Xenophobia, ethnic community, and immigrant youths' friendship network formation.

Tsai, Jenny Hsin-Chun

Xenophobia, ethnic community, and immigrant youths' friendship network formation are associated with the provision of social support, access to support, physiological responses, psychosocial development, health behaviors, and health status across the life span (Berman & Syme, 1979; Kuo & Tsai, 1986; Levy-Storms & Wallace, 2003; Schwartz, Dodge, Pettit, Bates, & The Conduct Problems Prevention Research Group, 2000; Seeman & Berkman, 1988). Children who have peer relationship difficulties are at risk of behavioral, emotional, and academic problems, and even adult psychopathology (Parker & Asher, 1987). Immigrant youth leave their relatives, friends, and other social ties behind in their home countries during their immigration processes (Aronowitz, 1984). In new countries, they face the challenges of rebuilding friendship and other social networks at school and in the community. A difficult transition in forming new friendship networks is likely to threaten immigrant youths' optimal psychosocial development and promote vulnerability to being marginalized. The purpose of this paper is to examine how sociocultural context enables immigrant youth to rebuild their friendship networks.

BACKGROUND

The Concept of Social Network

Social network, as an analytic concept, describes the complex interpersonal linkages in a social system and is generally divided into structural and interactional dimensions. Size or range, density, proximity, type of relationship, homogeneity, and reachability are components of the structural dimension. Characteristics of individual ties such as frequency of contact, intimacy, multiplexity, duration, reciprocity, and durability, are considered as the interactional dimension (Berkman, Glass, Brissette, & Seeman, 2000; Mueller, 1980; O'Reilly, 1988).

There are three types of networks: first order, second order, and extended (Mueller, 1980). Most studies have focused on the first order network, which consists of direct connections that individuals have with others (O'Reilly, 1988). Family, relatives, friends, neighbors, and co-workers are generally considered as members of the first order network. The second order network refers to indirect linkages that individuals develop through members of their first order network. The extended network is the connection individuals have with larger populations through the second order network.

How do people form social networks? Psychologists have identified several factors: physical proximity or propinquity, probability of interaction, frequency of exposure, reciprocity of liking and self-disclosure, individual judgement of the other person's characteristics (e.g., physical features, social skills, academic achievement, similarity), and developmental stages (Berndt & Ladd, 1989; Fehr, 1996; Furman, 1989; Hallinan & Williams, 1989). Anthropologists and sociologists have taught us that lives are structured by meanings, rules, conventions, or habits we adhere to as social beings. We draws on our cultural knowledge to organize our behavior, understand others and ourselves, and make sense of the world. In other words, formation of a social network is a sociocultural experience. For example, studies from the United States showed that children younger than five are already aware of language, food, skin color, hair differences, and facial characteristics as culturally defined "racial" markers. They use these markers to define their social boundaries and obtain control in their interactions with other children (Ramsey & Myers, 1990; Van Ausdale & Feagin, 1996). With age, "racial" attitudes become more crystallized and consistent across the cognitive, affective, and behavioral components and play an important role in shaping older children's cross-race relationships (Semons, 1991). In order to better understand youths' formation of their friendship networks, it is necessary to examine that experience in the macro sociocultural context that shapes the making and breaking of social networks. The factors identified by the psychologists, however, fall short of such a consideration.

Social Networks, Immigration, and Immigrant Youth

Loss of social networks due to immigration is not an uncommon experience for immigrants (Aroian, 1990; Aronowitz, 1984). Most literature on immigrant and social networks emerged from anthropology and sociology. Studies generally focus on the functions of immigrants' coethnic networks, including emotional and cultural support, economic survival, social mobility, social interactions, and ethnic identity (Chrisman, 1981; Hagan, 1998; Haines, Rutherford, & Thomas, 1981; Portes, 1995).

Few studies retrieved from the health science literature explored compositions of immigrants' social networks and relations with social support. Research with Middle-Eastern immigrants (May, 1992) and elder Korean immigrant women (Kim, 1999) living in the United States for more than 10 years showed that coethnics (foreign-born and American-born) accounted for the majority of their social networks and sources of support. Lynam's 1985 study provided further insight into factors influencing immigrant women's formation of social networks. In her study, she found that the women interacted with their kin, people from their ethnic or religious communities, and those outside their communities. Their decisions in establishing relationships were affected by several factors: similar backgrounds, common beliefs, hunger for friendships and meaningful relationships, desire to maintain privacy, and wanting to feel they belonged and were understood by people in Canada.

Literature on adult immigrants' social network formation is limited, and is even more scarce for immigrant youth. Information on immigrant youths' social networks (mostly friendship type networks) is usually embedded in the literature that investigates assimilation, ethnic identity, or adaptation. Nonetheless, several ideas from that literature are pertinent to immigrant youths' friendship network formation. Moran et al. (1993) in England found that Asian children were bullied more frequently than Caucasian children in junior and middle schools. Asian children's like or dislike of other children was associated with that experience. Moreover, name calling about one's color or race was the most common form of bullying for Asian but not Caucasian children. Closs et al.'s 2001 study noted similar findings with black refugee children in Scotland. As they described it, "Some reaching out towards friendship with success, but they also experienced racism, exclusion and struggle for socio-linguistic understanding. Trauma, loss, and grief were inhibitors to friendship for teenagers. Almost all, especially black children, experienced abuse/bullying" (p. 142). The refugee children often interacted with other non-Scots, if there were any. Older children experienced more difficulties in finding friends than those who were younger. Studies in the United States (Lee, 2001; Semons, 1991) and Canada (Phan, 2003) also suggested that racism was a salient factor in interfering with the first, or foreign born, and second generation immigrant youths' friendship development with Americans or Canadians. It was not unusual to find these youths spending most of their time with those who were racially or ethnically similar to them. Using U.S. national survey data on students in grades 7 to 12, Quiliian and Campbell (2003) found a consistent friendship pattern among students with white Hispanic, non-white Hispanic, or Asian backgrounds. Immigration generation order did not make any difference in that pattern. Additionally, they found that small racial minorities in a school increased the odds of same-race friend selection. They suggested that these students altered their friend-making behavior for "[their] desire for reasons of social support" (p. 560) when there were only few students of similar ethnic heritage in their school.

Background Summary

Relationships between social network and social support, health behaviors, and health outcomes have been studied for decades. Researchers have examined factors associated with social network formation; yet, there has been little research with immigrant populations. These limited studies with immigrant adults and youth suggest that, regardless of age, similarity of racial or ethnic background may be an important determinant in immigrants' choices of social network members. Coethnics in a nearby geographic area, such as school or ethnic community, provide opportunities for same-race interactions. When the number of coethnics shrinks, at least for immigrant youth, their likelihood of choosing friends similar to their own racial or ethnic backgrounds increase. Racism appears to be a salient factor that creates a sociocultural context in the school environment, which deters.: relationships between immigrant and native-born youths.

THE STUDY

Aims

The study examined how the societal context in which immigrant families live shape their immigration and resettlement experience. Specifically, this study describes recent Taiwanese immigrant families' adaptation process in the broader societal context of the United States and their psychosocial responses. Partial findings of the study are reported in this paper.

METHOD

Data were from a critical ethnographic study conducted between 1998 and 2000 in a northwest metropolitan area of the United States. Critical ethnography, a branch of ethnography, asks researchers to go beyond describing "what is" and explore how "what is" has been structured by societal ideologies and social control (Thomas, 1993).

Participants

A sample of 16 Taiwanese youths (male = 9, female = 7) participated in the study. Participants were from the community and selected for their potential for providing a range of information about the research questions. They had lived in the United States for 9 months to 10 years, with the majority (n = 8) less than five years. Their age ranged from 8 to 21 years (mean age = 16.1 years, SD = 3.7) Of the 16 youths, 12 had learned English prior to coming to the United States. All youths were in school. Two young adult participants also had a part-time job in restaurants.

Thirteen parents also participated to provide their accounts of their children's immigration and resettlement experience. They were all in their forties (mean age = 45.3 years, SD = 2.4). Nine of the 13 parents had at least 16 years of education. All were employed or had businesses in Taiwan; yet, only 7 of them were employed and 1 had a small business at the time.

Data Collection

Semi-structured in-depth interviews were the primary data sources. The youths were asked to describe what it was like to live in the United States, what their understandings or interpretations of those experiences were, and how they reacted in those circumstances. Parents were invited to provide their observations and perspectives about their child's post-immigration experiences.

The interviews were conducted primarily in Mandarin Chinese, the official language of Taiwan and the investigator's native language, and were tape-recorded with participants' consent. Participants were interviewed alone and, sometimes, with their counterpart (parent or child). Lengths of interviews ranged from 30 minutes to 3 hours with one participant and 1.5 to 3 hours with both parent and child. After the interviews, the investigator recorded her observations, personal reactions and reflections, and hypotheses.

Each participant completed the Demographic and Immigration Questionnaire (DIQ) at the end of the first interview. This questionnaire, developed in Mandarin Chinese, was designed specifically for this study in order to collect participants' demographic information and immigration history (e.g., year of arrival, immigration mechanism). All parents and three quarters of the youths chose the Mandarin Chinese over the English version.

In addition to the interviews and DIQ, the investigator had been involved in the local Chinese (including Taiwanese) community for two years. For example, she was a regular mentor in the youth and family program at a local Chinese agency. Anecdotes heard during those encounters provided additional valuable information about immigrant youths' experiences and helped generate guidelines for the interviews.

To enhance the scientific rigor of the study, parents were invited to participate in data generation triangulation. Their stories confirmed the youths' accounts and enriched the understanding of the youths' friendship formation experience. The investigator engaged in ongoing consultation with senior researchers and colleagues with diverse backgrounds to ensure that cogent alternative explanations were considered. Interpretation of the data with participants in various interviews were thus validated. The investigator consistently monitored her interactions with participants during the interviews. In addition, reflections on her interactions with the participants and how her points of view (i.e., biases, ideologies, and background) affected her work were documented in her field notes (Lowenberg, 1993; Thomas, 1993). This self-reflection heightened her awareness of her influence on data gathering and helped modify her approaches.

Ethical Considerations

Protection of participants was achieved through compliance with the procedures approved by the human subject committee at the investigator's affiliated university. Prospective participants were contacted, using the approved script, to confirm their interest and an interview appointment. At the beginning of each interview the investigator explained the study once again. Written consent for participation was obtained after their questions were answered. The verbal consent was offered as an alternative to those who did not read the consent form or were uncomfortable in signing it (Lipson & Meleis, 1989). No participant refused to provide a written consent.

Data Analysis

Descriptive statistics (e.g., frequency, mean) were used to analyze the DIQ data. Narrative analysis (Riessman, 1993) guided the interview analysis. The investigator and her research assistant, both of whom are first-generation Taiwanese immigrants, transcribed all interviews from audiotapes. The investigator validated the transcripts completed by her assistant and made corrections as needed prior to analysis. In order to include what was said (the content), what was not said, and how things were said in the interviews (Riessman, 1993), both verbal and nonverbal messages (e.g., emotional expression as well as pauses and repetitions) were preserved in the transcripts.

The investigator then read each transcript for preliminary analysis, which included examination of what participants presented in the interviews and the ways in which their stories were structured according to their everyday living context. The latter was the basis for answering the research questions. Specifics related to the research questions were identified for in-depth analysis. Comparisons within and between interviews were made during analysis to explore similarities and differences, identify overlap or duplications of codes, and make sure that no important codes were overlooked. The demographic information and field notes served as background information that helped in the interpretation of interview data.

FINDINGS AND DISCUSSION

All youths had developed friendship networks at the time they entered the study. Peers they had met at school or through family connections in the Chinese community were identified as their new social ties. The majority of their school peers were those who attended the English as a second language (ESL) program. Having non-immigrant peers might occur if a youth was taking only the mainstream classes. Regardless of their interactions at school, no high school and college-age participants considered "Americans" (referring to Euro-Americans primarily) as part of their friendship networks.

Xenophobia and Meaning of Limited English Proficiency (LEP) and ESL

LEP made social interactions with Americans difficult. Despite the fact that most of the youth had learned English in Taiwan, speaking to Americans was reported as nerve-wracking. Some teenagers perceived that their "poor English" could irritate American peers. One 16-year-old boy described his early days at school: "At that time, [you] don't want to ask too much. You're afraid of creating hassles for others.... You ask a lot of questions.... People would finally feel annoyed." When it came to interacting with native speakers, being alone and trying not to irritate others became a coping strategy to avoid nervousness and embarrassment. This may explain why Americans were rarely considered as part of their friendship networks by the youths.

Other data and further analysis, however, suggest that LEP not only influences the youths' perceptions about American peers' possible reactions to them and actions they took to stay at a distance from Americans, it creates an invisible reinforcement for segregation between immigrant and American youths. LEP becomes a label of "the other" (vs. Americans as "we")--the people who come from "other" countries. The ESL program is intended to promote immigrant youths' language proficiency and assimilation. In turn, it inadvertently perpetuates the identity of "otherness" that is imposed on immigrant youth and creates opportunities for them to be targets of discrimination.

Participants, including parents, reported that youths were mistreated or taken advantage of by American peers because of their LEP. Thus, interactions with American peers would occur only when "it's really necessary," as a 19-year-old boy said. The following account of a 14-year-old girl best illustrates how the labels "LEP" and "ESL" could foster immigrant youths' exclusion of Americans from their friendship networks. She said, when she had just arrived in the United States, some people in her elementary school would "trick" her because she spoke little English. They would purposely make jokes with unclear English. In middle school, she witnessed her school mates mocking ESL students by speaking English with an accent. Once she was out of the ESL classes, her experiences began to change. She was free from mistreatment and accepted by Americans in the mainstream classes because "they know that I am not in ESL." Her friends in the ESL classes, however, continued to be mistreated and separate from the mainstream group. Regardless of the fact that she spoke English like a native speaker when I met her, her close friends were two girls from her ESL classes.

The culturally defined "racial" markers already influence friendship formation among children younger than five years of age and continues to affect children's interactions with and choice of peers as they get older (Ramsey & Myers, 1990; Semons, 1991; Van Ausdale & Feagin, 1996). Same-race friendships were much more common than cross-race friendships (Closs et al., 2001; Phan, 2003; Quillian & Campbell, 2003). In her book, Tatum (1999) argued that in asking why a group of Black (or Asian or Latino) teenagers often sit together, we need to understand that "in a racially mixed setting, racial grouping is a developmental process in response to an environmental stressor, racism. Joining with one's peers for support in the face of stress is a positive coping strategy" (p. 62). Lee (2001) reported that Hmong high school youth who were ESL or former ESL students were frustrated with the social environment of the school, "where it is difficult to make 'American' friends" (p. 515). Foreign-born Hmong youth clustered with other foreign-born Asians at the side of the cafeteria. The findings of this study suggest that besides racism that is skin-color based discrimination, irrational hatred or fear of foreigners (xenophobia) should also be considered in immigrant youths' formation of friendship networks.

Most immigrants are distinct from the dominant White Americans in physical appearance on which "race" is based, but also their language skills and accent. Non-White Americans are considered as the "other" because of their "race" and the influence of racism (Bannerji, 1995). When comparing non-Whites to immigrants, however, the former is much more like "us" to the dominant white Americans than the latter. The label of "LEP" and "ESL" overtly signifies immigrant youths' "outsider" or "foreigner" status and defines the social boundary between immigrant and American youth. Immigrant youth may choose to exclude Americans from their friendship networks for their own psychological well-being.

Role of the Local Ethnic Community and Cultural Knowledge

The Chinese community in the area where this study was conducted was established in the 1860s and consists of Chinese from China, Hong Kong, and Taiwan as well as their American-born offspring (Davila, 2000). All participating families had connections with other Chinese families through their relatives or friends already in the area or directly through the local Chinese agencies. Through family connections or programs sponsored by the local Chinese agencies, the youth were exposed to other Chinese of similar ages.

All of the youth reported having friendships with Chinese from China, Hong Kong, or Taiwan. A few also had American-born Chinese (ABC) in their friendship networks. Bringing two individuals in contact through physical proximity is the first step in developing a social tie (Fehr, 1996). Consistent with the reviewed literature, existence of an ethnic community, such as this one, increases the likelihood of encounters with those who are alike. Because they were having difficulty making connections with Americans, the Chinese community became particularly important to them. In addition, it served as a support system, as Tatum (1999) suggested, for their psychosocial development during their early immigration and resettlement transition.

The existence of the Chinese community promoted the youths' same-ethnic friendship formation. Knowing about teen culture in several Chinese societies and the United States promoted friendships and provided common topics for conversation. One teen was encouraged by his mother to attend church with her so that he could meet more people. However, he quickly stopped going because "most people in that church are from Hong Kong. There is no topic at all to talk about with them. Really the topics are different (sounding bothered). I don't know the stuff they talked about." He had to talk about different topics with his friends from China in the beginning before he met other Taiwanese youths. "What was talked about in Taiwan was different from what was talked about with these classmates from China," he said. One mother used to encourage her two teens to participate in a Taiwanese youth group to make friends. They were unwilling to go because, as the son explained, "They [the youth group] have many ABCs. Their topics are different." Despite the fact that the youth shared the same ethnic heritage and were able to communicate with one another, cultural knowledge about their potential friends' country of origin had a role in determining their friendship selection.

Social networks provide a basis for social interactions, intimacy, companionship, cultural identification, access to resources and material goods, and social mobility (Berkman et al., 2000). Friendship networks, one type of primary social network, are associated with a youth's psychosocial development and mental health status (Hartup, 1996; Parker & Asher, 1987). The findings of this study suggest that we must rethink our understanding of immigrant youths' friendship network formation. Factors identified by psychologists are important in explaining how people form social networks; however, that framework is insufficient for capturing macro sociocultural contexts that shape formation processes and network structures. Racism has been identified as one salient contextual factor for first- and second-generation immigrant youths' friendship formation in white dominant countries. The findings of this study further show that xenophobia increases first generation, or foreign-born, immigrant youths' vulnerability to being marginalized and threatens their psychosocial development. Thus, macro sociocultural factors should be routinely considered in the psychosocial assessment of these youths. School nurses, psychologists, counselors, and teachers are in the best position to advocate for immigrant youth by raising schools' awareness of the effects of racism and xenophobia as well as the unintended consequences of ESL programs for first generation immigrant youths' psychosocial development. In addition, these professionals should work together to develop policies that will promote a friendly school environment for immigrant youths.

While this study found racism and xenophobia to be risk factors for immigrant youths' marginalization, ethnic communities, on the other hand, showed potential as a protective factor for the youths' psychosocial development, particularly during the early resettlement phase. Quiliian and Campbell (2003) found that small racial minorities in a school increased the likelihood of same-race friend selection because of "[their] desire for social support" (p. 560). This study was conducted in an area where there was an established Chinese community; thus, the generalization of the findings is limited. We need additional research to better understand the formation of friendship networks in immigrant youth who settle in areas where there are none or only a small number of coethnics available for interaction and support. Meleis (1999) states, "Cultural competence encompasses knowing the different ways in which marginalization in society tends to increase health risks for clients and communities and the potential range of outcomes due to being at the periphery of decision making" (p. 12). Longitudinal studies examining the psychosocial development of immigrant youth growing up with or without an ethnic community will be helpful during the critical time period for early intervention in order to prevent further marginalization and poor outcomes of immigrant youth.

REFERENCES

Aroian, K. J. (1990). A model of psychological adaptation to migration and resettlement. Nursing Research, 39, 5-10.

Aronowitz, M. (1984). The social and emotional adjustment of immigrant children: A review of the literature. International Migration Review, 18, 237-257.

Bennerji, H. (1995). Thinking through: Essays on feminism, marxism, and anti-racism. Toronto, Canada: Women's Press.

Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health: Durkheim in the new millennium. Social Science & Medicine, 51, 843-857.

Berman, L. F., & Syme, S. L. (1979). Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents. American Journal of Epidemiology, 109, 186-204.

Berndt, T. J., & Ladd, G. W. (Eds.). (1989). Peer relationships in child development. New York: John Wiley & Sons.

Chrisman, N. J. (1981). Ethnic persistence in an urban setting. Ethnicity, 8, 256-292.

Closs, A., Stead, J., Arshad, R., & Norris, C. (2001). School peer relationships of "minority" children in Scotland. Child: Care, Health and Development, 27, 133-148.

Davila, F. (2000, October 15). The struggle for the soul of Seattle's International District. The Seattle Times, pp. A1, A12-13.

Fehr, B. (1996). Friendship processes. Thousands Oaks, CA: Sage. Furman, W. (1989). The development of children's social networks. In D. Belle (Ed.), Children's social networks and social supports (pp. 151-172). New York: John Wiley & Sons.

Hagan, J. M. (1998). Social networks, gender, and immigrant incorporation: Resources and constraints. American Sociological Review, 63, 55-67.

Haines, D., Rutherford, D., & Thomas, P. (1981). The case for exploratory fieldwork: Understanding the adjustment of Vietnamese refugees in the Washington area. Anthropological Quarterly, 54(2), 94-102.

Hallinan, M. T., & Williams, R. A. (1989). Interracial friendship choices in secondary schools. American Sociological Review, 54, 67-78.

Hartup, W. W. (1996). The company they keep: Friendships and their developmental significance. Child Development, 67, 1-13.

Kim, O. (1999). Predictors of loneliness in elderly Korean immigrant women living in the United States of America. Journal of Advanced Nursing, 29, 1082-1088.

Kuo, W. H., & Tsai, Y. M. (1986). Social networking, hardiness and immigrant's mental health. Journal of Health and Social Behavior, 27, 133-149.

Lee, S. J. (2001). More than "model minorities" or "delinquents": A look at Hmong American high school students. Harvard Educational Review, 71, 505-528.

Levy-Storms, L., & Wallace, S. P. (2003). Use of mammography screening among older Samoan women in Los Angeles county: A diffusion network approach. Social Science & Medicine, 57, 987-1000.

Lipson, J. G., & Meleis, A. I. (1989). Methodological issues in research with immigrants. Medical Anthropology, 12(1), 103-115.

Lowenberg, J. S. (1993). Interpretive research methodology: Broadening the dialogue. Advances in Nursing Science, 16, 57-69.

Lynam, M. J. (1985). Support networks developed by immigrant women. Social Science and Medicine, 21, 327-333.

May, K. M. (1992). Middle-Eastern immigrant parents' social networks and help-seeking for child health care. Journal of Advanced Nursing, 17, 905-912.

Meleis, A. I. (1999). Culturally competent care. Journal of Transcultural Nursing, 10, 12.

Moran, S., Smith, P. K., Thompson, D., & Whitney, I. (1993). Ethnic differences in experiences of bullying: Asian and white children. British Journal of Educational Psychology, 63, 431-440.

Mueller, D. P. (1980). Social networks: A promising direction for research on the relationships of the social environment to psychiatric disorder. Social Science & Medicine, 14A, 147-161.

O'Reilly, P. (1988). Methodological issues in social support and social network research. Social Science & Medicine, 26, 863-873.

Parker, J. G., & Asher, S. R. (1987). Peer relationships and later personal adjustment: Are low-accepted children at risk? Psychological Bulletin, 102, 357-389.

Phan, T. (2003). Life in school: Narratives of resiliency among Vietnamese-Canadian youths. Adolescence, 38, 555-566.

Portes, A. (Ed.). (1995). The economic sociology of immigration: Essays on networks, ethnicity, and entrepreneurship. New York: Russell Sage Foundation.

Quillian, L., & Campbell, M. E. (2003). Beyond black and white: The present and future of multiracial friendship segregation. American Sociological Review, 68, 540-566.

Ramsey, P. G., & Myers, L. C. (1990). Salience of race in young children's cognitive, affective, and behavioral responses to social environments. Journal of Applied Developmental Psychology, 11, 49-67.

Riessman, C. K. (1993). Narrative analysis. (Vol. 30). Newbury Park, CA: Sage.

Schwartz, D., Dodge, K. A., Pettit, G. S., Bates, J. E., & The Conduct Problems Prevention Research Group. (2000). Friendship as a moderating factor in the pathway between early harsh home environment and later victimization in the peer group. Developmental Psychology, 36, 646-662.

Seeman, T., & Berkman, L. F. (1988). Structural characteristics of social networks and their relationship with social support in the elderly: Who provides support. Social Science & Medicine, 26, 737-749.

Semons, M. (1991). Ethnicity in the urban high school: A naturalistic study of student experiences. Urban Review, 23, 137-158.

Tatum, B. D. (1999). Why are all the black kids sitting together in the cafeteria? and other conversations about race. New York: Basic Books.

Thomas, J. (1993). Doing critical ethnography. Thousand Oaks, CA: Sage.

Van Ausdale, D., & Feagin, J. R. (1996). Using racial and ethnic concepts: The critical case of very young children. American Sociological Review, 61, 779-793.

The study was supported in part by the American Nurses Foundation, the Hester McLaws Nursing Scholarship Fund of University of Washington School of Nursing, Psi Chapter of the Sigma Theta Tau International, and Robert Gilbert Foundation of the Association of Child and Adolescent Psychiatric Nurses. The author gratefully thanks the families for participating in the study and Drs. Doris Boutain and Sally O'Neill for their thoughtful comments on earlier drafts of this paper.

Requests for reprints should be sent to Jenny Hsin-Chun Tsai, University of Washington, School of Nursing, Psychosocial and Community Health, Box 357263, Seattle, WA 98195-7263. E-mail: jennyt@u.washington.edu