Introduction

Acculturation related conflicts arising from incongruence between immigrant parents’ adherence to their heritage culture and children’s accelerated adoption of the mainstream dominant culture (Smokowski et al. 2008), is a topic that has received limited empirical attention but has been the focus of some debate over the last 20 years (Smokowski and Bacallao 2006). Prior research has yielded conflicting evidence regarding these associations with some findings suggesting that family conflicts emanating from parent–child acculturation disparities are significantly related to youth maladjustment, while others have found no such association. To illustrate, in several studies parent–child ethnic identity discrepancies were directly related to increased family conflict (Martinez 2006; Smokowski et al. 2008), and indirectly associated with youth internalizing distress and externalizing behavior problems (Schofield et al. 2008). In contrast, other studies have failed to find an association between parent and child ethnic identity discrepancies, family conflict, and increased risk for youth maladjustment (Lau et al. 2005; Pasch et al. 2006).

In addition to these inconsistent findings, there is a dearth of research examining the role that gender may play in influencing the relationship between parent and child ethnic identity discrepancies and adolescent psychosocial outcomes. Deviations from culturally prescribed gender roles are likely to be sources of tension––particularly between first- and second-generation parents and daughters––since many non-Western cultures afford males greater autonomy compared to females (Ahmed 2012; Sirin and Fine 2008). In all likelihood, first- and second-generation adolescent females experience distress as they negotiate the delicate and dynamic balance between behaving in ways consistent with parental norms and those of the heritage culture on the one hand, and peer group and host country norms on the other. Consistent with the definition used by Raffaelli et al. (2012), we refer to first generation immigrants as those who are foreign born and came to the US and second generation as those who are US born with at least one foreign born parent.

These two deficits in the literature: (1) inconsistencies in the field pertaining to these relationships, and (2) the dearth of research exploring gender differences vis-à-vis these associations––present a serious dilemma for mental health practitioners working with immigrant families. Thus, in the current investigation, we set out to explore gender differences in the association between parent and child ethnic identity discrepancies and adolescent psychosocial outcomes with the ultimate goal of informing practice applications. Our sample was drawn from an ethnically diverse population of youth, who also varied from each other in terms length of time in the US (with some US born included for comparison, while foreign born youth reported the number of years lived in the US). Such an ethnically heterogeneous sample was chosen to reflect the rapidly changing demographic landscape of the United States. Ultimately, our goal is to examine whether or not gender differences exist within these associations so as to identify a set of universal as opposed to culture-specific recommendations that might be relevant to practitioners working with first- and second-generation youth.

Parent–Child Ethnic Identity Discrepancies and Youth Psychosocial Adjustment

We have extended the definition used by Smokowski et al. (2008) from one strictly about ethnic identity to one that concerns parent–child ethnic identity discrepancies. For the current study, we have defined ethnic identity discrepancies as the incongruence between immigrant parents’ heritage culture and beliefs and those of the child’s who is likely to adopt the norms of the host country. In an effort to identify universal processes that affect acculturating individuals regardless of country of origin, we have cited together those references for which there is consensus in findings across different ethnic groups.

Several studies suggest that parent–child ethnic identity discrepancies are directly related to increased family conflict (Martinez 2006; Smokowski et al. 2008), and indirectly associated with child internalizing problems (Schofield et al. 2008), elevated levels of adolescent aggression (Smokowski and Bacallao 2006), and externalizing behavior problems (Gonzalez et al. 2006; Schofield et al. 2008; Smokowski and Bacallao 2006). Contrary to these findings, others report no such relationship. For example, Moon (2008) found no increased familial conflict as a result of disparities in parent–child patterns of adoption of the dominant culture. Furthermore, several studies failed to find an increased risk for youth maladjustment as a function of acculturation conflicts (Lau et al. 2005; Pasch et al. 2006). These inconsistencies in the field are directly related to disparate operational definitions of ethnic identity discrepancies across studies (Birman 2006; Telzer 2010), in addition to the complicated array of individual, familial (e.g. warmth, cohesion, socioeconomic status, length of stay in society of settlement, etc.), and host country (e.g. the level of cultural diversity and discrimination towards ethnic minorities) factors (Telzer 2010).

Gender Differences

Evidence suggests that adaptation to the host culture is associated with distinct psychosocial outcomes for young females and males. According to the findings of Berry et al. (2006), gender differences were apparent in their large scale study of youth struggling to acculturate: Female adolescents reported slightly worse psychological outcomes (e.g. life satisfaction, self-esteem, and internalizing distress) compared to males, while males exhibited poorer sociocultural outcomes (e.g. school adjustment and behavior problems) than females (Berry et al. 2006). This greater risk for females to demonstrate emotional distress in response to acculturation has been documented (Liebkind 1993; Berry et al. 2006), however this phenomenon is still relatively unexplored with regards to implications for mental health providers working with youth and families struggling with acculturation related conflicts.

The greater vulnerability to depression for acculturating females is likely a function of the clash of cultural expectations regarding gender roles since gender inequality in many non-Western cultures affords males more latitude in terms of autonomy as compared to females (Ahmed 2012; Sirin and Fine 2008). For instance, in a study of Vietnamese refugees living in Finland (Liebkind 1993), although there were disparities between parent and child acculturation levels reported for both sexes, females exhibited higher levels of depression and anxiety than males. The author attributed this greater distress to difficulties negotiating the traditional Vietnamese gender roles expected by their parents and those espoused by their more liberal Finnish peers. Likewise, parental views in accordance with traditional Islamic expectations regarding gender roles have been identified as a source of conflict for parents and their daughters (Abu-Ali and Reisen 1999). Tension and distance in the parent-daughter relationship are likely outcomes of disputes about culturally prescribed behaviors and are likely to play out as a theme when working with immigrant families (Ahmed 2012).

In addition to these acculturation stressors that selectively affect females, it is well documented that females tend to internalize distress more than males (Essau et al. 2000; de Matos et al. 2008; Hankin et al. 1998) and are twice as likely to be diagnosed with depression than are males (McGrath et al. 1990). The diathesis-stress model may partially explain these complex associations. This model purports that the greater vulnerability of females to develop depression interacts with social stressors to produce a greater incidence of depression (Garber and Hilsman 1992; Spence et al. 2002). For these reasons, we expected females with high parent–child ethnic identity disparities to be at greater risk for internalizing distress.

Study Overview: Theoretical Background and Operationalization of Constructs

Bronfenbrenner’s bioecological model (Bronfenbrenner and Morris 2006) provided the theoretical backdrop of the current investigation since it emphasizes the transactional and contextually sensitive nature of human development. Importantly, associations between parent and child ethnic identity discrepancies are not merely limited to the individual and family at the microsystemic level, but effects are likely to be occurring at the mesosystemic (e.g. interactions between parents and schools), exosystemic (e.g. neighborhood), as well as macrosystemic levels (such as cultural influences and laws) (Bronfenbrenner and Morris 2006). Since the intention here is to identify universal processes that are likely to be occurring across groups from different countries of origin, Bronfenbrenner’s model provides a robust foundation by recognizing the dynamic interplay of individual, context, and the role that culture may play in moderating those interactions without unduly emphasizing specific factors (e.g. countries of origin, path of immigration, etc.). This integrative model has been used in prior work examining the influence of acculturation on psychosocial adjustment (Torres et al. 2012) and on risky adolescent sexual behaviors (Raffaelli et al. 2012).

As has been done in other studies (Lee et al. 2000; Rick and Forward 1992), the current investigation operationalized parent–child ethnic identity discrepancies from the adolescent’s perspective. Difference scores between adolescents’ self-reports of ethnic identity and their perception of their parents’ ethnic identity were assessed on two dimensions––belongingness to one’s heritage culture as well as affiliation to the dominant group. Although some have argued against the use of this method of measurement to assess the acculturation disparities between parents and children (Birman 2006; Telzer 2010), we felt it was important to examine perceived parent–child ethnic identity discrepancies from the perspective of the child, since in clinical practice with families, this perception may play a more salient role in links to psychosocial distress among first- and second-generation youth.

Other predictors included age of child since this may affect the level of conflict between parent and child, as older adolescents may be less accepting of parents’ expectations and rules. Similarly ethnic background was included in analyses since different countries of origin may be more closely aligned with Western cultural norms and practices. Additionally, length of stay in the United States was included as a predictor given some evidence suggesting greater psychological adjustment with increased years lived in the host country (Berry et al. 2006) while other findings posit greater psychosocial maladjustment with increased time in the host country (García Coll and Marks 2012; Phinney et al. 2000).

Two sets of outcomes were of interest in this investigation: internalizing distress as indicated by depression and anxiety, and social adjustment measured in terms of social stress and adaptive interpersonal relationships. Hierarchical multiple regression analyses were conducted separately for males and females, as has been done in prior work where processes were expected to be different between groups (Ansary et al. 2011; Steinberg and Fletcher 1998). In this investigation we sought to explore two research hypotheses. First, we expected that a large disparity between parent and child ethnic identities would be associated with greater depression and anxiety, and that this effect would be stronger for females compared to males. Second, we expected that higher levels of child affiliation with the dominant group as compared to their parents’ ethnic group of origin would be associated with lower levels of social stress and more positive interactions with peers. Implications for clinical work with first- and second-generation families will be examined.

Method

Participants

Data collected were part of a larger cross-sectional study assessing adolescent psychological and social adjustment, ethnic identity, religiosity, and moral reasoning. High school students attending an ethnically and economically diverse regional high school in the Northeastern US were recruited for participation. The final sample consisted of 174 high school students (55.17 % female) and was comprised predominantly of minority youth (45 % Latino, 17.8 % Caucasian, 11.2 % African–American, 10.7 % of mixed background, 8.3 % Asian, and 7 % Other). It is likely that this sample generally mirrors the population of immigrants: According to Larsen (2004), Latinos and Asians comprise the largest portion (75 % in that study) of foreign-born individuals in the US. Ages of the participants ranged from 14 to 19 (M = 16.05, SD = 1.30). Although the majority of the sample was US born, approximately one-fifth of participants were born abroad (n = 38). The socioeconomic status of the sample was diverse with median incomes for the three towns served by the school as follows: $49,014, $56,815, and $86,246.

Measures

Predictors

Demographics

Students were asked to indicate their age, gender, ethnic background, parental education levels, number and relationship of others who lived in the participants’ homes, whether the participants had lived in the United States all their lives, and if not, how many years they had lived in the US. Participants were asked to respond to this question by indicating the appropriate category from among the following: 0–2, 3–5, 6–8, 9–11, 12–14 years and 15 years or more.

Ethnic Identity

The Multigroup Ethnic Identity Measure (MEIM; Phinney 1992) is a valid and reliable 20- item measure (see Ponterotto et al. 2003) that was utilized to assess both subjects’ and their parents’ ethnic identity. Respondents are asked to rate their level of agreement (1 = strongly disagree through 5 = strongly agree) on items related to one’s adherence to cultural practices, belief systems, and affiliation with one’s ethnic group, as well as with the dominant group. Of the four subscales that comprise the measure, two were used in the current study: the five item Affirmation and Belonging subscale and the six item Other Group Orientation subscale. Sample items from these subscales include, “I am happy that I am a member of the group I belong to,” and “I like meeting and getting to know people from ethnic groups other than my own” respectively. Students were asked to complete the MEIM once for themselves and immediately following, they were asked to complete an adapted version of the survey asking them to report on their parents’ ethnic identity related behaviors. Coefficients alpha for the Affirmation and Belonging subscale with students reporting on their own ethnic identity were .82 and .85 respectively for females and males. For student reports of parents’ ethnic identity related behaviors, on this subscale internal consistency coefficients were also .82 for females and .85 for males. In terms of student self-report measures of internal consistency for the Other Group Orientation subscale, alphas were .67 for females and .72 for males. Regarding students’ perceptions of their parents’ affiliation with the dominant group, measures of internal consistency for this subscale were .87 for females and .77 for males.

Two ethnic identity conflict variables were created to assess conflict between adolescent and parent. For the two subscales used (Affirmation and Belonging, and Other Group Orientation), parent values were subtracted from student values. These two difference scores were used in regression analyses as indicators of parent–adolescent ethnic identity discrepancies.

Outcomes

The Behavioral Assessment System for Children—Second Edition Self-Report-Adolescent (BASC-2 SRP-A; Reynolds and Kamphaus 2004) was used to assess the four outcomes of interest in this study: depression, anxiety, social stress, and interpersonal relations. For these four scales, respondents are asked to rate how often they experience thoughts and behaviors on a True/False scale as well as a four-point scale anchored by “never” and “always”. Psychometric studies of the BASC-2 SRP-A show high internal consistency reliabilities on each scale (between .7 and .8) as well as high test–retest reliabilities (Reynolds and Kamphaus 2004).

Depression

The Depression scale is comprised of 12 items of the BASC-2 assessing symptoms of depression such as isolation, sadness, and an inability to enjoy life. Sample items include: “Nothing feels good to me,” “I feel sad,” and “Nothing is fun anymore”. In this study, coefficients alpha were .88 for females and .78 for males.

Anxiety

Thirteen items of the BASC-2 adolescent version tap feelings of fear, worry, and nervousness. Items include: “I worry about little things,” “I worry but I don’t know why,” and “I often worry about something bad happening to me”. For this scale, the alpha coefficients were .84 for females and .87 for males.

Social Stress

Level of stress experienced when interacting with peers was assessed using this 10-item scale. Sample items include: “I am bothered by teasing from others,” “I feel out of place around people,” and “Other people are against me”. The internal consistency coefficients in the current study were .85 for females and .78 for males.

Interpersonal Relations

This 7-item scale assesses the adolescent’s ability to relate to others (peers and to a lesser extent adults) as well as the degree to which pleasure is garnered from these positive relationships. Sample items include: “My classmates don’t like me,” “I feel that nobody likes me,” and “I am slow to make new friends”. The internal consistency coefficients were .70 for females and .77 for males.

Procedure

This study was approved by the sponsoring institution’s IRB (Institutional Review Board) and also underwent an extensive approval process with the high school. Once active parental and student consent were obtained, the battery of measures was group administered to participants during a 60-min. block of time. All instructions and questions were read aloud in order to minimize differences in participant reading levels. A $10 gift card to a national chain bookstore was given to each participant for completing the survey.

Results

Means and standard deviations for all predictor and outcome variables were computed separately for females and males and are presented in Table 1. Gender differences were evident on the two indicators of internalizing distress. As can be seen in Table 1, in comparison to males, females scored significantly higher on the Depression, t (160) = 3.62, p < .001, d = .55, 95 % CI [1.29, 4.78], and Anxiety scales, t (169) = 3.56, p < .001, d = .78, 95 % CI [1.66, 5.78]. Cohen’s d for each test is greater than .5 indicating a moderate effect size (Cohen 1992).

Table 1 Means and standard deviations for males and females on predictor and outcome variables
Full size table

Hierarchical multiple regression analyses were conducted in order to ascertain the relative unique contribution of demographic variables (i.e. age and ethnicity), time lived in the US, and parent–child ethnic identity conflicts (i.e. Affirmation and Belonging, and Other Orientation), on the four indicators of adolescent psychosocial maladjustment. Each of these distinct dimensions was entered into steps and regressed on the outcomes. As noted earlier, analyses were conducted separately for females and males since the associations between acculturation processes and psychosocial adjustment were expected to be different for each sex. Results of regression analyses for females are presented in Table 2 and those for males are presented in Table 3.

Table 2 Hierarchical multiple regression analyses predicting depression, anxiety, social stress and interpersonal relations from age, ethnicity, time lived in the US and parent–child ethnic identity conflict for females
Full size table
Table 3 Hierarchical multiple regression analyses predicting depression, anxiety, social stress and interpersonal relations from age, ethnicity, time lived in the US and parent–child ethnic identity conflict for males
Full size table

Internalizing Distress

As expected, gender differences were detected in associations between the predictors and internalizing distress. Regarding females and links to depression, the parent–child ethnic identity conflict step was significant ∆R 2 = .22, p < .05 and this was primarily driven by the Other Group Orientation conflict β = .51, p < .05. Such associations between parent and child ethnic identity conflicts were not evident for males, although time lived in the US explained a significant portion of the variance in depression ∆R 2 = .18, p < .05. Additionally, for males, the step containing demographic variables was marginally significant ∆R 2 = .30, p < .10 and this revealed a positive link between age and depression.

When regressed on anxiety, the parent–child conflict predictors were not significantly associated with this outcome for either females or males, although a trend did appear in the data for girls. For females, the parent–child ethnic identity conflict step explained 18 % of the variance in anxiety but was slightly above our cutoff for marginal significance F∆ (2, 21) = 2.55, p = .102. This effect was primarily a function of the conflict produced by an Other Group Orientation, β = .45, p < .05. Regarding males, no significant associations were found between the predictors included in the steps and anxiety.

Social Outcomes

As was found for depression and anxiety, gender differences were also apparent on indicators of social adjustment. With respect to social stress, whereas a marginally significant parent–child ethnic identity conflict ∆R 2 = .20, p < .10––motivated by an other group orientation, β = .49, p < .05––predicted social stress for females, this finding did not emerge for males. Additionally among males, the significant demographics step ∆R 2 = .41, p < .05, indicated a positive association between age and social stress β = .63, p < .01.

With respect to quality of interpersonal relations, there were no significant associations between the predictor variables and this outcome for females. Among males, demographics ∆R 2 = .30, p < .10––with age (β = −.47, p < .05) driving the marginally significant effect––predicted quality of interpersonal relations. Older males in the sample reported poorer quality interpersonal relationships.

Discussion and Practice Implications

In the current investigation, gender differences in links between parent and child ethnic identity discrepancies and adolescent psychosocial maladjustment were apparent. Surprisingly, while males and females did not differ on mean levels of the two parent–child ethnic identity discrepancy domains, only females reporting higher discrepancies with their parents on affiliating with the dominant group, also reported elevated levels of depression, significantly higher levels of social stress, as well as a non-significant trend towards greater anxiety. In contrast, parent–child ethnic identity discrepancy did not significantly explain any of the four psychosocial outcomes for males. Importantly, this finding was obtained in spite of the fact that females in our sample reported living in the US about twice as long as their male counterparts. It is unclear whether this finding merely reflects gender differences in acculturation patterns or whether it is more broadly indicative of the “immigrant paradox”––a trend of increased maladjustment as acculturation into American culture progresses over time (García Coll and Marks 2012).

Our first hypothesis––which posited that conflicts would be predictive of internalizing distress for both females and males, with effects being stronger for females––was partially supported. Discrepancies regarding affiliation with the dominant group was the sole parent–child ethnic identity discrepancy variable among females associated with several outcomes including: depression, social stress, as well as a non-significant but notable association with anxiety. Alternatively, our second hypothesis, which stated an expectation of a positive association between parent and child conflict concerning affiliation with the dominant group and improved social relations with peers, was not supported at all. For females, higher levels of discrepancy with parents over a dominant group orientation, was actually linked with higher levels of social stress and not associated at all with the quality of interpersonal relationships.

For males, as noted earlier, neither parent–child ethnic identity discrepancy variable was associated with either psychological or social outcomes; however, our results suggest alternate correlates of psychosocial maladjustment for males. For males in this sample, the demographics block was marginally significantly associated with depression and social stress: This finding was primarily driven by age indicating greater depression and lowered quality of interpersonal relationships with maturation. Consistent with this trend, age was also significantly associated with social stress for males. Furthermore, the length of time adolescent males had lived in the United States was also a significant positive predictor of depression. Each of these findings and implications for mental health professionals working with acculturating youth and families will be discussed in turn.

Parent–Child Ethnic Identity Discrepancy and Adolescent Distress: Potentially Different Profiles for Females and Males

On a broad basis, the emergent gender differences in these associations partially support the findings of Berry et al. (2006). In that study, immigrant females appeared to be less psychologically adapted than their male counterparts while males demonstrated poorer sociocultural adaptation than females. However, in that study, sociocultural adaptation was operationalized as a composite of school adjustment and behavior problems—both domains in which females typically show greater adjustment than males (Broidy et al. 2003; Luthar and Ansary 2005; Newcomb et al. 2002). Thus, the findings from their study may also be interpreted not necessarily as gender specific psychosocial patterns of acculturation (although gender differences were found in the types of acculturation profiles adopted) but much more simply as normative gender specific ways of manifesting distress and coping. Stated otherwise, family conflicts emanating from disparate levels of parent–child acculturation patterns are likely to be associated with distress manifested by females via internalizing distress and among males by engaging in problem behaviors; much the way males and females typically cope with distress (Book et al. 2001; Papadakis et al. 2006). Although in the current investigation externalizing behavior problems (e.g. substance use, delinquency, etc.) were not examined, the gender differences obtained suggest that further investigation examining other problem behaviors, more typically exhibited by males, is warranted.

In addition to suggesting potential gender specific ways of manifesting acculturation related distress, the findings of the current study also imply the existence of possible gender differences in the nature of these conflicts as only parent–child discord associated with the dominant group orientation was associated with poorer psychosocial adjustment among females. However, parent–child ethnic identity discrepancies concerning affirmation and belonging to their heritage culture were not associated with female maladjustment. Greater gender equality witnessed in many Western countries may produce fundamentally different conflicts between girls and their parents as compared to boys and their parents (Ahmed 2012; Liebkind 1993). Thus, in all likelihood the nature of the parent–child ethnic identity discrepancies experienced by males and females organically differs in content, quality, and intensity. Future work would be well served to explore the nuances of the parent–child ethnic identity discrepancies as well as how gender may moderate their effects on myriad psychosocial outcomes.

In addition to the absence of significant findings regarding acculturation related conflicts and psychosocial adjustment among adolescent males, we did obtain interesting findings related to demographic variables predicting psychosocial adjustment of males. In these analyses, age was a predictor (although the steps were marginally significant for depression and interpersonal relationships) of three out of four of our maladjustment outcomes. This finding could be attributed simply to timing: the period of adolescence and the challenges associated with it is reaching its height in our sample of males, as the average age of our participants was slightly over 16.

Similarly, while not an a priori hypothesis, but contrary to the research evidence, for males in the current study, duration of time lived in the US was positively associated with depression. There are two potential explanations for this finding. First, and the most likely reason, is that age and time lived in the US are strongly related. Thus, just as prevalence rates of depression increase as all youth (regardless of acculturation) move into later adolescence, the association between time lived in the US and depression is likely to be a function of the adolescent males’ age. In terms of the second potential explanation–that there is, in fact, a more direct association between time lived in the US and depression––the literature is inconclusive. Although Berry et al. (2006) found that longer duration in the country of resettlement was associated with more positive adolescent outcomes, others have found the converse (García Coll and Marks 2012; Phinney et al. 2000). Given the cross-sectional design of the current investigation, we were not able to investigate if this was in fact the case among males in our sample.

Implications for Clinical Practice

The findings from the current investigation yield fruitful insights for social workers and mental health professionals––especially those working in the school setting––who provide services to first- and second-generation adolescents and their families. First, such providers must approach work with youth and families struggling to acculturate with cultural humility and a willingness to learn about the cultural beliefs and traditions of their clients (Moon 2008), especially as they pertain to gender role expectations. Following from this, and as the findings suggest, affiliation with the dominant group’s beliefs can be a source of tension for parents and daughters especially (Ahmed 2012) and should be explored as correlates of depression, social stress, and anxiety. Also implicit in these findings, is the salience of the peer context: parent–child ethnic identity discrepancies are likely to reflect the constant balancing between family and peer expectations that acculturating females must negotiate. Thus, working on adolescent female identity formation (i.e. movement toward biculturalism), in addition to working with parents to re-negotiate gender role expectations, are critical elements in working with adolescent females and their families struggling to acculturate.

With regard to males, the current findings suggest that although parent–child discrepancies do exist, these were not associated with psychosocial maladjustment. Though associations between parent–child ethnic identity discrepancies and externalizing behavior problems were not directly tested, these should be explored in clinical practice as possible correlates of conflicts between acculturating males and their families.

Limitations

Several limitations in the current investigation should be considered when evaluating the findings. The sample size is relatively modest, especially, the number of foreign-born adolescents (n = 38). Despite numerous initiatives including several mailings home to secure active parental consent, as well as hosting a number of discussion sessions for parents in order to garner their cooperation, of a potential 800 9th–12th graders, only 200 parents returned active parental consent for participation. Each mailing sent home to the parents of potential participants, described the study in the three languages most commonly spoken in the towns serviced by the school: English, Spanish, and Arabic.

Other researchers have discussed the difficulties involved when studying immigrant and ethnically diverse populations. For example Spring et al. (2003) note the problems associated with obtaining samples of Somali and Ethiopian refugees and immigrants in urban communities in Minnesota. This difficulty in garnering immigrant participants may arise from their unwillingness to disclose personal problems to others, especially if those personal matters involve mental health issues which are often stigmatized in such communities (Al-Krenawi and Graham 2000; Khan 2006; Khawaja 2007). Additionally, in spite of our efforts to provide the study information translated into the most common languages spoken––which included an explicit discussion of confidentiality––we believe immigrant families may not see the utility of such research and may be fearful of how the information may be used. Thus, replicating this study with a larger sample of diverse youth, would enable an exploration of gender differences in greater depth, as well as an examination of within group differences among youth struggling with acculturation (e.g. exploring first- and second-generation sub-group differences) thereby improving our understanding of acculturation related conflicts and distress. Furthermore, the cross-sectional design of the current study precluded explorations of changes in the association between perceived ethnic identity conflicts and psychosocial outcomes over time.

Additional study limitations also exist regarding our measurement of various constructs. First, though we were interested in the adolescent’s perception of ethnic identity discrepancies and measure this construct as others have done (Lee et al. 2000; Rick and Forward 1992), relying solely on the perception of the child can present a limitation in terms of disentangling acculturation differences and family conflict. Moreover, as Birman (2006) notes, child depression may negatively distort their perceptions of acculturation related discrepancies, and family conflict thus possibly inflating the relationships among these variables. Thus, as was done by Phinney et al. (2000), inclusion of parents as participants would have been beneficial in that parents could report on their own ethnic identity rather than having children report on their perceptions of parents’ ethnic identity. Second, a direct measure of family conflict would have been useful in order to more accurately assess potential gender differences in the nature and quality of the conflict. Last, the ‘time lived in the US’ variable only assessed number of years the adolescent had lived in the US compared to capturing the proportion of their life lived in the country of resettlement as Berry et al. (2006) have done. Notwithstanding these flaws, the current investigation is a strong cross-sectional design with findings that shed new light on an area that has been otherwise unexplored.

Conclusion

Our findings are important, both for the questions they answer as well as the queries they generate. Cumulatively, these findings imply that though females and males do not differ with respect to parent–child ethnic identity discord, such discrepancies are associated with substantially poorer psychosocial outcomes for females. In terms of practice implications, these results suggest that when working with first- and second-generation adolescents and families, social workers must be sensitive to the role that acculturation discrepancies may play in initiating and exacerbating psychosocial distress among females.