< ":Wfi'fii.’ . ,.. ,... m» Yfi‘fih. m: To‘l ’ 6: {W firm .q.» wwmmflfi- ,9; . £644 umm i- A b. 1 .p ‘3; $13332»? ~- #; 95247:: ' EH! }‘ 1? RH -'.‘.‘.:.tmv..m “M. ... . . . ‘ l °1v ‘ H" {H ‘7“ a??? ‘i I, , 0 Ln ‘ A; 1 :5 t 6) V l ‘i‘ 1. gait f - ' 'W‘fliflrv $5 I .Hu‘ '- fiéfiiéfifihz. ~ 1:?»‘3'2533‘ w-a» v: , .4533}: E“V “5' I . 0‘”. " "a. $5 . '.2,‘ m ".‘T‘... L :5 a, a? .5?“ a: .I I"; 4' . in...” “man. u.- WWlllllllllllllllllllHlllllllllllllllllllllllllllll 2132 3 1293 01405 This is to certify that the thesis entitled _‘The‘ Effects of Culture and acculturation on Child ‘ 'Behavifir and Emotional Problems: A Comparison of Jamaican Immigrants with Native U.S. and Jamaican Children presented by Mikhail Lyubansky has been accepted towards fulfillment of the requirements for r t“ ”A.- degree in PS; J‘ZQJLQJJPQ-y W Major professor Date [/1/7// '75 0—7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State Universlty PLACE ll RETURN BOX to removal”. Maximum your record. TO AVOID FINES mum on or before data duo. DATE DUE DATE DUE DATE DUE MSU Is An mum Action/Ema] Opportunity Institution Walla-9.! EFFECTS OF CULTURE AND ACCULTURATION ON CHILD BEHAVIORAL AND EMOTIONAL PROBLEMS: A COMPARISON OF JAMAICAN IMMIGRANTS WITH NATIVE U.S. AND NATIVE JAMAICAN CHILDREN By Mikhail Lyubansky A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1996 ABSTRACT EFFECTS OF CULTURE AND ACCULTURATION ON CHILD BEHAVIORAL AND EMOTIONAL PROBLEMS: A COMPARISON OF JAMAICAN IMMIGRANTS WITH NATIVE U.S. AND NATIVE JAMAICAN CHILDREN By Mikhail Lyubansky This study examined the prevalence of behavior problems in three different groups of non-clinic referred adolescents: Jamaican natives, U.S. natives, and Jamaican immigrants in the U.S. Two models of cultural influence (problem suppression-facilitation and adult distress threshold) were applied to the data to determine the most appropriate framework for understanding how immigration affects child behavior. Questions of interest included: 1) Do Jamaican immigrant children have higher problem severity than their U.S. and Jamaican counterparts, and 2) Do Jamaican immigrant children exhibit different problem types from the other two groups? Dependent variables included individual problem items, eight syndrome scores, internalizing and externalizing groupings of syndrome scores, and the total problem score. These were analyzed separately in a 3 (ethnic status) X 2 (gender) X 3 (age level) Analyses of Covariance (ANCOVAS). Socioeconomic status (SES) was used as a covariate. The findings indicated very modest differences among the problems reported by parents in the three ethnic groups. To my parents, who had the Vision and courage to leave behind everyone they knew, in order to start over in a land far away iii ACKNOWLEDGMENTS I am glad to finally have the opportunity to express my gratitude to several individuals. I am foremost indebted to my committee chairperson and research mentor, Dr. Michael Lambert, who enthusiastically read drafi afier draft of this manuscript. His attention to detail and thorough knowledge of the subject were invaluable in my development as a researcher. In addition, his trust and sense of humor made the long hours spent working on this thesis both enjoyable and fulfilling. I would also like to express my gratitude to my committee members: Dr. Gary Stollak, for the everpresent encouragement and support that allows graduate students to develop and pursue their own research interests and Dr. Jackie Lerner, for her generosity with her time and expertise. Additionally, this thesis would not have been possible without Dr. Gabriella Hohn, who graciously shared the data, which took her many long months to collect. Her encouragement and optimistic nature were also greatly appreciated. I am similarly grateful to Dr. Frank Knight, of the University of the West Indies, and Dr. Thomas Achenbach, of the University of Vermont for giving me access to their data. Finally, a special thank you is in order to my parents, Boris and Lydia Lyubansky, for encouraging my curiosity as a child and supporting my career choice. iv TABLE OF CONTENTS LIST OF TABLES .................................................. Vii INTRODUCTION ................................................... 1 CHAPTER 1 CULTURE ......................................................... 4 Historical Perspective ............................................. 5 Culture-Bound Syndromes ......................................... 7 Psychopathology in Children ...................................... 12 Clinic Samples ............................................. 12 General Samples ............................................ 14 Theories Models and Explanations .................................. 17 Problem Suppression Facilitation Model ......................... 17 Adult Distress Threshold Model ................................ 18 Embeddedness of Contexts .................................... 19 CHAPTER 2 ACCULTURATION AND MENTAL HEALTH .......................... 21 CHAPTER 3 METHOD ......................................................... 28 Sample ....................................................... 28 Native Jamaican Sample ...................................... 28 U.S. Sample ........... ' ..................................... 29 Jamaican Immigrant Sample ................................... 29 Research Design ................................................ 30 Data Collection Procedures ....................................... 3O CBCL .................................................... 31 J YC ...................................................... 32 CHAPTER 4 RESULTS ........................................................ 34 Individual Problem Scores ........................................ 35 Cultural Group Differences .................................... 35 Gender Differences .......................................... 36 Age Differences ............................................. 37 SES Effects ................................................ 37 Interactions ................................................ 37 Syndrome Scores ............................................... 38 Internalizing and Extemalizing Groupings ............................ 38 Extra Jamaican Items ............................................ 38 CHAPTER 5 DISCUSSION ..................................................... 41 REFERENCES .................................................... 48 APPENDICES ..................................................... 61 Appendix A: Table 1 ............................................ 61 Appendix B: Table 2 ............................................. 66 Appendix C: Table 3 ............................................. 67 Vi LIST OF TABLES Table l - Percentage of Variance Accounted for by Significant (p<.01) Effects for Culture Group, Gender, and Age on Behavior and Emotional Problem Scores ....................................................... 61 Table 2 - Comparison of Total Problem Scores on the CBCL for Native Jamaican, Native U.S. and Immigrant Jamaican children ........................ 66 Table 3 - Percentage of Variance Accounted for by Significant (p<.01) Effects for Culture group, Gender, and Age on Jamaican-specific Problems .......... 67 Vii INTRODUCTION The pOpulation of the United States is comprised of numerous immigrants and descendants of immigrants from across the world. Despite a projected decline in the population growth rate, the proportion of new immigrants to the United States is expected to keep growing (Culbertson, 1993). These immigrants bring with them their own sociocultural backgrounds and associated child rearing practices, which through the process of acculturation, often become integrated with those from the U.S. It has been documented that parenting style is often rooted in one's culture of origin. For instance, Lambert (1987) examined parental values of parents living in Italy, Greece, Portugal, Japan, United States, and Canada. He found several cross-cultural differences. For example, Italian parents expect more sex-role differentiations than Canadian parents, and Japanese parents oversocialize sons compared to North American parents, who tend to over socialize daughters. Hackett and Hackett (1993) also compared cultural differences in parental values. They found that Gujarati parents in England were less tolerant of physical aggression and bed-wetting but more tolerant of lying than their British counterparts. Other studies have also revealed cultural differences in parental values. Raina (1975) found that U.S. parents rated sense of humor, sincerity, and self-confidence as the top three qualities of ideal children, while Indian parents ranked completing work on time, courageous in convictions, and adventurous as first, second, and third, respectively. It has also been found that non-Anglo immigrants in Australia differed from other 2 groups in their strong emphasis on school performance, sex role performance, and Sibling relations (Burns, Homel, & Goodnow, 1984). Similarly, unlike U.S.-bom parents who favored developing autonomy in their children, immigrant parents from Cambodia, Mexico, the Philippines, and Vietnam rated conforming to external standards as being more important to develop in their children than autonomous behaviors (Okagaki & Stemberg, 1993). Changes in parental values following immigration were also documented by Kurian and Ghosh (1978), who found that immigrant Indian parents were less dominant and less formal in parent-child relationships than non-immigrant Indian parents. In Spite of the culturally based parenting practices discussed in the preceding paragraphs, parental values and attitudes toward child rearing are not completely rigid. That is, parental child rearing beliefs and behaviors are often influenced by the family's acculturation to the host country. For instance, in addition to examining cross-cultural differences in parental values, Lambert (1987) also sampled parents of comparable age and socioeconomic status who had been born in Italy, Greece, Portugal, or Japan but who had immigrated to Canada or the United States in early adulthood and were bringing up their families in North America. He found that persons from all four cultural groups adjusted their values in the direction of host-nation norms to some extent, but that all four also maintained some of the native-culture values. Each national group was found to have its own distinct pattern of adjusting (i.e. to the host culture). Portuguese parents were generally "large" adjusters while the Japanese and Italian parents were generally "small" adjusters. Lambert (1987) also documented that all four immigrant groups adjust 3 their sex-role expectations in the direction of North American parents. However, Japanese parents made the smallest changes. Numerous studies (e.g. Baumrind, 1971, 1980; Dombusch, Ritter, Leiderrnan, Roberts, & Fraleigh, 1987) have linked parenting style to psychological constructs like social competence, self reliance, impulsivity, and psychopathology. Parenting styles of immigrant families are influenced by their culture of origin and that of the host country. This poses certain challenges for research and clinical intervention with immigrant children and their families. These challenges include understanding the influence of the immigrant family’s culture of origin and the culture in which the family resides. Sensitivity to the effects of the host culture (i.e. magnitude of acculturation) on the child and family’s emotional and behavior development is also important. Consequently, the following review of the literature focuses on culture, acculturation, and their effects on child psychopathology. Chapter 1 CULTURE One of the Obstacles for progress in cross-cultural research has been confusion about the definition of culture (Brislin, 1983; Rohner, 1984). This is confounded by the fact that the concepts of race, ethnicity, and nationality, are still often used interchangeably with culture (Betancourt & Lopez, 1993). The issue of definition is sometimes made more complex when religion is involved. An example of such complexity may be found in the case of Jewish immigrants from the former Soviet Union who, upon migrating to the United States, find themselves referred to as "Russians" by people from the U.S. who perceive them as ethnically Russian people who practice the Jewish religion (Birrnan, in press). Rohner (1984) reviewed the elements found in anthropological and cross-cultural psychology literature. He proposed a definition of culture as "highly variable systems of meanings which are learned and shared by a people or an identifiable segment of the population....[and] transmitted from one generation to another." Although Rohner provided a good general definition, Triandis et al. (1980) proposed a more practical definition for the purpose of examining cross-cultural psychological research. Triandis distinguished between the physical culture which includes objects such as buildings and tools, and the subjective culture which includes elements such as social norms, roles, beliefs, and values. More specifically, familial roles, communication patterns, affective styles, and values regarding individualism, collectivism, spirituality, and religiosity all comprise Triandis's conception of subjective culture. 5 Like the definition of culture, the relationship between culture and psychological functioning has also evolved over time. The evolution of this association, beginning with a historical perspective, will be the focus of discussion throughout this paper. Historical Persmctive Psychologists have tried to determine the role of culture in the development of psychopathology Since the beginning of this century. For most of the early 1900's, the main question in the literature wasiwhether mental illness existed at all in preindustrial societies, or whether it was, as Freud had proposed, a by—product of "civilization and its discontents" (Freud, 193 0). Freud's opinion, which was quite popular at the time, might have originated from romantic Rousseauistic notions of the “happy primitive.” This perspective regarded mental disorder as a disease of civilization. Moreover, it argued that the prevalence of mental illness increased as a function of the advancement of civilization. Mead, a well-known anthropologist, was one of the first social scientists to study cultural influences on psychopathology. In fact, her now famous field research, which explored the role of culture in labeling, as well as creating psychopathology, provided initial support for the Rousseauistic View. Mead observed that the social and psychological difficulties expected in American society were not evident among girls in Samoa (Mead, 1928). She inferred that the behavior and emotional difficulties experienced in adolescence were the result of cultural artifacts and not caused by universal biological forces. Mead did not, however, agree with the notion that psychopathology was absent in “primitive” cultures. She concluded instead that human 6 nature is flexible and that cultural forces could be used to improve as well as to diminish psychological well-being. The research that followed (e.g. Malinowski, 1953; Opler, 1955), supported Mead's conclusions. Moreover, this research indicated that psychopathology was present across a wide range of cultural settings. The work of Benedict was also important in the early study of culture and psychopathology. Benedict based her case on various cultural accounts recorded by ethnographers starting in the late nineteenth century. Benedict (1934) argued that mental illness can only be defined in the context of cultural standards of normal and abnormal behavior. She claimed that each culture chooses a subset of all possible human behavior as socially appropriate. All behavior that does not fit into this subset is labeled “abnormal.” Benedict further suggested that within each culture there exist individual differences in both personality and behavior, and that across cultures, the standards of "normal" and "abnormal" are somewhat arbitrary. In addition to the ground-breaking work of Mead and Benedict, anthropologists such as Edgerton (1966), who studied four East African tribes, and Murphy (1976), who studied tribal groups in Alaska and Nigeria, helped establish the universality of psychopathology. Although this body of research contributed to the knowledge base, it was limited by the lack of standardized classification and diagnostic systems. Kraepelin and Meyer attempted to remedy this problem by creating a system for classifying mental disorders in the late 19th century. However, the system was plagued with deficits in both reliability and validity. The Diagnostic and Statistical Manual of Mental Disorders, third edition (DSM- 7 III) and the revision of the lntemational Classification of Diseases (ICD-9), were designed to address these deficits in the classification systems. In 1980, the American Psychiatric Association released DSM-HI. Unlike its two predecessors, the DSM-HI established explicit diagnostic criteria for most mental disorders. The standardized diagnostic guidelines provided by the newer DSM and ICD systems, formed the foundation for large-scale, cross-cultural psychiatric epidemiological surveys. Among the notable ones were those carried out by the World Health Organization (e.g., Sartorius, Jablensky, Gulbinat, & Emberg, 1980; World Health Organization, 1983). In addition to earlier research, these large scale surveys highlighted the presence of certain behavior and emotional problems that seemed universal (i.e., culture-general) and others that seemed culture-bound (i.e., culture specific). The universality and specificity of psychological problems are fully explored in the next two sections. Culture-Bound Syndromes Research in cross-cultural psychopathology has revealed examples of mental disorders that are not easily fitted into the European and American classification system. These disorders, which appear to be specific to a particular culture, are known as culture-bound syndromes, and they include Amok (SE Asia), Banga, and Koro (China), Latah (Malaysia), Misala, and Piblokto (Eskimo), Taijin kyofusho (Japan), Susto (Latin America), and Windigo (Northern Algonkian) (see Murphy, 1976; Rack, 1982). Although these syndromes may initially appear to be unlike the disorders found in the Western classification system, a closer examination reveals that most of them are local variants of disorders listed in the DSM system. 8 Amok, for example, is an outburst of murderous frenzy that occurs in South East Asia. Having reached the breaking point, the afflicted person goes on a rampage and tries to kill as many people as possible before being killed himself (Linton, 1956; Rack, 1982). Amok has been attributed to drug intoxication, psychogenic psychosis, and schizophrenia (Rack, 1982), and it has an undeniable similarity to a severe form of anti-social personality disorder or psychopathy. By implementing a unique intervention model, the Dutch society provided evidence that Amok is a form of hysteria or somatoforrn disorder. They successfully eliminated Amok by Specifying that no matter how many people an individual kills, the Amok runner was to be captured and given a life-sentence of labor, but was never to be killed (Linton, 1956). Further evidence that Amok might not be a culture-specific syndrome is provided by Westerrneyer, who pointed out that Amok may actually take place anywhere in the world and tends to increase during times of political, economic, and social upheaval (Westermeyer, 1973). Most of the other culture-bound syndromes also appear to be variations of Western disorders. Banga and Misala, in fact, appear to be merely local names for psychopathology of any kind (Rack, 1982). Others, like Windigo (a form of cannibalism unique to the Northern Algonkian culture) and Piblokto (convulsive seizures occurring among the arctic Eskimo), are related to culturally-determined forms of depression (Rack, 1982). Similarly, Latah, is clearly a Malaysian version of a somatoforrn disorder (Tseng & McDermott, 1981). One of the less exotic of the culture-bound syndromes is Taijin kyofusho (TKS), a common form of social phobia in Japan (Kinnayer, 1991). Like the disorders discussed 9 in the preceding paragraph, TKS may be viewed as a variation of psychopathology in other nationalities. However, the prevalence of this disorder in the Japanese society and the forms it takes are worthy of attention. Kasahara (1987) has described four different variations of TKS: 1) A transient type that manifests itself during a particular period in life, usually adolescence; 2) A phobic neurosis that closely resembles the DSMIII-R social phobia; 3) A severe social phobia that involves delusions of doing something that adversely affects others; and 4) A phobic disorder accompanied by schizophrenia. The last two types are much more rare, and most patients have a single dominant symptom like fear of blushing or of making eye contact, which has recently become more common (Reynolds, 1987). Kasahara (1987) noted that of 430 students receiving psychiatric care at the Kyoto University student health service, 18.6% could be classified as TKS. This disorder was the third most common diagnosis after depression (24%) and psychosomatic disorders (20%). It should be noted, however, that TKS symptoms usually develop between puberty and early adulthood and tend to diminish with age (Kirrnayer, 1991); thus the prevalence in the general population is probably lower. Regardless of the actual prevalence rate, it is clear that TKS is extremely common in Japan. With the exception of Korea, fear of making eye contact and other phobias usually associated with TKS have not been reported anywhere else in the world psychiatric literature (Kasahara, 1986). TKS can therefore be regarded as a unique disorder, formed at least in part by the Japanese culture. Like the culture-bound syndromes described earlier, it is possible to give a DSM 10 diagnosis to a patient with TKS. Tanaka-Matsumi (1979) performed a study in which 24 American psychologists and psychiatrists did blind DSMH diagnoses of six case descriptions of TKS taken from the Japanese literature and six case descriptions taken from standard U.S. textbooks. It was found that individual TKS cases were diagnosed with the same reliability as the U.S. cases. However, despite the inter-rater reliability, TKS cases were given a wide variety of diagnoses (e.g. schizophrenia, paranoid personality, phobic neurosis), indicating that Kasahara's four subtypes of TKS do not constitute a single coherent syndrome for U.S. mental health professionals. Conceptualizing the four subtypes of TKS as variations of the same disorder is based on the assumption that the TKS symptoms form a continuum of severity. The severity of the disorder ranges from mild phobic symptoms which do not have a profound effect on the individual's ftmctioning, to psychotic symptoms that are severely debilitating. This assumption differs from Western psychiatry, which categorizes constellations of symptoms representing distinct syndromes. It, however, matches empirically derived classification systems that quantify not only the presence of certain symptoms but also the severity of symptoms and syndromes (Achenbach, 1991a). Thus, it is possible that Taijin kyofusho and other “culture-bound syndromes” are merely cultural differences in diagnosis and not true culture-induced variations in psychopathology. Conversely, TKS may indeed be a product of the Japanese culture. For example, fear of making eye contact is quite logical when examined in the context of the social expectations of the Japanese culture. According to Kirmayer (1991), "the role of gaze in 11 nonverbal communication differs in Japan, where people who make too much eye contact are likely to be viewed as insensitive to others, unpleasantly bold, or aggressive." Given the above cultural standard, it is hardly surprising that a fear of hurting someone with one's glance is evident in Japanese society. The manifestation of TKS as a psychotic disorder is a little more difficult to explain. However, psychotic disorders are not completely immune to cultural influences. Tseng and McDermott (1981) indicated that while symptoms of schizophrenia do not differ across cultures, the content of the delusions and hallucinations is influenced by the culture. Since the Japanese culture places such a high premium on being sensitive and respectful, it would be surprising if these factors were not manifested in psychotic delusions and hallucinations. In addition, a study done by Waxler (1979), focusing on schizophrenia in Sri Lanka suggests that the outcome for schizophrenia might also be affected by cultural expectations. Although Waxler’s study dealt primarily with nonindustrial societies, the cultural expectations of a short-term illness followed by a quick return to normality, which is present in nonindustrial societies, is also found in Japan (Kirmayer, 1991). This underscores that this phenomenon may be a product of the cultural values of the society rather than being entirely a function of degree of industrialization. To summarize, there is evidence that despite its form, psychopathology is evident in practically every society. The literature also indicates that the forms some syndromes take are unique to some societies. However, variations of these syndromes may be observed in other societies. 12 Psychopathology in Children The discussion thus far has focused primarily on adults. This reflects the fact that most research on psychopathology conducted cross-nationally has focused almost exclusively on the adult population. This is surprising since the effects of culture are felt well before adulthood (Weisz et al., 1988). Recognition of this problem has led contemporary researchers to examine the effects of the cultural context on child behavior and emotional problems. In addition to individual problem items, many of these researchers (e.g., Achenbach et al., 1987a, 1987b; Lambert et al., 1989, 1994) also compared scores on two broad-band problem groupings known as internalizing (e.g., depression, shyness, fearfulness) and externalizing (e.g., stealing, fighting, threatening others). Studies examining the effects of the cultural context on number of problems and problem type (i.e., internalizing vs. externalizing) in both clinical and nonclinical samples are reviewed below. Clinic samples. Several studies have compared clinic-referred children across different cultures to determine cultural effects on the development of child behavior disorders. In a recent study, Lambert, Weisz, & Knight (1989) compared 360 clinic-referred Jamaican children with 360 clinic-referred U.S. children. Their findings supported their hypothesis that, because of the British based parental emphasis on compliance and respect for authority (Zigler, 1982), Jamaican children were more likely to present more internalizing and less externalizing problems than their U.S. counterparts when they were referred for clinical services. However, it was also found that despite the cultural emphasis on compliance, Jamaican children were still twice as likely to develop 13 externalizing problems than internalizing problems (Lambert et al., 1989). Another study, comparing children from the U.S. and Thailand, found that while there were no differences in the referrability of children for internalizing disorders, U.S. children were about four times more likely than Thai children to be referred for treatment of externalizing disorders (Weisz and Weiss, 1991). Cultural differences in the rates of behavior problems were also found between U.S. children and children from Puerto Rico (Thomas, Chess, Sillen, & Mendez, 1974). Unlike externalizing problems, cultural differences in the referrability of internalizing problems are, for the most part, either minuscule or nonexistent. Several studies (e.g. Weisz, Suwanlert, Chaiyasit, & Walter, 1987a; Weisz & Weiss, 1991; Lambert, Weisz, & Knight, 1989) reveal relatively small cultural differences in the prevalence of most internalizing disorders. One notable exception in the literature is the large cultural difference in the prevalence of somatic disorders. Lambert et a1. (1989) found that 19.8% of children in a Jamaican clinic population were referred for somatic problems, compared to 6.3% in the United States. Comparisons between Thai and U.S. children (Weisz et al., 1987a) revealed similar differences in somatic problems. The above studies indicate that culture may have a significant effect on the development of some child problems. However, the differences in number and type of child referral problems could be the result of culture-induced variation in parental attitudes about which behaviors require treatment and which do not. The availability and acceSSibility of mental health services and the attitudes of adults toward seeking these services may explain the cross-cultural differences found in clinic populations. AS l4 discussed below, recent studies revealed that these differences might not exist in the general population. General samples. In a study comparing Jamaican and U.S. children, Lambert and his colleagues found no cultural differences in parent-reports of internalizing, externalizing, or total problem scores in a sample of non-referred 6-11 year-olds (Lambert, Knight, Taylor & Achenbach, 1994). Cultural differences were also not evident in comparisons of U.S. children with children from Holland (Achenbach, Velhurst, Baron, & Akkerhuis, 1987b) and Thailand (Weisz et al., 1987b). These findings are noteworthy, since the methodology that was used is capable of detecting cultural differences, as evidenced by the fact that significantly more problems were found in Puerto Rican, Australian, and French samples than in demographically matched U.S. samples (Achenbach, Bird, et al., 1990; Achenbach, Hensley, Phares, & Grayson, 1990; Stanger, F ombonne, & Achenbach). Nevertheless, even when Significant cross-cultural differences were found, the differences in total problem scores were relatively small (Lambert et al., 1994). Few cultural differences in child behavior problems were also found in comparisons of mother-interviews of a total of 1227 randomly selected children from Sweden, Sudan, and Nigeria (see Cederblad, 1988). Although the studies revealed age and urban/rural differences, Cederblad found only one significant cultural difference (in a rural sample, 23% of Swedish children over age seven were characterized as behaviorally disturbed compared with 12% of Nigerian and 10% of Sudanese children). Based on his findings, Cederblad concluded that "the similarities of behaviorally disturbed children 15 were more striking than the differences." It is possible, however, that the cultural similarities may be artifacts of the samples utilized in some studies. For example, the earlier nonclinic Jamaican parent-report study (Lambert et al., 1994) surveyed relatively young children in primarily urban areas. By employing methodology similar to their 1994 study of 6-1 1 year-olds, Lambert, Knight, Lyubansky, & Achenbach (1996) found several cultural differences in their comparisons of Jamaican and U.S. adolescents (ages 12-18). Similar to their 1989 study of clinically referred children, the authors found no significant cultural differences between parent reports of total problem scores. However, parents of Jamaican adolescents reported significantly higher scores than parents of U.S. adolescents on internalizing problems and significantly lower scores on externalizing problems. A comparison of adolescent self-reports in the same sample also yielded similar findings to those obtained on parent-reports. Lambert and his colleagues interpreted their most recent findings in the context of the 1994 study of 6-11 year olds and their 1989 clinic study. They inferred that cultural effects on problem type exhibited by children may not become evident until adolescence, because it takes several years for the socializing influences of the culture to Shape the form of child problem behaviors. The pattern of cultural differences emerging in adolescence discussed in the Jamaican/U.S. comparisons is also evident in Dutch/U.S. comparisons. U.S. adolescents reported significantly more problems than their Dutch counterparts on the externalizing, internalizing, and total problem scales (V erhulst, Achenbach, Ferdinand, & KasiuS, 1993), whereas cultural differences were not found in Dutch/U.S. comparisons of 16 younger samples (Achenbach, Velhurst, Baron, & Akkerhuis, 1987b; Achenbach, Velhurst, Edelbrock, Baron, & Akkerhuis, 1987b 1987c). Cultural differences were also found in studies of teacher reports of non clinically-referred children. Lambert, Knight, Taylor, & Achenbach (1996) compared 6-1 l-year-old Jamaican and U.S. children, using the teacher form of the Child Behavior Checklist (TRF, Achenbach, 1991b). The authors found that J arnaican teachers rated their pupils significantly higher on 49 of the 51 items for which cultural effects emerged. Jamaican children were also rated higher than U.S. children on the externalizing, internalizing, and total problem scores. These findings were especially intriguing considering that the parent and teacher reports were obtained on the same sample of children. Moreover, it underscored the importance of obtaining information from a variety of informants in cross-national surveys of child behavior problems. A comparison of teacher reports in the adolescent sample also yielded higher scores for Jamaican adolescents on externalizing, internalizing, and total problem scales (Lambert et al., 1995). Taken together, the differences in parent and teacher report findings may be attributable to the lack of agreement between different informants (Lyubansky, Lambert, McCaslin, & Knight, 1995). Although this phenomenon is not unusual (e.g. Achenbach, 1989; Stanger and Lewis, 1993), it is possible that cultural factors also contributed to this profound difference. Lambert et a1. (1996) speculated that the tremendous emphasis Jamaican teachers and other adults place on academic achievement may cause many Jamaican children to find school exceptionally stressful. An alternative explanation is that child behavior is fairly consistent from home to school, 17 but that Jamaican teachers have higher expectations and less tolerance for even mildly inappropriate behaviors. To summarize, despite the variations across reporters, true differences may be caused by cultural factors such as attitudes regarding child rearing. Several theoretical models have been presented in efforts to explain the effects of attitudes and behaviors of adults such as parents and teachers on child problems. Some of these theoretical models are described below. " Theories. models, and explam Much progress has been made since Mead and Benedict first began to study psychopathology across cultures. It is now clear that childhood behavior and emotional disorders exist across cultures (e.g. Townsend, 1978; Cederblad, 1988). However, the precise role of culture in the development of these disorders has not yet been ascertained. Weisz et al. (1987a) suggested two possible methods through which culture could exert an influence on the development of childhood disorders. Since they describe very different cultural influences, each of them will be treated separately. Problem suppression-facilitation model. The problem suppression-facilitation model (Weisz et al., 1987a) predicts that the effects of culture are directly related to the incidence and prevalence of certain child disorders by suppressing (e.g., via punishment) some behaviors and facilitating (e.g. through reinforcement and modeling) other behaviors. This model is consistent with the previously described cultural differences in the prevalence of behavior disorders in the clinic-populations of the different cross-national studies. Many non-Western and several Western cultures value family ties 18 and respect for authority (e.g. Leff, 1981; Singer, Ney, & Lieh-Mak, 1978). According to the problem suppression-facilitation model, these cultures should suppress the development of externalizing behavior disorders, which are associated with independence and rebellion, and facilitate internalizing problems like withdrawal and somatic disorders. Research comparing parent reports of clinic-referred children in several different cultures (e.g. Lambert et al., 1989; Weisz and Weiss, 1991) support this model. However, as previously described, cross-cultural differences are often not evident when comparisons are made between pre-adolescent children in the general population (Achenbach et al., 1987b, 1987c; Lambert et al., 1994). Weisz's second model of cultural influence may provide one possible explanation of this phenomena. Adult distress threshold model. The adult distress threshold model (Weisz et al., 1988) proposes that culturally induced attitudes about childhood behavior help to set adult thresholds for distress over child problems and determine which problems are considered serious enough to be referred for treatment. There are two separate forms of this model. A general form of the model holds that adults in different cultures may differ in their thresholds for child problems. For example, as mentioned earlier, it is possible that Jamaican teachers report more child problems than U.S. teachers, because they have a lower tolerance for a wide range of child behaviors (Lambert et al., 1996). A pattern smcific form of the model, on the other hand, states that cultures differ with respect to the types of behaviors that are considered acceptable, with certain types of child problems arousing greater concern in some cultures than in others (Weisz et al., 1988). An example is that parents may only seek treatment for a particular child if they 19 consider that child’s behavior to be problematic. If this is indeed the case, then one would expect two different effects on prevalence rates of childhood disorders. First, given the traditional emphasis on respect and obedience found in Thailand, Jamaica, and many non-Westem cultures, adults in these cultures should have a lower tolerance for externalizing behavior than their counterparts in societies such as the U.S. These adults should be more likely to seek treatment for children exhibiting this type of behavior. Empirical studies completed thus far have failed to confirm this hypothesis. More specifically, these societies have a lower prevalence of extenralizing and a higher prevalence of internalizing disorders in their clinic referred youngsters. However, since culture-induced attitudes often determine which behaviors are referred for treatment, people in some societies are less likely to refer children to mental health professionals for treatment, regardless of the child's behavior. Research findings have consistently supported this prediction, especially in developing countries where mental health services are less readily available or accepted (e.g. Singer et al., 1978; Rack, 1982). Embeddedness of contexts. Szapocznik and Kurtines (1993) further elucidated the models proposed by Weisz et a1. (1987; 1988). Their model considers the family in the context of the culture, a phenomenon referred to as "the embeddedness of contexts" (Szapocznik & Kurtines, 1993). The Szapocznik and Kurtines model further argues that culture shapes family behavior in the same way that parents shape their children's behavior. The previously described cross-cultural differences in the prevalence of mental disorders are consistent with this model. An example of how this model explains the findings obtained in earlier cross-cultural research may further elucidate its application: 20 Since the Thai culture values respect for authority over independence (Weisz et al., 1987a), Thai parents expect their children to be respectful and obedient. Thai parents, therefore, suppress externalizing behavior and facilitate internalizing behavior as predicted by the suppression facilitation model. Thus, Thai children who are referred for clinical services are more likely than their U.S. counterparts to manifest internalizing problems (Weisz et al., 1987a). Parents are not the only agents of socialization. Teachers, peer groups, and other socializing factors within the classroom environment may profoundly influence child behavior. Schools are, therefore, considered to be an important socialization agent in many societies (V iondi, Fleming, & Mintz, 1983). More specifically, schools in many societies are charged with providing children with social and academic skills, which are usually deemed essential for successful adaptive development (Achenbach and Edelbrock, 1986). In some cases, however, the values of the school and the values espoused at home are not Similar. This phenomenon may be more salient for some immigrant families in societies like the United States. Moreover, differences in values may be more at odds in situations where child rearing values are different from the parents’ country of origin. The differences in values of some immigrant families and those espoused by the school context can often be a source of distress for immigrant children and their families. The construct of acculturation and its effects on behavior and emotional problems are therefore discussed below. Chapter 2 ACCULTURATION AND MENTAL HEALTH While cultural influences on mental health often vary from one country to another, this variability can also be observed between immigrants and natives in a given country. For example, the values and attitudes of immigrants arriving in a new country have been Shaped by their traditional culture, some of which are in sharp contrast to the values and attitudes of the new culture. The cultural diversity of values and attitudes can undoubtedly positively contribute to new ideas and creative innovations. However, the values of the traditional and new cultures sometimes clash and the resulting conflict can have deleterious effect on mental health. Consequently, it is important to explore how acculturation, a process of change and adaptation that results from continuous first-th contact between individuals or groups of different cultures, impacts immigrant mental health. One major hurdle in trying to determine the role of acculturation in the development of child behavior problems is that families to do not all respond the same way when they make contact with a new culture. Some families, for instance, may subscribe to the "When in Rome..." philosophy and try to blend into the new culture as much as possible. Other families continue to identify with, and maintain the values and attitudes of, their traditional culture. However, neither of these styles (which are respectively called high and low acculturation) appear to be optimal for adequate emotional adjustment. For example, high acculturation has been shown to be associated with elevated scores on the MMPI (Burnham, Hough, Kamo, Escobar, & Telles, 1987; 21 22 Sorenson & Golding, 1988). Furthermore, there is evidence linking high acculturation with difficulties in parent-child relationships and low levels of adjustment in children (Szapocznik, Kurtinez, & F emandez, 1980; Charron & Ness, 1981). Rumbaut's (1991) study of Indochinese adolescents in the United States provides another example of the deleterious effects of high levels of acculturation. According to his study, youngsters from these sociocultural groups who over identify with the American culture tend to be proportionately less successful academically than their less acculturated peers. Few if any studies examining low acculturation in children exist. However, there is empirical evidence that low levels of acculturation are associated with psychological distress in adults. For example, a variety of symptoms, including depression, withdrawal, somatization, post-traumatic stress disorder (PTSD), and obsessive-compulsive behaviors, have been positively associated with low levels of acculturation in adults (Escobar, 1983; Westermeyer, Bouafuely, Neider, & Callies, 1989). In addition to symptoms of psychopathology, it has also been documented that low acculturation and number of Negative Life Events such as divorce, hospitalizations, and death are also associated (Yu & Harburg, 1981). A slightly different approach to conceptualizing acculturation has been taken by Berry (1980; 1986) and Berry, Kim, Power, Young, & Bujaki (1989). In addition to high acculturation (which Berry called Assimilation) and low acculturation (Berry's term is Separation), Berry identified two additional styles of dealing with a new culture: integration and marginalization. Integration, also known as biculturalism, is the identifying with the new culture while maintaining traditional cultural identity. 23 Marginalization (or deculturation), on the other hand, is characterized by the rejection of both the new and the traditional cultures. The handful of studies which have examined these constructs indicate that integration or biculturalism is associated with the best levels of adjustment in adults. For example, Hispanic adults reported higher life quality, better emotional stability, lower levels of depression, and higher psychological adjustment than those who were either monoculturally Latino (separated) or monoculturally U.S. mainstream (assimilated; Lang, Munoz, Bernal, & Sorenson, 1982). Similarly, the same researchers noted that drug abuse was much more prevalent in monocultural individuals, Specifically in over-acculturated youths and under-acculturated mothers of Cuban families (Lang et al., 1982). Biculturalism was also found to be the optimal method of acculturation in a study comparing the satisfaction and acculturation of Southeast Asian and Hispanic adults. In that study, bicultural individuals reported the highest levels of satisfaction, followed by assimilated and separated individuals, respectively (Wong-Rieger & Quintana, 1987). A study of Indochinese adults (Rumbaut, 1991) yielded similar results. However, as Szapocznik et al. (1980) cautioned, it is not the retention of the old culture or the adaptation to the new culture that is in itself pathological. Rather it is the lack of biculturality that is maladjustive because it makes these individuals inappropriately monocultural in a bicultural context. Although clearly important, the styles of acculturation do not, in and of themselves, adequately explain the process of adapting to a new culture. Indeed, it is not 24 possible to integrate all of the research findings. A meta analysis of 30 studies of acculturation and mental health among Hispanic adults yielded an inconsistent overall pattern of direct, indirect, and curvilinear relationships (Rogler, Cortes, & Malgady, 1991). Rogler (1994) argued that the inconsistency may be due to the fact that the researchers failed to account for other important variables such as availability of social networks and changes in socioeconomic status. Indeed there is empirical evidence that these variables can have substantial effects on mental health. For instance, it has been extensively documented (e.g. Dohrenwend et al., 1980; Holzer et al., 1986; Neugebauer, Dohrenwend, & Dohrenwend, 1980) that there is a negative correlation between SES and mental health problems. This phenomenon can be especially salient in immigrant populations. Because of economic necessity, some immigrants are forced to accept employment that is below their level of education and professional experience. Thus, employment in their host country often results in a lower SES than they were accustomed to, which may lead to emotional stress and psychological problems (Rogler, 1994). Despite the potential for negative effects arising out of the immigration experience, several recent studies suggest that people who perceive a high degree of social support report greater emotional and physical well being than people perceiving a low degree of support, especially in times of stress (see Cohen & Wills, 1985). It is also possible that support from the ethnic community may be especially important in mediating the effects of stress and low SES. This theory, which has become known as the "ethnic density hypothesis" states that there is an inverse relationship between the incidence of mental illness in a particular ethnic group and its size relative to the total 25 population (Cochrane & Bal, 1988). Thus, the inconsistencies in findings from empirical studies may be attributed to the mediating effects of SES changes and the availability of social support for given immigrants. To summarize, there is a growing body of research that links the immigration experience to emotional adjustment in adults. However, few studies have focused on immigration effects on child adj ustrnent. Additionally, while there is a growing body of literature on sociocultural factors and their effects on child problems, few if any studies have examined these effects in immigrant and nonimmigrant populations in host countries and the countries of origin. As described below, the present study is a beginning attempt at bridging the gaps between these two important bodies of research. This exploratory research, therefore, focused on Jamaican children ages 7-12 residing in Jamaica, children of similar age who emigrated from Jamaica to the U.S. and currently reside in New York, and U.S. children of the same age group residing throughout the contiguous United States. This is an important sample, because Jamaicans comprise the largest group of West Indian immigrants, who now constitute one third of the total foreign born population and close to 20% of the entire student population in New York City public schools (Thomas & Lindenthal, 1990). Moreover, although Jamaicans are the 15th largest group of immigrants in the U.S. (See Rumbaut, 1994), little empirical psychological research has been done at this point with the Jamaican immigrant child population. Questions of interest included: 1) will J amaican immigrant children have higher problem severity than their U.S. and Jamaican counterparts? and 2) will Jamaican 26 immigrant children exhibit different problem types (i.e. syndrome scores and internalizing vs. externalizing groupings of syndromes) from the other two groups? The earlier discussion on stress and its association with behavior and emotional disorders in immigrant populations suggests that Jamaican immigrant children are more likely to receive higher problem ratings than their non-immigrant counterparts. However, the Jamaican immigrants’ propensity to stay in close proximity and support one another via extended families (I-lohn, 1996), as well as their strong sense of ethnic-cultural identity (Brice, 1982), may provide them with an adequate social support network to mitigate the effects of stress. The problem suppression-flacilitation model suggests that Jamaican immigrant children may receive higher ratings on externalizing problems than J arnaican non- immigrant children and higher ratings on internalizing problems than native U.S. children. According to the suppression-facilitation model, exposure to the U.S. culture which values independence and tolerates more externalizing behavior, might lead Jamaican immigrants to have proportionally more externalizing and fewer internalizing problems than native J arnaican children, whose parents place more emphasis on obedience and respect for authority and have little tolerance for externalizing behaviors. Conversely, due to their previous exposure to Jamaican values, immigrant Jamaican children should have proportionally fewer externalizing problems and more internalizing problems than U.S. natives. An alternative possibility is suggested by the previously described adult distress threshold model. According to the general form of this model, Jamaican parents may 27 have an overall lower tolerance for adolescent behavior problems. Consequently, the general model predicts that Jamaican parents should rate their children higher on all problems (higher total problem scores) than Jamaican immigrant parents. The immigrant adolescents, in turn, may have higher total problem scores than nonimmigrant U.S. adolescents. A pattern specific form of the model, on the other hand, predicts that adults in Jamaica should have a lower tolerance for externalizing behaviors than their counterparts in the United States. Jamaican parents, therefore, may be more likely to rate their children higher on these type of problems than U.S. parents. The pattem—specific form of the model further predicts that because of exposure to the host culture, Jamaican immigrant parents are more likely to rate their children lower on externalizing problems than the native Jamaican parents. The above patterns should also be evident in comparisons of syndrome scores across the two cultural groups (e.g., Jamaican parents should rate their children higher on the Agggpssive syndrome than U.S. parents). Chapter 3 METHOD Sample Native Jamaican Sample. The native Jamaican sample consisted of two separate samples collected at different times using similar methodology. The first sample was collected in 1990. It consisted of children from urban and suburban environments in Kingston, as well as rural areas in the Northeastern part of J arnaica. Three-hundred- forty-nine children, ranging in age from 7 to 11 years, were randomly selected from 29 randomly chosen schools. The schools included 27 public and 2 private schools, which is the approximate nationwide ratio for these two types of schools. From each school, classes from each grade level were randomly selected, and students were then randomly selected from each class. The second sample, consisting of 56 Jamaican 12-year-olds, was selected in 1994 from 16 schools located in urban, rural, and suburban areas throughout J arnaica. For both samples, classes from each school were randomly selected, and a maximum of one student was randomly selected from each class. Parents of the selected children in both samples were asked to participate; of those contacted 90% took part. Ninety percent of the total sample were of African descent, while the other ten percent consisted of other ethnic groups such as Chinese and East Indians. Although the total sample is not representative of the entire country (76% of the subjects resided in Kingston), it is still an important sample Since 27.2% of all Jamaicans reside in Kingston and 34.0% live in urban areas. Of the 405 total respondents, 71.5% were mothers, 9.9% were fathers, and the remaining 18.6% consisted 28 29 of other significant adults such as grandparents and guardians. U.S. Sample. The U.S. sample was collected in 1990. It consisted of subjects in a 3-year follow-up assessment, in which the follow-up rate was 90.7% and the completion rate of the initial survey was 92.1%. The sample of 993 children, ages 7-12, was selected to be representative of the U.S. population with regard to gender, ethnicity, SES, geographic region (Northeast, North Central, South, and West), and area of residence (urban, suburban, rural). Subjects were excluded from the study if they were mentally retarded, or physically handicapped, of if no English-Speaking parent (or parent surrogate) was available for the interview. The data was collected by Temple University's Institute for Survey Research (ISR). Respondents were either a parent of the target child (83.2% mothers and 14.7% fathers), a parent surrogate (e.g., foster parent, stepparent, legal guardian), or another adult who was most responsible for the child. The sample was 73.5% Caucasian, 15.7% African American, and 10.8% other (see Achenbach, 1991). Jamaican Immigmt Sample. The Jamaican immigrant sample was gathered in 1995. It consisted of 120 children (60 boys and 60 girls) between the ages of 7 and 12. The data was collected as part of several ongoing projects examining West Indian immigrant adjustment in New York City, under the auspices of the Caribbean Research Center at Medgar Evars College, City University of New York, Brooklyn. All participants were recruited from New York schools and churches in predominantly West Indian neighborhoods. Thus, the sample is not representative of all Jamaican immigrants in this country. The sample includes children born in the U.S. to Jamaican mothers, 30 children born in Jamaica who migrated to the U.S. with their mothers, and Jamaican children who migrated to the U.S. after a period of separation from their mothers due to nrigration. No attempt was made to select children on the basis of age at immigration or length of time in the United States. Research Desigp The three groups (i.e., Jamaican children, U.S. children, and Jamaican immigrants in U.S.) were matched according to gender and age-group so that each age X gender cell was the same size across the three groups. Each cell has an N of approximately 20 (range = 9 - 28). Data Collection Procedures Parent report instruments of the child’s problems (described below) were used to gather the data for all groups of children. Although it is recognized that parents are only one source of information on child problems, they are extremely important, Since parental distress regarding child problems often determines referral for treatment (see Achenbach, 1991a) All native U.S. interviews were conducted in the subj ects' homes. The Jamaican interviews (native and immigrant) were either conducted at home or at the child’s school (depending on parental preference) by one of two trained Jamaican interviewers. Some Jamaican immigrant interviews were also done at church. For all groups of parents surveyed, the interviewer determined whether the selected child was referred for mental-health related services during the previous year. Children were excluded from the sample if they had been referred for such services 31 during this time. For the native U.S. and Jamaican samples, the interviewer read out loud the Child Behavior Checklist (CBCL, Achenbach, 1991) and the Jamaican Youth Checklist (J YC) respectively, while the parent followed along on another copy. As the parent answered each question, the interviewer recorded the response. For the immigrant Jamaican sample, mothers completed the J YC on their own, under the interviewers supervision. If the mothers had poor reading or language skills, the questions were read to them by the interviewer. SEQ; In the United States, the problem report measure was the Child Behavior Checklist (CBCL), which consists of several demographic questions, 20 social competence items and 118 items describing behavioral/emotional problems (Achenbach, 1991a). Competence items consist of ratings for the amount and quality of the child's participation in sports and other activities, as well as the child's ability to get along with siblings, parents, and other children. Competence in jobs, chores, and academic performance are also included. The 1 18 problem items cover a wide range of problems for parents to report. Examples include "cruel to animals, sees things that other people don't see," and "unhappy, sad, or depressed." The parent scores each item by circling 0 if the item is no; true of the child, 1 if it is somewhpt or sometimes true. and 2 if it is very true or often m. In addition to the individual items, principal-components analyses of the CBCL have yielded eight syndromes for both genders and different age groups. The syndromes are designated Withdrawn, Somatic Complaints. Anxious/Depressed. Social Problems. Thought Problems, Attention Problems, Delinquent Behavior, and Aggessive BehaLior. 32 Second-order principal-factor analyses of these syndromes have yielded the broadband externalizing and internalizing groupings of the syndromes described earlier (Achenbach, 1991). The test—retest reliability of the problem items, computed with interclass correlations (ICC) between parent reports at l-week intervals was .95 (p<.01) (Achenbach & Edelbrock, 1983). The inter-parent reliability for mothers and fathers of clinically referred children is .98 for total behavior problems and .98 for social competencies (Mooney, 1984). AS documented elsewhere, the instrument is psychometrically sound (Achenbach, 1991a). LY; The Jamaican Youth Checklist (JYC) was designed to be similar enough to the CBCL to allow cross-national comparisons, while at the same time sufficiently sensitive to the Jamaican culture to detect subtle behavior patterns in Jamaican children (Lambert et al., 1994). Like the CBCL, the J YC consists of demographic items, followed by competency items, and problem items using the same 0-1-2 rating scale. The competency items are all the same as those of the CBCL, except for the question "Is your child in a special class?,” which was excluded because most Jamaican schools do not have special classes. All 118 of the CBCL problem items are included in the JYC in the same order. However, some items were slightly modified to better reflect the idiomatic expressions in Jamaica, while still retaining similar meaning in the United States (e.g. Item 56c, “Nausea, feels sick” was changed to “Nausea, bad feelings”). In addition, 35 extra items deemed clinically relevant to Jamaican children were added to the J YC, following the 118 problem items (see Lambert et al., 1994). The l-week test-retest ICC of the JYC was .89 (p<.01) for problem scores obtained from 30 parents and .79 (p<.01) 33 for inter-interviewer reliability for the two Jamaican interviewers who independently interviewed the same 30 parents. Preliminary data on the J YC indicate that it adequately discriminates between clinic and non-clinic children (see Lambert et al., 1994). For U.S. and Jamaican immigrant children, SES was scored according to Hollingshead's (1975) 9-Step scale for parent occupation. For calibration with the Jamaican SES scale, the 9-Step scale was divided into three categories: Hollingshead scores of 1.0 to 4.0 = lower SES; 4.5 - 6.5 = middle SES; and 7 - 9 = upper SES. Half-steps (e. g. 4.5) indicate that occupations not clearly scorable were given the mean of the two scores that seemed most appropriate. The U.S. mean SES score was 2.07, SD = .87, while the Jamaican immigrant mean was 2.28, SD=.75. A 5-step scale designed specifically for Jamaicans (Smith, 1984), was used to code SES for the native Jamaican respondents. The scores derived from the Jamaican SES scale were divided into the following three categories: Lower SES = 1.0 to 2.0; middle SES = 2.5 to 3.5; upper SES = 4.0 to 5.0. The mean SES score for native Jamaican children was 1.76, SD = .74. The regression slopes for the SES covariates in each sample were first compared via an Analysis of Covariance (AN COVA). The AN COVA was performed on total problem score (i.e., sum of ratings of the 120 problem items common to the CBCL and the J YC) with culture group as the independent variable and SES as the covariate. This showed no culture group X SES effect, F(2, 333) = .75, p > .05. SES was therefore retained as a covariate, and the same regression slopes were used in AN COVAs of all the dependent variables. Chapter 4 RESULTS Group differences were tested in the following categories: 1) scores for total problems (i.e., the sum of 1's and 2's across 118 of the 120 problem items; “asthma” and “allergy” were dropped according to the Achenbach 1991 procedure); 2) ratings on each of the 120 problem items; 3) scale scores for each of the eight syndromes; and 4) scores for the internalizing groupings of syndromes (i.e. Withdrawn, Somatic Complaints, and Anxious/Depressed) and the externalizing groups of syndromes (i.e. Delinquent Behavior and Aggressive Behavior). Scores for each of the 120 problem items, the 8 syndromes, the externalizing and internalizing scales, and the total problem score were compared Via 131 three (ethnicity: native U.S. children VS. native Jamaican children vs. Jamaican children in the U.S. ) X 2 (gender) X 3 (age group: 7 to 8, 9 to 10, and 11 to 12) ANCOVAS. To control for possible SES effects and to assess the association of each score with SES, the tripartite division of SES scores was used as a covariate. Given the high statistical power inherent in the large sample Sizes (120 in each sample), only differences reaching p < .01 were accepted as Significant. However, even with this conservative alpha criterion, Analyses of Covariance (ANCOVA) of these large samples could detect very small effects. Cohen’s (1988) criteria were, therefore, used to judge effect sizes of significant ANCOVA findings as follows: Effects accounting for l to 5.9% of the variance are small; 5.9 to 13.8% are medium; and >13.8% are large. Type I error was also reduced by identifying the five smallest Significant effects, where 34 35 five is the number expected by chance in a set of similar analyses using a p < .01 protection level (Field and Armenakis, 1974). Individual Problem Scores The results from the ANCOVAS of the 121 specific problems items are listed in Table 1. Five of the 121 E values could reach the .01 level of significance using a p < .01 protection level (see Field & Armenakis, 1974). Therefore, if five or less of the 121 E tests resulted in significant F values, they were considered as chance findings. All comparisons Showing Significant main effects are described below and listed in Table 1. However, the five smallest E values in each column are marked with the superscript ° because they are most likely to be chance findings. The numbers in Table 1 Show the percentage of variance accounted for by the significant differences in problem scores that were associated with culture group, gender, age, and SES. Cultural Group Differences. Significant cultural group effects occurred on 27 specific problems. The culture group effects and the associated percentages of variance are listed in the first column of Table 1. There were no large cultural group effects, but a medium effect emerged for six problems according to Cohen’s criteria. The strongest cultural group effect occurred on item 112, “Worries,” where the tendency of native Jamaican children to obtain higher scores than either the native U.S. children or the Jamaican immigrant children accounted for 12% of the variance. Cultural group effects accounted for 10% of the variance on item 74, “Showing off or clowning,” 9% of the variance on item 56f, “Stomach aches or cramps,” 8% of the variance on item 12, “Complains of loneliness,” and 7% and 6% of the variance on items 109, “Whining,” and 36 17, “Day-dreams,” respectively. All of the remaining 21 significant effects were small by Cohen’s criteria. There was no significant culture group difference in total problem scores. For all significant effects, Sheffe’s test was used to determine the direction of the effect (i.e., which group scored higher). Of the 27 problems showing significant culture group effects, native Jamaican children were significantly higher than the other two cultural groups (which did not statistically differ fi'om each other) on items 5, “Behaves like opposite sex,” 79, “Speech problems,” 81, “Steals at home,” 97, “Threatens people,” 106, “Vandalism,” 107, “Wets self during the day,” and 111, “Withdrawn, doesn’t get involved with others.” Native U.S. children were significantly higher than the other two groups (which were not statistically different) on items 2, “Allergy,” 12, “Lonely,” 31, “Fears doing something bad,” 35, “Feels worthless,” 45, “Nervous,” 58, “picks nose, skin, or other body parts,” 74, “Shows off or clowns,” 109, “Whining,” and 1 12, “Worries.” The Jamaican immigrant children were not statistically higher than both of the other groups on any of the items. However, the Jamaican immigrant children were higher than the native Jamaican children on items 17, “Day-dreams,” 29, “Fears things other than school,” and 56f, “stomachaches or cramps,” and higher than the U.S. children on items 20, “Destroys own things,” 37, “Fights a lot,” and 99, “Too concerned with neatrress.” Gender Differences. The percentages of variance accounted for by gender differences that were significant at p < .01 are listed in the second column of Table l. The superscripts F and M indicate higher scores for females and males, respectively. 37 Across the three cultural groups, males obtained significantly higher scores than females on five items, and females received a significantly higher score on one item. However, it is possible that five of the six significant effects were chance findings (see Table I). There was no Significant difference in total problem scores. All of the gender effects were small by Cohen’s criteria, which is consistent with the gender differences found in other cross-national samples. Age Differences. The third column of Table lindicates significant age differences across the three cultural groups. The only significant age effect occurred on item 29, “Fears certain animals, Situations, or places, other than school,” where younger children score significantly higher than older children. This was a small effect, accounting for 3% of the variance, and was likely due to chance. No significant age effect occurred for the total problem score. SES Effects. By using SES as a covariate in the ANCOVA analyses, SES effects for all cultural groups, gender, and age effects were partialed out. However, SES main effects on problem scores were also tested, and no significant SES effects were found for either the scores on the individual items or the total problem scores. Interactions. All combinations of culture group, gender, and age effects were also tested. A culture group X gender interaction emerged on item 89, “Suspicious,” accounting for 3% of the variance. Scheffe’s test for this effect indicated that boys received higher scores than girls in the native Jamaican sample, but were not statistically different in the other two samples. Since this was the only interaction found for the individual problem items and the total problem score, it is likely a chance effect. 38 Sypdrome Scores As with the individual problem items, scores for each syndrome were computed by ANCOVAS of culture group, gender, and age, with SES partialled out as a covariate. A significant culture group difference emerged for the Anxious Depressed syndrome, where U.S. children scored significantly higher than both native Jamaican and immigrant Jamaican children, who did not statistically differ from each other. This effect accounted for 6% of the variance and was a medium effect according to Cohen’s criteria. However, as with the individual items, we controlled for Type I error, and it is possible that this effect is due to chance. In addition to the culture group effect, the analyses of syndrome scores also yielded a gender effect. This effect emerged on the Delinquent syndrome, where boys scored significantly higher than girls. Gender accounted for 2% of the variance for this syndrome, which is a medium effect according to Cohen’s criteria, and it may also have been due to chance. No age, SES, or interaction effects emerged for any of the syndromes. Internalizing and Extemalizing Groupings Like the syndrome scores, internalizing and externalizing scores were analyzed separately in AN COVAS of culture group, gender, and age, with SES partialled out as a covariate. No culture group, gender, age, or interaction effects emerged for either of the broad band groupings. Extra Jamaican Items The J YC, which was administered to both immigrant and native Jamaican 39 children, had 35 additional items that were deemed clinically relevant to Jamaican children. Since the checklists given to U.S. children did not contain these 35 items, the U.S. children were excluded from this set of analyses, and the 35 items were compared Via 35 two (nonimmigrant VS. immigrant status) X 2 (gender) X 3 (age group) ANCOVAS, with SES as the covariate. The results of these analyses are listed in Table 4. Significant effects emerged for 11 out of the 35 problem items. Of these, seven effects (items 117, “Doesn’t answer when people talk to him/her,” 1 18, “Irritable,” 119, “Uncooperative,” 120, “Puts self in dangerous Situations,” 134, “Begs at home or on the streets,” 140, “Gossips,” and 147, “Arrives home too late from school”) occurred for immigration status, with native Jamaican children receiving higher scores than immigrant children on all seven items. All of these effects were small according to Cohen’s criteria, except item 147, “Arrives home too late from school,” which was a medium effect, accounting for 8% of the variance. In addition, because they have the lowest E values, it is likely that items 119, “Uncooperative,” and item 120, “Puts self in dangerous situations,” occurred by chance. In addition to the seven immigration status effects, there were also two significant gender effects, with boys receiving higher scores than girls on both problem items (138, “Plays too much,” and 143, “Plays too roughly”). Two significant effects also emerged for age, as younger children scored higher on item 134, “Begs at home or on the street,” while the converse was true for item 135, “Lazy.” The analyses also yielded one interaction, as an immigration status X gender effect emerged for item 129, “Talks to self.” A Sheffe’s test of this interaction indicated that native Jamaican boys scored 40 higher than immigrant boys, while girls did not differ across the two groups. From the other direction, native Jamaican boys scored higher than native Jamaican girls, while no gender difference was found in the immigrant sample. All of the gender, age, and interaction effects were small, according to Cohen’s criteria, and it is possible that all of them occurred by chance. Chapter 5 DISCUSSION The findings failed to support either the problem suppression-facilitation model or the adult distress threshold model. The former predicted that characteristics of a culture (e.g., parenting values, social norms) may suppress the development of certain types of child behavior problems and foster or facilitate the development of others. According to this model, Jamaican children should have received higher scores on internalizing problems than U.S. children, with the Jamaican immigrant children falling somewhere in between. The latter model, on the other hand, makes two different predictions. According to the general form, Jamaican children should receive higher total problem scores than U.S. children, because Jamaican parents have a lower tolerance for child problem behaviors than U.S. parents. In addition, according to the pattern Specific form, Jamaican children should receive higher scores on externalizing problems than U.S. children, because Jamaican parents are less tolerant of such behaviors than U.S. parents. The reverse is expected for internalizing problems, with Jamaican immigrants again expected to fall between the two groups, due to their exposure to both cultures. The findings, however, indicated very modest differences between the problems reported by parents for native Jamaican children, U.S. children, and Jamaican immigrant children living in New York. Significant cross-cultural effects occurred in only 22.5% (N=27) of the 120 problem items. Furthermore, there were no Si gnificant culture group differences in total problem scores (see Table 2) or in the internalizing and externalizing groupings. In addition, an examination of the syndromes yielded only one Significant 41 42 culture group effect, as U.S. children scored higher than both native and irnnrigrant Jamaican children on the Anxious Depressed syndrome. However, this effect may have been due to chance, which indicates that there is no consistent tendency for parents of one culture to report more problems than parents from another culture. Moreover, unlike findings on clinic-referred children (e.g. Lambert et al., 1989), parents of one culture did not (except for the Anxious Depressed syndrome) tend to report more problems of one type than parents of another culture. The Similarities across cultures in both number of problems and type of problems were consistent with similarities found by Achenbach et al. (1987b), Weisz et al., (1987b) and Lambert et a1. (1994), who utilized similar methodology in their respective comparisons of Dutch, Thai, and Jamaican children. The present findings differed, however, from the earlier Jamaican -- U.S. clinic comparisons, where Jamaican clinic-referred children received higher internalizing and lower externalizing scores than their U.S. counterparts (Lambert et al., 1989). These findings also differed from a comparison of non-referred J arnaican and U.S. adolescents (Lambert et al., 1996), which yielded the same internalizing differences as the study of clinically referred children. Unlike the present study, both of these studies supported the problem suppression-facilitation model. It has been argued that the differences between the findings of the clinic and nonclinic studies may have resulted from differences in their methodology. For example, the clinic data were gathered during intake interviews with clinicians using unstructured parent-reports of child problems, whereas the non-clinic data were gathered using 43 standardized checklists (see Lambert et al., 1994). This is a valid point, and studies using the same instruments and methodology employed here are currently under way and may shed some light on this issue. Additionally, factors such as parental thresholds toward problem behaviors and parental attitudes regarding mental health services have also been suggested as possible reasons for the differences between the clinic and non-clinic studies (Lambert et al., 1994). The reasons that the data do not support the problem suppression-facilitation or the adult distress threshold models may also lie in the methodology employed in this study. It has been documented (Aronowitz, 1984) that the age of rrrigration and whether the child migrates with or without the parent may affect the child’s acculturation and adjustment. These demographic data were not collected on the J arnaican immigrants. Thus, the potential confounding effects of these phenomena could not be partialled out. A future study that accounts for the effects of these variables may bolster or refute the absence of support for the models presented here. The similarity across the three groups may also reflect the fact that both Jamaican and U.S. parents have a similar level of concern for a wide range of child behaviors for young children. However, as the child gets older, the level of concern varies in each culture as a function of the child’s gender and age (Lambert et al., 1996). Longitudinal studies that include urban and rural children, account for the child’s age at immigration, and distinguish between children who accompanied their parents and those that joined parents who had migrated earlier, may further clarify these inferences. The factors outlined above do not, however, explain why the findings of the 44 present study are consistent with the comparison of Jamaican and U.S. children ages 6-1 1 (Lambert et al., 1994) but differ from the Jamaican -- U.S. comparisons of adolescents, ages 12-18 (Lambert et al., 1996). Such a pattern suggests that in non-clinic populations, the societal effects on problem type may be more evident in adolescents than in younger children, which may occur because adolescents have lived for longer periods in their respective societies and, consequently, are more exposed to the socializing effects of these societies than their younger counterparts. The fact that the present sample consisted primarily of children under age 12 may also limit the generalizability of these findings. The findings of this study are especially interesting in light of the challenges faced by immigrant children. Although these findings may not be generalizable to other immigrant groups or other geographic locations, this study suggests that Jamaican immigrant children living in New York City are exhibiting approximately the same number, and the same type, of problem behaviors as their non-immigrant counterparts in Jamaica and the United States. Thus, while it is possible that the conflictual values of the new and old cultures (e.g., friends and parents, respectively) may lead immigrants to experience unique difficulties in adolescence, there is no evidence here that the challenges associated with immigration (e.g., learning new societal customs) place pre- adolescent children at an increased risk for behavior and emotional problems. A comparison of the present findings with Lambert et al.’s (1994) study of Jamaican and U.S. children yields some evidence that Jamaican immigrant children became somewhat adapted to the U.S. culture. Specifically, while Lambert et al.’s study 45 found that native Jamaican children scored significantly higher than U.S. children on 27 individual problem items, the present study found that native Jamaican children scored higher than both the Jamaican immigrants and U.S. children on only 7 problem items. Further evidence that Jamaican immigrant children adapted to the U.S. culture is provided by the analyses of the extra Jamaican items. All seven of the significant Jamaican immigrant vs. non-immigrant differences indicated that native Jamaican children received significantly higher scores than the immigrant children. Because these items were specifically deemed clinically relevant to Jamaican children, these results indicate that children’s behavior clearly adapts to the norms of the new culture following immigration. A content analysis of the Jamaican problem items also suggests that Jamaican immigrants may have adapted to the U.S. culture. A review of the items on which J arnaican children received higher problem scores than Jamaican immigrants indicates that most problems (e.g., 119, “Uncooperative”; 120, “Puts self in dangerous situations”) represent the exhibition of distress in the environment and are, therefore, probably externalizing problems. Unfortunately, Since the individual problems have not yet been factor analyzed using a Jamaican clinic-referred sample, and since the CBCL does not include these items, it is difficult to compare native Jamaicans and Jamaican immigrants on syndromes and syndrome groupings that might include the extra items. If the research now in progress indicates that these Jamaican items load on externalizing groupings, the present findings may lend support to the adult distress threshold model. That is, Jamaican parents may have less tolerance for these J amaican-specific behaviors than 46 immigrant parents who have assimilated some U.S. parenting mores. The findings may also support the theory that immigrants who perceive a high degree of social support report greater emotional and physical well being than people perceiving a low degree of support (see Cohen & Wills, 1985). This is suggested by the fact that five of the nine problem items on which U.S. children scored higher than the two Jamaican groups were problems that make up the Depressed Anxious syndrome. Thus, while the total number of problem behaviors may not differ across the groups, according to the parent-reports, the Jamaican children appear to be less internally distressed than U.S. children about their own well-being. While we do not have direct support for the social support phenomenon, since we do not have an empirical measure of social support, such a hypothesis seems plausible given the propensity of Jamaican immigrants to stay in close proximity and support one another Via extended families (Hohn, 1996), as well as their strong sense of ethnic-cultural identity (Brice, 1982). Cross-cultural similarities in gender effects of the reported child problems are also noteworthy. Despite the popular assumption that boys present more problems than girls (see Weisz et al., 1987), this study revealed no Significant gender differences in total problem scores. The similarity of total problem scores also differed from earlier studies of West Indian immigrant children in Britain which found that girls exhibited more externalizing problems than boys (see Lambert et al., 1994 for review). However, the gender similarity in total problem scores is consistent with more recent comparisons of Jamaican and U.S. children and adolescents (Lambert et al., 1994; Lambert et al., 1996). To summarize, lack of Significant group differences in total problem scores and 47 the internalizing and externalizing scales, as well as the relatively few differences in individual problem scores and syndrome scores, suggest important Similarities among Jamaican immigrant and native Jamaican and U.S. children. In contrast to their older counterparts, the brief exposure of these children to the values and mores of their respective societies makes them less likely to exhibit differences in behavior. Longitudinal studies tracking large groups of immigrant and native children are required to more fully understand this phenomenon. LIST OF REFERENCES REFERENCES Achenbach, T.M. (1991a). Manual for the Child Behavior Checklist/4-18 and 1991 Child Behavior Profile. Burlington, VT: University of Vermont, Department of Psychiatry. Achenbach, T.M. (1991b). Manual for the Teacher's Report Form and 1991 TRF m Burlington, VT: University of Vermont, Department of Psychiatry. Achenbach, T.M. (1991c). Mual for the Youth Self-Report and 1991 YSR Brpfili Burlington, VT: University of Vermont, Department of Psychiatry. Achenbach, T.M., Bird, H.R., Canino, G., Phares, V., Gould. M.S., & Rubio-Stipec, M. (1990). Epidemiological comparisons of Puerto Rican and U.S. mainland children: Parent, teacher, and self-reports. Journal of American Academy of Child and Adolescent Psychiag, 2_9_, 84-93. Achenbach, T.M., & Edelbrock, CS. (1983). Manual for the Child Behavior Profile. Burlington, Vermont: University of Vermont, Department of Psychiatry. Achenbach, T.M., & Edelbrock, CS. (1986). Manual for the Teacher's Report Form and Teacher Version of the Child Behavior Profile. Burlington, Vermont: University of Vermont, Department of Psychiatry. Achenbach, T.M., Hensley, V.R., Phares, V., & Grayson, D. (1990). Problems and competencies reported by parents of Australian and American children. Journal of Child Psychology and Psychiatry. 3_1, 265-286. 48 49 Achenbach, T.M., Verhulst, F .C., Baron, G.D., & Althaus, M. (1987a). A comparison of syndromes derived from the child behavior checklist for American and Dutch boys aged 6-1 1 and 12-16. Journal of Child Psychology & Psychiatm & Allied Disciplines, _2_8_, 437-453. Achenbach, T.M., Verhulst, F .C., Baron, G.D., & Akkerhuis, M. S. (1987b). Epidemiological comparisons of American and Dutch children: I. Behavioral/Emotional problems and competencies reported by parents for ages 4 to 16. Journal of American Aaademv of Child and Adolescent Psychim, 2_6, 317-325. Achenbach, T.M., Verhulst, F.C., Edelbrock, C., Baron, G.D., & Akkerhuis, M. S. (1987c). Epidemiological comparisons of American and Dutch children: 11. Behavioral/Emotional problems reported by teachers for ages 6 to 11. Journal of American Academy of Child and Adolescent Psychiag, E, 326-332. American Psychiatric Association. (1987). Diagpostic and Statistical Manual of Mega] Disorders (3rd edirevfl. Washington DC: APA. Aronowitz, M. (1984). The social and emotional adjustment of immigrant children: A review of the literature. International Migration Review, fl, 237-257. Auerbach, J .G. & Lerner, Y. (1991). Syndromes derived from the Child Behavior Checklist for clinically referred Israeli boys aged 6-1 1: A research note. Journal of Child ngchology and Psychiatly, 32, 1017-1024. Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology Monoggrphs. 4 (1, Part 2). 50 Baumrind, D. (1980). New directions in socialization research. American Psychologist, 35, 639-652. Benedict, R. (1934). Anthropology and the abnormal. J oumalpf General Psychology, 10, 59-80. Berry, J .W. (1980). Acculturation as varieties of adaptation. In A.M. Padilla (Ed). Acculturation: Theory, Models, and Some New Findings. Boulder, CO: Westview Press. Berry, J .W. (1986). The acculturation process and refugee behavior. In C.L. Williams & J. Westerrneyer (Eds). Refugee Mental Health in Resettlement Countries. New York: Hemisphere. Berry, J .W., Kim, U., Power, S., Young, M., & Bujaki, M. (1989). Acculturation attitudes in plural societies. Applied Psychology: An International Review, 3, 185-206. Betancourt, H., & Lopez, SR. (1993). The study of culture, ethnicity, and race in American psychology. American Psychologist, 4_8_, 629-637. Birman, D. (in press). Acculturation and human diversity in a multicultural society. In E. Trickett, R. Watts, & D. Binnan, (Eds), Human Diversig: Perspectives of People in Context. San Francisco, J ossey Bass. Brislin, R.W. (1983). Cross-cultural research in psychology. Annual Review of Psychology, fl, 363-400. Bumham, M.A., Hough, R.L., Kamo, M., Escobar, J .I., & Telles, CA. (1987). Acculturation and lifetime prevalence of psychiatric disorders among Mexican Americans in Los AngeleS. Journal of Health and SociQBehavior, _9_, 105-130. 51 Burns, A., Homel, R., & Goodnow, J. (1984). Conditions of life and parental values. Australian Joumal of Psychology. 36, 219-227. Cederblad, M. (1988). Behavioural disorders in children from different cultures. Acta Psychiatrica Scandinavica, 78 (suppl. 344 1, 85-92. Charron, D.W. & Ness, RC. (1981). Emotional distress among Vietnamese adolescents: A statewide survey. Journal of Refugee Resettlement, 1, 7-15. Cochrane, R. & Bal, SS. (1988). Ethnic density is unrelated to incidence of schizophrenia. Britis_h Journal of Psychiatay, fl, 363-366. Cohen, J. (1988). Statistical ppwer analysis for the behavioral sciences (2nd ed.). New York: Academic Press. Culbertson, J .L. (1993). Clinical child psychology in the 1990's: Broadening our scope. Journal of Clinical Child Psychology, _2_2, 116—122. Dohrenwend, B.P., Levav, I., Shrout, P.E., Shwartz, S., Naveh, G., Link, B.G., Skodol, A.E., & Stueve, A. (1992). Socioeconomic status and psychiatric disorders: The causation-selection issue. Science, .22, 946-952. Dombusch, S.M., Ritter, P.L., Leiderman, P.H., Roberts, D.F., & Fraleigh, M.J. (1987). The relation of parenting style to adolescent school performance. _(Lild Development, fl, 1244-1257. Edgerton, RB. (1966). Conceptions of psychosis in four East African societies. American Anthropologist, fl, 408-425. 52 Escobar, J .I., Randolph, E.T., Puente, G., Spiwak, F ., Asarnen, J .K., Hill, M., & Hough, R. (1983). Post-traumatic stress disorder in Hispanic Vietnam veterans: Clinical phenomenology and sociocultural characteristics. J ougal of Nervous and Mental Health, _1_'Ll_, 586-596. Field, H.S. & Armenakis, AA. (1974). On use of multiple tests of Significance in psychological research. Psychological Reports, 3;, 427-431. Freedman, A.M., Kaplan, H.I., & Sadock, B.J. (1976). Modern Synopsis of Comprehensive Textbook of Psychiag/II. Baltimore: The Williams & Wilkins Co. Freud, S. (1930). Civilization and its Discontents. New York: Jonathan Cape & Harrison Smith. Hacket, L. & Hacket, R (1993). Parental ideas of normal and deviant child behaviour: A comparison of two ethnic groups. British Journal of Psychiam, 162, 353-357. Harkness, S., & Super, CM. (1990). Culture and psychopathology. In E.M. Lewis & S.M. Miller (Eds), Handbook of Developmental Psychopathology (pp. 41 -52). New York: Plenum Press. Hohn, GE. (1996). The Effects of Family Functioning on the Psychological and Social Adjustment of Jamaican Immigmt Children. Unpublished dissertation. Hollingshead, AB. (1975). Four Factor Index of Social Status. New Haven, CT: Department of Sociology, Yale University. 53 Holzer, C.E., Shea, B.M., Swanson, J .W., Leaf, P.J., Myers, J .K., George, L., Weissman, M.M., & Bemarski, P. (1986). The increased risk for specific psychiatric disorders among persons of low socioeconomic status. The American Journal of Social Psychiam, 6, 259-271. Kasahara, Y. (1986). Fear of eye-to-eye confrontation among neurotic patients in Japan. In T.S. Lebra & W.P. Lebra (Eds). Japanese Culture and Behavior (rev. ed., pp. 379-3 87). Honolulu: University of Hawaii Press. Kirmayer, L.J. (1991). The place of culture in psychiatric nosology: Taijin Kyofosho and DSMIII-R. The Journal of Nervous and Mental Disease, _1L9, 19-28. Kom, S.J., & Gannon, S. (1983). Temperament, cultural variation and behavior disorder in preschool children. Child Psychiatmand Human Development, Q, 203 -212. Kurian, G. & Ghosh, R. (1978). Changing authority within the context of socialization in Indian families. Social Science. 24-32. Lambert, M.C., Knight, F ., Lyubansky, M., & Achenbach, TM. (1996). Behavior and emotional problems among adolescents of J amaica and the United States: Parent, teacher, and self reports for ages 12 to 18. Under review. Lambert, M.C., Knight, F., Taylor, R., & Achenbach, TM. (1994). Epidemiology of behavioral and emotional problems among children of Jamaica and the United States: Parent reports for ages 6 to 11. Jourpal of Abnormal Child Psychology, 22, 113-128. Lambert, M.C., Knight, F., Taylor, R., & Achenbach, TM. (1996). Comparisons of behavioral and emotional problems among children of J arnaica and the United States: Teacher reports for ages 6 to 11. Journal of Cross-Cultural Psychology. 54 Lambert, M.C., Weisz, J .R., & Knight, F. (1989). Over-and undercontrolled clinic referral problems of Jamaican and American children and adolescents: The cultural general and the cultural specific. Journal of Consulting and Clinical Psychology, 51, 467-472. Lambert, W.E. (1987). The fate of old-country values in a new land: A cross-national study of child rearing. Canadian Psychology, 3, 9-20. Lang, J .G., Munoz, R.F., Bemal, G., & Sorenson, J .L. (1982). Quality of life and psychological well-being in a bicultural Latino community. Hispanic Journal of Behavioral Sciences, 4, 483-450. Leff, J. (1981). Psychiag Around the Globe: A Transcultural View. New York: Marcel Dekker, Inc. Linton, R. (1956). Culture and Mental Disorders. Springfield: Charles C. Thomas. Lyubansky, M., Lambert, M.C., McCaslin, S., & Knight, F. (1995). Aggeement among parents, teachers, and child remrting of behavior and emotional problems in a Jamaican sample. Poster session presented at the XXV Interamerican Congress of Psychology. San Juan, Puerto Rico. Malinowski, B. (1953). Sex and Repression in Savage Socieg. London: Routledge & Kegan Paul. Mead, M. (1928). Coming of Age in Samoa. New York: Morrow. 55 Murphy, H.B.M. (1976). Notes from a theory of latah. In W.P. Lebra (Ed.), Culture-Bound Smdromes, Ethnopsyhiatg, and Alternate Therapies. Honolulu: University of Hawaii Press. Murphy, J. (1976). Psychiatric labeling in cross-cultural perspectives (Yoruba and Eskimo). Science 1_9_1_, 1019-1028. Neugebauer, D.D., Dohrenwend, B.P., Dohrenwend, BS. (1980). The formulation of hypotheses about the true prevalence of functional psychiatric disorders among adults in the United States. In B.P. Dohrenwend, M.S. Gould, B. Link, R. Neugebauer, & R. Wunsch-Hitzig (eds). Mental Illness in the United States. New York: Praeger. Okagaki, L. & Stemberg, R.J. (1993). Parental beliefs and children's school performance. Child Development, Q1, 36-56. Opler, MK. (1955). Cultural perspectives in mental health research. American Journal of Orthopsychiatay, _2_4, 51-59. Pierloot, R.A., & Ngoma, M. (1988). Hysterical manifestations in Africa and Europe: A comparative study. British Journal of Psychiam, 1g, 112-115. Rack, P. (1982). Race Culture and Mental Disorder. New York: Tavistock Publications. Raina, MK. (1975). Parental perception about the ideal child: A cross-cultural study. Joumal of Marriage and the Family, 32, 229-232. 56 Reynolds, D.K. (1987). Japanese models of psychotherapy. In E. Norbeck & M. Lock (Eds). Health, Illness and Medical Care in Japan (pp. 1 10-129). Honolulu: University of Hawaii Press. Rohner, RP. (1984). Toward a conception of culture for cross-cultural psychology. Journal of Cross-Cultural Psychology. 3, 111-138. Rogler, L.H. (1978). Help patterns, the family, and mental health: Puerto Ricans in the United States. Intemational Migration Review. 2, 248-259. Rogler, L.H. (1994). International Migrations: A framework for directing research. AmericaaPsvchologiat, 92, 701-708. Rogler, L.H., Cortes, D.E., & Malgady, R.G. (1991). Acculturation and mental health status among Hispanics: Convergence and new directions for research. American Psychologist, fl, 585-597. Rumbaut, R. (1991). The agony of exile: A study of migration and adaptation of Indochinese refugee adults and children. In F. Ahearn & J. Athey (Eds). Reggea Children: Theog, Research, and Services. Baltimore, MD: The John Hopkins University Press. Rumbaut, R (1994). Data presented at the annual meeting of the American Association for the Advancement of Science. From U.S. Bureau of the Census. m Fofign Born Population in the United States: 1990. CPH-L-98, 1993. Sartorius, N., Jablensky, A., Gulbinat, W., & Emberg, G. (1980). WHO Collaborative Study: Assessment of depressive disorders. Psychosomatic Medicine, m, 743-749. 57 Singer, K., Ney, P.G., & Lieh-Mak, F. (1978). A cultural perspective on child psychiatric disorders. Comprehensive Psychiatfl, _l_9, 533-540. Smith, MG. (1984). Culture Race and Class in the Commonwealth Caribbean. Mona, Jamaica: University of the West Indies, Department of Extramural Studies. Sorenson, S.B. & Golding, J .M. (1988). Suicide attempts in Mexican Americans: Immigration and acculturation issues. Suicide & Life-Threatening Behavior. 18(4). Stanger C. & Lewis, M. (1993). Agreement among parents, teachers, and children on internalizing and externalizing behavior problems. Journal of Clinical Child Pychology, 2, 107-115. Stanger, C., Fombonne, E., & Achenbach, TM. (1993). Epidemiological comparisons of American and French children: Parent reports of problems and competencies for ages 6-11. European Child and Adolescent Psychiatry. Szapocznik, J ., & Kurtines, W.M. (1993). F arnily psychology and cultural diversity: Opportunities for theory, research, and application. American Psychologist, fl, 400-407. Szapocznik, J., Kurtines, W., & Fernandez, T. (1980). Bicultural involvement in Hispanic-American youths. Intemaflrrarl J oumal of Intercultural Relations, 4, 353-365. Tanaka-Matsunri, J. (1979). Taijin Kyofusho: diagnostic and cultural issues in Japanese psychiatry. Ml Medical Psychiatm. 3, 231-245. Taylor, E., & Sandberg, S. (1984). Hyperactive behavior in English schoolchildren: A questionnaire survey. Jourpal of AbnOrmal Child Psychology, Q, 143-156. 58 Thomas, A., Chess, S., Sillen, J ., & Mendez, O. (1974). Cross-cultural study of behavior in children with special vulnerability to stress. In D. Ricks, A. Thomas, & M. Roff (Eds). Life History Research in Psychopathology. (vol. 3, pp. 53-67). Minneapolis: University of Minnesota Press. Thomas, C., & Lindenthal, J. (1990). Migration and mental health among the people of the Caribbean, 1948-1980. International Review of Mental Health, L801), 92- 102. Townsend, J .M. (1978). Cultural Conceptions and Mental Illness. Chicago: University of Chicago Press. Triandis, H., Lambert, W., Berry, J ., Lonner, W., Heron, A., Brislin, R., & Draguns, J. (Eds). (1980). Handbook of Cross-Cultrmal Psvcfflqu, Vols. 1-6. Boston: Allyn & Bacon. Tseng, W., & McDermott. (1981). Culture, Mind & Therapy: an introduction to cultural psychiatpy. New York: Brunner/Mazel. Verhulst, F .C., Achenbach, T.M., Ferdinand, R.F., & Kasius, M.C. (1993). Epidemiological comparisons of American and Dutch adolescents' self-reports. Japanal of the American Academy of Child and Adolescent Psychiatpy, 32,, 1135-1144. Viondi, D.O., Fleming, N.J., & Mintz, S. (1983). Behavior problems of children as perceived by teachers, mental health professionals, and children. Psychology in the m, E, 93-98. Waxler, N., (1979). Is outcome for schizophrenia better in nonindustrial cultures? The case of Sri Lanka. The Journal of Nervops and Mental Disease. 167. 144-158. 59 Weisz, J .R., Suwanlert, S., Chaiyasit, W., & Walter, B.R. (1987a). Over-and undercontrolled clinic-referral problems among Thai and American children and adolescents: The wat and wai of cultural differences. Journal of Consulting and Clinical ngchology, L6,, 719-726. Weisz, J .R., Suwanlert, S., Chaiyasit, W., Weiss, 8., Achenbach, T.M., & Walter, B.R. (1987b). Epidemiology of behavioral and emotional problems among Thai and American children: Parent reports for ages 6 to 1 1. Jourpal of the American Academy of Chiltifid Adolescent Psychiag, 26, 890-897. Weisz, J .R., Suwanlert, S., Chaiyasit, W., Weiss, 8., Walter, B.R., & Anderson, W.W. (1988). Thai and American perspectives on over and undercontrolled child behavior problems: Exploring the threshold model among parents, teachers, and psychologists. Journal of Consulting and Clinical Psychology, §_6__, 601-609. Weisz, J .R., & Weiss, B. (1991). Studying the "referability" of child clinical problems. Journal of Consulting and Clinical Psychology, 2, 266-273. Westermeyer, J ., Bouafuely, M., Neider, J. & Callies, A. (1989). Somatization among refugees: An epidemiological study. Psychosomatics, fl, 34-43. Westermeyer, J. (1973). On the epidemicity of amok violence. Archives of General chhiag, 2_8, 873-876. Wong-Riegler, D. & Quintana, D. (1987). Comparative acculturation of southeast Asian and Hispanic immigrants and sojoumers. Journal of Cross-Cultural Psychology. fl, 345-362. 60 Yu, L. & Harburg, E. (1981). Filial responsibility to aged parents: Stress of Chinese Americans. lntemational Joumal of Group Tensions, 1_1_, 47-58. Zigler, E.F. & Child, IL. (1982). Socialization and PersonalityI Development. Philippines: Addison Wessley Publishing Co. Inc. APPENDICES APPENDIX A Table 1 Percentage of Variance Accounted for by Significant (p < .01) Effects for Culture Group, Gender, and Age on Behavior and Emotional Problem Scores Item Culture“ Genderb Age fl u—ov—Av—du—au—An—Iu—u—nu—hu—A \OOONQUIAUJNt-‘O 99°N9MPP’N . Acts too young Allergy Argues a lot Asthma Behaves like opposite sex Encopresis Bragging Can’t concentrate Can’t get mind of certain thoughts . Can’t sit still, restless, hyperactive . Too dependent . Lonely . Confused . Cries a lot . Cruel to animals . Cruel to others . Day-dreams . Harms self . Demands attention 61 3 2>1;2>3 d 6 2>3>t 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 62 Destroys own things Destroys others’ things Disobedient at home Disobedient at school Doesn’t eat well Poor peer relations Lacks guilt Easily jealous Eats nonfood Fears Fears school Fears impulses Needs to be perfect Feels unloved Feels persecuted Feels worthless Accident prone Fighting Is teased Hangs around children who get in trouble Hears things that aren’t there Impulsive Likes to be alone Lying and cheating Bites fingernails Nervous Nervous movements 5 l>2'. 3>2 5 3>l,2>1 3 2>1;2>3 d 47. 48. 49. 50. 51. 52. 53. 54. 55. 56a. 56b. 56c. 56d. 56e. 56f. 56g. 56h 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 63 Nightmares Not liked Constipated Too fearful or anxious Dizzy Feels too guilty Overeating Overtired Overweight Aches or pains Headaches Nausea, feels sick Eye problems Skin problems Stomach aches, cramps Vomiting . Other physical problems Attacks people Picking Plays with sex parts in public Plays with sex parts too much Poor school work Clumsy Prefers older children Prefers younger children Refuses to talk Repeats certain acts 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 64 Runs away from home Screams a lot Secretive Sees things that aren’t there Self-conscious Sets fires Sexual problems Showing off Shy or timid Sleeps too little Sleeps too much Smears feces Speech problems Stares blankly Steals at home Steals outside home Stores up un-needed things Strange behavior Strange ideas Stubborn, sullen, or irritable Moody Sulks a lot Suspicious Swearing or obscene language Suicidal talk Talks or walks in sleep Talks too much 3 l>2; l>3 31>2 d 4 l>2; l>3 94. 95. 96. 97. 98. 99. 100 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. Teases a lot Temper tantrums Thinks about sex too much Threatens people Thumb sucking Too concerned with neatness . Trouble sleeping Truancy Underactive Unhappy, sad, or depressed Unusually loud Uses alcohol or drugs Vandalism Daytime wetting Wets bed Whining Wishes to be of opposite sex Withdrawn Worrying Other problems Total problems 65 l>2;l>3 3 l>2;l>3 d 3 7 2>1;2>3 5 l>2;l>3 2>l;2>3 12 Note. Items are designated with their CBCL item numbers and summary labels for their content. Numbers in the body of the table indicate the percent of variance in item scores accounted for by each independent variable, where the effect was Significant at p < 0.01. "1 = Native Jamaican children; 2 = U.S. children; 3 == Jamaican immigrant children. bF = Females scored higher; M = Males scored higher. cY = Younger children scored higher. d = Not significant when corrected for the number of analyses. APPENDIX B Table 2. Comparison of Total Problem Scores on the CBCL for Native Jamaican, Native U.S. and Immigmt Jamaican children Culture group N X SD Jamaica 120 25.8 16.2 U.S. 120 28.8 14.0 Immigrant 120 25.5 18.6 Note: Means have been adjusted for SES by ANCOVA. The results are based on 120 items. The range of possible raw scores is O to 240. 66 APPENDIX C Table 3 Percentage of Variance Accounted for by Significant (p < .01) Effects for Culture Group. Gender, and Age on J amaican-spacific Problems Item Immigration Genderb Agec Status" 113. Absentrninded or forgets easily 114. Deliberately annoys others 115. Stays out late at night 1 16. Gambles -- -- 1 17. Doesn’t answer when people talk to him/her 5 ’ -- 1 18. Irritable 5 J -- 1 19. Uncooperative 3 1 d -- 120. Puts self in dangerous Situations 4 J -- 121. Does not respond to punishment 122. Selfish or won’t share 123. Lacks self confidence 124. Makes restless movements during sleep 125. Rude to others 126. Does not care or refuses to do schoolwork 127. Laughs inappropriately 128. Stays away from or leaves home without permission 129. Talks to self 130. Fainting 131. Stones people 67 68 132. Wanders off or walks aimlessly 133. Throws stones at objects 134. Begs at home or on the street 135. Lazy 136. Mischievous or naughty 137. Talks foolishness of nonsense 138. Plays too much 139. Plays too little 140. Gossips 141. Has excessive fantasies or belief in ghosts, etc. 142. Acts too old for his/her age 143. Plays too roughly 144. Seems to be in a trance 145. Has poor relationship with adults 146. Doesn’t attend to hygiene 147. Arrives home too late fiom school Note. Items are designated with their J YC item numbers and summary labels for their content. Numbers in the body of the table indicate the percent of variance in item scores accounted for by each independent variable, where the effect was Significant at p < 0.01. J = Native Jamaican children; I = Immigrant J arnaican children. bF = Females scored higher; M = Males scored higher. cY = Younger children scored higher; 0 = Older children scored higher d = Not Significant when corrected for the number of analyses. .. 2...... .. . . p.31 tr .1; .,.I' 1.4:: slit. Sizi : 2'11). . S., < wo53~