Parents' promotion of psychological autonomy, psychological control, and Mexican-American adolescents' adjustment.

Mexican-American adolescents are at an elevated risk for adjustment difficulties. In an effort to identify parenting practices that can affect the adjustment of Mexican-American youth, the current study examined parents' promotion of psychological autonomy and parents' psychological control as perceived by Mexican-American early adolescents, and explored their associations with adolescents' adjustment in the context of acculturation. In 5th grade, 134 (54.5% female) Mexican-American adolescents reported on their acculturation level and the parenting practices of their mothers and fathers. In 5th and 7th grade, adolescents also reported on their depressive symptoms, number of delinquent friends, and self-worth. Perceptions of promotion of psychological autonomy and of psychological control were positively correlated. However, perceptions of more promotion of psychological autonomy and of less psychological control predicted fewer depressive symptoms 2 years later. Perceptions of more promotion of psychological autonomy also predicted fewer delinquent friends two years later. Finally, perceptions of more promotion of psychological autonomy predicted higher self-worth only among less acculturated adolescents. The study underscores the roles that promotion of psychological autonomy and psychological control may play in Mexican-American children's well-being during early adolescence.

Introduction

Mexican–American youth face disproportionate risks for mental health problems (National Center for Chronic Disease Prevention and Health Promotion 2009). As this ethnic group represents one of the largest minority groups in the US (US Census Bureau 2007), identifying factors that relate to their adjustment is important to inform the planning of prevention and intervention programs. The current study aimed to address this issue by exploring two parenting practices, namely parents’ promotion of psychological autonomy and parents’ psychological control, and by examining the associations between these parenting practices and the well-being of Mexican–American early adolescents.

An important task for parents of adolescents in Western cultures is to enhance adolescents’ development of autonomy, i.e., to promote their self-governance and awareness of being separate, with independent thoughts, feelings, and behaviors (Allen et al. 1994; Grotevant and Cooper 1986; McLeod et al. 2007). The promotion of autonomy can take many forms, including adolescents’ privileges and responsibilities (Zimmer-Gembeck and Collins 2003). Among its distinct components, the promotion of psychological autonomy, namely the encouragement of adolescents’ self- exploration and self-assertion (Grotevant and Cooper 1998) has been examined widely (Boykin McElhaney and Allen 2001; Grotevant and Cooper 1998). In studies of European-American families, this aspect of parenting consistently predicts better social-emotional adaptation, as indexed by higher self-esteem, fewer internalizing and externalizing symptoms, and higher educational attainment of adolescents and young adults (Allen et al. 1994a, b; Best et al. 1997).

Another aspect of parenting that is considered central for adolescents’ development is parental control, which refers to demands and restrictions that parents put on their children to help them learn how to regulate their impulses and desires (Barber et al. 1994; Maccoby and Martin 1983). Parental control is conceived as a multifaceted construct with differential effects on adolescents’ well-being. For example, monitoring their adolescents is considered an adaptive component of parents’ control, while harsh punishment is considered maladaptive (Barber et al. 1994; Steinberg 1990). The psychological control component of parental control has received particular attention in the context of adolescents’ development. It refers to parents’ intrusion and manipulation of the psychological world of the child by pressuring the child to think, feel, and behave according to the parents’ desires (Barber and Harmon 2002; Assor et al. 2004). Psychological control is usually interpreted in the literature as inappropriate, especially in light of adolescents’ increased developmental need for autonomy. Since psychological control involves anxiety and guilt induction as well as conditional acceptance, it is thought to reduce adolescents’ sense of security and self competence, and thus restrict their social-emotional development (Barber 1996). Numerous studies on Western adolescents support this view, and indicate that this parenting practice is associated with more externalizing and internalizing behavior problems of adolescents (Albrecht et al. 2007; Barber et al. 1994; Hoeve et al. 2009; McLeod et al. 2007).

Psychological control and promotion of psychological autonomy might seem to be opposite ends of the same continuum, since the use of psychological control does “not permit the child to develop as an individual apart from the parent” (Schaefer 1965, p. 555). However, the absence of promotion of psychological autonomy does not necessarily mean that the parent is psychologically controlling. A parent who does not encourage adolescent’s self-exploration and self-assertion is not necessarily intrusive and manipulative. And vice versa, a parent who is not intrusive does not necessarily encourage the adolescent to explore and assert oneself (Silk et al. 2003). Indeed, previous studies indicated that the promotion of psychological autonomy and psychological control are separate constructs, and may have implications for different aspects of adolescents’ adjustment (Barber et al. 2002; Silk et al. 2003). For example, Silk et al. (2003) found that perceptions of adolescents regarding their parents’ promotion of autonomy and psychological control were only weakly negatively related to each other. Both were associated, in opposite ways, to the self-concept of the adolescents, yet only psychological control was related to more depression and anxiety among the adolescents. In summary, parents’ promotion of psychological autonomy and parents’ psychological control have been examined extensively in Western samples. Prior findings underscore the distinct roles that these parenting practices play in shaping the socio-emotional adaptation of European American adolescents. Yet, little is known about their effects on the development of Mexican–American adolescents. The goal of the current study was to begin to fill this void.

Promotion of Psychological Autonomy in Mexican–American Families

Parents’ promotion of psychological autonomy could be particularly important for the well-being of Mexican–American adolescents. Mexican–American adolescents tend to endorse collectivist oriented values (Delgado-Gaitan 1993; Phinney et al. 2000, 2005), such as “Familismo”, which is defined as “feelings of loyalty, reciprocity, and solidarity towards members of the family, as well as to the notion of the family as an extension of self” (Sabogal et al. 1987, p. 398) and “Respeto”, which refers to the importance of maintaining respectful hierarchical relationships determined by age, gender, and social status (Harwood et al. 2002). During adolescence, Mexican–American youth may face the challenge of integrating these traditional familial values with a developmental need for autonomy, which is encouraged by European American cultural values (Zayas et al. 2005). In support of this expectation, previous studies indicated that although the self-assertion behaviors of Mexican–American adolescents and their direct disagreements with parents were as frequent as their European American peers, Mexican–American adolescents were less likely than European Americans to believe that they should directly disagree with their parents (Fuligni 1998; Phinney et al. 2005). This suggests that Mexican–American adolescents may experience a gap between their perceptions of their actual behavior and their ideal behavior. Such a gap could lead to psychological distress and to lower self-worth (Harter 1985; Zayas et al. 2005). Perceptions of parents’ encouragement and acceptance of their psychological autonomy could be critical in reducing the gap and in enhancing the adolescents’ well-being (Zayas et al. 2005).

Immigrant families experience an ongoing process of acculturation, which refers to a cultural change that occurs when two or more cultural groups interact (Berry 1980). The meaning of “Familismo” and “Respeto” for Mexican–American adolescents may change with acculturation (Sabogal et al. 1987). Less acculturated adolescents might experience a larger gap between their set of values and their need for autonomy compared to more acculturated adolescents. In support of this expectation, Fuligni (1998) found that the gap between beliefs about the legitimacy of directly disagreeing with parents and actual self-assertion behaviors was larger for adolescents from families who had lived less than two generations in the US compared to adolescents from families who had lived in the US for more generations. If the dilemma between endorsement of familial values and the need for autonomy is more intense for less acculturated adolescents, then it is expected that the effects of parents’ promotion of autonomy on the well-being of the adolescents would be particularly strong for less acculturated adolescents. This suggests that the acculturation level of the adolescents should be included as a moderator when studying the implications of promotion of psychological autonomy for Mexican–American adolescents’ adjustment.

To date, only a handful of studies have examined the relationships between perception of promotion of psychological autonomy and the adjustment of adolescents in Hispanic (representing various Latin countries of origin) samples. These studies have neither yielded a consistent pattern of findings nor have they examined the moderating role of acculturation. In two studies (Bean and Northup 2009; Silk et al. 2003), perceptions of Hispanic late adolescents of their parents’ promotion of psychological autonomy were positively associated with the adolescents’ self-worth. In a third study, Love and Buriel (2007) failed to find significant associations between perceptions of this parenting practice and the depressive symptoms of Mexican–American early adolescents. The design and the sample of Love and Buriel’s study differed from those employed in the studies of Bean and Northup (2009) and Silk et al. (2003). Thus, it is unclear whether the link between the promotion of psychological autonomy and adolescents’ adaptation is evident only for Hispanic adolescents from non Mexican–American backgrounds, whether it is evident only in late adolescence, or whether a wider range of outcomes merit examination. Clearly more research is needed to determine the implications of parents’ promotion of psychological autonomy for Mexican–American adolescents, and to clarify the role of acculturation in these associations.

Psychological Control in Mexican–American Families

Parents’ psychological control is considered inappropriate and maladaptive for European American adolescents’ development (Barber and Harmon 2002). However, this may not be the case for Mexican–Americans. As suggested by Halgunseth et al. (2006), the developmental meaning of psychological control for Hispanic families may vary according to the motivations that underlie this parental behavior. Psychological control could reflect childrearing values of respect for parents as well as “Educación” (Education), namely parental responsibility to educate the child to internalize Mexican cultural values. In this context, psychological control could be perceived as legitimate, as an indicator of parental love and caring. Thus, while the perceptions of European American adolescents regarding psychological control and regarding promotion of psychological autonomy were found in previous studies to be weakly negatively associated (Silk et al. 2003), the perceptions of Mexican–American adolescents regarding the two parenting practices could be positively related. Also, the perceptions of psychological control may have only weak negative effects on the adjustment of Mexican–American adolescents.

If cultural values shape the meaning of psychological control, then the acculturation level of the adolescents can modify the effects of this parenting practice. More acculturated adolescents may interpret parents’ psychological control as less appropriate compared to less acculturated adolescents. Thus, a positive association between the perceptions of promotion of psychological autonomy and psychological control may be found only for less acculturated adolescents, while a weak negative association is expected for more acculturated adolescents. Acculturation may also modify the links between psychological control and poor adjustment of the adolescents. A weaker relation between psychological control and poor adjustment is expected for less acculturated adolescents, while a stronger association is expected for more acculturated adolescents.

As in the case of autonomy promotion, the studies that have examined psychological control among Hispanic American families did not assess the acculturation level of the adolescents. Moreover, these studies have not provided a consistent set of findings regarding the links between psychological control and adolescents’ adaptation. Silk et al. (2003) found that the perceptions of Hispanic late adolescents regarding psychological control and promotion of psychological autonomy were weakly negatively interrelated. Yet, two other studies suggest that Hispanic adolescents from ages 13 to 18 tend to interpret parents’ control practices (Crockett et al. 2007), and specifically parents’ psychological control (Mason et al. 2004), in a positive way, as signs of caring. As for the associations between psychological control and adolescents’ adjustment, in a previous study on a different component of parental control, parents’ behavioral control (i.e., using frequent commands), more parental control was related to fewer anxiety symptoms of Mexican–American early adolescents (Luis et al. 2008). Also, Walker-Barnes and Mason (2001) found that perception of higher levels of psychological control were associated with a decrease in gang involvement among Cuban–American late adolescents. Yet, in Silk et al.s’ (2003) study, psychological control was not related to deviant behaviors of Hispanic adolescents, and perceptions of more psychological control were related to lower self-worth and to more internalizing symptoms. Finally, Bean and Northup (2009) and Plunkett et al. (2007) found that perceptions of more psychological control were associated with lower self-worth of Hispanic late adolescents. In summary, the role that parents’ psychological control plays in the development of Mexican–American adolescents and the moderating role of acculturation remain unclear, and merit further examination.

The Current Study

In light of the paucity of previous work on parents’ promotion of psychological autonomy and psychological control in Hispanic families, the inconsistent pattern of results, and the lack of examination of the effects of acculturation, our study had two objectives. Our first goal was to re-examine the associations between perceptions of parents’ promotion of psychological autonomy, perceptions of parents’ psychological control, and the adaptation of Mexican–American early adolescents. In an effort to replicate extant research, we assessed similar aspects of adolescents’ adjustment to the ones used in previous studies (e.g., Plunkett et al. 2007; Silk et al. 2003; Walker-Barnes and Mason 2001), namely global self-worth, depressive symptoms, and number of delinquent friends. The last measure is considered a proximal risk factor for adolescents’ antisocial behavior and delinquency involvement (Henry et al. 2001; Scaramella et al. 2002). Our second aim was to explore the moderating role of adolescents’ acculturation in the association between the perceptions of the two parenting practices and in their links with adolescents’ adjustment.

There were three other unique aspects to our study. First, we focused on one subgroup of Hispanic adolescents, namely Mexican–Americans, in view of the recognition that the Hispanic ethnic category is heterogeneous and that specific subgroups within this ethnic group need to be evaluated separately (Halgunseth et al. 2006; Harwood et al. 2002). Second, we focused on early adolescence because it is considered to be the period of onset of an individuation process, in which the parent-adolescent relationship changes toward adolescents’ increased psychological autonomy (Holmbeck 1996; Paikoff and Brooks-Gunn 1991). Finally, we employed a longitudinal design, which enabled us to examine the predictive links between parenting practices and the adjustment of the adolescents 2 years later, while controlling for baseline levels of adolescents’ adjustment.

The following hypotheses guided our study. First, in accordance with previous work regarding the effects of Hispanic cultural values on the meaning of parenting practices (Crockett et al. 2007; Halgunseth et al. 2006; Mason et al. 2004), we hypothesized that the associations between the perceptions of parents’ promotion of psychological autonomy and parents’ psychological control would vary according to the level of adolescents’ acculturation. We hypothesized that a weak negative correlation between the perceptions of the two practices would be found for more acculturated adolescents. However, the perceptions of the two practices would be positively interrelated for less acculturated adolescents. Second, following most of the extant work on the associations between parents’ promotion of psychological autonomy and Hispanic adolescents’ enhanced well-being (Bean and Northup 2009; Silk et al. 2003; Zayas et al. 2005), we hypothesized that perceptions of more promotion of psychological autonomy would predict better adjustment 2 years later. Third, since it was suggested that less acculturated adolescents may face a larger gap between adhering to traditional values and the developmental need of autonomy, compared to more acculturated adolescents (Fuligni 1998; Zayas et al. 2005), we hypothesized a stronger association between parents’ promotion of psychological autonomy and better adolescents’ outcomes for relatively less acculturated adolescents. Fourth, in light of the concurrent associations found in some of the previous studies between parents’ psychological control and poorer adjustment of Hispanic late adolescents (Bean and Northup 2009; Plunkett et al. 2007; Silk et al. 2003), we hypothesized that perceptions of more parental psychological control would be related to poorer outcomes for early Mexican–American adolescents. However, since cultural values may shape the meaning that adolescents attribute to parents’ controlling behaviors (Crockett et al. 2007; Halgunseth et al. 2006; Mason et al. 2004), we hypothesized a weaker association between the perceptions of psychological control and poor adjustment for adolescents with low acculturation levels.

Method

Participants

One hundred and sixty-seven Mexican–American fifth graders (91 females, 54.5%) took part in the first phase of the study, as part of a larger longitudinal study of economic stress in Southern California (Parke et al. 2004). All participants were from dual-parent households, which included two biological parents of Mexican descent. The adolescents had attended school in the US for 5 years. Adolescents were followed annually for 3 years. At 7th grade, 33 (19.76%) adolescents were not located or refused to participate in the additional phase of the study, leaving a sample of 134 adolescents (54.5% females). There were no significant differences in the study’s measures between those who participated in the 7th grade phase and those who did not (all ps ≥ .41 except for depressive symptoms, for which p = .07).

At the time of initial recruitment, the mean age of the 134 adolescents who participated in fifth and in seventh grade was 10.83 (ranging from 10 to 13, SD = .63 in years), mothers’ mean age was 37.27 (ranging from 26 to 54, SD = 6.05), and fathers’ mean age was 38.75 (ranging from 26 to 56, SD = 5.93). The average total family income was $35,337 (SD = 25,039.01). In 24 families (17.9%), adolescents and both parents were born in Mexico. In 72 families (53.7%), adolescents were born in the US and both parents were born in Mexico. In 38 families (28.4%), the adolescent and at least one of the parents were born in the US.

Procedure

Interviews were held at a campus research center and at several alternate sites closer to family residences, including schools and community centers. When the adolescents were in fifth and seventh grade, they completed an identical series of questionnaires. Families were paid $200 for participation in each phase of the study.

Measures

All measures were available for participants to respond in English or Spanish. The Spanish questionnaires were translated into English and then back-translated. Only 8 adolescents (6%) chose to respond in Spanish.

Perceptions of Parenting Practices

To assess adolescents’ perceptions of their parents’ promotion of psychological autonomy and psychological control, the Block Child Rearing Practices Report (CRPR; Rickel and Biasatti 1982) was used. The CRPR includes 58 items, tapping various parenting practices. Adolescents responded to the CRPR twice, once reporting on their perceptions of their mothers’ practices, and once referring to their fathers’ practices. Studies of Hispanic families have usually reported good internal reliability and construct validity of the CRPR scales (Ceballo and Hurd 2008; Dixon et al. 2008; Ferrari 2002).

Promotion of Psychological Autonomy

We extracted a new scale from the CRPR, which included 5 items, and which reflected parental encouragement of adolescents’ self-expression (e.g., “My mother respects my opinions and encourages me to express them”) and of adolescent’s self-exploration (e.g., “My father encourages me to be curious, to explore, and to question things”). The items were chosen based on their resemblance to items used in extant measures of promotion of psychological autonomy (e.g. Bean and Northup 2009). The adolescents were asked to rate their perceptions of each parent’s promotion of psychological autonomy using a Likert scale ranging from 1 (“Strongly disagree”) to 5 (“Strongly agree”). Items were averaged, and a higher score reflected perception of more promotion of psychological autonomy. Since we did not have specific hypotheses regarding the perceived practice of each parent, and since we found high correlations between perceived maternal and paternal scores (r = .55, p = .000), we aggregated them by calculating their mean (Cronbach alpha of the scale was .78).

Psychological Control

We extracted a new scale from the CRPR, which included 11 items, reflecting anxiety and guilt induction (e.g. “My mother controls me by warning me about the bad things that can happen to me”); love withdrawal (e.g. “When my father is frustrated with me, he sometimes ignores me for a while”); and pressuring the adolescent to feel and behave according to the parents desires (e.g. “My mother does not allow me to get angry with her”). The items were chosen based on their face validity, i.e., their resemblance to items used in extant measures of psychological control (e.g., Plunkett et al. 2007; Silk et al. 2003). Also, a previous study of Hispanic mothers of 4th to 5th graders, mostly from Mexican–American backgrounds (Ceballo and Hurd 2008), used a similar psychological control scale, which was also extracted from the CRPR. The study documented good internal reliability of the scale (alpha = .76), and provided support for its validity for Hispanic samples, by showing that more acculturated mothers tended to report less use of psychological control.

The adolescents rated their perceptions of psychological control of each parent using a Likert scale ranging from 1 (“Strongly disagree”) to 5 (“Strongly agree”). Items were averaged, and higher scores reflected perception of more psychological control. Since we did not have specific hypotheses regarding perceived psychological control of each parent, and since we found high correlations between maternal and paternal scores (r = .50, p = .000), we aggregated them by calculating their mean (Cronbach alpha of the scale was .77).

Validation of the Scales of Promotion of Psychological Autonomy and of Psychological Control

As we were using new scales for assessing promotion of psychological autonomy and psychological control, we took several steps to examine their validity. First, to establish content validity, five researchers, experts in child and adolescent development and three graduate students in a developmental psychology doctoral program who were blind to the study aims and hypotheses, were provided with the definitions of parents’ promotion of psychological autonomy and of parents’ psychological control. They were asked to sort the selected 16 items into measures of promotion of psychological autonomy or psychological control. A uniform agreement (100%) among raters was found.

Second, to provide convergent and discriminant validity of the scales, we examined them in a sub-sample of the larger study (Parke et al. 2004), which consisted of 83 middle class European American 7th graders. The scales showed good internal reliability (α = .86 and α = .79 for promotion of psychological autonomy and psychological control respectively). Consistent with previous studies on European-American families, the scores of promotion of psychological autonomy and psychological control were not significantly interrelated (r = .04, p = .37). As expected, perception of more promotion of psychological autonomy was correlated with higher self-worth, fewer symptoms of depression and with having fewer delinquent friends. Perception of more psychological control was associated with having more delinquent friends and with parental reports of higher adolescent delinquency. The complete data is available from the first author.

Adolescents’ Adjustment

Self-Worth

The Global Self Worth Scale (GSW) from The Self-Perception Profile for Children (SPPC; Harter 1985) was used. The SPPC is a well-standardized measure for assessing self-worth in adolescence (Muris et al. 2003). The GSW provides a measure of a child’s global judgment of her/his own worth as a person. It includes 6 items, each consisting of two opposite descriptions (e.g., “Some kids are often unhappy with themselves” but “Other kids are pretty pleased with themselves”). Respondents chose the description that best fit them and indicated whether the description was “really true” or “sort of true”. Accordingly, each item was scored on a 4-point scale with a higher score reflecting a more positive self view. Items were averaged to create a global self-worth score. The GSW was used in studies of Mexican–American adolescents, demonstrating good internal reliability and construct validity (Hess and Petersen 1996; Michaels et al. 2007; Schmitz 2006). For our sample, the scale showed good internal reliabilities (α = .72 and α = .83 in 5th and 7th grade respectively).

Depressive Symptoms

Adolescents completed the Child Depression Index (CDI; Kovacs 1985). The CDI is a 27-item scale that assesses affective, cognitive, and behavioral symptoms of depression. For each item, adolescents chose one of three statements that best described their symptoms during the past 2 weeks (e.g. “I am sad”; “I feel like crying”) on scales ranging from 1 to 3, with higher score indicating more depressive symptoms. An item on suicidal ideation was excluded from the study because of ethical concerns. Therefore, adolescents responded to 26 items which were averaged to create a depressive symptoms score. The CDI has been used in studies of Mexican–American adolescents, demonstrating good internal reliability and construct validity (Love and Buriel 2007; Molina et al. 2009). For our sample, the CDI showed good internal consistency (α = .80 at both 5th and 7th grade).

Number of Delinquent Friends

Adolescents completed the short version of the Delinquent Peer Association Scale (Stoolmiller 1994), which includes 7 items tapping delinquent behaviors on the part of the adolescents’ friends. Adolescents rated how many of their friends have conducted delinquent behaviors during the last year using a scale ranging from 0 (“none of them”) to 5 (“all of them”). Items included, for example, “stolen something worth more than $50”, and “suggested you do something that was against the law”. Items were averaged to create a scale of the mean number of delinquent friends. The scale was validated in a longitudinal study of 4th to 8th grade boys, mostly from a European American background (Stoolmiller 1994). For the current samples, Cronbach alphas were .74 and .79 in fifth and seventh grade respectively.

Adolescents’ Acculturation

The Short Acculturation Scale for Hispanic Youth (SASH-Y; Barona and Miller 1994) was used. The scale includes 12 items regarding use of English and Spanish (e.g. “What language do you usually think in?”) and preferences for social relations (e.g. “My close friends are _______”). Adolescents responded to the social relations items on a scale ranging from “All Latinos” (1) to “Only non-Latinos” (5), and responded to the language-related items on a scale ranging from “Only Spanish” (1) to “Only English” (5). A global acculturation score, based on averaging 12 items was constructed. Higher scores reflected greater acculturation to English speaking society in the US. The SASH-Y has shown good internal consistency, split-half and test–retest reliabilities among Hispanic adolescents (Barona and Miller 1994; Serrano and Anderson 2003) and has been cross-validated across samples (Serrano and Anderson 2003). In our sample it showed good internal reliability (α = .83).

Data Preparation

Data were examined for nonnormality (i.e., skewness > 1, kurtosis > 2; see Garson 2010) and a log transformation was used to correct positive skewness and kurtosis of Family income (skew = 1.57 and kurtosis = 2.86 before transformation, skew = .89 and kurtosis = .25 after transformation), and the Delinquent peer association scores at 5th grade (skew = 2.03 and kurtosis = 5.19 before transformation, skew = 1.07 and kurtosis = .89 after transformation) and at 7th grades (skew = 1.63 and kurtosis = 2.95 before transformation, skew = .89 and kurtosis = .25 after transformation).

Results

Preliminary Analyses

Means and standard deviations for variables used in primary analyses are presented in Table 1. Adolescents from lower income families reported lower levels of acculturation (r = .26, p = .004). Adolescents from lower income families also reported lower self-worth (r = .18, p = .04) and more depressive symptoms (r = −.24, p = .01) at the 5th grade assessment. Significant gender differences in three of the study variables were found. In 7th grade, girls reported lower self-worth (M = 3.17, SD = .67) compared to boys (M = 3.38, SD = .52), t (130.01) = 2.05, p = .04. In 5th grade, girls reported having fewer delinquent friends (M = .11, SD = .10) than boys (M = .16, SD = .15), t (97.20) = 2.22, p = .03. Finally, in 7th grade, girls also reported having fewer delinquent friends (M = .12, SD = .12) than boys (M = .17, SD = .13), t (132) = 2.34, p = .02.

Table 1

Descriptive statistics

Variable	n	M	SD
Fifth grade
Acculturation	127	3.17	.68
Promotion of psychological autonomy	134	4.00	.66
Psychological control	134	3.38	.54
Self-worth	134	3.28	.63
Depressive symptoms	127	1.27	.20
Number of delinquent friends	126	.13	.13
Seventh grade
Self-worth	133	3.27	.61
Depressive symptoms	132	1.23	.19
Number of delinquent friends	134	.14	.13

Promotion of Psychological Autonomy, Psychological Control, and Acculturation

As shown in Table 2, parents who were perceived as higher in their promotion of autonomy were also perceived by their adolescents as more psychologically controlling. To examine whether acculturation moderated this link, and specifically whether more acculturated adolescents showed the same positive association between the two parenting practices, we conducted a regression analyses, in which promotion of psychological autonomy, acculturation and the interaction between them were regressed on psychological control. The independent variables were centered to reduce problems of collinearity, and the interaction term was constructed based on the centered scores (Holmbeck 2002). The Tolerance level ranged from .97 to .99, suggesting that there was no multicollinearity problem. The interaction term of promotion of psychological autonomy and acculturation was not significantly associated with psychological control (β = .07, p = .40), suggesting that the positive relation between promotion of psychological autonomy and psychological control did not vary across levels of acculturation.1

Table 2

Intercorrelations among the study variables (N = 126 to N = 134)

Variable	1	2	3	4	5	6	7	8	9
Fifth grade
1. Acculturation	–	−.08	−.03	.16†	−.19*	.03	.14	−.27**	.07
2. Promotion of psychological autonomy		–	.41***	.24**	−.17†	.02	.08	−.07	−.19*
3. Psychological control			–	−.09	.20*	.12	−.11	.20*	−.02
4. Self-worth				–	−.49***	−.13	.34***	−.25**	−.03
5. Depressive symptoms					–	.11	−.33***	.37***	.06
6. Number of delinquent friends						–	−.10	.08	.35***
Seventh grade
7. Self-worth							–	−.61***	−.08
8. Depressive symptoms								–	.20*
9. Number of delinquent friends									–

† p < .10, * p < .05, ** p < .01, *** p < .001

Promotion of Psychological Autonomy, Psychological Control, Acculturation, and Adjustment

Simple correlations of predictor and outcome variables are presented in Table 2. More acculturated adolescents reported fewer depressive symptoms at the two time points. Perception of more promotion of psychological autonomy was associated with higher self-worth at 5th grade and with fewer delinquent friends at 7th grade. Perception of more psychological control was associated with more depressive symptoms at the two time points.

Hierarchical regression analyses were conducted next to examine whether perceptions of parenting practices in 5th grade were predictive of adjustment in 7th grade after controlling for background variables and for adolescents’ baseline adjustment in 5th grade. These analyses also enabled the examination of the moderating role of adolescents’ acculturation. Three regression analyses were performed, one for predicting self-worth in 7th grade, one for depressive symptoms in 7th grade, and one for number of delinquent friends in 7th grade. In each regression equation the following variables were included. Block 1 included adolescents’ gender, family income, acculturation level, and the relevant 5th grade adjustment measure (e.g., self-worth in 5th grade was included in the regression equation predicting self-worth in 7th grade).

In light of the positive correlation found between the two parenting practices, in order to examine the unique roles which perceptions of promotion of psychological autonomy and perceptions of psychological control played in adolescents’ adjustment, we included both in the same regression equations rather than analyze separately the links between each parenting practice and adjustment. Thus, Block 2 included perceptions of promotion of psychological autonomy and perceptions of psychological control in 5th grade. To examine if acculturation moderated the links between parenting practices and adolescents’ adjustment, Block 3 included the interactions between each parenting practice and acculturation in 5th grade.2 In all analyses, parenting practices and acculturation were centered to reduce problems of collinearity. Interaction terms were constructed based on centered scores (Holmbeck 2002). When an interaction term significantly predicted adjustment, it was analyzed further by examining the relevant simple slopes (Schubert and Jacoby 2004). Because there were moderate to high correlations between some of the variables, the regression models were checked for multicollinearity. The examination of the Tolerance values for each independent variable did not reveal signs of multicollinearity, as the smallest value found was .71.

Self-Worth

As shown in Table 3, perceptions of promotion of psychological autonomy and perceptions of psychological control in 5th grade did not predict adolescents’ self-worth two years later. However, the interaction between promotion of psychological autonomy and acculturation in 5th grade was significant. This indicated that the link between perception of parents’ promotion of psychological autonomy and self-worth was different according to the various levels of adolescents’ acculturation. We next examined the simple slopes for the association between promotion of psychological autonomy and self-worth at 1 SD below the mean, at the mean score, and 1 SD above the mean of adolescents’ acculturation (Schubert and Jacoby 2004). As can be seen in Fig. 1, perception of more promotion of psychological autonomy predicted higher self-worth 2 years later only when the acculturation level of adolescents was low (β = .31, p = .03). When adolescents’ acculturation levels were medium or high, their perceptions of parents’ promotion of psychological autonomy did not significantly predict their self-worth (β = .05, p = .62 and β = −.21, p = .15 respectively). Finally, acculturation did not moderate the link between perceptions of psychological control in 5th grade and self-worth in 7th grade (see footnote 1).

Table 3

Hierarchical regressions predicting adjustment in 7th grade

Predictors	Adjustment in 7th grade
	Self-worth (n = 124)		Depressive symptoms (n = 123)		Delinquent peers (n = 124)
	β	∆R 2	β	∆R 2	β	∆R 2
Step 1		.15***		.22***		.17***
Gendera	−.17*		.09		−.17*
Family income	.04		.02		−.14
Acculturation	.07		−.26**		.05
Adjustment—5th gradeb	.30***		.33***		.31***
Step 2		.01		.04*		.04*
Autonomy—5th grade	.04		−.21*		−.23**
Control—5th grade	−.09		.20*		.05
Step 3		.05*		.01		.01
Autonomy × acculturation	−.25*		.13		.10
Control × acculturation	.06		−.07		−.10
Total R 2	.20		.27		.22
Final model	F (8,115) = 3.61***		F (8,114) = 5.33***		F (8,115) = 4.02***

* p < .05, ** p < .01, *** p < .001

aFor adolescents’ gender 1 = female 0 = male

bAdjustment refers to adolescents’ baseline adjustment in 5th grade (e.g., for the regression analysis predicting self-worth in 7th grade, adjustment refers to self-worth in 5th grade)

Fig. 1

Associations between parents’ promotion of psychological autonomy in fifth grade and adolescents’ self-worth 2 years later at low, medium, and high levels of adolescents’ acculturation in fifth grade

Depressive Symptoms

As can be seen in Table 3, perception of more promotion of psychological autonomy promoting in 5th grade predicted fewer depressive symptoms 2 years later. Perception of less psychological control in 5th grade predicted more depressive symptoms 2 years later. Acculturation did not moderate these links (see footnote 1).

In light of the positive correlations found between the perceptions of promotion of psychological autonomy and psychological control (see Table 2), we used a t-test, to examine if the coefficients of promotion of psychological autonomy and psychological control in the regression analysis predicting depressive symptom were significantly different from each other. The analysis revealed a significant difference between the two coefficients, t (116) = −3.07, p < .01, which provides additional support for the discriminant validity of the scales.

Number of Delinquent Friends

As shown in Table 3, perceptions of more promotion of psychological autonomy in 5th grade predicted having fewer delinquent friends 2 years later. Psychological control did not predict the number of delinquent friends 2 years later. Acculturation did not moderate the links between parenting practices and number of delinquent friends (see footnote 1).

Discussion

Parents’ promotion of psychological autonomy and parents’ psychological control are considered salient in parent-adolescent relationships in Western individualist cultures. Various studies have indicated that these parenting practices are distinct and have opposite effects on adolescents’ competencies during adolescence and early adulthood (Barber et al. 2002; Grotevant and Cooper 1998; Silk et al. 2003). However, the generalization of the findings to Mexican–American adolescents, who tend to endorse familial values based on their collectivistic cultural orientation (Phinney et al. 2000), is less clear, and to date is under-examined. The goal of our study was to address this gap in the literature. First, we examined the associations between the perceptions of Mexican–American early adolescents regarding parents’ promotion of psychological autonomy and psychological control, and the adjustment of the adolescents. Second, the acculturation experience of Mexican–American adolescents is a dynamic and ongoing process (Sabogal et al. 1987), which can affect their interpretations of their parents’ practices. Therefore, we also examined the moderating role of the acculturation level of the adolescents in the associations between the perceptions of parenting practices and the adjustment of the adolescents.

We expected that acculturation would alter adolescents’ perceptions of the two parenting practices. We hypothesized that a weak negative association between the perceived practices would be found for more acculturated adolescents, while the perceptions of the two parenting practices would be positively related for less acculturated adolescents. We found only partial support for this hypothesis. Regardless of their acculturation level, adolescents who perceived their parents as more promoting of psychological autonomy also tended to perceive them as more psychologically controlling. This is consistent with the studies of Crockett et al. (2007) and Mason et al. (2004), indicating that parental control may be perceived positively by Hispanic adolescents, as an indirect expression of parental caring. It suggests that even for relatively acculturated early adolescents, familial values may continue to be central and shape their perceptions of their parents’ childrearing practices.

The positive correlation between the parenting practices is contrary to Silk et al. (2003) who found a weak negative association between the promotion of psychological autonomy and psychological control among Hispanic American youth. The discrepancy could stem from the differences in the ages of participating adolescents in the two studies. Our study examined early adolescents, while Silk et al. studied adolescents in 9th through 12th grades. Perhaps by late adolescence, parents’ psychological control is perceived by Mexican–American adolescents as less appropriate. Indirect support for this interpretation comes from earlier studies by Fuligni (1998) and Phinney et al. (2005) indicating that older Mexican–American adolescents believe it is more appropriate to disagree with their parents than younger Mexican–American adolescents. Future research could benefit from examining adolescents’ perceptions of these parenting practices at different ages throughout adolescence.

Even though the perceptions of promotion of psychological autonomy and of psychological control were positively inter-related, they were associated with adolescents’ well-being in opposite directions. Controlling for baseline adjustment in 5th grade, perceptions of more promotion of psychological autonomy and of less psychological control in 5th grade predicted fewer depressive symptoms 2 years later. Perceptions of more promotion of psychological autonomy in 5th grade also predicted having fewer delinquent friends in 7th grade. These results support our hypotheses, and are generally consistent with Bean and Northup (2009) and Silk et al. (2003) who found that more autonomy promoting and less psychological control were concurrently associated with Hispanic adolescents’ better adaptation. Our results expand upon this previous work in several important ways. First, these earlier studies (Bean & Northrup and Silk et al.) were cross-sectional, while the current study was longitudinal and demonstrated predictive links between early autonomy promoting and psychological control and later adjustment. Second, our study extends to early adolescents the findings of Bean & Northrup and Silk et al. on late adolescents. Third, Bean & Northrup and Silk et al.’s samples consisted of adolescents of heterogeneous Hispanic origins. Following the call to separately examine developmental processes in specific Hispanic subgroups (Halgunseth et al. 2006; Harwood et al. 2002), we found similar results for Mexican–American adolescents.

A novel aspect of our study was the examination of adolescents’ acculturation as a moderator. We hypothesized a stronger relationship between the promotion of psychological autonomy and better adjustment for less acculturated adolescents, and a stronger link between parents’ psychological control and poorer adjustment for more acculturated adolescents. These hypotheses were not supported for the most part. The parenting practices generally had the same implications for the well-being of the adolescents, regardless of the level of the adolescents’ acculturation. However, acculturation did moderate the link between the promotion of psychological autonomy and adolescents’ self-worth, suggesting that this link was significant only for less acculturated adolescents. As we hypothesized, the perception of more promotion of autonomy was related to higher self-worth among less acculturated adolescents. Fuligni (1998) and Zayas et al. (2005) suggested that Mexican–American adolescents, especially when less acculturated, may experience a larger gap between their ideal behaviors, which are shaped by traditional values, and their actual behaviors, which are influenced by the developmental need for autonomy. Our finding is in accordance with this previous work, and may suggest that for less acculturated adolescents, parents’ support of psychological autonomy could be especially beneficial in reducing the ideal–actual behavior gap and, in turn, enhance self-worth.

Limitations and Conclusions

Several limitations of the study point to directions for future research. First, although our study was longitudinal, strong inferences about the causal relations between perceived parenting and adolescents’ adjustment cannot be drawn. Second, although the SASH-Y scale which we employed for assessing adolescents’ acculturation is a well validated measure, it does not directly assess adolescents’ values. Future research may benefit from examining changes in core cultural values of Mexican–American adolescents and the ways in which these values shape adolescents’ interpretation of their parents’ behaviors. Such work could shed more light on the processes by which acculturation affects the developmental meaning of promotion of psychological autonomy and psychological control. Finally, we focused on the adolescents’ perspectives, drawing on phenomenological (Rogers 1959) and symbolic interactionist approaches (Kinch 1972) as well as current adolescent research, suggesting that the adolescents’ subjectively experienced environment is a better predictor of their behavior and well-being compared to their parents’ subjective reality (Boyce et al. 1998; Powers and Welsh 1999). However, since this raises shared variance concerns, future research may benefit from observations of parent–adolescent communication to assess parents’ promotion of autonomy and psychological control in families of Mexican–American adolescents.

Adolescents’ development towards becoming more self-regulating, self-motivating and independent, while maintaining close relationships with their parents, is considered a central developmental task for European American adolescents (Zimmer-Gembeck and Collins 2003). The promotion of adolescents’ psychological autonomy in combination with low levels of psychological control are vital in this process, and have been found to facilitate adolescents’ well-being (Grotevant and Cooper 1998; McLeod et al. 2007a, b). Our results suggest that Mexican–American early adolescents may differ from European Americans in terms of their attitudes regarding promotion of psychological autonomy and psychological control. Nevertheless, they may resemble European Americans in the way that their adaptation is related to these parenting practices. As acculturation generally did not moderate the links between perceived parenting and adjustment, our findings suggest that even in the context of a relatively collectivist culture, promotion of psychological autonomy and limiting of psychological control can play important roles in the adjustment of early adolescents.

Finally, as Mexican–American adolescents are at relatively high risk for mental health problems, our findings may contribute to the development of prevention and intervention programs on behalf of this ethnic group. Programs aimed at enhancing Mexican–American early adolescents’ well-being may benefit from including parents. Such programs could be strengthened by emphasizing the value of promoting adolescents’ psychological autonomy through encouraging their self-exploration and self-expression in the family. In addition, such programs might also benefit from addressing the potential negative effects of psychological control practices, such as guilt and anxiety induction and conditional acceptance, and from highlighting their ineffectiveness in reducing adolescents’ affiliation with deviant peers. These programs could potentially improve family functioning and increase the long term well-being of Mexican–American youth.