Acculturation Gaps and Problem Behaviors among U.S. Southwestern Mexican Youth

Abstract

This article presents the findings of a study examining acculturation differences between adolescents and their mothers within Mexican immigrant families in the border region of the Southwest U.S. The main hypotheses of the study was that youth in mother-child dyads with mismatched acculturation strategies would report higher levels of externalizing problem behaviors than other adolescents, and that this relationship would be mediated by family conflict, acculturation conflict, family cohesion, and sense of familism.

The participants formed 142 dyads (N=284) of Mexican heritage mothers and their adolescent children. Regression analysis indicated that a gap or mismatch in acculturation strategies was associated with more externalizing youth problem behaviors.

Compared to the dyads where both mother and child were bicultural, only youth with more acculturated mothers demonstrated increased rates of externalizing problem behaviors. Family conflict and acculturation conflict mediated the relationship between acculturation gaps and externalizing behaviors. Implications for family-centered interventions as well as future research and policy implications are discussed.

Introduction

Children in immigrant families are the fastest growing group of children in the U.S., accounting for one-fourth of all children in the nation (Hernandez, Denton & Macartney, 2008). In Southwest border regions of the U.S., children of Mexican heritage increasingly represent a numerical majority in their schools and neighborhoods (Yabiku, et al., 2007). Many of these children embrace mainstream U.S values and often experience a clash of values with their immigrant or less acculturated parents, which can lead to family conflict and problems in family relationships (Choi, He, & Harachi, 2008). Due to the number of children affected, it is important to understand better how these cultural gaps affect the overall health and mental health of family members so that family interventions can be developed to prevent the erosion of protective factors from family and culture of origin, thereby reducing the risks associated with parent-child acculturation gaps.

Acculturation research has found an association between high levels of acculturation and adolescent externalizing problem behaviors, such as conduct problems and delinquency (Gonzalez, et al., 2008; Smokowski, Rose, & Bacallao, 2008). Acculturation gaps between parents and children can increase family conflict, which may increase the likelihood that youth will engage in deviant or externalizing problem behaviors (Mogro-Wilson, 2008; Szapocznick & Kurtines, 1993).

Acculturation gaps in immigrant families in the U.S. and their impact on family interactions and youth outcomes have been operationalized predominantly as a simple difference in degree of acculturation between parents and adolescents (Birman, 2006). To expand existing research, this study investigated the association between family acculturation gaps and youth self-reported externalizing problems using an expanded range of parent-child acculturation categories groups, and controlling for levels of family conflict.

Acculturation

Acculturation is a process of cultural adaptation characterized by changes in different psychosocial dimensions of individuals, families, and communities. It is triggered by intercultural contact which produces changes in attitudes, norms, behaviors, knowledge and identity (Berry, 2007). Acculturation is complex and multifaceted, and individuals may vary greatly in the rate and extent that they adopt the language, traditions, values, norms, and behaviors of a culture different than their culture of origin (Berry & Kim, 1988).

Acculturation is also multidirectional. As interactions among cultures may lead ethnic minority individuals to adopt cultural majority values, members of majority cultures may also acculturate into minority cultures, and individuals may acculturate from one minority culture into another. Individuals may also choose to maintain or regain a connection with their cultural heritage through enculturation, an effort to have a strong identification with a real and sometimes idealized culture of origin (Marsiglia & Kulis, 2009). Immigrant youth navigate these complex cultural processes while they transition from childhood into adolescence and then adulthood Hence, their regular developmental transitions are enriched and sometimes further stressed by acculturation experiences.

Theoretical Approach

To approach the acculturation phenomenon from a culturally-specific perspective, this study was guided by the eco-developmental approach (Szapocznik & Coatsworth, 1999) and Berry’s (2007) typology of four acculturation strategies. Berry’s schema is based on the degree of affiliation with origin and destination cultures: Assimilation (low on origin, high on destination), Separation/Withdrawal (high on origin, low on destination), Alienation (low on both), and Bicultural/Integration (high on both). In addition to Berry’s four acculturation strategies, a fifth strategy called Moderately Bicultural was added because recent research finds that a substantial proportion of Latino youth fall into this category (Coatsworth, et al., 2005). Moderately Bicultural individuals have affinities for both Latino and mainstream American cultures but not at the very high levels suggested by Berry’s Bicultural/Integration category. By utilizing this strategy, an expanded version of existing acculturation typologies (Berry, 2007) was possible, which addresses the unique characteristics of the fluid family acculturative environment present in border regions.

The eco-developmental approach (Szapocznik & Coatsworth, 1999) draws from Bronfenbrenner’s ecological perspective (1979) and provides a multidimensional framework to understand the effects of acculturation on Latino immigrant adolescents and their families. The approach stresses the interrelatedness of social environments and how these interactions produce behavioral outcomes (Szapocznick & Coatsworth, 1999). According to eco-developmental theory, acculturation creates a risk for Latino families at the macro system level, where differences in Latino and mainstream American culture may marginalize Latino immigrants or create acculturation differences between youth and their parents which may impact available social supports of parents, erode cultural protective factors, or create conflict (Szapocznick & Coatsworth, 1999). Due to the multidimensionality and variation of the acculturation process, it is important to consider the interrelatedness of such risk and protective factors that are culturally-based within Latino families.

Protective Factors: Familism and Family Cohesion

Culture can be a source of strength and moderate the potential negative impacts of acculturation on Latino families (Marsiglia, Nieri, & Becerra, 2008). Central to Mexican heritage and other Latino families is the concept of familism, the importance of family closeness, preserving good relationships, and contributing to each other’s well being (Marsiglia, Kulis, Wagstaff, Elek, & Dran, 2005; Santisteban, Muir-Malcolm, Mitrani, & Szapocznik, 2002).

Similarly, family cohesion is an important protective factor against deviant and problem behaviors among adolescents. Cohesion is the degree of connection between family members and the family unit. Less acculturated and more acculturated Latino families report higher cohesion, compared to families that are more acculturated to mainstream U.S. culture, or compared to bicultural families (Miranda, Estrada, & Firpo-Jimenez, 2000). Stronger adherence to traditional cultural beliefs concerning the prescribed roles of parent and child appears to create stronger emotional bonds among family members.

Potential Risk Factors: Acculturation and Family Conflict

In Mexican and other immigrant families, youth tend to acculturate at a faster rate than their parents due to environmental contexts in which they settle (Birman & Trickett, 2001; Szapocznik & Kurtines, 1993). Adults, on the other hand, tend to acculturate more slowly and are more likely to retain and pass on to their children the values of the culture of origin (Phinney & Vedder, 2006). As a result of the different pace of acculturation, an acculturation gap—where youth and their parent(s) are in different stages of acculturation—may develop. This gap can lead to acculturation conflict, which emerges when parents and children have different cultural compasses or values. This acculturation conflict can be a stressor for family life because it exacerbates the natural generational gap that exists between adolescents and their parents (Choi et al., 2008). In Mexican heritage families, acculturation conflict can erode the protective effects of familism and family cohesion against deviant adolescent behaviors (Gil, Vega, & Dimas, 1994). The stress caused by differences in the level of acculturation between parents and children can negatively affect parenting practices, the relationship between parent and child, and healthy youth development (Martinez, 2006).

Migration often adds to family stress because of the loss of social support in raising children. Immigration may lead to extended periods where parents are separated both from each other and from the larger extended family that provides assistance in child rearing. In addition, immigrant parents in the U.S. tend to spend long hours at work and away from their children (Pantin, Schwartz, Sullivan, Coatsworth, & Szapocnik, 2003). As youth acculturate to dominant U.S. values and affiliate less with their culture of origin, they report reduced family pride and begin to view parental involvement as controlling and authoritarian rather than a product of their parents’ desire to be involved and to be protective (Gil, et al., 1994; Lindahl & Malik, 1999).

Higher levels of parental acculturation influence children’s development and can predict negative outcomes (Vega, Gil, & Wagner, 1998). Low parental acculturation and low linguistic acculturation have also been studied as possible risks factors. Adults learn English more slowly than children because children are more likely to be exposed to English in schools and through the media. As immigrant children acculturate and English becomes their preferred language, protective family ties and effective parental communication and monitoring can erode (Marsiglia, Miles, Dustman, & Sills, 2002). In families where the parents rely on the youth for communication in English, traditional roles are challenged in that the parents are now dependent on their child for certain social and or operational interactions. This role reversal could impact the child – parent relationship and overall family functioning.

Family Conflict

Family conflict is defined as the emergence of frequent disagreements, fights, arguments, and anger between parents and children (Barber & Delfabro, 2000; Choi, et al., 2008). Family conflict increases the likelihood of emotional and behavioral problems across different ethnic groups (Ary et al., 1999; Szapocznik & Kurtines, 1993). Immigrant families may experience increased family conflict as children and parents acculturate and they acquire different values and behaviors (Birman, 2006; Szapocznick & Kurtines, 1993). Immigrant families appear to face greater value discrepancies than non-immigrant families. Large parent-child value differences have been reported when immigrant parents had a stronger affiliation to their culture of origin and the child was either assimilated or integrated within the host culture (Phinney & Veddar, 2006).

Previous studies with immigrant groups from various cultures have shown mixed results in supporting an association between parent-child acculturation gaps and family conflict. Some studies with Mexican-heritage samples found that family conflict and separation from parents mediate the relationship between acculturation processes and youth outcomes of aggression and deviant behavior (McQueen, Getz, & Bray, 2003). Other studies with similar populations found that parent-child dyads with a larger acculturation gap as measured by linguistic acculturation were not more likely to report family conflict (Pasch et al., 2006).

Acculturation Gaps and Youth Problem Behavior

Research about the experience of Mexican American youth and acculturation gaps is also inconclusive regarding externalizing problem behaviors. In some cases no connection between acculturation gaps and youth negative outcomes such as school misconduct or sexual behaviors has been reported (Pasch, et al., 2006). While similar findings were reported in a longitudinal study with preadolescents, the child’s perception of the quality of parent child relationship was found to moderate the association between acculturation gaps and reported youth negative behaviors (Schofield, Kim, Parke, & Coltrane, 2008). Findings with high risk Mexican American families did not support the acculturation gap-distress hypothesis but identified an unexpected group of parents who were more aligned with American culture than their at risk children (Lau et al., 2005).

Acculturation gap research is growing and diversifying its conceptual models but the assumption that acculturation differences between youth and their parent put children at greater risk for externalizing problems continues to be broadly accepted. A further assumption is that children are generally more acculturated to the host culture than their parents (Birman, 2006). In borderland regions these two assumptions need further exploration as the proximity to Mexico, the historical Mexican presence in the region, circular migration, and the constant influx of new immigrants may influence these processes in unique ways.

The ability of youth to incorporate both the host culture and the culture of origin appears to moderate intergenerational conflict (Miranda, Estrada, & Firop-Jimenez, 2000). To better understand the complex ways that acculturation gaps may impact youth in Mexican immigrant families, this study evaluated the effects of acculturation differences between Mexican heritage mothers and adolescents using the matched/mismatched acculturation style method of previous studies plus a modified bicultural scale (Coatsworth, et al, 2005). This modified scale represents a fuller range of individual mother and child acculturation strategies so that mother-child dyads can be explored based on an expanded set of Berry’s (1997) acculturation strategies: Separated, Alienated, Moderately Bicultural, Strongly Bicultural and Assimilated. This approach also allows for the exploration of children and mothers’ personal identity and their matching or mismatched cultural identities and how cultural conflict could be associated with problem behaviors and other health outcomes (Schwartz, Zamboanga, & Weisskirch, 2008).

From an eco-developmental perspective (Coatsworth, et al, 2005) it was hypothesized that adolescents more acculturated than their mothers will report a higher incidence of externalizing problem behaviors than adolescents who are bicultural and have bicultural mothers (either strong or moderate biculturalism). Family cohesion, familism and family conflict were treated as mediators between the acculturation mismatched mother-adolescent dyads and externalizing problem behaviors. It was further hypothesized that higher family cohesion and familism will decrease the incidence of externalizing problem behaviors among acculturating youth. Conversely higher family conflict and acculturation conflict will increase the likelihood of externalizing problem behaviors.

Methodology

Sample

This study uses data from a non-probability sample, the southwest component of the Latino Acculturation and Health Project, which examined acculturation and health outcomes. Data were drawn from the first, or baseline survey. The participants were 142 dyads of Mexican heritage mothers and adolescents, recruited in 2005 from ESL classes, community centers, local churches and community fairs in a large metropolitan area of the southwest U.S. Both the mother and adolescent in each dyad were interviewed separately at home. The university’s Institutional Review Board evaluated and approved the study’s protocol and its bilingual measures.

Measures

Participants self-identified their gender, age, time residing in the U.S., country of birth, and education. Gender was coded 0 for males and 1 for female adolescents. The education level for the mothers was skewed toward low levels of formal education, and thus was recoded into a dichotomous variable: 1 (less than a high school diploma) and 2 (high school diploma or higher). Academic achievement for the adolescents was measured by their self reported average grades in school. The original six response options were recoded into a dichotomous variable: 0 (other), 1 (mostly A’s and B’s).

Adolescent problem behaviors were assessed through the adolescents’ reports on the Youth Self Report (YSR) (Achenbach, & Rescorla, 2001). The YSR measures of adolescent problem behaviors produced acceptably reliable scales for externalizing problems (α = .72), which is a global measure of problems involving conflicts with others. The subscales of externalizing behaviors were also examined: aggressive behavior (α = .84), a 17 item scale measuring physical and verbal aggression; conduct problems (α = .65), a 15 item scale measuring lying, stealing, running away, setting fires, skipping school, threatening others, and lack of remorse for negative behaviors; oppositional defiant (α = .66), a 5 item scale measuring arguing and disobeying; and rule breaking (α = .71), a 15 item scale measuring adherence to rules in various settings.

Acculturation conflict within the family was measured using a 4 item scale (Vega, Alderete, Kolody, & Aguilar-Gaxiola, 1998) that assesses family problems caused by the youth’s choice of American values and norms over the values and norms of their family’s culture of origin, such as “How often have you had problems with your family because you prefer American customs?” (α = .86). To measure family conflict more generally apart from issues specifically related to acculturation, a 15 item scale was employed (Robin & Foster, 1989) that gauges family fights and lack of understanding among family members, such as “At least three times a week we get angry at each other” (α = .65).

Familismo (familism), was measured with a 6 item scale used in other studies of Latinos (Gil, et al., 2000) and found to be reliable in this sample (α = .86). An example item in the familism scale is “Family members feel loyal to the family.” The family cohesiveness scale (Olsen, 1991) includes 10 items that measure family bonding, how well they know each other, and how much they like spending time together (α = .89). An example cohesiveness item is, “Family members are supportive during difficult times.”

Acculturation was measured with the Bicultural Involvement Questionnaire (BIQ) (Szapocznik, Kurtines, & Fernandez, 1980). The BIQ measures the level of comfort with Latino and U.S. cultures in terms of language use, food, media consumption and cultural traditions. The BIQ Hispanicism scale has 20 items that are summed to indicate level of socialization to and affinity for a Latino way of life, and the Americanism scale has 20 items indicating socialization into and affinity for an Anglo-American way of life. Scores on the Hispanicism and Americanism scales were divided along the thresholds suggested by Coatsworth et al. (2005) to place each participating adolescent and mother into one of five acculturation groups: Alienated (low on both scales), Separated (high on Hispanicism, low on Americanism), Moderately Bicultural (moderately high on both), Strongly Bicultural (very high on both), and Assimilated (low on Hispanicism, high on Americanism). Rather than assuming a single underlying acculturation continuum, these groups represent a set of theoretically derived acculturation strategies (Berry, 2007).

Previous studies have measured the acculturation “gap” between parents and their children by subtracting the parent’s from the child’s overall acculturation score (Birman, 2006; Martinez, 2006). Problems with this approach include the confounding of distinct acculturation strategies for parent-child dyads that do not differ in acculturation (e.g., highly acculturated dyads are equated with highly unacculturated ones), and the need to simplify acculturation to a single continuum so that global scores for parent and child can be contrasted. Instead, this study created an acculturation gap typology based on the joint placement of adolescents and parents into the five theoretically meaningful acculturation groups. An advantage of this approach is that the large numbers of moderately bicultural respondents were identified.

Table 1 presents a cross-tabulation of the adolescent-by-mother placements into acculturation groups. Among both mothers and adolescents, the most common acculturation category was Moderately Bicultural, which described 35% of the mothers and 42% of the adolescents. An additional 16% of the mothers and 39% of the adolescents were Strongly Bicultural. Thus more than half the parents and four-fifths of the adolescents were moderately or strongly bicultural. Only one other acculturation category, Separated, described substantial proportions of respondents, including 44% of the mothers and 14% of the adolescents. Four percent of mothers and of adolescents were Assimilated, and less than one percent of each was Alienated.

TABLE 1.

Mother and Adolescent Joint Acculturation Status

MOTHER
		Alienated	Separated	Moderately Bicultural	Strongly Bicultural	Assimilated	Mother Total (%)
ADOLESCENT	Alienated	0	0	1	0	0	1 (0.7%)
	Separated	0	15	5	0	0	20 (14.1%)
	Moderately Bicultural	1	26	23	4	5	59 (41%)
	Strongly Bicultural	0	19	19	18	0	56 (39.4%)
	Assimilated	0	3	2	1	0	6 (4.2%)
	Adolescent Total (%)	1 (0.7%)	63 (44.4%)	50 (35.2%)	23 (16.2%)	5 (3.5%)	142 (100%)

As indicators of mother-adolescent differences and similarities in acculturation, the dyads were then assigned to one of four groups based on their joint placements in the five acculturation categories: Both Mother and Adolescent Bicultural (both mothers and adolescents were Moderately Bicultural or Strongly Bicultural; N=64 dyads); Both Mother and Adolescent Separated (N=15 dyads); Adolescent More Acculturated than Mother (the adolescents were Assimilated and the mother was not, or the adolescents were Moderately or Strongly Bicultural and the mothers were in the Alienated or Separated groups; N=52 dyads ); Mother More Acculturated Than Adolescent (the mothers were Assimilated and the adolescent was not, or the mothers were Moderately Bicultural and the adolescents were in the Alienated or Separated groups; N=11 dyads). There were no dyads where mothers and adolescents were both Assimilated. Thus over half the mothers and adolescents shared the same acculturation status, mostly when both were moderately or strongly bicultural. In dyads differing in acculturation, the adolescent was usually the more acculturated (assimilated or bicultural when mother was not).

Analyses

The analysis tested whether acculturation gaps between mothers and adolescents are associated with adolescent self reports of externalizing problem behaviors, including aggression, conduct problems, oppositional defiance, and rule breaking. Bivariate relationships with the outcome variables were explored through correlations. Two sets of multivariate ordinary least square regression models were estimated. The first model included as predictors of problem behaviors a set of dummy variables representing the four joint parent-child acculturation categories with Both Bicultural, serving as the reference group. All models included controls for the adolescent’s gender, age, school grades, time in the U.S., and the mother’s education. In order to determine if the effects of mother-adolescent acculturation gaps on youth problem behaviors were attributable to certain family and social network dynamics, a second set of models added family cohesiveness, familism, family conflict, and acculturation conflict. A subsequent test of mediation analysis was conducted using the Sobel test.

Results

Table 2 presents descriptive statistics for variables used in analysis, the acculturation groups, and demographic descriptors of the sample. Means for the outcome variables were far below the recommended YSR thresholds for classifying deviant behaviors as meeting clinical or borderline levels (Achenbach & Rescorla, 2001). The percentage of adolescents in the sample in the clinical range was 9% for externalizing problems and aggressive behavior, 7% for oppositional defiant behavior, and 11% for conduct problems. Another 7% of adolescents were in the borderline range for each of these problem behaviors. For rule breaking there were no adolescents in the borderline or the clinical range.

TABLE 2.

Descriptive Statistics

	N	Mean	Standard Deviation	Range
Outcomes
YSR Externalizing Behavior	142	11.96	7.31	1 – 37
YSR Aggressive Behavior	142	9.18	5.14	0 – 28
YSR Oppositional Defiant	142	2.49	1.83	0 – 8
YSR Conduct Problems	142	3.58	2.83	0 – 13
YSR Rule Breaking	142	4.44	2.93	0 – 14
Acculturation Scales
Adolescent BIQ Hispanicism	142	3.30	0.72	1 – 4
Adolescent BIQ Americanism	142	2.91	0.83	1 – 4
Mother BIQ Hispanicism	142	3.41	0.70	1 – 4
Mother BIQ Americanism	142	2.18	0.91	1 – 4
Joint Mother-Adolescent Acculturation Dummy Variables
Both “Separated”	142	0.11	0.31	0 – 1
Mother More Acculturated than Adolescent	142	0.08	0.28	0 – 1
Adolescent More Acculturated than Mother	142	0.37	0.48	0 – 1
Both Moderately or Highly Acculturated (Reference)	142	0.45	0.56	0 – 1
Hypothesized Mediators
Family Cohesiveness	142	37.21	8.14	14 –50
Familism	142	20.12	3.20	9 – 24
Family Conflict	142	3.47	3.83	0 – 15
Acculturation Conflict	141	7.23	3.12	4 – 17
Controls and Sample Demographics
Adolescent Gender (Male=0; Female=1)	142	0.62	0.49	0 – 1
Adolescent Age (years)	141	15.53	1.24	13 –18
Adolescent School Grades (0=Other; 1=Mostly As and Bs)	140	0.62	0.48	0 – 1
Adolescent Time in U.S.	141	11.50	5.36	0.2 –
Adolescent Born Outside USA (0=No, Yes=1)	142	0.44	0.56	0 – 1
Mother’s Education (0=<H.S.; 1=H.S.)	142	0.34	0.47	0 –1
Mother’s Age (years)	141	39.80	6.65	24 –60
Mother’s Time in U.S.	141	16.98	12.60	0.3 –60
Mother Born Outside USA (0=No, 1=Yes)	142	0.87	0.34	0 – 1

Means for the BIQ scales indicated that mothers were generally less acculturated than their adolescent child: the mothers reported somewhat higher Hispanicism but lower Americanism scores than their adolescent did. As reported above, mother-adolescent dyads most commonly were both bicultural (45%) or the adolescent was more acculturated than the mother (37%), with smaller numbers of dyads where both were relatively unaccultured or “separated” (11%), and a small group where mother was more acculturated than adolescent (8%).

Means on hypothesized mediating variables tended to be high—in the top quartile of the range—for family cohesiveness and familism, low for general family conflict, but in the mid-range for acculturation specific parent-child conflict.

Bivariate correlations between outcome and predictor variables indicated that the strongest relationships were with the hypothesized mediator variables. Family conflict (r = .41 to .56) and acculturation conflict (r = .18 to .23) were directly correlated with all of the problem behavior scales, while family cohesiveness and familism were inversely correlated (r = −.20 to −.43) with all outcomes. The correlations among these mediators were notable, with the largest being an r = .60 correlation between cohesiveness and familism, and r = −.58 between family conflict and familism. Due to the substantial correlations among the hypothesized mediators collinearity diagnostics were considered in all multivariate regression models, and no problematic levels of collinearity were found (all VIFs < 1.8). The dyadic acculturation gap measures were not strongly correlated with the outcomes. The largest relationship was with the dummy variable for dyads where parents were more acculturated than adolescents: here there were modest, marginally significant (p < .10) correlations with the outcomes (r = .06 to .16).

Table 3 presents the results of the multivariate ordinary least square regression models predicting the youths’ self reported aggressive behavior, oppositional defiance, conduct problems, rule breaking, and the global externalizing problem behaviors scale. The first models for each outcome did not include the hypothesized mediating variables. These models showed that, compared to the reference group where both mother and adolescent were bicultural, youth with mothers who were more acculturated than themselves reported significantly higher scores on the global externalizing problems scale. However, the larger group of dyads where the adolescent was more acculturated than the mother was not significantly different than the reference group on any outcome. These effects persisted when controlling for demographic variables (gender, mother’s education and time in the U.S.) that were significant predictors of the outcomes. Females reported higher scores than males on aggressive behavior and oppositional defiance. Youth with more highly educated mothers reported lower scores on conduct problems and rule breaking. Scores on all outcome measures were higher for youth who had lived in the U.S. longer.

Table 3.

Predictors of YSR Externalizing Problem Behaviors and Subscales

	Externalizing Behavior		Aggressive Behavior		Oppositional Defiant		Conduct Problems		Rule Breaking
Gender	2.026 (1.258)	2.360* (1.047)	2.283** (.844)	2.041** (.759)	.831** (.302)	.781** (.287)	.197 (.481)	.317 (.399)	.226 (.512)	.354 (.448)
Age	0.315 (.491)	.218 (.416)	.114 (.335)	.055 (.302)	.058 (.120)	.041 (.114)	−.009 (.191)	.021 (.158)	.247 (.203)	.279 (.178)
Grades	−0.137 (.358)	−.097 (.292)	−.029 (.240)	−.050 (.212)	−.060 (.086)	−.066 (.080)	−.039 (.137)	−.051 (.111)	−.036 (.145)	−.045 (.125)
Adolescent Time in U.S.	0.261* (.130)	−.042 (.114)	.171* (.087)	.005 (.083)	.105*** (.031)	.056+ (.031)	.084+ (.050)	−.047 (.043)	.091+ (.053)	−.035 (.480)
Mother’s Education	−1.037 (.523)	−.704+ (.432)	−.527 (.352)	−.300 (.313)	−.137 (.126)	−.064 (.119)	−.342+ (.201)	−.174 (.164)	−.529* (.214)	−.362* (.185)
Both Separated	2.071 (2.480)	−.506 (2.102)	1.752 (1.680)	.660 (1.525)	.542 (.600)	.227 (.577)	.497 (.958)	−.388 (.800)	−.367 (1.018)	−1.261 (.900)
Mother More Acculturated than Adolescent	5.068* (2.224)	2.537 (1.911)	2.779+ (1.534)	1.281 (1.386)	.964+ (.548)	.468 (.525)	2.061* (.875)	.972 (.727)	2.276* (.929)	1.188 (.818)
Adolescent More Acculturated than Mother	0.832 (1.330)	−.606 (1.137)	.861 (.904)	−.212 (.825)	.078 (.323)	−.258 (.312)	.451 (.516)	−.172 (.433)	.263 (.548)	−.348 (.487)
Both Bicultural (Reference)	--	--	--	--	--	--	--	--	--	--
Cohesiveness		−.045 (.080)		.050 (.058)		−.001 (.022)		−.063* (.030)		−.071* (.034)
Familism		−.160 (.207)		−.176 (.150)		−.007 (.057)		−.047 (.079)		.013 (.089)
Family Conflict		.872*** (.162)		.550*** (.117)		.159*** (.044)		.333*** (.062)		.330*** (.069)
Acculturation Conflict		.212 (.169)		.186 (.122)		.068 (.046)		.027 (.064)		.027 (.072)
N	139	138	139	138	139	138	139	138	139	138
Adjusted R2	.048	.345	.060	.275	.112	.234	.016	.356	.058	.310

Unstandardized regression coefficients and standard errors (in parentheses)

⁺p < .10

^*p < .05

^**p < .01

^***p < .001

The second set of regression models in Table 3 explored the possible mediating role of family cohesion, familism, family conflict, and acculturation conflict in the relationship between mother-adolescent acculturation gaps and youth problem behaviors. These regression models indicated that family conflict was significantly related to higher scores on all five youth problem behavior outcomes, and family cohesiveness had a significant negative relationship with conduct problems and rule breaking. However, after controlling for the hypothesized mediators, none of the mother-adolescent joint acculturation categories demonstrated significant relationships with any of the outcomes, suggesting the presence of a substantial indirect effect. These relationships were explored further in two ways. First, the four possible mediators were entered singly into the base models with joint acculturation categories and controls. In those analyses the dummy variable representing mothers more acculturated than adolescent did continue to have a significant positive relationship with the four externalizing behavior subscales (aggressiveness, oppositional defiance, conduct problems, and rule breaking) when familism or family cohesion were included, but not when acculturation conflict or family conflicts were included. A second additional analysis assessed the mediating relationship by applying the Sobel test of significance of the indirect relationship (Kline, 2005). Results indicated that both family and acculturation conflict were significant mediators connecting acculturation differences (mother more acculturated than adolescent) to the outcomes. However, the indirect effects through general family conflict (.05 for aggressiveness, .04 for oppositional defiance, .06 for conduct problems, and .12 for rule breaking) were more sizable than through acculturation conflict (.02 or .03 for all indirect effects). Because family conflict and acculturation conflict were only modestly correlated (r = .254), the

We explored an additional possible reason that youth who were less acculturated than their mothers reported more externalizing problem behaviors—that these youth may have arrived in the U.S. after their mothers had immigrated. However, after controlling for a difference between mother and adolescent in length of residence in the U.S., youth with more acculturated mothers continued to report significantly higher levels of adolescent problem behaviors as they had in the first models in Table 3 (results not presented in tables).

Discussion

Previous studies have found that adolescent problem behaviors are significantly associated with an acculturation gap between parent and child (Birman, 2006; Szapocznick & Kurtines, 1993). The results of this study did not support the hypothesis that Mexican-heritage youth who are more acculturated than their mothers would report more problem behaviors. Although this study found that acculturation gaps were associated with significantly higher reported levels of externalizing problems, these problems were reported at elevated levels by youth from dyads where the mothers were more acculturated than the adolescents. This finding from a low risk group of families (i.e., reporting low rates of adolescent externalizing problem behaviors) supports the previous findings from a high risk Mexican American sample (Lau, et al, 2005). Explorations of possible reasons for this relationship suggested that both family and acculturation conflict are implicated in youth reports of an array of externalizing problems when their parents are more acculturated. Of these two factors, only family conflict remained as a significant predictor of youth problem behaviors in full models and the indirect effect of the salient acculturation difference through family conflict was larger than its counterpart through acculturation conflict. These results indicate that although acculturation gaps may exist between adolescents and parents in Mexican immigrant families, and these youth may exhibit problem behaviors, those behaviors may not necessarily be due to the acculturation gap itself but rather to intra-familial conflict.

Some of these findings are different than the results found in other studies (Gonzalez, et al, 2008) and further studies are needed to examine the unique situation where a parent is more acculturated than the adolescent and how mismatched acculturation affects adolescent behaviors. For example, reported parent-adolescent difference in acculturation may reflect different perceptions and experiences during the acculturation process. Parents who made the decision to immigrate with their families and remain in the U.S. may have a more positive view of the culture of their adopted country and their own incorporation of its values. Perhaps because their children are better able to master English language acquisition and enter the social networks of the native born, they may be more aware of difficulties and trade-offs in the acculturation process, such as the lack of full acceptance of immigrants and Latinos in U.S. society.

This study was conducted with Mexican heritage families living in a border state. The proximity to Mexico facilitates a fluid movement across the border. More information is needed about the frequency, purpose, and length of the adolescents’ visits to Mexico. From an ecological perspective, family networks in Mexico might continue to be providing effective support to youth after migration to the U.S. (Gil, et al., 1994). In some cases, youth may be sent back to their towns of origin in Mexico as a strategy to correct externalizing behaviors and to distance them from U.S. peers who are perceived to have a negative influence. More information about this circular movement between the U.S. and Mexico may help explain the identified counterintuitive association between youth’s higher rates of externalizing behaviors and their parents’ relatively higher level of acculturation. Children born in the U.S. may have an easier time crossing back and forth than their immigrant parents born in Mexico, potentially resulting in children with stronger social and cultural ties to Mexico than their parents.

Another area requiring more in depth study is the unexpected finding that female adolescents reported higher rates on some externalizing behavior outcomes than their male counterparts. The greater sheltering of girls within traditional Mexican homes compared to boys, may not prepare them as well to cope with expanded opportunities to engage in risk behaviors in U.S. society. Gender role transitions for Mexican heritage girls may involve more abrupt or drastic changes than for boys, introducing conflict in families (Choi, et al., 2008). Parents may also have a more difficult time accepting the girls’ changes in norms and behaviors which may lead to more overt conflict.

Although this study provides important insights into possible causes of youth problem behaviors, it has several limitations. The sample size is modest and is not based on random probability sampling; therefore the results cannot be generalized to all Mexican immigrant groups. The modest sample size also resulted in small numbers in one of the acculturation gap groups, which may have affected the stability of the results. The study included only mothers and their adolescent children. Future research should apply innovative recruitment strategies to include Mexican immigrant fathers as well. More attention needs to be given to developmental differences between parents and their children and how the two generations interpret and answer questions gauging their acculturation strategies.

The findings suggest possible insights into the acculturation gap between children in immigrant families and their mothers in the U.S.-Mexico border region and advance knowledge about the experiences of Mexican families, their strengths and their needs. Policy makers can utilize these findings to inform the development of community and school policies that support language diversity and youth extracurricular activities. These findings can also inform the design of family interventions addressing the needs of children and parents facing the challenges of acculturation. The acculturation process is unavoidable but practitioners and researchers can assist immigrant Mexican origin parents and youth to navigate the sometimes turbulent waters of acculturation by relying on the strengths of the culture and the family. Children with externalizing problems may benefit from a combination of strategies applied within their immigrant families and with the support of practitioners.

Since much of the existing literature cites acculturation gaps between parents and children as a main factor in youth problem behaviors, teachers, social workers, psychologists and other professionals may incorrectly attribute all youth problem behaviors to acculturation gaps. Practitioners and researchers need to examine the issues affecting immigrant families in more depth because youth problem behaviors may be due to other family conflicts and not necessarily specifically due to acculturation conflict caused by acculturation gaps. Further research will be improved by involving youth and their families throughout the inquiry process and by adding a qualitative or narrative dimension. There is a need to better understand the meaning of the acculturation experience by different families and in different social contexts. Family histories and family constellations will help assess in greater depth which family conflicts would have come to surface in the country of origin even if the migration process did not occur.

The results of this study support previous findings indicating that family cohesiveness appeared to be a selective protective factor against externalizing youth problem behaviors (Volk, Edwards, Lewis, & Sprenkle, 1989). As a result, practitioners can utilize a strengths based approach to work with immigrant families to enhance the culturally valued concept of family cohesion. Replications of the current study comparing Mexico-based samples with Mexican-heritage U.S. samples will help advance knowledge about family dynamics versus acculturation in a globalized context such as the U.S.-Mexico border region (Marsiglia, Kulis, Martinez Rodriguez, Becerra, & Castillo, 2008). The U.S.-Mexico borderlands’ cultural and social context presents unique opportunities for future research with young people and their families immersed in Latino majority environments and engaged in enculturation processes (Marsiglia & Kulis, 2009) which are either by choice or a reaction to perceived discrimination or lack of acceptance by the larger society. Because the acculturation processes appears to be similar for other Latino subgroups as well as for other immigrant and migrant communities -with caution- these findings can be generalized to other populations.