rtvu 11 1 ' 1...”... ... r'-.'r v n . 4:11;! a . . ~4....;..';1.::'v\- ’ l' , '17-'3’.’ 1...: \ T1 “Cu“ '5 uC?1-.\u‘.k\x -.«~- fgi."s' 4 rim-... «431:. 11‘... ...-1.; .x ..x... .1.» 1...» ..2.\ V1.r\L..-1..u .... ““L ' ur‘: 1. :1 1 ‘: «v . :ru.‘ 1.1.. «~- 1 “ ‘L q a ..." m '4 .1- . 41:“, ,‘1 , hwy». ' 1 .127. w 5:13:55: ... . n .... ... ' ”‘1 ...,“ ‘-r3(1:.1‘ 1-- '35. l .. .... II _' n... “r, 5:. 7.5-: v ‘kvrié’? _ :§;,',', 71’": 1 ...._.:Z.,.. ‘1. . I. .... .5: fl] . .'.'.'."'J.' .. .4 . .r- ,.:,I,_.;'.r r, v ... 4.,— ,...1. . .u-.€.({‘v1-——fl - .... 4 22%.? .x‘. '.‘ .3."- ’4 - . .::..:;.__.:. 1:. W“... ... 4 .,.... "M” “‘1. - '45 art'wol: 4. V'- -v- wwwdw- v ... . IES \llllllllllllllllllllllllll \ This is to certify that the thesis entitled Underclass Poverty Effects on the Severity of Homelessness presented by Jonathan Blair VanGeest has been accepted towards fulfillment of the requirements for M.A. degree in Sociology wmgm ' ' Major professor [)ate 0-7639 MSU is an Affirmative Action/Equal Opportunity Institution r’ \ LIBRARY Michigan State University Lg I PLACE IN RETURN BOX to remove We checkout from your record. TO AVOID FINES return on or before due duo. DATE DUE DATE DUE DATE DUE msu le An Affirmative Action/Equal Opportunlty Institution ammo-n1 UNDERCLASS POVERTY EFFECTS ON THE SEVERITY 0F HOMELESSNESS BY Jonathan Blair VanGeest A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology 1991 ABSTRACT UNDERCLASS POVERTY EFFECTS ON THE SEVERITY OF HOMELESSNESS BY Jonathan Blair VanGeest While the relationship between extreme poverty and homelessness has been recognized, only limited research has examined the effects of a person's prior socioeconomic status on the severity of the homeless experience. This cross— sectional study of homelessness and underclass poverty, conducted in Benton Harbor and Grand Rapids, Michigan, represents an attempt to expand the discussion concerning these two related social phenomenon. Interviews were conducted of homeless persons classified as chronic, episodic, or transitional. Services for the homeless in each community ‘were categorized as primary, intermediate, or ultimate, depending on the amount.of mobility offered the client. This study confirmed a relationship between type of homeless and prior socioeconomic background. Additionally, people from an underclass background demonstrated a less hopeful life perception. Greater length of time homeless also negatively effected the individual range of services utilized and the hopefulness of life condition. Copyright by JONATHAN BLAIR VANGEEST 1991 ACKNOWLEDGMENTS I would like to take this opportunity to express my appreciation and gratitude to all those involved in the succesful completion of this thesis. The thesis committee: Dr. William Ewens, Dr. John Schweitzer, and Dr. Marilyn Aronoff whose insight and support encouraged the highest degree of learning. Ms. Vicki Peck and Mrs. JoDee Fortino, whose skills allowed for a successful and timely submission. Mr. Steven Cox for his assistance with graphics and layout. And finally my parents, grandparents, and a very special uncle, whose understanding and support allowed me to pursue my educational goals. LIST OF TABLE OF CONTENTS TABLES O O O O O O O O O O O O O O O 0 LIST OF APPENDICIES . . . . . . . . . . . . . Chapter I. INTRODUCTION . . . . . . . . . . . . . . Statement of the Problem . . . . . . . . Purpose of the Study . . . . . . . . . . II. REVIEW OF THE LITERATURE . . . . . . . . The Urban Underclass . . . . . . . . . . Defining Underclass Poverty . . . . . Process . . . . . . . . . . . . . . . Cultures, Family, an Migration . . . Economic Changes and Barriers to Participation . . . . . . . . . . . . The Homeless . . . . . . . . . . . . . . Defining the Homeless . . . . . . . . Process . . . . . . . . . . . . . . . Reductions in the Supply of Low-income Housing . . . . . . . . . . . . . . . Structural Changes in the Economy . . Reductions in Welfare Entitlements . . Mental Health . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . III. METHODOLOGY . . . . . . . . . . . . . . . Benton Harbor . . . . . . . . . . . . Grand Rapids 0 0 O O O O O O O O O O O The Instrument . . . . . . . . . . . . iv Page vi ix 10 10 11 17 17 23 32 32 37 38 41 44 46 47 49 49 52 59 Chapter IV. RESULTS . . . . . . The Characteristics of the Homeless Sample in Benton Harbor and Grand Rapids, Michigan Testing the Hypothesis Service Use . Hopelessness . Service Use . . Hopelessness Conclusion . V. SUMMARY AND CONCLUSIONS Implications of the Findings . Limitations of the Study . Conclusions . . REFERENCES . . . . . . . APPENDIX A O O O O O O 0 APPENDIX B . . . . . . 62 62 67 72 78 80 89 93 94 96 99 101 103 112 119 Table LIST OF TABLES Change in Number of Jobs in Selected Central Cities and Suburban Rings, by Occupational Sector, 1970 - 1980 Change in Number of Jobs in Selected Central Cities and Suburban Rings, by Industrial Sector, 1970 - 1980 Change in Number of Central-City Jobs, by Education Level of Jobholders, 1970 - 1980 Percentage of Central-City Jobs, by Education Level of Jobholders, 1970 - 1980 Constant-Dollar Average Monthly Transfer Payments. 1968-1985 Benton Harbor Greater Area Population Statistics 1960 to 1980 Benton Harbor Greater Area Black Population, 1960-1980 Number of Manufacturing Establishments and Employees, Benton Harbor and St. Joseph, 1966 - 1986 Percent Net Gain/Loss of Manufacturing Establishements and Employees, Benton Harbor and Saint Joseph, 1966-1986 Employees and Establisments by Industry, Kent County, 1969-86 Statistics by Selected Industry Group, Grand Rapids, 1967 - 1982 Key to Conceptual Service Organization Transitional, Episodic, and Chronic Homeless in Benton Harbor and Grand Rapids vi Page 26 27 28 29 45 50 50 51 52 54 55 57 64 Table Page 4.2 Reported Work Histories for Homeless in Benton Harbor and Grand Rapids 65 4.3 Reasons for Homelessness in Benton Harbor and Grand Rapids 66 4.4 Perceived Availability of Family and Friends in Time of Personal Crisis 67 4.5 Crosstabulation of Type Homeless by Prior Socioeconomic Status 68 4.6 Regression Analysis of Time Homeless by Prior Socioeconomic Status 70 4.7 Crosstabulation of Type Homeless by Child Socioeconomic Status 71 4.8 Anova for Differences in Number of Services Used by Prior Economic Status 73 4.9 Anova for Differences in Primary Service Use by Prior Socioeconomic Status 74 4.10 Anova for Differences in Intermediate Service Use by Prior Socioeconomic Status 75 4.11 Anova for Differences in Ultimate Service Use by Prior Socioeconomic Status 76 4.12 Breakdown of Service use by Prior Socioeconomic Status 77 4.13 Regression Analysis of Hopelessness by Prior Socioeconomic Status 79 4.14 Results of the Prediction of Hopelessness by Thirteen Predictors Reflecting Poverty Conditions 81 4.15 Hopelessness by Socioeconomic Status 82 4.16 Anova for Differences in Number of Services Used by Length of Time Homeless 83 4.17 Anova for Differences in Primary Service Use by Length of Time Homeless 85 4.18 Anova for Differences in Intermediate Service Use by Length of Time Homeless 86 vii Table Page 4.19 Anova for Differences in Ultimate Service Use by Length of Time Homeless 87 4.20 Breakdown of Service Use by Amount of Time Homeless 88 4.21 Crosstabulation of Length of Time in Crisis by the Type of Service at the Place of Interview 90 4.22 Regression Analysis of Hopelessness by Length of Time Homeless, Homeless History, and Number of Times Homeless 91 4.23 Hopelessness by Time Homeless 92 viii LI ST OF APPENDICIES Appendix Page A. Selected Population Characteristics 112 B. Personal Questionnaire 119 ix CHAPTER I INTRODUCTION As we struggle to redefine extreme poverty and understand today's homelessness, much of the human element or detail associated.with the subject is lost in our abstractions” IMuch can be learned from the varied individual life experiences of those who have endured such a fate. £2221 Henry was interviewed at a community health clinic in Grand Rapids, Michigan. A white gentleman, about sixty-five years old at the time of the interview, he had spent the last fifteen years in the Grand Rapids area. A high school dropout, Henry served as an army cook in Korea, retiring just prior to Vietnam. Since leaving the army he has only been employed on a temporary basis, working in low paying jobs such as bus boy, parking attendant, and short order cook. Henry is an alcoholic and he blames alcohol for his intermittent employment following his retirement from the army. Henry has been in and out of local missions for the last year and a half. He has been assaulted and robbed more times 2 than he can remember, and his nose was broken in the spring of 1990. About three months ago he offered to help prepare the evening meal at one of these local missions. While working, he was repeatedly told not to cook food that would taste too good for fear that homeless people would come from other missions or from across town. Disgusted by the living conditions and the attitude of the mission staff, Henry eventually left» Henry has no family and few friends in the area that can help. At the time of the interview, he was unsure of where he was going to go next. 3.12.1152; Michael was greatly distressed at the time of his interview in the Benton Harbor Soup.KitchenJ He had just sold the last of his possessions (a tackle box with miscellaneous fishing equipment) and was preparing to leave the city in an effort to find work. Michael, a black man, is thirty years old and has lived in Benton Harbor all his life. He was reasonably well kept at the time of the interview. Michael lost his job at a local factory two years ago. He has graduated from high school but admits that he has few skills valued by employers. His marriage fell apart soon after he became unemployed. Last year he spent the entire winter sleeping in his van. JMeals are taken at the local soup kitchen and he receives some help from friends. He did not think that he could spend another year in the van and decided to move before winter set in. Michael is unsure of where he 3 will go. He suggests "south" but settles for "wherever the money will take me." Anything is considered far better than a van in a Michigan winter. 2531a gag Aggie Paula was interviewed at a Benton Harbor shelter for women and families of women. Paula, a white woman, was fourty-one at the time of the interview. She has been in and out of shelters (in five different states) over the past ten years. This is the second time in two years that Paula has stayed at this particular shelter. Paula graduated from high school, but has only held occasional part-time employment. By self-report, it has been over 10 years since she had last held meaningful employment. She has never been married. Angie is Paula's eleven year old daughter. Because of the many moves, Angie has never stayed in one school over the course of any given year. Of late, Angie has not been going to school at all because she is eight months pregnant. Angie has had limited prenatal care while in the various shelters. The father of the child is believed to be one of her mother's old boy friends. 33:! Mary, a ‘white 'woman, was in. her mid-thirties when interviewed in the Benton Harbor Soup Kitchen. Mary is mentally impaired, having been released from a local institution in the mid-1980's. Mary herself is not exactly 4 sure of how long ago it was that she was released. A room in a local hotel was secured for her when she left the institution, but she no longer stays there regularly and is unsure why this benefit is no longer provided. Mary spends a great amount of her time on the streets of Benton Harbor, occasionally sleeping at the local single room occupancy hotel if money is available. When money is not available, she often stays with friends or seeks the assistance of local churches. Lunch at the soup kitchen is the part of the day that Mary looks forward to ‘the most. She often chats with friends, assists the volunteers, and offers encouragement to anyone who passes her way. Volunteer staff at the Soup Kitchen express the concern that more care be given to Mary considering the severity of her emotional and physical handicaps. let-L! Betty, a black woman, was twenty-four years old at the time of her interview in a downtown Grand Rapids rescue mission. Betty has three children under the age of seven. Repeatedly abused by the children's father and unable to gain access to an area safe shelter, Betty is keeping distance between herself and her husband by moving between local missions“ There is numerous evidence of abuse and Betty talks freely of the family's situation. Pregnant in her senior year of high school, Betty dropped out to have the baby and get married. She has not returned to 5 school, has never held a steady job, and has few other resources on which to drawn Her family moved out of the Grand Rapids area a couple of years ago. She has not retained contact with her parents, and is not sure if she would be welcome in their home. There.is a constant fear that her husband will locate the family and Betty does not think they will be safe until she can save enough to move away from the city. She hopes that her stay in the mission will give her time to get a job and save the needed money. Bob and gagen Bob and Karen were interviewed in an intermediate shelter facility in Grand Rapids, Michigan. IBob and Karen are married and have just moved into the area with their three children. Bob is twenty five years old. He graduated from high school, but has not held a steady job for over two years. The family moved to Grand Rapids hoping that Bob could find work in construction. Karen had their first child before she was eighteen years old and did not finish high school. Karen has never held steady employment. Bob's mother stays with the family in the small two bedroom facility. The family had lived in a trailer in a southern community before coming to western Michigan. The trailer was owned by Bob's.mother and.was sold to finance their move. ‘The children are all under six years old. The oldest has not started school because of the change. 6 Both Karen and Bob find the transitional housing project to be far better than their previous shelter accommodation within a local mission" Bob is able to remain with the family at this facility and there is a great deal more privacy. This facility will allow the family to stay for only thirty days. If Bob does not find work, the family will have to move back to the mission and live apart until further arrangements can be made. The stories and observations, as recorded by the author within the context of this present study, assist in attaching a human face to a social problem that is, for a vast majority of Americans, all too distant and ill-understood. Homelessness is a diverse problem, not solely a product of specific personal traits held by those afflicted. Homelessness is also not monocausal in a societal context (e.g. homelessness not solely a housing problem). All of the faces presented above share one condition that supersedes all others - aggravated extreme poverty. The decade of the eighties has seen a dramatic increase in the ranks of the homeless, particularly in our nation's cities. Single women with children, teenage runaways, families, and single men are all significantly represented in this most diverse population. This past decade has also witnessed an increase in the numbers of the urban underclass, a group characterized by conditions of extreme poverty; hopelessness and despair; high incidence of single parent 7 families, teenage pregnancies, criminal victimization; and near total social and economic isolation. Extreme poverty is the root cause of both homelessness and underclass formation. This poverty is the result of the overall loss of low-skilled job opportunities, the reduction in welfare entitlements, and the greatly diminished supply of low-income housing. There is a relationship between the extremely poor and the homeless, as the extremely poor constitute the greatest pool from which the homeless are drawn. For many homeless, periods of domiciled extreme poverty existed for years prior to their literal homeless participation. It is the major hypothesis of this analysis that periods of domiciled - underclass - poverty have the capability to affect the severity of the individual's homeless experience based on an ability to restrict needed resources; possibly extending the time homeless, and increasing the relative hardship experienced by the participant. Statement of the Problem It is a small jump between extreme poverty and homelessness. The economic and social isolation that characterizes the urban underclass, and the individual resources there-in, increases their probability of homeless participation. Both time (duration/extent of poverty) and place (patterns of movement) are part of the gradual downward drift that often precedes the condition. Time homeless and location can also affect service use and the severity of the 8 homeless experience. To assist in analysis, homeless persons are recognized as chronic, episodic, or transitional based on their length of time and number of times homeless. The emergency service delivery system for the homeless is designed to provide first tier emergency services (soup kitchens, food pantries, shelters), followed by second tier transitional housing, and third tier permanent housing. Theoretically, individuals are supposed to proceed through the various levels of service to achieve permanent housing. This study classified services at each level of the delivery network as primary, intermediate, or ultimate depending upon the amount of social or economic mobility offered the client. Although homeless participants are supposed to advance through the three tiers to achieve permanent housing, a significant number remain in first tier primary services that provide for life maintenance without offering avenues out of the homeless condition. When examining the uneven use of services among the homeless, questions are raised about the degree to which poverty, associated with an individual prior to service use, affects their movement through, and use of, the service network. .Are those most deprived (underclass) more likely to be chronically homeless, with the middle- and lower-classes more likely to exit the homeless population prior to the chronic state? Do homeless people from an underclass background utilize a narrower range of services and have a 9 less hopeful perception of their life condition? Purpose of the Study The purpose of this study is to begin the examination of the relationship between the social phenomena of homelessness and underclass poverty. The focus of the research will be on the homeless, as there is a great deal more information on this population than there is on the urban underclass or extreme poverty in general. Emphasis will be given to the findings from a 1990 study of homelessness and the urban underclass conducted by Dr. John Schweitzer and myself. Confirmation of the expectation that large numbers- of the chronic homeless have underclass backgrounds may call to question.many of the current policies and programs designed to address the needs of the homeless. New'policies would.have to be developed that specifically take into account the underclass backgrounds of many of the homeless in our inner cities. CHAPTER II REVIEW OF LITERATURE Th ban de 1 55 Discussion of the urban underclass has brought the subject of poverty back into the forefront of sociological discourse. In the mid-1960's, liberal urban analysts had begun to discuss the rise of social dislocations among the ghetto "underclass", a large subpopulation of low-income families and individuals whose behavior obstructed full participation in the mainstream economy (Wilson, 1987; Patterson, 1986). High rates of joblessness, teenage pregnancies, out-of—wedlock births, female-headed families, welfare dependency, and high crime rates characterized identified pathologies destined to reach catastrophic proportions within this subpopulation by the mid-1970's (Wilson, 1987). Early conservative attempts at explanation attributed the development of inner-city pathologies to the interconnection between cultural traditions, family history, and individual behavior (Wilson, 1987; Lemann, 1987; Patterson, 1986). Liberal critics of conservative thought 10 ll countered, believing that ghetto specific behavior was largely due to segregation, limited opportunities, and external obstacles against education as determined by historical circumstance (Wilson, 1987; Patterson, 1986). Liberal scholars in the 1960's argued that cultural values do not ultimately determine behavior or success. Rather, cultural values emerge from specific social circumstances and life chances and reflect one's class and racial composition (Wilson, 1987). Urban field studies in the late 1960's supported the probability of this relationship between culture and social structure. Criticism of the ‘culture of poverty' thesis and the debate over the disintegration of the black family prompted many liberal urban analysts to refocus their position in the early 1970's. Liberals became reluctant to continue research on those social dislocations associated with the underclass, focusing instead on the strengths of the inner- city community (Ellwood, 1988; Wilson, 1987; Hammer, 1986). In the face of a recessed liberal argument, the conservative perspective came to the fore in the mid-1970's, charging that social welfare programs - the very product of liberal social policy and the war on poverty - had the perverse consequence of increasing poverty by exacerbating ghetto- specific cultural tendencies and pathologies (Kaus, 1986; Murray, 1984). The extraordinary rise in inner-city poverty following the passage of the most sweeping 12 antidiscrimination and antipoverty legislation in the nation's history was used as evidence in support of the conservative agenda. ConserVative vs. liberal debate flowered anew in the mid-1980's following the reorganization of the liberal argument in response to contemporary policy dictates. The focus of the new debate on poverty was now an urban underclass so pervasive as to become synonymous with poverty as a whole. As can be expected from the wide differences in perception resulting from the varying philosophical perspectives on the problem, the underclass phenomenon has been noted for its complexity and general lack of precise concepts (McLanahan & Garfinkel, 1989; Hammer, 1986). Defining Underclass Poverty Much disagreement exists on the causes for the recent growth of the underclass, and even to the extent to which growth has occurred (Sullivan, 1989; Hammer, 1986). A major problem in assessing growth or causation is the wide range in definition used to describe the phenomenon. William Julius Wilson (1987), in his seminal work on the concept, defined the ghetto underclass as persons who live in neighborhoods "populated almost exclusively by the most disadvantaged segment of the black urban community, that heterogeneous grouping of families and individuals who are outside the mainstream of the American occupational system. Included in this group are individuals who lack 13 training and skills and either experience long-term unemployment or are not members of the labor force, individuals who are engaged in street crime and other forms of aberrant behavior, and families that experience long-term spells of poverty and/or welfare dependency" (Wilson, 1987, p8). In an attempt to operationalize this rather unwieldy concept, Wilson designated as underclass those urban census tracts where at least twenty percent (20%) of the people were below the poverty level. Ghetto underclass areas were further divided into three categories to reflect the varying degrees of concentration of poor people. Mild poverty areas are defined as those tracts with between 20 and 29% poor people; moderate poverty areas are those with between 30 and 39% poor people; and extreme poverty areas are those with 40% or more poor people (Wilson, 1987). Using Chicago census data, Wilson illustrated a phenomenal growth and spread of poverty areas between 1970 and 1980. Accompanying this growth was a heavy and increasing concentration of poor people in these areas, with fewer and fewer poor people living in non-poverty areas. Definitions of underclass poverty have been so broad as to include all persons both urban and black (Hammer, 1986). Others are rooted in the pathological manifestations associated with either the individual (Kasarda, 1989), or place (Sawhill, 1988). John D. Kasarda (1989) defined the 14 urban underclass as an "immobilized subgroup of socially isolated, persistently poor ghetto dwellers characterized by substandard education and high rates of joblessness, mother- only households, welfare dependency, out-of-wedlock births, and crime" (Kasarda, 1989, p29). The vast variations in definition reflect the difficulty in isolating that specific social phenomenon to which the term "underclass" refers. It has been questioned whether historic or institutional racism is important when defining the urban underclass (Wilson, 1987), when the economic position of the inner-city poor blacks actually deteriorated during the very period in which the most sweeping antidiscriminatory legislation and programs were enacted and implemented. It has also been contended that if contemporary discrimination was the main culprit, why were the most severe problems encountered in that same decade following the civil rights legislation and affirmative action programs (Wilson, 1987). Rather, race-neutral economic factors (Wilson, 1987; Wilson, 1980) are presented as responsible for underclass formation. Wilson (1987) gave weight only to "historic racism" when explaining why such factors disproportionately effect the black minority. When examining race-neutral economic factors in the nation's five largest cities and defining underclass poverty, Wilson still attributed the extraordinary increase in both the poor and nonpoor populations in the extreme 15 poverty areas between 1970 and 1980 solely to the changes in demographic characteristics (e.g. increasing unemployment, female headed households, etc.) of the black population. Loic J.D. Waquant and Wilson (1989), Nicholas Lemann (1986), and Thomas Hammer (1986) also used similar or more directed race-specific strategies in their definition and analysis. Using Wilson's five-city study or similar data, authors have criticized the race-specific nature of many underclass definitions (Bonacich, 1989; Billingsley, 1989; Marks, 1989; Jenks, 1988). The focus of the varied arguments is on a substantial (24%) increase in the concentration of poor whites in poverty areas and the emergence within this subpopulation of similar social dislocations as found within the black urban underclass. When examining economic processes, Billingsley (1989) asserts that "the major factors pushing blacks into poverty, into inner cities, and into what Wilson calls the ‘tangle of pathology'... are the same forces causing these problems among low- and moderate- income whites" (Billingsley, 1989, p26). The question of race in underclass definition continues to be hotly debated. While there is recorded a 24 percent increase in the concentration of poor whites in poverty areas, the total black population in extreme poverty areas increased by 148 percent between 1970 and 1980 (Wilson, 1987). Whatever the reasons which attributed to this increase, it is clear that the responsibility of underclass 16 poverty participation rests disproportionately on the shoulders of the inner-city black population (Kasarda, 1989: Wilson, 1987; Hammer, 1986). While considering this responsibility, it is still wrong to define the concept in race-specific terms. The economic processes at work effectively isolate both black and white poor (Billingsley, 1980; Jenks, 1988). Racism, and the dynamics of race and economic change (Jenks, 1988), must be confronted when attempting to understand the disproportionate participation of blacks in the urban underclass. Wilson's census-based operationalization of the‘ underclass definition has been criticized as being too broad (Billingsley, 1989; Ellwood, 1988). It is questioned whether the development of the necessary social/economic isolation and the concentration of poor people can occur within census tracts with only 20 percent of their population below the poverty line (Billingsley, 1989). Ellwood (1988) suggests a ghetto poverty area as a more natural measure of the urban underclass. Such an area in a moderate-size or large city would designate as "underclass" only those neighborhoods characterized by a poverty rate of 40 percent or higher and a high degree of behavioral troubles as measured by dropout levels, unemployed prime- aged males, welfare recipients, and female-headed households. 17 Questions of definition are not purely rhetorical, as differences dictate changes in population parameters affecting the magnitude of the condition. Being a census- based measure, Ellwood's ‘ghetto poverty area' would effectively examine only those extreme poverty areas (with a poverty rate of at least 40%) as delineated by Wilson. A review of only extreme poverty areas in the nation's ten largest cities reveals the underclass to represent only 7 percent of the total poor (Ellwood, 1988). If the analysis - was to include only black poor, only 5 percent of the total poor would be labeled as underclass. This percentage would be much lower than that defined by Wilson. Definitial difficulties abound, but however one defines the condition, it must be remembered that underclass poverty represents only a fraction of our nation's poor (Ellwood, 1988). EEQQ§§§ Much disagreement exists on the causes of the recent growth of the underclass. Analysis of urban underclass formation have generally followed the liberal vs. conservative debate of the 1960's, with greater emphasis given to economic processes or the impact of welfare by the respective camps. Cultures, Eamily, and Migration Nicholas Lemann (1986), while examining the bifurcation of black America, contends that the clearest line between the black middle-class and the underclass is family 18 structure - single parent family structure being a product of ghetto or underclass culture. Further, Lemann states that "every aspect of the underclass culture in the ghetto is directly traceable to roots in the South - and not to the South of slavery but the South of a generation ago. What happened to make the underclass grow so much in the seventies can best be understood by thinking less about welfare and unemployment than about demographics - specifically, two mass migrations of black Americans" (Lemann, 1986, p35). The first migration was from the rural South to the urban North, the second a migration out.of the ghettos by members of the black working and middle classes. The key division in Lemann's analysis is between rural black migrants and migrants coming from Southern towns and cities. Black Southern migration into Chicago prior to 1950 is described as coming from a more advanced Southern urban social structure, with working class values. Lemann contends that the migration directly affecting growth in the underclass came later - between 1950 and 1970 - and was from the rural South, bringing with them a share cropper mentality and an ‘ethic of dependency'. In the city - away from family, religious and social structures of small-town life back home - all the migrants experienced a loosening of the constraints on their behavior; a process exacerbated by the rapid exodus of the working black population. 19 Female-headed families sanctioned by the Southern black cultural heritage and an implied predilection of blacks toward out of wedlock pregnancy, created an underclass culture in the city characterized by poverty, crime, low educational achievement, and a low work effort. The spread of underclass culture in contemporary society (following the stagnation of migration in the 1970's) is the function of the increased gang recruitment of youth and their subsequent participation in illicit gang activity. William Julius Wilson (1980) also assigned special significance to black migration between the turn of the century and 1970. Critically important was the sudden growth and disproportionate percentage of young persons between fourteen and twenty-four in the central-city black population; their number rising by 78 percent between 1960 and 1970. A review of statistics show that persons under the age of twenty-one have greater rates of violent crime and out-of-wedlock births (Wilson, 1987; Hammer, 1986). Demographic change alone within this population can be expected to disproportionately effect those social dislocations associated with the extremely poor. Significantly, studies of migration, welfare recipients, and poverty status (Long, 1974) have revealed black migrant families to have lower welfare participation and poverty rates than non-migrant families. In support of this analysis, Wilson (1987) reported data from a second 20 study (Long & Heltman, 1975) which employed utilized measures of income as the dependent variable and considered the effects of education, labor-force participation, occupation, and extent of unemployment. Lemann argues that welfare receipt among non-migrants is a testament to the destructive effects of Northern ghetto life, but his argument is anecdotal (Hammer, 1986) and lacks the empirical documentation necessary for causation. Sullivan (1989) argues that culture does play a role in shaping local responses to teenage pregnancy, female-headed households, and welfare - a role tempered by the interrelated effects of structural economic change and social ecology. Wilson believed that cultural behaviors were largely the result of opportunity and/or repression, and not ultimately determining individual success or failure. Contrary to that proposed by Lemann, as social and economic opportunities change, new behaviors emerge and become reinforced by developing social norms. In his analysis, Lemann (1986) described welfare receipt as a trap, its tidal pull toward dependency stronger for people on the rolls than those who are working and considering their options. While expenditures on social welfare programs increased enormously between 1960 and the early 1970's, little improvement in poverty was recorded. Conservatives charged that antipoverty programs had failed, possibly even reducing the initiative of the poor to seek 21 self-sufficiency (Kaus, 1986; Murray, 1984). Charles Murray (1984) advanced the dependency thesis, concluding that welfare generosity had the unintended effect of increasing poverty. Murray believed that more people became dependent on government assistance as more money was poured into the war on poverty. Increasing rates of joblessness, crime, out-of—wedlock births, and female-headed households are also explained as products of a system that effectively changed the rewards and penalties that govern human behavior. Both Murray and Kaus (1986) address the underclass specifically as a product of welfare's ability to undermine autonomy, motivation, and personal responsibility. After reviewing welfare use and poverty in America, Ellwood and Summers (1986), Burtless (1986) and Danziger et al. (1986) conclude that welfare expenditures were in fact too small to ever have much effect on measured poverty. Cash assistance moves only 5 percent of poor persons out of poverty. Spreading the $20 billion dollars spent on cash assistance across 30.6 million poor persons yields an average cash benefit of slightly more than $50 per poor person per month (Ellwood & Summers, 1986, p86). More importantly, only modest changes in expenditures were recorded between 1970 and 1980. Over the entire decade, annual real expenditures per non-elderly poor person rose only $93 - hardly enough to effect movement of poor people across the poverty line. 22 Comparisons of changes in family structure over time, with changes in the welfare system and of differences in family structures across states, to differences in welfare benefits across states were used to examine the effects of welfare receipt on family structure. Trends linking increases in both the fraction of all children living in female-headed households and number of children on AFDC diverge sharply in the late 1960's. Since 1972, the fraction of all children living in female-headed households rose from 14 percent to almost 20 percent. During that same time period, the fraction of all children in houses collecting AFDC held almost constant at 12 percent. The figures are even more dramatic for blacks. Between 1972 and 1980, the number of black children in female-headed families rose nearly 20 percent, while the number of black children on AFDC actually fell by 5 percent (Ellwood & Summers, 1986, p94). If welfare was effectively tearing families apart, it is difficult to understand why the numbers of children on AFDC would remain nearly constant at the very time dramatic increases in female-headed households were recorded. Conservative charges of welfare dependency are less easily challenged. While most people who receive AFDC benefits stay on the program for a relatively short time, Wilson (1987), Hammer (1986), Ellwood & Summers (1986), and Ellwood (1988) reviewed studies which identified a very small proportion (1% -2%) of Americans in poverty as 23 persistently poor. A vast majority of the persistently poor are identified as coming from female-headed households. Disagreement exists on the extent to which the persistently poor contribute to underclass poverty. No definitive work exists on the extent to which pathological dependency occurs or on the role of welfare in creating dependency (Wilson, 1987; Ellwood & Summers, 1986; Hammer, 1986). Ellwood and Summers suggest that such dependency is easily explained: self-sufficiency for single mothers is only gained through work and marriage, conditions hampered by the presence of small children. Economig Changes and Barriers to Participation Wilson (1987) portrayed the dynamics of the underclass as an interplay between ghetto-specific cultural characteristics and economic opportunity. Structural economic changes shifted major urban industrial areas from centers of production and commodity distribution to centers of administration, information, exchange, and advanced service provision. Rapid increases in poverty followed the onset of this transition with the nation's five largest cities recording an increase of some 22 percent between 1970 and 1980. An additional byproduct of economic transformation was an increased concentration of poor people in territorial enclaves delineated by acute social and economic marginalization (Jenks, 1988; Wilson, 1987). Between 1970 24 and 1980, the population residing in poverty areas within the nation's five largest cities grew by 40 percent overall. Furthermore, the population living in high poverty areas (with a poverty rate of at least 30%) increased by 69 percent and extreme poverty areas (with a poverty rate of at least 40%) recorded a staggering 161 percent increase. Waquant and Wilson (1989) describe the rapid increase in inner-city poverty and resultant concentration of poor people as a process of hyperghettoization. Among those structural economic shifts associated with this process of hyperghettoization are "the decentralization of industrial plants, which commenced at the time of World War I but accelerated sharply after 1950, and the flight of manufacturing jobs abroad, to the Sunbelt states, or to the suburbs and exurbs at a time when blacks were continuing to migrate en masse to Rustbelt central cities; the general deconcentration of metropolitan economies and the turn toward service industries and occupations promoted by the growing separation of banks and industry; and the emergence of post-Taylorist, so-called flexible forms of organization and generalized corporate attacks on unions - expressed by, among other things, wage cutbacks and labor contracting" (Waquant & Wilson, 1989, p11). The very sustenance of the ghetto is threatened as traditional blue-collar employment becomes rapidly scarce. Faced with intense competition in a shrinking labor-market, 25 blacks suffered a deterioration in their economic position on nearly all major labor-market indicators. John D. Kasarda (1989) further defined the implications of urban industrial transition for the changing nature of jobs available in the inner-city. Table 2.1 shows that while all six of the largest Northern cities gained employment in every occupational classification, the central city lost substantial numbers of jobs in clerical, sales, and blue-collar occupations; substantial growth only being recorded in managerial, professional, and high-tech functions. A cross-classification of blue-collar employment change by industry (Table 2.2) establishes the greatest percentage of inner—city blue-collar losses in traditional urban manufacturing sectors. Central-city employment increases in managerial, professional, and high-level technical and administrative support occupations that occurred concurrently with precipitous drops in blue-collar and other jobs requiring lower levels of education, contributed to major changes in the educational composition of occupants of central city jobs (Kasarda, 1989, p30). The nature of this change in educational composition is presented in Tables 2.3 and 2.4. Net change and percent change in the number of inner-city jobs by education of employees is revealed in table 2.3. Table 2.4 presents the changes in educational structure of city employment corresponding to the urban economic 26 1’1.’. I‘-‘ |‘ | | Illllll II III 1‘ . l‘l ll o ”a. vino.mwu m: on: .moucmo m u o smmusm .OOHOEEOO O unmEuummm. .momH .ocuuuax. chad .madsum daouo sucsoo «ma ..oanaxomma .oadeum a am .oaam madeum «unwound: on: .m.mluuoouoom I, llnl. J oom.HsH oom.m~ oem.om omm.mm om~.om mnuanam 666.65: oo~.m>n coo.¢mu omm.mm oeo.n~ suflo Huuucmo mandamcmflflnm omo.mm¢ omo.s~ oeo.am oom.oH~ ooa.oom mausnsm omo.mmu oom.H>Hn omm.smau om>.nsfi 66¢.om sufio Hmuucmo h xuo» 302 _ . omm.mmH o~n.m~ o¢~.m¢ oom.~o oom.am mnuaasm J oom.¢oau oom.mmu o4m.mnu owm.ma ooh.¢ sufio Hmuucmo vacuumo M 666.56 oom.m~ oom.mH oma.m~ ova.om mausnsm m o~>.~¢u omm.¢nn om~.m~u o¢~.¢a oom.m mufio Hmuucmo m ocwao>mao _ M owo.ono oom.sn~ oom.mHH omo.o~H oma.6ma mausnzm _ omm.mmu oom.maflu own.mmu ooe.wo oem.flm sufio Hmuucmo m oomOfiso _ owm.mwn 64¢.GHH owq.mm omm.mn 666.cofl mnuanam _ om¢.m¢u oom.~mu 664.64: con.om oma.w~ mufio Hmuucmo _ coumom h IIAdeHIIIIMMdflmeonm mwamm can uuoomom Hmcowwwwuoum Maud ” Hmoflumau m>fiuouumflcfleo< can cmuflaomouuoz _ can HMOwcnomB HMflummmcmz . coma I obma .uouomm Hmcofiumooooo am .mmcfim cannonom can mmwuflo Hmuucmu oouomamm :a moon mo umnaoz cw amouno .H.~ manna 27 Ammqa .mouonmxv .mameum muouo sucsoo am" ..cuauuommfl .madeum max and on: .moocoo 0 u 0 smousm unmousom oom.m~ ooh.oa oop.a mausnsm oo~.m>u oem.au cam.mau om~.Hn omm.mu oms.w¢u aufio Hmuucmo adamaocaawnm omo.s~ om~.ma oon.oan owm.ma omm.>n oom.mnu mausnsm oom.H>Hu oo¢.¢ owm.m¢u ooo.o own.¢~n omo.moau mufio Hmuucwo Rho» to: omn.m~ own.m can.» own.m ome.ma oea.man mnusnsm omm.mmn oem owm.ean om~.6u omm.mau omm.nmn auflo Hauucoo vacuuoo oom.m~ ovm.a omm omm omv.ma owe.» «banana omm.vnu omm.an oom.mu oemu oo~.mu omo.o~n huge annuaoo QQIHO>0HO oom.>m~ oem.- omm.om oo>.ma omm.mm omN.~HH mnusnsm oom.mafln omm.m oem.m~u omm.¢ omm.o~u omn.¢mu muwo Hmuucmo oudowno 644.6HH oom.ma o~>.6 oom.na oem.e~ o-.6m mnuansm oom.~on oem.ou o-.mu oo~.~u om>.>a oms.mmu muwo Hmuucmo novoom HGHOB Houowm m00fi>me waH>me OGMHB OCwOSCOHm MGHG owansm umaomcoo Moosooum mooou amuwaomouumz H mwfluflo Hmuucmo omuomamm :fi mach uwHHoOI owma I obma .uouomm HmfluumoocH hm .mmcflm cannonom can msam mo amassz :H mmcmno .~.~ manna .mmcoco mmmucmoumm ouocoo mommsucmumn :H mommucmoumm "mag. .mmaH .uuuamox. .opmfi .madaum ozone sucsoo was ..uuna «coma .mfldeuw d s .um Ah 0 mean upo..uoflw on: 04 .:m o .m.oo¢ mod on: .momcoo o u bisoous" "mmousom In.¢uv A¢.smv Am.oov AH.HHnV A~.>env va _ omm.fisu ocm.mm om~.m¢ ovo.amu omo.¢vaa awnmaoouawnm_ Am.muv Am.sev Ao.Hmv Am.mauv Ae.oeuv Awe oeo.HoHn omn.owm omm.>n~ omH.HoHu oom.me¢n snow to: Ao.o~uv In.mnv A¢.mev A>.m~uv Ao.mmuv va oma.moan omn.- omn.mn 66¢.mmu oom.>oau uaouuoo Am.afiuv Ao.anc Am.nmc Ao.¢auv A~.menv Awe oo6.~eu omm.mfl oom.6~ o-.o~u 066.66: cauao>oao Am.ouv A».wmv Am.m¢V Am.mauv Am.H¢uv Awe _ oom.mmu oom.~HH omm.am omo.amu oo¢.HH~u omuuwno as.muv A¢.Hsv Am.~mv Am.w~uv As.mmuv Awe _ om~.mea om~.mm ooh.m~ omm.m¢u oo~.omu «canon. sauce mmmmmmuo muwaaou >Hco Hoonom nu“: suflo Hmuucmw” mmoHHOO waom Hoonom nmwm cmnu mmmg _ ommH I Chad .mumpaonOH no Hm>mq cowumooom hm .mQOH >uwunawuucou mo Hmnaoz :w mmcmso .m.~ OHQMB 29 Table 2.4. Percentage of Central-City Jobs, By Education Level of Jobholders, 1970 - 1980 Less than High School Some College Central City High School Graduate College Graduate Baltimore 1970 48.3 29.2 10.2 12.2 1980 29.6 32.3 19.4 18.6 Boston 1970 29.4 36.4 ' 16.8 17.5 1980 13.4 28.6 24.7 33.2 Chicago 1970 37.5 32.3 15.4 14.7 1980 23.4 28.2 23.8 24.7 Cleveland . 1970 35.4 38.0 13.0 13.6 1980 20.7 36.8 22.5 20.1 Detroit 1970 37.3 36.8 13.9 12.0 1980 21.1 32.8 25.8 20.3 New York 1970 35.8 33.1 12.7 18.4 1980 22.0 28.8 21.2 28.0 Philadelphia 1970 39.9 37.0 10.4 12.6 1980 23.2 36.3 18.4 22.0 St. Louis " 1970 43.4 33.0 11.0 12.5 1980 25.4 33.5 22.1 19.0 Washington D.C. 1970 22.7 31.9 17.7 27.8 1980 11.3 24.1 24.0 40.6 Souibézi aurazu’o?“£ne ééfiéis,viaéh n. Readable Pub- é‘uéé u criiata Sample File, 5% A Sample, 1980;1bid., 15% County Group Sample, 1970 (Kasarda, 1989) 30 transformation of the 1970's. Simply, the economic transformation of major cities from centers of production to centers of information processing and exchange has produced a marked decrease in the number of jobs typically filled with a high school education (Jenks, 1988). Ghetto residents are, on the whole, less educated than other inner-city residents, and it is this population that is most negatively affected by the increased competition for those traditional blue-collar jobs that remain (Kasarda, 1989; Wilson, 1987). Furthermore this same population is spatially confined to a limited labor market by below average automobile ownership (Kasarda, 1989), reducing again their overall competitiveness in the dispersed metropolitan economy. The rise in male joblessness associated with the changing nature of economic opportunity in the inner-city is directly linked to the rise in never—married parenthood (Testa, Astons, Kroyh & Neckerman, 1989; Wilson, 1987). Single-mother families themselves are more likely than other families to be poor, dependent on welfare, and urban. Marital disruption, and the lower socio-economic status associated with the condition, negatively impact the years of school completed by the involved child (McLanahan, 1985); contributing to further social isolation (McLanahan & Garfinkel, 1989), persistent poverty (Nichols-Casbolt, 1988), and the greater evidence of social disorganization as 31 found in crime, drug abuse, and unemployment. Causal relationships between poverty and family form have been hotly debated (Sullivan, 1989). On the one hand, the increase in female-headed families is considered an outgrowth of women's increasing economic independence and attractive alternatives to marriage as a means of support (McLanahan & Grafinkel, 1989; Farley, 1988). On the other hand, Testa et al. (1989) found a significant impact of male economic status on whether men will marry the mother of their child. Debate continues to rage on the relative impact of culture, economic parity between genders, welfare incentives, or a mixture between culture and economic factors. Social mobility for the ghetto poor is thwarted by a dearth of resources available to those impoverished. Economic resources have been explored. Underclass persons also possess lower volumes of social capital as indicated by the average number of social ties, and the value of those ties as measured by the social position of their partner, parents, siblings, best friends, etc (Kasarda, 1989; Auslander & Litwin, 1988). This is especially true of female-headed households (McLanahan & Garfinkel, 1989). Isolation remains key, as "poverty concentration has the effect of devaluing the social capital of those who live in its midst" (Waquant & Wilson, 1989, p24). 32 W The homeless have emerged as an important social problem in the 1980's. While not solely an urban phenomenon, they are especially evident in all major urban areas in this country. There is consensus that, while a growing problem, the nature of the homeless is also changing. Estimates of the size of the population range from 250,000 to 3 million (Rossi & Wright, 1989). No longer the stereotypical middle-aged, white, alcoholic male often associated with the skid row of the 1930's, the homeless are now characterized by a far greater diversity (Stefl, 1986; Wright, 1989). The new homeless of the 1980's are younger, more likely to be minority, and are often women and/or members of families. This great diversity itself has created problems for scholars and advocates as they try to address the needs of the population. Defining The Homeless The absence of a widely accepted definition of homelessness is the primary obstacle to developing credible data on the population (Rossi, 1989; Peroff, 1987). Definitions of homelessness range from highly inclusive characterizations covering all persons not adequately housed (Jahiel, 1987) to highly restrictive definitions focusing only on those individuals who lack a permanent and regular residence within a conventional dwelling unit (Rossi, 1989). 33 Choice of definition is problematic, greatly influencing both the scope and nature of the population to be studied. Definitial questions researchers must resolve include: 1) Are homeless people only those who live on the streets or who occupy overnight shelter, or does homelessness include those evicted or displaced and forced to stay with family or friends (Rossi, 1989; Johnson, 1989; Peroff, 1987); 2) When defining shelter, what is considered ‘regular' and ‘customary' access (Rossi, 1989; Rossi & Wright, 1989), and should conventional dwelling units include single room occupancy hotels, long-term detox centers, halfway houses, transitional housing, cars, trucks, tents, etc. (Rossi, 1989; Peroff, 1987); and 3) Should researchers focus on the number homeless at any given time, or consider the number of persons experiencing homelessness for any length of time (duration) in any given time frame (Caton, 1990; Johnson, 1989; Sosin & Westerfelt, 1989; Sosin, Colson & Grossmann, 1988; Peroff, 1987). Alice K. Johnson (1989) proposed that residential experiences, measured by the interaction of time (duration and/or repetitive nature of experience) and place (a continuum between substandard housing and literally homeless), be used as a frame—work for conceptualizing the severity of homelessness in terms of degree of depervation. Measuring homelessness as a variable rather than an absolute phenomenon, she hypothesized that "the shorter the 34 experience of homelessness and the further away the experience is from literal homelessness, the less severe it will be. Inversely, the longer the experience of homelessness and the closer it is to being literal, the more severe it will be" (Johnson, 1989, p3). Amount of time homeless and the relative hardship experienced by the participant have also been used to classify homelessness as either benign or malignant (Jahiel, 1987) or to make distinctions between street people, persons in shelters, and persons marginally housed (Stefl, 1987). Homelessness as a variable proves to be a difficult and often arbitrary measure (Rossi & Wright, 1989). Contemporary surveys of the homeless have tended to define the condition in more pragmatic terms. ‘Homelessness' usually refers only to people with nowhere to go other than the shelters provided by public and private agencies (Caton, 1990; Rossi, 1989) and is distinct from that subpopulation residing in physically inadequate housing or overcrowded conditions. Peter Rossi (1989), in what is considered one of the first attempts to apply modern social science methodology to the study of the homeless, found it useful to distinguish between (1) the literally homeless, persons who would be homeless by any conceivable definition of the term; and, (2) precariously, or marginally, housed persons with tenuous or very temporary claims to a conventional dwelling, and 35 running a high risk of becoming literally homeless for varied periods of time (Rossi, 1989; Rossi & Wright, 1989; Wright, 1989). While not solving all definitial problems, the distinction between literally homeless and marginally housed does reduce sampling frames and cost, and provides for an acceptable measure of the homeless population (Rossi, 1989). Rossi has been criticized for the low numbers produced by the narrow definition of the homeless (Appelbaum, 1987). ' To enlarge the population beyond the pragmatic definition used by Rossi would again produce those definitial ambiguities associated with the condition. In theory, there is agreement that contemporary homelessness lies along a continuum between persons not adequately housed and those literally homeless (Rossi & Wright, 1989). Definitial ambiguities still have led some authors to believe that it is impossible to categorically define homelessness within this continuum (Johnson, 1989; Wright, 1989). The rhetoric of definition is extremely important, as the number of homeless and existential conditions of the homeless depend on how narrowly or broadly the condition is defined. MM Attempts to count the homeless population are hampered by varied definitial boundaries (Caton, 1990; Peroff, 1987) which blur the nature and extent of the population. Once identified, the number homeless at any given time represents 36 only a small fraction of the total homeless population (Rossi, 1989; Peroff, 1987). Three strategies (Johnson, 1989) used to identify and enumerate the homeless are: 1) expert judgement or key-person surveys; 2) service utilization data; and 3) census or survey data. Inherent biases exist within each of the enumeration strategies. Key informant surveys are likely to overestimate or underestimate numbers depending upon the accountability constraints and care used in defining geographical boundaries. Estimates based on agency contact will only count that specific homeless subpopulation-which use social services. Sources of bias in survey data include: number of locations utilized, process by which locations are selected, and seasonal variations in the homeless population (Johnson, 1989). Recognizing those biases associated with each strategy, methods for enumeration should be selected based on the precision required and the projected use of the data. The conventional quantitative assumption that people can be enumerated within customary dwelling units (Rossi & Wright, 1989) does not hold valid when attempting to sample the homeless. Attempts to create a representative probability sample of the population must contact persons in nondwelling units and conduct interviews at times when the distinction between the homeless and the housed is at a maximum (Rossi, Wright, Fisher & Willis, 1987). Sampling 37 procedures must also recognize that this population is not readily identifiable (Rossi, 1989) and is often geographically concentrated (Rossi, 1989; McMurry, 1988). Knowledge of spatial distribution and a method for identification are important elements in any sampling exercise. Large data sets based on self-report or agency assessment (Wright & Weber, 1987), first generation prevalence studies utilizing detailed examinations of small samples (Johnson, 1989), and second generation prevalence studies utilizing larger sampling frames and standardized diagnostic measures (Piliavin, Sosin & Westerfelt, 1989), have also been used to design a sampling frame that would assign equal probability of selection. While methods of sampling the homeless have improved, equal probability of selection has not been achieved (Johnson, 1989). Selection remains biased simply because the outlying parameters of the population remain unknown. £222§§§ As the ranks of the homeless grew in the 1980's, so did the quantity and quality of research attempting to explain the phenomenon. The decade's first wave of literature on the homeless tended to fall into three categories (Wolch, 1990): government documents, advocacy reports, or collections of varying coverage. 38 Since the early 1980's a second wave of research emerged (Wolch, 1990) in sharp contrast to the decade's earlier works on the homeless. Homelessness was no longer being explained by use of monocausal explanations (e.g. homelessness as a product of deinstitutionalization vs. homelessness as a consequence of inadequate low—cost housing). There was also clear emphasis on the new homeless and their related characteristics, as opposed to the nature of earlier skid row residents. Contemporary authors (Caton, 1990; Rossi, 1989; Wright, 1989; Ropers, 1988) principally focus on reductions in low-income housing, structural changes in the economy, and a decline in welfare entitlements when explaining the growth in today's homelessness. Reductions in the supply of low-income pousing Homelessness is directly and indirectly related (Rossi, 1989) to the shortage of inexpensive housing which accelerated in the 1980's. Directly, reductions in low-cost housing effectively place shelter out of the reach of the majority of homeless. Using 40 percent of the poverty line ($265.00 in 1983) as a conservative measure of maximum affordable gross monthly rent (Wright, 1989), reveals a 30 percent decline in affordable housing between 1978 and 1983 (Rossi, 1989; Wright, 1989; Wright & Lam, 1987). The decline in single room units - normally the last resort for the extremely poor - is even more precipitous. Single room 39 occupancy hotels (SRO's) declined by approximately one million units between 1970 and 1980; almost 40 percent of the total units available. At the same time that cities were recording a roughly 30 percent decrease in affordable housing, the number of persons living below the poverty line grew by 36 percent (Caton, 1990; Wright, 1989). Increases in poverty, with concurrent decreases in available low-cost housing, predestines an increase in the number without shelter (Wright, 1989; Carliner, 1987). Inflation in consumer prices for all commodities (Caton, 1990; Wright, 1989) factors heavily in the decline of low-cost housing. The trend in total number of units nationwide renting for $80 or less can be used to illustrate the effects of inflation (Wright, 1989). In 1970, there were some 5.5 million units renting for $80 or less. By 1983, the number of units had fallen to 650,000. Rent-to-income ratios can also be used as an indicator of affordable housing. Theoretically, one-fourth of one's annual income is considered the maximum average a household should pay for rent. Reality was quite different for the over seven million people in 1980 paying over 50 percent of their income in rent (Caton, 1990). Over half of the very poor in 1980 (earning less than $3000 per year), were paying over 70 percent (Caton, 1990) of their annual income on shelter. Single room occupancy hotel rooms renting in 40 Chicago for $195 per month in 1984 ($240 per month if rented by the day) (Rossi, 1989), represented over 35 percent of a $5,000 dollar annual income. The monthly median income of homeless persons in Chicago (Rossi, 1989) was only $100 dollars. Available government support is simply not enough to make up the difference (Carliner, 1987). There has also been an absolute loss in low-income housing as urban areas are gentrified or re-zoned for more profitable use (Caton, 1990; Rossi, 1989; Wright, 1989). Between 1974 and 1979, the net loss in low-income rental units averaged 360,000 units per year (Wright & Lam, 1987). Shifts in net migration flows (Carliner, 1987) and the increase in the numbers of non-family and childless households, further reduced the available low-cost housing. While these housing units were being systematically reduced from the nation's low-cost housing stock, nowhere near enough units were being built to replace the losses. The indirect effects of housing shifts on the homeless are felt through the increased financial burdens (Rossi, 1989; Wright & Lam, 1987) placed on poor families. A tight housing market will allow for larger units to command disproportionately higher prices, forcing many poor families to move to smaller dwellings. This market will also increase chances of homeless participation among the very poor, while making it harder to double-up in an effort to retain shelter (Caton, 1990; Rossi, 1989). The smaller 41 housing units occupied by the poor will either: 1) spatially restrict a family's ability to house adult children (Rossi, 1989; Wright & Lam, 1987), or 2) strain personal relations (Rossi, 1989; Lee, 1987) to effectively limit an ability to provide prolonged shelter. Sppugtpral changes in the egonomy The homeless today are better educated, younger, and more heavily dominated by racial and ethnic minorities than in years past (Rossi, 1989; Wright, 1989). The economic processes that helped create the new homeless are the same contributing to urban underclass formation. Decentralization of industrial plants, the flight of industry out of the inner-city, and the shift of metropolitan economies from manufacturing to advanced service industries and occupations (Waquant & Wilson, 1989) have created sharp increases in urban poverty (Wright, 1989; Wilson, 1987) whose burden fell primarily on young men and minorities (Rossi, 1989; Wilson, 1987). In 1980, the percentage of people living below the poverty level reached 13 percent, the highest figure recorded since 1969. The poverty rate continued to climb reaching 15.2 percent by 1983 (Wright & Lam, 1987). Structural economic changes also transformed the nature and availability of occupations (Kasarda, 1989), increasing competition between high school educated persons for those blue-collar jobs that remained. This competition led 42 directly to an increase in the average education among the homeless in the 1980's. Job histories of the literally homeless, especially minority participants, suggest that they have been among the extremely poor for years (Rossi & Wright, 1989; First, Roth & Arewa, 1988). As in the formation of underclass poverty, the impoverished state of young males had a direct effect on household formation (Rossi, 1989; Burt & Cohen, 1989; Wilson, 1987). Homeless women are rapidly increasing in representation. Of this growing subpopulation, only about half of homeless single women and women with children have ever been married. A majority of these urban homeless women, and almost all of the homeless families are black (Rossi, 1980; Burt & Cohen, 1989; Sullivan & Damrosch, 1986). The uncertain labor market situations of black men have endangered black family formation (Rossi, 1989; Wilson, 1987) and exposed young black women with children to periodic bouts of homelessness. Economic changes, and homeless participation itself, greatly effected the kin networks - the social capital - available to individuals in time of crisis (Rossi, 1989; Lee, 1987). From the Chicago data, Peter Rossi (1989) concluded the average life of tolerance by family social networks of dependent adult children to be about four years; the period that the homeless were without employment before becoming homeless. The socioeconomic condition of the 43 social network (Rossi, 1989) and the disruptive life events, age, and mental health of the homeless individual in question (Rossi, 1989; Lee, 1987) influenced the degree of tolerance available to the individual. The poor also had considerably smaller social networks available than the near poor (Auslander & Litwin, 1988; Piliavin & Sosin, 1987). Findings concerning frequency of contact (Auslander & Litwin, 1988) with network ties closely parallel network availability. Distance, above both race and socioeconomic status, was the deciding factor in kinship involvement (Mennen, 1988). Finally, economic situation affected the degree of criminality among the homeless. The homeless have a higher overall arrest rate than the housed population, although a majority of their offenses are for public intoxication, disturbing the peace, burglary, and other non-violent crimes (Snow, Baker & Anderson, 1989; Fisher, 1988). A vast majority of crimes are committed by persons between 17 and 34 years of age. Most arrests were composed of people who were either: 1) breaking into buildings in search of shelter, or 2) shoplifting (usually food and/or cigarettes). A relationship between amount of time homeless and extent of criminality have led Snow et a1. (1989) to believe that such activity represents "a gradual expression of one's adaptive repertoire to include various criminal activity" (Snow et al., 1989, p542). 44 'ons ' W are nt't emen 5 Changing economic conditions and the loss of low income housing are exacerbated by the decline in the economic positions of poor families, as expressed by the continued inability of means tested income transfers to move people out of poverty. Similarly, the declining level of benefits directly affects the capacity of poor individuals to take care of themselves apart from parents (Rossi, 1989). In an era of huge budget deficits, there is little support for expanded initiatives to combat poverty. Where states have programs to support single unattached persons, none provide over $4,000 per year (Rossi, 1989). Many states do not even have programs to support this segment of the population. Of the three kinds of programs established by the United States to reduce poverty - means tested income transfers, social insurance, and targeted education programs (Burtless, 1986) - only social insurance has expanded repeatedly since its inception. Means tested transfers, the program most likely used by working-aged adults in need of assistance, have actually declined by 20 percent since their peak in 1976 (Burtless, 1986). Total real spending on cash assistance has also decreased by 15% since the mid-1970's. Table 2.5 details the average dollar values (in constant 1985 dollars) of several national income transfer programs active between 1968 and 1985. 45 .uOumAumo a mo pom: xmoca m>wum HOESmcoo «onwaaooxmmma Ca c3onmumuoz ..amma .Amuomc "mousom mmn mwn ¢m¢ omm uamd How mmm Nam bmm huwawnmmflo MOM Hmm vhw mum ¢mm mwm onwam man How Hmm ¢oa boa «ma ham comm may now OEOOCH Houcwamammom mnv new man mcmcmuo\mzoow3 aufiuaomm Hmfloom vmv now «me wuwawnmmwm xuwuoomm Hmfloom anew 644m ¢H¢m mmmm acmemufluou mam 6H0 sufiusomm ammoom mama omma mums moma amumoum “mumcmna mucwshmm womam>< wanucoz Hmcowumz mama I mead .mucmaamm Houwcmua aanucoz womam>¢ umHHoolusmumcou .m.m manna 46 Major declines in constant-dollar values were recorded for both AFDC and General Assistance. Reductions in AFDC increased the possibility of homelessness for single-parent families (Rossi, 1989), while cuts of over 50% in General Assistance payments greatly reduced the possibility of independent living on the part of clients. Mgptal Health If the origins of homeless are in extreme poverty, resulting from reduced blue-collar job opportunity, a decline in the supply of low income housing, and reductions in welfare entitlements, what distinguishes the homeless from the millions of people in poverty who have managed to remain domiciled? Rossi (1989) compared the housed and homeless in extreme poverty (general relief clients and AFDC recipients vs. the homeless) and determined the homeless more vulnerable because of disabilities which reduce their ability to compete in housing or labor markets. Disabilities included: physical and mental handicaps, time in prison or jail, depression, and substance abuse problems. The extensive evidence presented of various forms of disability create the impression that the homeless are on the street because of their own personal inadequacies (Wolch, 1990). Many of the disabilities are themselves related to conditions of extreme poverty (Wolch, 1990; Wilson, 1987). Of those least likely to be a product of poverty, physical and mental handicaps, mental health, and 47 the condition of related services, are most often mentioned as the causes of today's homeless (Bean, Stefl & Howe, 1987; Snow, Baker, Anderson & Martin, 1986). Significantly, Snow et a1. (1986) found only 10 to 15 percent of the homeless population to be mentally ill. Using a sample of health clinic participants, Wright (1988) conceded that the mentally ill do not represent a majority of the homeless, but contended that estimates of around 33 percent would be more realistic for the homeless population. Piliavin et a1. (1989) examined the argument made by both authors and, noting the effects of choice based sampling on the representativeness of Wright's sample, concluded in favor of the lower percentages. Mental illness may increase the possibility of homeless participation but only for a small proportion of the homeless. u ar A poorly understood mixture of cultural, social, demographic, and economic factors have contributed to the emergence of a socially and economically isolated ghetto underclass in our nation's cities. This population is most likely to be composed of minorities, and is characterized by lower levels of education, high levels of unemployment, female-headed households, crime, and welfare dependency. At the same time, a new homeless has also emerged, and has been linked to those same economic processes which have defined extreme poverty. But not all persons in extreme poverty are 48 underclass, and the homeless represent a more heterogeneous population than that found in ghetto neighborhoods. None of the studies examined have explored relationships between underclass poverty and homelessness. The following chapter outlines the method employed by this study to evaluate possible relationships between underclass participation and severity of homeless experience. CHAPTER III METHODOLOGY The study of homelessness and the urban underclass was conducted in Benton Harbor and Grand Rapids, Michigan. The cities were selected because each represent a different - polar -response to the macro structural economic changes that started in state during the 1960's and accelerated between 1970 and 1980. Benton Harbor Once a flourishing regional industrial manufacturing center, Benton Harbor has come to exemplify the ultimate and most devastating end that could await small industrial cities as they confront the process of deindustrialization (Hurst & Lang, 1987). In less than twenty years, demographic and deindustrial transformations have left the city a virtual ghost town; economically and socially isolated from the rest of the region. Benton Harbor is located in the far southwest region of Michigan, ninety miles southeast of Chicago. The St. Joseph river forms the city's east-west boundary and separates Benton Harbor from its "twin city" of St. Joseph. To the west of the river, and adjoining the south city limit, lies 49 50 St. Joseph Township. To the east of the river, and adjoining the east city limit of Benton Harbor, lies Benton Township. Tables 3.1 and 3.2 illustrate the demographic changes taking place in the region between 1960 and 1980. Table 3.1. Benton Harbor Greater Area Population Statistics 1960 tO 1980 YEAR BENTON BENTON SAINT ST. JOSEPH BERRIEN HARBOR TOWNSHIP JOSEPH TOWNSHIP COUNTY 1960 19,136 19,914 11,755 7,418 149,865 1970 16,481 19,034 11,042 9,605 163,940 1980 14,707 19,120 9,622 9,219 .l7l,276 1980 Black Population Number 12,693 7,507 171 305 24,817 ‘ Percent 86.3 39,3 1.8 3.3 14.5 E a Source: 1960 - 1980 0.8. Census Bureau (Hurst & Lang, 1987) Table 3.2. Benton Harbor Greater Area Black Population, 1960-80 YEAR BENTON HARBOR SAINT JOSEPH BERRIAN COUNTY PERCENT NUMBER PERCENT NUMBER PERCENT NUMBER 1960 25.2 4,817 0.6 68 6.9 10,374 1970 58.8 9,687 1.3 140 11.2 18,283 1980_ 86:3__UH12,693_W_‘ 1.8 171 14.5 24,817 , Southwest Michigan Commission, 1985 (Hurst & Lang, 1987). The Benton Harbor/St. Joseph population data present a clear trend of racial segregation and social isolation. The 51 river is the demarcation line (Hurst & Lang, 1987), with the aggregated black population of Benton Harbor and Benton Township accounting for 81.4 percent of the county's entire black population. Macro - national - economic shifts in the 1970's created local economic transformation which conspired to make this socially isolated population the poorest in the region. Tables 3.3 and 3.4 show that between 1966 and 1986, the city's industrial core had a net loss of 23.7 percent of it's labor force and a net decline of 22.6 percent in the number of manufacturing firms (Hurst & Lang, 1987). Table 3.3. Number of Manufacturing Establishments and Employees, Benton Harbor and St. Joseph, 1966 - 1986 BENTON HARBOR SAINT JOSEPH TOTAL AREA YEAR EMPLOYEES FIRMS EMPLOYEES FIRMS EMPLOYEES FIRMS 1966 7,975 119 9,069 40 17,044 159 1971 9,473 117 9,499 37 18,972 154 1976 6,993 99 9,646 40 16,639 139 1981 6,000 95 8,235 35 14,235 130 1986 5,472 88 7,529 35 13,001 123 1966-86. Source: Harr s D rectory of Mich gan Manufactur ng F rms, (Hurst & Lang, 1987) 52 Table 3.4. Percent Net Gain/Loss of Manufacturing Establishments and Employees, Benton Harbor and Saint Joseph, 1966-1986 BENTON HARBOR SAINT JOSEPH EMPLOYEES FIRMS EMPLOYEES FIRMS EMPLOYEES FIRMS GAINED GAINED GAINED GAINED GAINED GAINED = 1966-71 18.8 -7.5 11.3 1 1971-76 1976-81 1981-86 1966-86 Source: Harris Directory of Michigan Manufacturing F rms, 19.6-86. (Hurst & Lang, 1987) The four principal industrial sectors of transportation, electronic, machinery, and primary metals suffered the greatest loss in numbers of firms and employees. In 1985, Benton Harbor had an unemployment rate of almost 45 percent. Employment trends continued to decline as two major manufactures - Clark Equipment Company and Whirlpool - either closed or divested its manufacturing interests in the region. Grand Rapids The city of Grand Rapids is located in Kent County, seventy miles west of Lansing, Michigan. Kent County has responded well to the changing economic conditions of the 1970's and 1980's. Overall, between 1969 and 1986, the County experienced consistent growth (as measured in number of employees and total number of establishments) in areas of 53 finance, sales (retail and wholesale), manufacturing, and service. Table 3.5 outlines the growth experienced within the various occupations. A majority of the economic growth within the city occurred in finance and service related occupations. While manufacturing increased in the County, opportunity plummeted within the city (see Table 3.6) as the economic transformation to a center of administration, information, and advanced service provision was attempted. The burden of the city's transformation from a center of industry to an economy based on information and service provision fell on families who had traditionally relied on blue-collar employment for their lively-hood. Between 1970 and 1980, the number of families living below the poverty line in Grand Rapids increased by 10 percent. Furthermore the number of families living in high poverty areas (with a poverty rate of at least 30%) increased by 76 percent. Extreme poverty areas within the city (with a poverty rate of at least 40%) recorded a staggering 178 percent increase. Of the families living in underclass poverty tracts, 74.6 percent were headed by single women. A majority of the households were black. 54 oucuso ommusmoumm ouosmo mononusouum ca noaoucoouoa «muoz cowumHsmom 0:» mo usmcmo .m.o “mousom nunuununnnnnnu Aa.o~+v Am.ma+v mood «so.~s mew mmn.om oszoaommm meuwm mwmaonsm meuflm mmmhonsm mauflm mmmaonam auumsosH oomucmouom omma moma omumwma .mucsoo ucmx .>uumsosH an mucofinmwanmumm can mmohonsm .m.m OHQMB 55 Table 3.6. Statistics by Selected Industry Group, Grand Rapids, 1967 - 1982 NUMBER OF FIRMS INDUSTRY (Type) 1967 1982 %CHANGE Food and Kindred Prod. 256 21 -(91.8) Lumber/Wood Fixtures 39 25 -(35.9) Furniture/Fixtures 68 48 -(29.4) Paper/Allied Products 42 11 -(73.8) Printing/Publishing 421 101 -(76.0) Chemicals/Allied Products 130 14 -(89.2) Rubber/Plastics 53 15 -(71.7) Stone/Clay/Glass Prod. 81 0 -(100.0) Primary Metals 147 21 -(85.7) Fabricated Metal Prod. 456 70 -(84.7) Machinery (not electrical) 690 87 -(87.4) Electrical Equipment 99 11 -(88.9) Transportation Equip. 101 0 -(100.0) Instruments/Related Prod. 35 0 -(100.0) Misc. Manufacturing 111 12 -(89.2) =1 { Source: U.S. Census Bureau The purpose of this study is to determine if ghetto underclass participation prior to homelessness influences the severity of the homeless experience, as measured by time homeless (duration and/or repetitive nature of the experience), the nature of the resources utilized, and the hopefulness of current life condition. 56 The following hypotheses were tested for the combined data from the two cities: 1) 2) 3) There is a relationship between type of homelessness and socioeconomic background, with the underclass more likely to be chronically homeless and the middle- and lower-classes more likely to exit the homeless population prior to the chronic state; Homeless people from an underclass background will utilize a narrower range of services and will have a less hopeful perception of life condition; As the length of time of the current episode of homelessness increases for an individual, their range of services utilized will decrease, and their personal perception of their homeless condition will become less hopeful. Services for the homeless are categorized as primary, intermediate, or ultimate depending on the amount of mobility out of the homeless condition offered the client. Service use is further classified into five categories — health, financial, employment, transportation, and food and shelter. Table 3.7. A key to service organization is presented in In this study it was impossible to use conventional random sampling techniques because there are no directories of homeless from which to draw a sample - they cannot be Table 3.7. 57 Key to Conceptual Service Organization II. III. IV. Health A. PRIMARY (Shower, Hospital Emergency) B. INTERMEDIATE (Health Clinic, Counseling) C. ULTIMATE (Nutrition, Hospital/Dental Care) Financial A. PRIMARY (Direct Cash Handouts) B. INTERMEDIATE (Budgeting, Public Assistance) C. ULTIMATE (Credit Counseling, Loans) Employment A. PRIMARY (Mail Reception Service) B. INTERMEDIATE (Telephone, Hiring Hall) C. ULTIMATE (Training, Placement, Child Care) Transportation A. PRIMARY (Bus Passes, Dial-A-Ride) B. INTERMEDIATE (Referrals, Car Pooling) C. ULTIMATE (Car Loan, Repair Service) Food and Shelter A. PRIMARY (Soup Kitchen, Emergency Shelter) B. INTERMEDIATE (Food Stamps, SRO, Cooking Facilities) C. ULTIMATE (Transitional Housing, Housing Placement) 58 located in conventional dwelling units. Various methodologies have been developed in attempts to ensure equal probability of selection among the homeless (Johnson, 1989). A second generation prevalence study, utilizing a random sample of persons in and around agencies located in areas of known homeless concentration, is used to achieve a representative sample of homeless persons in Benton Harbor and Grand Rapids, Michigan. A person is defined as homeless if that person lacks 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 to accept a traditional residential setting for other reasons. The operational definition of homelessness covers the literally homeless and, to the degree that they cannot regularly sleep and receive mail at their present location, those persons doubling up or living in substandard quarters. Homeless individuals are also grouped into three categories: chronic, episodic, and transitional (Michigan Task Force on the Homeless, 1986). The chronic homeless are those persons who have been homeless for more than twelve months. Episodic homeless have been homeless for less than twelve months, but have at least one prior episode of homelessness. The transitional homeless have been homeless for less than twelve months and have no prior history of 59 homelessness. This study did not try to enumerate the homeless population in either city. In Grand Rapids, blocks were first selected based on an expected concentration of homeless people. This stratification was accomplished with the cooperation of the Grand Rapids Police Department. Because service agencies are located throughout the community, Benton Harbor was treated as one single area. Only one area of homeless concentration was identified within Benton Harbor. Interviews were conducted in all agencies serving the homeless in Benton Harbor. In Grand Rapids, a purposive sample was drawn from all primary, intermediate, and ultimate service agencies located in the city. Interviewees were drawn systematically from all persons present at the various agencies. Street interviews were conducted within those same blocks selected for the agency survey. Cooperating respondents were paid $10.00. A total of 200 persons were interviewed - 100 in each city - between June and October, 1990. W All persons encountered in and around the various service agencies were not assumed to be homeless by the operational definition. To determine homelessness, a preliminary questionnaire was employed prior to the administration of the survey questionnaire. Individuals 60 determined to be homeless by the preliminary questionnaire were then interviewed to obtain data on their employment and residence histories, service use, and socio-demographic and socio-economic characteristics. The survey questionnaire was divided into five sections (see appendix II) - childhood, situation just prior to homelessness (prior situation), current situation, use of services (services), and future. A number of questions, across all five sections of-the questionnaire, were asked in order to determine prior underclass participation. Urban underclass participation was determined by: 1) The extent to which social dislocations (e.g. high school dropouts, single-parent families, crime, etc.) were identified within the neighborhood of their childhood and/or the neighborhood in which they lived just prior to their homeless experience; and 2) Personal and family employment history, their present degree of available social capital, their educational history, present and childhood family structure, and evidence of substance abuse and/or criminal activity. Questions asked about current homeless situation addressed their reasons for homelessness, their length of time homeless, any history of homelessness, their marital status, and the number and ages of dependent children. Data on use of services was obtained for 36 general types of services organized into the five categories mentioned earlier. 61 With respect to the future, hopefulness was determined by whether the respondent believed that 1) they would need a lot of help to turn their life around, 2) their current situation was only a temporary setback in their life, 3) they could overcome their current situation, 4) they would be homeless for much of the rest of their life, 5) in a year from now they would be much better off, 6) it would take a miracle to overcome their current situation, 7) their present condition would get much better in the future,~and 8) they better get used to their current situation because that was the way it was going to be. The results obtained through use of the questionnaire are presented in chapter IV. CHAPTER IV RESULTS The Characterispics of phe Hemeless Sample in Behpoh nepbop ehd Grand Rapids. Michigan Because the survey of homeless in Benton Harbor and Grand Rapids, Michigan was not an attempt to census the population, or to provide for equal possibility of assignment throughout the community, limitations must be recognized in attempts to generalize for the whole of the condition. This sample effectively includes only service users in either community. Nevertheless, the sample drawn does closely resemble that obtained by Rossi et al in their probability sample of the homeless in Chicago. Homelessness has been described as a predominantly male condition. Three out of five (59%) of the 200 persons interviewed in Benton Harbor and Grand Rapids were men. Blacks constituted a disproportionate share of the homeless (60%), with whites and Hispanics proportionately underrepresented when compared to housed population statistics. While over 60 percent of the sample had children, the burden of care for dependent children fell 62 63 primarily on single mothers. The average age of the homeless sample was 33.9 years. Sixty percent of the sample had never married. The typical homeless individual was also a high school graduate. While informative, homelessness described solely in terms of average population characteristics can obscure the heterogeneous nature of the population. It is important to remember that forty percent of the subjects were female, a majority with full-time care responsibilities for dependent children. Also significant was a minority of young black women (22% of the homeless), typically homeless with young children. Older males were usually white and often chronically homeless. Important demographic characteristics are outlined in Appendix I. All of the homeless interviewed described a situation of extreme poverty. The average length of time homeless was between 2 and 4 months. Twenty-six percent (26.5%) of the sample would be considered chronically homeless (Table 4.1) by definition; their time homeless greater than one year. Forty-seven percent of the respondents had been homeless an average of 1.94 times prior to their present experience. Over seventy percent had been forced to stay with family or friends an average 2.8 times since adulthood. Over forty percent had stayed with family or friends just prior to their present homeless experience. Interestingly, while homeless by definition, 27 percent did not consider 64 Table 4.1. Transitional, Episodic, and Chronic Homeless In Benton Harbor and Grand Rapids ; Cum i o ss e Freguehey Repeepp Eereept : TRANSITIONAL 77* 38.5 38.5 g EPISODIC 70 35.0 73.5 ; CHRONIC 53 26.5 100.0 ' TOTAL 200 100.0 Valid Cases 200 Missing Cases 0 * Mode themselves homeless at the time of the interview. Many of those interviewed had experienced extreme poverty for a number of years prior to their present homeless condition. Persistent poverty is reflected in Tables 4.2 and 4.3. Thirty-seven percent characterized their adult life as continued experiences with poverty and unemployment. Thirty-six percent had grown up in neighborhoods described as predominately poor. Forty-four percent had lived in high poverty neighborhoods just prior to their bout with homelessness. As can be expected, poverty in childhood was significantly related (.001) to reported poverty during the two years prior to homelessness. A majority of those growing up in high poverty neighborhoods had only one parent - usually the mother - for most of their youth. 65 Table 4.2. Reported Work Histories for Homeless in Benton Harbor and Grand Rapids Last Time Steadily Valid Cum Employed Frequency Percent Percent PRESENTLY WORKING 28 14.1 14.1 1-3 MONTHS AGO 35 17.6 31.7 3-6 MONTHS AGO 35 17.6 49.2 6-9 MONTHS AGO 20 10.1 59.3 9-12 MONTHS AGO 8 4.0 63.3 OVER 1 YEAR AGO 73* 36.7 100.0 TOTAL 200 100.0 Valid Cases 199 Missing Cases 1 *Mode Social isolation accompanied homelessness in both cities. Over sixty percent of those homeless never married; of those ever married, most were either separated or divorced. While a majority had surviving family members, only 17 percent thought there were many family members or friends to whom they could turn to in a time of crisis (Table 4.4). Over fifty percent (64.5%) had also been forced to move at least two times within the two years preceding their homelessness. The lack of social networks contributes to the continued vulnerability of homeless persons to changes in employment, income, and/or health. 66 Table 4.3. Reasons for Homelessness in Benton Harbor and Grand Rapids Reason for Homelessness Frequency Percent EVICTION 51 25.6 PERSONAL CONFLICT ' 72 36.2 FAMILY CONFLICT 59 29.6 SPOUSE ABUSE 16 8.0 ECONOMIC REASONS 91 45.7 BUILDING CONDEMNED 11 5.5 FIRE 9 4.5 RELEASED FROM PRISON 15 . 7.5 RELEASED FROM HOSPITAL 13 6.5 P OVERCROWDING 19 9.5 DIVORCE 10 5.0 NEW TO AREA 42 21.1 OTHER 21 10.6 Respondents could check more than one reason for homelessness. Column percentages will not add to 100.0% 67 Table 4.4. Perceived Availability of Family and Friends in Time of Personal Crisis erceived P Valid Cum Level of Support Frequency Percent Percent 'ALMOST NO ONE TO TURN TO' 82 41.0 41.0 ' 'SOME FAMILY AND/OR FRIENDS' 84 42.0 83.0 'MANY FAMILY AND/OR FRIENDS' 34 17.0 100.0 TOTAL 200 100.0 u_lid Cases===320 Missing Cases 0 I 7 Testing The Hypotheses 1. There is a relationship between type of homeless and socioeconomic background, with the underclass more likely to be chronically homeless and the middle- and lower - classes more likely to exit the homeless population prior to the chronic state. A crosstabulation of type homeless (i.e. Transitional, Episodic, and Chronic) by prior socioeconomic status (Table 4.5) shows a significant relationship between the two variables. A majority of the chronically homeless come from either underclass or lower-class backgrounds. Sixty-eight percent of those homeless persons identified as underclass were found to be chronically homeless. Homelessness for persons identified as lower - or lower-middle class tended to be episodic in nature. Forty-three percent of those 68 Table 4.5. Crosstabulation of Type Homeless by Prior Socioeconomic Status COUNT UNDER- LOWER- LOWER- MIDDLE- “ COL % CLASS CLASS MIDDLE CLASS ROW CLASS TOTAL TRANS- 1 35 32 7 75 ITIONAL 5.3 39.8 43.8 43.8 38.3 EPISODIC 5 33 24 4 66 26.3 37.5 32.9 25.0 33.7 CHRONIC 13 20 17 5 55 68.4 22.7 23.3 31.3 28.1 COLUMN 19 88 73 16 196 TOTAL 9.7 44.9 37.2 8.2 100.0 ‘ == CELLS WITH CHI-SQUARE D.F. SIGNIFICANCE MIN E.F. E.F.<5 19.80948 06 .0030 4.490 1 OF 12 69 homeless identified as middle-class were transitional. As expected, a majority of the homeless had lived in poverty areas prior to their homeless experience. The relationship between type Of homeless and socioeconomic status (SES) can also be explained by examining the effect of SES on the amount of time homeless. Simple regression analysis (Table 4.6) shows a significant relationship between prior SES and the length of time homeless. While significant, only three percent of the variance in the amount of time homeless is explained by knowledge of a person's socioeconomic background. Because underclass poverty is often described in intergenerational terms, type of homeless was also compared with a measure Of child socioeconomic status that resembled the prior SES scale. Table 4.7 reveals that child socioeconomic status (the socioeconomic status of the person's household while growing up) is not significantly related to type homeless. When examining the possible relationships between child SES and type Of homeless, it is important to note the age of the homeless sample. The average homeless person in the sample was 33.9 years Old. The economic transformations which affected both Benton Harbor and Grand Rapids would have had their greatest impact at the time this population was entering the workforce, not while they were an adolescent and dependent upon their family 70 Table 4.6. Regression Analysis Of Time Homeless by Prior Socioeconomic Status Oijrr‘ O y: ’ O ’ G’ S _ON 0 . ad TIME PRIOR HOMELESS SES TIME 1.000 HOMELESS PRIOR -.167 1.000 SES Prediction of Length of Time Homeless by Prior Socioeconomic Status Standardized Significance Predictor Regression T-Value Level Variable Coefficient Prior SES -.50147 -2.359 .0193 * R Square: .02790 Dependent variable: length of time homeless. * Significant at 0.05 level. Table 4.7. Crosstabulation of Type Homeless by Child Socioeconomic Status COUNT UNDER- LOWER- LOWER- MIDDLE- COL % CLASS CLASS MIDDLE CLASS ROW CLASS TOTAL TRANS- 5 13 22 29 69 ITIONAL 7.2 18.8 31.9 42.0 .36.9 EPISODIC 7 20 14 23 64 10.9 31.3 21.9 35.9 34.2 CHRONIC 3 14 21 16 54 5.6 25.9 38.9 29.6 28.9 COLUMN 15 47 57 68 187 TOTAL 8.0 25.1 30.5 36.4 100.0 E CELLS WITH CHI-SQUARE D.F. SIGNIFICANCE MIN E.F. E.F.< 5 7.33041 06 .2914 4.332 1 OF 12 72 for support. The blue-collar employment in both cities offered Opportunity to parents - even poor parents - throughout the 1950's and 1960's. 2. Homeless people from an underclass background will utilize a narrower range Of services and will have a less hopeful perception of their life condition. §epvice Use The analysis of variance (ANOVA) test (Table 4.8) shows no significant relationship between prior socioeconomic status and the total number of services used while hOmeless. Tests were also conducted to specifically examine possible relationships between prior SES and primary, intermediate, and ultimate service types. Analysis of variance (ANOVA) tests for primary service use by prior SES (Table 4.9), intermediate service use by prior SES (Table 4.10), and ultimate service use by prior SES (Table 4.11) show no significant relationships between the variables. Table 4.12 breaks down service use by prior socioeconomic status and sex of respondent. Overall, the total number of services used while homeless by sex is consistent across the SES scale; ranging from 10.9 to 15.9 (the one female middle-class respondent not included in the analysis). Examination of the percentage use of primary, intermediate, and ultimate services does suggest some 73 .mes mmow>umm no Hones: Heuou ”manmfiue> useccommo cmw. owm. Hmo.HH th.nm no mmm HOHHA mmH.Nv omb.wmmm mmd HMHOB www.mc www.mmam mad Houum coo. cmm. CON. Hmo.HH www.mm no H0602 mumsvm wuflaflncnoum msHm> mumsvm mmucsvm EOOOORE cofiucwum> m m m COOS no How mo memumeo mo mousom msueum oflaocoom uofium an Como mmow>umm mo nonssz Ca mmocmueumao now m>o=< .w.v OHQMB 74 .omm: mmofi>u0m wumsfiuo mo pcmouom "OHQOwHO> unmocmomo new. wmm. mac. wmo. no mam MOMMA cmo. mam.m «ma HOUOB mmo. owm.m HwH HOHHm woo. new. «mm. mHo. mmo. no H060: mucoom xuaawnenonm mon> mumovm mmumsvm Eoooouh sofiumflum> m m m one: no saw no moonoeo mo mousom msucum oflaocoomofloom Howum an on: OOH>H0m wnmswwm a“ mmocmumuufio now e>oc¢ .m.¢ OHDMB 75 .Omms mOOfi>wom ouewOOBHOusfl no uceonmo ”manmwum> unwocmomo who. Obo. moo. moo. no mam HOfiHm who. cmm.w «ma HvaB wno. ocm.o Hmd HOHHW Hoo. Ohm. one. moo. moo. no H060: ORODOm huflafinwnoum OOHO> Ohmsvm monsoon Soommnm cofiumflue> m m m one: no saw no mmmummo mo mouoom msucum Oflaocoomowoom HOMHQ an own oofl>umm oumfiomaumusH Ca mmocmumuuflo Hon 0>oc¢ .OH.¢ OHnMB 76 .Omo OOH>HOm eumawuas mo useouoo "maamwne> ucwocmooo mmm. mom. :5. NnH. no mum “~0me cco. th.h ah." HMHOB coo. 0235 mud HOMHH hHo. mom. wma. «co. NMH. no HOUOS mumsvm auwaflnmnoum OOHO> mumovm mmumsvm aooewum cowucwhm> m m m new: no Bow no mmenomo mo mouoom moumum OflEOCOOOOHOOm Mofium ha 0m: OOH>uOm OUOEwuHD a“ mwOCOHOMMHo Rom m>oc4 .HH.¢ OHOOB 77 .meoa>umm mandadm>< no on: moeuceoumm euoceo mmmmcucouem cw newsman «ouoz 16.6 V Aa.n~o xo.mav Am.~n. AH.6HV Ao.mmv .¢.H~V .~.~Hv mumsfluao A¢.mav .m.¢¢v Am.mmv A~.6¢v As.snv AO.N¢V 15.nmv .¢.smv OunmcmsumucH AH.HHV Am.amv Ao.mvo A¢.Hoo Am.omv Am.mmv A~.>vv Am.smv mumsfium o.m m.ma m.Ha m.ma s.HH m.ma w.oH m.HH 60m: umnssz «seeds was: .wdmswm 0H6: («Adams mam: («Adams mam: cummlmmmwmmmw mmmqo mqooHs moooHanmzoo mmaqo mmzoo mmmqwmmmmml moucum ofiSOCOOOOHOOm HOAMO >9 mm: mow>nmm uo czooxmmum .NH.¢ dance 78 expected patterns. First, primary service use was consistent for both sexes across the socioeconomic scale. Primary agencies remain a principle entry point into the homeless service network. Second, percentage use declined from primary to intermediate to ultimate services, revealing a heavy concentration of homeless in services that Offer little social or economic mobility to the client. Third, while men consistently reported greater use of primary and intermediate services, women showed consistent use of all three categories of service; possibly a reflection of the nature and availability Of services designed for women and families of women in both cities. Finally, underclass men and women reported the lowest percentage use of intermediate and ultimate service facilities. Hopelessness Simple regression analysis (Table 4.13) shows prior socioeconomic status to be significantly related to hopelessness. However, only six percent of the amount of variance in hopelessness is explained by prior SES. While significantly related to prior socioeconomic status, hopelessness is very much a product of variables which reflect or aggravate the extreme poverty that is homelessness. Important to understanding the impact of the present condition on hopelessness are: service use, age, education, recent employment successes or failures, past experiences with homelessness, sex of respondent, and the 79 Table 4.13. Regression Analysis of Hopelessness by Prior Socioeconomic Status N F EG E S ON F PRIOR HOPELESSNESS SES HOPELESSNESS 1.000 PRIOR -.236 1.000 SES Prediction of Hopelessenss by Prior Socioeconomic Status Standardized Significance Predictor Regression T-Value Level Variable Coefficient Prior SES -.l8863 -3.355 .0010 * R Square: .0559 Dependent variable: Hopelessness of current life condition * Significant at 0.05 level. 80 stability and physical quality of the current residence. Multiple regression analysis (Table 4.14) shows a significant relationship between the above mentioned variables (including prior SES) and hopelessness. Together, they account for thirty-seven percent of the variance in hopelessness. Table 4.15 illustrates the patterned change in hopelessness for men and women over the four socioeconomic groups. Interestingly, men reported higher levels of hopelessness in each of the socioeconomic categories. 3. As the length of time Of the current episode.of homelessness increases for an individual, their range Of services utilized will decrease, and their personal perception of their homeless condition will become less hopeful. w The analysis of variance (ANOVA) test (Table 4.16) does show a significant relationship between the total number of services utilized and the amount of time homeless. However, by asking for the total number of services used since becOming homeless, this measure does not reflect the range Of services in use at the time of the interview. It can be expected that the total number of services used since becoming homeless will increase as the length Of time homeless (or repetitive nature of the condition) increases. Analysis of variance tests were also conducted to 81 Table 4.14. Results of The Prediction of Hopelessness by Thirteen Predictors Reflecting Poverty Conditions QQBBELAIIQE§ STABILITY CONDITION PRIOR EDUCATION SEX OF OF SES RESIDENCE RESIDENCE HOPELESS -.2627 -.0321 -.2733 -.1194 .1962 NUMBER PRIMARY INTERMED SERVICES SERVICES SERVICES WORK DEPENDENT USED USED USED HISTORY CHILDREN HOPELESS .0897 .1695 .1503 .1194 -.2478 SIMILAR TIMES SITUATION HOMELESS BEFORE AGE HOPELESS .3212 .0803 .1613 Ahaiysis of Varience SUM OF MEAN F F R D.F. SQUARES SQUARE VALUE PROB SQUARE Regression 13 13.98135 1.07549 3.64486 .0002 .3662 Residual 82 24.19573 .2950? Dependent variable: Hopelessness of current life condition. 82 . c mmwfiv AOOQD wwwflv QOKOJ N\\N . .nd 6622:8264 2.... 888 6:622 I L a.“ n 3 m msomom Baoooooowoom km mmOOmmBomom 13.6 3an 83 .H050H mo.o an acuosmscmsm . .mmwaoaon ocflsoomn mocfim owns mmofi>umm no Honsoz umdnswus> unmocmmwo vmo. moa.~ mom.ew mam.o~m mo mmmamsom mafia hmm.~¢ mum.we¢w mma HOHOB me.Hv bmo.mvmb nma uonnm « «mo. moa.~ mom.om mam.omm mo H0002 zuflawnmnoum Ozae> mucsom mmumovm Sooomum sowumflnc> m m coo: no saw no mmmuooo «o mousom mmOHOSOm OEHB mo cumcmq an coma mmow>umm no nonssz :fi mmocmumuufio pom m>os4 .oa.¢ manna 84 examine possible relationships between length of time homeless and percent of primary, intermediate, or ultimate service use. Analysis of variance (ANOVA) tests for primary service use by length of time homeless (Table 4.17), intermediate service use by length of time homeless (Table 4.18), and ultimate service use by individual length of time homeless (Table 4.19) show no significant relationship between the variables. Table 4.20 breaks down service use by amount of time homeless and sex of respondent. For each length of time homeless, percentage use of primary services is much greater than the percent use of intermediate and/or ultimate services. Overall, intermediate and ultimate service use is most active for persons homeless between 0 and 12 months. As a measure of the changes in the range of services utilized, percent use of the three service groups suffers from the same problem associated with the total number of services utilized; percentages are calculated from the number of services used over the whole of the homeless experience. The agency of contact with the service community service network for homeless was the primary service agency for a majority Of the sample. One way to determine if the amount of time homeless has an affect on the range of services utilized is to examine the nature of the service at which the individual was interviewed. A crosstabulation of 85 who. .Oms mow>nmm mumswuo mo useoumo ”manmwus> useosoomo Hmm.H moa. mam. oo mmmdmfiom OEwB wmo. HNH.OH mad HMHOB mmo. nom.m owH HOHHM Hmm.H med. mam. oo H0002 05Hm> mucovm mmnmovm Booemnm coflumwum> m one: «0 saw no memummo mo mouoom mmmaoeom mafia no numcon an on: mofi>u0m >usswnm Cw mooceueuufio Mom s>oc¢ .bH.¢ OHQMB 86 .Oms 00w>uow ouewomsumusw no useoumo "OHQOHHO> unmocwmeo omo. wmo.H moo. mac. oo wmo. bmo.b bod hmo. mom.o Hod mmo. omo. wmm.H moo. mac. mo OMOSOm unflawnmnoum moam> oncomm monsoom Soommum m m m one: no saw no mmmummo mmmamsom mafia mo summon >8 mm: OUH>HOW OHMflUOEHOUCH Cfl MOUCOHOHHflQ HOW M>OQ¢ mmmamsom mafia HSHOB HOHHW cowuewus> mo mouoom .mH.¢ OHQMB 87 .Omo mofi>u0m ousawuHo mo unmoumm "Tanmwus> useocmomo mom. cmN.H omo. mmn. mo mmmamfiom OEHB eco. Hem.b Hoa HmuOB «co. mmo.b mud Houhm Nwo. NON. vm~.H omo. mmm. mo HUGO! mucsmm aufiawnmnowm woam> oncoom mmumsmm Eoomwum cofiusfiuc> m h m one: no saw no mmmummo mo mousom mmmaosom mafia no numcmq an Omo mow>umm muwsfluao Ca mwocmummuflo Rom e>os< .ma.¢ manna 88 .mmo unmoumm ouoceo mononuceusm sq newsmam «muoz AH.mmo A~.¢~v Luv AH.6HV AH.smv AH.OHC mumsfiuap As.m¢v As.mev “no 16.44o Am.¢mv Am.omv museumsumucH A¢.oso Am.smv any Am.mmv As.mev as.mmc sumsflum 6.4H m.¢H : «.ma o.m~ ¢.HH 60m: .62 ..-..mmmm----mmm:--mfim----mmm::mmmm:--mmm ......... mmmmwmmm wmwmmmlmmlmmmm mmszoz as n ma mmezos ma u «H ......w...:. 4.38 31mm... ..... A .....mv.--l.....m6 1...: is ......... m may 14.6mv A~.m¢o 16.44v 16.6mm Am.Hev Am.mev Am.~mo 1~.H4V mumficmsumucH Am.mmv AH.HOV 14.44o Ae.mmo AH.mmv Am.omv A¢.mvo A».Hmv sumsflum m.~a 5.4H m.~a 4.4a m.~H o.mH ~.oH m.~H ammo .oz . wmmzmw::. moms .iuoazmmuunummmw!:::mmmmmw .wmmm momznm. mqnmm no szooxmmwm .o~.¢ manna 89 length of time homeless by the type of service at place of interview (Table 4.21) shows a significant relationship between the two variables. Of those chronically homeless, only nineteen percent were interviewed in service facilities with an ability to directly affect individual mobility out of the homeless condition. All of the chronically homeless contacted in mobility oriented services were interviewed within intermediate-type agencies. Twenty-one percent of persons homeless between 6 to 12 months were interviewed in mobility-oriented service facilities; all of which were classified as intermediate service agencies. By comparison, fully fifty-one percent of those homeless less than 6 months were interviewed in mobility- oriented service agencies; of which 10 percent were contacted in ultimate service facilities had been homeless less than 3 months. W Multiple regression analysis (Table 4.22) shows length of time homeless (duration and/or episodic nature of the condition) to be significantly related to a person's hopeful perception of their current life condition. The graph in Table 4.23 illustrates the change in hopelessness for men and women over the various lengths of time homeless. The 90 TABLE 4.21. CROSSTABULATION OF LENGTH OF TIME IN CRISIS BY THE TYPE OF SERVICE AT THE PLACE OF INTEREVIEW COUNT NOT MOBILITY MOBILITY ROW COL % ORIENTED ORIENTED TOTAL 0 TO 6 52 55 107 H MONTHS (48.6) (51.4) (53.5) 6 TO 12 30 8 38 MONTHS (78.9) (21.1) (19.0) 12 TO 18 12 2 14 MONTHS (85.7) (14.3) ( 7.0) OVER 18 33 8 41 MONTHS (80.5) (19.5) (20.5) COLUMN 127 73 200 TOTAL (63.5) (36.5) (100.0) CELLS WITH CHI-SQUARE D.F. SIGNIFICANCE MIN E.F. E.F.< 5 91 Table 4.22. Regression Analysis of Hopelessness by Length of Time Homeless, Homeless History, and Number of Times Homeless T RI 0 EG ESS ON F C E S NUMBEROF HOPELESS TIME HOMELESS HISTORY TIMES HOPELESS 1.000 TIME .348 1.000 HOMELESS HISTORY -.011 -.149 1.000 NUMBER OF .239 .095 .564 1.000 TIMES Prediction of Hopelessness by Length Of Time Homeless, History of Homelessness, and Number of Times Homeless Standardized Significance Predictor Regression T-Value Level Variable Coefficient Times Homeless .07651 2.742 .0070 * Length .08141 3.530 .0006 * History -.16974 -l.211 .2281 R Square: .1738 Dependent variable: Hopelessness Of current life situation * Significant at 0.05 level. Table 4.23 by Time Homeless Hopelessness 93 greatest increase in hopelessness is at one year, the same point at which intermediate and ultimate service use declines. Qohciusion Chapter IV has organized the results of the survey around the three hypotheses upon which the research was centered. Various statistical procedures were used to test the hypotheses and provide information on: 1) relationships between extreme poverty and type of homelessness; 2) the affects of extreme poverty participation on emergency service use and hopelessness; and 3) the affects of the amount of time homeless on service use and hopelessness of the individual life condition. The final chapter presents a discussion of these findings and the implications for research and service provision. CHAPTER V SUMMARY AND CONCLUSIONS This study sought to explore and to expand upon possible relationships between the severity of the homeless experience and prior - domiciled - underclass poverty. There is support for the first hypothesis which suggested that persons from an underclass background would most likely become chronically homeless when compared to individuals from other socioeconomic backgrounds. Sixty- eight percent of those defined as underclass prior to their homeless experience were chronically homeless at the time of their interview. The percentages for the other socioeconomic levels are: lower class - twenty-three percent; lower middle class - twenty-three percent; and middle class - thirty-one percent. Homelessness for forty- two percent Of those from either lower, or lower middle class, backgrounds was transitional in nature. The second hypothesis stated that persons from an underclass background would utilize a narrower range of services and have a less hopeful perception of their life condition. Differences in overall service use between 94 95 socioeconomic groups was not significant. However, homeless men and women from an underclass background had the lowest percentage use of the mobility oriented intermediate and ultimate service types. Prior socioeconomic status was significantly related to hopelessness, with homeless persons from an underclass background likely to record the highest levels of despair when compared to the other SES groups. In the analysis, men Consistently reported higher levels Of hopelessness than women. The third hypothesis suggested that as the length of time of the current episode of homelessness increased for an individual, their range of services utilized will decrease and their personal perception of their homeless condition will become less hopeful. Differences in percentage use of primary, intermediate, or ultimate services by length of time homeless was not significant. This measure was hampered, in part, by questions which asked for the total number of services used since becoming homeless, not for a range of services currently in use at the time of the interview. In order to analyze a range of services currently in use by the participants, the nature of the service provided at the place of interview was recorded for individuals of varying lengths of time homeless. The relationship between the amount of time homeless and the type of service at the 96 location of interview was significant, with a majority of persons homeless for more than one year currently using primary service agencies. Only transitional homeless were interviewed within ultimate service agencies. An individual's amount of time homeless was also significantly related to the hopelessness of their current life condition. The greater their length of time homeless, the greater their despair. The greatest increase in hopelessness was recorded for persons homeless for more than one year; the same point at which individuals recorded their greatest decrease in mobility oriented service use. Impiications of the Findings While the first hypothesis was supported, the analysis of variance in service use by people from various socioeconomic backgrounds provides little explanation for the varied patterns of homelessness. It is quite possible that the severity of the homeless experience is partially dependent upon the level of social and economic capital possessed by the individual prior to homelessness; factors diminished for the underclass participant by the extreme - isolated - nature of their previous poverty experiences. The ghetto participant is often the least educated or trained, with few of the skills necessary to survive in the increasingly competitive and changing economies of our urban regional and industrial centers. The three tiered service delivery system designed to help the homeless is simply not 97 equipped with the tools needed to reintegrate this most disadvantaged population. Persons with adequate individual capital to successfully utilize the limited mobility-oriented services do so, and most Often permanently escape the homeless condition. Those who do not have sufficient capital fail in their use Of mobility-oriented services and often become reliant upon those primary services which maintain life but offer little assistance out of the homeless condition. The findings also suggest that the high degree of social isolation that accompanies homelessness can effectively limit a persons ability to escape the condition. The longer a person is homeless, the more likely they are to depend upon services which offer little in the way of economic or social mobility. It is possible that extensive periods of time homeless (approximately one year) may actually decrease the social and/or economic capital available to the participant. Social capital, while still important to individual survival, is socially devalued as other homeless people begin to play important roles within the resource network. With only limited opportunity for skill training, or even present work experience, homelessness is hardly a condition that favors the development Of individual economic capital. As individual capital declines over lengthy periods of homelessness, the successful use of intermediate and 98 ultimate type services arrested. Overall, hopelessness is important as a measure reflective of individual capital. Where resources exist to assist movement out of the homeless condition, there is hOpe for positive change. Homeless persons from an underclass background recorded the highest level of hopelessness when compared to other SES levels. The high levels of hopelessness for the underclass is possibly a reflection of the extreme nature of their poverty which preceded homelessness and the level of resources at their disposal through which they can escape the condition. Hopelessness is also greatest for persons experiencing an extensive amount of time homeless, possibly a reflection of the diminished personal resources and the limited success with mobility-oriented services over time. The largest increase in hopelessness were recorded for men and women at the same time that use of intermediate and ultimate services declined; approximately one year into the homeless experience. Differences in levels of hopelessness for men and women are a possible reflection of the nature of basic services available to single men as compared to the more private accommodations used by single women and families of single women in either city. Higher levels of hopelessness for men might also be a reflection Of the nature and availability of welfare entitlement for the majority of single, unattached, 99 men as compared to the majority of women on AFDC. When addressing possible impacts on local service delivery networks, it is important to remember that the causes of chronic homelessness are many; the most visible being the mental and physical handicaps which complicate integration into permanent independent housing. Still, many become chronically homeless because of a lack of available resources.’ Declines in welfare entitlements, reductions in low-income housing, and on-going structural changes in urban economies will continue to create intense competition for increasingly limited Opportunities. Until local service delivery networks account for the lower levels of education and training associated with underclass participants, persons from extreme poverty backgrounds will continue to be likely candidates for chronic homeless participation. Limitations of the Studv A principle problem associated with this study of homelessness and the urban underclass centers around the measure used to determine individual socioeconomic background. Traditional measures comparing reported income with a national poverty line remain controversial. A preliminary study of homelessness in Benton Harbor, conducted by the Michigan State University Benton Harbor Project, reported difficulties in the self-report Of income in the years prior to the homeless experience. Because of the sporadic work histories of the homeless, and the 100 limitations associated with the self-report of income, socioeconomic status in this report was based on the individual's perception of the degree of poverty and economic Opportunity within their neighborhood prior to their homeless experience. The perceptions are relative, and may or may not represent the reality of the environment. Great care was given to the creation of uniform standards on which Opportunity and poverty were judged. Persons defined as underclass in this study were more likely to come from extreme poverty areas as opposed to Wilson's more highly inclusive underclass poverty tracts. As already discussed, questions concerning use of services addressed the whole of the homeless experience. This was particularly problematic when trying to ascertain levels of service use across different lengths of time homeless. The cross-sectional nature of the study also limited the analysis of changes in use of services over the individual amount of time homeless. The cross-sectional survey also limited analysis Of the different homeless patterns associated with the four socioeconomic categories. Finally, a major limitation of the study is the very nature Of the subject matter in question. Because the outlying population parameters of the homeless remain unknown, equal probability Of selection could not be achieved. The definition of the homeless used in this study, along with the sampling procedure employed in it's 101 operationalization, have created biases which must be recognized before any generalization of the findings may be made. 92112125120 The extremely poor constitute the greatest pool from which the homeless are drawn. Within extreme poverty, there exists an increasingly marginalized urban underclass; a subpopulation characterized by extremely low levels of social and economic capital. Economic transformations within traditional manufacturing centers increase the level of competition for remaining blue-collar employment to a level beyond the reach of this already disadvantaged subpopulation. As low-cost housing disappears, and the real value of formal support networks diminish, some of the underclass become homeless and confront a three-tiered service delivery system designed to arrest their slide and assist in their attempts to regain permanent housing. Extreme poverty, associated with the individual prior to homelessness, does affect individual movement through, and use of, the emergency network. It is possible that the low levels Of social and economic capital possessed by the underclass limit the effectiveness of those services which allow for the greatest degree of movement out Of the homeless condition. As length of time homeless increases, individual resources are further diminished and effective service use lessened. Hopelessness services as an adequate 102 gauge for the condition and extent of the resources held by the individual. Limitations have been identified and must be recognized when drawing conclusions or making generalizations from this exploratory study. It is hoped that future research in this area will reduce the limitations and more fully examine the relationship between the social phenomena of homelessness and underclass poverty. References Anderson, E. (1989). "Sex Codes and Family Life Among Poor Inner-City Youths". The Annals of the American Academy of Political and Social Science, vol 501:1989; Ihe_§heppe Uhgerclass: Social §cience Pepspecpives [edited] by W. Wilson. ’ Appelbaum, R. (1987). gountihg phe homeless. University of Chicago, Center for Urban Research and Policy Studies. Auslander, G. and Litwin, H. (1988). "Social Networks and the Poor: Toward Effective Policy and Practice". Soeiei Fork, ;;(3), 234-238. Bean, G., Stefl, M. and Howe, S. (1987). "Mental Health and Homelessness: Issues and Findings". Sociai Work, ;e(5), 411-416. Billingsley, A. (1989). "The Sociology of Knowledge of William J. Wilson: Placing Ihe_1:giy_flieegyehpegeg in its Socio-Historical Context". gehphel_ef_§eeielegy_ehgo SQEin_flsl£§I§. 19(4): 7'39- Block, F., Cloward, R., Ehrenreich, B. and Piven, F. (1987). an O : The ck O ar . Pantheon Books: New York. 103 104 Bonacich, E. (1989). "Racism in Advanced Capitalist Society: Comments on William J. Wilson's The Ipuiy Disagvehpageg". Jeurhal of Sociology and §eciei Weifape, i§(4), 41-55. Burt, M. and Cohen, B. (1989). "Differences Among Homeless Single Women, Women With Children, and Single Men". §eeiai Problems, 25(5), 508-524. Burtless, G. (1987). "Public Spending for the Poor: Trends, Prospects, and Economic Limits". Fighting Beveppy: What Works and What Doesn't. [edited] by S. Danziger, and D. Weinberg. Harvard University Press. Carliner, M. (1987). "Homelessness: A Housing Problem?". The Homeless in Contemporary Soeiety. [edited] by R. Bingham, R. Green, and S. White, Sage Publications: Newbury Park. Caton, C. (1990). Homeless in America. Oxford University Press: New York. Danziger, S., Haveman, R. and Plotnick, R. (1987). "Antipoverty Policy: Effects on the Poor and the Nonpoor". ove ° Wor n' . [edited] by S. Danziger, and D. Weinberg. Harvard University Press. Ellwood, D. and Summers, L. (1987). "Poverty in America: Is Welfare the Answer or the Problem?". Fighting Eovephy: at orks Wha esn't. [edited] by S. Danziger, and D. Weinberg. Harvard University Press. Ellwood, D. (1988). oor u or ° overt ' the Ame ' an ami . Basic Books, Inc: New York. 105 Farley, R. (1988). "After the Starting Line: Blacks and Women in an Uphill Race". Demogpephy, 2§(4), 477-495. First, R., Roth, D. and Arewa, B.D. (1988). "Homelessness: Understanding the Dimensions of the Problem for Minorities". Social Work, ;;(2), 120-124. Fischer, P. (1988). "Criminal Activity Among the Homeless: A Study of Arrests in Baltimore". Hos 't u Psychiatry, 12(1), 46-51. Hagen, J. (1987). "Gender and Homelessness". o i W k, ee(4), 312-316. Hammer, T. (1986). Sources, Characteristics and Policy Implications of the Urban Underclass. The Institute for Urban Economic Development. Evanston, Il. Hurst, R., & Lang, M. (1987). "Local and regional implications of manufacturing divestment and relocations in Southwest Michigan: The case with Benton Harbor, Michigan." Papers & proceedings of Applied Geography Conferences, ig, 1987. Jahiel, R. (1987). "The Situation Of Homelessness". The Homeless in gohtemporary §eeiety. [edited] by R. Bingham, R. Green, and S. White. Sage Publications: Newbury Park. Jenks, C. (1988, June). "Deadly Neighborhoods: How the Underclass has Been Misunderstood". Ihe_hem_3ephhiie. Johnson, A. (1989, December). "Measurement and Methodology: Problems and Issues in Research on Homelessness". fieeiei ngk Q Abstzggts. 106 Kasarda, J. (1989). "Urban Industrial Transition and the Underclass". The Annals of the American Academy of Political and Social Science, vol 501:1989; The_§heppe Uhderelass: Sociel Seience Repspeetives [edited] by W. Wilson. Kaus, M. (1986, July). "The Work Ethic State: The Only Way to Cure the Culture of Poverty". The New Republie. Lee, B. (1987). "Homelessness and Community". Paper presented at the Annual Meeting of the American Sociological Association, August, 1987. Lemann, N. (1986, June). "The Origins of the Underclass". The Atlantic Monthly. Long, L. (1974). "Poverty Status and Receipt of Welfare Among Migrants and Nonmigrants in Large Cities". Amepieem Sociological Review, 22, 46-56. Marks, C. (1989). "‘Occasional Laborers and Chronic Want': A Review of The Truiy Disedvahhege ". {emphai ef Secioiogy ehg Social Weifare, 15(4). 57-68. McLanahan, S. (1985). "Family Structure and the Reproduction of Poverty". hmemican Joupnal of §OCioiegy, 29(4), 873-901. McLanahan, S. and Garfinkel, I. (1989). "Single Mothers, The Underclass, and Social Policy". he als e ° n 't a O ' c‘e , 59;,1989; The ghetto Underciass: Sociei §eienee Rerspectives [edited] by W. Wilson. 107 McMurry, D. (1988). "Hard Living on Easy Street". thonicies, ;z(8), 15-19. Mennen, F. (1988). "The Relationship of Race, Socioeconomic Status, and Marital Status to Kin Networks". u a f gocioiogy ahd Secial Welfere, i§(4), 77-92. Michigan Task Force on the Homeless (1987). Lifie_ih Trensit. A Report to the Governor's Human Services Cabinet. Murray, C. (1984). Losing ground; American §ociel Reiiey. 1250-;980. Basic Books: New York. Patterson, J. (1986). America's Sppuggle hgainsp Reveppy: i990 - lee . Harvard University Press. Peroff, K. (1987). "Who Are the Homeless and How Many Are There?". The homeless ih Conpemporary Society. [edited] by R. Bingham, R. Green, and S. White. Sage Publications: Newbury Park. Piliavin, I. and Sosin, M. (1987). "Tracking the Homeless". Eeeme, 1Q, 4:20-24. Piliavin, I., Westerfelt, H. and Elliot, E. (1989). "Estimating Mental Illness Among the Homeless: The Effects of Choice-based Sampling". Soeiel Epehlems, ;§(5), 525- 531. Piliavin, I., Sosin, M. and Westerfelt, H. (1989). ... . s ., '-u I . . mo .- -, om-_essl- s' 4 Emplepepezy_§phgy. University Of Wisconsin-Madison, Institute for Research on Poverty. 108 Ropers, R. (1988). The invisible homeless: h Rem HIDQE Reoiogy. Human Services Press, Inc.: New York. Rossi, P., Wright, J., Fisher, G. and Willis, G. (1987). "The Urban Homeless: Estimating Composition and Size". Science z;_, 1336-1341. Rossi, P. (1989). Withoup Sheltep: hemeiessmess in She 12§Q§- Priority Press Publications: New York. Rossi, P. (1989). Qowh and Out in America: The Qrigihs of Romelessness. The University of Chicago Press: Chicago, Il. Rossi, P. and Wright, J. (1989). "The Urban Homeless: A Portrait of Dislocation". The Annals of the Academy of Political and Social Science, vol 501:1989; The Ghette Umderclass: Social Science Rerspectives [edited] by W. Wilson. Sawhill, I. (1988). Chailenge to Leadepship: Economic end Secial Issues To; the Neg; Decede. Urban Institute: Washington, D.C. Snow, D., Baker, 8., Anderson, L. and Martin, M. (1986). "The Myth of Pervasive Mental Illness Among the Homeless". Secie; Rpobiems, S;(5), 408-423. Snow, D., Baker, S. and Anderson, L. (1989). "Criminality and Homeless Men: An Empirical Assessment". Seeiei EIQDISES. 35(5): 532-549~ 109 Sosin, M., Colson, P. and Grossman, S. (1988). Romeiessness ih thcego: Roverty end Rathology, Sociel Tmstitutiens ehg o ia n e. University of Chicago, School of Social Service Administration. Stefl, M. (1987). "The New Homeless: A National Perspective". The hemeiess ih Contemporapy Seciepy. [edited] by R. Bingham, R. Green, and S. White. Sage Publications: Newbury Park. Stoner, M. (1989, December). "Beyond Shelter: Policy Directions for the Prevention of Homelessness". Soeiel WorR & Abstrects. Sullivan, P. and Damrosch, S. (1987). "Homeless Women and Children". The Homeless im Compempopery Society [edited] by R. Bingham, R. Green, and S. White. Sage Publications: Newbury Park. Sullivan, M. (1989). "Absent Fathers in the Inner City". The Annals of the American Academy of Political and Social Science, vol 501:1989; The Ghetto therclass: Seciel Seience Rerspectives [edited] by W. Wilson. Testa, M., Astone, N., Krogh, M. and Neckerman, K. (1989). "Employment and Marriage Among Inner-City Fathers". The Annals of the American Academy Of Political and Social Science, vol 501:1989; The Ghehto Uneepeiass: Seeiei Seiemee Repspeepives [edited] by W. Wilson. 110 VanGeest, J. (1988). Renton Harbor's Rroject tor the Homeiess, Benton Harbor Project, Michigan State University. Waquant, J. and Wilson, W. (1989). "The Cost of Racial and Class in the Inner City". The Annals of the American Academy of Political and Social Science vol 501:1989; The Ghetpo Ungerclass: Social Science Perspectives [edited] by W. Wilson. Wilson, W. (1980). The Qeciinimg Sighifiieence o: Reee: acks d a i Amer'ca I st'tu 'o s. University Of Chicago Press. Wilson, W. (1987). The Truiy Disadvahteged: The Inme; City, the Underclass, and Public Policy. University of Chicago Press. Wolch, J. (1990). "Homelessness in America: A Review of Recent Books". gournal o: Upban hffieirs, 12(4): 449-463. Wright, J. and Lam, J. (1987). "Homelessness and the Low- Income Housing Supply". Social Pol'c , 17:4:48-53. Wright, J. and Weber, E. (1988). me ssness h. McGraw-Hill, Inc.: Washington D.C. Wright, J. (1989). d ress nkno ' e Hom ss n Rmepiee. Aldine de Gruyter: New York. APPENDIX A SELECTED POPULATION CHARACTERISTICS 112 SEX OF RESPONDENTS Valid Cum Value Label Freguencv r t r e t Repcent MALE 117 58.5 59.4 59.4 FEMALE 80 40.0 40.6 100.0 TOTAL 200 100.0 100.0 Valid Cases 197 Missing Cases 3 RACE OF RESPONDENTS Valid Cum Value Label Ereguencv .Pegcent Rerceht Rercent BLACK 132 66.0 66.7 66.7 WHITE 57 28.5 28.8 95.5 HISPANIC 2 1.0 1.0 96.5 NATIVE AMERICAN 5 2.5 2.5 99.0 OTHER ' 2 1.0 1.0 100.0 (-999999) 2 1.0 MISSING TOTAL 200 100.0 100.0 Valid Cases 198 Missing Cases 2 113 EDUCATION OF RESPONDENTS Valid Cum Value Label Rreguencv Perce t er Esrssnt 8TH GRADE OR LESS 25 12.5 12.5 12.5 SOME HIGH SCHOOL 61 30.5 30.5 43.0 HIGH SCHOOL GRADUATE 71 35.5 35.5 78.5 SOME COLLEGE OR TRADE SCHOOL 26 13.0 13.0 91.5 TRADE SCHOOL GRADUATE 7 3.5 3.5 95.0 COLLEGE GRADUATE 10 5.0 5.0 100.0 TOTAL 200 100.0 100.0 Valid Cases 200 Missing Cases 0 114 MARITAL STATUS Valid Cum Value Label Epeguencv Pepeemt Rezeehp Fem SINGLE 121 60.5 60.5 60.5 MARRIED 25 12.5 12.5 73.0 WIDOWED 8 4.0 4.0 77.0 SEPERATED 17 8.5 8.5 85.5 DIVORCED 29 14.5 14.5 100.0 TOTAL 200 100.0 100.0 Valid Cases 200 Missing Cases 0 RESPONSIBLE FOR CARE OF DEPENDENT CHILDREN Valid Cum Value Label Fpeguencv .Percenp Perceht EEIQEEE NOT RESPONSIBLE 88 44.0 51.5 51.5 YES, PART TIME 24 12.0 14.0 65.5 YES, FULL TIME 59 29.5 34.5 100.0 (-999999) 29 14.5 (MISSING) TOTAL 200 100.0 100.0 Valid Cases 171 Missing Cases 29 CROSSTABULATION: CHILD 115 SEX OF RESPONDENT BY RESPONSIBLE FOR CARE OF DEPENDENT NOT YES, YES, Row RESPONSIBLE PART-TIME FULL-TIME Total MALE 57 21 15 93 (55.4) FEMALE 28 3 44 75 (44.6) Column 85 24 59 168 Total (50.6) (14.3) (35.1) (100.0) Shi-Sguare D.F. Significance 36.13460 2 .0000 Lambda (Symmetric) = .28481 Cramer's V .46377 Number of Missing Observations 32 116 LENGTH OF TIME HOMELESS Valid Cum Value Label Rreguencv Rerceht Rercent Percent 0 TO 3 MONTHS 73 36.5 36.5 36.5 3 TO 6 MONTHS 34 17.0 17.0 53.5 6 TO 9 MONTHS 15 7.5 7.5 61.0 9 TO 12 MONTHS 23 11.5 11.5 72.5 12 TO 15 MONTHS 9 4.5 4.5 77.0 15 TO 18 MONTHS 5 2.5 2.5 79.5 - OVER 18 MONTHS 41 20.5 20.5 100.0 TOTAL 200 100.0 100.0 Valid Cases 200 Missing Cases 0 117 NUMBER OF TIMES HOMELESS Valid Cum Value Label Rpegpencv Pepcent Rezeemp Rezcent 0 22 11.0 15.1 15.1 1 33 16.5 22.6 37.7 2 33 16.1 22.6 60.3 3 14 7.0 9.6 69.9 4 12 6.0 8.2 78.1 5 13 6.5 8.9 87.0 6 4 2.0 2.7 89.7 7 2 1.0 1.4 91.1 8 1 .5 .7 91.8 9 12 6.0 8.2 100.0 (-999999) 54 (MISSING) TOTAL 200 100.0 100.0 Valid Cases 200 Missing Cases 54 CROSSTABULATION: CITY BY TYPE OF HOMELESS == Count TRANSITIONAL EPISODIC CHRONIC Row Row % Total BENTON 36 36 28 100 HARBOR (36.0) (36.0) (28.0) (50.0) GRAND 40 33 27 100 RAPIDS (40.0) (33.0) (27.0) (50.0) 200 _(19°r°) APPENDIX B PERSONAL QUESTIONNAIRE 119 PRELIMINARY QUESTIONS: 1. Where are you currently living? 2. How long have you lived there? 3. How would you describe your current residence? emergency shelter single room occupancy hotel abandoned building outdoors transitional housing home of a family member home of a friend other (please specify) 4. Are you able to regularly sleep and receive mail at this location? no yes 5. Would you describe yourself as a migrant farmworker? no yes 6. Would you describe yourself as homeless at this time? no yes 7. Have you ever had any experience of being homeless? no yes If yes, probe for detail and decide if person will be interviewed 120 CHILDHOOD: We would like to ask some questions about your childhood and the neighborhood where you grew up. For each question, check the answer that is most like your own experience. 1. 2. 3. 6. When you were a child, did your family: own their own home. rent their home. move around a lot between family and friends. Did you live with: a single parent (mother or father) for most of my youth. a single parent for part Of my youth. both parents for all of my youth. When growing up, what was the employment status of the family? ‘ at least one of my parents has a full-time job. at least one Of my parents had part-time or temporary full-time employment. no one in my family was regularly employed. Was there any problem in your family of alcohol and/or drug abuse? alcohol and/or drug abuse played no part in my family. alcohol and/or drug abuse was present in my family. alcohol and/or drug abuse was a significant part Of my family life. Which of the following best describes the neighborhood where you grew up? most of my friends and neighbors were poor. some of my friends and neighbors were poor. only a few of my friends and neighbors were poor. How would you describe the neighborhood where you grew up with respect to crime and/or drug abuse? the neighborhood had little crime and/or drug abuse. the neighborhood had some crime and/or drug abuse. the neighborhood had a lot of crime and/or drug abuse. 121 7. In your neighborhood, what was the rate of high school dropouts? there were many high school drop outs. there were some high school drop outs. there were few high school drop outs. 8. How many single-parent families were there in this neighborhood? there were a lot of single-parent families. there were some single-parent families. there were few single-parent families. PRIOR SITUATION: We would now like to ask questions about your neighborhood and your personal situation for the two years prior to your use of emergency services. Think about your situation and where you lived before you needed emergency services. 1. Where did you live prior to your use of emergency services? Grand Rapids Benton Harbor Other (please specify) 2. Where did you live in this community? (please note address or approximate street location) 3. How long did you live at this residence? 4. Which of the following best describes the neighborhood where you lived? most of my friends and neighbors were poor. some of my friends and neighbors were poor. only a few of my friends and neighbors were poor. 5. How would you describe the area with respect to crime and/or drug abuse? there was very little crime and/or drug abuse. there was some crime and/or drug abuse. there was a lot of crime and/or drug abuse. 10. 11. 12. 122 What was the high school drop out rate like in this area? there were many high school drop outs. there were some high school drop outs. there were few high school drop outs. ' What was the nature Of opportunities for kids and young adults? there were many positive opportunities. there were some opportunities for kids. there few Opportunities for constructive behavior. How many single-parent families were there in this neighborhood? there were a lot of single-parent families. there were some single-parent families. there were few single-parent families. Did you move frequently in the two years prior to your present situation? I had only one address. I had two or three residences. I had four or more residences. When a major crisis did arise, were friends and family able to provide support? there was almost no one I could turn to for support. there were some family members and/or friends to whom I could turn to for support. there were many family members and/or friends to whom I could turn to for support. In the two year period prior to becoming homeless: I owned my own home. I was renting. I often stayed with family or friends. What was the employment situation in that two year period? at least on person had a full-time job. at least one person had part-time or temporary full-time employment. no adult in my family was regularly employed. 123 13. In the two years prior to this crisis situation: alcohol and/or drug abuse was a significant part of my life and/or those around me. there was some alcohol and/or drug abuse. there was little or no alcohol and/or drug abuse. 14. In general, how would you describe your changes in the two years prior to the present crisis situation? I was doing well economically but had a crisis that immediately forced me to this situation. I was doing well economically but things kept getting worse and I wound up in my present situation. Things hadn't been going well economically for some time and the resources finally ran out. The last question in this section addresses you whole adult life. Please select the response that best describes you as an adult. 15. How would you characterize you adult life? continued experiences w/ poverty and unemployment. some experiences w/ poverty and unemployment. little experience with poverty and unemployment. CURRENT SITUATION: We would like to ask you some questions concerning your current crisis situation in which you are living in an emergency shelter facility, transitional housing, or single room occupancy hotel. 1. What are the reasons for your present crisis situation? (check all that apply) ___eviction ___fire ersonal conflict ___released from prison ___family conflict ‘___released from hospital spouse abuse ___overcrowding economic reasons d1vorce new to area building condemned other (please specify) 124 How long have you been in this crisis situation? 0 to 3 months 3 to 6 months 6 to 9 months 9 months to 1 year 1 year to 1 year and 3 months 1 year and 3 to 6 months over 1 year and 6 months Have you been in a similar crisis situation before? no yes if yes, how many times? Have you ever had to stay with family or friends because of an economic situation? no yes if yes, how many times? Did you stay with family or friends immediately before your present crisis situation? no yes if yes, how long was this stay? 0 to 3 months 3 to 6 months 6 to 9 months 9 months to 1 year over 1 year How old are you? Are you: male female How much education have you completed? 8th grade or less some high school high school grad or GED some college or trade school trade school graduate college graduate 10. 11. 12. 13. 125 What is your marital status? single married widowed separated divorced Do you have children? no ____yes if yes, what are their ages? Are you responsible for caring for your children? no yes, part-time yes, full-time What is your race? Black White Hispanic Native American Asian other When was the last time you worked? presently working 1 to 3 months 3 to 6 months 6 to 9 months 9 to 12 months over 1 year ago SERVICES: Now we would like to ask you about services that may be available or needed by people like yourself in this community. For each service listed, please answer the following three questions: 1. To your Knowledge, is this service readily available to you at this time? 2. IS this service needed by people like you in this community? 3. Have you used this service since becoming homeless? 126 Service Available Needed Used YES NO YES NO YES NO Hospital emerg. health care Community health clinic Nutritional education Hospital care Counseling Substance abuse asst. Dental care Com. shower facilities Budgeting assistance Credit counseling Direct cash handouts Public assistance Loans Job training Job placement Day use of telephone Day labor hiring hall Child care Mail reception service Formal education Available Needed Used Service YES NO YES RQ YES RQ Bus passes or dial-a-ride Car loan Referrals for service Auto repair service 127 Car pooling Soup kitchen Food pantry Food stamps Emergency shelter Transitional housing Long term housing Cooking facilities Single room occupancy hotel Clothing;coat,shoes,etc. Housing location assistance FUTURE: Now we would like to ask you some questions about what you think your future condition will be like. For each of the following statements, please indicate your feelings using the following: SA = Strongly agree A = Agree N = Not sure D = Disagree SD = Strongly disagree 1. I will probably need a lot of help from many people to turn my life around. SA A N D SD 2. My current situation is only a temporary setback in my life. SA A N D SD 3. I am confident that I can overcome my current crisis situation. SA A N D SD 4. I will probably be homeless for much of the rest of my life. SA A N D SD 128 5. In a year from now I expect to be much better off than I am now. SA A N D SD 6. It would take a miracle for me to overcome my current situation. SA A N D SD 7. My present condition will get much better in the future. SA A N D SD 8. I better get used to my current situation because this is the way it's going to be for me. SA A N D SD INCARCERATION: 1. Have you ever been arrested? no yes if yes, how many times? 2. What offense(s)/crime(s) have you been charged with? (check all)] ___Breaking and entering ___Robbery ___Assault ___Murder ___Auto theft ___Sexual Assault ___Larceny (theft) ___Trespassing ___Shoplifting ___Domestic Violence ___Bad checks ___DUI ___Fraud ___other 3. What was your original Offense/crime reduced down to? 4. Have you ever been in a: Jail: yes no Prison: yes no Half-way house yes no 5. Were you provided assistance from a community agency upon your release from jail, prison, or half-way house? no yes if yes, was this for: Housing Employment Personal Counseling Medical Substance Abuse other 129 What agency was most helpful after your release? Has anyone in your immediate family ever been arrested? no yes if yes, were they incarcerated? yes no Did you have a permanent place to stay upon release from jail, prison, or half-way house? no yes if yes, was this: Rented home Own home Home of a relative Home of a friend Single room occupancy hotel other "7'111111111111111