k33 3§b§h§$§§§% .00........ ........00. .0......00. ............ ........0.. ....0...... 33333333339 33333333333 3535553555§ mhfififihfifififik 33333333333 1 -. 33333333333 33®QQQQQQQQ ms m. . di...0..... mm mw ....0.....0 Ehhhhfibbhbu ......0.... .......00.. ......0.... ..am ......00... ........... ........0.. .......0... .....3V.... ..... ..... ......3V... .00... 0... .........0. 0......0... ..........0 .0......... ..33...000. .. .0000... ......0.... .....0..... ........... fiéfihbéfiéflfih ..3w '0 ...) ....P.’. D... 0. 0 «m.nw «.+w maLoum Lavmapu umn.ucouco+m I43 .QJOLw m+_poc_E_Lu £U_I . m oamklo Low maLoum wamawu .m oLDm_L wamapu co_+ou_e_+u_> mucw_wcoLh u_Lve_£ume poc_a_Lu ......0.0.0...0.0...0.. .. - .0.....0...0.0.0.0.... 20.00.000.000... 6‘ nhfihhfifikhfifi $§fi§§fi€€fi€€ €€€é€333333 3353$b§§33% .....0.0... ........00. r!?99999?93 ........0.. ......000... ........0.. ......0.... ......0.... .0......00. . .0... .0... ¢d€d€€¢€€€€ 33333333333 NM NV fiddddfmiddl ..........Pu ......0.... ..........0 - - .........0. ........... . . 0......0000 . . 0 0 .....0... . .00.. .00 N m¢ ........... ............. .. . ........0.. mm wt .........0._ ..sm mNWmm mmLoum Lmvmawu uwN_uLouco+m 144 .QJOLw mcoLh LU—I I m+_poc_e_Lu £U_I 5 enable Low moLoum Luemapu .m 0L3m_m La+mapu cattzzkzwu_> monetahri. ugueo:h&$3_ woe—erhu .QQQQQQQQQQ. - - B¢ m§€§€§$$§%w 33333393333 .0........ ...... .... ......0...... ........... .0....... . . ........n%. .........0.. ...33..0.0. ......03... .0.....0... .0. ..0.... ........ .. ............ .......0... .....3V.... ........... .03V....... ...... .... ......0..... ... ......0 .0.....0... .0......... .4%.3%..... ........... ........... o. 0. 0.... .0......... 33333333333 fibfifififlfifififi€ 333V3333333 0 0 I. vm D? ................... . m 0* am .........0. ......0.... .......0... .........3V 0.0000000 . ......0...0 01%........ vflfiflflflfififlfiw 0 0000. ..... ........0 .0000000000000000 mmAmm ..aw .0 uflfiflfifiv.00% 335$3‘3h333 .00.}.0.P.0 0.0.0.)... :1 8h nb.mm maLoum La+uapu ton—ucouco+m 145 Table 51 Number of Members and Homogeneity of O-Types Derived from the Typological Analysis Overall O-Type Number of Members Homogeneity 1 57 91 2 13 82 3 9 78 4 20 .86 5 5 .93 6 5 93 7 10 88 83 2 .94 aSubsequently dropped from further analysis because p was less than 5. 146 Table 52 Means, Standard Deviations, and Homogeneity of the Eight Derived O-Types O-Types Cluster 1 Cluster 2 Cluster 3 Cluster 4 (Criminal) (Psychiatric) (Transiency) (Victimization) 1 M 45.42 44.17 46.04 46.28 _§p 4.78 4.90 3.62 3.25 M .88 .87 .93 .95 2 M 48.03 48.68 46.11 71.48 pp 7.33 4.81 2.38 6.88 5 .68 .88 .97 .73 3 M 51.25 51.93 73.42 45.17 _§p (51.19)a (50.99)a (76.92)a (48.91)a u 4.48 6.27 8.59 4.53 .89 .78 .51 .89 4 M 48.16 65.11 46.65 47.10 §p 5.89 6.33 2.61 4.48 .y .81 .77 .97 .89 5 g 43.72 61.31 64.10 45.83 l§p 3.48 5.73 2.88 1.37 y .94 .82 .96 p .99 6 M 65.25 42.52 46.59 45.20 §p 4.61 3.60 3.58 3.15 5 .89 .93 .93 .95 7 M 63.66 49.90 58.38 47.27 50 (69.82)6 (48.98)a (59.53)a (48.94)a F‘ 7.68 3.59 3.36 3.18 ’ .64 .93 .94 .95 8b M 52.50 75.04 45.62 88.19 50 1.80 0.00 1.53 6.12 E” .98 1.00 .99 .79 a Reflects cluster mean after cases initially ppt included in O-Types were added upon examination of cluster scores. One case was added to O-Type 3, and two cases were added to O-Type 7. b Subsequently dropped from further analysis because of low p. .QJOLm muco_>wo L010; a newbie Low moLoum Lw+mawu .m mLJm_m L0¥MJPU co_+oN_e_vu.> muco_wCOLh u_Lvo_£ummm poc_a_Lu ......0.0.0...0.0.0.0.. , 0 - 8.? “...“..."exe "as...“wxxx. .0000000000. 0000000000.. 0000000000 00000000000 ..0...000000000.000.00. 5.0.0.050. L3.0.0.01. 00.000000... . ....0..0......0..00....., .2 #0 N¢ . mu mm.mv es.w< 1 am 1 so 1 as am wLoum Lw+mzpu uwN_uLouco+m 139 .% w%...... ca_+o~_e_+u_> .saote co_+o~_e_+u_> ss_: N wamhio Low meoum Lw+mawu .¢ 0L3m_m Lw+mspu, mucw_wc0Lh u_L+o_£uanm Poc_s.Lu ....0 ... 0.0.0.. 0.... .3V3333V3V3 .. 0 .. .0 ..........0. 0... .0.... .... ....... ......0.... ..........0. ........... . 0.0.0.0... .0........ .......0.... .........0. ..0.....0... ........0.. ........0... .....0.... .........0.. ......0.... ......0... . .......0.0 .0... 0.. .... ...0. .... 0...... ......0.... .00. 0.0.... .0. 0 ...0. . 0. ...... .... .0... ..0..00.... .0...... ...00.... ......0... . ......0. .....0.. Qfifififififiv... ......3V. . kxfiflfifiv. 0 .0 00 A '8 .0 c€€¢£wdddi 0. . .0. .0. aw 0. 0 % kfiflfifi‘. ...%€€€€€€ 0.00.000... 333333 .003V0 ufififififififiwfifi .".x.".”.“.”.x.w.u 0.0.0.0.... 0...... 0.0. «H.wv am MQLOUM 1.09%: PU th _ muLOmuC0¥m 140 co_+ou_e_+u_> .QJOLw mace—mCOLH £a_I m unable Low meoum Lw+mawu Lw+m3wu Juan—wCOLh u_L+o.£ummm poc.s.Lu .. ......‘.....0........ 33333333333 .....0..... .0......000. ........... .0.....000 . 333333333V; .......0.0 333333. .000... .000... .0. 0 0.. ....0 . .0.... ....00.. .00...... .0....... ......0. 0.000... .0. . 0. .00.... .0..... 00.0... .0...... ......0. .0...... 0.0.00. .3‘333‘ .0. .000 .0000000. 0.- 0... 329’. %€.%??v .00.... 0 «IVQW. 0...... 0.0.. 0... 333% .%€€€€€€§§& .3 .m 0L3m_m nu am so an an mmcoum Lm+mapu vow—uLouco+m 141 .QJOLw u_L+o_£ummm £U_I u wumhio Low mOLoum Lw+mapu Lw+mapu co_+oN_e_vu_> muca_uc0Lh u.Leo_£umwm Poc_e_Lu 0.0.0.0.. ...... - 0 0 - -. 0 .0 0.. 33333.. afifififififififififl hfififififififibfi$ ........00. .......000. ......00... ......0..... 0.0.0.0.... .0000...0.0 .0000...0.0 .00....00.. .000...00.. .....000.... .0.....0.0. .%....00.0. .....0..... .0000....0. 0.0.0.0... .....0000. . ......0...0 ......0.... ........0.. ........... ....03V000. ..0...0..0.. .......0.0. .0000. 0... ........... . .. 0.000000000000000... . 0 a 0 I 0 I I D . I . a nu mm mu . an 0v 1 N~.mw .w wLam_L av am so an maLoum Lw+mapu uwN_uLouco+m 142 .QJOLw Lummm £U_I i muco_mc0Lh £U_I m enable Low maLouu La+u3~u .n 0L:m_m Loemawu co_+ow_e_ru_> mucw_wceth u_Lvo_£umum Poc_6.Lu ..0.0.0.0.0.0.0-0.0. 0.. 33333333333 0.0.0.0.... . .0........ ...00000... . 00.0.00... xhhhfihhbbbfi .000000000000000000000. mmA3. ,ae 0.0.0. .0.0.0.0. .0.‘ €6€é€d€d€¢3 .0......... .0......00. ..00000.0.. ........0.. ......00. . 00.00.00... .......00.. 0 00.0.0... .0 ...... . 0.0..00.00. ......0...0 00.0.0.0... 3333333V033 ...... v . 0 . 0 .0 . 0 .0 33333% bkbbb‘ 00 00000 00000000000 0 0 000 000 00”0000 0 0 déddddiwdda . 0.0 00... 0.00.000... ...... .... .00... .00. 00........ 00.00.0000. .00....0000 ...00000... e. 00...... 0. 0 00 0. ......000. 0 000.0 00. .000000.... bkbbbhhfibbu fihhhfihhhhhfi V —.¢w am moLoum Laemzpu umN_uLouco+m 143 .QJOLw m+_—oc_E_Lu LU_I . m mamhio Low maLoum Lw+mawu .m 0L3m_m Lw+mapu corroflervu; mucflwcob. Ska—Luann. woe—6.5 ,sv . ......... 0.0.. 000.000.0000.. 2.0000000...» $fixxxxxxy3x €¢£33333333 Dbéfléébéééi 0 0 0 0 0 0 0 0 0 0 0 0.0000000000000001000. .0000 0 0 0 0 0 0 0 . ......00.... .....AV.... 00000000000 5 .00000 0000. .......... . mm N0 0.0.0.0.... ..00.0..0 . .000000000. ......0... . .000...000. fidddtdddddl N .mv ......0......0000..0... fidtdddddddt .000000...0. am ..80 mfiflmo .185 muLoum Lm+mapu uwN_uLouco+m 144 .aaogm ace: :3: .. 3.28.56 :3: n 25.—.10 Low meoum vam: 6 .m 0.59.”. Laemapu 5.005330; 30:129.... Utah—comma .05.-TO . ... ......0...0...... - 0 - Ev 00.00....00 ....000000 héébbéééfibty fifléfibfiéfihfiv 0 . 0 . . 0 0 . 0.. . v 0.0.0.0000... .00... ......0...000.000.000o. 0.. . 0 0 0 0 0.0 0 0 ......0... . 0....JV0... 9w!)322¥333 dddddigdlda 0m€0.0.0... 000.000.... .......0.0.. 00.00.00... 33333?33333 JwI2333E3?3 . afifififififififififl . .Lam cm 0? ....0....0000000.000.. m at ......0.... ......0.... 000.000.... 00.0.0.0043 0.0.0.00. . .0......000 .1V....000. .0 000.00.. 00.0.0.0... 00.00.00... .0... 0.... .0... 00.00 $§fifi€fi€€fifi€ .00000000000000000000. mmgflm ..sm uoLoum Lo+u3pu th_tLovco+m 145 Table 51 Number of Members and Homogeneity of O-Types Derived from the Typological Analysis Overall O-Type Number of Members Homogeneity 1 57 91 2 13 82 3 9 78 4 20 .86 5 5 .93 6 5 .93 7 10 88 8a 2 94 aSubsequently dropped from further analysis because 2 was less than 5. 146 Table 52 Means, Standard Deviations, and Homogeneity of the Eight Derived D-Types O-Types Cluster 1 Cluster 2 Cluster 3 Cluster 4 (Criminal) (Psychiatric) (Transiency) (Victimization) 1 M 45.42 44.17 46.04 46.28 50 4.78 4.90 3.62 3.25 5 .88 .87 .93 .95 2 g 48.03 48 68 46.11 71 48 59 7.33 4.81 2.38 6.88 H .68 .88 .97 .73 3 M 51.25 51.93 73.42 45.17 _59 (51.19)a (50.99)a (76.92)a (48.91)a 5 4.48 6.27 8.59 4.53 .89 .78 .51 .89 4 M 48.16 65.11 46.65 47.10 §g 5.89 6.33 2.61 4.48 5 .81 .77 .97 .89 5 n 43 72 61.31 64.10 45.83 §g 3.48 5.73 2.88 1.37 H .94 .82 .96 _ .99 6 g 65.25 42.52 46.59 45.20 £9 4.61 3.60 3.58 3.15 5 .89 .93 .93 .95 7 M 63 66 49.90 58.38 47.27 _59 (69.82)a (48.98)a (59.53)a (48.94)6 H 7.68 3.59 3.36 3.18 ’ .64 .93 .94 .95 8b g 52.50 75.04 45.62 88.19 so 1.80 0.00 1.53 6.12 E‘ .98 1.00 .99 .79 a Reflects cluster mean after cases initially not included in O-Types were added upon examination of cluster scores. One case was added to O-Type 3, and two cases were added to O-Type 7. b Subsequently dropped from further analysis because of low E. 147 AN-H "amenav m.¢ ~.m H.¢ N.m N.e v.m ~.¢ “.8 boo —mpopw m.m~ o.mm o.om o.om o.m N.mm m.om m.m~ gauzoP—o» cow umccaumg & N.ofi m.NH m.mH ¢.NH m.wH N.¢H o.mH o.mH whou um mxmo m.HH m.HH m.HH w.HH m.~H m.HH o.HH ¢.HH cowumuzuw M «.mm m.m¢ m.qm w.©m N.o¢ H.Nm 0.0m o.mm mm< ~.m~ o.ooH 0.0m o.ow o.mN o.mm o.cw m.w~ prco:_z N w.om o.o o.oH o.om o.mm ¢.¢q q.mH H.Hm muwzz n mumm m.om 0.0 0.0 o.om 0.0m m.mm N.mo ¢.oc mFmEmw & N.mo o.ooH o.ooH o.ow 0.0m p.00 m.om o.mm mpms & xmm mNH m CH m ON a ma Km m 4 aucmw>mo ewco-wz saw: mcmch-_= cm.: 59.: saw: c6264 wmxkuo wnxpuo >5 corumEcowcH.msomcmmeumwz ccm uwgamcmoemo mm m—amk 148 .Lmum=_u :xucm_m:wcp= mcu mm:_5mu mpnm_cm> mwcp a ANufi "mmcmmv 5.5 w.m 5.5 5.m o.m N.m m.m m.5 boo 5c_m=o: m.H m.~ m.o 5.5 5.5 5.5 m.o 5.5 5855 :5 $325: 35.5 .5 o.5 m.m N.m m.5 5.5 5.m N.m m.m mmzucos o 5855 um>wp mmumpg N o.~5 o.mN 0.0m o.ooH o.mm m.mm m.om 5.5N mmo_meo; >5maow>mca mam N 5.0K m.No o.ow o.ow 0.00 m.mw m.Nm N.mo mmm_mEO; an 65 55mm cmuwmcou N w.0N 5.0 o.ow o.o5 0.05 5.55 m.N5 ~.H5 c» pma_ 555\mmsoz =5 56>5_ »_pmoe N 5.H m.~ m.H 0.5 5.5 5.5 m.5 m.H 5Lma» 5665 LmNngm cw mmswu N mNH 5 OH m ON 5 m5 Nm m 4<505 mcmcpuw: chwewcu cuxma-_: cuxma ucmwmcwck Ewuuw> xucm5>mo 5558-5: :55: m=6c5-5I :55: 555: £55: 56354 mmx5-o maxh-o an mmpnmwcm> mcwmzo: 5m mpamk 149 0.0m 5.55 5.55 5.55 5.5 5.5 5.55 5.55 555555 N 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5555555 N 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5553 N 5.55 5.5 5.5 5.5 .5.5 5.55 5.55 5.55 5555>55 N 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5.55 555555 555555 N 5.55 5.5 5.5 5.55 5.55 5.5 5.5 5.5 5555 N 555505 c 5555 525555 50 mmu5zom 5.55 5.55 5.55 5.55 5.55 5.555 5.55 5.55 55555 5 5555 5555 msouc5\z N 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5.55 555555 5 5555 555535 555503 N 555 5 55 5 55 5 55 55 a 45:0.» £50.:an _.w:_.E_Lu 29357.5: comma ucmwmcmfi. 5.50.; hucmgmo 5555-5: 555: 55555-5: 555: 555: 555: 55355 5555-5 5555-o 55 55555555 5:5 555: mm wpnmp 150 .555505u 5555555555»: 555 5555550 5505555> 55:5 0 .5555055 5550555: 555555055555 5:5 5505550 5555555> 55:5 5 55-5 855555 0.5 0.5 0.5 5.5 0.m 0.5 5.5 0.5 500 mmoc5cwm\x5oz 0.05 5.5m 0.05 0.05 0.5 5.55 5.mm 0.0 5505555 N 5.55 5.55 0.0 0.05 0.05 m.mm 5.5 0.0 02550055055 N 0.5 0.0 0.0 0.0 0.0 0.55 0.0 0.5 55555505 . .555550 N 555555 505555 0 5555 550505 50 5555000 555 5 55 5 55 5 55 55 m 5<505 5:55h-51 55055550 05555-5: 55555 505555555 25555> >5055>50 5555-5: 555: 55555-5: 555: 555: 555: 55555 50x5-0 5055-0 50 55505055 005 5503 55555555555 55 55555 151 .5555555 5555555555555>= 555 5555555 5555555> 5555 5 55-5 5505550 5.5 5.5 5.5 5.5 5.5 5.5 5.5 5.5 555 555555 55. 55.5 55. 55. 55. 55. 55.5 55. 55555555 55555 5 55. 55. 55. 55. 55. 55. 55.5 55. . 55555555555 55555 5 55. 55. 55. 55. 55. 55. 55.5 55. 5555555555 55555 5 55. 55. 55. 55. 55. 55. 55. 55. 555555 55555 5 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5.55 5555555555 5 555 5 55 5 55 5 55 55 m 5<505 55555-5: 55555550 55555-5: 55555 555555555 55555> >5555>50 5555-5: 5555 55555.5: 555: 5555 5555 55355 5555-5 5555-0 55 555555 55 55555 152 w.NH m.NH o.oH o.o~ o.m¢ N.NN o.o 0.0 muwaoguocuama umowgummga\3 N o.Nm m.NH o.o o.ow o.m~ ¢.q¢ m.wm o.¢H m»g0pmwg Pmp_amo; uwgum_;uxma\z a ©.m~ m.No 0.0m o.o¢ o.oH N.~N H.mm m.mm mgoe go x;\xH mcmshwgme meEm u ~.ok m.Nm o.ooH o.oq o.mo m.- m.mm N.NN mngu mm__w umm: Lm>m n m.mm m.mo o.om o.o~ o.o~ ¢.¢q w.om o.¢~ xz\xH cmzu mgos Fecou_m xcwgu N N.@m m.No o.o~ o.ow o.ow N.©@ N.©¢ o.¢H mgooa mm ;u_mm; acwpmg N mNH m CH m om @ mH Rm m .355... mcmghn w: EFF—=20 .5mean zuxmn— ucmwmcmfr ESQ; Xucmgwo EPLU-?I cow: mcmgp-wz cow: saw: gov: gmzoA mmxh-o maxk-o xn mm_nmwgm> $_mm Km mFQmH 153 .gmpm:_u =>Loumwz uwgpmvcu>m¢= ms“ mmcwwmv mpnowgm> mwzp m Akna ”mmcmmv maxh-o x3 mmpnmwgm> »_mm Aumzcwpcouv Km m_omh o.v ¢.¢ m.¢ N.m m.m m.m m.¢ H.m Joe wpom .o.~ o.~ H.H m.m w.~ m.m o.m o.H m_mumn=m cowumeucsom o.N ¢.H m.H m.N m.~ m.m N.H m.H «Pmumnzm uwaoca ~.N o.m H.m N.m N.m w.m m.~ m.~ mPGUmnzm cowmmmgamo Amufi ”mmcmmv e.m N.N ~.H ¢.N o.m w.m m.~ m.m oH-Jum mNH m CH m ON a ma Km m 4 xucmw>mo s_gu-w1 saw: mcmLF-PI saw: gmwz saw: LszJ mmxp-o 154 m.o v.0 N.o N.o o.H w.o e.H m.o m>wpmmmc « e.m m.~ ~.N ¢.H o.m m.~ m.m o.m Pmcowposm * H.~ w.H m.N v.0 ¢.N m.H m.H N.N owe? a muw>um * w.~ m.m m.~ o.H m.m m.o m.m o.m awzmcowcmaeou * mgmugoaaam meuom $0 mgmnE=z m.om 0.0 0.0 o.o o.mH o.o N.o¢ H.wm guppmcm um :mLu—wcu\3 u o.Hv m.No 0.0m o.oN o.mm N.oo m.wm H.N¢ ma pr mpcmgoa spon\z um>wp u N.Hm m.~m 0.0m o.ow o.om ¢.¢¢ m.mm N.mo umwggme Lm>mc N mNH m CH m om m mH Rm m 4 aucm_>mo swgu-?: ;m_: mcmgp-w: saw: saw: ;m_: Lm304 mmxp-o mmxp-o x3 mwpnmwgm> ugoamzm meuom mm mpnmp 155 AN-H "mmcmmv v.¢ N.m m.¢ m.e N.¢ H.m m.c ¢.¢ Joe mg:m_m4 \mucmucmamucH AN-~ ”mmcmmv N.¢ m.m m.¢ m.m o.m m.m m.¢ m.¢ Joe xpwsmu AN-H "mmcmmv N.¢ N.¢ m.¢ ~.¢ m.¢ o.¢ m.¢ o.m goo gnaw uom m.om m.mm o.wN o.o¢ m.oH m.om N.mm m.om mucmNLN N ~.Nm o.mm ¢.mo o.¢ q.¢o m.~¢ N.mm ¢.wm cwx N q.N¢ o.mfi m.mm o.¢ N.mm o.He H.N¢ ¢.Hm xpwEmN me_u=c N mgmugoaazm mo mawzmcowpmpmm mNH m OH m cm m mfi Nm m 4 aucmw>mo E_LU-PI saw: mcmgp-_: now: now: ;m_: LszA mmxh-o maxh-o x5 mm~nmwgm> “Loaazm meuom num=:_pcouv mm m_nmh 156 .mem:_u =»p__mc_swgu= mg“ mmc_wmu m_nmwgm> mwck a w.m o.m~ o.om o.o o.m H.HH «.mH m.m zuce umm~ _mmm~pw meoucw cwms\z N mm. mN.H cm.H oo. mo. mm. mH. oo. cemwga cw cmmn was?» N mm. mm.m om.H 00. em. NH. Hm. HM. «mem cw :mma Mme?» N o.mm o.ooH o.ooH o.om o.mm m.mw N.ov H.mm mumummggm N mNH m CH m ON 0 MH Km m 4 xu:m_>ma EPLU-PI saw: mcmgh-wz gov: ;m_z saw: gm304 mmxh-o mqak-o xn Low>mzmm _mcwENLu mm m_nmp CHAPTER V FOLLOW-UP RESULTS In this study, a 2 x 2 factorial design was used to make compari- sons between different methods of eliciting returns for follow-up. The two factors which were varied were type of payment and type of appoint- ment card received. At the time of the initial interview, 86.6% of the participants indicated that they "definitely" or "probably" would return follow-up appointment. However, only a total of 23.3% (n = 28) of the participants actually returned for their scheduled follow-up interview. Relationships between return for the second interview and a number of demographic variables were examined. Pearson correlations indicated no statistically significant relationships between race, age, or gender and whether participants returned for follow—up. However, having a history of psychiatric hospitalization was negatively correlated to return for follow-up (5 = .16, p < .05). In the following sections, the experimental hypotheses are examined. Analysis of Hypotheses One and Two The first two experimental hypotheses asked: 1. Are cash incentives or incentives of material goods more effective in obtaining returns for follow-up appointments? 2. Are participants more likely to return if they are given permanent-type appointment cards instead of traditional paper appointment cards? 157 158 A total of 26.7% of the participants who were paid in cash returned for their follow-up appointment, as compared to 20.0% of those who received material goods payments. Similarly, 26.7% of the those who received a permanent card returned for follow-up, compared to 20.0% of those who received a regular appointment card. The rates of return by condition are presented in Figure 10. In order to determine whether either of the varied factors had an effect on level of follow-up returns, a Chi-Square analysis was per- formed. The Chi-Square, with return for follow-up as the dependent variable and type of payment and type of appointment card as the two independent variables, failed to detect any statistically significant differences in the cell frequencies (i.e., p < .05 or better). Thus, neither Hypothesis One nor Hypothesis Two was supported. Analysis of Hypothesis Three The final experimental hypothesis asked: 3. Are participants who are more satisfied with services they have received within the interview setting more likely to return for a subsequent interview at the same location? Satisfaction with COTS Shelter A Shelter Satisfaction Scale was developed to assess how positively or negatively participants felt about their experiences at COTS. Scores on the scale could range from five to twenty. The average score on the Shelter Satisfaction Scale was 15.9 (S9 = 2 6), indicating a high over- all level of satisfaction with the shelter. 159 .vcmo ucmsycwoaam yo max“ ucm acmexma we max“ xn scape; a:-3oF_ow ucoocmm .oH mczm_d Awm n av ANH " av Aofl n mv Nm.mm No.0N NN.©~ Sacco Aofi u mv Am u mv AN u mv NN.0~ No.0m Nm.mm ocaeaEcaa ANH u mv Am n mv Am ” av No.ow No.0H NO.Om canmam acau we max» 412 to 18 years ....... over 18 years ........ 12. (If have children aged 12 or younger) HHERE DO YOUR CHILDREN (aged 12 or younger) LIVE? (Indicate number in each category - 1-6 I code actual number 7 I 7 and above 8 I not applicable no kids with respondent ....... with spouse/other parent . . . with other relative ..... foster care ......... doesn't know ......... other (specify) Social Support Questionnaire - Part I - page 2 239 Participant ID‘_ THE NEXT QUESTIONS HAVE TO DO HITH HHAT YOU DO IN THE COMMUNITY. 13. HAVE YOU ATTENDED RELIGIOUS SERVICES VOLUNTARILY DURING THE PAST MONTH? (because you wanted to) 1 I yes 2 I no ( go to Q 15) I4. HOH OFTEN HAVE YOU ATTENDED SERVICES DURING THE PAST MONTH? (Include voluntary attendance only!) I I every day more than once a week once a week 2 or 3 times in past month once in past month n/a hasn't attended services in past month OW‘UN N I N I 15. DO YOU BELONG NOH TO ANY VOLUNTARY GROUPS OR ORGANIZATIONS - LIKE CHURCH GROUPS. SOCIAL CLUBS. PARENT GROUPS. AND THE LIKE? (“Voluntary' means because you want to.) 1 I yes ----1 1-I-HDH MANY? (do not code this response) 2 I no (go to next measure) 16. TO HHAT GROUPS DO YOU BELONG? (If more than 3 groups. list groups in which respondent is most active.) (Gather following information on each group) HHAT TYPE OF ORGANIZATION Name of IS THAT? (i.e. social. How active?‘ Organization church. type of club. etc.) (see below) 1. 2. 3. ' HOH ACTIVE ARE YOU IN THE AFFAIRS OF (group)? HOULD YOU SAY THAT YOU ARE: 1 I VERY ACTIVE OR ATTEND MOST MEETINGS 2 I FAIRLY ACTIVE OR ATTEND MEETINGS FAIRLY OFTEN. OR 3 I NOT ACTIVE. THAT 15 YOU BELONG BUT HAROLY EVER GO TO MEETINGS (or group has no meetings) Bla Social Support Questionnaire - Part 1 - page 3 240 Participant 10 Social Support Questionnaire ; Part g "I'MEIKTOASKYmeIESTIMSABflITPENLEWME PARTNYMLIF HHO PROVIDE YOU HITH HELP OR SUPPORT. AS I ASK EACH QUESTION. I HANT YOU TO NAME ONLY THOSE PEOPLE HHO COME TO MONO QUICKLY. “ To Interviewer: Do not list more than 10 names per question. Be sure to record the tfiE'first initial of the person's last name. even 1 i? that person's name comes up more than once! If respondent indi- ates “nobody:‘_be sure to indicate {hat on line 'a.‘) f respondent provides a name of an or anization, see if there s a key rson within that organization. I? not. record name of I i organization. aaaaaaaaaq aaaaaaa If respondent does not know person's last name. indicate that next to the name. Do this even though you have arbitrariTy assigned a last ifiTEial to that person. aaaaq mama: Social Support Questionnaire - Part 2 I page 1 241 Participant 1 I. THE FIRST COUPLE OF QUESTIONS HAVE TO DO HITH 'COMPANIONSHIP.’ III DO rnu USUALLY SPEND TIE HITH? (If participant needs a reference period. say “recently.“ or “the way things are going now.') a f) b) 9) C) h) d) - i) e) J) __ 2. IN AN AVERAGE HEEK. HHO DO YOU ENJOY CHATTING HITH? a) fI b) 9) c) h) d) I) e) I) __ 1 1 : Hand participant Card #3) : 3. IN GENERAL. I“ IN) vnu FEEL ABINIT THE MINT W CWANIGISHIP THAT YOU HAVE; DO YOU FEEL: I I ncumncn (mam euascn) 2 I PLEASED 3 I msm smsncn 4 I MIXED (ABINIT swam SATISFIED mun DISSATISFIED) s I nosm nIssmersn a I mm 7 I 1mm: _ 4. IN GENERAL. HOH DO YOU FEEL ABOUT THE QUALITY OF COMPANION- SHIP THAT YOU HAVE; DO YOU FEEL: 1 I mmu: a I mm s I msm DISSATISFIED 4 I, MIXED (mm comm SATIS‘IEDAND DISSATISFIED) 3 I mnsm smsncn z I ruascn I I ncumncn (amour mason) _ Social Support Questionnaire - Part 2 - page 2 242 Participant ID __ 5. now I'M some TO ASK YIN! ABIXJT A DIFFERENT KIND W IELP THAT vnu MY RECEIVE FR“ OTHERS CALLED 'ADVICE "D INFNMATIM.’ U10 CAN YOU CDINIT OI FIR ADVICE (II INFIRMATIM ABOUT PERSONAL MTTERS (Fm EWLE. PRULEMS HITH YINIR CHILNEM. FRIENDS. (NI SPGISE; DEALING HITH A PERSONAL SITUATION. THINGS LIKE THAT)? I) r) b) 9) e) h) d) i) e) j) __ __ I 6. I” CAN YOU RELY M F“ ADVICE W INFINIMATION YOU NEED ABOUT NESINNICES; rm: EXMPLE. ABINIT FIIIING A m a: A PLACE TO STAY. ABOUT HHERE TO APPLY FIN! HELFARE/FCNM) STWS. THINGS LIKE THAT? a) f) b) 9) c) h) a) i) e) J) __.__.' 7. IN GENERAL. INN DO YIN! FEEL ABGJT THE WUNT or ADVICE A_ND_ II'INIMATIM THAT YIN! RECEIVE; nn YIN! FEEE— 7 I TERRIBLE 6 I mm s I ImSTLI DISSATISFIED 4 I MIXED (Aamn EDDALLI SATISFIED mm DISSATISFIED) 3 . IIDSILv SATISFIED z I PLEASED I I DELIGHTED (mREIIELv PLEASED) _ D. IN GENERAL. anv no YOU FEEL ADDUT THE ALITY or ADVICE _A_M_D_ IIIDAIIATIDN THAT rnu RECEIVE; Do run I L: _— I I DELIGHTED (EXTREKLY PLEASED) Z I,PLEASEO 3 I MSTLY SATISFIED 4 I MIXED (AMT EMLLY SATIS’IED mm DISSATISFIED) s I IIDSTLI DISSATISFIED s I mm 7 I TERRIBLE __ Social Support Questionnaire I Part 2 - page 3 243 Participant 1 9. THE NEXT WLE (I: WESTINS HAVE TO no HITH AMTHER TYPE U WT CALLED “PRACTICAL ASSISTANCE.’ INICANYWCWNTMTDBEKPEDABLEHENYOUNEEDIELP? I) f) b) 9) C) h) d) i) I) .i) _‘ IO. WCANYWIXMMTUTONAFAVINIFGYW (MEMLE. TAKIK YW SKPLACE YIN! NEED TO no. LDANIK m GIVIK YIN] A ”LL HOUNT W mNEY. HATCHING YINR KIDS. LMNIK YOU WING YIN! NEED. ETC.)? a) f) b) 9) c) h) d) i) e) J) __.. 11. INGENERAL. WWWFEELABWTTHEWUNTW PRACTICAL ASSISTANCE TINT Yw RECEIVE;W FEEL: I I DELIGHI’ED (man: PLEASED) 2 I PLEASED 3 I NDSTu SATISFIED 4 I IIIxED (Aanm EquALLv SATISFIED mm DISSATISFIED) s . nnSTLv DISSATISFIED s I mm 7 I TERRIBLE __ 12. IN GENERAL. I" W van FEEL ABGlT THE ALITY (I: HACTICAL ASSISTANCE THAT rnu RECEIVE; FEEL: 7 I TERRIBLE B I WPY 5 I HISTLY DISSATISFIED 4 I MIXED (ABINIT EWALLY SATIS’IED m DISSATISFIED) 3 I MSTLY SATISFIED 2 I PLELSED I I DELIG'ITED (EXTREKLY PLEASED) _ Social Support Questionnaire - Part 2 I page 4 I3. 14. IS. 16. 244 IN I'M HIING TO ASK YIN! A CWPLE G' MSTIWS ABOUT 'ENTIMAL WT.’ INICMYWCINMTNTOLISTENTDYIRIIENYINIINTTOTALX ABOUT “THING PERSONAL? Participant It I) f) b) 9) C)' h) d) i) e) J) an REALLY CARES ABOUT YOU? I) f) b) 9) c) h) d) 1) e) .1) IN GENERAL. I" no YW FEEL ABWT THE WUNT IN: ENTIINIAL SlPPMT TINT Yw RECEIVE; WFEEL: 7 I TERRIBLE 6 I HAPPY 5 I NSTLY DISSATISFIED 4 I MIXED (AMT EwALLY SATI§IED MD DISSATIS'IED) 3 I mSTLY SATISFIED 2 I PLEASED I I HLIGTTED (EXTREKLY PLEASED) IN GENERAL. I“ W YIN! FEEL ABGJT THE ALITY W ENTINAL WT THAT Yw RECEIVE; no L: I I DELIE‘ITED (EXTRE‘LY PLEASED) 2 C PLENSED 3 O ”STLY SATISFIED B I MIXED (W EwALLY SNTI§IED m DISSATISFIED) 5 I ”STLY DISSATISFIED O 3 WY 703 MS.“ Social Support Questionnaire I Part 2 I page 5 245 Participant ID 17. mm MAKES YOUR LIFE DIFFICULT; SUCH AS SOIEONE HHO EKPECTS mmmmmmcsmnmnawmsnuvw. satanic HHO YOU HISH HIRILO LEAVE You ALONE on man mo YOU mum LIKE TO AVOID? I) f) b) 9) e) h) d) i) c) J) __ _ * I 18. IN ALL. ABOUT NDH MANY CLOSE FRIENDS HOULD YOU SAY YOU HAVE? (pe0ple you feel at ease with and can talk with about what is on your mind) close friends I ( 19. NOH. FOR THE LAST QUESTION. HOH DO YOU FEEL OVERALL ABOUT THE AMOUNT AND QUALITY OF THE SOCIAL SUPPDR RECEIVE? I I DELIGHTED (EXTREMELY PLEASED) 2 I PLEASED 3 I MOSTLY SATISFIED 4 I MIXED (ABOUT EQUALLY SATISFIED AND DISSATISFIED) S I MOSTLY DISSATISFIED 6 I UNHAPPY 7 I TERRIBLE __. I (I companionship) ;_ __ (0 advice and info) I I (0 practical assist) I I (l emotional) __ __ (0 negative) _ _ (0 total positve) __ _ Social Support Questionnaire I Part 2 I page 6 246 BLANK PAGE 247 alhw H N — rvl‘ W n G n N 4. a1. N ~uar‘thVl| .0 :o;onun.n~. Rates .83 .TI. ..ovson.n-§.nu. :soncn~_§.~_§a s :Bmuhwa.n~_ :15 .88 i levllllll. an:on.n~.fiacn~_ Re.on.n-3:~_: o :o;ononcn- Ancét. .89 VIII. Rasonwnudsvnud :zoncnfi~8:~_3n.on a Seamsonqn~_ Annucnv .uonn VII. “Shoncnu_a§n- Rena."~_83-:n.fi . :Smnton.n~. 5.93 .89 -Tl. AgnunanNHStn~_ Rm.on.n~.:8~:nufi n zeoononanS RE: .89 -Tl. :stoncnun 83 n a. :_:n.n-G:~12: n :20ohonanN. :To: .89 5:0“.nndaan. Eon.n- E~_:.$ll._ 291.. oufifimhon.n- 2.28 3383 .23 . : 1.5.5.! SUD—=0. 5:5» :6 352368.S 28:239.; 35m...“ 3...! :3 Lo autogrfid mgfitwfizs: ngfizia £3. Einfitxi 31: 135828: Ease—SEPI— All...EEu 3333.85.28 “8:32.622: 28.52 Stacey! 5:2...h2 urging; .33628 Euchssilxun ...—65.3335 5:9ng 3:2 9::5fi8.8 58:3..uflx.“ 28308 East-Eu pgbifie 683.8 328.31.. 2.8.8 3.4.8.5: US$53»; Stat-$98.3 22.-5:8. .1332: p 239.8 333 — £296 a 3 28' u £3392qu Pa:— Adau £9.3- :38 5399—050 Jean—8 a. al.—23380 u co 2:515 H:85 22:3 :50 u 8 333 u no 62:... “as 5:22 . 6 38 258 can: 2.: .5: ...—Data 3! 5! m. :3. Egg: 383 23: A3 I 38 3-: on “553:9; .5?“qu a .63.... H mus. quuuwa a 139.0. 248 Participant ID Social Support Questionnaire ; Part 3 (Participant ID) I 1 I If no names were provided, proceed to Demographic and 1 1 Bacfground Information. I 1 I (To interviewer: If participant named a total of three or fewer people on the last set of social support questions. ask the following questions about each of those individuals. Do not ask them to provide the names of the 3 most important pe0pl3_Tn the network.) (F (ALL) THE PEOPLE Ill IE HAVE TALKED WT (repeat nus). IMICH TIRED ARE MST INTANT TO YOU? (Remember to include first initial of last name!) 1. 2. 3. HHICH if THESE PEOPLE IS THE _flfl INTANT TO YOU? DURING THE PAST MONTH, HON OFTEN HAVE YOU HAD CONTACT HITH (name 1. name2. or name3). INCLUDE TIMES YOU MAY HAVE TALKED ON THE PHONE. (Confirm answer by repeating coded response. For eque. you can say. 'So you saw him about once a week in the past month.') I.Personll 123456-8. ..... 2. Person '2 l 2 3 4 5 6 - O ...... 3.Personl3 123456-8. ..... (Use the following categories to code these items. If necessary, probe for answers or confirm responses with following categories.) I . every day 2 - several days a week 3 . about once a week 4-2or3 times in pastmnth 5 - once in st month Genotatal inpastnnth 8 - not applicable (no friend named) Social Support Questionnaire - Part g,- page 1 Blank (8- (u (12 (1‘ (1! (u (17 2419 Participant IO Demographic and Background Information NOH I HOULD LIKE TO ASK YOU SOME QUESTIONS ABOUT YOUR BACKGROUND. 1. HHAT IS YOUR DATE OF BIRTH? / / month day year NOH I'M GOING TO ASK YOU SOME QUESTIONS ABOUT SCHOOL AND HORK. 2. HHAT IS THE HIGHEST GRADE OR YEAR OF SCHOOL YOU COMPLETED? 1-11 . code actual grade 12 I High School grad/GED 13 c Vocational training 14 2 Some college 15 . College graduate 16 c Other (specify) 98 . Never attended sEbool 99 a Doesn't know/missing __ 3. HHEN DID YOU LAST HORK FOR PAY AT A JOB LASTING 159 HEEKS IN §_§Q! OR LONGER? [code most recent] 1 - currently working 5 . >1 to 2 years ago 2 - within last month 6 . >2 to 3 years ago 3 - >1 to 6 months ago 7 c more than 3 years ago A - >6 to 12 months ago 9 u never worked for pay I90 to Q 5] A. [If worked] HHEN YOU HORK, HHAT TYPE OF HORK DO YOU USUALLY 00? [Be as specific as possible.] NOH I'M GOING TO ASK YOU SOME QUESTIONS ABOUT YOUR HEALTH. 5. IN GENERAL. HOH HOULD YOU RATE YOUR HEALTH? IS IT: I . EXCELLENT 4 . POOR 2 - GOOD 5 - don't know (gg_ggt 3 . FAIR read this response) Demographic and Background Information - page I 250 Participant ID HOH DOES YOUR HEALTH NOH COMPARE TO YOUR HEALTH THO YEARS AGO? IS IT: I I BETTER 2 I THE SAME 3 I HORSE NOH I'M GOING TO ASK SOME QUESTIONS ABOUT YOUR CONTACTS HITH DOCTORS AND HOSPITALS. 7. IO. 11. 12. IN THE PAST YEAR. HOH MANY TIMES HAVE YOU GONE TO SEE A DOCTOR. FOR ANY REASON? Number of times: (If none. code '00“) HAVE YOU EVER BEEN HOSPITALIZED FOR EMOTIONAL PROBLEMS OR MENTAL ILLNESS? I I yes 2 I no (go to Q 14) [If yes] HHAT HOSPITAL HERE YOU IN DURING YOUR MOST RECENT HOSPITALIZATION FOR EMOTIONAL PROBLEMS OR MENTAL ILLNESS?__ I Northville Detroit Receiving Kingswood Kirwood I Sinai 7 I other (specify) . B I not applicable -_hisn't been in hospital I 2 3 a 5 HHAT STATE IS THAT HOSPITAL IN? I I Michigan 2 I other state (specify) 8 I not applicable - hasn'ffibeen’in‘hospital HHEN HAS THE LAST TIME YOU HERE HOSPITALIZED FOR EMOTIONAL PROBLEMS OR MENTAL ILLNESS? (Do not include hospitalizations for alcoholism) (Code date entered hospital) Date / 9998 I n/a - hasn't month year been in hosp. HOH OLO HERE YOU AT THE TIME OF YOUR FIRST PSYCHIATRIC HOSPITALIZATION? Age: 98 I n/a - hasn't been hospitalized Demographic and Background Information . page 2 251 Participant ID 13. HOH MANY TIMES IN ALL HAVE YOU BEEN HOSPITALIZED FOR PSYCHIATRIC PROBLEMS? Number of times: (42 (If hasn't been hospitalized. code as '00“) THE NEXT QUESTIONS HAVE TO DO HITH MEDICATION YOU MIGHT BE TAKING. IA. DURING THE LAST SIX MONTHS. HAVE ANY MEDICINES BEEN PRESCRIBED FOR YOU-BY A DOCTOR? l I yes 2 I no (go to 0 18) (44 IS. HHAT MEDICINES HAVE BEEN PRESCRIBED POR YOU? (If says 'psychotropics.' find out which ones) Med. 1 Med. A Med. 2 Med. 5 Med. 3 Med. 6 16. ARE YOU TAKING YOUR MEDICIHE(S) HHEN YOU ARE SUPPOSED TO; THAT IS. ACCORDING TO THE OOCTOR'S INSTRUCTIONS? 1 m yes (go to Q 18) 2 I no (indicate which one(s) next to names above) 8 I n/a - no prescription __ (45 17. [If no] HHAT IS THE MAIN REASON YOU AREN'T FOLLOHING YOUR PRESCRIPTION? I ran out of medications I don't like side-effects I don't have Medicaid card I doesn't help I no longer required (e.g. illness over) 6 I other (specify) 8 I not applicable - no prescription __ (45 crime-no.- NOH I'M GOING TO ASK YOU ABOUT YOUR USE OF ALCOHOL. lBa. HAVE YOU EVER HAD ALCOHOLIC BEVERAGES. SUCH AS BEER, HINE, OR LIQUOR? II S - 2-3 (go too 19) . __ (a. Demographic and Background Information - page 3 252 Participant H 18b. HOH OFTEN DID YOU DRINK ALCOHOLIC BEVERAGES, SUCH AS BEER. HINE. OR LIQUOR IN THE LAST MONTH? every day more than once a week once a week 2 or 3 times in past month once in past month not at all in past month n/a - doesn't drink can't determine. doesn't know @QOU'MDUNe—l 18c. NOH I'M GOING TO READ YOU A LIST OF PROGRAMS THAT SOMETIMES HELP PEOPLE HITH DRINKING PROBLEMS. TELL ME HHETHER OR NOT YOU HAVE EVER PARTICIPATED IN EACH OF THESE TYPES OF PROGRAMS. (Read each program and indicate number 9: times received help/admissions.) (Do not include drug treatment programs) (Coding: O I 32 help received) 1-6 = code actual number 7 I 7 and above) OETOX PROGRAM ...................... INPATIENT REHABILITATION PROGRAM ............ OUTPATIENT PROGRAM ................... HALFHAY HOUSE ...................... ANY OTHER ALCOHOL TREATMENT PROGRAM? (i.e. in correctional setting) (Specify) “GOOD 18d. (If has been in one of above programs) HHEN HAS THE LAST TIME YOU HERE IN AN ALCOHOL TREATMENT PROGRAM? currently under treatment under treatment within last 6 months, but no longer under treatment >6 months to 1 year ago more than a year ago not applicable/never in program Gibb-D NH N IBe. HAVE YOU EVER BEEN IN AA (Alcoholics Anonymous)? 1 I yes, currently involved 2 I yes. no longer participating 3 I no Demographic and Background Information - page 4 253 Participant NOH I'M GOING TO ASK YOU A FEH QUESTIONS ABOUT DRUGS. 19. HAVE YOU EVER SMOKED MARIJUANA? 1 I yes 2 I no (go to Q 21) 20. IN THE PAST MONTH, HOH MANY TIMES HAVE YOU SMOKED MARIJUANA? every day more than once a week once a week 2 or 3 times in past month once in past month not at all in past month n/a - has never smoked marijuana can't determine. doesn't know IOQOUYAwNI-I 21. HAVE YOU EVER USED ANY OTHER DRUGS OR NARCOTICS. SUCH AS COCAINE. HEROIN, LSD, SPEED, OR OTHER THINGS LIKE THAT? (Do not include drugs prescribed by physician) 1 I yes 2 I no (go to Q 24) 22. IN THE PAST MONTH, HAVE YOU USED ANY OTHER DRUGS OR NARCOTICS? l I yes 2 I no (go to Q 23b) 23. HHAT DRUGS HAVE YOU HOH OFTEN HAVE YOU USED USED IN THE PAST MONTH? (drug) IN THE PAST MONTH? (use categories below) every day more than once a week once a week 2 or 3 times in past month once in past month not at all in past month can't determine, doesn't know \DOImwaI-I NNMNIINII 23b. HAVE YOU EVER BEEN IN A DRUG TREATMENT PROGRAM? I yes 1 2 no (go to Q 24) Demographic and Background Information - page 5 2.5.4 Participant IO 23:. HHEN HAS THE LAST TIME YOU HERE IN A DRUG TREATMENT PROGRAM? I I currently under treatment 2 I under treatment within last 6 months. but no longer under treatment 3 I )6 months to 1 year ago 4 I more than a year ago B I not applicable/never in program NH I'M GOING TO ASK YOU QUESTIONS ABOUT TIMES HHEN YOU MAY HAVE BEEN A VICTIM or A CRIME DURING THE LAST 2; MONTHS; THAT IS. BETHEEN I. 198_ AND TODAY. 24a. BETHEEN I. 19 AND TODAY. DID ANYONE TAKE SOMETHING—M I TO TAKE SOMETHING DIRECTLY FROM YOU BY. USING FORCE. SUCH AS BY A STICKUP. MUGGING OR THREAT? I I yes-II! II- NOH MANY TIMES? * 2 I no (go to O 25) 24b. (THINK ABOUT THE LAST TIME THAT THIS HAPPENED.) HHERE HERE YOU HHEN YOU HERE ROBBED? HHAT HAPPENED? (Describe what happened): 01 I at or in own dwellin 02 I near own fififie s1 ewalk. driveway. on street immediately adjacent to home. apartment hall or laundry room - not parking lots) 03 . at. in. or near E'Triend/relative/neighbor's home (see '2 above) 04 I on the street (other than immediately adjacent to own/friend/relative/neighbor's home) I on public transportation I parking lot I temporary shelter I other (specify) 98 I not applicable/nit’attacked 24c. HAS THE PERSON HHO ROBBED YOU SOMEONE YOU KNEH. OR A STRANOER YOU HAD NEVER SEEN BEFORE? I I stranger 2 I known: HHAT HAS THEIR RELATIONSHIP TO YOU? (ie spouse. friend. etc.) (SPecify) ' B I not applicable/not‘ittaéked Demographic and Background Information . page 6 # (61} (62) (6E 255 Participant 25. (other than any incidents already mentioned) DID ANYONE BEAT YOU UP. ATTACK YOU HITH A HEAPON OR HIT YOU HITH SOMETHING. SUCH AS A ROCK OR BOTTLE? I I yes-III! 1-- HOH MANY TIMES? * 2 I no (go to Q 28) 263. THINK ABOUT THE LAST TIME THAT THIS HAPPENED. HOH DID THE PERSON ATTACK YOU? (probe: ANY OTHER HAY?) (Describe): (Coding: I I yes ZITIO 8 I n/a - not attacked) raped ............................ tried to rape ........................ shot ............................. knifed ............................ hit with object held in hand or thrown object ........ hit, punched. slapped, knocked down, grabbed. held. etc. . other (specify) ....... 26b. HHERE HERE YOU HHEN YOU HERE ATTACKED? (Describe): __fi 01 02 at or in own dwellin near own Fame sidewalk, driveway. on street imnediately adjacent to home. apartment hall or laundry room I not parking lots) 03 I at. in. or near 3_friend/relative/neighbor's home (see '2 above) . 04 I on the street (other than immediately adjacent to own/friend/relative/neighbor's home) OS I on public transportation 06 I parking lot 07 I temporary Shelter 08 I other (specify) 98 I not applicable/not attacked 27. HAS THE PERSON HHO ATTACKED YOU SOMEONE YOU KNEH. OR A STRANGER YOU HAD NEVER SEEN BEFORE? 1 I stranger 2 I known: HHAT HAS THEIR RELATIONSHIP TO YOU? (ie spouse. friend. etc.) (specify) * 8 I not applicable/not attacked Demographic and Background Information I page 7 256 Participant ID _____ 28. (Other than 22y incidents already mentioned) DURING THE LAST SIx MONTHS. OIO ANYONE THREATEN TO BEAT YOU UP OR THREATDT’VOU HITH A KNIFE. Gum ME OTHER HEAPON. NOT 'INCLuo' ING THREATS OVER THE TELEPHONE? 1 I yes ...1 OIIHOH MANY TIMES? I 2 I no __ (Describe most recent time): 29. SINCE 1. 198 . DID ANYONE STEAL THINGS THAT BELONGED IO YOU FROM INSIDE YOUR HOME OR THE PLACE HHERE YOU HERE STAYING? 1 I yes-II! IIIHOH MANY TIMES? * 2 I no (go to O 31) __ 30. THINK ABOUT THE LAST TIME THIS HAPPENED. HHAT TYPE OF PLACE HERE THESE—THINGS—STOLEN FROM? HAS IT FROM YOUR OHN HOUSE OR APARTMENT. FROM A FRIEND'S PLACE. OR FROM SOME OTHER PLACE? (Describe what happened): 1 I Own house/apartment 2 I friend or relatives house/apartment temporary shelter on street . other (specify) _ not applicable/nOEhing stolen ___ 00!.“ m m w (Pir—tTcipant TD) ... _ _ NOH I'M GOING TO ASK YOU SOME QUESTIONS ABOUT PLACES HHERE YOU HAVE LIVED OR STAYED. 31. HHEN HAS THE FIRST TIME YOU STAYED IN A SHELTER OR MISSION? (e.g. COTS. Detroit Rescue Mission. Harbor Light. Day House) (If this is first shelter month year stay. code current date) \ 'Date: 32a. NOH MANY TIMES HAVE YOU STAYED IN A SHELTER OR MISSION DURING THE PAST YEAR. INCLUDING THIS TIME? Number of times (If this is the Tirst shelter stay. code as '01") _"-— Demographic and Background Information I page 8 257 Participant 1 32. ABOUT HOH MANY TIMES IN ALL HAVE YOU STAYED IN A SHELTER. )NQLUOING THIS TIME? Number of times (estimate if necessary) (If this is the first shelter stay. code—EE'VQE") 33. DID YOU SLEEP HERE AT COTS LAST NIGHT? 1 I yes 2 I no (go to O 33) 34. [If yes] HOH MANY NIGHTS HAVE YOU SLEPT HERE AT THE SHELTER DURING YOUR PRESENT STAY? Number of nights: 1-97 I code actual number 98 I not applicable. didn't sleep at shelter last night NOH I'M GOING TO ASK YOU SOME QUESTIONS ABOUT THIS SHELTER. 35. HOH SAFE DO YOU FEEL IN THIS SHELTER? DO YOU FEEL: VERY SAFE SOMEHHAT SAFE SOMEHHAT UNSAFE VERY UNSAFE wao-I M 36. HOH SATISFIED ARE YOU HITH THE AMOUNT OF HELP YOU HAVE RECEIVED HERE FROM THE STAFF AT COTS? ARE YOU: I I QUITE DISSATISFIED 2 I MILDLY DISSATISFIED OR INDIFFERENT 3 I MOSTLY SATISFIED 4 I VERY SATISFIED 37. DO YOU FEEL THAT THE OTHER PEOPLE HHO STAY HERE AT THE SHELTER ARE: 1 I A LOT LIKE YOU 2 I PRETTY MUCH LIKE YOU 3 I NOT MUCH LIKE YOU A I NOT LIKE YOU AT ALL H A HAT DO YOU THINK THE CHANCES ARE THAT YOU HILL EVER STAY 38. T A SHELTER LIKE THIS AGAIN? DO YOU THINK THAT YOU: DEFINITELY HILL PROBABLY HILL PROBABLY HON'T DEFINITELY HON'T bumb- u M II Demographic and Background Information I page 9 39. A0. A1. A2. A3. 258 Participant ID __ IF YOU DO HAVE TO STAY AT A SHELTER AGAIN SOMETIME. HHAT DO YOU THINK THE CHANCES ARE THAT YOU HILL COME BACK HERE TO COTS? DO YOU THINK THAT YOU: I I DEFINITELY HILL 2 I PROBABLY HILL 3 I PROBABLY HON'T A I DEFINITELY HON'T HOH DO YOU FEEL ABOUT THE PEOPLE HHO HORK HERE AT THE SHELTER: DO YOU: I I LIKE ALL OF THE PEOPLE HMO HORK HERE 2 I LIKE MOST OF THE PEOPLE HHO HORK HERE 3 I DISLIKE MOST OF THE PEOPLE HHO HORK HERE A I DISLIKE ALL OF THE PEOPLE HHO HORK HERE IN AN OVERALL. GENERAL SENSE. HOH SATISFIED ARE YOU HITH THE SERVICES YOU HAVE RECEIVED HERE AT COTS? ARE YOU: I VERY SATISFIED I MOSTLY SATISFIED INDIFFERENT OR MILDLY DISSATISFIED I QUITE DISSATISFIED OWNH N HHAT DO YOU LIKE BEST ABOUT THIS SHELTER? HHAT DO YOU LIKE LEAST ABOUT THIS SHELTER? NOH I'M GOING TO ASK YOU SOME QUESTIONS ABOUT PLACES YOU HAVE LIVED IN THE LAST YEAR. 44., A5. HOH LONG HAVE YOU LIVED IN DETROIT? O-I month >1I3 months >3I6 months >6I12 months ........... 5 I more than 12 months ---}50 t0 0 A6 .wNe-I NNNN (If 6 Months or less) HOH MANY DIFFERENT CITIES HAVE YOU LIVED IN DURING THE PAST YEAR? Amber of cities: 98 I not applicable/lived in Detroit over 6 months Demographic and Background Information I page 10 A6. 47. A8. A9. 2E59 Participant ID HHAT TYPE or PLACE HAVE YOU HOSTLV BEEN LIVING IN DURING THE PAST s15 MONTHS? house, apartment, mobile home room, hotel group living (halfway house, AFC home, etc.) hospital, nursing home correctional facility shelters/missions on street other (specify) muombwwu nmnmnunn HHAT TYPE OF PLACE DID YOU STAY AT THE NIGHT BEFORE YOU CAME TO THIS SHELTER? house, apartment, mobile home - room, hotel .. ........... . group living (halfway house, AFC home, etc.) hospital, nursing home correctional facility -go to shelters/missions 0 50 on street ONOHJIDwNa-o IIIIIIII other (specify) HON MANY QIEEE PEOPLE HERE YOU LIVING HITH? (Include family members) Number of peOple (If living along, code as '00“) HHO HERE YOU LIVING HITH? (probe: ANYONE ELSE?) (List number in each category - Hake sure adds up to number of people living with) (Coding: 1 - 6 I code actual number 7 I 7 and above 8 I n/a alone or not in house or apartment spouse, grown children, or parents ............. bayfriend/girlfriend .................... other relatives . .‘o . .'. . . . . . . . ...... friends . , ........................ Other (specify) ......... Demographic and Background Information - page 11 (2‘ (2: (2 260 Participant ID __ 50. HOH LONG DID YOU LIVE OR STAY THERE? 0-1 month >1-3 months >3-6 months >6-12 months more than 12 months not applicable communis- IIIII Hum 51. HHY DID YOU LEAVE THE LAST PLACE YOU HERE STAYING? (probe: ANY OTHER REASONS?) (Describe reasons - Be clear! - th couldn't s/he stay at this place any longer? - I? stayed at this place only a short time, also get information on why left last permanent type housing situation. Get information gn_thg general circumstances leading 39 the shelEEF'stay!) type place how long stayed Reasons left,,etc. Who—With (Get enough information to code following reasons - Use these reasons as probes.) (Coding: l I yes 2 I no) Economic reasons/couldn't pay rent ............ Interpersonal conflict with household member(s) (specify what and with whom) Thrown out/EVTCtion—by"lanHTOrd .............. Disaster (assault, fire) (specify what) . . I Discharged or released from hospital or jail/prison . . . . Program terminated .................... Had stayed at shelter maximum days allowed . . . . . . . . .(Name and city of shelter: ) On street and needed shelter ............... Other (specify) llllll Demographic and Background Information - page 12 52. 2,61 Participant ID BEFORE YOU CANE TO THIS SHELTER, mu NANY DIFFERENT FENCES HAD YOU LIVED IN OR STAYED AT DURING THE LAST _S_I_X_ MONTHS? THAT IS. SINCE , 198_. IN OTHER HORDS, HOH MANY TIMES DID YOU GO FROM ONE PLACE HHERE YOU HERE STAYING TO A NEH PLACE? INCLUDE TIMES YOU STAYED IN A SHELTER BEFORE THIS, OR ON IHE STREET. (Do not include present shelter stay]) (KeeE'Tn mind what participant has already told you!) Number of places: General locations 0? places (e.g. brother‘s house in Chicago, etc.): 53a. 53b. 53c. DO YOU CONSIDER YOURSELF TO BE HOMELESS NOH? I I yes 2 I no HAVE YOU EVER BEEN HOMELESS IN THE PAST? (i.e. before now) I I yes 2 I no (go to Q 54) (If yes) NOH MANY TIMES HOULD YOU SAY YOU HAVE BEEN HOMELESS BEFORE (THIS TIME)? Number of times: Comments: THE NEXT MAY HAVE 5A. 55. FEH QUESTIONS HAVE TO DO HITH JAIL OR PRISON, AND TIMES YOU BEEN ARRESTED. HAVE YOU EVER BEEN ARRESTED AS AN ADULT? (age 17 or older in Nichigan) (Do not include traffic violations, such as speeding or EFTving without a permit.) I I yes 2 I no (go to Q 74) (probe - 'SO you ' haven't been in jail or prison or on probation?) ABOUT HON MANY TINES IN ALL HAVE YOU BEEN ARRESTED? Number of times: 98 I n/a never arrested Demographic and Background Informotion - page 13 262 Participant 10 56. HOH OLD HERE YOU THE FIRST TIME YOU HERE ARRESTED AS AN ADULT? Age: 98 I n/a never arrested (52-5 57. HOH MANY TIMES HERE YOU ARRESTED DURING THE PAST YEAR? 0 I none/hasn't been arrested in last year I-6 I code actual number 7 I 7 and above 8 I n/a never arrested (54) 58. HAVE YOU EVER BEEN SENTENCEO TO SERVE TIME IN A COUNTY JAIL? (do not include time awaiting disposition or sentencing) (Includes Detroit House of Correctons-DEHOCO) I I yes 8 I n/a never arrested 2 I no (go to O 62) __ (55) 59. RON NANY TINES IN ALL HAVE YOU SERVED TIME IN COUNTY JAIL? Number of times: 98 I n/a never in Jail (56-E 60. HOH MANY TIMES HAVE YOU SERVED TIME IN COUNTY JAIL DURING THE LAST YEAR? 0 I none/hasn't been in Jail in past year I - 6 I code actual number 7 I 7 or more 8 I n/a never in Jail (58) 61. Obtain the following information for each of the three most recent periods of incarceration in a county Jail. Offense convicted of Length of time Date of release Name of (be specific) served (month/year) Jail (ibst recent)* (2ndfimost recehETE (Bid mOSt recent) 62. HAVE YOU EVER BEEN SENTENCEO TO A STATE OR FEDERAL PRISON? I I yes 8 I n/a never arrested 2 I no (If has been in Jail, go to Q 66 (If ES? 393 been in Jail, go to O 73) ___ (59) Demographic and Background Information - page 14 ‘263 Participant ID 63. HOH MANY TIMES IN ALL HAVE YOU BEEN SENTENCEO TO PRISON? l - 6 I code actual number 7 I 7 and above 8 I n/a never in prison 6A. Obtain the following information for each of the three most recent periods of incarceration in a state or federal prison. (6O Offense convicted Length of Name of prison Date released of (be specific) sentence where began (month/year) (min to max) sentence (iOSt recent) (2nd most recent) (3rd most recent) 65. HAVE YOU EVER BEEN ASSIGNED A MICHIGAN PRISON NUMBER? 1 I yes (please indicate number ) 2 I no 8 I not applicable/never arrested 66. HHAT TYPE OF PLACE DID YOU FIRST LIVE IN THE LAST TIME YOU HERE RELEASED FROM INCARCERATIDN? (either Sail or prison) (Record followigg information 12: reference): Offense: Date of release: _Jail or __pri son 01 I house, apartment, mobile home 02 I room, hotel ------- . 03 I group living (halfway house, AFC home, etc.) 04 I hospital, nursing home OS I correctional facility Igo to 06 I shelters/missions O 68 07 I on street. 08 I other (specify) . 98 I not applicable/never incarcerated Demographic and Background Information - page 15 (61 (62 254 Participant I 67. HHO HERE YOU LIVING HITH? (probe “Anyone else?") (Coding: first options take priority over latter) spouse, grown children, or parents boyfriend/girlfriend other relatives or other relatives and friends friends only alone other (specify) n/a no prison; not in house or apartment ooascnwa—o II II II M II I. II LONG DID YOU LIVE THERE? I o I 68. O-l month >1-3 months >3-6 months >6-12 months ----------- more than 12 months ----- 3‘50 to 0 72 not applicable mmowmu H M M II M M 69. (If less than 6 months) HHY DID YOU LEAVE THE LAST PLACE YOU LIVED? (Record reason): (GEt enough information to code following reasons) (Coding: 1 I yes 2 I no 8 I not applicable/never incarcerated) Economic reasons/couldn't pay rent ............ Interpersonal conflict with household member(s) (specify what and with whom) Thrown out/eviction’by landlord .............. Disaster (assault, fire) (specify what) Discharged or released from hospital or Jail/prison . . . . Program terminated .................... Had stayed maximum number of days at shelter ....... (Name and city of shelter: ) On street and needed shelter ............... Other (specify) Demographic and Background Information - page 16 265 70. (If less than 6 months) HHAT TYPE OF PLACE DID YOU LIVE Participant ID IN NEXT? 01 I house, apartment, mobile home 02 I group living (halfway house, AFC home, etc.) 03 I room, hotel 04 I hospital, nursing home 05 I correctional facility 06 I shelters/missions 07 I on street 08 I other (specify) 98 I not applicable 71. HOH LONG DID YOU LIVE THERE? l I 0-1 month 2 I >1-3 months 3 I >3-6 months A I >6-12 months 5 I more than 12 months 8 I not applicable 72. (No Item I 72) 73. ARE YOU ON PROBATION OR PAROLE NOH? 1 I yes, probation: Offense: Date sentence began: 2 I yes, parole: Offense: Date parol§_5egan: 3 I no (Participant ID) NOH I'M GOING TO ASK YOU SOME QUESTIONS ABOUT MILITARY SERVICE. 74. HAVE YOU EVER BEEN IN THE MILITARY SERVICE? 1 I yes 2 I no (go to O 79) 75. HHEN DID YOU LEAVE THE SERVICE? [If still in service, code current date] I month year Demographic and Background Information - page 17 9998 I n/a - not in service 266 Participant ID 76. HOH LONG HERE YOU IN THE SERVICE? 6 months or less >6 months to 12 months >12 months to 2 years 2 years or more not applicable - not in service (11) cps-uni.- M N I! M 77. HERE YOU EVER IN ACTIVE COMBAT? 1 I yes 8 = n/a - not in service 2 I no (go to Q 79) (12) 78. HHERE DID YOU SERVE ACTIVE COMBAT? I Korea Vietnam Europe/Pacific - HHII other (specify) 1 2 3 4 ——I-———r 8 not applicable - not in serVIce (13) THE NEXT QUESTIONS HAVE TO DO HITH YOUR SOURCES OF MONEY. 79. DURING THE LAST SIX MONTHS, HAVE YOU GOTTEN MONEY FROM: (Have participant respond to each response!) (Coding: 1 I yes 2 I no 8 I refused to answer) HORK (either yourself or spouse) ............ __ (14) YOUR FAMILY (not counting spouse) ........... . __ (15) ALIMONY/CHILD SUPPORT ................. . ___ (I6) PENSION/RETIREMENT ................... __. (17) $51, $501, SOCIAL SECURITY (Circle type received). . . . __ (18) PUBLIC ASSISTANCE, SUCH AS ADC, FOOD STAMPS, HELFARE, GA, AND THE LIKE (Circle types received). . .‘__ (19) VA BENEFITS ...................... __ (20) UNEMPLOYMENT COMPENSATION ............... ___ (21) PANHANDLING; THAT IS, ASKING STRANGERS FOR MONEY . . . . __ (22) SAVINGS ........................ .‘__ (23) ILLEGAL SOURCES .................... . __ (24) ANYTHING ELSE? ANY OTHER HAYS YOU GET MONEY OR SUPPORT? . ___ (25) (specify: I Demographic and Background Information - page 18 26.7 Participant IO 80. HHICH OF THESE HAS BEEN YOUR LARGEST SOURCE OF MONEY OVER THE PAST MONTH? (RepeaE sources cited above) 01 I work (self or spouse) 02 I your family (not counting spouse) 03 I alimony/child support 04 I pension/retirement OS I SS], 5501. social security O6 I Public Assistance (ADC, food stamps, welfare, GA, etc) 07 I VA benefits 08 I unemployment compensation 09 I panhandling ID I savings 11 I illegal sources 12 I other (specify) (26-27) 98 I refused to answer """ 81. COUNTING ALL MONEY YOU GOT FROM (read sources cited in O 79), HAS'VOUR TOTAL FAMILY OR HOUSEHOLD INCOME DURING THE PAST YEAR: (yourself and other members of your family who you lived with - If respondent lived alone or independently, then Just include his or her income from all sourcesJ (Read categories) 01 I LESS THAN $1,999 I - (less than SI66/month) 02 I SZOOO TO $2.999 - - I(SIGG - SZSO/nonth) 03 I $3,000 TO A.999 - - -(SZSI - $416/month) DA I $5.000 TO 7,999 - - -(SAI7 - $666/month) OS I 38.000 TO $9.999 - -($667 - $833/month) 06 I $10,000 TO 14,999 - I(8834 I $1,250/month) 07 I $15,000 To 19,999 - I(SI,251 - $1.666/month) 08 I OVER 820,000 I I I I(over $1.666/month) 98 I (refused to answer) __ __ 82. NOH MUCH OF THIS TOTAL DID !QU YOURSELF ACTUALLY EARN, SUCH AS BY HORKING? (Do not include money received as gifts or loans, from pubITE'assistance, etc.<- if has worked, make sure you get amount earned) OI I NONE 02 I 31 TO $1,999 03 I 32000 TO 82.999 04 I 83,000 TO 4,999 OS I $5,000 TO 7,999 06 I $8,000 TO $9,999 07 I $10,000 TO 14,999 08 I 915,000 TO 19.999 09 I OVER 520,000 __ ___ (30-31) 98 I (refused to answer) (28-29) Demographic and Background Information I page 19 2658 Participant ID NOH I'M GOING TO ASK YOU ABOUT HHAT YOU HILL DO HHEN YOU LEAVE THE SHELTER. 83. DO YOU KNOH HHERE YOU HILL STAY AFTER YOU LEAVE THE SHELTER? 1 I yes (must know exactly 2 I no (go to O 86) where will stay) 84. IN HHAT TYPE OF PLACE HILL YOU BE STAYING? 01 I house, apartment, mobile home 02 I room, hotel --------- 03 I group living (halfway house, AFC home. etc.) i 04 I hospital, nursing home 05 I correctional facility Igo to 06 I shelters/missions Q 86 07 I on street 08 I other (specify) 98 I not applicable/doesn't know where will stay '—’_' 85. [Ij_house pp apartment] HHO HILL YOU BE LIVING HITH? (probe ”Anyone else?") (Coding; first options take priority over latter) spouse, grown children, or parents boyfriend/girlfriend other relatives or other relatives and friends friends only alone other (specify) not applicable -_35esn't know where will stay, or ‘__ not house or apartment mam‘the-l II M M I I M I 86. HILL YOU BE STAYING IN THE DETROIT AREA? 1 I yes 3 I doesn't know 2 I no I HHERE HILL YOU BE STAYING? ___ ---87: NOH ONE LAST QUESTION, HOH DO YOU FEEL ABOUT YOUR LIFE AS A HHOLE. DO YOU FEEL: . DELIGHTED (EXTREMELY PLEASED) PLEASED I MOSTLY SATISFIED MIXED (ABOUT EQUALLY SATISFIED AND DISSATISFIED) MOSTLY DISSATISFIED UNHAPPY TERRIBLE __ THAT COMPLETES THE QUESTIONS THAT I HAVE FOR YOU. DO YOU HAVE ANY QUESTIONS? THANK YOU VERY MUCH FOR YOUR HELP. NOM‘WNP—P M Time interview finished: Blank Demographic and Background Information I page 20 (32) (33-34) (35) (36) (37) (38) 269 Participant ID Surveyors Impressions Rate respondent's grooming l I body and clothes neat and clean 2 I body clean, but clothes dirty 3 I clothes clean, but body dirty A I unwashed, unkempt, dirty clothes (39) Rate respondent's attire I I appropriate to weather and place 2 I inappropriate (specify ) (40) Rate level of attention 1 I Attentive and responsive 2 I some lapses of attention 3 I paid no attention much of the time (41) Rate manner of speech: often sometimes never talked in digressive or rambling manner I 2 3 __ (42) talked or muttered to self 1 2 3 __ (43) refused to answer 1 2 3 __ (44) illogical/nonsensical 1 2 3 ___ (45) disorganized/incoherent I 2 3 __ (46) Rate emotional state: often sometimes never flat affect 1 2 3 __ (47) angry or hostile 1 2 3 __ (48) sad, depressed I 2 3 ___ (49) anxious, apprehensive I 2 3 ___ (50) hallucinating 1 2 3 __, (51) Rate attitude toward interviewer I I cooperative 2 I neutral 3 I uncooperative ___ (52) 270 Participant ID 7. Rate attitude toward interview interested neutral bored (53) (AND-0 MMII E valid do you feel this person's responses are overall? valid questionable not valid (54) “NH "MM Specify problem areas in interview: 9. Has this interview tape-recorded? 1 I yes 2 = no __ (55) 10. If no, why not? participant refused recorder not working supervisor said not necessary other (specify) n/a interview recorded __ (56) mtha-I N in ii A N 11. Did participant state that s/he would be leaving town soon after the shelter stay? 1 2 yes (Specify: I 2 I no ___ (57) 12. (If in material goods condition) Hhat did participant choose for payment? I I toilet articles 8 I not applicable/ 2 I cigarettes received cash __ (58) SUMMARY OF INTERVIEH/DESCRIPTION AND IMPRESSIONS OF PARTICIPANT: (Length of interview) __ __ __* (30-32) 2171 Participant ID INTERVIEH EVALUATION Hhat did you think about this interview? Has it: a . VERY INTERESTING b. SOMEHHAT INTERESTING c. SOMEHHAT BORING d. VERY BORING How satisfied are you with your payment for participating in these interviews? Consider both the payment you received now, and the payment you will receive when you return for the next interview in six weeks. a. VERY DISSATISFIED b. SOMEHHAT DISSATISFIED c. SOMEHHAT EBIIEEIEQ d. VERY SATISFIED How useful do you feel the payments are which you are receiving for your participation in these interviews? a. VERY USEFUL b. SOMEHHAT USEFUL c. A LITTLE BIT USEFUL d. NOT AT ALL USEFUL Hhat did you think of the person who interviewed you? Did you: a. 591 LIKE HER/HIM AT ALL b. LIKE HER/HIM JUST A LITTLE BIT c. LIKE HER/HIM PRETTY MUCH d. LIKE HER/HIM A LOT How likely is it that you will return in six weeks for your second interview? Do you think that you: a. DEFINITELY EILL RETURN b. PROBABLY HILL RETURN c. PROBABLY 595;: RETURN d. DEFINITELY HON'T RETURN APPENDIX E RESPONSE CARDS 272 1 2 3 4 5 6 7 2 7'3 2 D m 987654321 274 CARD 3 l 2 3 A 5 6 7 DELIGHTED PLEASED MOSTLY MIXED MOSTLY UNHAPPY TERRIBLE (EXTREMELY SATISFIED (ABOUT DISSATIS- PLEASED) EQUALLY FIED SATISFIED AND DISSATISFIED) '275 CARD A I 2 3 A 5 6 7 NOT NOT VERY SOMEHHAT MIXED SOMEHHAT VERY EXTREMELY IMPORTANT IMPORTANT flIMPORTANT (ABOUT IMPORTANT IMPORTANT IMPORTANT AT ALL EQUALLY IMPORTANT AND UNIHPORTANT) 276 Card 5 I 2 3 A 5 NOT AT ALL A LITTLE BIT MODERATELY QUITE A BIT EXTREMELY APPENDIX F COTS' RECORD DATA COLLECTION FORM 277 COTS' RECORDS DATA Name: Case ID Date entered COTS: / / month’* day year Date left COTS: I / month’7 day year Birthdate: / / month ’day year Race: Marital status: Number of children with guest I married I Single I widowed I divorced I state custody mwao-I Under parole: __ 1 I yes __ 2 I no Past psychiatric hospitalization 1 I yes (Name of hospital Date of discharge - 2 I no Reason for leaving last permanent address: Objectives for the client: Has guest: discharged terminated I th? 278 Forwarding address indicated: __IIyes __2Ino Hhat will city of residence be? _ I I Detroit _ Z I other (specify ) __ 9 I missing/can't determine Hhat type of housing is indicated? _ 1 I house, apartment __ 2 I hospital (specify ) __ 3 I other (specify ) __ 9 I missing Record information on all incident reports received during stay: Date Reason for report (e.g. missed curfew, fighting, etc.) I. 2. 3. A. Total incident reports this stay: Hhat are client's goals? Hhat does he/she need to get permanently resettled? Other information/comments: LIST OF REFERENCES 279 LIST OF REFERENCES Agar, M. (1970). The professional stranger: An informal introduction to ethnography. New York: FAcademic Press. Amour, J. & Bedell, J. (1978). Facilitating return rates Of mailed program evaluation questionnaires. Evaluation and the Hea th Professions, 1, 81-88. Andrews, F. & McKennell, A. (1980). Measures Of self-reported well-being: Their affective, cognitive, and other components. Social Indicators Research, 8, 127-155. Andrews, F. & Withey, S. (1976). Social indicators of well- beingg Americans' perceptions Of life quality. New YOrk: Plenum Press. Appleby, L. and Desai, P. ‘(1985). DOCUmenting the relationship between homelessness and psychiatric hospitalization. Hospital and Community Psychiatry, pg, 732-737. Arce, A., Tadlock, M., Vergare, M., & Shapiro, S. (1983). A psychiatric profile Of street people admitted to an emergancy shelter. Hospital and Community_Psychiatry, 33, 812-817. Arce, A. & Vergare, M. (1984). Identifying and characterizing the mentally ill among the homeless. In H.R. Lamb (Ed.), Identifying and Characterizing the Mentally Ill Among the Homeless (pp. 75-89), Washington, D.C.: American Psychiatric Association. Bachrach, L. (1984a). Interpreting research on the homeless mentally ill: Some caveats. Hospital and Community Psychiatry, gs, 914-917. Bachrach, L. (1984b). Research on services for the homeless mentally ill. Hospital and Community Psychiatry, §§, 910- 913. Bachrach, L. (1984c). The homeless mentally ill and mental health services: An analytical review of the literature. In H.R. Lamb (Ed.), The Homeless Mentally Ill (pp. 11-53), Washington, D.C.: American Psychiatric Association. 280 Bachrach, L. (1984d). Deinstitutionalization and women: Assessing the consequences Of public policy. American Psychologist, gs, 171-1177. "___—“’— Bahr, H. (1973). Skid row: An introduction to disaffiliation. New York: Oxford University Press. Bahr, H. & Caplow, T. (1973). Old men drunk and sober. New York: New York University Press. Ball, F.L.B. & Havassy, B. (1984). A survey Of the problems and needs Of homeless consumers of acute psychiatric services. .Hospital and Community Psychiatry, §§, 917-921. Barrera, M. & Ainlay, S. (1983). The structure Of social support: A conceptual and empirical analysis. Journal Of Community Psyghology, 11, 133-143. - Bassuk, E. (1984). The homelessness problem. Scientific American, 251, 40-45. Bassuk, E., Rubin, L., & Lauriat, A. (1984). Is homelessness a mental health problem? American Journal Of Psychiatny, 141, 1546-1550. Baxter, E. 8 Hopper, K. (1981). Private lives/Public spaces: Homeless adults on the streets Of New York City. New YOrk: Community Service Society. Baxter, E. & Hopper, K. (1982). The new mendicancy: Homeless in New York City. American Journal of Orthopsychiatgy, §§, 393-408. Beels, C. (1981). Social support and schizophrenia. Schizo- phrenia Bulletin, 1, 58-72. Belle, 0. (1983). The impact Of poverty on social networks and supports. Marriage and FamilysReview, s, 89-103. Berkman, L. & Syme, L. (1979). Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents. American Journal of Epidemiology, 100, 186-204. Blumberg, L., Shipley, T., and Barsky, S. (1978). Li uor and poverty: Skid row as a human condition. Rutgers: Rutgers Center Of Alcohol Studies. Bogat, G. A., Chin, R., Sabbath, R., Schwartz, C. (1983). The Adult's Social Support Questionnaire. (Technical ReporE_NO. 2),7East Lansing: Michigan State University. 281 Brugha, T., COnrOy, R., Walsh, N., Delaney, W., O'Hanlon, J., Dondero, E., Daly, L., Hickey, & BOurke, G. (1982). Social networks, attachments and support in minor affective disorders: A replication. British Journal Of Psychiatry, 131, 249-255. Bruhn, J. 8 Philips, 8. (1984). Measuring social support: A synthesis of current appraoches. Journal of Behavioral Medicine, 1, 151-169. ' Bubolz, M., Eicher, J., Evers, S., & Sontag, S. (1978). A human ecological approach to quality Of life: Conceptual framework and results of a preliminary study. Social Indicators Research, 1, 103-136. Campbell, A. (1981). The sense of well-being in America. New York: McGraw-HillT’ Campbell, A., Converse, P., & Rodgers, W. (1976). The quality of American life. New York: Russell Sage Foundation. Christoff, C. (1986, May 25). Linking with millions will be a family affair. Detroit Free Press, p 15A. Cohen, C. & Sokolovsky, J. (1978). Schizophrenia and social networks: ex-patients in the inner city. Schiggphrenia \s Bulletin, 5, 546—560. Cohen, C. & Sokolovsky, J. (1983). Toward a concept Of homeless- ness. Journal of Gerontology, 38, 81-89. Cohen, S. & Hoberman, H. (1983). Positive events and social supports as buffers Of life change stress. .Jgppngl_gf Applied Social Psycholggy, 13, 99-125. Collin, R. (1984). Homelessness: The policy and the law. .ng Urban Lawyer, lg, 317-329. 'Comic Relief' for the Homeless: Comedians, HBO launch charity effort. (1986, January 16). 13g Washington Post, p 812. Congress of the United States House Of Representative Committee on Government Operations. (1985, April 22). Congressional committee declares homelessness a national emergency. Press release. Washington, D.C.: Congress of the United States Crystal, 5. (1984). Homeless men and homeless women: The gender gap. Urban and Social Change Review, 11(2), 2-6. 282 Darnton-Hill, I. (1984). Some past and present influences on the health of homeless men. (abstract) Community Health Studies, 8, 262. Davies, M., Rose, 5., 8 Cross, K. (1983). Life events, social interaction and psychiatric symptoms in general practice: A pilot study. Psychological Medicine, 1;, 159-163. Davis, F. & Yates, B. (1983). Collecting mental health self- evaluations: Effectiveness and cost Of third-party techniques. Evaluation and the Health Professions, Z, 91- 97. Depp, F. & Ackiss, V. (1983, July). Assessin6 needs among shel- tered homeless women. Paper presented at the conference ‘HOmelessness: A Time for New Directions, Washington, D.C. Derogatis, L. & Melisaratos, N. (1983). The Brief Symptom Inventory: An introductory report. Psycholggical Medicine, 13, 595-605. Dillman, D. (1978). Mail and telephone surveys: The total design method. NEw_YOFk: JOhOTWiTey BfiSons. Donald, C. & Ware, J. (1982). Thesguantification Of social contacts and resources (Report NO. R-2937-HHS)?’ Santa .‘MOhica: The—Rand’Corporation. Ell, K. (1984). Social networks, social support, and health status: A review. Social Service Review, §§, 131-149. Felice, J. (1984). An SRO hotel owned and Operated by Franciscans. Hospital and Communisy Psychiatry, gs, 1146- 1147. Fischer, P. (1984, August). Health and social characteristics Of Baltimore homeless persons. Paper presented at the meeting of the American Psychological Association, Toronto. Fischer, P., Shapiro, S., Breakey, W., Anthony, J., & Kramer, M. (1986). Mental health and social characteristics Of the homeless: A survey Of mission users. American Journal Of Public Health, 16, 519-524 Five million Join hands for needy. (1986, May 26). The Detroit News, plA. ' Flanagan, J. (1978). A research approach to improving our quality Of life. American Psychologist, 3;, 138-147. 283 From sea to sea, a show Of heart. (1986, May 26). Detroit Free Press, p 1A. Fruensgaard, K., Benjaminsen, 5., Joensen, S., & Helstrup, K. (1983). Social Psychiatry, 18, 129-135. Fustero, S. (1984, February). Home on the street. Psychology Today, 56-63. Futrell, C. & Lamb, C. (1981). Effect on mail survey return rates ' of including questionnaires with follow-up letters. Perceptual and Motor Skills, 88, 11-15. Garoogian, A. (1984). The homeless in America: A selected bibliggraphy. Public Administration Series: Bibliography, NO. P 1461, Monticello, Ill: Vance Bibliographies. - Garrison, V. (1978). Support systems Of schizophrenic and nonschizophrenic Puerto Rican migrant women in New York City.. Schizophrenia Bulletin, 4, 561-596. Gore, S. (1978). The effect of sOcial support in moderating the health consequences of unemployment. Journal of Health and Social Behavior, 18, 157-165. Gottlieb, 8. (Ed.). (1981). Social networks and social sgpport. Beverly Hills: Sage Publications. Grove, L. (1984, December 14). Birth of a street person. Wash- ington Post, pp. C2, C8, C9. Hammer, M. (1981). Social supports, social networks, and schizophrenia. Schizophrenia Bulletin, 1, 45-57. Hammer, M., Makiesky-Barrow, S., & Gutwirth, L. (1978). Social networds and schizophrenia. Schizoprenia Bulletin, 3, 522- 545. Heitzmann, C. 8 Kaplan, R. (1983, August). AssesSment of methods for measuring social support. Paper presented at the meet- ing Of the American Psychological Association, Anaheim, CA. Heller, K. & Monahan, J. (1977). Psychology and Community Change. Homewood, Ill.: The Dorsey Press. ' Henderson, 5., Duncan-Jones, P., McAuley, H., & Ritchie, K. (1978). The patient‘s primary group. British Journal Of Psychiatgy, 132, 74-86. 284 Hinrichs, J. (1975). Factors related tO survey response rates: Effects of sampling, follow-up letters, and commitment to participation on mail attitude survey reSponse. Journal Of Applied Psychology, 60, 249- 251. Holden, C. (1986, May 2). Homelessness: Experts differ on root causes. Science, pp 569-570.- Hombs, M. & Snyder, M. (1982). Homelessness in America: A forced march to nowhere. Washington, D.C.: The Community for Creative Nonviolence. Hope, M. & Young, J. (1984). From back wards to back alleys: deinstitutionalization and the homeless. Urban and Social Change Review, 17, 7-11. Hopper, K. (1984). Whose lives are these, anyway? (A comment on the recently issued Report on the Homeless and Emergency Shelters by the Department Of Housing and Urban Develop- ment). Urban and Social Change Review, 17(2), 12-13. Hopper, K., Baxter, E., Cox, 3., G Klein, L. (1982). One Year Later: The Homeless Poor in New York City. New YorE: Com- munity—Service Society. House, J. & Kahn, R. (1984). Measures and conCepts of social support. In S. Cohen & L. Syme (Eds.), Social Support and Health. New York: Academic Press. - House, J., Robbins, C., & Metzner, H. (1982). The association Of social relationships and activities with mortality: Prospective evidence form the Tecumseh Community Health Study. American Journal of Epidemiology, 116, 123-140. Hull, C. & Nie, H. (Eds.). (1981). SPSS update 7-9: New procedures and facilities for releases 7-9. New York: McGraw-Hill. Human Services Research Institute. (1985). Homelessness needs assessment study: Findings and recommendations for the Massachusetts Department of Mental Health. Boston: Author. Jones, R. (1983). Street peOple and psychiatry: An introduction. Hosital and Community Psychiatpy, 88, 807- 811. Jones, W. (1979). Generalizing mail survey inducement methods: Population interactions with anonymity and sponsorship. Public Opinion Quarterly, 38, 102-111. 285 Kanuk, L. & Berenson, C. (1975). Mail surveys and response rates: A literature review. Journal of Marketing Research, .12, 440-453. Kasinitz, P. (1984). Gentrification and homelessness: The single room occupant and the inner city revival. Uppap and Social Change Review, llll), 9-14. Katz, L. (1984, November 12). No person lives on the streets by choice. USA Today, p. 13A. Kaufman, N. (1984). Homelessness: A comprehensive policy approach. Urban and Social Change Review, 11(1), 21-26. Kenton, C. (1983). The Homeless. National Library of Medicine .Literature Search No. 83-35, Washington, D.C.: USDHHS Public Health Service. Krauthammer, C. (1985, January 4). For the homeless: Assylum. Washington Post, p. A15. Kroll, J., Carey, K., Hagedorn, C., Fire Dog, P., & Benavides, E. (1986). A survey of homeless adults in urban emergency shelters. Hospital and Community Psychiatry, 31, 283-286. LaHarpe, F., Blickle, J., & Gerhart, C. (1983). Ameliorer le suivi du malade alcoolique. Revue de l'Alcoolisme, 22, 153- 157. Lamb, R. & Grant, R. (1983). Mentally ill women in a county jail. Archives of General Psychiatpy, 39, 363-368. Laufer, W. (1981). The vocational interests of homeless, unem- ployed men. Journal of Vocational Behavior, 18, 196-201. Leavy, R. (1983). ‘Social support and psychological disorder: A review. Journal of Community Psychology, 11, 3-21. Lehman, A. (1983a). The well-being of chronic mental patients. Archives of General Psychiatry, pp, 369-373. Lehman, A. (1983b). The effects of psychiatric symptoms on quality of life assessments among the chronic mentally ill. Evaluation and Program Planning, 6, 143-151. Lehman, A., Ward, N., & Linn, L. (1982). Chronic mental patients: The quality of life issue. American Journal of Psychiatry, 13.2. 1271-1276. Leo, J. (1985, March 11). Harassing the homeless. Time, p. 68. 286 Levine, I. (1984). Service programs for the homeless mentally ill. In H.R. Lamb (Ed.), The Homeless Mentally Ill (pp. 173-200), Washington, D.C.: *American Psychiatric Associa- tion. Levinson, B. (1957). The socioeconomic status, intelligence, and psychometric pattern of native-born white homeless men. The Journal of Genetic Psycholpgy, 91, 205-211. I Liem, R. & Liem, J. (1978). Social class and mental illness reconsidered: The role of economic stress and social support. Journal of Health and Social Behavior, 12, 139- 156. Lin, N. & Dean, A. (1984). Social support and depression: A panel study. SoCial Psyghiatry, 12, 83-91. Lin, N., Dean, A., & Ensel, W. (1981). Social support scales: A methodological note. _Schizophrenia Bulletin, 7, 73-89. Linsky, A. (1975). Stimulating responses to mailed uestion- naires: A review. (Public Opinion Quarterly, §_, 83-101. 'Lipton, F., Cohen, C., Fischer, E., & Katz, S. (1981). Schizo— phrenia: A network crisis. Schizophrenia Bulletin, 7, 144- 151. Lipton, F. & Sabatini, A. (1984). Constructing support system for homeless chronic patients. In Lamb, R. (Ed.), The Homeless Mentally Ill, (pp. 153-172), Washington, D.C.: American Psychiatric Association. Lipton, F., Sabitini, A., & Katz, s. (1983). Down and out in the city: the homeless mentally ill. Hospital and Community Psychiatry, 33, 817-821. Marsella, A. & Snyder, K. (1981). Stress, social supports, and schizophrenic disorders: Toward an interactional model. Schizophrenia Bulletin, 2, 152-163. McFarlane, A., Neale, K., Norman, 6., Roy, R., & Streiner, D. (1981). Methodological issues in developing a scale to measure social support. Schizophrenia Bulletin, 1, 90-100. McGarrell, E. & Flanagan, T. (Eds.). (1985) Sourcebook of criminal justice statistics - 1984. U.S. Department of Justice, Bureau of Justice Statistics. Washington, D.C.: USGPO. McKay, P. (1986, Feerary, 16). My home is a lonely bed in a dreary D.C. shelter. Washington Post., pp. C1, C3. 287 McKennell, A. & Andrews, F. (1983). Components of perceived life quality. Journal of Community Psycholpgy, 1;, 98-110. Milburn, N. & Watts, R. (1984, August). A neglected §pecial pop- ulation: The homeless. Paper presented at the meeting of the American Psychological Association, Toronto. Miller, R. (1982). The Demolition of Skid Row. Lexington, Mass.: D.C. Heath and CBmpany. Miner, M. (1983). Preliminary contact with a mailed follow-up survey: Effect on rate of response of former mental patients. Evaluation Review, 7, 385-396. — Mitchell, R. (1982). Social networks and psychiatric clients: The personal and environmental context. American Journal of Community Ppychology, 19, 387-401. ' Monroe, S. (1983). SoCial support and disorder: Toward the untangling of cause and effect. American Journal of Community ngcholpgy, ll, 81-97. Mowbray, C. (1985). Homelessness in America: Myths and ‘ realities. -American Journal of Orthopsychiatry, §§, 4-8. Murrell, S. & Norris, F. (1983). Quality of life as the . criterion for need assessment and community psychology. Journal of Community Psychology, 11, 88-97. Murrell, 5., Schulte, P., Hutchins, G., & Brockway, J. (1983). Quality of life and patterns of unmet need for resource dec15ions. American Journal of Community Psychology, 11, 25-39. Nguyen, T., AttkisSon, C., & Stegner, B. (1983). Assessment of patient satisfaction: Development of a Service Evaluation Ouestionniare. Evaluation and Prpgram Planning, p, 299-314. Norbeck, J., Lindsey, A., & Carrie, V.‘ (1983). Further development of Norbeck Social Support Questionnaire: Normative data and validity testing. Nursing_Research, 32, 4-9. Pattison, E. & Pattison, M. (1981). Analysis of schizophrenic psychosocial network. Schizophrenia Bulletin, 1, 135-143. Perrucci, R. & Targ, D. (1982). Mentalpatients and Social net- works. Boston: Auburn House. 288 Phillips, S. (1981). Network characteristics related to the well-being of normals: A comparative base. Schizophrenia 'Bulletin, 1, 117-124. Pianin, E. (1984, November 17). City Forms Unit to Aid Homeless: Barry Says D. C. Still Challenges Initiative 17. Washington Post, pp. A1, A11. Pichirallo, T. (1986, June 21). 'Snyder, 26 others start new hunger strike. Washington Post, p. D3. Rabinowitz, D. ‘(1985, December 30). Degradation of the homeless. Washington Times, p. 30. Raspberry, W. (1986, January 6). The homeless, without jobs. Washington Post, p A13. Reagles, K. (1979). A handbook for follow-pp studies in the human services. New York: ICD Rehabilitation and Research en er. Reid, T. (1984, December 25). For homeless, the inn has room. Washington Post, p. A3. Roberts, J. (1985, December 9). Helping the homeless. New York. 118-121. Robertson, M. and Cousineau, M. (1986). Health status and access to health services among the urban homeless. American Journal of Public Health, 16, 561-563. Ropers, R. & Robertson, M. (1984, January). The inner-city homeless of Los Angeles: An empirical assessment. Paper presented at the Growing with the Needs of the Homeless Forum, sponsored by the Los Angeles County Task Force for the Homeless, Los Angeles. Roth, 0., Bean, J., Lust, N., & Saveanu, T. (1985). Homelessness in Ohio: A study of people in need. Columbus, Ohio: Ohio Department of Mental Health. Ryan, L. (1976). ‘Blaming the victim (rev. ed.). New York: Vintage. Sandler, I. & Barrera, M. (1984). Toward a multimethod approach to assessing the effects of social support. American Jour- nal of Community Psychology, 12, 37-52. — Sarason, I., Levine, J., Basham, R., & Sarason, B. (1983). Assessing social support: The Social Support Questionnaire. Journal of Personality and Social Psychology, 44, 127-139. 289 Schaefer, C., Coyne, J., & Lazarus, R. (1981). The health-related functions of social support. Journal of Behavioral Medi- - cine, 3, 381-406. . Segal, 5., Baumohl, J., & Johnson, E. (1977). Falling through the cracks: :Mental disorder and social margin in a young vagrant population.’ Social Problems, 33, 387-400. Sexton, P. (1982). Bibliography on homelessness. Unpublished manuscript, New York University, Department of Sociology. Shadish, W. (1984). Lessons from the implementation of deinsti- tutionalization. American Psychologist, 33, 725-738. Simpson, J., Kilduff, M., & Blewett, C.D. (1984). Struggling to survive in a welfare hotel. New York: Community Service Society. Smith, S. (1984, December 9). Thousands living from crisis to crisis. .Detroit Free Press, pp. 1A, 14A, 15A. Solarz, A. & Mowbray, C. (1985a, October). An examination of physical and mental health problems of the homeless: Implications for policy. Paper presented at the meeting of the Evaluation Research Society, Toronto. Solarz, A. & Mowbray, C. (1985b, November). An examination of physical and mental health problems of the homeless. ’Paper presented'at the meeting 6? the American Public Health Association, Washington, D.C. Solarz, A., Mowbray, C., & Dupuis, S. (1986). Michigan Human Services Task Force on the Homeless: Executive summary and report. LanSifig, MI: Michigan Department of Mental Health. Solenberger, A. (1914). One Thousand Homeless Men (2nd ed.). ’ New York: The Russell Sage Foundation. ‘ ‘ . Stafford, J. (1966). Influence of preliminary contact on mail returns. Journal of Marketing Research, 3, 410-411. Stern, M. (1984). The emergence of the homeless as a public problem. Social Service Review, 33, 291-301. Stokes, J. (1983). Predicting satisfaction with social support from social network structure. American Journal of Community Psyghology, 33, 141-152. 290 Stokes, J. & Wilson, 0. '(1984). The Inventory of Socially Supportive Behaviors: Dimensionality, prediction, and gender differences. American Journal of Community Psychology, 11, 53-69. Stoner, M. (1983). The plight of homeless women. Social Service Review, 31, 565-581. Stoner, M. (1984). An analysis of public and private sector provisions for homeless people. Urban and Social Change Review, 11(1), 3-8. Sutherland, E. & Locke, H. (1936). Twenty thousand homeless men: A study of unemployed men in the Chicagp shelters. Chicago: JJB. Lippincott. Talbott, J. & Lamb, H.R. (1984). Summary and recommendations. In H.R. Lamb (Ed.). The Homeless Mentally Ill, (pp.1-10). Washington, D.C.: American Psychiatric Association. Taube, C. & Barrett, S. (Eds.). (1985). Mental health, United States2 1985. OHHS Pub. No. (ADM) 85:1378T* Washington, Thoits, P. (1982a). Conceptual, methodological, and theoretical problems in studying social support as a buffer against life stress. Journal of Health and Social Behavior, 13, 145-159. Thoits, P. (1982b). Life stress, social support, and psychologi- cal vulnerability: Epidemiological considerations. Jour- nal of Community P§ychology, 19, 341-362. Thomas, E. (1985, February 4). Coming in from the cold. Time, pp. 20-21. Tolsdorf, C. (1976). Social networks, support, and coping: An exploratory study. Family Process, 13, 407-417. Tryon, R. 8 Bailey, 0. (1970). Cluster analyses. New York: McGraw-Hill. United Community Services of Detroit. (1985). Toward a better life: A new approach to serve the unemployed and the workipg poor. Detroit: Author. United States Bureau of the Census. (1983). .1980 Census of population: General social and economic characteristics for 'Michigan. Washington, D.C.: U.S. Department of Commerce. 291 United States Bureau of the Census. (1985). Statistical abstract ‘ of the United States: 1986. Washington, D.C.: U.S. Department of Commerce. United States Department of Justice, Bureau of Justice Statistics. (1983). Criminal victimization in the United Sates, 1981: A National Crime Survey report (NCJ-90208). Washington, D.C.: U.S. Department of Justice. Walsh, E.. (1986, May 21). Homeless initiative ruled valid. Washipgton Post, pp. A1, A9. . Warner, J., Berman, J., Weyant, J., & Ciarlo, J. (1983). Assessing mental health program effectiveness: A comparison of three client follow-up methods. Evaluation Review, 1, 635-658. Widgery, R. (1982). Satisfaction with the quality of urban life: A predictive model. American Journal of Community Psychology, 13, 37-48. ‘ Zautra, A. (1983). Social resources and quality of life. Ameri- can Journal of Community Psychology, 11, 275-290. ((311110HlHIli l 93 10766 3407