The Nature and Correlates of Mexican-American Adolescents’ Time With Parents and Peers

Abstract

Drawing on cultural –ecological and person –environment fit perspectives, this study examined links among Mexican-American adolescents’ time with peers and parents, parents’ cultural orientations, and adolescents’ psychosocial adjustment and cultural orientations. Participants were 492 Mexican-American adolescents (Ms = 15.7 and 12.8 years for older siblings and younger siblings) and their parents in 246 families. Family members described their family relationships, cultural orientations, and psychosocial functioning in home interviews, and time-use data were collected during a series of nightly phone calls. Mexican-American adolescents spent the majority of their peer time with Mexican youth. Some support was found for the hypothesis that the mismatch between parents’ cultural orientations and adolescents’ peer involvement is linked to adolescents’ psychosocial functioning.

Adolescents’ daily activities are important contexts for development (Bronfenbrenner, 1979). Grounded in an ecological perspective, a growing body of work documents that how and with whom youth spend their time has important implications for their psychosocial functioning (e.g., Larson & Verma, 1999; McHale, Crouter, & Tucker, 1999; Osgood, Wilson, O’Malley, Bachman, & Johnston, 1996). As yet, however, we know little about the nature and developmental significance of Latino adolescents’ activities, despite the fact that they are part of the largest and fastest growing ethnic minority group in the United States (Ramirez & Patricia de la Cruz, 2002).

Scholars interested in minority youth and families have called for more research on normative socialization processes, noting that most research on minority youth focuses on risk and pathology (García Coll et al., 1996; Hagen, Nelson, & Velissaris, 2004; McLoyd, 1998). In addition, these researchers highlight the importance of ethnic-homogeneous designs, designs that illuminate the diversity of experiences within cultural groups, and the cultural and ecological factors that give rise to within-group variations in youth development. Ethnic-homogeneous designs move the field away from pathologizing minority youth (García Coll et al., 1996; McLoyd, 1998) as often happens when a mainstream culture group is treated as the standard of comparison. This study focused on Mexican-American adolescents’ time use. Specifically, we examined the social contexts of youth’s daily activities and the links between youth’s time in different social contexts and both their psychosocial functioning and their cultural orientations. Although the links between adolescents’ time use and psychosocial adjustment have been investigated in other ethnic groups, we found no prior research examining these associations in Mexican-American youth, nor has prior work investigated the connections between adolescents’ everyday activities and their cultural orientations.

Social Contexts of Mexican-American Adolescents’ Daily Activities

Ecological and cultural frameworks posit that activities are building blocks of development (Bronfenbrenner, 1979): Daily activities provide opportunities for developing interpersonal bonds, for learning skills, and for forming a self-identity (Bronfenbrenner, 1979; Larson & Verma, 1999). Elaborating on ecological and contextual paradigms, Szapocznik and Kurtines (1993) drew attention to the multicultural settings in which individuals’ and families’ lives are embedded and underscored the significance of the “culturally pluralistic milieu” (pp. 401 –402) that is an everyday reality for minority youth. To better understand the social milieu of Latino youth, our first goal was to describe Mexican-American adolescents’ time spent in Mexican, non-Mexican, and ethnically heterogeneous peer groups as well as adolescents’ time with parents and to explore the family correlates of these daily activity patterns. We focused on Mexican-American adolescents from two-parent families, a group that currently represents the modal family structure of Mexican families raising children in the region of the country where the data were collected (U.S. Census Bureau, 2000). Seventh graders and their next older siblings participated, allowing us to examine potential differences in adolescents’ time use during early and middle adolescence. Given the increasing salience of peers as youth move through adolescence (Richards, Crowe, Larson, & Swarr, 1998), we expected that older adolescents would spend more time with peers than younger adolescents. We also anticipated gender differences in adolescents’ time with peers and parents. Specifically, the greater autonomy and fewer family responsibilities of Mexican-American boys versus girls (e.g., Azmitia & Brown, 2002; Valenzuela, 1999) led us to hypothesize that boys would spend less time with parents and more time with peers.

Our conceptual grounding in a cultural – ecological perspective also directed attention to the ecological conditions that underlie adolescents’ choices about their companions (García Coll et al., 1996; McAdoo, 1993). We examined a range of cultural background/context characteristics, including parents’ place of birth, parents’ cultural orientations, and the ethnic composition of families’ places of residence, and their links with adolescents’ time with Mexican, non-Mexican, and ethnically heterogeneous peer groups as well as their time with parents. We hypothesized that higher levels of parental enculturation (i.e., involvement in Mexican culture) would be positively related to adolescents’ time with Mexican peers, and that parent acculturation (i.e., involvement in U.S. culture) would be linked to adolescents’ time in non-Mexican peer groups. Given the emphasis in Mexican culture on family cohesiveness and interdependence (Cauce & Domenech-Rodriguez, 2002), we also expected that adolescents would spend more time with parents when parents reported closer ties to Mexico.

The role of neighborhood ethnic composition and families’ economic resources is also important to consider in understanding the opportunities for and constraints on adolescents’ choice of companions. Reasoning that adolescents’ time in different types of peer groups may depend on the availability of peers, we predicted that adolescents would spend more time with Mexican peers when there was a higher percentage of Latinos in their neighborhoods. In addition, drawing on research on culture and socioeconomic status (McLoyd, 1998; Negy & Woods, 1992), we hypothesized that adolescents from families with more resources would spend more time in non-Mexican and ethnically heterogeneous peer contexts.

Parents’ Cultural Orientations and Adolescents’ Peer Involvement: Links with Adolescents’ Psychosocial Functioning

Our second goal was to examine the links between the social contexts of Mexican-American adolescents’ daily activities and both their psychosocial functioning and parent – adolescent relationship experiences. In addressing this goal, we drew on a person – environment fit perspective (Eccles et al., 1993; Lerner & Lerner, 1983), which highlights the importance of congruence across different facets of youth’s experiences. The sometimes conflicting values and traditions of the two cultures in which Mexican-American youth spend their time (García Coll et al., 1996; Szapocznik & Kurtines, 1993) mean that issues of congruence may be especially pertinent to understanding their psychosocial functioning. In their work with Cuban Americans, for example, Szapocznik and Kurtines (1993) proposed that discrepancies between parents and youth in acculturation contribute to problematic parent – adolescent relationships and youth maladjustment. Other researchers have highlighted the potential negative implications of intergenerational discrepancies for parent – child relationships in immigrant families (Phinney, Ong, & Madden, 2000; Rumbaut, 2005; Vega, Khoury, Zimmerman, Gil, & Warheit, 1995).

Building on these ideas, we hypothesized that incongruence between parents’ cultural orientations and adolescents’ peer time would be associated with more problematic adjustment and less parental acceptance. Specifically, we examined two types of mismatches between parents’ cultural orientations and adolescents’ time spent with peers: parents’ Mexican orientations and adolescents’ time with non-Mexican peers and parents’ Anglo orientations and adolescents’ time with Mexican peers. Testing a goodness-of-fit prediction, we predicted that adolescents’ psychosocial functioning would be optimal when their choice of peer groups matched their parents’ cultural orientations. Extant research and theory proposes that problems may arise when youth adapt to U.S. culture at a faster rate than their parents (e.g., Szapocznik & Kurtines, 1993), suggesting that parents’ stronger ties to Mexican culture, in combination with youth’s stronger ties to Anglo culture, may result in strained parent – child relationships and youth involvement in maladaptive behavior. That a mismatch defined by parents’ stronger orientations to Anglo culture in combination with youth’s stronger ties to Mexican peers would result in strained parent – adolescent relationships and problematic youth adjustment is also consistent with a person – environment fit perspective, but has been given little empirical attention. We explored the implications of both types of mismatches in the current study.

In addressing our second goal, we also expanded on extant research, which focuses largely on mothers, to consider both mothers’ and fathers’ cultural orientations and their links to adolescent well-being (Gonzales, Knight, Morgan-Lopez, Saenz, & Sirolli, 2002). Mothers and fathers may be important in youth adjustment for different reasons. For instance, research highlights mothers’ domestic responsibilities and their role in the transmission of cultural knowledge and values (Buriel & Cardoza, 1993; Cauce & Domenech-Rodriguez, 2002; Ramirez & Arce, 1981). Experiences with fathers, in contrast, are more leisure-oriented and peer-like, making father – offspring relationships especially relevant to youth’s peer competence and involvement (Parke & Buriel, 1998). Given parents’ differing family roles, incongruence between adolescents’ activities and the values and expectations of their fathers may have different implications than incongruence involving mothers’ expectations.

Parents’ Cultural Orientations and Adolescents’ Time with Peers and Parents: Links to Adolescents’ Cultural Orientations

Our final goal was to explore whether mothers’ and fathers’ cultural orientations, in concert with youth’s daily activities, have implications for adolescents’ cultural orientations. Perspectives on the development of cultural orientations highlight the bi-dimensional nature of cultural adaptation and specify two independent and concurrent processes: acculturation, or the process of involvement in and adoption of the values and beliefs of the host culture, and enculturation, or involvement in the culture of origin (Berry, 2003; Knight, Bernal, Cota, Garza, & Ocampo, 1993; Phinney, 1990). Theorists also draw attention to the multifaceted nature of cultural adaptation, noting that change occurs along multiple dimensions (e.g., knowledge, attitudes, behaviors, self-concept). Moreover, Knight and colleagues (under review) propose that cultural adaptation processes differ across development, becoming more complex and differentiated in adolescence. Furthermore, in adolescence, youth assume a more active role in the development of their cultural orientations (García Coll & Pachter, 2002), have more opportunity to explore the world outside the home, and have more autonomy in choosing how to spend their time.

Our focus on the role of parents and peers in adolescents’ cultural orientations is consistent with both cultural and developmental perspectives. From a cultural perspective, for example, Berry (2003) directed attention to the importance of both vertical (i.e., parent) and horizontal (i.e., peer) socialization experiences in the transmission of cultural values and knowledge. From a developmental perspective, although adolescence is a period when peers become important sources of companionship and social support, relationships with parents remain important (Steinberg & Silk, 2002). In the case of Latino youth, research documents the significance of friendships (Stanton-Salazar & Spina, 2005; Suárez-Orozco & Suárez-Orozco, 1995; Way, Cowal, Gingold, Pahl, & Bissessar, 2001); the cultural emphasis on family support and interdependence, however, means that Mexican-American parents also play an influential role in youth’s well-being and development (Cauce & Domenech-Rodriguez, 2002; Vega, 1990).

In this study, we extended the investigation of adolescents’ cultural orientations in two ways. First, we examined the interaction between parents’ cultural orientations and adolescents’ time with Mexican and non-Mexican peers in youth’s cultural orientations. Specifically, we drew on a person – environment fit perspective to hypothesize that congruent socialization experiences (i.e., matches between parents’ cultural orientations and adolescents’ peer involvement) would be associated with stronger cultural orientations in adolescents. As such, adolescents whose parents were oriented toward Mexican culture and who also spent more time with Mexican peers should report stronger Mexican orientations; in contrast, adolescents with more acculturated parents, who spent more time with non-Mexican peers, should report stronger Anglo orientations. We also examined parents’ cultural orientations and adolescents’ time with parents, testing the prediction that parents’ enculturation, in combination with more time with parents, would be associated with stronger orientations toward Mexican culture in youth. This prediction is consistent with the idea that parents play a primary role in enculturation processes (Knight et al., 1993).

Method

Participants

The data came from a study of family socialization and adolescent development in Mexican-American families (McHale, Updegraff, Shanahan, Crouter, & Killoren, 2005; Updegraff, McHale, Whiteman, Thayer, & Delgado, 2005). The 246 participating families were recruited through schools in and around a southwestern metropolitan area. Given the goal of the larger study, to examine normative family, cultural, and gender role processes in Mexican-American families with adolescents, the criteria for participation were as follows: (1) mothers were of Mexican origin; (2) seventh graders were living in the home and were not learning disabled; (3) an older sibling was living in the home; (4) biological mothers and biological or long-term adoptive fathers lived at home (all nonbiological fathers had been in the home for a minimum of 10 years); and (5) fathers worked at least 20 hr/week. Most fathers (i.e., 93%) also were of Mexican origin. Importantly, our sampling criteria and our focus on a local population mean that our sample was not designed to be representative of Mexican-American families in general. Instead, the overall study goals directed our attention to two-parent families (so that we could examine the role of fathers) with two siblings in a circumscribed age range (so that we could compare the experiences of older and younger adolescents). Paternal employment was a criterion, given our interest in the larger contexts within which family dynamics emerge and research documenting the role of work in defining parents’ roles in Mexican-American families (Baca Zinn, 1980; Coltrane & Valdez, 1993).

To recruit families, letters and brochures describing the study (in both English and Spanish) were sent to families, and follow-up telephone calls were made by bilingual staff to determine eligibility and interest in participation. Families’ names were obtained from junior high schools in five school districts and from five parochial schools. Schools were selected to represent a range of socioeconomic situations, with the proportion of students receiving free/reduced lunch varying from 8% to 82% across schools. Schools also differed in the percentage of students who were Latino (range = 6–63% for public junior high schools). Letters were sent to 1,856 families with a Latino seventh grader who was not learning disabled. For 396 families (21%), the contact information was incorrect and repeated attempts to find updated information through school personnel or public listings were unsuccessful. An additional 42 (2.4%) families moved between the initial screening and final recruitment contact, and 146 (10%) refused to be screened for eligibility. Eligible families included 421 families (23% of the initial rosters and 32% of those we were able to contact and screen for eligibility). Of those who were eligible (n = 421), 284 (or 67%) agreed to participate, 95 (23%) refused, and we were unable to recontact the remaining 42 families (10%) who were eligible to determine whether they would participate. Interviews were completed by 246 families. Those who agreed but did not participate in the final sample (n = 38) were families that we were unable to locate to schedule the home interview, that were unwilling to participate when the interview team arrived at their home, or that were not home for repeated interview attempts. Because we had surpassed our target sample size (N = 240), we did not continue to recruit the latter group.

Families represented a range of education and income levels, from poverty to upper class (see Table 1). The percentage of families that met federal poverty guidelines was 18.3%, a figure similar to the 18.6% of two-parent Mexican-American families living in poverty in the county from which the sample was drawn (U.S. Census Bureau, 2000). Most parents had been born outside the United States (71% of mothers and 69% of fathers); this subset of parents had lived in the United States an average of 12.4 (SD = 8.9) and 15.2 (SD = 8.9) years, for mothers and fathers, respectively. About two thirds of the interviews with parents were conducted in Spanish. Older siblings were 50% female and 15.70 (SD = 1.6) years of age on average. Furthermore, 47% had been born outside the United States, and 82% were interviewed in English. Their younger siblings were 51% female (n = 125) and 12.8 (SD = .58) years of age on average. Of these youth, 38% had been born outside the United States, and 83% were interviewed in English.

Table 1.

Background Characteristics (N = 246)

	M	SD
Family income	$53,183.00	$45,381.00
Years of educationa
Mothers	10.34	3.74
Fathers	9.83	4.40
Age
Mothers	39.00	4.63
Fathers	41.70	5.77
Older siblings	15.70	1.54
Younger siblings	12.77	0.58
No. of children in household	3.79	1.60

^aNote. 12 = high school graduate, 16 = college graduate.

Procedures

Data were collected using two procedures. First, during in-home interviews lasting an average of 3 hr for parents and 2 hr for adolescents, family members reported on their family relationships, cultural values and backgrounds, and psychosocial adjustment. Interviews were conducted individually using laptop computers by bilingual interviewers in separate locations in the home. The majority of questions were read aloud (due to variability in parents’ and adolescents’ reading levels) and interviewers entered family members’ answers into the computer. For questions that were sensitive in nature (e.g., risky behaviors, delinquent peer affiliations), adolescents were able to enter their answers directly into the laptop computers. Informed consent was obtained before the interview.

During the 3 – 4 weeks following the home interviews, families were telephoned on 7 evenings (5 weekday evenings and 2 weekend evenings) and reported on their activities during the day of the call; adolescents participated in all 7 calls and parents participated in 4 calls each. Families who did not have phones were provided cell phones to complete the calls. Using a cued-recall strategy (McHale, Crouter, & Bartko, 1992), adolescents reported on their involvement in 86 daily activities, including how long each event lasted, who else participated, and the ethnicity of all participants (i.e., Mexican, non-Mexican, both Mexican and non-Mexican). From these data, we calculated adolescents’ time spent in activities with peers in Mexican, non-Mexican and ethnically heterogeneous peer groups, and with parents. Pilot work was conducted to ensure that the range of activities was representative for this population. Families were paid a $100 honorarium for their participation in the home interviews and an additional $100 for participating in phone interviews.

Measures

All measures were forward and back translated into Spanish (for Mexican dialect in the local area) by two independent translators following the procedures outlined by Foster and Martinez (1995). All final translations were reviewed and discrepancies were resolved.

Background factors and cultural characteristics

Mothers and fathers reported their annual incomes. Parents reported their places of birth and all four family members completed the Acculturation Rating Scale for Mexican Americans – II (ARSMA – II; Cuéllar, Arnold, & Maldonado, 1995). On this 30-item scale, respondents use a 5-point rating scale to show how often (1 = not at all to 5 = extremely often or always) each experience applied to themselves during the past year. We adapted eight items to increase comprehensibility for adolescents (e.g., “My thinking is done in the English language” was adapted to “I think in English”). The ARSMA – II includes Mexican and Anglo orientation subscales. In this sample, Cronbach’s αs ranged from .70 to .90 for family members’ reports in English and Spanish.

Neighborhood ethnicity

We used geographic information systems (GIS) to link families’ addresses to census data defining neighborhoods by police grids (½ mile × ½ mile areas, which coincided with census block groups in almost all cases). We then calculated the percentage of Latinos residing in families’ neighborhoods using census data (range = 0 – 88.9%; M = 32.3%; SD = 21.4%). Families resided in 142 different census block groups, with the majority of block groups (65%) including only one family in the study.

Time spent with peers and parents

The time adolescents spent with Mexican peers, non-Mexican peers, and in heterogeneous groups of peers, as well as their time with parents, were measured in hours/7 days. Specifically, during each phone call, adolescents reported on the durations of and their companions in (e.g., sibling, parents, peers), their daily activities during nonschool hours as well as the ethnic composition of the social group involved in each activity (i.e., Mexican, non-Mexican, or both Mexican and non-Mexican individuals). From these data, we calculated adolescents’ unsupervised time with Mexican peers, with non-Mexican peers, and in heterogeneous peer groups (i.e., the time adolescents spent with peers outside the presence of adults). We focused on unsupervised time with peers because the majority of the time youth spend with peers does not include adults (Osgood et al., 1996) and because this time reflects socialization opportunities directed by youth rather than by adults. Peers were defined as both unrelated and related youth (e.g., cousins) but not siblings. We also calculated the total amount of time adolescents spent with parents. The correlation between older and younger adolescents’ reports of their shared time, r = .90, p<.001, provides strong evidence for the reliability of adolescents’ time use estimates. In the analyses, square root transformations were applied to all time measures to correct for skewness.

Risky behavior

Adolescents rated the frequency with which they engaged in each of 24 problematic or high-risk behaviors during the past year (e.g., skip a day of school, got drunk or high) on a 4-point scale (never to more than 10 times) using a measure developed by Eccles and Barber (1990) for ethnically diverse youth. Higher scores indicate more risky behaviors. αs ranged from .85 to .92 for older and younger adolescents in English and Spanish.

Peer delinquency

Adolescents reported on their close friends’ involvement in delinquent activities using a 5-item scale for multiethnic samples (Barrera et al., 2001; Mason, Cauce, Gonzales, & Hiraga, 1995) from the Denver Youth Survey (Huizinga, Esbensen, & Weiher, 1991) and National Youth Survey (Elliot & Ageton, 1980). Sample items tap how many close friends “… used force (e.g., threats or fighting) to get things from people” or “… have gotten drunk or high” Responses were rated on a 5-point scale (none to almost all). αs were above .74 for all adolescents except Spanish-speaking younger adolescents (α = .58). This low α is likely due to the small number of items and the small sample of Spanish-speaking younger adolescents (n = 39). The reliability of this scale has been established with Spanish-speaking adolescents in other work (Barrera et al., 2001).

Grade point average (GPA)

Adolescents’ GPAs were indexed by their most recent school report card grades in four subjects: English, Social Studies, Math, and Science. Grades were converted to a 5-point scale (A = 4.0; B = 3.0; C = 2.0; D = 1.0; F = 0) and averaged. School report card grades were available for 228 younger siblings and 188 older siblings.

Parental acceptance

Older and younger adolescents completed the short form of the Children’s Report of Parental Behavior Inventory (Schwarz, Barton-Henry, & Pruzinsky, 1985) to describe their relationships with their mothers and fathers (at separate points in the interview). Each of eight items was rated on a 5-point scale ranging from almost never to almost always. Higher scores on this summed scale reflected greater warmth and acceptance. This scale has been shown to be reliable and valid with Latino populations in English and Spanish (Knight, Virdin, & Roosa, 1992). Cronbach’s αs ranged from .83 to .93 for adolescents’ reports of their relationships with their mothers and fathers in English and Spanish.

Results

The results are organized around our study goals: (a) to describe the nature and correlates of younger and older adolescents’ time spent with peers and parents; (b) to test the hypothesis that mismatches between parents’ cultural orientations (i.e., Mexican/Anglo) and adolescents’ time spent with peers (i.e., Mexican vs. non-Mexican) would be linked to adolescent adjustment; and (c) to test the hypothesis that matches between parents’ cultural orientations and adolescents’ time with peers would be linked to adolescents’ cultural orientations and to explore whether parents’ Mexican orientations, coupled with time spent with parents, were linked to adolescents’ Mexican orientations. Sample size varied for the analyses because complete phone data were missing for 12 older adolescents and 7 younger adolescents. In a preliminary analysis, bivariate correlations between background and cultural characteristics were calculated (see Table 2).

Table 2.

Correlations Between Family Background and Cultural Characteristics^a

		Birthplace		Mexican orientation		Anglo orientation
	Family income	Mother	Father	Mother	Father	Mother	Father	Neighborhood ethnicity
Family Income	—
Birthplaceb
Mother	−.36**	—
Father	−.28**	.71**	—
Mexican orientation
Mother	−.22**	.65**	.62**	—
Father	−.34**	.67**	.63**	.62**	—
Anglo orientation
Mother	.46**	−.72**	−.64**	−.50**	−.59**	—
Father	.47**	−.62**	−.58**	−.48**	−.56**	.68**	—
Neighborhood ethnicity	−.39**	.26**	.26**	.18**	.30**	−.36**	−.34**	—

^a**Note. p<.01.

^bU.S. born=1; Mexican born=2.

Goal 1: Describing the Nature and Correlates of Adolescents’ Social Ecologies

To address our first goal, we conducted a series of mixed-model ANCOVAs. First, we examined adolescents’ time spent with Mexican, non-Mexican, and ethnically heterogeneous groups of peers. Next, we examined adolescents’ time with parents. Because adolescents’ time spent in different social contexts may vary as a function of the ethnic composition of their residential area and family financial resources, we treated neighborhood ethnicity (i.e., the percentage of Latinos residing in the neighborhood) and family income as covariates. We calculated Cohen’s d (Cohen, 1988) as a measure of effect size for all analyses; adjusted ds were computed for within-group analyses (Cortina & Nouri, 2000). Next, we explored the family background and cultural correlates of adolescents’ time spent in these social contexts.

Time spent with peers

First, we conducted a 2 (older adolescent sex) × 2 (younger adolescent sex) × 2 (sibling: older adolescent vs. younger adolescent) × 3 (peer time: Mexican, non-Mexican, and heterogeneous) mixed-model ANCOVA with Sibling and Peer Time as the within-groups factors and neighborhood ethnicity and family income as covariates. The dependent variable was adolescents’ peer time in hours (across the 7 calls) with Mexican peers, non-Mexican peers, and in heterogeneous peer groups. Untransformed means are reported in Table 3 for descriptive purposes. A Peer Time effect, F(2,227) = 16.10, p<.0001, in combination with follow-up tests (i.e., paired t tests with Bonferroni corrections), showed that adolescents spent the most time with Mexican peers, followed by non-Mexican peers, and the least amount of time in heterogeneous peer groups (see Table 3). Specifically, there were significant differences between Mexican and non-Mexican peer time, d = .80, and Mexican versus heterogeneous peer time, d = 1.13. This main was qualified, however, by interactions between Peer Time and Neighborhood Ethnicity, F(2, 227) = 14.41, p<.001, and Peer Time and Family Income, F(2, 227) = 9.39, p<.001. To follow-up, we used median splits to create high- and low-income groups and high and low neighborhood ethnicity groups: Adolescents spent more time with non-Mexican peers and in ethnically heterogeneous groups when families had higher (as compared with lower) incomes; ds ranged from .36 to .75. Adolescents spent less time with non-Mexican peers when neighborhoods had a higher versus lower percentage of Latinos, d = .91.

Table 3.

Untransformed Means and Standard Deviations for Older and Younger Adolescents’ (Nonschool) Time Spent (in hours/7 days) With Mexican Peers, Non-Mexican Peers, in Ethnically Heterogeneous Peer Groups, and With Parents as a Function of Family Income and Neighborhood Ethnicity

			Neighborhood ethnicitya				Incomeb
	Overall		Low % Latinos		High % Latinos		Low income		High income
Social context	M	SD	M	SD	M	SD	M	SD	M	SD
Mexican peers
Older adolescents	4.24	5.33	2.99	3.95	5.55	6.24	5.54	5.87	2.98	4.44
Younger adolescents	2.94	3.40	2.33	3.23	3.56	3.47	3.76	3.84	2.14	2.68
Non-Mexican peers
Older adolescents	1.74	3.95	2.79a	5.02	0.63b	1.80	0.65c	2.32	2.79d	4.82
Younger adolescents	1.03	2.36	1.62a	2.96	0.44b	1.28	0.50c	1.52	1.56d	2.87
Heterogeneous groups
Older adolescents	1.37	3.23	1.60	3.30	1.13	3.15	0.75c	1.57	1.97d	4.18
Younger adolescents	0.84	1.85	0.93	2.04	0.76	1.65	0.72c	1.62	0.97d	2.06
Time with parents
Older adolescents	8.82	6.76	9.72a	7.13	7.86b	6.24	8.50	7.08	9.12	6.45
Younger adolescents	10.39	7.03	11.30a	6.87	9.46b	7.09	10.31	7.63	10.46	6.40

^abNote. Means with different superscripts differ significantly by neighborhood ethnicity group.

^cdMeans with different superscripts differ significantly by income group.

Time spent with parents

We next conducted a 2 (older adolescent sex) × 2 (younger adolescent sex) × 2 (sibling: older adolescent vs. younger adolescent) ANCOVA with neighborhood ethnicity and family income as covariates, with time with parents as the dependent variable (see Table 3 for means). This analysis revealed a Sibling effect, F(1, 228) = 6.92, p<.01, d = 1.87, indicating that younger siblings spent more time with parents than did older siblings. There was also a significant Younger Sibling Sex effect, F(1, 228) = 6.04, p<.05, d = .29, revealing that parents spent less time with their offspring when they had younger sons versus daughters (M = 10.54, SD = 6.54 for younger daughter families; M = 8.64, SD = 6.38, for younger son families). There was a similar trend for older siblings’ sex, F(1, 228) = 3.24, p<.08. Finally, a significant neighborhood ethnic composition effect, F(1, 228) = 6.63, p<.05, d = .29, revealed that adolescents spent more time with parents when there was a lower percentage of Latinos in the neighborhood (M = 10.51; SD = 6.62) as compared with a higher percentage of Latinos (M = 8.70; SD = 6.31)

Correlations Between Adolescents’ Time Use and Background and Cultural Characteristics

We next conducted correlations between both family background/cultural characteristics and parents’ cultural orientations and adolescents’ time spent in different social contexts (see Table 4). Beginning with the links between family background characteristics and adolescents’ peer time, family income was negatively associated with time spent with Mexican peers and positively correlated with time spent in non-Mexican and heterogeneous peer groups, as was predicted. For older siblings, age was positively associated with all measures of peer time.

Table 4.

Correlations Between Adolescents’ Time Spent With Mexican Peers, non-Mexican Peers, and in Heterogeneous Peer Groups and Parents’ Background and Cultural Characteristics

	Younger adolescents			Older adolescents
	Mexican peers	Non-Mexican peers	Heterogeneous groups	Mexican peers	Non-Mexican peers	Heterogeneous groups
Family income	−.20**	.28**	.09	−.22***	.32***	.18**
Adolescent age	.03	.03	.00	.16**	.18**	.14*
Birth placea
Mother	.19**	−.42***	−.21***	.21***	−.39***	−.24***
Father	.21***	−.37***	−.14*	.25***	−.42***	−.14*
Mexican orientation
Mother	.18**	−.31***	−.18**	.21***	−.32***	−.17**
Father	.23***	−.44***	−.13*	.24***	−.40***	−.22***
Anglo orientation
Mother	−.27***	.41***	.17**	−.35***	.41***	.16**
Father	−.25***	.41***	.10	−.29***	.36***	.21***
Neighborhood ethnicityb	.23***	−.30***	.02	.24***	−.34***	−.09

^aNote. U.S. born=1; Mexico born=2.

^bNeighborhood ethnicity scored as the percentage of Latinos living in the neighborhood (range = 0–88%).

^*P<.05

^**P<.01

^***P<.001.

Cultural characteristics also were linked to adolescents’ time spent in different peer groups, and similar patterns emerged for older and younger adolescents. In general, when parents were Mexican born and when they reported stronger Mexican orientations, adolescents spent more time with Mexican peers and less time in non-Mexican and heterogeneous peer groups. In contrast, stronger orientations toward Anglo culture were linked to less time with Mexican peers and more time in non-Mexican and heterogeneous peer groups. In addition, when there was a higher percentage of Latinos in the areas where families lived, adolescents spent more time with Mexican peers and less time with non-Mexican peers. As an additional step, we conducted partial correlations between parents’ birthplace and cultural orientations and adolescents’ time use, controlling for family income and neighborhood ethnicity. The pattern of correlations remained the same.

For the correlations between background and cultural characteristics and adolescents’ time with parents, two significant associations emerged: a negative correlation between older adolescents’ age and time spent with parents, r = −.34, p<.01, and a negative correlation between neighborhood ethnic composition and younger adolescents’ time with parents, r = −.16, p<.05.

Summary Goal 1

All but one of our hypotheses were supported. Adolescents’ age was positively associated with time spent with peers (in all ethnic contexts), parent enculturation was associated with more time with Mexican peers, and parent acculturation was associated with more time with non-Mexican peers and in ethnically heterogeneous peer groups. In addition, when adolescents resided in neighborhoods with high percentages of Latinos, they spent more time with Mexican peers and when families had more economic resources, adolescents spent more time in non-Mexican and ethnically mixed peer contexts. Inconsistent with our hypothesis, stronger ties to Mexican culture were not linked to adolescents’ time with parents.

Goal 2: Parents’ Cultural Orientations, Adolescents’ Peer Time, and Adolescent Adjustment

We tested person – environment fit hypotheses regarding the joint role of parents’ cultural orientations and adolescents’ peer time in adolescents’ psychosocial functioning and parental acceptance using multilevel modeling (MLM), an approach that extends multiple regression to incorporate nested data, in our case, siblings within families (Bryk & Raudenbush, 1987). Our two-level model partitioned variance into: (a) between-sibling (or within-family) and (b) between-family components. At Level 1, the between-sibling model, independent variables included birth order, adolescents’ sex, age, and time spent in particular peer contexts. At Level 2, the between-family model, control variables (neighborhood ethnicity and family income), and mothers’ and fathers’ cultural orientations were included. In addition, these models included two cross-level interactions: Mothers’ cultural orientation × adolescents’ peer time and fathers’ cultural orientation × adolescents’ peer time. Continuous variables were centered before creating interaction terms to reduce multicollinearity, and dichotomous variables were effect coded (.5 vs. −.5). The equations for the complete model are presented in the appendix (see Model A). Because our hypotheses focused on mismatches between parents’ cultural orientations and adolescents’ peer time, we conducted two sets of analyses: The first focused on parents’ Anglo orientations and adolescents’ time in Mexican peer contexts, and the second focused on parents’ Mexican orientations and adolescents’ time in non-Mexican peer contexts. Dependent variables included adolescents’ risky behavior, peer delinquency, GPA, and parental acceptance. Follow-up analyses for interactions were conducted according to guidelines by Aiken and West (1991). Because significant effects emerged only in the first set of models, we focus on these. The findings for this goal are displayed in Table 5.

Table 5.

Hierarchical Linear Modeling Results Predicting Dependent Variables from Background Variables, Parents’ Anglo Orientations, and Adolescents’ Time in Mexican Peer/Kin Contexts

	Adolescents’ risky behaviors		Peer delinquency		Grade point average		Maternal acceptance		Paternal acceptance
Model	γ	SE	γ	SE	γ	SE	γ	SE	γ	SE
Intercept	1.42***	02	1.87***	.03	2.62***	.05	3.97***	.04	3.71***	.05
Neighborhood ethnicity	0.001	.001	0.00	.002	0.002	.002	−0.002	.002	−0.003	.002
Log income	−0.05	.03	−0.06	.06	0.36***	.09	0.11	.07	0.13	.08
Birth order	0.03	.05	−0.12	.11	0.11	.12	0.03	.10	0.02	.12
Gender	0.13***	.03	0.14*	.07	−0.34***	.08	−0.03	.07	0.08	.08
Age	0.05**	.02	0.09**	.03	0.004	.03	0.01	.03	−0.07*	.03
Mothers’ Anglo orientations	0.09**	.03	0.12*	.05	0.13	.08	0.06	.06	−0.15*	.07
Fathers’ Anglo orientations	0.00	.03	−0.02	.05	−0.18*	.08	−0.11	.06	0.06	.07
Time with Mexican peers	0.01***	.002	0.02***	.004	−0.01**	.005	−0.01*	.004	−0.01*	.005
Time with Mexican peers × mothers’ Anglo orientations	0.00	.002	−0.006	.005	0.01	01	0.007	.005	0.02**	.006
Time with Mexican peers × fathers’ Anglo orientations	0.001	.002	0.01*	.005	−0.02*	.01	−0.006	.005	−0.01*	.007

^*Note. p<.05

^**p<.01

^***p<.001.

Before testing these models, we examined the correlations between the outcome variables for older and younger siblings. Adolescents’ GPAs were negatively related to risky behavior and peer delinquency (r = −.13, p<.05 to r = −.29, p<.01) and positively related to parental acceptance (r = .15, p<.05 to r = .22, p<.01). Parental acceptance was negatively related to risky behavior and peer delinquency (r = −.20, p<.01 to r = −.31, p<.01). Risky behavior and peer delinquency were positively correlated for both older and younger siblings (r = .60, p<.01 and r = .63, p<.01, respectively).

Adolescents’ risky behaviors

There were three Level 1 main effects in this model: A main effect for sex indicated that boys reported higher levels of risky behaviors than did girls, a main effect for age revealed that older adolescents reported more risky behaviors than did younger adolescents, and an effect for unsupervised time with Mexican peers suggested that more time in this context was linked to higher levels of risky behavior. At Level 2, mothers’ Anglo orientations were also positively associated with adolescents’ risky behaviors, but the predicted interaction was not significant.

Peer delinquency

This model also revealed three Level 1 main effects. Boys’ peers were more involved in delinquent activities than girls’peers; older adolescents’ friends were more involved in delinquent activities than younger adolescents’friends; and time with Mexican peers was positively associated with peer delinquency. At Level 2, mothers’ Anglo orientations were also positively associated with peer delinquency, and the predicted cross-level interaction between fathers’ Anglo orientations and adolescents’ time with Mexican peers was also significant: Consistent with our mismatch hypothesis, youth who spent more time with Mexican peers but had fathers who were more Anglo-oriented reported the highest levels of peer delinquency (M = 2.08). In contrast, youth who spent less time with Mexican peers and who had more Anglo-oriented fathers reported the lowest levels of peer delinquency (M = 1.61).

GPA

There were Level 1 main effects for adolescents’ sex and time spent with Mexican peers: girls had higher GPAs, and time with Mexican peers was associated with lower GPAs. At Level 2, family income was positively associated with GPA. The predicted cross-level interaction between fathers’ Anglo orientation and adolescents’ time with Mexican peers was also significant: When fathers were more acculturated but adolescents spent more time with Mexican peers, adolescents’ GPAs were low (M = 2.17) compared with those of other youth (Ms ranged from 2.74 to 2.91).

Parental acceptance

Time with Mexican peers was associated with lower levels of maternal and paternal acceptance; for the model predicting fathers’ acceptance, there was also a main positive association with adolescent age. We also found a negative association between mothers’ Anglo orientation and paternal acceptance. Finally, there were two cross-level interactions: fathers’ Anglo orientation × adolescents’ time with Mexican peers, and Mothers’ Anglo orientation × adolescents’ time with Mexican peers. Consistent with a fit hypothesis, paternal acceptance was lowest when fathers’ Anglo orientation was high and adolescents’ time with Mexican peers was high (M = 3.55) and paternal acceptance was highest when fathers’ Anglo orientation was high and adolescents’ time with Mexican peers was low (M = 3.98). For the interaction between mothers’ Anglo orientation and adolescents’ time with Mexican peers, however, the pattern was inconsistent with our hypothesis: paternal acceptance was lowest when mothers’ Anglo orientation was low and adolescents’ time with Mexican peers was high (M = 3.63) and paternal acceptance was highest when mothers’ Anglo orientation and adolescents’ time with Mexican peers were low (M = 4.09).

Summary Goal 2

The mismatch between fathers’ Anglo orientations and adolescents’ time with Mexican peers was linked to adolescent adjustment, with interactions emerging for three of the four dependent variables. We did not find evidence that incongruence between parents’ Mexican orientations and adolescents’ time with Anglo peers was linked to adolescent adjustment.

Goal 3: Parents’ Cultural Orientations, Adolescents’ Time Use, and Adolescents’ Cultural Orientations

Our final goal was to investigate how parents’ cultural orientations, in combination with adolescents’ time use, were linked to adolescents’ cultural orientations. The first set of models tested the hypothesis that congruence between parents’ cultural orientations and adolescents’ peer time would be associated with stronger orientations toward Mexican or Anglo culture. For the second set of models, we predicted that parents’ Mexican orientation, in combination with more parent – adolescent shared time, would be associated with stronger Mexican cultural orientations in youth. Again, we used an MLM approach, partitioning variance into: (a) between-sibling (or within-family) and (b) between-family components (see appendix Model B for equations). At Level 1, the between-sibling model, explanatory variables included birth order, adolescents’ sex, age, and time spent with Mexican or non-Mexican peers. At Level 2, the between-family model, control variables (the percentage of Latinos in each Census tract and family income) and mothers’ and fathers’ cultural orientations were included. Finally, two cross-level interaction terms were included (i.e., mothers’ cultural orientation × adolescents’ peer time and fathers’ cultural orientation × adolescents’ peer time). Continuous variables were centered before creating interaction terms, and dichotomous variables were effect coded (.5 vs. −.5). In the second set of models, we replaced adolescents’ peer time with time spent with parents.

Adolescents’ enculturation

The first model examined parents’ Mexican orientations and adolescents’ time with Mexican peers as predictors of adolescents’ enculturation in Mexican culture (see Table 6). A Level 1 effect showed that girls were more Mexican-oriented than boys, and at Level 2, family income was negatively associated with adolescents’ enculturation, and mothers’ and fathers’ enculuration scores were positively related to adolescents’ enculturation.

Table 6.

Hierarchical Linear Modeling Results Predicting Adolescents’ Mexican Orientations from Background Variables, Parents’ Mexican Orientations, and Adolescents’ Time in Mexican Peer/Kin Contexts

Model	γ	SE
Intercept	3.66***	.03
Neighborhood ethnicity	0.00	.00
Log income	−0.22***	.05
Birth order	−0.09	.07
Gender	−0.18***	.05
Age	−0.02	.02
Mothers’ Mexican orientations	0.34***	.06
Fathers’ Mexican orientations	0.33***	.05
Time with Mexican peers	0.005	.003
Time with Mexican peers × mothers’ Mexican orientations	0.003	.005
Time with Mexican peers × fathers’ Mexican orientations	0.001	.005

^***Note. p<.001.

The model testing parents’ enculturation and adolescents’ time with parents as predictors of adolescents’ enculturation revealed all the main effects described above (see Table 7): Girls were more enculturated than boys, family income was negatively associated with adolescents’ enculturation, and mothers’ and fathers’ enculturation scores were positively related to adolescents’ enculturation. In addition, time spent with parents was positively associated with adolescents’ Mexican orientations. Again, the interactions were not significant.

Table 7.

Hierarchical Linear Modeling Results Predicting Adolescents’ Mexican Orientations from Background Variables, Parents’ Mexican Orientations, and Adolescents’ Time With Parents

Model	γ	SE
Intercept	3.66***	.03
Neighborhood ethnicity	0.00	.00
Log income	−0.23***	.05
Birth order	−0.08	.07
Gender	−0.18***	.05
Age	−0.01	.02
Mothers’ Mexican orientations	0.35***	.06
Fathers’ Mexican orientations	0.33***	.05
Time with parents	0.012*	.005
Time with parents × mothers’ Mexican orientations	0.002	.001
Time with parents × fathers’ Mexican orientations	0.001	.001

^*Note. p<.05

^***p<.001.

Adolescents’ acculturation

The first model, which included parents’ acculturation and adolescents’ time with non-Mexican peers, revealed that adolescents’ age was negatively and time with non-Mexican peers was positively associated with adolescents’ acculturation (see Table 8). At Level 2, mothers’ acculturation was positively associated with adolescents’ acculturation. Finally, there was evidence for the predicted interaction between mothers’ acculturation and adolescents’ time with non-Mexican peers. Follow-ups revealed that adolescents were most acculturated when they spent more time with non-Mexican peers and their mothers were more acculturated (M = 4.17); in contrast, adolescents were least acculturated when mothers’ acculturation and adolescents’ time with non-Mexican peers were both low (M = 3.66).

Table 8.

Hierarchical Linear Modeling Results Predicting Adolescents’ Anglo Orientations from Background Variables, Parents’ Anglo Orientations, and Adolescents’ Time With Non-Mexican Peers/Kin

Model	γ	SE
Intercept	4.00***	.03
Neighborhood ethnicity	0.00	.00
Log income	0.17***	.06
Birth order	−0.08	.07
Gender	0.03	.05
Age	−0.05*	.02
Mothers’ Anglo orientations	0.10*	.05
Fathers’ Anglo orientations	0.09	.05
Time with non-Mexican peers/kin	0.02***	.005
Time with non-Mexican peers/kin × mothers’ Mexican orientations	−0.01*	.005
Time with non-Mexican peers/kin × fathers’ Mexican orientations	0.00	.01

^*Note. p<.05

^***p<.001.

Summary Goal 3

Findings revealed that congruence between mothers’ acculturation and adolescents’ time with non-Mexican peers was associated with adolescents’ acculturation. All other patterns revealed the importance of parents and peers for adolescents’ acculturation, but did not support our expectations about incongruence.

Additional Analyses

To better understand findings on links between adolescents’ time with Mexican peers and their adjustment problems (i.e., r = .30, p<.01 for both adolescents’ involvement in risky behaviors, r = .20, p<01 for both adolescents’ peer delinquency, r = −.28, p<.01 for older siblings’ GPAs and r = −.19, p<.01 for younger siblings’ GPAs, and lower levels of parent acceptance, r = −.22, p<.01 for younger adolescent maternal acceptance, r = −.16, p<.05 for older adolescent paternal acceptance), we recoded the time use data to create an additional variable that indexed adolescents’ supervised time with Mexican peers (i.e., one or more adults were present). Correlations between adolescents’ supervised time with Mexican peers and adolescent psychosocial functioning were not significant; it appears that only unsupervised time with Mexican peers is linked to adolescent risk (e.g., Osgood et al., 1996). Analyses focused on total unsupervised and supervised time with peers revealed a similar pattern: Total unsupervised time with peers (Mexican plus non-Mexican) was linked to higher levels of risky behavior and delinquent peer affiliations and less parental acceptance. The correlation for adolescents’ GPAs was in the expected direction, but not significant. Total supervised time with peers (regardless of ethnic context), in contrast, was not associated with adjustment.

Discussion

This study drew on cultural – ecological and person – environment fit perspectives to explore the nature and correlates of Mexican-American adolescents’ everyday activities with peers and parents. In doing so, we used an ethnic-homogeneous design to examine cultural and ecological factors that were linked to within-group variability in well-being among Mexican-American youth. Our multimethod, multi-informant approach alleviates potential concerns about shared method variance and reporter biases, thereby increasing our confidence in the study’s findings. Furthermore, our findings contribute to emerging literature on normative developmental and family processes in minority youth.

Describing the Social Contexts of Mexican-American Adolescents’ Daily Activities

A primary contribution of this study was its descriptive information on Mexican-American adolescents’ time spent with parents and in different peer groups. In summarizing our findings, we address three themes and their implications for future work on Mexican-American youth’s time use: (a) the diversity of adolescents’ peer involvement; (b) the role of family, cultural, and contextual factors in adolescents’ time use; and (c) developmental and gendered patterns of time use. Importantly, our results are derived from a specific sample of Mexican-American families (two-parent families, mostly from immigrant backgrounds, in the Southwest). A direction for future work is to study Mexican-American families with different immigration histories, from different geographic locations, and from a wider range of family structures.

Demographic trends highlighting the rapid growth in minority populations in combination with declines in majority populations in the United States (U.S. Census Bureau, 2000) suggest that consideration of the time that both minority and majority adolescents spend in ethnically homogeneous and heterogeneous peer contexts is important information in its own right. In this sample, older and younger adolescents spent twice as much unsupervised time outside school with Mexican peers (i.e., 3 – 4 hr per 7 days) than with non-Mexican peers or in ethnically heterogeneous groups. Adolescents’ time in these different peer groups varied across social contexts, however. Consistent with a cultural ecological emphasis on contextual affordances and constraints, we found that adolescents spent more time with non-Mexican peers and in ethnically heterogeneous groups when their families had more economic resources and more time with non-Mexican peers when their neighborhoods had lower percentages of Latinos.

A second theme that emerged was the important role of parents’ cultural backgrounds and orientations in adolescents’ time with peers. When parents had stronger ties to Mexican culture, adolescents spent more time with Mexican peers and less time with non-Mexican peers and in ethnically heterogeneous peer groups, but when parents reported stronger orientations toward Anglo culture, adolescents spent less time with Mexican peers and more time in non-Mexican and ethnically heterogeneous peer groups. These patterns suggest that parents play an important role in shaping adolescents’ involvement with peers, and future work should be directed at the mechanisms underlying the associations between parents’ cultural orientations and adolescents’ time with peers. One possibility is that parents’ cultural orientations influence those of adolescents and in turn, adolescents’ time in different ethnic peer contexts. Parents may also actively guide adolescents’ peer experiences through supervision, monitoring, and support (Parke & Buriel, 1998). Future efforts should also focus on how parents’ biculturalism may be linked to youth’s peer involvement.

In contrast to the clear pattern of correlates of adolescents’ time spent with peers, adolescents’ time with parents was not associated with the family, cultural, or contextual factors that we examined. We found no support for our hypothesis that parents’ ties to Mexican culture would be linked to adolescents’ time with parents. We used global measures of parents’ enculturation, however, and future studies should examine how specific cultural values (e.g., familism values) are linked to the time adolescents spend with their parents.

A third theme from our descriptive data pertained to gender and developmental differences. We found some support for our prediction that boys would spend less time with parents but not for our prediction that boys would spend more time with peers. Given the gender differences in adolescent adjustment found in this and other studies, more research on gendered parenting processes is warranted. As predicted, we also found positive associations between older adolescents’ age and time spent with peers in all ethnic contexts, and negative associations between older adolescents’ age and time with parents. Older adolescents also reported more risky behavior and delinquent peer affiliations than younger adolescents. Exploring the world beyond the family is an important part of adolescent development but one that brings challenges.

Adolescents’ Peer Involvement, Parents’ Cultural Orientations, and Adolescent Psychosocial Adjustment

In studying the implications of parents’ acculturation and youth time use as “main effects”, our findings replicated results from prior work. For instance, consistent with data suggesting that acculturation places youth at risk for adjustment problems (Gonzales et al., 2002), we found links between mothers’ acculturation and youth’s risky behavior, delinquent peer affiliations, and paternal acceptance. In accord with prior time use research (e.g., Osgood et al., 1996), we also found that unsupervised time with Mexican peers was associated with risky behavior, involvement with delinquent peers, lower GPA, and lower parental acceptance; importantly, the same pattern of findings emerged when we examined youth’s total unsupervised peer time. Our findings showing no links between adjustment problems and time with Mexican peers in the company of adults add to a body of work on the importance of monitoring and supervision during this developmental period (Parke & Buriel, 1998).

Our primary interest was in the combined effects of parents’ cultural orientations and youth’s time with peers. Consistent with a person – environment fit perspective, we found some evidence of the negative implications of a mismatch between fathers’ Anglo orientations and adolescents’ time with Mexican peers. The extant literature on intergenerational acculturation gaps has focused on discrepancies that arise when youth acculturate at a faster rate than their parents (Rumbaut, 2005; Szapocznik & Kurtines, 1993; Vega et al., 1995), and little is known about mismatches that result when parents but not youth have strong ties to Anglo culture. Our findings highlighted the role of fathers but not mothers in such dynamics. Fathers who have strong ties to Anglo culture are also likely to have more financial resources and may spend more time themselves in Anglo and culturally diverse settings. In these families, youth involvement with Mexican peers may occur for different reasons than in less acculturated families where youth may have fewer opportunities to interact with peers from other ethnic groups. One possibility is that youth involvement with Mexican peers in these highly acculturated families represents problematic involvement with peer groups (e.g., gang affiliations) and disengagement from school and school-related activities than in less acculturated families. That this group of youth had the highest level of involvement with delinquent peers and lowest GPAs supports this potential explanation.

Another possibility is that when fathers are highly acculturated, they strive to create economic, social, and educational opportunities for their offspring. When offspring spend their time with Mexican peers, however, fathers may see adolescents as being unwilling or unable to take advantage of these opportunities: Youth may gravitate toward Mexican peer groups in the face of parents’ negative reactions and institutional barriers and discrimination may prevent youth from spending time with non-Mexican peers or heterogeneous groups. The resulting discord may lead to poor adjustment and low levels of acceptance in the father – adolescent relationship. In addition, families may have limited support for dealing with this type of mismatch to the extent that it is less common than intergenerational discrepancies resulting from youth acculturating more quickly than parents.

The link between mothers’ Anglo orientations, adolescents’ time with Mexican peers, and paternal acceptance did not support our hypotheses. When mothers reported low Anglo orientations and adolescents were less involved with Mexican peers, paternal acceptance was the highest. An important part of understanding the links between mothers’ cultural orientations and qualities of the father – adolescent relationship will be to explore matches between parents’ orientations and their implications for family dynamics and youth development. More generally, the overall pattern of results, highlighting more consistent connections between fathers’ (as compared with mothers’) cultural orientations and youth outcomes, is consistent with the notion that mother – and father – adolescent relationships operate in different ways (Parke & Buriel, 1998). The results are also consistent with ideas about the traditional family role of Mexican-American fathers (Cauce & Domenech-Rodriguez, 2002): To the extent that fathers’ more traditional roles are those of disciplinarians and figures of authority in Mexican-American families, incongruence between fathers’ cultural orientations and adolescents’ peer time may be more important for youth adjustment. More detailed information about the different dynamics of mother – adolescent and father – adolescent relationships in Mexican-American families in general and in Mexican-American families that differ in the traditionality of their parenting roles will be important for understanding these patterns. In addition, because the role of fathers’ cultural orientations in adolescents’ psychosocial functioning has been largely ignored (Gonzales et al., 2002), future studies should examine mothers’ and fathers’ orientations, alone and in combination.

We did not find evidence that the mismatch between parents’ Mexican orientations and adolescents’ time with non-Mexican peers was associated with adolescents’ psychosocial functioning. Given the background characteristics of this sample (i.e., parents were predominantly born in Mexico and spoke Spanish) and the limited involvement of these adolescents with non-Mexican peers, it will be important to test these ideas in more acculturated families. The significance of different types of mismatches between parents and youth may depend on the broader cultural and economic context in which families’ lives are embedded and the opportunities these contexts afford youth. More generally, our findings are consistent with the conclusion of García Coll and Pachter (2002) that, because of minority families’ unique circumstances, developmental processes and outcomes need to be “evaluated vis-à-vis the goodness of fit between parenting practices and the context in which it occurs” (p. 12).

Parents’ Cultural Orientations, Adolescents’ Time With Peers and Parents, and Adolescents’ Cultural Orientations

Our final goal was to extend understanding of the role of normative socialization experiences in the development of Mexican-American adolescents’ cultural orientations. We hypothesized that congruence across social contexts would promote adolescents’ orientation toward Mexican or Anglo culture, and found one significant effect supporting this hypothesis: When mothers were low on Anglo orientation and adolescents spent less time with non-Mexican peers, adolescents reported the lowest levels of acculturation. The remainder of our findings highlighted the main effects linking parents’ Mexican orientations to adolescents’ Mexican cultural orientations, and linking parents’ Anglo cultural orientations and time with non-Mexican peers to adolescents’ Anglo orientations. In addition, time with parents was positively related to adolescents’ Mexican cultural orientations. This pattern of results highlights the role of parents in adolescents’ Mexican orientations and the roles of both parents and peers in adolescents’ Anglo orientations.

A body of work has examined the family’s role in Latino youth’s cultural orientations (e.g., Bernal & Knight, 1993), but we know less about the role of peers. In this sample, Mexican-American adolescents spent considerable time with peers, and time with peers was linked to adolescents’ cultural orientations. These findings are consistent with other work that documents the significance of friendships for Latino youth (e.g., Phinney, Romero, Nava, & Huang, 2001; Stanton-Salazar & Spina, 2005; Suárez-Orozco & Suárez-Orozco, 1995; Way et al., 2001), and suggest that examinations of peer influences on acculturation and enculturation merit future study. Because our data are cross-sectional, we were unable to draw conclusions about direction of effect: Adolescents with stronger orientations toward Mexican culture may choose to spend time in the company of Mexican peers or the other way around. Thus, longitudinal studies of Mexican-American youth’s peer experiences are an important direction for future study.

Conclusion

Responding to the call of scholars interested in minority youth and families for investigations of normative socialization processes (García Coll et al., 1996; McLoyd, 1998), this study took an important first step in describing the nature and correlates of Mexican-American adolescents’ time with peers and parents. How children and adolescents spend their time reflects normative developmental processes, and daily activities are an important influence on youth development (Bronfenbrenner, 1979). The findings from this study also show that, for Mexican-American adolescents, daily activities are a concrete manifestation of what it means to grow up in multiple worlds.

Appendix

Model A

Level 1

$$\begin{array}{ccc}y\hfill & =\hfill & {\mathrm{\beta }}_0+{\mathrm{\beta }}_1(\text{birth order})+{\mathrm{\beta }}_2(\text{gender})\hfill \\ \hfill & \hfill & +{\mathrm{\beta }}_3(\text{time with Mexican peers}/\text{child kin})+{\mathrm{\beta }}_4(\text{age})+R.\hfill \end{array}$$

Level 2

$$\begin{array}{ccc}{\mathrm{\beta }}_0\hfill & =\hfill & {\mathrm{\gamma }}_{00}+{\mathrm{\gamma }}_{01}(\text{percent Latino})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{02}(\text{log of family income})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{03}(\text{mothers' Anglo orientation})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{04}(\text{fathers' Anglo orientation})+u_0,\hfill \end{array}$$

$$\begin{array}{ccc}{\mathrm{\beta }}_1\hfill & =\hfill & {\mathrm{\gamma }}_{10},\hfill \end{array}$$

$$\begin{array}{ccc}{\mathrm{\beta }}_2\hfill & =\hfill & {\mathrm{\gamma }}_{20},\hfill \end{array}$$

$$\begin{array}{ccc}{\mathrm{\beta }}_3\hfill & =\hfill & {\mathrm{\gamma }}_{30}+{\mathrm{\gamma }}_{31}(\text{mothers' Anglo orientation})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{32}(\text{fathers' Anglo orientation}).\hfill \end{array}$$

Model B

Level 1

$$\begin{array}{ccc}y\hfill & =\hfill & {\mathrm{\beta }}_0+{\mathrm{\beta }}_1(\text{birth order})+{\mathrm{\beta }}_2(\text{gender})\hfill \\ \hfill & \hfill & +{\mathrm{\beta }}_3(\text{time with Mexican}/\text{Non-Mexican peers}/\hfill \\ \hfill & \hfill & \text{child kin}/\text{Time with Parents})\hfill \\ \hfill & \hfill & +{\mathrm{\beta }}_4(\text{age})+R\hfill \end{array}$$

Level 2

$$\begin{array}{ccc}{\mathrm{\beta }}_0\hfill & =\hfill & {\mathrm{\gamma }}_{00}+{\mathrm{\gamma }}_{01}(\text{percent Latino})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{02}(\text{log of family income})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{03}(\text{mothers' Mexican}/\text{Anglo orientation})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{04}(\text{fathers' Mexican}/\text{Anglo orientation})+u_0,\hfill \end{array}$$

$$\begin{array}{ccc}{\mathrm{\beta }}_1\hfill & =\hfill & {\mathrm{\gamma }}_{10},\hfill \end{array}$$

$$\begin{array}{ccc}{\mathrm{\beta }}_2\hfill & =\hfill & {\mathrm{\gamma }}_{20},\hfill \end{array}$$

$$\begin{array}{ccc}{\mathrm{\beta }}_3\hfill & =\hfill & {\mathrm{\gamma }}_{30}+{\mathrm{\gamma }}_{31}(\text{mothers' Mexican}/\text{Anglo orientation})\hfill \\ \hfill & \hfill & +{\mathrm{\gamma }}_{32}(\text{fathers' Mexican}/\text{Anglo orientation}).\hfill \end{array}$$

Note. Birth order and gender were effect coded: first born = −.5, secondborn = .5; female = −.5, male = .5.