Ethnicity & Health

This article was downloaded by:[University of Arizona] On: 6 December 2007 Access Details: [subscription number 767968599] Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Ethnicity & Health Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713421971

Bicultural Stress and Adolescent Risk Behaviors in a Community Sample of Latinos and Non-Latino European Americans

Andrea J. Romero a; Daniel Martinez a; Scott C. Carvajal ab a Mexican American Studies & Research Center, Department of Psychology, College of Social and Behavioral Sciences, University of Arizona, Tucson, AZ, USA b The Melvin and Enid Zuckerman College of Public Health, University of Arizona, Tucson, AZ, USA Online Publication Date: 01 November 2007 To cite this Article: Romero, Andrea J., Martinez, Daniel and Carvajal, Scott C. (2007) 'Bicultural Stress and Adolescent Risk Behaviors in a Community Sample of Latinos and Non-Latino European Americans', Ethnicity & Health, 12:5, 443 - 463 To link to this article: DOI: 10.1080/13557850701616854 URL: http://dx.doi.org/10.1080/13557850701616854

PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Downloaded By: [University of Arizona] At: 17:06 6 December 2007

Ethnicity and Health Vol. 12, No. 5, November 2007, pp. 443463

Bicultural Stress and Adolescent Risk Behaviors in a Community Sample of Latinos and Non-Latino European Americans Andrea J. Romeroa, Daniel Martinezb & Scott C. Carvajala,c a

Mexican American Studies & Research Center, Department of Psychology, College of Social and Behavioral Sciences, University of Arizona, Tucson, AZ, USA; bMexican American Studies & Research Center, Department of Psychology, College of Social and Behavioral Sciences, University of Arizona, Tucson, AZ, USA & cThe Melvin and Enid Zuckerman College of Public Health, University of Arizona, Tucson, AZ, USA

Objectives. The study examined the relation between adolescent risk behaviors and bicultural stress due to discrimination, immigration, and acculturation factors. We hypothesized bicultural stress would be related to increased risk behavior and depressive symptoms independent of socioeconomic status, ethnic self-identification, and acculturation. Design. Middle school student participants (n519; median age 14) completed a self-report questionnaire on their risk behaviors, psychosocial antecedents, and sociodemographic factors. Latino (304) and non-Latino European American (215) students were surveyed through a large, urban, West Coast US school district. Results. More bicultural stress was significantly related to reports of all risk behaviors (i.e. smoking, drinking, drug use, and violence) and depressive symptoms. Further, bicultural stress was a robust explanatory variable across sub-groups, and appears largely independent from depressive symptoms.

Correspondence to: Andrea J. Romero, PhD, Mexican American Studies and Research Center, University of Arizona, Ce´sar Cha´vez Building No. 23, P.O. Box 210023, Tucson, AZ 85721-0023, USA. Tel: 1 520 626 8137. Email: [email protected] ISSN 1355-7858 (print)/ISSN 1465-3419 (online) # 2007 Taylor & Francis DOI: 10.1080/13557850701616854


444 A. J. Romero et al.

Conclusion. The hypotheses were supported. Bicultural stress appears to be an important underlying factor for health disparities among US adolescents. Future research may consider promoting well-being in majority, as well as minority adolescents, through targeting sources of bicultural stressors or examining ways to moderate their effects on adolescent risk behaviors. Keywords: Bicultural; Adolescent; Risk Behavior; Health Disparities Adolescents’ risk behaviors are central indicators of their health (Jessor 1998; CDC 2004). Latinos now represent the largest ethnic minority in the US, and they are disproportionately represented in younger populations (US Census Bureau 2001). Latino ethnicity is a pan-ethnic label that typically refers to a common language (Spanish) or a common ancestry (Americas south of the US), encompassing the same ethnics groups as the label ‘Hispanic’, which is more frequently used in US government records (Marıń & Marıń 1991). While for some health outcomes, Latinos are healthier than would be expected based on ethnic trends in education and income (the ‘Hispanic epidemiological paradox’; see De La Torre & Estrada 2001; Farley et al. 2005), this has not been the case regarding adolescent risk behaviors. Rather, disparities research has identified that, relative to other US youth, Latino youth report early age of onset of substances (Ellickson et al. 2004; Kandel et al. 2004; De La Rosa et al. 2005; Johnston et al. 2005; Stueve & O’Donnell 2005) as well as high rates of violence and depressive symptoms (Hovey & King 1996; CDC 2004). Additionally, some subgroups of Latino youth based on nationality and/or acculturation are at further disparate risk; however, much of this research has been inconsistent due to the limited measures of cultural change (e.g. lack of orthogonal or multidimensional assessments) utilized in national surveys (Carvajal et al. 2002; Unger et al. 2004a; Borrell 2005). One way to advance health disparities research is by examining plausible psychosocial factors that may better explain disparities than broad ethnic labels (e.g. Walsh et al. 2000; Nazroo 2001; Bradby 2003). Scholars have increasingly articulated the negative impact of stress on health outcomes, particularly stress related to discrimination and prejudice (Karlsen & Nazroo 2001; Williams et al. 2003; Mays et al. 2007). The current work aims to elucidate factors underlying disparities in Latino and non-Latino European American youths by examining the role of bicultural stressors and considering multiple potentially relevant socio-demographic variables, including current measures of cultural orientation within Latinos. Research on Acculturative Stress There is substantial literature on immigrant populations’ perception of stress due to such factors as prejudice, discrimination, minority status, and pressure to learn English (Cervantes et al. 1991; Hovey 2000; Berry, 2003). Some researchers define


Ethnicity & Health 445

prejudice as the everyday negative interactions between ethnic groups that may result from ethnocentrism, and suggest that it may be a major source of stress for many minority groups (Allison 1998; Clark et al. 1999; Sellers & Shelton 2003; Meyer 2004). In adolescence, peer discrimination may be particularly stressful because positive acceptance by peers is highly salient (Williams-Morris 1996; Romero & Roberts 1998; Arnett 2002; Spears-Brown & Bigler 2005). Stress may come not only from pressure to adopt the US majority culture, in fact, some acculturative stress models suggest adolescents may perceive pressure to maintain their culture of origin from peers within their own ethnic group. For example, same-ethnic group youth may use derogatory terms, such as ‘oreo/apple/banana/coconut’, to imply that youth act ‘White’ despite their exterior ethnic phenotype (Oetting & Beauvais 19901991; LaFromboise et al. 1993). Acculturation-related stressors experienced by adolescents occur not only within peer contexts, but in other contexts such as school or family (Compas et al. 1987). For example, intergenerational gaps and conflicts within the family may be a source of stress in families, where adolescents are acculturating faster than their parents (Szapocznick et al. 1978; Gil et al. 2000). Additionally, dichotomies between language at home and school may create bilingual environments such that youth feel pressure to speak English and Spanish fluently (Romero & Roberts 2003a). While many acculturation models emphasize cultural exchange between cultures (see Zagefka & Brown 2002; van de Vijver & Phalet 2004), most acculturation stress research has been guided by linear models of acculturation with a focus on stress experienced by immigrant groups acculturating to majority cultures (Carvajal et al. 2002). It was posited that minority ethnic groups experience unequal pressure to assimilate due to less power and lower status; whereas majority ethnic groups have minimal or no pressure to acculturate to the minority culture (Berry 2003). Thus, it is implied that stress from navigating more than one culture would be predominantly experienced by minority groups. However, orthogonal models of acculturation have demonstrated that individuals can adhere to more than one culture independently (Oetting & Beauvais 19901991; LaFromboise et al. 1993; van de Vijver & Phalet 2004), and related research has demonstrated that some majority youth also experience cultural context stressors similar to those described in acculturative stress models (e.g. Romero et al. in press). Conceptualizing and Measuring Bicultural Stress While the preponderance of the focus of acculturative stress research is on stresses experienced by immigrant groups, there may be intra and inter-cultural pressures that impact broader populations of youth. For example, ethnocentrism among minority groups may lead to negative stereotypes and discriminatory behavior of European Americans, the majority group (Romero & Roberts 1998). In a previous study of discrimination among a multiethnic group of adolescents, perceived racial prejudice reports were generally higher for African American and Latino youth,



though all adolescent groups, including European Americans, reported some level of perceived prejudice (Fisher et al. 2000). Based on the conceptualizations of orthogonal acculturation models, we define ‘bicultural stress’ as the perception of stress due to everyday life stressors that result from pressure to adopt the majority culture as well as pressure to adopt minority cultures for youth in multiethnic environments (Romero & Roberts 2003a). To address cultural stressors for minority and majority adolescents, Romero et al. (2003a) developed, validated and examined a general measure of bi-cultural stress using a subjective, context-specific (Lazarus 2000) model. Examples of stressors include those due to perceived discrimination, negative stereotypes, intergenerational acculturation gaps, and pressure to speak multiple languages (Romero et al. 2003a). This measure addresses the cultural context of stressors that are not only increasingly present for US youth, but will be increasingly prevalent throughout industrial nations where there are substantial demographic changes in immigrant or minority youth populations (Arnett 2002; Zagefka & Brown 2002). Though bicultural stress is higher among minority groups, such stressors have been reported among European American individuals (e.g. Romero et al. in press). We believe this perspective holds important contributions to the understanding of youth stressors globally, not only for immigrants, but also for minority and majority ethnic groups (e.g. Bourhis et al. 1997), as adolescents increasingly navigate between multiple cultures and languages in school and at home. Also, living in an environment with more than one culture may create stresses from negotiating more than one set of values, norms, and identities (Stonequist 1935; LaFromboise et al. 1993). Furthermore, regions in the US that do not have one ethnic majority are becoming increasingly prevalent, which may provide unique opportunities to study more multidirectional exchanges of culture and perceptions of stress (US Census Bureau 2001). Bicultural Stress and Health Disparities While the relation between bicultural stressors and risky health behaviors has not been examined using the Bicultural Stress scale (Romero et al. 2003a), some related stress indicators have been associated with risk behaviors (e.g. Booker et al. 2004; Bennett et al. 2005; Krieger et al. 2005). Also, Latino adolescents who experience acculturative stress and/or discrimination report more negative mental health outcomes (e.g. Cervantes et al. 1991; Vega 1998; Samaniego & Gonzales 1999; Hovey 2000; Szalacha et al. 2003; Krieger et al. 2005). Measures of subjective bicultural stress and/or bicultural conflict have been examined in Latinos, and positively associated with depressive symptoms (Benet-Martinez et al. 2002; Romero & Roberts 2003a; Romero et al. in press). Lazarus (1997) argues that the subjective component of stress is important to represent psychological reactions to acculturation, and to more clearly link acculturative stress to mental health outcomes. Further, utilizing



subjective measures clarifies that not all individuals may experience equivalent levels of stress due to bicultural experiences (LaFromboise et al. 1993). The Current Study The aims of the present research are to examine the relations between a new measure of bicultural stress (Romero & Roberts 2003a; Romero et al. in press) and markers of adolescent health in a diverse urban West Coast US community sample. This sample is further unique in that there is no ethnic majority in the target population, with the largest ethnic groups and those included in the current study being Latino and non-Latino European American adolescents. Guided by general subjective stress and coping models (e.g. Folkman & Lazarus 1980; Lazarus 2000) we expect that the subjective perception of bicultural stress will be associated with more frequent risk behaviors and more depressive symptoms. Further, we will contextualize bicultural stress with ethnicity as well as other socio-demographic factors (Phinney et al. 1998; Goodman et al. 2003; Weden & Zabin 2005). Research indicates the perception of stress associated with acculturation is subjective, and may vary based on demographics and language preferences (Padilla et al. 1986; Rogler et al. 1991; LaFromboise et al. 1993; Lazarus 1997), though we expect bicultural stress to be a stronger underlying explanatory variable for disparities than demographic variables (Walsh et al. 2000). Based on theory and previous empirical findings we posit the following hypotheses: Hypothesis 1: Bicultural stress will be related to higher levels of reporting all risk behaviors and depressive symptoms. Hypothesis 2: The relations of bicultural stress to risk behaviors and depressive symptoms will account for much of the disparities attributable to other sociodemographics. Additionally, the generality and robustness of the findings will be further tested in exploratory analyses. Specifically, the relations among bicultural stress, risk behaviors, and depressive symptoms will be examined for consistency consistent across subgroups (e.g. gender or ethnic differences in the role of stress; Zambrana & Silva-Palacios 1989; Booker et al. 2004). Also, as depressive symptoms are a wellknown factor related to adolescents’ health, we will examine whether bicultural stress is distinct from depressive symptoms in relation to risk behaviors. This approach may further extend the potential contribution of bicultural stress to understanding health disparities research. Finally, the potential influence of school ethnic and other demographic characteristics (Sampson et al. 2005) on the estimation of the relations between stress and the health outcomes will be examined.



Methods Participants and Procedures Participants (n 519) included self-identified Latinos (n304) and non-Latino European Americans (n215) from a larger cross-sectional sample. The larger sample was of 8th grade students who were surveyed from a large urban school district in the West Coast of the US. Trained staff collected the data at the schools in the fall of 2000. Active parent and student consent procedures were employed, and participants were recruited for this sample through randomly selected classrooms from all seven middle schools of that district. The median age of participants was 14 years, and 54% were female. There was evidence that our study sample well represents the local population. The ethnic variation of the larger cross-sectional survey matched the records of the participating, no ethnic-majority, K12 school district (Carvajal et al. 2004). Also, the educational background reported for the participants’ parents was consistent with the educational background as reported on the US Census (2000) for the county in which the participating district resides (e.g. respondents indicated 38% of their mothers and 41% of their fathers attained a bachelors degree; the county Census data, collected about one year prior to our survey, showed a rate of 40.4% for all adults). While not specific to the Latino and non-Latino European American sub-samples employed for this study, the response rate for our larger study was 61%. Additional aspects of the random sampling and data collection procedures for our cross-sectional data are detailed elsewhere (Carvajal et al. 2002, 2004; Romero et al. in press). Coding of ethnicity followed the guidelines of the National Longitudinal Study of Adolescent Health (2001, 2001a) study principally funded by the US National Institute of Child Health and Human Development, and one of the most comprehensive surveys of youth’s health status. Youth were categorized as Latinos based on affirmative responses as follows (a) pan-ethnic label Latino/a or Hispanic, (b) one of the specific Latino sub-groups (e.g. ‘Mexican American’, ‘Cuban American’, or ‘Puerto Rican), (c) youth who reported more than one ethnic group including Latino and either European American or American Indian. Mexican American (78%) was the most frequent specific Latino sub-group, followed by those of Central American descent (9%), and Puerto Rican descent (7%). Youth were categorized as non-Latino European American when they identified as ‘White (Caucasian/nonHispanic) and no other specific group. It should be noted that ethnic representation varied substantially between schools, the lowest and highest proportion of Latinos and European Americans ranged from 16 to 81%, 6 to 52%, and 10 to 39%, respectively. Also, 35% of all respondents of the larger survey were excluded from the current investigation as they were from another ethnic group or multiple ethnic groups. They were not included because we did not have sufficient sample size to generate stable ethnic specific risk behavior estimates*the primary aims of the current study.



Measures Participants completed a self-report questionnaire (English and Spanish versions offered). The variables of focus in the current study included assessments of sociodemographics, bicultural stress, risk behaviors and depressive symptoms. Bicultural stress Bicultural stress was measured from a 20-item scale (Romero & Roberts 2003a). This scale (see Appendix A) addresses everyday stressors within schools, peers, and family contexts, and include factors related to discrimination, intragroup pressures (Rodriguez et al. 2002; Romero & Roberts 2003a), intergroup conflict, and acculturation. The range of responses to each stressor are ‘Does not apply’ (coded as 0), ‘Not stressful at all’ (coded as 1), ‘A little bit stressful’ (coded as 2), ‘Quite a bit stressful’ (coded as 3), and ‘Very stressful’ (coded as 4). These items provide information on whether youth experienced the stressors when dichotomized (does not apply or is present) or the intensity of stressfulness if kept continuous (‘Not stressful at all’ to ‘Very stressful’ if it was experienced). An average score was taken across all items and used in the analyses of the current study. Internal consistency of the full scale was high within each ethnic group (European Americans a0.94 and Latinos a0.94). Socioeconomic status (SES) Socioeconomic status (SES) has a pervasive impact on health and its adequate measurement is critical in ethnic health disparities research (Karlsen & Nazroo 2002). To best represent SES from the available youth self-reports, we used three indicators: parental education, perceived SES, and home characteristics. Data from adolescents on their parent’s education level is generally considered to be a marker of social class with relatively high validity (Wills et al. 1995). Parental education was measured by the average of the mother’s and father’s education. Responses ranged from 1 to 4, where 1 represented less than high school, and 4 represented graduating from college. Perceived socioeconomic status and household structure have proven to be particularly useful proxy measures for SES among adolescent samples, as many may not reliably know their parents’ education or income (Gore et al. 1992). Perceived SES (14) was the average of three questions (‘Compared to others in your ‘‘city’’/‘‘school’’/‘‘community’’ do you think your family’s income is ‘‘somewhat less’’/ ‘‘a little less’’/‘‘a little more’’, ‘‘somewhat more’’ than most others’) where higher scores indicate higher SES. Home characteristics (02) reflected the sum of two dichotomous items reflecting whether they lived in a house (yes 1) and whether their rooms to resident ratio in their home (calculated from separate questions) was greater than or equal to one (yes 1). These indicators of SES were significantly (pB0.05) positively associated with each other overall and within all ethnic groups;



thus we created an overall SES variable by equally weighting parental education, perceived SES and household SES into a composite variable (Romero et al. in press). Generational status Generational status categories for Latinos were based on the generation definitions from Cue´llar et al. (1995). Generational status is proxy for acculturation that has been found associated with language acculturation and cultural values among adult samples (Cue´llar et al. 1995), as well as with adolescent populations (Carvajal et al. 2002; Romero & Roberts 2003b). Latinos were identified as first generation if they indicated they were born outside the US. Latinos were identified as second generation if they were not first generation, but reported at least a parent was born outside the US. Latinos were identified as third generation or higher if participants marked they and their parents were born in the US. Acculturation Language acculturation among Latinos was measured by an 8-item version of the Bidimensional Acculturation scale (Marıń & Gamba 1996), modified for use in multiethnic adolescent populations (Carvajal et al. 2002). Responses to each item ranged from ‘Never’ (coded as 1) to ‘Always’ (coded as 4). Four items from this scale measure English orientation (a0.76) and four items from the scale measure language other than English orientation (a0.90). The English and Other language orientation summed scales were dichotomized into ‘high’ or ‘low’ based on the median score (Cue´llar et al. 1995), necessary given the high ceiling of the English orientation scale for the Latino participants. Parallel to related work (Romero & Roberts 2003a; Romero et al. 2004; Romero et al. in press), three language preference groups best represented these Latino youth: (1) Other language preference (low English); (2) Bilingual preference (high English and high other language); (3) English preference (high English and low other language). Risk behaviors Risk behaviors assessments paralleled those in US national surveys, including the Youth Risk Behavior Survey (CDC 2004), the Monitoring the Future study (Johnston et al. 2005), and the National Longitudinal Study of Adolescent Health (2001). These included self-reports of (1) ‘any drinking in lifetime’; (2) ‘heavy drinking in the past month’ (having five or more drinks in a row in the past 30 days); (3) ‘any smoking in lifetime’; (4) ‘smoking in last 30 days’; (5) ‘illegal drug use in lifetime’ (marijuana or other illicit substances); (6) ‘weapon carrying in the past 30 days’; and (7) ‘having been in a physical fight in the past 12 months’. These variables were dichotomized prior to conducting inferential analysis, where no engagement of such behavior was coded as ‘0’, and any engagement of such behavior was coded as ‘1’.



Depressive symptoms An aim of the current study was to relate stress to somatic and affective symptomatology rather than to categorize individuals with clinical levels of depression (Crockett et al. 2005). Thus, depressive symptoms were measured with an 8-item shortened version of the CES-D (Radloff 1977). The measure includes assessments of the frequency of experiencing four major symptom domains of depression within the past week: negative affect, absence of positive affect (reversed scored), somatic complaints, and interpersonal problems (Rushton et al. 2002). There is evidence supporting the measures applicability for ethnic minority adolescents (Carvajal et al. 2002; Crockett et al. 2005). Items referred to the past seven days; respondents exhibiting four or more symptoms in the previous week were considered high for depressive symptoms (Carvajal et al. 2002). The scale also showed high internal consistency for all respondents and across each ethnic group in our data (a0.81 overall; a0.78 within Latinos; a0.84 within European Americans). Results Socio-demographic Characteristics and Their Associations to Bicultural Stress Within the full sample (n519), 57% reported that a parent had a college degree, 79% lived in a house, and 33% lived in a home with greater or equal number of rooms as residents (see Table 1). Analysis revealed significant associations between ethnicity and our composite SES measure (p B0.01), though the magnitude of their relations also shows there is substantial non-overlap (h2 0.244). Overall, Latinos reported lower SES than European Americans. Latinos were closely evenly divided between those who did or did not have parents born outside the US (49% were third generation or higher, 15% were first generation), and 74% were bilingual-oriented or other language-oriented (Table 1). Latinos reported significantly (h0.18; p B0.05) higher bicultural stress (M 12.5) than non-Latino European Americans (M7.9). Also, Latinos of lower SES (r 0.27, p B0.01), and those Spanish-oriented (versus bilingual or English-oriented Latinos) (h0.24, p B0.05) reported more stress. There were no differences in stress due to gender, age or generational status within Latinos. Risk Behaviours and Depressive Symptoms in the Full Sample The first set of analyses employed univariate logistic regression to estimate associations between ethnicity, SES (standardized in all logistic models to facilitate the interpretation of odds ratios), and bicultural stress (standardized in all logistic models to facilitate the interpretation of odds ratios) in Latinos and non-Latino European Americans (n519). Table 2 presents the unadjusted and adjusted proportions of each risk behavior associated with the study variables. The univariate models (yield unadjusted odds ratios) show Latinos at increased risk for smoking,

452 A. J. Romero et al. Downloaded By: [University of Arizona] At: 17:06 6 December 2007

Table 1 Sample characteristics (n 519) Latino (% or mean (SD)) Age (range 1214) Female First generation Second generation Third generation Spanish-oriented (language acculturation) English-oriented (language acculturation) Bilingual (language acculturation) Parent graduate college Perceived SES score (range: 14) Live in a house Rooms/resident ratio SES composite score (raw; range: 13.7) Bicultural stress score (raw; range: 06.5) Base

European American (% or mean (SD))

Total (% or mean (SD))

14.1 (0.4) 57 15 36 49 31

14.2 (0.4) 56 n/a n/a n/a n/a

14.1 (0.4) 56 n/a n/a n/a n/a

26

n/a

n/a

43

n/a

n/a

39 2.61 (0.67) 69 19 2.28 (0.51)

81 2.86 (0.58) 90 49 2.84 (0.43)

57 2.71 (0.65) 79 33 2.51 (0.55)

12.4 (14.0)

7.9 (11.3)

10.4 (13.0)

304

215

519

Perceived SES was the average of three questions (e.g. ‘Compared to others in your city you think your family’s income is . . .). SES composite was derived from by equally weighting parental educational level, perceived SES and home structure.

illegal drugs use and fighting, and those of lower SES are at increased risk for smoking, illegal drugs use and depressive symptoms. Bicultural stress (standardized in logistic models) is significantly (p B0.05) associated with higher levels of all risk behaviors and depressive symptoms. In the multivariate models (yield adjusted odds ratios), stress accounted for nearly all ethnic and SES differences, except (1) SES and stress were both independently associated with ever smoking and self-identification as Latino, and (2) stress was related to fighting. Not presented in the tables, from the multivariate models older age was associated with the smoking variables (Ever: OR 1.8; 95% CI 1.03.0; last 30 days: OR2.4; 95% CI 1.15.1), boys reported higher levels of the violence indicators (Weapon: OR3.2; 95% CI 1.38.0; Fight: OR5.0; 95% CI 3.08.3), and females reported more depressive symptoms (Female as referent, OR0.6; 95% CI 0.40.9). Risk Behaviours and Depressive Symptoms in Latinos Univariate logistic regression was employed to estimate associations between acculturation variables, SES and bicultural stress in the sample of Latinos (n 304). Table 3 shows the unadjusted and adjusted proportions of each risk behavior associated with other study variables. Few acculturation or SES differences emerged: those Spanish-oriented reported more lifetime drinking than the English-oriented in


Table 2 Crude and adjusted odds of risk behaviors and depressive symptoms associated with ethnicity, SES and bicultural stress in Latino and non-Latino European American adolescents (n 519) Latino SES bicultural stress

%

COR (CI)

AOR (CI)

COR (CI)

AOR (CI)

COR (CI)

AOR (CI)

Ever drank Drank heavily in the last month Ever smoked cigarettes Smoked a cigarette in the last month Ever used illegal drugs Carried a weapon in the last month Been in a fight in the last year Depressive symptoms

58 4 27 9

1.3 1.1 2.0 1.4

(0.91.9) (0.43.0) (1.33.0) (0.82.7)

1.1 0.8 1.4 1.3

(0.71.8) (0.22.6) (0.82.3) (0.62.8)

0.8 0.8 0.6 0.7

(0.71.0) (0.51.2) (0.50.7) (0.61.0)

0.9 0.9 0.7 1.0

(0.81.2) (0.51.6) (0.60.9) (0.71.5)

1.7 1.8 1.6 1.8

(1.42.2) (1.32.6) (1.31.9) (1.42.3)

1.7 1.8 1.4 1.8

(1.32.1) (1.22.7) (1.11.8) (1.42.4)

13 6 23 21

1.7 1.6 2.0 1.2

(1.02.9) (0.73.5) (1.33.2) (0.81.8)

1.3 1.4 1.9 0.8

(0.72.6) (0.53.8) (1.13.3) (0.51.5)

0.6 0.8 0.8 0.7

(0.50.8) (0.61.2) (0.71.0) (0.60.9)

0.8 1.4 1.1 0.8

(0.61.1) (0.82.3) (0.91.5) (0.61.0)

1.6 2.3 1.7 1.5

(1.22.0) (1.63.2) (1.42.2) (1.21.9)

1.4 2.1 1.9 1.5

(1.11.9) (1.52.9) (1.52.3) (1.21.8)


Crude odds ratios (COR) reflect associations of ethnicity (referent is European American), SES (lower to higher, standardized prior to this analysis), and stress (lower to higher, standardized prior to this analysis) to risk behaviors and depressive symptoms. Adjusted odds ratios (AOR) were derived from models that including those variables as well as age and sex as covariates. Significant covariates not presented above were as follows: older age was associated with the smoking outcomes, boys reported higher levels of weapon carrying and fighting, and females reported more symptoms (estimates reported in the text).


Acculturation SES bicultural stress

%

Ever drank Drank heavily in the last month Ever smoked cigarettes Smoked a cigarette in the last month Ever used illegal drugs Carried a weapon in the last month Been in a fight in the last year Depressive symptoms

54 5 27 10 13 7 23 23

COR (CI) ns ns 2.1 (1.14.0)b ns ns ns ns ns

AOR (CI) 2.4 (1.15.0)a ns 2.9 (1.36.4)b ns ns ns ns ns

COR (CI) 0.8 0.6 0.6 0.8 0.7 0.9 1.0 0.8

(0.71.1) (0.31.2) (0.50.8) (0.51.2) (0.51.0) (0.51.4) (0.71.3) (0.61.1)

AOR (CI) 0.9 0.6 0.6 0.9 0.7 1.3 1.1 1.0

(0.71.2) (0.21.3) (0.50.9) (0.51.4) (0.51.1) (0.72.4) (0.81.5) (0.71.4)

COR (CI) 1.7 1.8 1.6 1.8 1.6 2.3 1.7 1.5

(1.42.2) (1.32.6) (1.31.9) (1.42.3) (1.22.0) (1.63.2) (1.42.2) (1.21.9)

AOR (CI) 1.7 1.8 1.4 1.8 1.4 2.1 1.9 1.5

(1.32.1) (1.22.7) (1.11.8) (1.42.4) (1.11.9) (1.52.9) (1.52.3) (1.21.8)

Crude odds ratios (COR) reflect associations of acculturation (those identified as Spanish-oriented, English-oriented or bilingual), SES (lower to higher, standardized prior to this analysis), and bicultural stress (lower to higher, standardized prior to this analysis) to risk behaviors and depressive symptoms. Adjusted odds ratios (AOR) were derived from models that included those variables as well as generational status, age and sex as covariates. For the acculturation variable, each of the three classifications were compared to one another, and significant (pB0.05) odds ratios are presented. Significant covariates not presented above were as follows: higher generation status was related to more symptoms and boys were more likely to have been in a fight (estimates reported in the text). a Spanish-oriented with English-oriented as referent. b English-oriented with bilingual as referent.


Table 3 Crude and adjusted odds of study outcomes associated with acculturation, SES and total bicultural stress in Latino adolescents (n 304)



the multivariable model, and more lifetime smoking was reported by those of lower SES and those English-oriented relative to bilingual-oriented in univariate and multivariate models. However, bicultural stress was again associated with all outcomes, more stress was significantly related to all risk behaviors and depressive symptoms. Also, in the multivariable models, higher generation status was related to more depressive symptoms (OR1.9; 95% CI1.23.2), and boys were more likely to have been in a fight (OR4.9; 95% CI 2.69.8). Exploratory Logistic Regression Models to Evaluate the Robustness of the Relations Between Bicultural Stress and Risk Behaviours We tested two series of logistic regression models based on the previous models to further verify the generality of the findings. The first set examined if stress was more relevant for specific sub-groups. To do this, we evaluated gender and ethnicity by bicultural stress interactions in the full sample. None of these 16 models approached significant improvement (p 0.15) as expressed in the change in the log likelihood of the multivariate models. Next, we added depressive symptoms as continuous covariate in the multivariable models predicting risk behaviors. Bicultural stress remained the single most predictive variable of the majority of risk behaviors, and was only non-significant for a single outcome in these models, predicting ever smoking (i.e. in this model OR1.3; 95% CI 1.01.5; p 0.058). Did Region Specific Demographics of Schools Influence Bicultural Stress? As some schools had majority European Americans and other schools majority Latinos, we were concerned that the variability in the proportion of each ethnic group between schools may have impacted the respondents’ personal experience of bicultural stress (Romero et al. in press). To address this concern, we used variance components analyses using maximum likelihood to evaluate the potential presence of school level variance in our bicultural stress measure, and if so, estimate the magnitude via the intra-class correlation (ICC). As covariates may frequently impact the magnitude of ICC estimates (Murray 1998), the sociodemographic variables employed in our multivariable logistic models were also employed in this analysis. The results suggest the intra-school correlation in bicultural stress was B0.001 in this model. (Previous research suggest minimal ICCs for risk behaviors within these schools; Carvajal & Granillo 2006.) These magnitudes of school-level variation would have a negligible influence (B0.01%) on all parameter estimates and test statistics for the individual variation-focused analyses conducted, and support the analytic models used to test study hypotheses (Murray 1998). These multivariable variance components analyses provide evidence that school-level characteristics were not substantively influential on the experience of bicultural stress, and suggest the findings generalizability to other non-majority and majority regions.



Discussion Researchers have increasingly emphasized the need to go beyond descriptive analyses of health disparities, and attempt to identify underlying psychosocial and policy relevant variables to guide public health promotion efforts (Walsh et al. 2000; Nazroo 2001; Myers & Rodriguez 2003; Unger et al. 2004a). The purpose of this study was to investigate the subjective perception of bicultural stress and its relation to risk behaviors among Latino and non-Latinos European American youth from a community sample. Bicultural stress was defined as the perception of stress due to everyday life stressors that result from navigating between more than one culture, and the assessments include items relating to discrimination, negative stereotypes, intergenerational acculturation gaps, and pressure to speak more than one language (Romero et al. 2003a; Romero et al. in press). As hypothesized, higher bicultural stress was consistently associated with more risk behavior engagement and higher depressive symptoms. Also, as hypothesized, higher bicultural stress accounted for the preponderance of the differences in risk behaviors due to ethnic self-identification, socio-economic status and acculturation factors within Latinos, all of which are factors more frequently presented in ethnic health disparities research. The findings also show bicultural stress was generally independent of depressive symptoms in relating to risk behaviors. Also, interactions between ethnicity and stress in relating to risk behaviors or depressive symptoms were not observed in the current study, nor was there evidence for school influences on the experience of bicultural stress despite substantial school variability in ethnic composition. These findings suggests that the experience of cultural and acculturative stressors by majority as well as minority groups may impact health, and suggest substantial generality to bicultural stress as a potential antecedent of multiple adolescent risk behaviors. These findings have several implications for health disparities research that may advance the work beyond descriptive analyses toward a better understanding of risk behavior etiology, and improved intervention potential. Moreover, certain culturally grounded adolescent prevention interventions are already beginning to address some related issues in order to decrease health disparities, such as adolescent substance use (Schinke et al. 1988; Unger et al. 2004b; Kulis et al. 2005; Hecht & Krieger 2006). Additionally, in terms of issues of discrimination in multicultural environments, recent modifications to the Contact Hypothesis, such that increased contact will decrease prejudice, demonstrated with empirical evidence that improved intergroup relations may be achieved through interventions (VanOudenhoven et al. 1996; Voci & Hewstone 2003; Paolini et al. 2006). Findings of the current work suggest that these interventions may be of relevance to both minority and majority adolescents. The study has a number of limitations; one being that it was region specific. However, due to demographic changes and increasing rates of immigration from Latin America, the current study’s sample characteristics may foreshadow the increasingly ethnic pluralistic nature of many communities within the US and other industrial nations (Nazroo 2001; Arnett 2002; Zagefka & Brown 2002). Also, though



ethnic variation varied substantially by school, our analyses revealed the experience of stress was primarily tied to individual or family factors, and suggests these daily youth stressors are likely to be experienced by adolescents across many regions and settings. Another limitation was that the modest community-sized sample precluded other ethnic comparisons (e.g. African Americans, Native Americans, Asian Americans, multi-ethnic youth) or more specific within Latino categories based on common nativity, history, immigration, acculturation, socioeconomic status, and language characteristics (Marıń & Marıń 1991; de la Torre & Estrada 2001). Additionally, in our data, we did not have visible markers of ethnicity (including skin color and phenotype) that could impact the experience and perception of bicultural stress (Montalvo 2004). Finally, like most surveys monitoring US levels of adolescent risk behavior (CDC 2004; Johnston et al. 2005), there were no biological markers of disease or behaviors, and the data were cross-sectional, limiting interpretations of causal influence. Despite these limitations, there are important strengths. The study employed many relevant socio-demographic variables and a current acculturation scale (Marıń & Gamba 1996). As our study was before high-school and comprehensive within a district, the sample was not susceptible to the exclusion of high-risk dropouts. This issue may be particularly problematic for studies conducted through US public high schools and confound health disparities research, since Latino males have among the highest drop out rates (CDC 2004). Also, our stress measure included everyday stressors related to discrimination, health relevant aspects to experiencing stress that are under-identified in the literature (Meyer 2004). Further, the measure of bicultural stress included subjective components of stress (Lazarus 2000), and was youth focused (Romero et al. 2003a); discrimination events in employment and financing typically assessed in adult populations (Meyer 2004) would probably not be relevant to middle school aged youth. In conclusion, to our knowledge, no national survey of adolescent health implements a current measure of subjective bicultural change nor a theory-based model of adolescent bicultural stressors (Romero et al. 2003a, in press). Given the consistent and robust findings, the current study highlights the importance of considering the measurement of bicultural stressors to address adolescent risk behavior disparities, and their antecedents in diverse populations. Also, a growing area of research has identified positive cultural buffers, such as a positive and bicultural ethnic identity, that may help individuals cope with some of those stressors (Benet-Martinez et al. 2002; Walters & Simoni 2002; Romero & Roberts 2003b). Given many common etiological process for health behaviors and inequities across nations (Nazroo 2001; Vega et al. 2002), it is likely that similar stressors are experienced by many other youth in increasingly multi-ethnic youth contexts. Interventions that address bicultural stress and potential coping processes may have wide applicability to youth health promotion efforts in diverse communities.



Acknowledgements This research was supported by grants from the National Institute on Drug Abuse (R03 DA14371), National Institute on Alcohol Abuse and Alcoholism (R21AA12803), National Center on Minorities and Health Disparities (P60MD000155), and the California Tobacco-Related Disease Research Program (7KT-0151) through ETR Associates of Scotts Valley, CA. We also thank the participating school administrators, staff and teachers, and acknowledge Karin Coyle, PhD, and Carrie Hanson, MA, at ETR Associates. K. Coyle and C. Hansen were instrumental to the data collection and project management of the initial survey funded by the California Tobacco-Related Disease Research Program, but were not involved in the current study’s conceptualization, analysis or writing. References Allison, K. W. (1998) ‘Stress and oppressed social category membership, in Prejudice: The Target’s Perspective, eds J. K. Swim & C. Stagnor, Academic Press, New York, pp. 149170. Arnett, J. J. (2002) ‘The psychology of globalization’, American Psychologist, vol. 57, pp. 774783. Benet-Martinez, V., Leu, J., Lee, F. & Morris, M. (2002) ‘Negotiating biculturalism: cultural frame switching in biculturals with ‘‘oppositional’’ vs. ‘‘compatible’’ cultural identities’, Journal of Cross-Cultural Psychology, vol. 33, pp. 729750. Bennett, G. G., Wolin, K. Y., Robinson, E. L., Fowler, S. & Edwards, C. L. (2005) ‘Perceived racial/ ethnic harassment and tobacco use among African American young adults’, American Journal of Public Health, vol. 95, pp. 238240. Berry, J. W. (2003) ‘Conceptual approaches to acculturation’, in Acculturation: Advances in Theory, Measurement, and Applied Research, eds K. M. Chun, P. O. Balls & G. Marıń, American Psychological Association, Washington, DC, pp. 1737. Booker, C. L., Gallaher, P., Unger, J. B., Ritt-Olson, A. & Johnson, C. A. (2004) ‘Stressful life events, smoking behavior, and intentions to smoke among a multiethnic sample of sixth graders’, Ethnicity & Health, vol. 9, pp. 369397. Borrell, L. N. (2005) ‘Racial identity among Hispanics: implications for health and well-being’, American Journal of Public Health, vol. 95, pp. 379381. Bourhis, R. Y., Moı¨se, L. C., Perreault, S. & Senećal, S. (1997) ‘Towards an interactive acculturation model: a social psychological approach’, International Journal of Psychology, vol. 32, pp. 369 386. Bradby, H. (2003) ‘Describing ethnicity in health research’, Ethnicity & Health, vol. 8, pp. 513. Carvajal, S. C. & Granillo, T. M. (2006) ‘A prospective test of distal and proximal determinants of smoking initiation in early adolescents’, Addictive Behavior, vol. 31, pp. 649660. Carvajal, S. C., Hanson, C. E., Romero, A. J. & Coyle, K. K. (2002) ‘Behavioural risk factors and protective factors in adolescents: a comparison of Latinos and non-Latino whites’, Ethnicity & Health, vol. 7, pp. 18193. Carvajal, S. C., Hanson, C. E., Downing, R. A., Coyle, K. C. & Pederson, L. L. (2004) ‘Theory-based determinants of youth smoking: a multiple influence approach’, Journal of Applied Social Psychology, vol. 34, pp. 5984. Centers for Disease Control and Prevention (2004) ‘Surveillance summaries. May 21, 2004’, Morbidity and Mortality Weekly Report, vol. 53 (No. SS-2). Cervantes, R. C., Padilla, A. M. & Salgado de Snyder, N. (1991) ‘The Hispanic Stress Inventory: a culturally relevant approach to psychosocial assessment’, Psychological Assessment, vol. 3, pp. 438447.


Ethnicity & Health 459 Clark, R., Anderson, N. B., Clark, V. R. & Williams, D. R. (1999) ‘Racism as a stressor for African Americans: a biopsychosocial model’, American Psychologist, vol. 54, pp. 805816. Compas, B. E., Davis, G. E., Forsythe, C. J. & Wagner, B. M. (1987) ‘Assessment of major and daily stressful events during adolescence: the Adolescent Perceived Events Scale’, Journal of Consulting & Clinical Psychology, vol. 55, pp. 534541. Crockett, L. J., Randall, B. A., Shen, Y.-L., Russell, S. & Driscoll, A. K. (2005) ‘Measurement equivalence of the Center for Epidemiological Studies Depression Scale for Latino and Anglo Adolescents: a national study’, Journal of Consulting and Clinical Psychology, vol. 73, pp. 4758. Cue´llar, I., Arnold, B. & Maldonado, R. (1995) ‘Acculturation Rating Scale for Mexican AmericansII: a revision of the original ARSMA Scale’, Hispanic Journal of Behavioral Sciences, vol. 17, pp. 275304. De La Rosa, M. R., Holleran, L. K., Rugh, D. & MacMaster, S. A. (2005) ‘Substance abuse among U.S. Latinos: a review of the literature’, Journal of Social Work Practice in the Addictions, vol. 5, pp. 120. de la Torre, A. & Estrada, A. (2001) Mexican Americans and Health: Sana! Sana!’, University of Arizona Press, Tucson, AZ. Ellickson, P. L., Orlando, M., Tucker, J. S. & Klein, D. J. (2004) ‘From adolescence to young adulthood: racial/ethnic disparities in smoking’, American Journal of Public Health, vol. 94, pp. 293299. Farley, T., Galves, A., Dickinson, L. M. & de Jesus Diaz Perez, M. (2005) ‘Stress, coping, and health: a comparison of Mexican immigrants, Mexican-Americans, and non-Hispanic Whites’, Journal of Immigrant Health, vol. 7, pp. 213220. Fisher, C. B., Wallace, S. A. & Fenton, R. E. (2000) ‘Discrimination distress during adolescence’, Journal of Youth & Adolescence, vol. 29, pp. 679695. Folkman, S. & Lazarus, R. S. (1980) ‘An analysis of coping in a middle aged community sample’, Journal of Health and Social Behavior, vol. 21, pp. 219239. Gil, A. G., Wagner, E. & Vega, W. A. (2000) ‘Acculturation, familism, and alcohol use among Latino adolescent males: longitudinal relations’, Journal of Community Psychology, vol. 28, pp. 443 458. Goodman, E., Slap, G. B. & Huang, B. (2003) ‘The public health impact of socioeconomic status on adolescent depression and obesity’, American Journal of Public Health, vol. 93, pp. 18441850. Gore, S., Aseltine, R. H. & Colton, M. E. (1992) ‘Social structure, life stress, and depressive symptoms on a high school aged population’, Journal of Health and Social Behavior, vol. 33, pp. 97113. Hecht, M. L. & Raup Krieger, J. L. (2006) ‘The principle of cultural grounding in school-based substance abuse prevention’, Journal of Language and Social Psychology, vol. 25, no. 3, pp. 301319. Hovey, J. D. (2000) ‘Acculturative stress, depression, and suicidal ideation in Mexican immigrants’, Cultural Diversity & Ethnic Minority Psychology, vol. 6, pp. 134151. Hovey, J. D. & King, C. A. (1996) ‘Acculturative stress, depression, and suicidal ideation among immigrant and second-generation Latino adolescents’, Journal of the American Academy of Child & Adolescent Psychiatry, vol. 35, pp. 11831192. Jessor, R. (1998) New Perspectives on Adolescent Risk Behavior, Cambridge University Press, New York. Johnston, L. D., O’Malley, P. M. & Bachman, J. G. (2005) Monitoring the Future National Results on Adolescent Drug Use: Overview of Key Findings, National Institute on Drug Abuse, Bethesda, MD (NIH Publication No. 05-5726). Kandel, D. B., Kiros, G. E., Schaffran, C. & Hu, M. C. (2004) ‘Racial/ethnic differences in cigarette smoking initiation and progression to daily smoking: a multivariate analysis’, American Journal of Public Health, vol. 94, pp. 128135.



Karlsen, S. & Nazroo, J. Y. (2002) ‘Relation between racial discrimination, social class, and health among ethnic minority groups’, American Journal of Public Health, vol. 92, pp. 624631. Krieger, N., Smith, K., Naishadham, D., Hartman, C. & Barbeau, E. M. (2005) ‘Experiences of discrimination: validity and reliability of a self-report measure for population health research on racism and health’, Social Science & Medicine, vol. 61, pp. 15761596. Kulis, S., Marsiglia, F. F., Elek, E., Dustman, P., Wagstaff, D. A. & Hecht, M. (2005) ‘Mexican/ Mexican American adolescents and ‘‘keepin’ it REAL’’: an evidence based substance use prevention program’, Children & Schools, vol. 27, no. 3, pp. 133145. LaFromboise, T., Coleman, H. & Gerton, J. (1993) ‘Psychological impact of biculturalism: evidence and theory’, Psychological Bulletin, vol. 114, pp. 395412. Lazarus, R. S. (1997) ‘Acculturation isn’t everything’, Applied Psychology: An International Review, vol. 46, pp. 3943. Lazarus, R. S. (2000) ‘Toward better research on stress and coping’, American Psychologist, vol. 55, pp. 665673. Marıń, G. & Marıń, B. V. (1991) Research with Hispanic Populations, Sage Publications Inc., Thousand Oaks, CA. Marıń, G. & Gamba, R. (1996) ‘A new measurement of acculturation for Hispanics: the bidimensional acculturation scale for Hispanics (BAS)’, Hispanic Journal of Behavioral Sciences, vol. 18, pp. 297316. Mays, V. M., Cochran, S. D. & Barnes, N. W. (2007) ‘Race, race-based discrimination and health outcomes among African Americans’, Annual Review of Psychology, vol. 58, pp. 201225. Meyer, I. H. (2004) ‘Prejudice as stress: conceptual and measurement problems’, American Journal of Public Health, vol. 93, pp. 262265. Montalvo, F. F. (2004) ‘Surviving race: skin color and the socialization and acculturation of Latinos’, Journal of Ethnic & Cultural Diversity in Social Work, vol. 13, pp. 2543. Myers, H. F. & Rodriguez, N. (2003) ‘Acculturation and physical health in racial and ethnic minorities, in Acculturation: Advances in Theory, Measurement, and Applied Research, eds K. M. Chun, P. B. Organista & G. Marıń, American Psychological Association, Washington, DC, pp. 163185. Murray, D. M. (1998) Design and Analysis of Group-randomized Trials, Oxford University Press, New York. National Longitudinal Study of Adolescent Health (2001) In School Questionnaire Code Book, Carolina Population Center, University of North Carolina, Chapel Hill. Nazroo, J. Y. (2001) Ethnicity Class and Health, PSI Press, London. Oetting, E. R. & Beauvais, F. (19901991) ‘Orthogonal Cultural Identification Theory: the cultural identification of minority adolescents’, The International Journal of the Addictions, vol. 25, pp. 655685. Padilla, A. M., Alvarez, M. & Lindholm, K. J. (1986) ‘Generational status and personality factors as predictors of stress in students’, Hispanic Journal of Behavioral Sciences, vol. 8, pp. 275288. Paolini, S., Hewstone, M., Voci, A., Harwood, J. & Cairns, E. (2006) ‘Intergroup contact and the promotion of intergroup harmony: the influence of intergroup emotions’, in Social Identities: Motivational, Emotional and Cultural Influences, eds R. Brown & D. Capozza, Psychology Press/Taylor & Francis, Hove. Phinney, J. S., Madden, T. & Santos, L. J. (1998) ‘Psychological variables as predictors of perceived ethnic discrimination among minority and immigrant adolescents’, Journal of Applied Social Psychology, vol. 28, pp. 937953. Radloff, L. S. (1977) ‘A CES-D scale: a self-report depression scale for research in the general population’, Applied Psychological Measurement, vol. 1, pp. 385401. Rodriguez, N., Myers, H. F., Mira, C. B., Flores, T. & Garcia-Hernandez, L. (2002) ‘Development of the multidimensional acculturative stress inventory for adults of Mexican origin’, Psychological Assessment, vol. 14, pp. 451461.


Ethnicity & Health 461 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, vol. 46, pp. 585597. Romero, A. J. & Roberts, R. E. (1998) ‘Perception of discrimination and ethnocultural variables in a diverse group of adolescents’, Journal of Adolescence, vol. 21, pp. 641656. Romero, A. J. & Roberts, R. E. (2003a) ‘Stress within a bicultural context for adolescents of Mexican descent’, Cultural Diversity & Ethnic Minority Psychology, vol. 9, pp. 17184. Romero, A. J. & Roberts, R. E. (2003b) ‘The impact of multiple dimensions of ethnic identity on discrimination and adolescent’s self-esteem’, Journal of Applied Social Psychology, vol. 33, pp. 22882305. Romero, A. J., Robinson, T. N., Haydel, K. F., Mendoza, F. & Killen, J. D. (2004) ‘Associations among familism, language preference, and education in Mexican-American mothers and their children’, Journal of Developmental & Behavioral Pediatrics, vol. 25, pp. 3440. Romero, A. J., Carvajal, S. C., Valle, F. & Orduna, M. (2007) Adolescent bicultural stress and its impact on mental well-being among Latinos, Asian Americans and European Americans, Journal of Community Psychology, vol. 35, no. 4, pp. 519534. Rushton, J. L., Forcier, M. & Schectman, R. M. (2002) ‘Epidemiology of depressive symptoms in the National Longitudinal Study of Adolescent Health’, Journal of the American Academy of Child and Adolescent Psychiatry, vol. 41, pp. 199205. Samaniego, R. & Gonzales, N. A. (1999) ‘Multiple mediators of the effects of acculturation status on delinquency for Mexican American adolescents’, American Journal of Community Psychology, vol. 27, pp. 189210. Sampson, R. J., Morenoff, J. D. & Raudenbush, S. (2005) ‘Social anatomy of racial and ethnic disparities in violence’, American Journal of Public Health, vol. 95, pp. 224232. Schinke, S. P., Botvin, G. J., Trimble, J. E., Orlandi, M. A., Gilchrist, L. E. & Locklear, V. S. (1988) ‘Preventing substance abuse among American Indian adolescents: a bicultural competence skills approach’, Journal of Counseling Psychology, vol. 35, no. 1, pp. 8790. Sellers, R. M. & Shelton, J. N. (2003) ‘The role of racial identity in perceived racial discrimination’, Journal of Personality and Social Psychology, vol. 84, no. 5, pp. 10791092. Spears-Brown, C. & Bigler, R. S. (2005) ‘Children’s perceptions of discrimination: a developmental model’, Child Development, vol. 76, pp. 533553. Stonequist, E. V. (1935) ‘The problem of marginal man’, American Journal of Sociology, vol. 7, pp. 1 12. Stueve, A. & O’Donnell, L. N. (2005) ‘Early alcohol initiation and subsequent sexual and alcohol risk behaviors among urban youths’, American Journal of Public Health, vol. 95, pp. 887893. Szalacha, L. A., Erkut, S., Coll, C., Alarcon, O., Fields, J. P. & Ceder, I. (2003) ‘Discrimination and Puerto Rican children’s and adolescents’ mental health’, Cultural Diversity and Ethnic Minority Psychology, vol. 9, pp. 141155. Szapocznick, J., Scopetta, M. A., Kurtines, W. & Aranaldi, D. (1978) ‘Theory and measurement of acculturation’, Revista Interamericana de Psicologia, vol. 12, pp. 113130. Unger, J. B., Baezconde-Garbanati, L., Shakib, S., Palmer, P. H., Nezami, E. & Mora, J. (2004a) ‘A cultural psychology approach to ‘‘drug abuse’’ prevention’, Substance Use & Misuse, vol. 39, pp. 17791820. Unger, J. B., Chou, C.-P., Palmer, P., Ritt-Olson, A., Gallaher, P., Cen, S., Lichtman, K., Azen, S. & Anderson Johnson, C. (2004b) ‘Project FLAVOR: 1-year outcomes of a multicultural, school-based smoking prevention curriculum for adolescents’, American Journal of Public Health, vol. 94, no. 2, pp. 263265. United States Census Bureau (2000) ‘Census 2000’. US Department of Commerce Economics and Statistics Administration. Available at: http://www.bayareacensus.ca.gov/counties/SantaClara County.pdf.



United States Census Bureau (2001) ‘The Hispanic population in the United States’. US Department of Commerce Economics and Statistics Administration. Available at: http://www.census.gov/ population/socdemo/hispanic/p20-535/p20-535.pdf. van de Vijver, F. J. R. & Phalet, K. (2004) ‘Assessment in multicultural groups: the role of acculturation’, Applied Psychology: An International Review, vol. 53, pp. 215236. VanOudenhoven, J. P., Groenewoud, J. T. & Hewstone, M. (1996) ‘Cooperation, ethnic salience and generalization of interethnic attitudes’, European Journal of Social Psychology, vol. 26, pp. 649661. Vega, W. A., Alderete, E., Kolody, B. & Aguilar-Gaxiola, S. (1998) ‘Illicit drug use among Mexicans and Mexican Americans in California: the effects of gender and acculturation’, Addiction, vol. 93, pp. 18391850. Vega, W. A., Aguilar-Gaxiola, S., Andrade, L., Bijl, R., Borges, G., Caraveo-Anduaga, J. J., DeWit, D. J., Heeringa, S. G., Kessler, R. C., Kolody, B., Merikangas, K. R., Molnar, B. E., Walters, E. E., Warner, L. A. & Wittchen, H. (2002) ‘Prevalence and age of onset for drug use in seven international sites: results from the International Consortium of Psychiatric Epidemiology’, Drug and Alcohol Dependence, vol. 68, pp. 285297. Voci, A. & Hewstone, M. (2003) ‘Intergroup contact and prejudice toward immigrants in Italy: the mediational role of anxiety and the moderational role of group salience’, Group Processes & Intergroup Relations, vol. 6, no. 1, pp. 3752. Walsh, M., Smith, R. A., Morales, A. & Sechrest, L. (2000) Ecocultural Research: A Mental Health Researcher’s Guide to Studying Race, Ethnicity, and Culture, Human Research Services Center, Cambridge, MA. Walters, K. L. (2002) ‘Reconceptualizing Native women’s health: an ‘‘indigenist’’ stress-coping model’, American Journal of Public Health, vol. 92, pp. 520524. Weden, M. M. & Zabin, L. S. (2005) ‘Gender and ethnic differences in the co-occurrence of adolescent risk behaviors’, Ethnicity & Health, vol. 10, pp. 213234. Williams, D. R., Neighbors, H. W. & Jackson, J. S. (2003) ‘Racial/ethnic discrimination and health: findings from community studies’, American Journal of Public Health, vol. 93, pp. 200208. Williams-Morris, R. S. (1996) ‘Racism and children’s health: issues in development’, Ethnicity & Disease, vol. 6, pp. 6982. Wills, T. A., McNamara, G. & Vaccaro, D. (1995) ‘Parental education related to adolescent stresscoping and substance use: development of a mediational model’, Health Psychology, vol. 14, pp. 464478. Zagefka, H. & Brown, R. (2002) ‘The relationship between acculturation strategies, relative fit and intergroup relations: immigrant-majority relations in Germany’, European Journal of Social Psychology, vol. 32, pp. 171188. Zambrana, R. E. & Silva-Palacios, V. (1989) ‘Gender differences in stress among Mexican immigrant adolescents in Los Angeles, California’, Journal of Adolescence Research, vol. 4, pp. 426442.



Appendix A English Version of the Bicultural Stress Scale 20. Please indicate how stressful the following experiences have been for you. If you have never had the experience please circle ‘1’: Never happened to me. Please fill in only one answer for each item. Never Not at happened all to me stressful a. I have been treated badly because of my accent b. I have worried about family members or friends having problems with immigration c. I do not feel comfortable with people whose culture is different than my own d. I feel uncomfortable when others make jokes about people of my ethnic background e. I have had problems at school because of my poor English f. I do not like it when others put down people of my ethnic background g. I have felt that others do not accept me because of my ethnic group h. I feel that I can’t do what most American kids do because of my parent’s culture i. I feel that belonging to a gang is part of representing my ethnic group j. I do not understand why people from a different ethnic background act a certain way k. I feel that it will be harder to succeed because of my ethnic background l. Because of family obligations, I can’t always do what I want m. I have felt pressure to learn Spanish n. I have felt that I need to speak Spanish better o. My friends think I’m acting ‘White’ p. My parents feel I do not respect older people the way I should q. I have had to translate/interpret for my parents r. I have felt lonely and isolated because my family does not stick together s. I have had to help my parents by explaining how to do things in the US t. I have argued with my boyfriend/girlfriend over being too traditional

A little bit stressful

Quite a Very bit stressful stressful

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1 1

2 2

3 3

4 4

5 5

1 1

2 2

3 3

4 4

5 5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5