.. < '7'. m1: _-.3 ‘4 r1") Swag? w,- 5. ma 7.5. . 9mm. . r? . 3'. Nail-.311: #1.. r ’2 n r. 5:915 ‘w‘n I .f. - 7 . “31;; amiuufinfi ‘\|ru . 1 A w‘ 3-3, . .1 6-3 . ‘ 4. ‘ V . I | 1'. 3.1"”; ,I- ‘!.‘v .\ a“ 4n . .43, 4‘.‘.‘ w- . . . n 1-‘1 . :rra . ‘, fir" .‘ a ,1 v‘ U41; My ‘ 'A u ' ' A1. ‘ r V 1‘ n'lflxw,“ .....)1A'., '0 ’4' ,- ~. - I; 1'}Vf" r .3, , ,-. ‘ a 3:9 - ,1 ‘nrv .,=. . 3-) ‘04 ”av n.“ l4 1“...» i". tt‘fi‘fin dink: ‘ ' x»- . ‘- k. n.3‘; ”Assign-5.. 3w. _ .11": . . m . «p...- us: x u""“£;“¢~ ‘ « an": d“... u \m .... .f{ *c. mswfliflu ‘C‘ .v . 4.. . :AACx‘CIT‘J.‘ 7'“ ‘ I u - w v - £323.. 3“, h‘ V “A 0‘ . .. - w.......m .. M. 3. cum I L 4.3;...” 2.3-, .. ~ 4: 1R, “ I _ 2 arm hm “Wang “‘3: 1:1" 7% 1 .,u kw - 1' , ‘L‘ 7 n, «v 1‘ r - ~. 1' (My . um :' an. .7 ‘ r .x‘ ‘ 2"” 1 9'1! a! Wu." J" ‘ a. m «I " YEI'IJI‘ t 5,. 4n. vw-fi «43:5 Q. if v r, . ill " If; ' w r .-A - 34:31:27 "ff-"IR. >--— . 1,. . '3" w ‘ ”'33?th N" 4‘ 3:; NIVERSITY LIB BIRAR ES lllllllllllllllllllIlllllllllIH| ll llllzl 1| 31293 008990 This is to certify that the thesis entitled DEMOGRAPHIC AND PSYCHO—SOCIAL PREDICTORS OF DEPENDENCY ON PUBLIC ASSISTANCE AMONG SOUTHEAST ASIAN AND EASTERN EUROPEAN REFUCEES- presented by Cheribeth Tan has been accepted towards fulfillment of the requirements for M.A. Psychology degree in W1» Afl/gwbfl Major professor Date I”)! 1" ( MQR 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution ~— LIBRARY Michigan State University J wT—T‘u —— 1p PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. ‘ DATE DUE DATE DUE DATE DUE JAN 1 9 _ g§5 i;fi134‘m’L _A ‘ MSU Is An Attlrmative ActioNEqual Opportunity lnmttution owns-pd DEMOGRAPHIC AND PSYCHO-SOCIAL PREDICTORS OF DEPENDENCY ON PUBLIC ASSISTANCE AMONG SOUTHEAST ASIAN AND EASTERN EUROPEAN REFUGEES BY Cheribeth Tan A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1992 ABSTRACT DEMOGRAPHIC AND PSYCHO-SOCIAL PREDICTORS OF DEPENDENCY ON PUBLIC ASSISTANCE AMONG SOUTHEAST ASIAN AND EASTERN EUROPEAN REFUGEES BY Cheribeth Tan This study was conducted to determine the various demographic as well as psycho-social factors that contribute to the utilization and dependency on public assistance among resettled refugees, and to discover differences among the ethnic groups that relate to their economic adjustment. Sixty Hmong, Vietnamese and Polish household heads were interviewed. Results showed that household size, length of stay in refugee camps, acculturative stress, amount of government assistance received, and favorable attitudes towards government assistance relate positively to welfare dependency. Education, occupational experienCe, and utilization of job assistance sources were other determinants of welfare utilization. Additional findings indicated significant ethnic differences on the above variables. Hmong refugees were at a disadvantage with respect to both acculturation and economic adjustment compared to the other ethnic groups. They also experienced the greatest amount of emigration stress. To my parents, Elizabeth and Chiao Kiat Tan iii ACKNOWLEDGEMENTS So many people have helped me make this project possible. First, I would like to thank my chairperson, Dr. Tom Reischl, for guiding me through the whole research process. His ready availability, technical help, sound advice and patience are greatly appreciated. I also would like to thank my other committee members, Dr. Barry Stein, Dr. William Davidson and Dr. Brian Mavis, for their willingness to work with me on this project, and for their help and unique contributions to this study. Many thanks to Nguyen Que and Chow Vue for their invaluable help in interpreting, clarifying issues, facilitating entry into the refugee communities, and providing insights into the people and the culture. Thanks also to Kirstin Rummel for her help in the interviews, and for our numerous discussion and brainstorming sessions. Special thanks to Patricia Hepp and the staff of Refugee Services Office for their whole—hearted support of the study, without which this project would not have been possible. Finally, my thanks to the refugees in the camps in the Philippines and in this study who have touched me in many ways. Their courage, resiliency, warmth, hospitality, and appreciation of simple joys serve to continually inspire me. iv TABLE OF CONTENTS LIST OF TABLES CHAPTER 1: INTRODUCTION U.S. Resettlement Issues Public Assistance Programs for Refugees Secondary Migration and Impactment The Welfare Dependency Issue Economic Adjustment Economic Status Predictors of Economic Self-sufficiency Psycho-social Factors Acculturation Model Emigration and Acculturation Stress Expectations and Welfare Dependency Rationale for Current Study The Present Study Hypotheses CHAPTER 2: METHOD Sample Data Collection Procedures Interview Measures and Operationalization of Variables CHAPTER 3: RESULTS Demographic Information V 10 10 12 20 21 22 29 31 32 33 38 39 44 45 51 51 English Ability Emigration and Acculturative Stress Economic Condition Employment History Utilization of Government Assistance Attitudes Towards Government Assistance Predictors of Government Assistance Utilization Relationships Among Predictor Variables CHAPTER 4: DISCUSSION Major Findings Economic Adjustment Predictors of Long-term Utilization of Government Assistance Ethnic Differences Methodological Issues Sample Response Validity Future Research Implications APPENDIX A: APPENDIX B: APPENDIX C: APPENDIX D: APPENDIX E: REFERENCES Southeast Asian and Eastern European Refugees Survey Interview Schedule Refugee Camps Where Respondents Stayed Before Resettlement to U.S.A. First-Aslyum Countries Where Respondents Stayed Before Resettlement to U.S.A. Occupation of Refugees in Their Home Country Jobs Held by Refugees in U.S.A. 54 56 62 67 71 71 72 86 87 87 87 89 92 96 96 98 99 101 105 127 128 129 131 133 Table 10 11 12 13 14 15 LIST OF TABLES Summary of Sampling and Tracking Procedures Demographic Information on Refugee Household Heads Mean English Ability Scores Emigration Stress Information Percentage of ”Yes" Responses to Emigration Stress Items in Order of Frequency Summary of Acculturative Stress Information Means and Standard Deviations of Acculturative Stress Items in Order of Seriousness Economic Situation Home Country Occupational Experience Reason(s) Given for Not Working Sources of Past and Future Job Assistance Differences Between Refugees Who Got Off Government Assistance Within the First 18-month Period and Those Who Continued to Receive Government Assistance Correlation Matrix of Hypothesized Variables Results of Discriminant Analysis of Variables Related to Dependency on Government Assistance After 18 Months Hit Rates Using the Discriminant Function to Predict Dependency on Government Assistance After 18 Months vii 42 52 55 56 58 61 63 65 67 68 70 73 76 81 85 CHAPTER 1 INTRODUCTION People have always been uprooted by persecution, conflict and famine in all ages. However, it is in the 20th century that refugee migrations have reached worldwide proportions. By 1990, the world refugee pOpulation is approximately 17 million, of whom 87% had found asylum in developing countries (UNPD, 1991). The recognition of an individual as a refugee is crucial as it contains implications for a country's immigration and asylum policies as well as the assistance and benefits given to the person (Suhrke, 1983; Moore, 1987a). A refugee as defined in the United Nations (1967) 9 o 01 ‘1 0! :- .7' . . 9- . a -n or :7 -ose: is "any person who owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or owing to such fear, is unwilling to avail himself of the protection of that country." It is important to consider the difference between refugees and other types of migrants. Despite their ethnocultural diversity, all refugees share a common predicament: they differ from other migrants in the involuntary nature of their migration. Kunz (1973) in his kinetic model of the refugee in flight suggests that immigrants are pulled to a new country: refugees, on the 1 2 other hand, are pushed out of their homeland rather than pulled out, mainly because of fear for their own safety. The validity of such fear, however, often can not really be tested. It is the person's interpretation of events and perception of danger that motivate the flight. ‘ The international community relies on three solutions to address the refugee problem: (1) voluntary repatriation, the return of refugees to their countries; (2) local integration, establishing new homes and settling in the country of first asylum; and (3) third country resettlement, transporting and transplanting refugees from countries of first asylum to other countries where they have the opportunity to begin a new life (Lacey, 1987; Moore, 1987a; Stein, 1981). Voluntary repatriation is the most desirable solution and most commonly advocated by the international community but then is often the most difficult to achieve. Normally, for refugees to be willing to return home, the conditions in the country of origin must have changed so much that the refugees no longer believe their lives or freedom to be threatened. Permanent local integration is also often unlikely as first asylum countries are among the poorest, and often find the responsibility of taking care of the influx of refugees a great burden. The third option, resettlement, ideally should be the last option to consider. The resettlement process is difficult and expensive, and demands tremendous effort on the part of the refugees and the host countries and communities. 3 The existence of refugees is not only a global problem but a domestic concern in the United States as well. The U.S. share of refugee resettlement is approximately 40% of the worldwide total (Moore, 1987b). Since World War II, the U.S.A. had responded to one Cold War refugee crisis after another, for example, Hungary in 1956, Cuba in 1960, Czechoslovakia in 1968, and Indochina in 1975. In the last 40 years, over 2 million refugees have been resettled in the U.S.A. (Goodwin-Gill, 1987: UNPD, 1991). Perhaps the most recognizable group of refugees to come to the U.S.A. over the last 15 years has been the Southeast Asian refugees from Vietnam, Laos, and Cambodia. From 1975 to 1990, of the 1.4 million refugees admitted to U.S.A., 68% were from Southeast Asia (ORR, 1991). Most of these refugees are from five ethnic groups: Vietnamese, ethnic Chinese primarily from Vietnam, Khmer of Cambodia, lowland Laotians, and Hmong and other tribal highlanders of Laos. Other refugees in the U.S.A. include: Soviet Jews, Eastern Europeans (Rumanians, Hungarians, Poles, Czechoslovakians), Haitians, Cubans, Ethiopians, Afghans, Iranians, Iraqis, and other smaller groups from Latin America, Africa and Asia. In 1990 alone, approximately 122,000 or an average of 334 refugees a day arrived in the U.S.A., 42% of whom were refugees from Southeast Asia, and 47% were from Eastern Europe and the Soviet Union (ORR, 1991). Several problems, however, abound in the resettlement of refugees and in the refugees' own adjustment to their new environment. 4 H1§1_B§§§§§l§mfinfi_1§§22§ In the U.S.A., resettlement is considered as the process by which a refugee is offered third country protection of a lasting nature, a sense of security, and the chance of a normal life (Goodwin-Gill, 1987). The objective of resettlement programs is simply to bring refugees to economic self-sufficiency as quickly as possible. Thus, resettlement efforts consist mainly of finding homes and jobs. These have historically been mainly carried out by non-government agencies. Over the years, these agencies became institutionalized as the mechanism linking the federal government and local resettlement. A recurring question in resettlement, however, is its objectives and the problem of defining when the goals have been met (Wright, 1981). Important events such as the different type of refugees being resettled, changes in the government's refugee admissions and resettlement policies, the welfare program, and secondary migration by the refugees have taken place that significantly affected resettlement. The legal basis for the U.S. resettlement programs is the Refugee Act of 1980. Before this Act was passed, there was no single policy guiding the entry of refugees into the U.S.A. Rather, federal refugee admissions and resettlement policies and programs developed in an ad hoc, reactive way (Wright, 1981): Zucker, 1983). With the coming of Southeast Asian refugees, the size of the resettlement resulted in the need for interagency coordination (Keely, 1983). 5 The Refugee Act of 1980 was a serious attempt to come up with a coherent, humane, comprehensive and efficient policy on the admissions, resettlement and assistance to refugees (Zucker, 1983). It was the most significant piece of legislation on the refugee issue which considered both international and domestic concerns (Keely, 1983: Wright, 1981; Zucker, 1983). Since 1980, however, there still have been shifts in federal refugee resettlement philosophy, policy and practice. Revised regulations on the eligibility of refugees for cash assistance, for example, cut the time period when refugees would be allowed special status from 36 months to 18 months, and further down to 12 months. WW As stated above, a major goal in the resettlement of refugees is self-sufficiency in the U.S. economic environment. Refugee resettlement policies have generally assumed, however, that a short-term utilization of public assistance is unavoidable (Wright, 1981). Thus, income support are made available to the refugees through the basic structure of federal and state cash assistance programs existing for other impoverished populations. §a§n_g§§1§tgngg. The joint federal-state program, Aid to Families with Dependent Children (AFDC), provides cash assistance to single-parent families with minor children. Several states also have an AFDC-UP (unemployed parent) program which covers intact families when one parent is unemployed. A refugee family can receive benefits under 6 this program if all other eligibility requirements for the AFDC program are met. Federal Supplemental Security Income (SSI) is the cash assistance program for the aged, blind and disabled. Refugees qualify for SSI under the same conditions as non-refugees. If a refugee family meets a state's income and resource eligibility requirements, but not the criteria on family composition, they are still eligible for the special Refugee Cash Assistance Program (RCA). Intact families in states without an unemployed parent program, and single individuals ineligible for AFDC can receive benefits from RCA (ORR, 1991). Initially, under the Refugee Act of 1980, refugees were entitled to 36 months of specialized public assistance under the RCA, after which refugees must qualify for cash assistance according to the regular AFDC and SSI rules (North, Lewin & Wagner, 1982). The 36-month period for RCA decreased to 18 months in 1982, and to 12 months in 1989 (ORR, 1991). When refugees cease to be eligible for the RCA and also fail eligibility requirements for AFDC or SSI, they would have to qualify under an existing state or local General Assistance (GA) program on the same basis as other needy residents (ORR, 1991). Needy refugees are eligible for food stamps under the same standard as non-refugees. uggiga1_g§§i§tangg. Medicaid authorized under the 7 Social Security Act, is the major medical assistance program for poor people (North, et a1., 1982: Wright, 1981). Refugees who fulfill all applicable Medicaid eligibility criteria can receive medical services under this program. Those who are eligible for RCA are also eligible for Refugee Medical Assistance (RMA), which is provided in the same manner as Medicaid is for other needy residents. Refugees may also be eligible for only medical assistance if their income is slightly above that required for cash assistance eligibility and if the medical expenses they incur bring their net income down to the Medicaid eligibility level (ORR, 1991). Refugees, thus, have a choice of either seeking English language and/or other training, or seeking immediate self- sufficiency through employment. Not only is the welfare system made available to the refugees but significant exceptions of some of the rules are also made in recognition of their special needs. WWW While voluntary agencies are responsible for resettlement, the U.S. government puts considerable pressure on them to disperse refugees throughout the country and thus prevent the development of large ethnic communities that might create tensions or strain the local community resources (Kelly, 1986). While refugees usually have little control over their initial resettlement, many of them relocate. The movement is often from rural to urban areas, 8 from northeastern states to southern and western states. The government takes a dim view of secondary migration, and yet puts no constraints over this movement (Kelly, 1986). Various explanations for secondary migration have been suggested: greater employment opportunities: more generous welfare benefits: better training opportunities: reunification with relatives and friends: the attraction offered by a established ethnic community: and a warmer climate (ORR, 1991). The development of impacted areas is exacerbated by latter-wave refugee influx, most of whom are now sponsored by relatives and compatriots. Impactment creates enormous financial and service drains in the social services which are further compounded by the reduction in federal assistance in human services. Competition for community-aid services, low-income housing, jobs, and scarce resources with other economically marginal residents has resulted in racial tensions and culture clashes between the refugees and the host communities (Zucker, 1983). Secondary migration has caused much consternation among sponsors, refugee-service providers and policy-makers who expected that dispersal could speed their absorption into U.S. society. The underlying concern is the slow process by which refugees attain economic self-sufficiency. We Approximately 49% of eligible refugees who have been in the U.S.A. two years or less are still dependent on public 9 assistance (ORR, 1991). The continued utilization and long-term dependence on public assistance by refugees is often considered a major problem in U.S. resettlement efforts (Kelly, 1986: RPG, 1987). The level of welfare dependency is a thorny issue. The high rate, the great number of individuals affected, and the high cost alarm public officials (Kerpen, 1985). Kerpen, however, argued that comprehending dimensions of welfare dependency for refugees is a complicated process. First, researchers are frustrated that government agencies assisting refugees with financial assistance programs often do not collect adequate, comparative data. The methodologies used to come up with estimates have been questioned as well. Second, a national rate does not reveal the important variations in the dependency rates among refugees from state to state. Dependency rates, for example, vary from a low 3% for the District of Columbia to a high 80% for California (ORR, 1990). Third, refugee's utilization of public assistance also fluctuates markedly over time. Newly arrived refugees opt for public assistance in large numbers but the rate of usage declines over time. Fourth, national backgrounds, ethnicity, and other social characteristics also differentiate recipients of public assistance (Kerpen, 1985). For example, the dependency rate among Polish refugees is roughly 14% but around 61% for Laotians including the Hmong (ORR, 1990). The reliance on welfare assistance also has little 10 association with employment and thus is not likely to be remedied by finding jobs but is more a function of the kind of employment open to newcomers (Kelly, 1986). Given the complexity of this issue, then, Kerpen (1985) concluded that it is hard to evaluate if a chronically dependent population of refugees is being created. EQQDQmi£_5912§&m§D§ Several field research studies have been conducted on the economic adjustment of resettled refugees. A variety of survey methodologies, including personal interviews, use of indigenous field workers, cluster samples of refugees, mail and telephone surveys were utilized. Most were done with the objectives of determining the economic status of the refugees and establishing reliable correlates of economic adjustment. mm For many refugees, living in the United States has meant unemployment, underemployment, downward social and occupational mobility, and dependency on the American welfare system (Caplan, et al., 1985: Kelly, 1986: Stein, 1979: Stepick & Portes, 1986). An extensive study on the economic self-sufficiency of refugees conducted by Caplan, Whitmore and Bui (1985) reported that of the 1,384 Southeast Asian households surveyed in five sites across the country, 50% fell below the federal poverty level. A majority of the households (65%) received some kind of cash assistance; 45% of those 11 who have been in the U.S.A. for over three years still received some cash assistance. This study also showed that only 44% of the 4,160 Southeast Asian adult refugees were in the job market: among these refugees, the unemployment rate was a high 42%. The unemployment rate is, however, an aggregate figure which includes all refugees seeking work from the moment they enter the U.S.A. The unemployment rate dropped from 86% for refugees in their first months of resettlement to around 30% for refugees with residence of more than three years in the U.S.A. The figures are slightly different in the Fall 1990 annual survey by the Office of Refugee Resettlement (ORR) of Southeast Asian refugees who had been in the U.S.A. less than five years. The results indicated a labor force participation rate of 36% for those in the sample aged 16 years and older as compared with 66% for the U.S. population. Of those in the labor force, either working or seeking work, approximately 92% were employed as compared with 95% of the U.S. population (ORR, 1991). Those who were working tend to have low status and low wage jobs and little direct transfer of their previous skills (Caplan, et al., 1985: ORR, 1991: Starr & Roberts, 1982). Stepick and Portes (1986) in their survey of 499 Haitian refugees in South Florida who arrived in the U.S.A. in the early 1980's also reported similar findings. A high 63% of the refugees were jobless: 29% received welfare aid. Those who managed to find jobs have also experienced 12 significant downward occupational mobility. W Factors influencing economic self-sufficiency of various groups of refugees identified in several studies include the following: (1) the length of time in resettlement country, (2) the size and composition of the household, (3) the utilization of multiple-wage earner strategy, (4) English proficiency upon arrival in the U.S.A., (5) education, (6) occupational skills and experience, (7) the prevailing labor market and economic conditions, (8) the type of sponsorship, and (9) ethnicity. Lgngth_gf_re§1dgngg. Employment among the refugees vary directly with the length of time they have been residents of the country they resettled in. Virtually all the refugees in the study by Caplan, et al. (1985) received cash assistance during the early months in the U.S.A. Measures of cash assistance dependency, unemployment and poverty level all showed a steady progress over time. In another study, Montgomery (1986) found that the unemployment rate of his sample of 537 Vietnamese refugees resettled in Alberta, Canada dropped from 50% for those in Canada 18 months or less to as low as 16% for those who have been in Canada for over four years. y-_=-99 e = - 1!! 9' u. 9 '-.e-- ‘2 1' =t { --:. Caplan, et al. (1985) reported that the manner by which Southeast Asian refugee households get ahead is by increasing the number of its occupants who are working and 13 bringing in earned income - that is, by a multiple-job strategy. Thus, households made up of extended families were somewhat better than nuclear families on overall economic status. The main element is the number of employable adults available in the household. The ORR survey reported a similar pattern. Households that receive no cash assistance have an average of four members and two wage-earners. Households depending on cash assistance average six members with no wage-earners (ORR, 1991). The number of children in refugee families has also been found to be an important factor in several studies (Bach & Carroll-Seguin, 1986; ORR, 1991: Rumbaut & Weeks, 1986). Bach and Carroll-Sequin (1986), for example, found the number of children to be a major predictor of labor force participation in their analyses of data derived from two national ORR surveys of Southeast Asian refugees conducted in 1982 and 1983, and their own telephone survey of 1,500 Southeast Asian refugee households in the U.S.A. Rumbaut and Weeks (1986) in turn found the number of children to be a major predictor of welfare dependency in their survey of 739 Southeast Asian adults in San Diego County in California. Apparently the need to support large families with several dependent children deplete the meager resources of refugees. Larger families also increase family responsibilities. At least one adult is more likely to remain at home with childcare responsibilities (Bach & Carroll-Seguin, 1986). 14 Eng11§n_§kill§. The ability to obtain and hold work in any occupations other than those that require unskilled, routinized labor depends on communication skills in English. Several studies show a strong relationship between English skills and economic adjustment of refugees. Caplan, et al. (1985) found arrival English to be a better predictor of economic self-sufficiency than current English, prior occupation, or predisposing characteristics such as educational level, sex and age. Strand's (1984) survey of 800 Southeast Asian household heads in the San Diego area in California indicated that the lack of English language communication skill is a major barrier to refugee employment. Similarly, the ORR survey (1991) found refugees who spoke no English to have a labor force participation rate of only 5% and an unemployment rate of 13% as compared to the labor force participation rate of 45% and the unemployment rate of 0% for refugees who claimed to speak English fluently. English facility is also a necessary skill for occupational mobility among Vietnamese refugees in Alberta (Montgomery, 1986). The refugees themselves in Nicassio and Pate's (1984) survey of 1,638 Southeast Asian refugees in Illinois have acknowledged the need to acquire English skills. English language training is found to be the most important need of 17-22 year old Southeast Asian refugees in a national survey of 249 refugees conducted by the Center for Applied Linguistics (1988). Fluency in the language of the host 15 culture is one of the best determinants of both economic and cultural adaptation since it is the key to learning the norms and expectations of the host culture. EWWHW. Samuel and Woloski (1985) in their analysis of the archival records of a sample of 3,687 immigrants including refugees to Canada in 1979 found out that years of schooling correlated positively with earnings and negatively with weeks of unemployment. Bach and Carroll-Seguin's study (1986) also found education to be a significant predictor of labor force participation. Skills and experience were the strongest correlates of employment among Haitian refugees in Florida (Stepick & Portes, 1986). A common explanation given is that education enables a refugee to obtain the skill facilitating adjustment to a new environment. Refugees with a higher level of education and/or skills may also be expected to command a higher wage rate due to their higher level of human capital than refugees with lower education and less skills (Samuel & Woloski, 1985). Montgomery (1986), however, found that education or training level was not significantly associated with employment among the Vietnamese refugees in Alberta. It was nevertheless significantly related to upward occupational mobility. ‘Lgga1_1ah91_ma;ket_ggnditign§. The resettlement and integration of new arrivals always occurs in a particular context. Samuel (1984) in his analysis of the economic 16 adaptation of refugees in Canada during the last quarter century using archival data found out that when the economic climate is favorable, even refugees with poor proficiency in the local languages found jobs without much difficulty as seen from the experiences of Czechoslovak and Hungarian refugees who arrived in Canada in the 50's and 60's. Among 70 Hmong communities, variation in self- sufficiency was reported by a 1985 government-commissioned report on Hmong resettlement in 30 states (Ranard, 1988). While a majority of the Hmong still depend on welfare, a growing number of small communities are becoming self-sufficient. One factor to account for this was the type of jobs available in a given locale to the refugees. In areas of high Hmong employment, refugees are able to find jobs that provide health benefits and wage levels adequate for the families to support themselves (Ranard, 1988). The extent to which an area offers jobs for women is also a critical factor of a Hmong community's economic independence. Without a second source of income, large Hmong families would have great difficulty in attaining self-sufficiency. The availability of low-cost housing in an area is another factor in the economic progress of the Hmong refugees (Ranard, 1988). Iype_9f_§pgn§9;§hip. A sponsor generally assists in the resettlement of a refugee entering the U.S.A. Sponsorships can be private U.S.-born families, formerly related or unrelated resettled refugees of the same ethnic 17 group, church congregation, or an organizational sponsor. The results of Bach and Carroll-Seguin's study (1986) showed at that those Southeast Asian refugees who participated in the labor force were significantly more likely to have been resettled by American families or church congregations. Relative and agency sponsors were more prevalent among those who were not in the labor force. One explanation given was“” that the networks and information offered by an ethnic community in which the members are themselves mainly in low wage jobs and having a relatively high rate of public assistance utilization, are less economically helpful than those offered by private American families and church congregations or by highly differentiated ethnic community with a strong entrepreneurial class which have access to more resources. Economic opportunities through host community networks is an important factor with Eastern European groups as well. In Samuel's study (1984) of various groups of refugees in Canada, the Hungarian refugees who arrived in the 1950's were absorbed quite rapidly into the labor force partly because of the presence of a large and very closely organized Hungarian community which served as a job information channel and also provided a large number of Hungarian-born employers willing to hire the refugees. Etnn1§1_y. Traditionally, refugee groups entering the U.S.A. came from developed areas of the world and often possessed marketable skills. This is still true of many of 18 the refugees from Eastern Europe. Such refugees were resettled relatively easy, and relatively few of them utilized public assistance before becoming economically self-sufficient (Stein, 1986: Wright, 1981). In the last 15 years, however, the refugee system has included primarily Southeast Asian refugees, the latter waves of which consisted largely of a great number of refugees who are poor, from rural areas, or members of tribal minorities. Such refugees tended to be less educated and even illiterate (Tayabas & Pok, 1983). They also tended to have had little exposure to Western culture (Strand, 1984). The resettlement of the more rural and tribal refugees in a modern technological society poses major problems. One of the processes by which refugees become more self-reliant in U.S. society is the movement from welfare dependency which is often their initial status to wage employment. However, wage employment is specific to modern culture, and some refugees have had minimal experience to prepare them for it (Fass, 1986). Few of these refugees have held jobs in the Western sense of the term. These refugees are structurally at a disadvantage in the labor market because employers perceive high risks in hiring workers who are unable to communicate in English well, are poorly literate or numerate, and have no clear previous work histories (Fass, 1986). Wright (1981) further stated that the risk of creating dependency on the part of this group of refugees is 19 increased by the fact that the benefits the refugees receive under the welfare system enable them to live at a standard higher than they have experienced in their homeland. They may even be unaware of or insensitive to existing pressures and sanctions to prevent the abuse of the welfare system. In Samuel's (1984) analyses of archival data on refugees in Canada, Asian refugees who were originally residing in and later fled Uganda were the ones who most easily adjusted economically to their new country. The Tibetans have the most difficulty settling down in Canada despite special measures to help them adapt due to cultural, linguistic, educational and health factors. Among refugees in the U.S.A., Stein (1979) using archival data from government surveys, reports to the Congress, Congressional hearings, and published studies on various refugee groups, found out that compared to other migrant groups, the Vietnamese are worse off than the 1970 immigrants, the 1956 Hungarian refugees, and the refugees from Nazism. A factor that he believes could partly explain the group differences is the degree of disparity in culture and level of development between the refugees and the host society. Among the 602 Vietnamese refugees in a survey conducted by Desbarats (1986) in Chicago, San Francisco, and Orange County in California, ethnicity constitutes an independent dimension of adaptation. Even after variables on personal characteristics and resettlement context were controlled 20 for, the Sine-Vietnamese refugees in the study were still at a disadvantage with respect to both acculturation and economic adaptation compared to ethnic-Vietnamese. An explanation provided by Desbarats was the apparent mismatch of the Sino-Vietnamese culturally defined attitudes and behaviors, such as the Chinese traditional resistance to acculturation, with contemporary American context. Schein (1987) and Habarad (1987) found the Hmong and the In Mien refugees depending on welfare as a means to protect themselves against the immediate need to find jobs that would force them to cope with the complexities of American life. The dependency rates calculated by nationality show relatively high dependency among the Southeast Asians compared with most other groups. The estimated nationwide dependency rates by nationality are about 56% for Vietnamese, and 61% for Laotians which include the Hmong. Among other nationality groups, the dependency rates are about 56% for refugees from Afghanistan, 45% for Ethiopian refugees, and 40% for Soviet refugees. Eastern Europeans other than Polish refugees show a dependency rate of about 25%, while Polish refugees have the lowest dependency rate at roughly 14% (ORR, 1990). s -s ' cto 5 Studies on the various correlates of economic self- sufficiency among refugees have often focused on socio- demographic variables as seen from above. An analysis of 21 other variables including such psycho-social constructs as acculturation process, stress, and learned helplessness might contribute towards a better understanding of the processes of economic adjustment of the refugees. AQQBlEBIQEiQn_HQ§§1 The term acculturation has various definitions. However, the following definition given by the Social Sciences Research Council is generally agreed upon among social scientists: acculturation is defined as ”culture change that is initiated by the conjunction of two or more autonomous cultural systems" (Berry, 1986). Typically, acculturation phenomena take place in a situation in which one cultural group is dominant over the other, resulting in various degrees of conflict and processes of adaptation. According to Berry (1986), adaptation can be conceptualized as occurring on both the group and individual levels of analysis. Modes of adaptation may be classified based on two questions: (1) whether the migrant group decides to retain its own cultural identity, and (2) whether positive relations are sought with the host culture. The four types of adaptation resulting from positive or negative responses to the questions are assimilation, integration, rejection, and deculturation. Integration consists of both retention of the migrant group's cultural identity and movement to become a part of the dominant culture. Assimilation, on the other hand, refers to a rejection of one's cultural identity and the adoption of the host culture 22 norms and values. Rejection refers to movement away from the host culture and the retention of one's old cultural identity. Deculturation involves rejection of both the old and host culture norms and values, reflecting a state of alienation (Berry, 1986: Nicassio, 1985). As to which mode of adaptation a refugee might engage in may depend on both individual and host community factors. The nature of the acculturating group such as their past experiences with the dominant culture, and psychological characteristics and individual differences will affect the mode of acculturation. Host cultural factors such as cultural compatibility between the migrant's background and the host society, and the general receptiveness of the host community to the migrants, also affect the mode of adaptation chosen by the newcomer (Berry, 1986). Employment is one of the first steps in the larger process of acculturation. Stein (1979) hypothesizes that this might even be the critical factor in the movement of the refugees into the mainstream of the host community. NW: The nature of migration may have tremendous effect on the refugee's well-being. In his kinetic model of refugee movements, for example, Kunz (1973) distinguishes flight and settlement pattern of refugees as either anticipatory or acute. Anticipatory refugees recognize the danger early and plan orderly departures before the deterioration of the situation in their own country. The movement of acute 23 refugees, on the other hand, result from an overwhelming push. Acute refugees flee often in panic with only the primary objective of reaching safety in a neighboring country that will grant them asylum. If the migration has been prolonged, unexpected and traumatic, it would exert an influence on the psychological well-being of the individual. Although the emigration experiences of the refugees varied markedly, and in spite of differences in demographic, social and cultural backgrounds, similar sources of stress are present. In the process of emigration and acculturation, refugees often experience numerous traumatic events. Salient features of the migration experiences of many refugees include high levels of unpredictability, and lack of personal autonomy (Lin, 1986). Various empirical studies discussed by Wortman and Brehm (1978) in a review of related literature on responses to uncontrollable outcomes suggested that individuals are motivated to maintain control over their environment, and that even perceived control is generally beneficial to them. Seligman (1975) in his learned helplessness model has suggested that a belief in one's ability to control the environment influences one's behavior-outcome expectancies. From this perspective, individuals with control over their environment expect that their own instrumental responses can produce desirable outcomes. But when individuals have been conditioned to feel that their actions do not make a difference to their state of well-being, feelings of 24 hopelessness, behavioral passivity and dependency may emerge and persist. Refugees often find themselves in situations where they have to surrender significant control over their daily lives and their fate to more powerful people or institutions. This state of unpredictability and loss of control can go on for years. Several have spent years in refugee camps with minimal control over their personal welfare and destiny, and without knowing when and even if a third country would accept them. W. The situation of the refugees who stay in camps or settlements for protracted periods of time differs widely. Certain common characteristics of refugee camps, however, have been identified by Murphy (1955). These are: ”segregation from the host population, the need to share facilities, a lack of privacy, overcrowding and a limited restricted area within which the whole compass of daily life is to be conducted." Opportunities for economic independence such as farming and other forms of employment are rare. Refugee camps have become semi-permanent holding centers for many who cannot go home and are not accepted for resettlement elsewhere (Bousquet, 1987). Data collected by Bousquet (1987) and Lacey (1987) stress the fact that the experience of living in a camp for sometime works against the future, for its unique environment has little relationship to the world outside. As Kunz (1973) describes 25 it, the refugees have "arrived at the spiritual, spatial, temporal, and emotional equidistant no-man's land of midway-to-nowhere and the longer he remains there, the longer he becomes subject to its demoralizing effects.” Habarad (1987) in his extensive study of the In Mien refugees found that the In Mien while in Southeast Asia had for the most part been autonomous, self-reliant and geographically dispersed. In camps, they came to manifest patterns of large-scale settlement, intra-community political centralization, and for the first five years, overall economic dependency. Habarad thus argues that factors leading to the stark contrast between earlier self-reliance to dependency in camps and resettlement are situational, not intrinsic or cultural. Rather, they are a result of the way in which camps and resettlement were planned and run to administer aid which discouraged individual action, initiative, or sense of responsibility. Q;1§1§_Qf_19§§. Central to the plight of refugees is the sense of loss in their lives. Refugees suffer from a crisis of loss of what is external - home, country, material possessions, work and status, family and friends - and loss that is less tangible - self-esteem, familiar social and cultural milieu, cultural mores, learned roles, personal identity, and self-respect (Lin, 1986: Rumbaut, 1985). Not only has the emigration process for many led to substantial and often multiple material, cultural and psychological losses, it has also caused widespread disruption in families 26 and other social support systems (Lin, 1986: Nicassio & Pate, 1984). Nicassio and Pate (1984) found that resettled refugees viewed being separated from other family members, having painful memories of the war and their flight from Indochina, and homesickness as more serious problems than practical obstacles to resettlement such as learning to speak English, having adequate financial resources, and receiving medical care. The severity of traumatic experiences vary among groups, however. Q;1§i§_gfi_lgag. It is also generally recognized that transcultural adjustment is often times a very stressful, difficult and frustrating process (Cohon, 1979: Lin & Masuda, 1983; Nicassio & Pate, 1984: Westermeyer, 1986). Just as the refugees suffer from a crisis of loss, so do they suffer from a crisis of "load" as they are required to come to terms with the present and the immediate future imposed by the need to survive (Rumbaut, 1985). Culture shock results from a sudden change from an expectable environment to an unpredictable one (Cohon, 1981). Refugees experience ”primary culture shock" as they realize how foreign everything is, often accentuated by their inability to satisfy and/or make their needs known because of lack of information or communication barriers (Lin, 1986: Rumbaut, 1985). Information on a person's relationship with the environment is slowly acquired through continuity and consistency of exposure to similar events in the course of 27 the person's socialization experiences. In the process of emigration and resettlement, however, refugees are uprooted from the environment they know. Often the migration experience results in cultural displacement as they end up living in a culture which contrasts greatly with the culture they have internalized (Aylesworth & Ossorio, 1983). A culturally displaced person loses the support of social and geographical familiarity, of established relationships, and of learned values internalized over time. The experience also leads to a disorganization of the individual's role system inducing cognitive stress. Organized social patterns through which one's needs are fulfilled may no longer be effective. Living in a dissimilar environment may also mean that many economic practices such as various farming techniques may no longer be applicable (Scudder & Colson, 1982). New response patterns are required. The culturally displaced person may not only be inadequately prepared to participate in the social and economic forms of the host culture, but problems may also arise from the fact that such a person is well prepared and strongly disposed to act and behave in accordance to what the culture of origin has defined as right or natural. Participating in the host culture may require a deviance from what one feels is natural and real (Aylesworth & Ossorio, 1983). According to Berry (1986), acculturation stress will be highest when the cultural distance, or cultural and 28 behavioral disparity between the two groups is greatest, and when it is involuntary on the part of the individual. A variety of demographic, social and psychological characteristics of the individual can modify the acculturation-stress relationship. Knowledge of English and education are correlated with low stress (Berry, 1987). Higher education leads to an increase in intellectual and even perhaps social resources to deal with the change. For many educationally disadvantaged refugees who have had minimal prior exposure to Western culture, there will be greater demands for cultural learning and social readjustment creating additional obstacles to their psychological and economic adjustment (Nicassio, 1985). Having lived in an urban (usually culturally plural) as opposed to a rural (usually unicultural) setting is another predictor of lower stress (Berry, 1987). The refugee experience is, thus, a high-demand, low-control situation that greatly tests the refugees' emotional resilience and coping resources. In addition to illnesses, financial insecurities, lack of shelter and other problems that are potential threats to refugees and other migrants and residents as well, refugees in particular are forced to confront the challenges of culture change from a more disadvantageous position as they encountered significant stressors during emigration and new stressors in the acculturation process that may continue to lead to subjective distress and frustration (Nicassio, 1984). The 29 psychological impact of this experience can even diminish the level of economic motivation (Cox, 1985). A common strategy to cope with the stress of migration to an unfamiliar environment is to cling to the familiar and change no more than what is necessary, to avoid activities that involve risks and hence might increase still further the level of stress (Scudder & Colson, 1982). The utilization of public assistance may then be one way refugees cope with their new environment by postponing the highly stressful experience of entering and competing in the work force of the host culture while they learn to cope with other internal and external stressors. WWW From the Hungarian refugees in the 1950's to the Hmong in the 1980's, resettled refugees oftentimes come to the United States with high, even unrealistic expectations of their new lives especially with regard to their economic adjustment (Stein, 1981: Westermeyer, 1986). There is also a tendency to dwell on refugee success stories which are not representative of the experiences of the group. Such stories set a standard of expectation that add to the refugee's frustrations (Stein, 1981). Many learn of government assistance even before arriving in the U.S.A. In his survey of 100 Eastern European refugees in Detroit, Krolewski (1988) found out that 68% of the respondents already knew of government assistance and benefits they can get here prior to their 3O arrival in the U.S.A. If not, they learn soon enough upon arrival through the "refugee grapevine" and come to expect the benefits. Refugee attitudes towards public assistance are reported by several voluntary agencies and by public welfare personnel as directly affecting an increase in cash assistance use (North, et al., 1982). One commonly observed belief among some refugees is that they are owed something by someone (Stein, 1981: 1986). Cash assistance is but to be expected. Skinner and Hendricks (1979) described a large number of refugees as viewing public assistance to be an entitlement stemming from their status as refugees or their belief that the U.S. government has an obligation to them, and thus are not reluctant to accept help on this basis. Refugees are also said to be sophisticated in their knowledge of public assistance benefits and of their entitlement to them. Some refugees even believe that the use of public assistance is fully acceptable (North, et al., 1982). Refugees may also not want to work when they know that at some point they are going to lose their benefits. For example, in 1987, a special pilot project in Fresno, California exempted all families, including refugees on welfare, from the 100 hour rule which states that a welfare recipient who worked 100 hours or more a month lost all of his cash assistance and his medical benefits. As a result, more refugees were willing to make the transition from 31 welfare to employment (Ranard, 1988). W In examining the various studies on refugees, several limitations were apparent. First, there is a need to directly compare different refugee groups within the same time period. Most of the empirical studies on refugees have been conducted with one ethnic group or another. Refugees are a very heterogeneous group, coming from different countries, ethnic backgrounds, religions, and social classes. Despite this, there are still some generalizations that can be made. However, very few studies have extended their studies and analysis beyond the experience of a specific ethnic group (Kunz, 1973, 1981: Stein, 1979, 1981). Two exceptions are Stein's (1979) systematic comparison of employment patterns among refugees from Southeast Asia, Nazi Europe, Hungary, Cuba and two immigrant groups, and a similar study by Samuel (1984) of the economic adaptation of Hungarian, Czechoslovakian, Uganda Asian, Chilean, and Southeast Asian refugees and other displaced persons in Canada. Both studies, however, relied heavily on archival records and published studies on the economic adjustment of different groups of refugees in varying time periods. Second, there is a need to explore the psychological aspects affecting the economic adjustment of refugees. At present, there is little systematic studies on the relationship between psychological variables such as stress, attitudes and expectations with employment status. Most of 32 the studies on economic adjustment are macrolevel studies which describe and relate refugees' demographic characteristics to their participation in economic life. An analysis of psychological factors and processes might help us understand better the relationship of refugees' socio- demographic characteristics and their economic behavior. The issue of refugees' dependency on public assistance is as crucial as ever. The search for solutions to dependency of refugees will influence greatly the direction of U.S. refugee policy and resettlement efforts. Understanding welfare dependency may help in the search for a better solution to the refugee problem. W The major objective of the study was to answer the following questions: (1) what are the factors that contribute to the utilization and long-term dependency on public assistance among refugees? (2) what are the major sources of variation among the different refugee groups which are related to their economic adjustment? Examining all potential predictors of refugees' dependency on public assistance is, however, extremely difficult given the complexities of the historical experience and cultural backgrounds of the various refugee groups, and the local environmental conditions of their resettlement areas. The study instead focused on the demographic factors identified from a review of the literature to be related to economic adjustment. It also 33 explored the relationship of psycho-social variables to the economic behavior of refugees. W The main hypotheses in the study were that (1) there are significant ethnic differences on various demographic and psycho-social variables, and (2) dependency on public assistance is influenced directly and indirectly by the background characteristics and personal resources of the refugees (ethnicity, occupational experience, educational level, English ability, and household size), behavioral and attitudinal variables (migration history and emigration stress experienced by the refugee, acculturative stress and attitudes towards government assistance), and community resources available to the refugee (type of sponsorship, access to employment assistance services, and amount of government assistance received). The specific hypotheses are as follows: 1. Ethnic Differences Ethnicity as to whether a refugee is Hmong, Vietnamese or Polish, is hypothesized to be significantly related to various variables. 1.1 Ethnicity is significantly related to household size. Polish households are smaller with fewer dependent children than Hmong or Vietnamese households. There are no significant differences in household size between Hmong and Vietnamese refugees. 34 Ethnicity is significantly related to educational level. Hmong refugees have the lowest average educational level, followed by Vietnamese refugees. Polish refugees have the highest average educational level. Ethnicity is significantly related to occupation in home country. Hmong refugees have the highest percentage of low-skilled occupational experiences among the three ethnic groups. Polish refugees have the highest percentage of high-skilled occupational experiences. Ethnicity is significantly related to emigration stress. Both Hmong and Vietnamese refugees experienced greater amount of emigration stress than Polish refugees. There are no significant differences in emigration stress between Hmong and Vietnamese refugees. Ethnicity is significantly related to length of stay in refugee camps. Both Hmong and Vietnamese refugees stayed longer in refugee camps than Polish refugees. There are no significant differences in length of stay in refugee camps between Hmong and Vietnamese refugees. Ethnicity is significantly related to acculturative stress. Hmong refugees experience the greatest amount of acculturative stress. Vietnamese refugees experience greater amount of acculturative 35 stress than Polish refugees. 1.7 Ethnicity is significantly related to attitudes toward public assistance utilization. Both Hmong and Vietnamese refugees have more favorable attitudes toward public assistance utilization than Polish refugees. There are no significant differences in the attitudes toward public assistance utilization between Hmong and Vietnamese refugees. 2. Predictors of Long-term Utilization of Public Assistance 2.1 Demographic Characteristics and Personal Resources 2.1.1 Ethnicity is significantly related to dependency on public assistance. Hmong refugees have the highest public assistance dependency rate. Vietnamese refugees have higher public assistance dependency rate than Polish refugees. The nature of one's occupation in home country is significantly related to dependency on public assistance. Refugees with lower-skill occupational experiences are more likely to be dependent on public assistance than refugees with higher-skill occupational experiences. Educational level correlates negatively with dependency on public assistance. English proficiency skill correlates 36 negatively with dependency on public assistance. Household size correlates positively with dependency on public assistance. Psychological and Behavioral Factors 2.2.1 The amount of emigration stress experienced correlates positively with dependency on public assistance. The length of stay in refugee camps correlates positively with dependency on public assistance. The amount of acculturative stress experienced correlates positively with dependency on public assistance. Refugee's favorable attitudes towards public assistance utilization correlate positively with dependency on public assistance. Community Resource Availability 2.3.1 2.3.2 The type of sponsorship for a refugee is significantly related to dependency on public assistance. Refugees sponsored by members of the host communities get off public assistance sooner than refugees sponsored by relatives or members of the same ethnic group. Access to employment assistance services correlates negatively with dependency on 37 public assistance. 2.3.3 The amount of government assistance received correlates positively with dependency on government assistance. Relationship Among Predictor Variables 3.1 Educational level correlates negatively with acculturative stress. English proficiency skill correlates negatively with acculturative stress. The amount of emigration stress experienced correlates positively with acculturative stress. The length of stay in refugee camps correlates positively with acculturative stress. The length of stay in refugee camps correlates positively with favorable attitudes towards public assistance utilization. Refugee's favorable attitudes towards public assistance utilization correlate negatively with their utilization of refugee employment assistance services. The type of sponsorship for a refugee is significantly related to acculturative stress. Refugees sponsored by members of the host communities experience less acculturative stress than refugees sponsored by relatives or members of the same ethnic group. CHAPTER 2 METHOD The study was conducted with the help of the Office of Refugee Services located in Lansing, Michigan. This office is the local affiliate of the United States Catholic Conference which has been one of the most active national voluntary agencies contracted by the government to handle refugee resettlement. Since 1979, the Refugee Services Office has been responsible for the resettlement of refugees consisting of families or single adults in the Mid-Michigan area. This agency has resettled a total of 2,076 refugees (703 cases) from 1979 to 1989 in the Greater Lansing Area. For the first three years, only Vietnamese, Lowland Laotians and Hmong were resettled. Polish refugees started to be resettled by the office in 1982. By 1984, the agency's resettlement efforts have expanded to include Cambodians, Rumanians, Hungarians, Bulgarians, Ethiopians, Iranians and Afghans. The Office of Refugee Resettlement of the Department of Health and Human Services (1991) reported that 49% of the refugees who had been in the U.S.A. 24 months or less were receiving some form of cash assistance. Among eligible refugees in Michigan, a dependency rate of 32% was reported (ORR, 1990). One of the major problems regularly cited by the state of Michigan in their refugee resettlement effort 38 39 reports is the need to continue to reduce dependency rates among all refugees. mm The sample for this investigation came primarily from the population of refugees who were resettled by the Refugee Services Office from 1984 to mid-1989, and who were still residing in the Greater Lansing area. The main criterion for determining eligible cases was the refugee's length of stay in the U.S.A. That is, respondents should have already been in the U.S.A. for more than 18 months at the time of the interview. They should then be ineligible for any special refugee assistance programs, and thus have to find other means of income. Data collection lasted for over a year. The sample was initially limited to refugees resettled within a four-year period, from 1984 to 1988. After a few months, it was decided to expand the time period to five and a-half years, from 1984 to mid-1989, to get the necessary sample size. The five and a-half year time range was short enough to ensure homogeneity and thus control for the effect of local economic conditions on the adjustment of the refugees. However, it was long enough for variation in economic adaptation among the refugees to take place. Only ethnic groups with at least 20 refugee households resettled in the Greater Lansing Area within the above time period were included in the study. Thus, the sample was limited to Hmong, Vietnamese and Polish refugees. Together, 40 they comprised 79% of the total refugee population resettled by Refugee Services Office. Several strategies were undertaken to generate a list of potential respondents. First, names of Hmong, Vietnamese and Polish refugees who resettled in the U.S.A. within the established time period were obtained from the masterlist of all resettlement cases processed by the Refugee Services Office. As several of these refugees have moved away or could not be located, the list of potential respondents had to be supplemented with additional names obtained from the list of refugees who sought job assistance services at the Refugee Services Office and from the directory of Vietnamese residents of Lansing compiled by the Vietnamese refugee caseworker. The fourth source for potential respondents were the Hmong interpreter and other refugees who participated in the study. Attempts were made to contact each eligible case. Several strategies were also used to track the potential respondents. First, addresses and phone numbers of eligible cases were obtained from their files in the Refugee Services Office. However, in most cases, these were the initial addresses and phone numbers of the refugees and were inaccurate at the time of the study. Other cases had no known addresses or phone numbers. Updated addresses and phone numbers were then obtained from phone directories, from the interpreters' personal knowledge or from their acquaintances, and from other refugees participating in the 41 study. In some cases, forwarding phone numbers or addresses of potential respondents were obtained from residents at the potential respondents' old address or phone number. Despite the various strategies utilized, there were still a small number of potential respondents who could not be located. Table 1 summarizes the results of the sampling and tracking procedures utilized in the study. The number of refugees obtained from the list of refugees resettled by the Refugee Service Office reflects the number of households and not necessarily the actual cases of refugees resettled. For example, there were a total of 65 Polish cases resettled by the office within the established time period. However, six cases were family reunification cases, i.e. family members joining another member (case) resettled earlier. Thus, the final number of Polish eligible cases derived from the Refugee Services Office was 59. Those cases that cannot be located either did not have any known address or phone number in the Refugee Service Office, or have transferred from a given address but have not been confirmed to have moved out of Greater Lansing Area. The primary respondent for each case was the household head. A household head was the current or most likely to be the primary income-provider for the household. In many cases, the potential primary employment-seeker was the member of the household attending adult education or skills 42 I I I Con- l l | I I | Source of Potential | N of Ifirmed ICannot | Unsuc-I I I I Respondents I eligi-I moved I be IcessfulIRefusedI Inter-I I I ble |out of |located|contact| I viewedI I I cases J QLAF I I I I I I l I I | I I I I Hmong I I I I I l l I I I I l I I I I Lists of refugees | 27 | 6 I 1 I 2 | 1 I 17 | I resettled by RSO** I | | I | | | I from 1984 to mid-1989I I | | I I | I I | I I l I I I Interpreter's I 3 | 0 | 0 | 0 I 0 | 3 I I knowledge I I I I I I I l I ------- I ------- I ------- I ------- I ------- I ------- I I Hmong total I 30 I 6 I 1 | 2 | 1 | 20 | I ---------------------------------------------------------------------- I I Vietnamese I | I I I | | I | I I I l | | I Lists of refugees | 23 I 8 | 4 I 0 I 4 | 7 | I resettled by R80 | | I | | | | I from 1984 to mid-1989I | I I | | | I l I | I I I | I Lists of refugees I 16 | 3 I 5 | 1 I 2 I 5 | I seeking Job I I I I | I | I assistance at RSO I I | | | I | I | I I | I I I I Lists of refugees | 8 | O | O I 0 I 0 | 8 | I from RSO Vietnamese | I I I | | | I caseworker | I I I | | I I l ------- I ------- I ------- I ------- l ------- I ------- I | Vietnamese total I 47 | 11 I 9 I 1 I 6 I 20 | I ---------------------------------------------------------------------- I I P011811 I I I I I | I I I l l I I I I I Lists of refugees | 59 | 17 | 14 | 8 | 2 | 18 | I resettled by R80 I | | | | | I I from 1984 to mid-1989I I | I I I | I I l | | I I I I Polish respondents I 2 I O | 0 I 0 I 0 | 2 | I I """"" I ------- I ------- I ------- I ------- I ------- I | Polish total I 61 I 17 | 14 I 8 | 2 | 20 I I ---------------------------------------------------------------------- I I Grand Total I 138 | 34 | 24 I 11 I 9 I 60 | I l I I I I I I Note. * Greater Lansing Area ** Refugee Services Office 43 training classes. A household is defined as the nuclear family and considered as one case by social service agencies. In five cases where extended families lived under one roof, elderly members of the household received government assistance separate from the rest of the family. Household and financial data used in the analyses were limited to the primary nuclear family. Data from the elderly members were excluded. Two or more unrelated refugees sharing a common dwelling but not common expenses were defined as separate cases by government agencies and were thus considered as separate households. Of the 138 eligible cases, only 69 (50%) were successfully contacted. Refusal rate among those contacted was 13%. A total of 60 (43%) refugees were interviewed with 20 cases for each ethnic group. However, data from two Polish and one Vietnamese respondents were excluded from further analysis as the interview data revealed that at the time of the arrival of these respondents to the U.S.A., at least one family member had resettled in U.S.A. for more than eight years. These family members were the primary sources of economic support for the first few years of the three respondents. This made the economic situation of these respondents different from the rest of the sample. Ninety-one percent of the 57 respondents were the primary existing or potential income earner of the households. Nine percent were not the potential primary income earner but were chosen to represent the household in 44 the interview. For these few cases, as much data as possible on the potential primary income earner were obtained from the household representative. W The source of data were structured interviews conducted by two interviewers in English or with the assistance of interpreters when necessary. Interpreters were utilized in interviews with Hmong and Vietnamese respondents. The two interpreters were refugees themselves. They were recruited to assist in the study upon the recommendations of Refugee Services Office. The interpreters received training sessions in interviewing and in which the intent and meanings of all the items on the questionnaire and the instructions were made clear to them. Initial consents were obtained before a respondent was interviewed. Before the start of each interview, respondents were also told of the purpose of the study, of the measures utilized to ensure confidentiality of their responses and of the voluntary nature of the interview. Interview sessions lasted on the average for a little over an hour. Special attention was given to assuring the respondents that their answers were confidential and that the study was in no way associated with any agencies such as the Department of Social Services or the Refugee Services Office. With two Polish cases, another member of the household present during the interview spoke better English than the 45 household head. Such household member was asked to aid in the interview as a translator. However, only the responses of the household head was officially recorded. 1 -‘ - .zaz-.-: .92 09- a or. . o .- . as -; The questionnaire contained the following sections: (1) general information, (2) household information, (3) English proficiency, (4) emigration history, (5) emigration stress scale, (6) activities in the first 18 months in U.S.A., (7) employment history, (8) family income and utilization of public assistance programs, (9) expectations and attitudes towards public assistance utilization, and (10) acculturative stress scale (see Appendix A). figng;a1_1nfgzmatign. This section provided data on the following variables: ethnicity, educational level, and type of sponsorship. Ethnicity was measured by a question asking the respondent to identify his/her ethnic origin. Educational level was tapped by the question asking for the number of years of formal education the respondent has had in the home country. Type of sponsorship was measured by asking the respondent to identify the sponsor either as a family member, a relative, a friend or acquaintance, an American family, a church-based organization or congregation, or an agency. This section also provided additional demographic information such as age, sex, religion, marital status, birthplace, last place of residence in home country and length of stay in the U.S.A. Eggsgnglg_1nfgxmatign. This section asked the 46 respondent to identify and give the age, sex, occupation and length of stay together of people in the household. From this information, three variables were calculated: the number of dependent children and household size at the time of the interview, and household size within the first 18 months in U.S.A. Eng11§n_ngfigign§y. This section had two parts. The first part was a self-assessment on a five point scale of the refugee's ability to understand, read, write and speak English at two time periods: (1) upon arrival in the U.S.A., and (2) at the time of the interview. Both English ability scales showed high reliability. Cronbach's alphas were 0.97 for the respondents' retrospective rating of their English skills to the time of their arrival in U.S.A., and 0.91 for the respondents' rating of their English skills at the time of the interview. This section also provided information on English classes taken by the respondent. The second part consisted of the interviewer's assessment, again on a five point scale, of the respondent's ability to comprehend and answer the interview questions in English. This second assessment was used to cross-validate the respondent's self-assessment of English ability at the time of the interview (£30.79, p<.01). Emigratign_n1§;g:y. This section asked for the escape experiences of the refugees including the length of stay in refugee camps, operationalized in terms of the total number of months of refugee camp experiences, from first asylum 47 camps to refugee processing centers. Emigratign_§t;g§§_§g§1g. The Emigration Stress Scale consisted of events which may or may not have been experienced by the refugees in the emigration process. The scale score indicates the level of emigration stress experienced. The pool of items for this measure came from the Emigration Stress Index (ESI) developed by Nicassio, Solomon, Guest and McCullough (1986). Cronbach's alpha for the ESI was 0.77. Another source of items was the Life Events and Social History Questionnaire developed by the Indochinese Psychiatry Clinic in St. Elizabeth's Hospital in Boston by Mollica (1987) and partly used to survey the number of traumatic events experienced by the refugees during the war, escape and stay in refugee camps. Reliability analysis of the 22-item Emigration Stress Scale using data from the present study showed Cronbach's alpha also at 0.77, suggesting an acceptable level of internal consistency given the nature of the scale items. A refugee who experienced one stressful event (item) may not necessarily also experience another item. Scale items with low item correlation to the total score were generally those of events differentially experienced by various ethnic groups. Removing such items to improve the scale's reliability would only bias the scale towards a particular group. MW This section asked the respondent what they did in the first 18 48 months of their stay here in the U.S.A. The answer to this section was used to substantiate responses in other sections, particularly on employment history. Two variables were also calculated: whether a respondent worked at some point during this time period, and/or attended classes. Emplgyment_ni§tgzy. This section provided information on the respondents' occupation in their home countries and their work experiences, if any, after arriving here in U.S.A. The use of employment assistance services were measured in terms of which employment assistance sources (agencies, friends, relatives, sponsor, etc.) the refugee utilized, which ones actually helped the refugee obtain employment, and which sources they would go to in the future. Other questions asked for skills training efforts, and reasons for being unemployed. .u' ! ou- 1!! - ' 1t ., . sup g,='» .1 nggrams. This section gave information on the family income, the utilization of various forms of public assistance, and other sources of support received by the refugees. This section measured the criterion, dependency on public assistance, which was operationalized in terms of a dichotomy, that is, whether the respondent was economically independent or continued to use any form of public assistance after 18 months. From data on the monthly income received from work, income from government assistance, and the time periods when such income were received, three additional variables were 49 calculated: proportion of income from work-related sources, proportion of income from government assistance and amount of monthly aid received within the first 18 months. Households are classified as being above or below the poverty level using the poverty index from the 1990 Statistical Abstract of the United States published by the Bureau of Census. A?‘ e ,‘17, o]! -. -°.‘,'f '1: °.~‘- 9-..‘ , 1-: E 1! o This section consisted of items tapping attitudes toward public assistance. Several of the items were constructed by the investigator. Other items were adapted from two scales constructed by Hinckley and Hinckley in the late 1930's: Attitude towards Receiving Relief and Attitude towards Earning a Living. Both scales are Thurstone-type scales with split-half reliability measures reported at 0.90 and with apparent good content validity (Shaw & Wright, 1967). Other questions in this section asked where and how the refugee first learned about government assistance. Principal components factor analysis with varimax rotation was performed on the scale in order to identify possible underlying dimensions. Results of the analyses showed no factor solution that provided conceptually distinct and coherent factor structure, and a decision was made to retain it as one scale. However, several items with very low corrected item-total correlations were excluded from the final scale. The resulting scale used in the analysis consisted of 11 items with Cronbach's alpha at 50 0.73. Aggultgzatiye_§t:§§§_§ga1g. The Acculturative Stress Scale consisted of problems encountered in the process of resettlement and acculturation. The scale score indicates the level of acculturative stress experienced. The items for this measure came from the Acculturative Stress Index (ASI) also developed by Nicassio, Solomon, Guest and McCullough (1986). Cronbach's alpha for the ASI was a high 0.91. Other questions asked about respondents' efforts to bring other family members to the U.S.A. and their desire to go back to their home countries. Initial reliability analysis of the 27-item Acculturative Stress Scale using data from the current study showed several items with very low corrected item—total correlations. The final scale is composed of 21 items with Cronbach's alpha at 0.79. CHAPTER 3 RESULTS MW Demographic information for the whole sample and for each ethnic group is summarized in Table 2. The sample of potential or present household income earners was quite young. Their ages ranged from 19 to 64 with an average of 34.16 years. Seventy-five percent were male while 25% were female. A majority of the respondents were married. Marital status, however, differed significantly among the ethnic groups. Most of the Hmong and Polish respondents were married while more than half of the Vietnamese respondents were either single male or single mothers. Household sizes ranged from a single individual to as large as nine members, with a mean of 3.88 members, again with significant ethnic differences. Scheffe‘ post-hoc test showed that Hmong households were significantly larger (p<.05) than the other ethnic groups; Hmong households also have on the average more than twice the number of dependent children than either of the Vietnamese or Polish households. Religion also differed among the three ethnic groups. The Polish respondents were mainly Catholics. The Hmong refugees either had converted to Protestantism or continued to practise ancestor worship or animism. The Vietnamese were either Buddhists or Catholics, or had no religion. 51 52 Table 2 u-fl.. ,.,° 1 .,u, .g ., ; - ,. .- . -:d- Ethnic Group Demographic ----------------------- Variable Total Hmong Viet. Polish Statistics (N=57) (n=20) (n=19) (n=18) Sex chi2=0 . 36 Male 75.4% 80.0% 73.7% 72.2% Female 24.6% 20.0% 26.3% 27.8% Marital status chi§=l9.20** Single 26.3% 10.0% 57.9% 11.1% Married 59.6% 80.0% 31.6% 66.7% Divorced 10.5% 5.0% 5.3% 22.2% Widowed 3.5% 5.0% 5.3% 0% Religion chi2=77 . 21*** Catholicism 40.4% 0% 31.6% 94.4% Protestantism 21.1% 60.0% 0% 0% Buddhism 14.0% 0% 42.1% 0% Ancestor 8.8% 25.0% 0% 0% worship Animism 3.5% 10.0% 0% 0% No religion 12.3% 5.0% 26.3% 5.6% Sponsor chi2=28 . 99*** Agency 38.6% 5.0% 31.6% 83.3% American(s) 31.6% 35.0% 42.1% 16.7% Family/rel. 29.8% 60.0% 26.3% 0% 53 Table 2 (cont'd). Ethnic Group Demographic ----------------------- Variable Total Hmong Viet. Polish Statistics (N=57) (n=20) (n=19) (n=18) Homeland residency chiA=30.07*** City 61.4% 15.0% 89.5% 83.3% Town 21.1% 40.0% 5.3% 16.7% Country 17.5% 45.0% 5.3% 0% § age 34.16 35.95 32.89 33.50 2:0.59 (9.28) (12.05) (9.19) (5.07) etaé=0.02 M household 3.88 5.60 3.26 2.61 E?13.09*** size (2.28) (2.26) (2.05) (1.20) etafl=0.33 H number of 1.89 3.40 1.27 0.89 E=17.10*** dependent (1.81) (1.85) (1.48) (0.68) etafi=0.39 children M months 34.18 42.20 31.89 27.67 E;5.66** in U.S.A. (14.88) (18.89) (11.86) (7.40) eta%=o.17 H years of 8.14 2.70 7.58 14.78 £352.90*** home education (6.12) (2.98) (5.12) (2.02) eta5=0.66 ** g < .01. *** p < .001. NOSE- Numbers in parentheses are the standard deviations. 54 The respondents have been in U.S.A. on the average for about three years with the Hmong being here the longest, followed by the Vietnamese, and then, the Poles. Almost all the refugee households in the study came through the Refugee Services Office with about a third of them sponsored by their family members or relatives. Another third had American sponsors. Around 39%, however, did not have any other sponsor except the Refugee Services Office or the Lutheran Refugee Office which took care of their arrival and resettlement in the Lansing Area. A majority of the Hmong respondents were sponsored by their family or relatives while most of the Polish respondents came as free cases with only Refugee Services Office as their main sponsor. The respondents varied widely in the number of years of formal education they had in their homelands. Scheffe‘ post hoc test revealed the Polish sample as the most educated, followed by the Vietnamese. The Hmong sample had significantly lower years of formal education. The Vietnamese and Polish respondents lived primarily in urban areas in their homelands prior to their escape while the Hmong respondents were mainly rural people. All the Hmong, 95% of the Vietnamese and 39% of the Polish refugees had some English training in refugee camps or in their own homelands prior to their arrival in U.S.A. In spite of having some amount of prior English training, however, most of the respondents recalled arriving in U.S.A. with very limited English ability. 55 average English scores of the sample. Table 3 shows the Table 3 WW Ethnic Group English ----------------------- Rating Total Hmong Viet. Polish Statistics (N=57) (n=20) (n=19) (n=18) Retrospective 1.95 1.55 2.25 2.07 E;2.29 self-rating (1.09) (0.87) (1.04) (1.25) etaz=0.08 of English skill to time of arrival Self-rating 3.22 2.69 3.37 3.65 E:7.73** of English (0.87) (0.77) (0.90) (0.65) eta2=0.22 skill at time of interview Interviewer's 3.32 2.60 3.45 4.00 E=7.15** rating (1.27) (1.37) (0.96) (1.07) eta2=0.21 N229. ** p < .01. Numbers in parentheses are the standard deviations. All three ethnic groups generally rated their English skills at the time of their arrival in U.S.A. as poor, averaging only 1.95 on a scale where 1=very poor to 5=excellent. However, they perceived their English ability to have improved, with self-ratings of their English skills at the time of the interview averaging 3.22 (fair) and the interviewer's rating of their English ability averaging 3.32. The Polish respondents had the highest ratings while 56 the Hmong sample had the lowest ratings. Younger respondents also had better English skills (gs-.35, 3:57, p<.05). All the Hmong, 95% of the Vietnamese, and 72% of the Polish respondents received additional English training in U.S.A. c an cc lt t v ess All the respondents experienced some amount of emigration stress, although in varying degrees. Table 4 gives a summary of the emigration stress experienced by the three groups. Table 4 Ethnic Group Emigration ----------------------- Variable Total Hmong Viet. Polish Statistics (N=57) (n=20) (n=19) (n=18) § emigration 8.15 11.45 6.61 6.11 zs18.21*** stress score (3.87) (2.84) (3.73) (2.40) eta%=0.40 E months in 31.39 76.45 10.63 3.22 z:101.53*** refugee camps (37.76) (28.42) (4.55) (7.87) etaé=o.79 E months in 37.37 76.45 10.79 22.00 §=71.11*** lst-asylum (34.47) (28.42) (4.49) (12.36) eta%=0.72 countries *** p < .001. Ngtg. Numbers in parentheses are the standard deviations. Emigration stress scores ranged from 1 to 18 with an average of 8.52 (possible range = 0 to 22). All respondents 57 also had to spend several months in some first-asylum countries before coming here, mostly in various refugee camps in Southeast Asia and Western Europe (please see Appendices B and C). The three ethnic groups varied greatly in the amount of emigration stress they experienced. The emigration stress scores of the Hmong respondents were almost two standard deviations higher than either of the other ethnic groups. The Hmong group also stayed the longest in refugee camps, averaging more than six years, while the other groups spent less than two years in first-asylum countries, and less than a year in refugee camps. Because of the highly significant ethnic differences in the emigration stress scores, individual emigration stress items were analyzed. Table 5 shows the stressful events experienced by the respondents while emigrating. Most of the respondents were separated from their loved ones, had to spend several years in refugee camps, and feared being captured while escaping. More than half also slept worse than before and had lost weight. Hmong respondents suffered the greatest amount of emigration stress. All the Hmong respondents feared being captured or of death while escaping, and stayed in refugee camps. Significantly higher percentages of the Hmong group experienced several other stressful and traumatic events such as walking for days through jungles and mined fields, being hungry and losing a lot of weight, seeing corpses or 58 Table 5 " ‘1 2°? 0 ' ‘=' fi901= - O u 0 '01 := ‘ns fn QIQEI.Q£.EI§QB§D§¥ Ethnic Group Item ------------------- Total Hmong Viet.Polish Chi2 (N=57)(n=20)(n=19)(n=18) Got separated from people 94.7 95.0 89.5 100.0 2.06 they felt very close to while escaping, in a camp or moving to the U.S. Had family member(s) left 84.2 90.0 68.4 94.4 5.48 behind. Stayed in a refugee camp 77.2 100.0 100.0 27.8 36.49*** before coming to U.S.A. Had another close 57.9 55.0 73.7 44.4 3.35 relative(s) they wanted to take with them left behind. Had a constant fear of 54.4 100.0 26.3 33.3 26.03*** being captured and/or of death while escaping. Slept poorly and 52.6 65.0 42.1 50.0 2.12 significantly worse than before leaving their homeland. Lost a lot of weight 50.9 80.0 26.3 44.4 11.67** more than 5 kilos). Were hungry much of the 49.1 85.0 36.8 22.2 16.66*** time. Walked for days through 33.3 95.0 0 0 52.73*** jungles and/or mined fields. Saw other people killed. 28.1 65.0 10.5 5.6 20.93*** Table 5 (cont'd). 59 Ethnic Group Item ------------------- Total Hmong Viet.Polish Chi2 (N=57)(n=20)(n=19)(n=18) Were completely alone 26.3 15.0 21.1 44.4 4.64 while escaping, in a camp or moving to the U.S. Were shot at or had people 24.6 50.0 21.1 0 12.97** tried to kill them while escaping. Saw other people beaten 22.8 25.0 31.6 11.1 2.28 up. Had family member(s) died. 21.1 40.0 5.3 16.7 7.38* Had another close 21.1 40.0 0 22.2 9.40** relative(s) died. Had to lie and deceive 21.1 5.0 10.5 50.0 13.44** people to escape. Left in a boat and spent 19.3 5.0 52.6 0 20.49*** several days at sea. Had new problems with 19.3 10.0 26.3 22.2 1.81 health. Thought of harming or 15.8 40.0 0 5.6 13.80** killing themselves. Felt as though other 15.8 35.0 5.3 5.6 8.55* people controlled most of their actions. Swam across a river. 12.3 35.0 0 0 14.76** Other people threatened 10.5 10.0 10.5 11.1 0.01 or tried to harm them in the camp or while moving to U.S. * p < .05. ** p < .01. *** p < .001. 60 witnessing the death of other people, experiencing the death of loved ones, being shot at or having other close brushes with death, and having to swim across rivers. Higher percentages of the Hmong group also thought of the possibility of having to kill themselves rather than be captured, and felt as though other people controlled most of their actions. Higher percentages of the Polish group had to lie and deceive people to escape: most of the Polish respondents were able to go to Western Europe on various pretexts, primarily as tourists and subsequently seeking political asylum in such countries. Half of the Vietnamese respondents were boat people who took the risk of leaving their homeland in small boats and reaching first-asylum countries in Southeast Asia: all Vietnamese respondents, however, had to spend some time in refugee camps. In terms of the acculturative stress, all respondents also experienced some amount of such stress. Table 6 shows the acculturative stress scores and other related information. The acculturative scale scores ranged from 0.05 to 1.23 with the average score at 0.54 (possible range = 0 to 2.0). No significant differences among the ethnic groups existed. There were, however, ethnic differences on other related acculturation variables. Most of the Vietnamese, and half of the Hmong sample but only 28% of the Poles wanted to and tried to bring other family members to U.S.A. When asked if they would want to go back to their homelands, a majority of Table 6 61 Summar1_2f_A22ulturatixe_§tre§s_lnfgrmation Ethnic Group Acculturation ----------------------- Variable Total Hmong Viet. Polish Statistics (N=57) (n=20) (n=19) (n=18) ; accultura- 0.54 0.53 0.59 0.49 E;0.74 tive stress (0.28) (0.21) (0.35) (0.24) etaS=0.03 score Attempted to 50.9% 50.0% 73.7% 27.8% chiss7.8o* bring family to U.S.A. Desire to go chi2=29.65*** back to one's homeland Would go 33.3% 75.0% 21.1% 0% back to live in homeland if situation changes Would go 56.1% 20.0% 57.9% 94.4% back only for a visit No desire 10.5% 5.0% 21.1% 5.6% * p < .05. *** p < .001. Not . Numbers in parentheses are the standard deviations. 62 the Hmong respondents as compared to only 21% of the Vietnamese and none from the Polish group, would go home if the situation in their homelands changes and there were no further threats to their lives. Most of the Polish and Vietnamese refugees would like to go back to their homelands but only for a visit and then come back to U.S.A. For descriptive purposes, means and standard deviations of each acculturative stress scale item were determined. Table 7 contains such information. The most common and most serious problem experienced by the respondents was the difficulty in finding a job, followed by learning to speak and understand the language. Feeling homesick and being separated from loved ones were other serious problems the respondents experienced. Common practical problems include their poor financial situation, and subsequently the difficulty in finding adequate housing. Cross-cultural difficulties such as lack of understanding of the behavior of Americans and Americans in turn not understanding their cultural ways were also other common problems. Eggngmig Condition The refugee households were generally impoverished. Table 8 summarizes the refugees' economic condition. The annual household income ranged from $0 to $40,800 with an average of only $12,565. The majority of the households were classified as below the official poverty level. There were, however, significant group differences 63 unsafe neighborhood. Table 7 !-',l: 21'. 0.192 '_ .’ . 'I‘ ' i ‘1 ‘IL‘ W Item Item Item Mean S.D. Difficulty in finding a job. 1.53 0.73 Learning to speak and understand English. 1.37 0.82 Feeling homesick. 1.14 0.75 Not having enough money for basic 0.97 0.71 necessities. Being separated or isolated from family 0.77 0.79 members or other loved ones. Finding adequate housing. 0.60 0.68 Understanding the behavior of Americans. 0.56 0.63 Americans not understanding their 0.54 0.73 cultural ways. Difficulty in dealing with government 0.47 0.73 agencies Having to get used to a different climate. 0.46 0.66 Feeling unable to do anything about the 0.39 0.59 events in their homeland. Not understanding rules and regulations. 0.35 0.55 Experiencing a big change in their role 0.33 0.64 in the family. Not knowing what was/is happening in 0.30 0.54 their homeland. Having to live in an undesirable or 0.30 0.60 64 Table 7 (cont'd). Item Item Item Mean S.D. Being/looking different from other people. 0.25 0.43 Conflict over American and their own 0.24 0.47 cultural ways of behavior. Knowing how to take care of daily needs. 0.21 0.53 Being unsure of how to act in public 0.18 0.38 places. Feeling afraid or anxious when they meet 0.18 0.47 Americans. Difficulty raising children because of 0.16 0.49 confusion over American and their own cultural values and behaviors. 65 Table 8 Economic_§ituation Ethnic Group Economic ----------------------- Variable Total Hmong Viet. Polish Statistics (N=57) (n=20) (n=19) (n=18) R annual 12565 12000 8807 16951 =7.24** income at (7173) (2817) (4771) (9957) etafl=o.21 time of interview Below poverty 62.5% 85.0% 72.2% 27.8% chi%=14.31*** level § proportion 0.42 0.32 0.30 0.65 3:3.77* of income (0.46) (0.46) (0.39) (0.46) eta5=0.12 from work i proportion 0.54 0.68 0.59 0.34 £s2.97 of income (0.46) (0.46) (0.44) (0.45) eta5=0.10 from govt. assistance Continued to 68.4% 95.0% 73.7% 33.3% chil=17.04*** receive aid after 18 mos. x amount of 618.91 837.50 533.17 461.78 £;11.64*** monthly aid (302.15)(258.50)(302.18)(198.07) eta5=0.31 received in 1st 18 mos. Worked within 43.9% 5.0% 47.4% 83.3% chi%=23.75*** lst 18 months Had a job at 50.9% 35.0% 47.4% 72.2% chil=5.39 the time of interview Received asst. 70.2% 60.0% 78.9% 72.2% chifi=l.72 from non-govt. sources * p < .05. ** p < .01. *** p < .001. Ngtg. Numbers in parentheses are the standard deviations. 66 on income, and proportion of the ethnic group living below poverty level. While the Vietnamese had the lowest annual income, Hmong households with the highest number of dependents were the most impoverished despite receiving the highest amount of government assistance within the first 18 months. Polish households were financially better off. In spite of their generally poor economic condition, 32% of the sample managed to send money from time to time to their families and relatives in their home countries. Household income were mainly from work or government assistance. On the average, 54% of their income came from government assistance while 42% came from work-related sources. The Polish households had on the average a higher proportion of their income from work than either of the other two groups. A significantly higher proportion of the Polish respondents worked at some point within the first 18 months than either the Hmong or the Vietnamese respondents. At the time of the interview, about half of the respondents were employed with no significant differences among the ethnic groups. Most households also received some form of assistance from non-governmental sources such as their sponsors or from private donors through Refugee Services Office, usually in the initial months of their stay here. Such assistance includes furniture (54%), food items (35%), clothes (28%), household items (28%) or appliances (18%). 67 W Prior to coming to U.S.A., the refugees had very diverse occupations, from managerial-professional jobs (e.g. engineers, scientists, managers), skilled jobs (e.g. mechanic, electrician, plumber, hairdresser) to military jobs and unskilled labor (e.g. farming, dishwasher, factory worker). Table 9 contains a summary of the respondents' occupational experience in their homelands. Appendix D contains a more detailed description of the various occupational experiences of the respondents. Classification of the home country job experiences were based on the job classification found in the 1990 Statistical Abstract of the U.S.A. by the Bureau of Census. Table 9 H g I i I' J E . Percentage Occupational Category -------------------------------- Total Hmong Viet. Polish (N=57) (n=20) (n=19) (n=18) Professional-managerial 21.1 5.0 10.5 50.0 Skilled labor 49.1 30.0 73.7 44.4 Unskilled labor 29.8 65.0 15.8 5.6 Note- Chiz=27.62, p < .001. Most of the Polish refugees with several years of formal education have had professional jobs in their home 68 country. Hmong refugees were farmers and/or guerilla fighters. Vietnamese refugees were mostly in occupations requiring skilled labor. In the first 18 months of their stay here, 91% of the respondents pursued various areas of education, from English instruction, basic Adult Education, high school or college to specific job skills training. Less than half of the sample (44%) worked at some time within this time period. However, that a refugee worked within the first 18-month period did not necessarily lead to independence from‘ government assistance. Among the 25 respondents who did work, 7 (28%) were still on assistance after 18 months. At the time of the interview, only 51% of the total sample were employed. The main reason for not working was the need for them to go to school (see Table 10). Table 10 WW9 Reason n % Attending school 20 35.1 Poor English 12 21.1 Difficult to find jobs 12 21.1 Keeping house 6 10.5 Can live on government 3 5.3 assistance Others 4 7.0 69 Most of the refugees (77%) did look for jobs: 70% approached several sources for job assistance. Only 23% did not attempt to find jobs: 6% of the Polish sample, 16% of the Vietnamese sample and 45% of the Hmong sample (152(2,N=57)=9.17, p<.05) . Table 11 shows job search and employment assistance information. The most common sources for job assistance utilized by the refugees were the Refugee Services Office and their friends or acquaintances. About a third (35%) undertook other self-initiated actions such as going through the yellow pages of the telephone directory or job ads in newspapers and bulletin boards. Self-efforts proved most successful in providing jobs for the refugees with 35% finding jobs this way. When asked who the respondents would go to for job assistance in the future, more than half would depend on their own efforts first rather than on any other individuals or agencies. Friends or acquaintances and the Refugee Services Office were other sources respondents would still approach for job help in the future. Very few would go to any formal job assistance service agencies. More than half of the refugees (61%) have had some work experience since their arrival in the U.S.A. Most jobs found by the refugees (53%) were in service jobs such as cleaning, waiting in restaurants or washing dishes, or labor intensive jobs such as working in factories, warehouses, or supermarkets: other jobs held by refugees (42%) required some skills such as welding, drafting, clerical and 70 Table 11 §QBI22E_QI_Ei§E_éD§_EB§BI§_IQD_A§§1§§§n£§ % of Refugees Source Approached Received Would Go to Source Successful Source in Assistance the Future Refugee Services Office 45.6% 17.5% 31.6% Friend/acquaintance(s) 38.6% 19.3% 42.1% Sponsor(s) 10.5% 5.3% 14.0% Relative(s) 3.5% 0% 3.5% Government job assistance 10.5% 0% 8.8% service agency Private job assistance 12.3% 3.5% 3.5% service agency School 3.5% 5.3% 0% Self-initiated 35.1% 35.1% 54.4% 71 administrative support jobs, and construction work. Only around 11% found professional higher-paying jobs. These were all Polish refugees who already had such professional occupations in their home country. More than a third (39%) had never been employed, either they could not find a job that paid enough for their basic needs and/or chose to concentrate on furthering their skills through formal education. Appendix E contains the different types of jobs held by refugees since their arrival in U.S.A. HI'JI II E E I E . I Majority of the heuseholds (68%) received some amount of government assistance for more than 18 months (see Table 8), but there were significant ethnic differences. The highest percentage was for the Hmong group with 95% still on government assistance at the end of 18 months. The average total amount of aid received by a household per month within the first 18 months was $618.91. Most households (65%) received Aid to Families with Dependent Children while 18% received General Assistance at some point in their first 18 months here. About a fifth (19%) stated receiving the Refugee Cash Assistance for a few months. W The mean score on the Attitudes Towards Government Assistance Scale was 1.63 with scores ranging from 1.09 to 2.00 (possible range = 1.0 to 2.0). The mean score indicates that most refugees believed in their entitlement to some form of assistance from the government and looked at 72 such assistance positively without the shame and stigma other people would attach to it. However, the three groups differed in their scores on this scale (£(2,54)=8.98, p<.001: eta§=.25). The Polish sample had significantly smaller mean than the other two groups at 1.49 (§&Q;:0.20) as compared to 1.70 (s. =0.14) for the Hmong sample and 1.70 (figgg=0.17) for the Vietnamese group. Most refugees (61%) learned about government assistance only here in the U.S.A. from the Refugee Services Office staff (23%), their sponsors (25%) or from their friends and relatives (14%). More than one-third (37%) learned about government assistance prior to their arrival here mainly from workers at refugee camps or other first-asylum countries (25%). This was particularly true for the Vietnamese sample with 63% already aware while still in refugee camps of the existence of special government assistance for resettled refugees as compared to only 10% of the Hmong and 39% of the Polish samples (ghif(4,N:57)=14.30, p<.01). Almost half (48%) expected to receive cash assistance from the government when they arrived here. 3 ‘2'. 0 ‘ -_ 0 ‘ Lu‘! 4-8" a ‘ , a 0 Out of the twelve variables hypothesized to predict long-term utilization of public assistance, nine were significantly related to the criterion variable. Table 12 summarizes the differences on the hypothesized variables between refugees who continued to receive assistance from those who got off assistance after 18 months. 73 Table 12 'V‘QLH'! Off GA Still on Predictor Total After 18 GA After Statistics Variable Months 18 Months (N=57) (n=18) (n=39) Ethnicity chiz=17 . 04 um Hmong 35.1% 5.6% 48.7% Vietnamese 33.3% 27.8% 35.9% Polish 31.6% 66.7% 15.4% Sponsor chiS=5.71 Agency 38.6% 61.1% 28.2% American grp/ 31.6% 22.2% 35.9% individuals Family/rel. 29.8% 16.7% 35.9% Occupational chiz=8 . 93 * experience Professional 21.1% 44.4% 10.3% Skilled labor 49.1% 38.9% 53.9% Unskilled labor 29.8% 16.7% 35.9% i years of 8.14 12.50 6.13 _s4.42*** home education (6.12) (4.76) (5.66) § self-rating 1.95 2.15 1.85 t=0.94 of English (1.09) (1.15) (1.06) skill at time of arrival Table 12 (cont'd). 74 Off GA Still on Predictor Total After 18 GA After Statistics Variable Months 18 Months (N=57) (n=18) (n=39) E household 3.40 1.67 4.21 ts-6.07*** size in the (2.19) (1.03) (2.13) first 18 mos. E emigration 8.15 6.78 8.78 ts-1.89 stress score (3.87) (3.67) (3.84) E months in 37.37 20.11 45.33 t=-3.46** 1st-asylum (34.47) (17.52) (37.51) countries E acculturative 0.54 0.42 0.59 _=-2.36* stress score (0.28) (0.25) (0.27) H amount of 618.91 372.83 735.47 _=-5.55*** monthly aid (302.15) (206.16) (269.76) received within 1st 18 months Sought job 70.2% 94.4% 59.0% chii=5.81* assistance ' Attitudes towards 1.63 1.48 1.70 t=-4.40*** government asst. (0.19) (0.19) (0.15) score * p < .05. ** p < .01. *** p < .001. Ngtg. Numbers in parentheses are the standard deviations. 75 Refugees who were off assistance after 18 months were more likely to (1) be Polish, (2) have professional- managerial or skilled labor occupational experience, (3) have several years of home education, (4) have smaller household size, (5) spend less than two years in first- asylum countries, (6) experience less acculturative stress, (7) receive less amount of assistance per month, (8) seek job assistance from various sources, and (9) have less favorable attitudes towards government assistance. The extent of the intercorrelations among the various predictor variables and the criterion variable is shown in the correlation matrix in Table 13. Since the variables are interrelated, the next step would be to conduct multivariate analysis such as discriminant analysis that incorporate the dependencies and also control the experimentwise error rate or the overall risk of Type I error associated with the use of multiple univariate tests. A discriminant analysis would help us understand better and determine the set of variables that best characterize the differences between those refugees who got off government assistance from those who did not by 18 months. The resulting discriminant function can be used to predict future membership in either the two groups, and identify refugees who need special assistance and interventions towards economic independence. Given the small sample size of 57 and the subsequent concern with weakened power of the statistical tests, it was 76 Table 13 t Va Still Being Being Being Agency on asst. Hmong Viet. Polish sponsor Still on asst. 1.00 Being Hmong 0.42* 1.00 Being Vietnamese 0.08 -0.52** 1.00 Being Polish -0.51** -0.50** -0.48** 1.00 Agency as sponsors -0.31 -0.51** -0.10 0.62** 1.00 Amer. individuals/ 0.14 0.05 0.16 -0.22 -0.54** groups as sponsors Family members/ 0.20 0.48** -0.05 -0.44** -0.52** rel. as sponsors Homeland occupation 0.35* 0.53** -0.04 -0.51** -0.45** Home education -0.49** -0.66** -0.07 0.74** 0.54** Arrival English -0.13 -0.27 0.20 0.08 0.10 Household size in 0.54** 0.51** -0.17 -0.35* -0.35* the lst 18 months Emigration stress 0.24 0.63** -0.28 -0.36* -0.36* Months in lst- 0.34* 0.84** -0.55** -0.31 -0.37* asylum countries Acculturative 0.29 -0.03 0.15 -0.13 -0.23 stress Monthly aid 0.57** 0.54** -0.20 -0.36* -0.43** received within lst 18 months Sought job asst. -0.36* -0.32 0.05 0.28 0.28 Attitudes towards 0.55** 0.25 0.24 -0.50** -0.35* government asst. Table 13 (cont'd). 77 Amer. individuals/ groups as sponsors Family members/ rel. as sponsors Homeland occupation Home education Arrival English Household size in the 1st 18 months Emigration stress Months in lst- asylum countries Acculturative stress Monthly aid received within lst 18 months Sought job asst. Attitudes towards government asst. Household size in the lst 18 months Emigration stress Months in 1st- asylum countries Amer . Family Home sponsor sponsor occu. -0.44** 0.18 -0023 0.07 0.30 -0034 -0017 0.28 0.38* 0.56** Emig. stress 1.00 -0.65** -0.25 0.26 0.16 0.37* -0.24 0.46** Months in 1.00 0.41* -0.42* -0.37* -0.48** -0.03 -0.47** 0.43** -0.49** Accult. stress Arrival English 1.00 -0.07 -0.14 -0.26 0.15 -0.13 Monthly aid 0.53** 0.59** 1.00 78 Table 13 (cont'd). Hhold Emig. Months Accult. Monthly size stress in stress aid country Acculturative 0.06 0.04 -0.09 1.00 stress Monthly aid 0.80** 0.09 0.50** 0.20 1.00 received within lst 18 months Sought job asst. -0.30 -0.15 -0.32 0.05 -0.30 Attitudes towards 0.29 0.10 0.16 0.22 0.46** government asst. Sought Govt. job asst. Asst. attitude Sought job asst. 1.00 Attitudes towards -0.15 1.00 government asst. two-tailed significance: * p < .01. ** p < .001. 79 deemed necessary to decrease the number of predictor variables and thus exclude one or two of the hypothesized variables. Occupational experience was highly related to years of education (z=-.65, 3:57, p<.001), increasing the threat of multicollinearity if included. Including the occupational background would necessitate adding two more dummy variables as compared to only one predictor variable for home education. Thus, it was decided to exclude home country occupational experience from the discriminant analysis. For the 57 cases and 13 selected predictors (including dummy-coded variables for ethnicity and type of sponsorship), assumptions of normality, linearity, and homogeneity of variance-covariance matrices were first evaluated. Examination of sample variances for the variables revealed no gross discrepancies between the two groups. The threat of multicollinearity was further protected by the computer program through checks of tolerance. Investigation of the z-scores for each group failed to show the existence of any outstanding outliers. The distributions of each of the variables were also examined for normality. No serious threat to discriminant analysis existed. A hierarchical discriminant function analysis was performed. Step 1 of the analysis utilized predictor variables tapping background information on the refugees upon arrival in the U.S.A.: (1) being Hmong, (2) being 80 Polish, (3) level of education in home country, (4) retrospective self-rating of English ability at the time of arrival, (5) total number of people in the household within the first 18-month period, and (6) length of stay in first asylum countries. There was a statistically significant separation of the two groups from the background predictors alone, E(6,49)=5.79, p<.001 with Wilks's Lambda at 0.59. Step 2 of the analysis added the following variables to the background predictors: (7) having Americans as sponsors, (8) having family members/relatives as sponsors, (9) emigration stress score, (10) acculturative stress score, (11) amount of assistance received within the first 18 months, (12) whether one sought job assistance or not, and (13) attitudes towards government assistance score. The resulting discriminant function using all 13 predictors improved with the value of Wilks's Lambda decreasing to 0.39, distributed as a 32(13,n=57)=44.92, p<.001. We can reject the null hypothesis of equality of group means. The eigenvalue of 1.57 is large enough and again indicates the goodness of the function with all 13 predictors. The canonical correlation B; of .78 shows the association between the discriminant scores and group membership. Since the overall function is statistically significant, the contributions of the individual variables to the differentiation of the two groups can be evaluated for significance. Table 14 shows the results of the discriminant analysis. It contains both the standardized 81 Table 14 - ~ : . if; u a, ,, a - . . -‘ x . . D "1!’! . -v in I . =' 2QC . t o Is Standardized Predictor Discriminant Structure Variable Function Coefficient Coefficient Being Hmong .05 .38 Being Polish -.67 -.47 Amer. individuals/grps as sponsors -.39 .10 Family members/rel. as sponsors -.69 .17 Home education .34 -.45 Arrival English skills -.22 -.11 Household size in the 1st 18 mos. .50 .51 Emigration stress .34 .21 Months in 1st-asylum countries .02 .30 Acculturative stress .41 .27 Monthly aid received within .25 .55 1st 18 months Sought job assistance -.38 -.30 Govt. assistance attitudes .49 .52 Canonical R .78 Eigen value 1.57 Wilks' lambda 0.39 cni2 (g;=13) 44.92 Equivalent E(13,42) 5.09 82 discriminant function coefficients and the structure coefficients. One indication of the importance of each predictor variables is the absolute magnitude of the standardized discriminant function coefficients. Such coefficients measure the unique contribution of any predictor variable while all other predictor variables are controlled for. Being Polish, having family members or relatives as sponsors, household size in the first 18 months and attitudes towards government assistance would appear to contribute the most to the differentiation of the two groups. However, the magnitude of the discriminant function coefficients can be misleading. Significant intercorrelations among these variables reduce the extent to which the coefficients and the subsequent importance of individual variables can be assessed. Actual signs of the coefficients are also affected by intercorrelations and are thus arbitrary. For example, being Hmong is significantly related to several of the predictor variables: so are total months in first asylum countries and amount of monthly aid received. Being Hmong, total months in first asylum countries, and monthly aid received have small discriminant function coefficients, much smaller than the coefficients for other variables such as having family members or relatives as sponsors or acculturative stress which might be actually less important in the discrimination. An alternative to the discriminant function coefficient 83 for assessing the relative contribution of a variable is the structure coefficient. Structure coefficients are the correlations between predictor variables and the discriminant function, and are also known as the discriminant loadings. While the discriminant function coefficients reflect the unique contribution of a predictor variable over and above that of the remaining predictor variables, the structure coefficients reflect the total contribution of any predictor variable to the discrimination without taking into consideration their relationship to or redundancy with the other predictor variables. The structure coefficients suggest that the primary predictors (with loadings of at least .50) for distinguishing the two groups are the household size, the amount of monthly aid received within the first 18 months, and attitudes towards government assistance. The correlations of .51 of household size, of .55 of monthly aid received, and .52 of government assistance attitudes with the discriminant function indicate that larger household size, higher amount of aid received, and favorable government assistance attitudes are associated with higher ”function values reflecting continued use of assistance. Group centroids were -1.79 for the group who got off aid within 18 months and 0.85 for the other group who were still on government assistance after 18 months. Two other predictor variables with loadings of at least 0.45 are whether one was Polish and home education. Being Polish and 84 higher education contribute to higher chances of being off government assistance. Using both the discriminant function coefficients and the structure coefficients indicate that three variables, being Polish, household size in the first 18 months, and attitudes towards government assistance are relatively more important than other predictor variables in differentiating the two groups. For the classification procedure, the unequal sample sizes of 38 (68%) and 18 (32%) were used to modify the probabilities with which cases were to be classified into groups. Results concerning the accuracy of the discriminant function in classifying the two groups are presented by the cross-tabulation in Table 15. With only the background information as predictors, 79% of the refugees were correctly classified. After the addition of the other variables, the function with all 13 predictors in the equation resulted in correct predictions being made for 91% of the cases. The improvement in classification, however, did not quite reach the criteria for statistical significance (p=.08). If our only alternative strategy is chance prediction, only 56% would be correctly classified by using prior probabilities of 0.68 for the group still dependent on government assistance and 0.32 for the smaller group of refugees off assistance, or if we assign every individual to the bigger group, only 68% would be correctly classified. 85 Table 15 3‘9‘19'! 01 0V‘,!h71 :=_ - a! ’ i ‘ 1 99! 2: Actual Group ---------------------------- Total Off GA Still on GA Step 1: Background predictors only: ethnicity, home education, household size, months in 1st-asylum countries, English skills upon arrival Off GA n 12 6 18 % 66.7% 33.3% 31.6% Still on GA n 6 33 39 % 15.4% 84.6% 68.4% Total n 18 39 57 % 31.6% 68.4% Percent of "grouped" cases correctly classified: 78.95% Step 2: Background and other predictors: emigration and acculturative stress, govt. assistance attitudes, type of sponsorship, monthly aid received, and whether one sought job assistance or not Off Ga n 15 3 18 % 83.3% 16.7% 32.1% Still on GA n 2 36 38 % 5.3% 94.7% 67.9% Total n 17 39 56 % 30.4% 69.6% Percent of "grouped" cases correctly classified: 91.07% 86 The obtained value of 91% correct prediction is a statistically significant improvement over the values of 56% (1=4-17. p<.001) or 68% (i=3.04, p<.01). BMW The other specific hypotheses of the existence of relationships of acculturative stress to the following variables: educational level, English proficiency skill, amount of emigration stress experienced, length of stay in refugee camps and type of sponsorship failed to reach statistical significance. Neither were there any significant relationships of the length of stay in refugee camps with favorable attitudes towards public assistance utilization, nor of favorable attitudes towards public assistance utilization with the utilization of refugee employment assistance services as hypothesized. CHAPTER 4 DISCUSSION W This survey was conducted to find out the factors that contribute to the utilization and long-term dependency on public assistance among refugees, and to discover sources of variation among the different refugee groups which relate to their economic adjustment. 1 'us nt Consistent with other studies of the economic conditions of refugees (Caplan, et al., 1985: Kelly, 1986: Stein, 1979: Stepick & Portes, 1986), for most of the respondents, life in the United States has meant living in poverty, dependency on government assistance, unemployment and underemployment. More than half (63%) live below poverty: 85% of the Hmong households, 72% of the Vietnamese households but only 28% of the Polish households. A high percentage of the respondents were still dependent on government assistance after 18 months: 95% of the Hmong respondents, 74% of the Vietnamese respondents and 33% of the Polish respondents. Even though resettlement programs and policies are geared towards helping refugees gained economic independence through employment, only 44% worked at some point within the first 18 months of their stay in U.S.A.: 51% were employed at the time of the interview. 87 88 Most (70%) have tried looking for jobs. Among the 25 respondents who worked within the 18-month time period, 72% were off government assistance at the end of 18 months. This appears to provide strong support to the current policy and practice of Refugee Services Office to get refugees into the work force as quickly as possible with the belief that such action would be most effective in moving them towards economic independence. However, being employed may not be adequate. Several of the refugees did find work in their first months in U.S.A. but often only on a sporadic basis. These jobs were also usually poorly—paid, high turnover, entry-level posts, often temporary in nature that were easily affected by downturns in the economy (e.g. janitorial jobs, dishwashing, assembly line worker, etc.). Several refugees reported being discouraged by the kind of jobs they found and eventually quit such jobs to concentrate on improving their skills through schooling, while utilizing government assistance and student loans or grants for their daily needs. Some refugees who did work still needed help due to low wages. Several of the educated and skilled refugees expressed frustrations at their inability to find jobs that paid decent wages commensurate to their skills. Most of the Polish respondents experienced downward occupational mobility, because of the apparent nontransferability of their training and credentials earned in Poland. A common complaint against the Refugee Services Office 89 heard from the more educated refugees was their perception that the office pressured them to take any job that came along, which were mainly entry-level menial jobs that pay low wages. To view having a job as the sole solution to welfare usage is thus misleading. What is clear is that having some work experience in the U.S does facilitate economic independence. ‘2 o e - s .- -qu . _-.,- 0V‘QLU‘I. .:~Ts .1 - Several variables were found to be significantly related to the long-term utilization of government assistance among refugees as hypothesized. Continued utilization of government assistance was most prevalent among Hmong refugees and least among Polish refugees. Long-term utilization of government assistance was associated with lower educational attainment in their home country, larger household sizes, longer stay in first-asylum countries, greater acculturative stress, more favorable attitudes towards government assistance and higher amounts of aid received within the first 18-month period. However, households receiving higher amounts of assistance were not necessarily economically better off. In fact, at the time of the interview, 84% of those on assistance lived below poverty as compared to only 17% of those off assistance. Refugees with professional-managerial occupational experience and/or those who approached various sources for job assistance were more likely to get off government assistance. Only three hypothesized predictors -- English 90 ability upon arrival, amount of emigration stress experienced, and the type of sponsorship -- failed to relate significantly to long-term utilization of government assistance in univariate analyses. However, based on the discriminant function coefficients, the type of sponsorship appeared to contribute to the differentiation between continued utilization of government assistance and economic independence when the other predictor variables were controlled for. Such finding might be misleading, however, as the correlation matrix showed high intercorrelations among the predictor variables, but not for the sponsorship dummy variables. From the discriminant loadings or the structure coefficients, the best predictors for long-term utilization of government assistance, regardless of the predictors' relationship with other predictor variables, appeared to be the amount of aid received, attitudes towards government assistance, household size within the first 18 months, being a Polish refugee, and home education. Attitudes towards government assistance have often been observed and reported by refugee resettlement workers to relate to their utilization of public assistance. A contribution of the current study to the body of literature on dependency on government assistance among refugees is its actual measurement and establishment of the relationship of such attitudes and the utilization of government assistance. Refugees who consider receiving government assistance as an 91 acceptable behavior in the U.S. society would tend to continue being on assistance. Household size as an important predictor is a finding consistent with several other studies (Bach & Carroll- Seguin, 1986: Rumbaut 8 Weeks, 1986). For refugee families that were already large upon arrival, cash assistance is a necessary safety net. Income from low-wage jobs would not be enough to support large families with several dependent children. Larger families also have greater family responsibilities with at least an adult required to remain at home with childcare responsibilities, decreasing the chances of having a second wage-earner contributing to the household's income. From anecdotal reports, it appeared that for refugee families with large numbers of children, welfare income is considered as a fact of life until the children are grown and can contribute to the family's income. Home education is another important predictor. Higher education results in less dependency on assistance as found in other studies (Samuel & Woloski, 1985: Bach 8 Carroll- Seguin, 1986: Stepick & Portes, 1986). A refugee with higher education is likely to possess greater intellectual and social resources that facilitate adjustment to a new environment. Poorly educated refugees, for example, not only face language problems but also inadequate literacy skills to deal with an overwhelming amount of paperwork required even for government assistance. Refugees with 92 higher levels of education and skills also are more likely to find higher wage jobs. The fact that 91% of the refugees attended school within the first few years of their stay in U.S.A. is a recognition of the need for language improvement, acquisition of new skills or training for new occupation, and retraining or certification of their credentials. W As hypothesized, there were significant ethnic differences on several variables that relate to their economic adjustment: (1) household size, (2) educational level, (3) occupational background, (4) emigration stress experienced, (5) length of stay in refugee camps, (6) attitudes towards public assistance utilization, and (7) dependency on public assistance. Only one hypothesis, that the amount of acculturative stress experienced is significantly related to ethnicity, was not confirmed. Hmong and Vietnamese refugees are more likely to continue being on welfare after 18 months than Polish refugees. Polish refugees are structurally in a better position to attain economic independence than the other ethnic groups. Polish respondents are highly educated with an average of 14.8 years of formal education. Most had professional or highly skilled occupations. They came from urban areas and stayed only an average of 3.2 months in refugee camps. They did stay an average of 22 months in first-asylum countries prior to resettling in U.S.A., and 93 suffered separation from loved ones and other emigration stress. They have the smallest household size with an average of only 2.6 members including 0.9 children. The English ability of Polish respondents at the time of the interview was also generally rated good. Vietnamese respondents have an average of 7.6 years of schooling. They were also mostly from urban areas and had work experiences in service occupations or in skilled jobs. Household size averaged 3.3 members with 1.3 dependent children. They experienced some amount of emigration stress mostly separation from loved ones, and stayed an average of 10.6 months in refugee camps. Their English ability was, on the average, rated fair by the interviewer. Hmong refugees, on the other hand, averaged only 2.7 years of formal education. They came from rural areas and were farmers and/or guerilla fighters in their home country. They experienced a great amount of emigration stress including separation from and the loss of loved ones, and close brushes with death. They also stayed the longest in refugee camps, averaging 76.5 months. They have the largest households with an average of 3.4 dependent children. They generally have poor English skills. All the refugees suffer from a great sense of loss (of material possessions, country, family members and familiar social and cultural norms, etc.) and the stresses of having to come to terms with the present and the immediate future in a new environment imposed by the need to survive. 94 However, the emigration experiences of most Vietnamese and Polish refugees can be described as fitting Kunz' (1973) anticipatory refugee description. They left their homelands in a more orderly way with some preparation. Kunz (1973) theorized that the prospects are good for their satisfactory adjustment to life in a new land. On the other hand, escape stories of the Hmong fit Kunz description of acute refugee movements. These refugees flee en masse, or in bursts of groups, emigrating out of fear and not from a rational desire to resettle elsewhere. Kunz suggested that this class of refugees might be expected to face more difficult problems of adjustment. Hmong refugees suffered the greatest emigration stress among the ethnic groups. They also spent the longest time in refugee camps with minimal control over their daily lives and destiny. Hmong refugees, too, had minimal prior exposure to Western culture, thus facing greater demands for cultural learning and social readjustment. Polish refugees are closest to the majority of Americans culturally and racially. It is expected that they would have greater ease in adjusting to the host culture than the other refugees whose cultures have greater disparity with the host culture. Resettlement in a modern Western society poses greater difficulties for the rural Hmong. Their skills and occupational experience which is limited to farming and warfare are less transferable to the new environment. They also have the least experience of wage employment, or of 95 operating in a society where literacy is a crucial skill. The Hmong refugees are structurally at a disadvantage and would be expected to have considerable difficulty in participating in the U.S. labor market. A common strategy to cope with the stress of migration to an unfamiliar environment is to cling as much to the familiar and change as little as possible (Scudder 8 Colson, 1982). Hmong and other less skilled refugees would be more likely to delay having to cope with the complexities of American work force by utilizing welfare and choosing to concentrate on getting basic literacy and other employment skills. Although there were no significant differences in the acculturative stress scores among the ethnic groups, other results indicate that Hmong respondents might be having the greatest difficulty in adapting to their new environment. A majority of the Hmong respondents would prefer to go back and live in their homeland again if the situation there improves, while Polish and Vietnamese respondents expressed little if any desire to go back and live in their home countries. Previous expectations of life in U.S.A. may both facilitate or discourage economic independence. The risk of creating dependency on assistance among refugees is increased if the benefits the refugees receive under the welfare system enable them to live at a standard higher than they have experienced in their homeland. Thus, such refugees would be quite satisfied with their economic state 96 and the amount of income they received from government assistance. On the other hand, expectations of a much better life and a higher standard of living can serve to motivate refugees to seek additional sources of income other than from government assistance. Polish refugees, for example, have generally moved from their initial place of residence to better housing in more desirable neighborhoods even if it means paying higher rent. Polish standards for housing appears much higher than the standards of the Hmong and Vietnamese refugees who were mostly clustered in low-income apartment complexes. Attitudes towards government assistance also differ among the three groups. Hmong and Vietnamese respondents were more likely than Polish respondents to believe in their entitlement to some form of assistance from the government and the acceptability of the use of such assistance. Clearly, the three ethnic groups are vastly different, not only on background characteristics and personal resources but on psycho-social variables as well. Such differences influence their utilization and long-term dependency on public assistance. MW 3.2111219 Originally, it was proposed that the sample for the study would be randomly chosen from the list of refugees in the Greater Lansing Area resettled withina four-year time period, from January, 1984 to the end of 1987. Due to the 97 high mobility of the sample, it was necessary after several months in the interview period, to get more potential respondents by including those resettled in 1988 and 1989. This increased the heterogeneity of the sample. Several of the later Vietnamese arrivals came through the Orderly Departure Program and were part of the Amerasian program. Their emigration history was vastly different from the other refugees who undertook great risks and escaped from their homelands. The whole sampling and tracking process was itself a lesson in patience and perseverance. Refugee Services Office did not have the staff nor is it necessarily their job to constantly follow up refugees and update their addresses or phone numbers. Refugees tend to move several times in the first few years of their stay here. Even the Vietnamese interpreter who assisted in the study moved three times in the time period he was on call for the study. In addition, several of the potential respondents were not easily accessible because of the nature of their work (i.e. as truck drivers who were frequently out of town) or because of having to work at odd shifts and attend school at the same time. Even the best sampling design would have produced an imperfect sampling outcome with this population. Due to the problems in getting enough cases for the study, no randomization procedure was done. Attempts were made to contact all possible respondents just to be able to interview 20 cases from each ethnic group. Thus, given the 98 small sample size, the diversity in the refugee sample, and the lack of random sampling, generalizability of the study's findings to the broader refugee population is limited. W Another methodological issue that arose in the study was in the validity of the responses. The applicability of survey techniques to non-Western cultures is still a subject of academic debates. Conducting a multi-cultural survey always poses greater conceptual and linguistic problems. First, several respondents in the sample had no previous experience answering in a survey. There was always the danger that respondents provided what they thought were the expected answers or answers that satisfy the interviewers, rather than their true opinions. Second, there might have been contextual differences in the interpretation of questions due to the different life experiences between the refugees and the researchers. Involving refugees in the construction and refining of the questionnaire, as well as in some of the interviews as interpreters were strategies employed to minimize problems in response validity. Other strategies utilized were simplifying the questions as much as possible, doing additional probing during the interview process, and finally excluding some questionnaire items where researchers doubted the validity of the responses. Another issue that came up in the study was how to deal with community suspicion and local politics. At the start of the study, established refugee leaders such as officers 99 of organized associations and refugee caseworkers were approached and asked to comment, provide suggestions and assist in the project. In the course of the interviews, however, the researchers became aware of local political "going-ans" within the different refugee communities. For instance, a member of one ethnic group (not a potential respondent) who considered herself knowledgeable of the issues in the refugee community, took it upon herself to call one of the researchers to determine the real purpose of the study and to express her views and personal opinions on specific individuals. Information such as negative opinions towards certain organizations or individuals in the organizations were used to modify the approach, introduction of the researchers and description of the project to potential respondents. Suspicion that survey information would be used by the welfare system of the state of Michigan to discredit those who were still on assistance had to be allayed as well. The interpreters provided great assistance with their knowledge of the "ins and outs" of the ethnic group, and with the culturally appropriate ways to deal with the respondents. EDLQI§_B§§§§IED This study hoped to add to the increasing literature on refugee issues. It answered some questions regarding the factors that affect continued utilization of public assistance. However, the validity of the discriminant function that was generated in the present study needs to be 100 established. The percentage of cases classified correctly by the discriminant function is an inflated estimate of the true performance in the population. The discriminant function should be cross-validated and applied in a new sample to determine the stability of the weights and the actual predictive accuracy of the equation beyond the sample from which it was derived. As was found out in the study, refugees arrived in different waves. Even within an ethnic group, different sets of refugees arrived at certain times such as those who came through the Orderly Departure Program. Their different emigration experiences possibly affected the results of the study. For a more useful research, a longitudinal approach is recommended, tracing a refugee household over time. Such a study would require painstaking care and perseverance, and enough financial support to execute. Dependency on government assistance has been shown to relate to differences among the refugee groups on various variables. The small sample sizes for each group, however, preclude coming up with meaningful analysis of the relationship of these variables to dependency on government assistance within each group. A similar study is recommended but with larger samples for each ethnic group to be able to determine what variables are predictive of long- term utilization of assistance for each ethnic group, and compare such variables across the different groups. Most studies on refugees have been of the very small 101 percentage of refugees resettled in industrialized countries. More studies are needed of refugees in first- asylum countries and of those who were repatriated back to their homelands. These would also require tremendous resources and perseverance but are necessary if we aim to improve our existing knowledge and understanding of the refugee issue, and eventually help lessen this problem that affects millions of people. 11121152811285 The results of this study provide several directions for intervention and policy-changes. First, there is the challenge involved in designing effective resettlement policies, programs and services that recognize differences among and within ethnic groups, and are thus tailored to specific groups or needs. Services which enhance the educational and job-related skills of refugees should continue to receive major emphasis. Although the emphasis on getting refugees gainfully employed as soon as possible should be continued, provisions must also be made to ensure that refugees be adequately trained and prepared with job skills for current and future job markets beyond the entry-level low-wage posts. Refugees should also be encouraged to delay long-term training until they understand the American occupational market better and have some work experience in the U.S.A. Innovative solutions must also be arrived at to address the dependency issue. Economic independence can be 102 facilitated if refugees are able to utilize existing skills. For example, an agricultural cooperative might be created for more rural refugees as an alternative to entering the labor force and competing for industrial jobs in an urban setting. Another area for intervention is to come up with strategies to counteract current favorable attitudes towards the utilization of government assistance. Implementation of such strategies would have to start prior to their resettlement in U.S.A. such as in the English or cultural orientation classes in refugee camps. Expectations have to be clarified and misconceptions corrected. Childcare and other services to meet the special needs of dependent children of refugees must ideally be made available to enable employable adults in the refugee household to attend English and job-skills training, or join in the labor force. Culturally appropriate health and family planning services to provide more choices in terms of family size to some refugees might also be necessary. Refugee resettlement policy should be based on a comprehensive multidimensional and long-term understanding of the diverse historical and cultural backgrounds of the refugee population, and the complex processes of forced migration and adaptation to new environments. However, the refugee problem is not mainly having to meet the immediate physical survival needs of a group of people. There is a danger of correlating admission flows to costs, and 103 perceiving refugees as burdens to the community. Too often overlooked are the refugees' assets. Refugees infuse vitality into their new communities, and make permanent cultural, social, and economic contributions. There are several million refugees in the world today, most of them from Third World countries seeking refuge in another Third world country. Most refugees will not resettle in industrialized countries or return home soon. Moreover, the problems still continue in spite of the end of the Cold War which has made possible the settlement of some conflicts and the democratization of authoritarian nations. Refugees are certainly among the world's most disadvantaged people. Although it is beyond the scope of this study, refugee issues should be looked at as part of a larger, more complex picture. The refugee problem should be linked with foreign policy. Ways must be found to prevent people from becoming refugees in the first place. This means addressing the deeply rooted causes of the refugee problem: political oppression: violence and armed conflicts: violations and abuse of human rights: ethnic and religious conflicts: the great disparity between the rich and poor nations, and between the rich and the poor in a nation: and the grinding poverty brought on by various natural and human causes. This means addressing both the economic and humanitarian aspects of mass migration at their roots, in the countries of origin as well as the policies and practices of powerful nations. 104 There are no easy solutions to the global and domestic refugee problems. Whatever the political debates, the policies and controversies that determine their states, the ultimate issue is that these are real individuals whose basic right to a decent life is at stake. APPENDICES Appendix A SOUTHEAST ASIAN AND EASTERN EUROPEAN REFUGEES SURVEY INTERVIEW SCHEDULE Respondent Code No.: Interviewer: Date: Time Started: Time Ended: Hello. My name is . I am doing a study on refugees resettled here in Lansing. Hopefully, the study will be useful to the Refugee Services Office and other government agencies as they continue to prepare and help refugees resettle in U.S.A. The interview will take around one hour. This interview is CONFIDENTIAL. Your answers will be kept secret. Nobody aside from me will know about you or link you with your answers. You may also choose NOT to answer any of the questions if you do not want to. Do you have any question? I will now ask you first for some general information. GENERAL INFORMATION Sex of respondent. 1 FEMALE 2 MALE What is your nationality? 1 HMONG 2 VIETNAMESE 3 POLISH How old are you? Where were you born? Where did you live before you left your home country? 105 10. 11. 12. 13. 106 Did you live in a city, a town or out in the country? 1 CITY 2 TOWN 3 COUNTRY What is your religion? NO RELIGION ANCESTOR WORSHIP BUDDHISM CONFUCIANISM CATHOLICISM MOHAMMEDANISM PROTESTANTISM OTHERS (SPECIFY) @QQMhUNH Are you single or married? (Probe.) SINGLE MARRIED SEPARATED DIVORCED WIDOWED OTHERS (SPECIFY) mUI-huNH How many years did you go to school in your home country? Did you graduate from school? 0 NO (Go to no. 12) 1 YES Please state the 1 A FAMILY MEMBER (SPECIFY RELATION 2 A RELATIVE (SPECIFY RELATION u BACKGROUND AN AMERICAN FAMILY AN AGENCY (SPECIFY) A FRIEND/ACQUAINTANCE OF THE SAME ETHNIC A CHURCH-BASED ORGANIZATION OR CONGREGATION 4mm:- OTHERS (SPECIFY) 107 HOUSEHOLD INFORMATION Do you live alone here? 0 NO 1 YES (Go to the next section) Who live with you? Your spouse, son, daughter, friend, etc.? (Fill up the first column of the table. Then ask the next set of questions and fill up the rest of the table.) a. Sex : 1 = FEMALE 2 = MALE b. How old is ? c. Is working and/or studying? Work status: 1 = NOT WORKING 2 = WORKING PART-TIME 3 = WORKING FULL-TIME d. If working, what is his/her job? e. How long has lived with you? RELATIONSHIP WORK LENGTH OF TO SEX AGE STATUS JOB(S) TIME LIVING RESPONDENT WITH R. 108 ENGLISH ABILITY When you first arrived in U.S., how was your ability to 1-1 understand Americans talk in English 1 VERY POOR 2 POOR 3 FAIR 4 GOOD 5 EXCELLENT 1-2 have others understand your spoken English VERY POOR POOR FAIR GOOD EXCELLENT Ul-bMNH 1-3 read in English VERY POOR POOR FAIR GOOD EXCELLENT UlbUNH 1-4 write in English VERY POOR POOR FAIR GOOD EXCELLENT 0101.010!" Did you have English training either in a refugee camp and/or here in U.S.A.? 0 NO (Go to no. 4) 1 YES 109 4. What were the English classes you took (ESL, Vocational English, etc.)? How long did you take each one? 5. How is your ability at present to 5-1 understand Americans talk in English 1 VERY POOR 2 POOR 3 FAIR 4 GOOD 5 EXCELLENT 5-2 have others understand your spoken English 1 VERY POOR 2 POOR 3 FAIR 4 GOOD 5 EXCELLENT 5-3 read in English VERY POOR POOR FAIR GOOD EXCELLENT U'thND-I' 5-4 write in English VERY POOR POOR FAIR GOOD EXCELLENT UluhuNH 1. 5. 110 I will now ask you questions on your experiences in leaving the country. EMIGRATION HISTORY How did you escape or leave your country? Did you stay in another country before you came here? 0 NO (Go to the next section) 1 YES Did you stay in a refugee camp before you came here? 0 NO (Go to no. 5) 1 YES (Go to no. 4 and then to the next section) What were these refugee camps? Where are they located? How long did you stay in each camp? What were the countries you went to before coming here? How long did you stay in each country? | HOST COUNTRY | NO. OF MONTHS | 111 EMIGRATION STRESS SCALE The next set of questions deals with events that might have happened to you IMMEDIATELY BEFORE and DURING YOUR ESCAPE or DEPARTURE FROM YOUR COUNTRY, and in YOUR TRAVEL TO U.S. For each of the sentences, please answer yes or no. 10. 11. 12. A member of your own family (spouse, son, daughter, father, mother, brother, sister) was left behind. Another close relative (grandfather, grandmother, uncle, aunt, cousin) whom you wanted to take with you was left behind. A member of your own family (spouse, son, daughter, father, mother, brother, sister) died. Another close relative (grandfather, grandmother, uncle, aunt, cousin) died. Other people shot at you or tried to kill you in your escape from your homeland. You left your homeland in a boat and spent several days at sea. You walked for days through jungles and/or mined fields. You swam across a river. You were separated from people (family members, relatives, etc.) whom you felt very close to while escaping, in a refugee camp or moving to the U.S. You had new problems with your health. There was a constant fear of being captured and/or of death while you were escaping. You were hungry much of the time. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 112 You saw other people beaten up. Other people threatened you or tried to harm you in the camp or during other parts of your move to U.S. You were completely alone while escaping, in a camp or during other parts of your move to the U.S. You lost a lot of weight (more than 5 kilos). You slept poorly and significantly worse than before you left your homeland. You thought of harming or killing yourself. You felt as though other people controlled most of your actions. You saw other people killed. You had to lie and deceive people to escape. (Did respondent stay in a refugee camp before coming here?) 113 EMPLOYMENT HISTORY AND ACTIVITIES IN THE FIRST 18 MONTHS How did you usually earn money in your country? (What was your main job?) How did you earn money in the first asylum country or refugee camp before coming here? Did you have job training in a refugee camp or here? 0 NO (Go to no. 5) 1 YES What were the vocational training classes you took (ex. automechanic, etc.)? How long did you take each one? 114 5. What did you do in your first year and a half here in the U.S.A.? Did you WORK FULL-TIME WORK PART-TIME ATTEND ESL CLASSES ATTEND JOB/VOCATIONAL TRAINING PROGRAM GO TO COLLEGE GO TO HIGH SCHOOL LOOK FOR JOBS BUT DIDN'T FIND ONE DID NOT WORK OR GO TO SCHOOL OTHERS \OQQGMhUNP II II II II II II I II II lst month: 18th month: (month/year) (month/year) 6. Are you working now? 0 NO 1 YES (Go to no. 9) 7. Have you worked here? 0 NO 1 YES 10. 115 Why are you not working now? (or) Why haven’t you here before? (Check as many as applicable.) worked 1 ATTENDING SCHOOL (SPECIFY ) 2 KEEPING HOUSE 3 POOR HEALTH 4 POOR ENGLISH 5 CAN LIVE ON WHAT THE GOVERNMENT GIVES US 6 HAVE MEANS OF SUPPORT NOT FROM THE GOVERNMENT 7 DIFFICULT TO FIND JOBS 8 DISCOURAGED TO LOOK FOR JOBS FROM REPEATED FAILURES TO FIND THEM 9 OTHERS (SPECIFY) Did you try to go to for help in getting a job? NO YES 9-1 Refugee Services Office 0 1 9-2 government employment assistance 0 1 service agency 9-3 private employment assistance 0 1 service agency 9—4 friend(s) 0 1 9-5 relative(s) O 1 9-6 your Sponsor(s) O 1 9-7 others (specify) 0 1 How many jobs have you had since you arrived here? 116 11. What were these jobs? When did you have each job? How much was the usual amount of money you earned from each job? Who helped you get these jobs? (If the answer to no. 10 is more than five jobs, limit the jobs to those the respondent had until the time they stopped receiving government assistance. For source of help, use the following coding system: THE REFUGEE SERVICES OFFICE A GOVERNMENT EMPLOYMENT SERVICE AGENCY A PRIVATE EMPLOYMENT SERVICE AGENCY FRIEND/ACQUAINTANCE(S) RELATIVE (S) SPONSOR(S) NOBODY OTHERS) mQO)U|-wal-' POSITION EMPLOYMENT INCOME OF HELP 117 12. If you have to look for a job now or sometime in the future, where or to whom would you go to for help first? the second? and the third? GOV’T EMPLOYMENT ASSISTANCE SERVICE AGENCY 3. PRIVATE EMPLOYMENT ASSISTANCE SERVICE AGENCY 118 I will now ask you questions on your family income. G. FAMILY INCOME AND UTILIZATION OF PUBLIC ASSISTANCE 1. How much money does your family earn or receive each month? 2. Have you received money or any other help from the government like food stamps? 0 NO (Go to no. 5) 1 YES 3. Does/did your family receive from the government? When did you start receiving ? When did you stop receiving ? How much do/did you receive? (Go through each one and fill up the table.) 4. If you receive(d) money from the government, what kind(s) of cash assistance is(was) it? AID TO FAMILIES WITH DEPENDENT CHILDREN (AFDC) SUPPLEMENTAL SECURITY INCOME (SSI) REFUGEE CASH ASSISTANCE (RCA) GENERAL ASSISTANCE (GA) OTHERS (SPECIFY) U'IthH 5. Have you received money or any other help that is not from the government? 0 NO (Go to no. 7) 1 YES 119 6. Does/did your family receive ? From whom? How often do/did you receive ? How much do/did you receive? (Go through each one and fill up the table.) | ASSISTANCE IREC’DI SOURCE | FREQUENCY | AMOUNT | 1. MONEY 7. Do you send money regularly to your family or relatives in your homeland? 0 NO 1 YES 120 Next, I would like to ask what you think and believe about the help you get from the government. EXPECTATIONS AND ATTITUDES TOWARDS GOVERNMENT ASSISTANCE Where did you first learn of government assistance (welfare)? hUNH IN MY HOME COUNTRY IN A REFUGEE CAMP OR IST ASYLUM COUNTRY HERE IN THE U.S.A. OTHERS (SPECIFY) How did you learn about government assistance (welfare)? \lmmuhUNH H C p. H H FROM RELATIVE(S) FROM FRIEND(S) FROM MY OWN COUNTRY FROM TEACHER(S)/WORKER(S) AT THE REFUGEE CAMP FROM U.S. OFFICIAL/STAFF AT THE REFUGEE CAMP FROM THE REFUGEE SERVICES OFFICE STAFF HERE FROM MY SPONSOR(S) OTHERS (SPECIFY) now read you several sentences. Please answer whether you agree or disagree with each statement. DISAGREE AGREE The U.S. government has a 1 2 responsibility to help us. Everyone should earn their living 1 2 by working. Receiving money from the 1 2 government is okay. I would rather take a job that 1 2 pays low wages than depend on money from the government. A refugee should learn English 1 2 first before looking for a job. I would prefer government 1 2 assistance (welfare) to work. Receiving food stamps from the 1 2 government is okay. As long as the government would 1 give me food stamps or money, I would receive it. 121 One of the most important duties of every person is to earn a living. I would accept government assistance (welfare) even if I don't need it. I would like to stay on government assistance (welfare) as long as I could. A refugee should immediately look for a job upon arriving here. I would starve before I would accept money from the government. Please think back now to your first year and a half year and answer yes or no. 3-14 I expected to receive money from the government when I arrived here. It was difficult for me to get a job that paid enough money. As a refugee, I felt entitled to government assistance. I was glad to accept government assistance (welfare). Even without cash assistance from the government, my family would still be able to live well. I accepted government assistance (welfare) without shame. The income we earned was not enough for our normal expenses. I felt forced to receive government assistance (welfare). 122 ACCULTURATIVE STRESS The next set of questions are on problems which you may or may not have encountered here. Please problem as to how serious it was for you. 0 - NOT SERIOUS 1 - SOMEWHAT SERIOUS 2 - VERY SERIOUS Learning to speak and understand English. Difficulty in finding a job. Understanding the behavior of Americans. Finding adequate housing. Difficulty raising children because of confusion over American and your own cultural values and behaviors. Knowing how to take care of daily needs. Not being able to communicate with family members or other loved ones in your homeland. Not having enough money for basic necessities. Americans not understanding your cultural ways. Having to live in an undesirable or unsafe neighborhood. Having to get used to a different climate. A loss of your social status or prestige. Not understanding rules and regulations. Feeling homesick. rate each NS SS VS 123 NS SS 1-15 Difficulty with your children being 0 1 in American schools or in relating to the American educational system. 1-16 Being unsure of how to act in public 0 1 places. 1-17 Difficulty in dealing with 0 1 government agencies 1-18 Being/looking different from O 1 other people. 1-19 Not knowing what was or is happening 0 1 in your homeland. 1-20 Feeling that Americans are 0 1 unfriendly or prejudiced toward you. 1-21 Experiencing a big change in your 0 1 role in your family. 1-22 Difficulty in practising your 0 1 religion. 1-23 Conflict over American and your own 0 1 cultural ways of behavior. 1-24 Being separated or isolated from 0 1 family members or other loved ones. 1-25 Difficulty obtaining familiar food. 0 1 1-26 Feeling unable to do anything about 0 1 the events in your homeland. 1-27 Feeling afraid or anxious when you 0 1 meet Americans. 2. Do you try to bring other family members out of your country? 0 NO 1 YES 3. Do you think about going back to your country? 0 NO 1 YES, BUT ONLY FOR A VISIT 2 YES, I WOULD LIKE TO GO BACK AND LIVE IN MY COUNTRY AGAIN IF THE SITUATION THERE CHANGES VS 124 J. Overall, how have you adjusted to life here in U.S.? K. What suggestions or comments do you have about the policies or services given to refugees? Thank you for your time and patience! Is there anything else you would like to say? II. INTERVIEWER’S RATING OF RESPONDENT’S ENGLISH ABILITY 1. How well did the household head understand the questions in English? 1 2. How well English? 1 2 UNABLE TO UNDERSTAND THE QUESTIONS IN ENGLISH AT ALL OR UNDERSTOOD ONLY LESS THAN 10 QUESTIONS UNDERSTOOD SOME QUESTIONS WITH SEVERAL REPETITIONS OR REPHRASING UNDERSTOOD MOST QUESTIONS WITH SEVERAL REPETITIONS OR REPHRASING UNDERSTOOD MOST QUESTIONS WITH SOME REPETITIONS OR REPHRASING UNDERSTOOD MOST QUESTIONS WITH HARDLY ANY REPETITIONS OR REPHRASING did the household head answer the questions in UNABLE TO ANSWER IN ENGLISH AT ALL OR ANSWERED ONLY LESS THAN 10 ITEMS ANSWERED ALMOST ALL THE TIME IN FRAGMENTED ENGLISH AND/OR WITH SEVERAL ERRORS ANSWERED HALF OF THE TIME IN FRAGMENTED ENGLISH AND/OR WITH SEVERAL ERRORS ANSWERED MOST OF THE TIME IN APPROPRIATE ENGLISH WITH SOME ERRORS ANSWERED EASILY IN APPROPRIATE ENGLISH WITH VERY FEW ERRORS Note: 125 The items below were removed from the final scales due to very low corrected item-total correlations. I. Attitudes Towards Government Assistance Scale II. 1-12 I would rather take a job that pays low wages than depend on money from the government. A refugee should learn English first before looking for a job. I would prefer government assistance (welfare) to work. One of the most important duties of every person is to earn a living. A refugee should immediately look for a job upon arriving here. I would starve before I would accept money from the government. I expected to receive money from the government when I arrived here. It was difficult for me to get a job that paid enough money. The income we earned was not enough for our normal expenses. .I felt forced to receive government assistance (welfare). Acculturative Stress Not being able to communicate with family members or other loved ones in your homeland. A loss of your social status or prestige. DISAGREE AGREE 126 Difficulty with your children being in American schools or in relating to the American educational system. Feeling that Americans are unfriendly or prejudiced toward you. Difficulty in practising your religion. Difficulty obtaining familiar food. Appendix B REFUGEE CAMPS WHERE RESPONDENTS STAYED BEFORE RESETTLEMENT TO U.S.A. Refugee Camp n % Phanat Nikhom, Thailand 21 36.8 Ban Vinai, Thailand 20 35.1 Refugee Processing 17 29.8 Center, Philippines Pulau Bidong, Malaysia 5 8.8 Galang, Indonesia 3 5.3 Latina, Italy 4 7.0 Others 4 7.0 No camp experience 13 22.8 127 Appendix C FIRST-ASYLUM COUNTRIES WHERE RESPONDENTS STAYED BEFORE RESETTLEMENT TO U.S.A. lst-Asylum Country n % Southeast Asia Thailand 23 40.4 Philippines 17 29.8 Malaysia 5 8.8 Indonesia 3 5.3 Cambodia 1 1.8 Western Europe Greece 4 7.0 Holland 4 7.0 Italy 4 7.0 West Germany 3 5.3 Austria 1 1.8 France 1 1.8 Spain 1 1.8 Eastern Europe 1 1.8 128 Appendix D OCCUPATION OF REFUGEES IN THEIR HOME COUNTRY Ethnic Group Occupation ........................... Hmong Viet. Polish Total (n=20) (n=19) (n=18) (N=57) Professional specialty occupations Engineer 0% 0% 11.1% 3.5% Physical scientist 0% 0% 11.1% 3.5% Social worker/counselor 0% 0% 11.1% 3.5% Social scientist 0% 0% 5.6% 1.8% Teacher 0% 0% 5.6% 1.8% Computer programmer 0% 0% 5.6% 1.8% Managerial occupations 0% 0% 5.6% 1.8% Clerical & administrative support 0% 5.3% 11.1% 5.3% Service occupations Restaurant cook 0% 5.3% 0% 1.8% Waiter 0% 5.3% 0% 1.8% Hairdresser 0% 5.3% 0% 1.8% Entertainer 0% 5.3% 0% 1.8% Precision production, craft, and repair occupations Mechanic 0% 21.1% 0% 7.0% Construction worker/carpenter 0% 0% 11.1% 3.5% Plumber 0% 0% 5.6% 1.8% Electrician 0% 5.3% 0% 1.8% 129 130 Ethnic Group Occupation --------------------------- Hmong Viet. Polish Total (n=20) (n=19) (n=18) (N=57) Farming and fishing 80.0% 10.5% 0% 31.6% occupations Operators and laborers Train supervisor 0% 0% 5.6% 1.8% Factory worker 0% 0% 5.6% 1.8% Machine operator 0% 0% 5.6% 1.8% Bus conductor 0% 5.3% 0% 1.8% Ship navigator/seaman 0% 0% 11.1% 3.5% Military Soldier/guerilla fighter 25.0% 5.3% 0% 10.5% Military officer 5.0% 10.5% 0% 5.3% Small business operator/vendor 0% 26.3% 5.6% 10.5% Student 5.0% 10.5% 5.6% 7.0% Appendix E JOBS HELD BY REFUGEES IN U.S.A. Ethnic Group Hmong Viet. Polish Total (n=20) (n=19) (n=18) (N=57) Job in U.S.A. Professional specialty occupations Engineer 0% 0% 5.6% 1.8% Physical scientist 0% 0% 11.1% 3.5% Teacher 0% 0% 5.6% 1.8% Computer programmer 0% 0% 5.6% 1.8% Accountant/bookkeeper 0% 0% 5.6% 1.8% Service occupations Food preparation and service 5.0% 42.1% 5.6% 17.5% Cleaning except household 10.0% 5.3% 11.1% 8.8% Private household occupations 0% 0% 5.6% 1.8% Nurse aide 0% 0% 5.6% 1.8% Massage therapist 0% 0% 5.6% 1.8% Precision production, craft, and repair occupations Welder 10.0% 0% 5.6% 5.3% Construction worker/carpenter 0% 0% 16.7% 5.3% Electrician 0% 0% 5.6% 1.8% Technician 0% 0% 5.6% 1.8% Technical draftsman 0% 5.3% 0% 1.8% Auto mechanic 0% 0% 5.6% 1.8% 131 132 Ethnic Group Hmong Viet. Polish Total (n=20) (n=19) (n=18) (N=57) Job in U.S.A. Sales representative 0% 0% 11.1% 3.5% Administrative support 10.0% 5.3% 5.6% 7.0% including clerks Operators, fabricators and laborers Factory worker/assembler 10.0% 10.5% 16.7% 12.3% Machine operator 5.0% 0% 11.1% 5.3% Seamstress/tailor 5.0% 5.3% 0% 3.5% Warehouse laborer 5.0% 0% 5.6% 3.5% Truck driver 0% 0% 11.1% 3.5% Cab driver 0% 0% 5.6% 1.8% Delivery person 0% 10.5% 0% 3.5% Grocery bagboy 0% 5.3% 0% 1.8% Vendor 0% 5.3% 5.6% 3.5% No job 60.0% 36.8% 16.7% 38.6% REFERENCES REFERENCES Aylesworth, L.S. 8 Ossorio, P.G. (1983). Refugees: cultural displacement and its effects. Advenees in DQSQIIESiYQ Psychology, 1, 45-93. Bach, R.L. 8 Carroll-Sequin, R. (1986). Labor force participation, household composition and sponsorship among Southeast Asian refugees. International Mignatien Review, gg, 381-405. Berry, J.W., Kim, U., Minde, T. 8 Mok, D. (1987). Comparative studies of acculturative stress. Internenienal Mignenien Review, 2;, 491-511. Berry, J.W. (1986). The acculturation process and refugee behavior. In C.L. Williams 8 J. Westermeyer (Eds.), Rafngaa mental naalnn in.raaaftlanann nannnriaa (PP- 25-37). Washington: Hemisphere. Bousquet, G. (1987). Living in a state of limbo: a case study of Vietnamese refugees in Hongkong camps. In S.M. Morgan 8 E. Colson (Eds.), People in unneave; (pp. 2-53). New York: Center for Migration Studies. Caplan, N., Whitmore, J. K., 8 Bui, Q. L. (1985). Economic self-sufficiency among recently-arrived refugees from Southeast Asia. Eaanania anlnak H_A. Third Qnartar 60- -63. ----- (1985)- Sannnaaat Aaiana aalfzanffiaianax annaxa fine; zenenn (Report for the Office of Refugee Resettlement, U.S. Department of Health and Human Services). Center for Applied Linguistics (CAL, 1988). Yenng egnln nefugee enrvey: Fine; ISEQIL (Report for the Bureau for Refugee Programs, U.S. Department of State). Washington, DC: author. Cohon, J. D. (1981). Psychological adaptation and dysfunction among refugees. International Migratian BaYiaY 15(1). 255- 275- Cox, D. (1985). Welfare services for migrants, can they be better Planned? Internaninnal Migranian. 23. 73-94- 133 134 Desbarats, J. (1986). Ethnic differences in adaptation: Sino-Vietnamese refugees in the United States. Internatisnal Migratinn BsYisE. 29. 405-427- Fass, S. (1986). Innovation in the struggles for self- reliance: the Hmong experience in the United States. Intannatianal Migratian BsYisY. 29. 351- 380. Finnan, C.R. (1981). Occupational assimilation of refugees. WI 1.5: 292-309 0 Goodwin-Gill, G. (1987, December). North American evolution of the right of asylum. Befngeee, 15, pp. 36-39. Habarad, J. (1987). Refugees and the structure of opportunity: Transitional adjustments to aid among U.S. resettled Lao Iu Mien, 1980 - 1985. In S.M. Morgan 8 E. Colson (Eds.), Peenle in npneeyel (pp. 66-87). New York: Center for Migration Studies. Keely, C.B. (1983). Current status of U. S. immigration and refugee policy. In M. M. Kritz (Ed. ), QeSL Inmigzati_n and refuses 221192; filnnal ans Canastis issnss (PP- 339-359). Lex1ngton: D.C. Heath. Kelly, G.P. (1986). Coping with America: Refugees from Vietnam, Cambodia, and Laos in the 1970s and 1980s. Annals sf tns Ansrisan,Assssiatisn sf Balitisal and §Q§i§l SCie C 5, $81, 138-149. Kerpen, K.S. (1985, Winter). Refugees on welfare: is the dependency rate really a problem? £22112 Welfare, pp. 21-25. Krolewski, M.A. (1988). The Eastern European refugee community-1988 (Metropolitan Detroit), unpublished manuscript. Kunz, E. F. (1981L Exile and resettlement: refugee theory. Intsrnatianal Eigratisn BsYisY.1s. 42-51. Kunz, E.F. (1973). The refugee in flight: Kinetic models and forms of displacement. Intannatisnal Migratian Bsxisu. l. 125-146- Lacey, M. (1987). A case study in international refugee policy: Lowland Lao refugees. In S.M. Morgan 8 E. Colson (Ede-). Esanls in.unhsaYal (PP- 17-33). New York: Center for Migration Studies. Lin, K.M. (1986). Psychopathology and social disruption in refugees. In C.L. Williams 8 J. Westermeyer (Eds.), Esfngss mental health in rsssttlsnsnt snantriss (PP- 61-73). Washington: Hemisphere. 135 Lin, K.M. 8 Masuda, M. (1983). Impact of the refugee experience: Mental health issues of the Southeast Asians. In R. Morales (Ed.), Bridging enltnnee; Sentneast Asian refnsees in Aneriea (PP- 32-52)- Los Angeles: Asian American Community Mental Health Training Center. Liu, W.T. 8 Cheung, F. (1985). Research concerns associated with the study of Southeast Asian refugees. In T. C. Owen (Ed ). Sentneast Asian mental health; Treatmenti nzeyenniene eervices, tneining eng neeearcn (DHHS Publication No. ADM-139, pp. 487- 516). Rockville, MD: National Institute of Mental Health. Mollica, R.F., Wyshak, G., 8 Lavelle, J. (1987). The psycho-social impact of war trauma and torture on Southeast Asian refugees. Anenieen EQBIEQl 9f Eexshiatzx. iii. 1567-1572- Montgomery, R. (1986). The economic adaptation of Vietnamese refugees in Alberta: 1979-84. Innennenienel Migratien. 2A. 749-767- Moore, J. (1987a, July). Refugees and foreign policy: Immediate needs and durable solutions. Denennnen; 9; State snlletin. PP- 70-74. Moore, J. (1987b, November). Proposed refugee admissions for FY 1988. Department sf State talletin. PP- 47-52- Murphy, H.B.M. (1955). The extent of the problem. In H.B.M. Murphy (Ed.), Elign; end resettlemenn. Paris: UNESCO. Nicassio, P.M., Solomon, G.S., Guest, S.S. 8 McCullough, J.E. (1986). Emigration stress and language proficiency as correlates of depression in a sample of Southeast Asian refugees. Innexnenignel Jgnznel g; Sesial EsstiatrY. 32. 22-28- Nicassio, P.M. (1985). The psychosocial adjustment of the Southeast Asian refugee: an overview of empirical findings and theoretical models. Jennnei ef Qnoss- snltnral EsYenelng. 1a. 153-173- Nicassio, P.M. 8 Pate, J.K. (1984). An analysis of problems of resettlement of the Indochinese refugees in the United States. £29111 Eeyeniegny, 12, 135-141. Nicassio, P.M. (1983). Psychosocial correlates of alienation: Study of a sample of Indochinese refugees. learnal sf tresszenltnral RsYeneleg . 11. 337-351. 136 North, D.S., Lewin, L.S. 8 Wagner, J.R. (1982). Kaleidessenea The resettlement of refngees in the United States hY the YelnntarY ageneies (Report for the Bureau for Refugee Programs, U.S. Department of State). Office of Refugee Resettlement, U.S. Department of Health and Human Services (ORR, 1991). Repgp; pg Qengpeeei Refngee Resettlement Rregraml 9ffiee of Refngee BSSSLLISESRL- Office of Refugee Resettlement, U.S. Department of Health and Human Services (ORR, 1990). Report pg Congpese, Refngee Resettlement Programl.9ffise of Refngee Reseptienenp. Pecora, P.J. 8 Fraser, M.W. (1985). The social support networks of Indochinese refugees. Joupnai g; Soeigiogy ans Seeial Eelfare. 12(4). 817-849- Ranard, D.A. (1988, November). The Hmong. In Anepiee; RersnestiYes en Refugee Resettlement. l. 1-12- Refugee Policy Group (RPG, 1987). ERLBES gipeepigne in pne R,S. Refingee Reeettiement Rppgpen (Report for the Office of Refugee Resettlement, U.S. Department of Health and Human Services). Rumbaut, R. 8 Weeks, J.R. (1986). Fertility and adaptation: Indochinese refugees in the United States. International Migration ReYieY. 29. 428-465- Rumbaut, R.G. (1985). Mental health and the refugee experience: A comparative study of Southeast Asian refugees. In T.C. Owan (Ed.), Sputheast Reign mental health; Ireetnenn, ppevenpion, services, ppeining eng reseeren (DHHS Publication No. ADM-139, pp. 433-486). Rockville, MD: National Institute of Mental Health. Samuel, T.J. 8 Woloski, B. (1985). The labor market experiences of Canadian immigrants. Internatienei MiQEQEiQR. 21. 225-249- Samuel, T.J. (1984). Economic adaptation of refugees in Canada: experience of a quarter century. Ingernapionei Migration. 22. 45-55. Schein, L. (1987). Control of contrasts: Lao-Hmong refugees in American contexts. In S.M. Morgan 8 E. Colson (Ede-). Resale in unhanal (PP- 88-107)- New York: Center for Migration Studies. Schorr. A-E- (1990)- Refnsee ans immigrant resenrse dipeepgpy, i920-i921. Alaska: The Denali Press. 137 Scudder, T. 8 Colson, E. (1982). From welfare to development: a conceptual framework for the analysis of dislocated people. In A. Hansen 8 A. Oliver-Smith (Ede-). InYolontarY migration and resettlement (PP- 267-287). Boulder: Westview. Seligman, M.E. (1975). e s e . San Francisco: Freeman. Shaw, M.E. 8 Wright, J.M. (1967). fieeiee ggp‘pne neeenpenen; p; eppipngee. New York: McGraw-Hill Book Company. Skinner, K.A. 8 Hendricks, G.L. (1979). The shaping of ethnic self-identity among Indochinese refugees. m1 2: m2 $113.5. 1(3)! 25-41- Starr, P.D. 8 Roberts, A.E. (1982). Community structure and Vietnamese refugee adaptation: the significance of context. International Migration R V' W. 19(3). 595- 618. Stein, B.N. (1986). The experience of being a refugee: Insights from the research literature. In C.L. Williams 8 J. Westermeyer (Eds.), Regugee nenpel neeipn in resettlement oomntries (PP- 5-23). Washington: Hemisphere. Stein, B.N. (1981). The refugee experience: Defining the parameters of a field of study. inpepnepignei migration R 'eW. 19(1). 320-330- Stein, B.N. (1979). Occupational adjustment of refugees: the Vietnamese in the United States. Intepnepional Migration ReYieY. 19. 25-45- Stepick, A. 8 Portes, A. (1986). Flight into despair: a profile of recent Haitian refugees in South Florida. International Migration ReYieY. 29. 329-350. Strand, P.I. (1984). Employment predictors among Indochinese refugees. International Migration Rexien, 19(1). 50-64. - Suhrke, A. (1983). Global refugee movements and strategies of response. In M.M. Kritz (Ed.), R,S. innigration end refugee pgiicy: gigpei eng demestie issues (pp. 157- 171). Lexington: D.C. Heath. Tayabas, T. 8 Pok, T. (1983). The arrival of the Southeast Asian refugees in America: an overview. In R. Morales (Ed-). Bringing ooltnresa Sontheast Asian refnoees in Anepiee (pp. 3-14). Los Angeles: Asian American Community Mental Health Training Center. 138 United Nations (1967)- 9onYention Relating to the Statns of Refugeesi 29 inlY 12911 as amended hY the Rrotoool Relating to the Status of Refooeesi Rem YorRl 21 danoarY 1991- United Nations Population Division (UNPD, June 1991). The World's Refugee Populations. Repnlntign NSESIQLLSI. 11! pp. 1-8' U.S. Bureau of the Census (1990). fitdtiStile npeppeep p; the United States- U.S. Committee for Refugees (USCR, 1989). nglg Refingee Suzzexg 1299 in BSYIQEE Westermeyer, J. (1986). Migration and psychopathology. In C.L. Williams 8 J. Westermeyer (Eds.), Rejngee menpe; health in resettlement oonntries (PP- 39-59)- Washington: Hemisphere. Wortman, C. 8 Brehan, J.W. (1975). Responses to uncontrollable outcomes: an integration of reactance theory and the learned helplessness model. In L. Berkowitz (ed-) AdYanoes in Ernerimental Sooial Reyenglegy. New York: Academic Press. Wright, R.G. (1981). Voluntary agencies and the resettlement of refugees. lnpepnepignel Rigpepipn 892192. 19(1). 157'174- Zucker, N.L. (1983). Refugee resettlement in the United States: Policy and problems. Anneie p; pne Anepieen Assooiation.of Rolitioal and Sooial Soienoes. lol. 172- 186.