Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Considerations for Assessing Asian-American Youth

Asian-American youth in the USA are a heterogeneous group originating from approximately 52 different Asian countries from the Asia continent, Southeast Asia, and the Indian subcontinent, with over 100 spoken languages and dialects. These youth originate from different social classes, speak different languages with varying degrees of fluency, practice different religions, and adhere to diverse cultural values (Huang, 1994). One of the fastest growing ethnic/racial minorities in the USA, the Asian-American population has increased at a higher rate than all other race groups in the country, growing by 46 % from 11.9 million in 2000 to 17.3 million in 2010 (Hoeffel, Rastogi, Kim, & Shahid, 2012). Extant research has contradicted earlier notions of Asian-Americans as the “model minority,” and literature has documented various social, educational, health and mental health disparities among this population along with comparable mental health illness and psychological distress prevalence rates to the general population (see Lee, Lei, & Sue, 2001 for a review). Over 12 % of Asian-Americans live in poverty, higher than the 9.9 % rate of poverty among non-Hispanic whites (US Census, 2010), and the illiteracy rates of Asian-Americans is 5.3 times that of non-Hispanic whites (Le, 2013). Southeast Asians have the highest high school dropout rates in the country, 33 % of Asian-American students in public high schools drop out or do not graduate on time, and 24 % of Asian-Americans over the age of 25 do not have a high school degree equivalent (Le, 2013). Relatively little is known about the mental health status and needs of Asian-American youth, but existing research suggests that there are higher rates of depression, anxiety, and suicide among this group. Thirty percent of Asian-American girls in grades 5 through 12 reported depressive symptoms, higher than the rates reported in White girls (22 %), Latino girls (27 %), and African-American girls (17 %) (Schoen et al., 1997). Seventeen percent of Asian-American boys in grades 5 through 12 reported physical abuse, as compared to 8 % among White boys (Schoen et al., 1997). Further, generational differences and conflict regarding family values and beliefs may cause additional strain, stress, anxiety problems, and behavioral difficulties for these youth. Equally concerning is the evidence that Asian-American families tend to delay professional service seeking even when clinical problems arise (USDHHS, 2001), and Asian-American youth have consistently demonstrated a higher level of unmet mental health needs and lower levels of mental health service engagement than non-Hispanic White youth (Li & Seidman, 2010).

In order to address Asian-American youths’ emotional/behavioral problems, accurate appraisal of their mental health status is essential to identifying those who could benefit from psychological intervention and to designing effective treatment plans. There are numerous assessments that exist for the evaluation of youth emotional and behavioral problems in the USA. However, many of these have not been explicitly validated on Asian-American youth populations. Most existing assessments have demonstrated their psychometric properties on predominantly non-Hispanic white samples, and their results may not be readily generalized to Asian-American youth due to varying cultural factors that are salient to the psychological development of youth from different cultures. Some measures have been translated for use with indigenous populations, but it is not always clear from existing psychometric studies whether these translated versions have achieved cross-cultural equivalence at the individual test item level and/or at the construct level. In addition, there is a lack of psychometric information regarding how these translated versions of assessments that have originated from Asian countries might be used with Asian-American populations.

For example, one cannot assume that a questionnaire that is developed and normed on one cultural group can be used to assess the same construct in another cultural group (Arnold & Matus, 2000). When administering any instrument as part of a clinical assessment, one must consider how the respondent may interpret the items. This is particularly important if the respondents are immigrants, nonnative English speakers, or originate from a culture different from the one in which the instrument was developed and validated (Gee, 2004). Individuals from varying cultural groups may ascribe different meanings to symptoms than were intended by the authors of the original instrument. For example, symptom expression has been shown to vary in different Asian-American ethnic groups (USDHHS, 2001; Paniagua, 2000), and several culture-bound syndromes have been documented in different Asian populations in the DSM-IV-TR (APA, 2000). For instance, Vietnamese Americans have been shown to endorse symptoms of depression that include somatic complaints (e.g., pains, poor appetite) as well as other symptoms that could not be readily translated into an English language equivalent (Kinzie et al., 1982). For Asian-American youth, symptoms of anxiety may also consist of somatic complaints, along with sleep disturbance, and poor school performance (Huang, 1997). Lack of awareness or attention to these potential cultural differences is likely to lead to misclassification or underdiagnosis of emotional/behavioral disorders in Asian-American youth.

Another important issue to consider that bears implications for cross-cultural equivalence is the use of multiple informants in the assessment of psychopathology in Asian-American youth. Weisz, McCarty, Eastman, Chaiyasit, and Suwanlert (1997) asserted that the study of youth emotional/behavioral problems “is inevitably the study of two phenomena: the behavior of the child, and the lens through which adults view child behavior” (p. 569). The use of several reporters for youth psychopathology assessment is common for many existing evaluation tools (evidenced by the availability of different informant forms for the same questionnaire), and the information gained in this way is highly valuable. However, reports from youth, parents, and teachers have consistently shown low cross-informant agreement (Achenbach, McConaughy, & Howell, 1987), and the research base suggests that inter-informant discrepancies may vary systematically by race/ethnicity (Lau et al., 2004). These racial/ethnic differences suggest that there may be problems in meeting one or more of the conditions set forth by Marsella and Kameoka (1989) when evaluating cross-cultural equivalence and interrater reliability: (1) Linguistic equivalence and conceptual equivalence require that raters have a shared understanding of the characteristic being rated, and an understanding of the range of behaviors that are representative of that characteristic; (2) Metric equivalence requires that raters have a shared metric so that they may accurately scale behaviors that are related to the characteristic being rated; and (3) Normative equivalence requires that raters can determine the presence and absence of behaviors relevant to the characteristic being rated.

Closer examination suggests that youth–teacher and parent–teacher disagreements may be explained in part by cultural differences and/or racial ethnic biases (Lau et al., 2004). There is evidence that culture may influence adults’ attitudes toward youth behaviors and the determination of whether an observed behavior constitutes a problem necessitating professional intervention (Lau et al., 2004). For example, consistent with traditional Buddhist values, Thai parents rated both undercontrolled problems (e.g., disobedience, fighting) and overcontrolled problems (e.g., shyness, fear) as less serious, less worrisome, and more likely to improve with time compared to American parents (Weisz et al., 1988). Another example utilizes two traditional Chinese concepts emerged from Confucian thought: (1) “Chiao Shun,” which refers to the Chinese youth behaving in ways consistent to his or her training (i.e., teaching the child appropriate conduct by exposing him/her to explicit examples of proper behavior and restricting exposure to undesirable behavior in the context of a supportive, highly involved, and physically close parent–child relationship), and (2) “Guan,” which equates parental care, concern, and involvement with firm control and governance of the child (Chao, 1994, 2000). Traditional parents may not view some internalizing behaviors on Western assessment scales such as passivity, obedience without question, and lack of assertiveness as clinical problems but rather desired behaviors in their child consistent with the concepts of “Chiao Shun” and “Guan.” Similarly, differing views of desired attachment styles in traditional Japanese families may lead Japanese mothers to view some internalizing behaviors as normative and expected in their children. For example, the indigenous Japanese concept of “Amae” which refers to relationships that involve both attachment and dependence (Doi, 1989; Emde, 1992) have been shown in research to translate to more expressed feelings of sadness due to separation between Japanese mothers and children than between American mothers and children (Mizuta, Zahn-Waxler, Cole, & Hiruma, 1996). These findings suggest that concepts of sadness and loss have different meanings to parents and children in Japan than in the USA, and can potentially significantly affect assessment and intervention of attachment and youth emotional problems (Rothbaum, Weisz, Pott, Miyake, & Morelli, 2000).

Complicating the multi-informant assessment picture are linguistic challenges when assessing families in which parents may speak only one language (from their country of origin) proficiently, and in which the youth may speak only English proficiently or are bilingual. In 2011, the foreign born from Asia represented over one-fourth of the total foreign-born population in the USA (Gryn & Gambino, 2012), which suggests that a substantial amount of these foreign-born Asian-Americans are not native English speakers. For these individuals, issues regarding acculturation status and relatedly, linguistic considerations, must be considered when conducting a multi-informant assessment. For example, when translated versions are available, is it appropriate to administer the scale in one language to the parent, and in another language to the youth? When administered in different languages, how can we ensure that the raters have a shared, or culturally equivalent, understanding of the characteristic being rated? If there are intergenerational differences between the parent and youth regarding level of acculturation to the mainstream American culture, these cultural differences may affect the determination by the parent and youth of whether a behavior constitutes a problem necessitating professional intervention. Specifically, parents who are less acculturated to mainstream American culture compared to their child may rate a particular behavior as non-problematic, whereas their child may rate a particular behavior as problematic, or vice versa, leading to larger parent–youth disagreement on an assessment. Further, if only and English version of an assessment is available, would it be appropriate to request the youth to serve as an interpreter? If so, how might this affect the validity of the parent’s report of the youth’s emotional/behavioral problems provided that the youth is hearing these responses and also doing the written transcribing when it is a written measure? The existing literature is limited in providing guidelines on how to handle the above outlined issues, particularly with multi-informant assessments with children, and the field would benefit from more research in this area.

The perceptions of adult observers who are not part of the nuclear family may be particularly subject to the influence of race-related beliefs about base rates of youth psychopathology (Lau et al., 2004). Teacher expectancies for student achievement and classroom behaviors have been related to student race (Dusek & Joseph, 1983). Teachers may have the tendency to view Asian-American youth as model students who are diligent and respectful, yet passive with low assertiveness and poor social competence (Bannai & Cohen, 1985; Schneider & Lee, 1990). Similarly, teachers may rate overcontrolled behaviors such as worry and shyness as more representative of Asian students than students of other ethnicities (Chang & Sue, 2003). Intergenerational differences may also affect youths’ perception of their own behaviors. For example, youth from immigrant families may adopt dominant American cultural values, norms, and behaviors more easily than their parents (Szapocznik & Truss, 1978), leading to more pronounced parent–youth disagreements about youth emotional/behavioral problems (Lau et al., 2004).

Another crucial issue to contemplate is whether syndromes of child psychopathology in the USA readily translate to those of various Asian cultures. Syndromes, groups of emotional/behavioral problems that co-occur, often are the bases for clinical assessment, yet may not be similar across cultures. For example, the Achenbach scales (e.g., Achenbach, 1991, 1995) utilize syndromes as organizational units for its subscales. Although information on youth psychopathology has grown quickly in recent decades, most studies have been conducted in Western countries. This restricted range of cultures sampled in empirical research may limit our cross-cultural understanding of psychopathology as findings from Western countries may not generalize to other cultures (Weisz, Weiss, Suwanlert, & Chaiyasit, 2003). Weisz and colleagues (2003) found that some child psychopathology syndromes may not match up very well between Thai and US children. For example, the narrowband syndromes did not show strong concordance, and only two of the eight broadband comparisons showed substantial or almost perfect agreement. Research is limited in this area, but if child psychopathology syndromes differ systematically across cultures, this would undermine efforts to develop a common classification system to be used for all cultures. Relatedly, the construct validity of assessment tools developed using syndromes from a common classification system as a basis may be questionable when applied to different cultures. The appropriateness of clinical assessments conducted in Asian cultures using scoring categories developed from US samples may also be called in to question. Thus, it may be wise to assess syndromal similarity across cultures before applying the syndrome-based categories of one culture to another.

Without an accurate assessment of the clinical picture, motivating and engaging the youth and his/her family for professional services and designing and implementing effective treatment plans become even more of a challenge for this already underserved population.

Recommendations

Many Asian cultural groups prefer more indirect ways of communication that involve a combination of verbal, explicit, nonverbal, and implicit responses. For example, traditional Chinese groups tend to rely on indirect, more complex methods of communication like indefinite comments while Westerners tend to adopt direct and simple methods (Su, Wang, Fan, Su, & Gao, 2008). These differences in communication and response style may render direct responses to face-valid surveys (the majority of most emotional/behavioral assessment tools) more difficult for Asians compared to their American counterparts. Further, many Asian groups both in the USA and internationally tend to avoid mental health services because of their fear and rejection of mental health disorders (USDHHS, 2001). Thus, the adoption of a more indirect communication style coupled with the stigma against mental illness may significantly influence how Asian groups respond to items on a face-valid mental health assessment. They may be more likely to underreport symptoms and this has been demonstrated consistently in the literature (USDHHS, 2001). Misunderstandings may also ensue as symptom expression is often different for various Asian groups (e.g., the tendency to somaticize symptoms for many Asian groups; USDHHS, 2001).

The way the items on an instrument are constructed may also alter response patterns for different Asian groups. For example, many Asian languages bear linguistic structure differences to English. In Korean, the verb comes at the end of the sentence, as do positive and negative valence words. Therefore, Korean respondents may give more attention to the last portions of sentences. In English, the verb does not always come at the end of the sentence. Therefore, the way in which an item question is structured may inadvertently direct an individual’s attention towards or away from an item’s focal content as conceptualized by the original test developers.

For the reasons above, interviews with the youth’s primary caregivers may be a very helpful supplement, as they may provide information regarding the meaning of specific symptoms and shed light on how cultural factors may play a role in the expression of these emotional/behavioral problems. Of course, the information gained by adult informants needs to be weighed according to what is known from the existing research regarding informant disagreements regarding youth emotional/behavioral problems and the role that culture may play in whether a problem is viewed as worthy of clinical attention, and the preliminary findings that syndromes established with Western samples may not be similar across different Asian cultures. For the above reasons, it may also be important to assess the acculturation levels and/or ethnic identity of both the parents and youth in order to gain some additional context around the degree to which each informant may ascribe to more westernized conceptualizations of psychopathology and syndromes.

Overall, an integrative approach should be utilized for clinical assessment which combines both standard and ethnocultural assessment strategies such as the one described by Huang (1994). An ecological approach to assessment provides multiple sources of data essential to a holistic picture of the youth, and combines the evaluation of domains of a standard assessment (e.g., physical appearance, language, affect, family composition, demographics, communication patterns) as well as an ethnocultural assessment (e.g., generational status, acculturation level, ethnicity and self-concept, migration history, salience of ethnicity, history with cultural differences; Huang, 1994).