SCHOLASTIC ACHIEVEMENT OF ADOLESCENT REFUGEES FROM CAMBODIA AND CENTRAL AMERICA.

Rousseau, Cecile ; Drapeau, Aline

ABSTRACT

This study investigated the relationship of emotional disturbance and pre- and postmigration environment to the scholastic achievement of adolescent refugees of very different cultural backgrounds. One hundred fifty-two Central American and Cambodian students in six Canadian high schools, as well as their parents, were interviewed to assess the students' emotional problems (using the Youth Self-Report and Child Behavior Checklist) and to determine the pre- and postmigration family environment. The findings indicated that the relationship between the emotional problems and scholastic achievement of teenaged refugees was tenuous. It was concluded that a connection between young refugees' symptomatology and their functional capacity should not be assumed. Nonetheless, certain pre- and postmigration variables, particularly trauma experienced in the homeland, seem to be associated with the academic achievement of some refugees.

The lives of most adolescent refugees have been seriously disrupted by the trauma of war and cultural uprooting. As a result, refugee children may exhibit signs of major psychological distress (Kinzie et al., 1986, 1989). Since research on North American children and adolescents has found a relationship between emotional problems and trouble at school, as well as between the social and family environment and scholastic achievement (Salyer et al., 1991; Weinberg et al., 1989), young refugees, who run a high risk of experiencing mental health problems, might also be expected to have serious academic problems.

Initial investigations of clinical samples of young refugees supported this hypothesis: problems at school were among the main reasons for psychiatric consultation and constituted potential signs of major emotional disturbance (Williams & Westermeyer, 1983; Irwin & Madden, 1985). However, the results of later studies on the scholastic achievement of refugee children and adolescents were more ambiguous. Sack et al. (1995), for instance, found no differences in overall scholastic achievement between teenaged Khmer refugees with and without posttraumatic stress syndrome (PTSS) or, indeed, any psychiatric diagnosis (however, school records could be obtained for only a small fraction of their total sample). The same research team also reported that there was no consistent relationship between English proficiency and a diagnosis of PTSS among Khmer adolescents (Clarke et al., 1993).

Sack et al. (1986) hypothesized that the lack of relationship between emotional problems and scholastic achievement for Southeast Asian refugee children and teenagers can be explained, in part, by the specific form taken by disturbances observed in this population, especially the predominance of internalized symptoms. However, this hypothesis has not been systematically tested, with regard to adolescents, using a transcultural research protocol. Research with younger children indicates that the influence of culture on the relationship between emotional disturbances and scholastic achievement in teenaged refugees may be more complex than it first appears. A study of refugee children from Central America and Southeast Asia found that the relationship between their emotional problems and scholastic achievement was weak in both groups, despite their very different cultural origins (Rousseau et al., 1996). Furthermore, remedial measures implemented by the schools were more strongly associated with externalized sy mptoms than with scholastic achievement per se. Since such symptoms were more prevalent in Central American students, they were the ones more often identified by the school as having learning disabilities, despite the fact that their scholastic achievement was comparable to that of their Asian peers.

Another line of investigation suggests that the scholastic achievement of adolescent refugees is directly influenced by factors specific to their pre- and postmigration environments, such as degree of acculturation, without being mediated by psychopathological symptoms (Rumbaut, 1991; Jupp & Luckey, 1990). More research in this area is needed, however.

In order to develop and implement appropriate support programs for refugee children and adolescents, it is important both to delve further into the possible relationship between psychological symptoms and scholastic achievement in refugees of various cultural backgrounds and to determine the extent to which their specific pre- and postmigration environments influence that achievement. This study attempted to answer two questions: Is there a relationship between scholastic achievement and the symptoms, if any, exhibited by adolescent refugees from two vastly different cultural groups? How do the pre- and postmigration environments relate to the adolescents' scholastic achievement?

METHOD

The target population consisted of Central American and Cambodian teenagers attending Montreal-area schools with large numbers of students of different ethnic origins and an especially high proportion of Central Americans or Cambodians. Potential subjects were selected by cluster sampling, with the sampling base consisting of the enrollment records of six high schools. Two hundred thirteen students met the following selection criteria: (1) enrolled in regular or special education classes in the seventh or eighth grade; (2) of Central American (Honduras, Guatemala, or El Salvador) or Cambodian origin; (3) born outside Canada; (4) neither physically nor mentally disabled. Of the 195who could be contacted, 158 were interviewed after they and their parents had signed a consent form. The total acceptance rate was 81% (158/195): 75% (82/110) of the Central American students and 89% (76/85) of the Cambodian students. Report cards could not be obtained for 6 students, so the final sample was 77 Central Americans and 75 Cambodians.

Interviewers were of the same ethnic origin as the respondents (the adolescents and their mothers, fathers, or guardians). Three categories of variables were considered: scholastic achievement, mental health problems, and pre- and postmigration variables.

Scholastic Achievement

Two variables reflected the adolescents' scholastic achievement: scholastic failure and relative scholastic achievement. Scholastic failure was defined as a failing grade in the current year in French, English, or math, which are compulsory courses in seventh and eighth grades. A measure of relative scholastic achievement (rather than raw grades) was used, because grading varied from class to class and from school to school. First, a score for each compulsory course was determined by calculating the ratio of the adolescent's final grade to the class average. Scores were then summed. A total score of less than 3 indicated that the teen had performed below average scholastically, and a score higher than 3 meant that the teen's performance was above average.

Mental Health Problems

The emotional functioning of each adolescent was assessed from both the parent's and adolescent's points of view. The Child Behavior Checklist (CBCL) was administered to the parent or guardian and the Youth Self-Report (YSR) was utilized with the adolescent (Achenbach & Edelbrock, 1987, 1993). Khmer and Spanish versions of the CBCL and a French version of the YSR were employed.

The CBCL and YSR are often used in transcultural settings and are the assessment tools most widely recommended when taking an empirical approach to mental health problems (Bird, 1996; Verhulst & Achenbach, 1995). The present study employed the internalization and externalization indices of the Cross-Informants Symptom Checklist (CISC), comprising items common to the subdimensions of the CBCL and YSR. Alphas were between .78 and .87 for all scores except the adolescent's degree of internalization as reported by Cambodian parents ([alpha] = .56). In the Cambodian sample, correlations between the parents' reports and those of the adolescents were .26 for the internalization index and .05 for the externalization index. In the Central American sample, the correlations were .28 and .31, respectively. These are fairly weak correlations, but similar discrepancies between the assessments of parents and children have been observed in a number of other studies (Achenbach & Edelbrock, 1987).

Pre- and Postmigration Variables

In addition to the usual sociodemographic variables (adolescent's age and sex; household income, size, and type; and parents' employment status, education, and language proficiency), two other types of context variables were analyzed: the premigration experience (trauma experienced by the family and significant separations experienced by the adolescent in his or her homeland or on the way to Canada) and the postmigration experience (adolescent's degree of acculturation and number of years in Canada). There were two measures of trauma--related to war or violent political conflict--experienced by the adolescent's family prior to and during migration (reported by the parent). These were developed in the course of earlier research (Rousseau et al., 1989, 1996) and are based on the Breslau and Davis model (1987). The first counts the number of traumas (e.g., torture, execution, disappearance, imprisonment) experienced by the family before the adolescent subject's birth. The second counts the traumas experienced a fter the subject's birth, including war- or conflict-related acts of violence that he or she may have witnessed, and whether time was spent in a refugee camp. There were also two measures of separation: from parental figures (mother, father, or other family member who had taken care of the subject for more than six months before he or she left for Canada) and from attachment figures (other caring people who were part of the subject's circle). The number of separations experienced in the homeland or as a result of leaving for Canada were counted (Rousseau et al., 1996).

Family environment, as reported by parents, was measured using the Family Environment Scale (FES; Moos & Moos, 1986). The 90-item FES assesses 10 dimensions of the family environment, but only cohesion (i.e., commitment, help, and support family members provide one another) and conflict (i.e., openly expressed anger, aggression, and conflict among family members) were examined here. There are nine statements for each, with the respondent indicating whether or not they applied to his or her family. The Spanish version of the FES, developed and validated by Szapocznik et al. (1983), was utilized. The Khmer version that was used had been translated by two Cambodian professionals and validated though back-translation in the course of earlier research (Rousseau et al., 1996).

Parental depression was assessed using the 20-item Self-Rating Depression Scale (SRDS; Zung, 1969), with the respondent indicating the degree to which he or she has experienced each item: none or little of the time, some of the time, a good part of the time, or most or all of the time. The Spanish version, validated by Zung, and a Khmer version, which was checked by means of back-translation (Rousseau et al., 1996), were employed.

The Behavioral Acculturation Scale (BAS; Szapocznik et al., 1978), designed for adolescents, was used to measure acculturation. The BAS has been used with Hispanic and Southeast Asian populations (Celano & Tyler, 1991).

Statistical Analysis

Differences between mean scores and between percentages, using confidence intervals (degree of confidence = .95), were calculated and analyzed. In addition, correlational analyses (by sex and ethnic origin) were conducted. The Spearman rank correlation coefficient was used, rather than the Pearson product-moment correlation, because some categories contained relatively few subjects and the distribution of some variables was asymmetrical.

RESULTS

The family environment of the 75 Cambodians was comparable to that of the 77 Central Americans in terms of household income, size and type, and parents' employment status (see Table 1). However, the Cambodian parents tended to have less education compared with the Central American parents. The Cambodian parents also tended to be less proficient in French or English, but because of the small number of subjects at the extreme ends of the scale (no proficiency and good proficiency), it was impossible to confirm the reliability of this observation.

The two groups had similar proportions of males and females, but other sociodemographic characteristics varied by ethnic origin. The mean age of the Cambodian adolescents was slightly lower than that of the Central Americans, while the mean length of time they had been in Canada and mean BAS score were higher for the Cambodians (see Table 1).

The mean number of traumas experienced by the family before the adolescent subject's birth was higher in the Cambodian sample (4.6) than in the Central American sample (0.4), but the mean number of traumas experienced after was lower among the Cambodians (3.6) than among the Central Americans (5.4). The mean number of separations from parental figures and attachment figures was lower in the Cambodian sample (0.3 and 1.2, respectively) than in the Central American sample (1.9 and 2.2).

The percentage of adolescents who failed at least one course (English, French, math) during the current academic year did not vary significantly by ethnic origin, but was especially low (21%) among Cambodian females (see Table 2). The mean scholastic achievement score was virtually identical for Cambodians and Central Americans of either sex, and was close to that of their classmates. The course in which the adolescents did best or worst tended to vary by sex and ethnic origin, but this was not confirmed statistically.

The teenagers' internalization and externalization indices depended on whether the emotional profile was based on the parent's point of view or the teen's. As reported by the parents, the mean internalization and externalization indices were higher for the Central American adolescents (11.7 and 7.7, respectively) than for the Cambodians (5.3 and 4.2). This intercultural difference disappeared when the teenagers' point of view was taken into account: the differences between the mean internalization and externalization indices of Central American adolescents (14.8 and 13.2) and those of the Cambodians (17.4 and 15.5) were not statistically significant.

The relationship between emotional profile and relative scholastic achievement was fairly weak (r ranged from -- .20 to .16)--except for that between scholastic achievement and internalization as reported by the parents of Central American males (r = -.33, CI = -.57: -.03). Similarly, adolescents who failed one or more courses could not be distinguished from those who did not on the basis of their emotional profile scores.

The relationship between premigration experiences and scholastic achievement was also generally quite weak (r ranged from -.20 to .19). However, for Central American males, there was a significant negative correlation (r = - .31) between scholastic achievement and trauma experienced by the family after the birth of the adolescent subject (see Table 3). Moreover, the mean number of traumas experienced before the birth of the adolescent subject was lower among the Cambodians who had failed at least one course ([micro] = 3.2) compared with those who had not failed any courses ([micro] = 5.3), and this did not substantially vary by sex (see Table 4).

A few postmigration variables were significantly associated with scholastic achievement, especially among Central American females (see Table 3). For these females, scholastic achievement was associated with household income (r = .41) and degree of acculturation (r = .34). Family cohesion, at r = .32, approached statistical significance. It should be pointed out that, among Cambodian females, the association with household income was reversed (r = - .37). Among the males, even the highest correlations between scholastic achievement and postmigration variables were not significant, namely degree of acculturation (r = .22) and household income (r = .22) for the Cambodians, and number of years in Canada (r = .22) and family conflict (r = -.24) for the Central Americans.

With only two exceptions, adolescents who failed one or more courses could not be distinguished from those who did not on the basis of the postmigration variables. In the Cambodian sample, 70% of the adolescents whose parents could speak neither English nor French had failed at least one course, while only 30% of the adolescents whose parents' language proficiency was poor and 37% of those whose parents' language proficiency was good had failed at least one course. This association between parents' proficiency in English or French and adolescents' scholastic failure was particularly noticeable among the males. In the Central American sample, academic failure was related to parents' education: 75% of Central American teenagers whose parents had gone no further than elementary school had failed at least one course, while the same was true for 40% of those whose parents had completed high school or vocational school, and 50% of those whose parents had a university degree.

DISCUSSION

The scholastic achievement of the adolescent refugees corresponded by and large to the average achievement of their classmates, a finding that seems to contradict the general perception that refugees have more academic problems (Latif 1988). It must be kept in mind, however, that the schools were chosen for their ethnic mix, and thus the academic performance of those enrolled may not be typical of the entire student population.

Cambodian females fail less often at school than do Central American females or males of either group. According to sources in the Cambodian community, females' greater scholastic success may stem from their close ties to their parents. Often females are responsible not only for providing their parents with emotional support, but also for maintaining their family's honor. Tobin and Friedman (1984) underscored this emotional closeness between Southeast Asian parents and their daughters and the fact that their sons had a harder time identifying with their parents in the new cultural environment.

No significant associations were found between failure at school and internalized or externalized mental health symptoms, whether reported by parents or the adolescents themselves. Academic achievement was associated with internalized symptoms only among Central American males; this association has also been reported for a sample of Central American refugees in elementary school (Rousseau et al., 1996). On the whole, these results confirm the findings of other studies (Sack et al., 1995; Clarke et al., 1993), namely that there is no significant correlation between the scholastic achievement of adolescent refugees and psychological symptomatology. The relationship between scholastics and emotional profile seen in the general population is not found among refugees. This difference may be due to the origin of the emotional problems observed in refugee populations, who are more preoccupied with adapting and grieving, and in the host population, for whom problems less related to acute environmental factors genera lly play a greater role. This hypothesis is in line with Terr's (1983) observations of traumatized North American children, whose scholastic achievement was not affected despite the persistence of severe symptoms.

Results concerning the relationship between scholastic achievement and certain pre- and postmigration factors highlight variations between the two cultural groups and between females and males of the same background. While traumatic family experiences that occurred after the adolescent subject's birth were associated with more scholastic problems for both cultural groups, and especially for Central American males, trauma experienced before the Cambodians were born was, on the contrary, associated less academic failure. Overcompensation, similar to that found among some children of Holocaust survivors (Solkoff, 1992), might explain these results among the Cambodians: after the massacres perpetrated by the Pol Pot regime, children of survivors may feel that they have an obligation to make a success of their lives, because of all those who never had the same chance. Jupp and Luckey (1990), discussing the success of Southeast Asian refugees in mathematics, mention their extraordinary determination to succeed. Th e results here suggest that on top of the value traditionally placed on education by people from Southeast Asia, historical factors involving the transmission of trauma from generation to generation influence the high degree of determination to succeed.

The two variables that pertain to familiarity with the host country--acculturation (BAS score) and number of years spent in Canada--seemed to help the Central Americans avoid scholastic failure. Nonetheless, while the BAS score was significant for females, in the case of males it was the length of time they had been in the country that stood out. These variables were not significant for the Cambodians. While the link between acculturation, length of time in host country, and scholastic aptitude found for the Central Americans has been frequently reported in the literature (Berry et al., 1986), the absence of links between these variables for the Cambodian group is more surprising. Rumbaut (1991) has stated that, for adolescents from Southeast Asia, acculturation may be synonymous with departure from traditional values, which translates into more problems at school. Jupp and Luckey (1990) found that improvement in scholastic achievement among Southeast Asian adolescents was associated with the length of time in the host country, but pointed out that their families had not assimilated. The differences found here between the two cultural groups confirms that the effect of acculturation on scholastic success is not homogeneous, and that complex interactions between the community of origin (its history and values) and the host community (its educational system and the prospects it offers) must be taken into account.

This study had a number of limitations that must be considered when interpreting the results. First, the sample was relatively small, and may not be representative of refugee youths in schools with high numbers of students from the host country. Second, the two groups had different trauma histories: the Cambodian families experienced more traumas before the birth of the adolescent subjects, while the Central American families experienced many more traumas afterward. The contrasting trauma histories may have had an impact on the difference between Central Americans and Cambodians with regard to the risk and protective factors associated with scholastic achievement. Third, the BAS score can in no way be taken as a measure of overall acculturation. In the field of education, a means of determining acculturation in terms of the values that largely determine the motivation to learn would be a welcome addition to the available scales. Fourth, the internalization and externalization indices are based on factor anal yses derived from North American samples; indices derived from Central American and Cambodian samples might include different items.

Despite these limitations, the results suggest that the relationship between emotional disturbance and scholastic achievement in adolescent refugees is tenuous. Further, the findings indicate that scholastic failure and relative scholastic achievement do not measure exactly the same dimension, since different variables were associated with each. In addition, pre- and postmigration factors associated with failure at school may not be universal, but rather specific to a given cultural context. The results also suggest that a family history of trauma, too often considered solely as a risk factor, may play a complex role in scholastic achievement, and may even be a protective factor. Last but not least, the findings point the way for further research into the relationship between psychological symptoms and functional capacity in immigrants and refugees, and indicate the potential danger inherent in assuming that refugee children are a dysfunctional group in the school setting.

Aline Drapeau, Department of Psychiatry, Montreal Children's Hospital, Montreal, Quebec, Canada.

Reprint requests to Cecile Rousseau, Department of Psychiatry, Montreal Children's Hospital, 4018 Ste. Catherine Street West, Montreal, Quebec, Canada H3Z 1P2.

REFERENCES

Achenbach, T. M., & Edelbrock, C.(1987). Manual for the Youth Self-Report and Profile. Burlington, VT: University of Vermont, Department of Psychiatry.

Achenbach, T. M., & Edelbrock, C. (1993). Manual for the Child Behavior Checklist and Revised Child Behavior Profile. Burlington, VT: University of Vermont, Department of Psychiatry.

Berry, J. W., Trimble, J. E., & Olmedo, E. L. (1986). Assessment of acculturation. In W. J. Lonner & J. W. Berry (Eds.), Field methods in cross-cultural research. Beverly Hills, CA: Sage.

Bird, H. (1996). Epidemiology of childhood disorders in a cross-cultural context. Journal of Child Psychology and Psychiatry, 37, 35--49.

Breslau, N., & Davis, G. C. (1987). Post-traumatic stress disorder: The etiologic specificity of wartime stressors. American Journal of Psychiatry, 144, 578--583.

Celano, M. P., & Tyler, F. B. (1991). Behavioral acculturation among Vietnamese refugees in the United States. Journal of Social Psychology, 131,373--385.

Clarke, G. N., Sack, W. H., Ben, R., Lanham, K., & Him, C. (1993). English language skills in a group of previously traumatized Khmer adolescent refugees. Journal of Nervous and Mental Disease, 181, 454--456.

Irwin, A. D., & Madden, C. (1985). A psychoeducational assessment procedure for Southeast Asia refugee students. In T. C. Owen (Ed.), Southeast Asian mental health: Treatment, prevention, services, training and research. Washington, DC: U.S. Department of Health and Human Services Utilization.

Jupp, J. J., & Luckey, J. (1990). Educational experience in Australia of Indochinese adolescent refugees. International Journal of Mental Health, 18, 78-91.

Kinzie, J. D., Sack, W. H., Angell, R., Clarke, G., & Ben, R. (1989). A three-year follow-up of Cambodian young people traumatized as children. Journal of the American Academy of Child and Adolescent Psychiatry, 28, 501-504.

Kinzie, J. D., Sack, W. H., Angell, H. R., Manson, S., & Roth, B. (1986). The psychiatric effects of massive trauma on Cambodian children: I. The children. Journal of the American Academy of Child and Adolescent Psychiatry, 25, 370-376.

Latif, G. (1988). L'ecole quebecoise et les communautes culturelles au Quebec [School and the etbno-cultural communities in Quebec]. Education Department, Government of Canada.

Moos, H. R., & Moos, S. B. (1986). Family Environment Scale manual. Palo Alto: Counsulting Psychologists Press.

Rousseau, C., Corin, E., & Renaud, C. (1989). Conflit arme et trauma: Une etude clinique chez des enfants refugies latino-americains [Armed conflict and trauma: A clinical study of Latin American refugee children]. Canadian Journal of Psychiatry, 34, 376-385.

Rousseau, C., Drapeau, A., & Corin, E. (1996). School performance and emotional problems in refugee children. American Journal of Orthopsychiatry, 66, 229-251.

Rousseau, C., Drapeau, A., & Corin, E. (1997). The influence of culture and context on the pre- and postmigration experience of school-aged refugees from Central America and Southeast Asia in Canada. Social Science and Medicine, 44, 1115-1127.

Rumbaut, R. G. (1991). The agony of exile: A study of the migration and adaptation of Indochinese refugee adults and children. In F. L. Ahearn, Jr., & J. L. Athey (Eds.), Refugee children: Theory, research, and services. Baltimore: Johns Hopkins University Press.

Sack, W. H., Clarke, G. N., Kinney, R., Belestos, G., Chanrithy, H., & Seeley, J. (1995). The Khmer Adolescent Project: II. Functional capacities in two generations of Cambodian refugees. Journal of Nervous and Mental Disease, 183, 177-181.

Sack, W. H., Richard, H. A., Kinzie, J. D., & Roth-Ber, D. J. (1986). The psychiatric effects of massive trauma on Cambodian children: II. The family, the home and the school. Journal of the American Academy of Child and Adolescent Psychiatry, 25, 377-383.

Salyer, K. M., Holmstrom, R. W., & Washpitz, J. D. (1991). Learning disabilities as a childhood manifestation of severe psychopathology. American Journal of Orthopsychiatry, 61, 230-240.

Solkoff, N. (1992). Children of survivors of the Nazi Holocaust: A critical review of the literature. American Journal of Orthopsychiatry, 62, 342-358.

Szapocznik, J., Kurtines, M. W., & Foote, H. F. (1983). Some evidence for the effectiveness of conducting family therapy through one person. Journal of Consulting and Clinical Psychiatry 51, 889-899.

Szapocznik, J., Scopetta, M. A., Kurtines, W., & de los Angeles Aranalde, M. (1978). Theory and measurement of acculturation. Interamerican Journal of Psychology, 12, 113-130.

Terr, L. (1983). Life attitudes, dreams, and psychic trauma in a group of "normal" children. Journal of the American Academy of Child and Adolescent Psychiatry, 22, 221-230.

Tobin, J. J., & Friedman, J. (1984). Intercultural and developmental stresses confronting Southeast Asian refugee adolescents. Journal of Operational Psychiatry, 15, 39-45.

Verhulst, F. C., & Achenbach, T. M. (1995). Empirically based assessment and taxonomy of psychopathology: Cross-cultural applications. European Journal of Child and Adolescent Psychiatry, 4, 61-76.

Weinberg, W. A., McLean, A., Swider, R. L., Nickols, A. S., Rintelmann, J. W., & Erwin, P. R. (1989). Depression, learning disability and school behavior problems. Psychological Reports, 64, 275-283.

Williams, L. C., & Westermeyer, J. (1983). Psychiatric problems among adolescent Southeast Asian refugees: A descriptive study. Journal of Nervous and Mental Disease, 171, 79-85.

Zung, W. W. K. (1969). A cross-cultural study of symptoms in depression. American Journal of Psychiatry, 126, 154-159.