How are Mismatched Parent-Adolescent Autonomy Beliefs Related to Psychological Adjustment among Immigrant Chinese Canadian Families?
by
Lauren Julia Chance
B.Sc., University of Waterloo, 2007 A Thesis Submitted in Partial Fulfillment of the
Requirements for the Degree of MASTER OF SCIENCE in the Department of Psychology
© Lauren Julia Chance, 2009 University of Victoria
All rights reserved. This thesis may not be reproduced in whole or in part, by photocopy or other means, without the permission of the author.
How are Mismatched Parent-Adolescent Autonomy Beliefs Related to Psychological Adjustment among Immigrant Chinese Canadian Families?
by
Lauren Julia Chance
B.Sc., University of Waterloo, 2007
Supervisory Committee
Dr. Catherine L. Costigan, Supervisor (Department of Psychology)
Dr. Bonnie J. Leadbeater, Departmental Member (Department of Psychology)
Dr. Marion F. Ehrenberg, Departmental Member (Department of Psychology)
Supervisory Committee
Dr. Catherine L. Costigan, Supervisor (Department of Psychology)
Dr. Bonnie J. Leadbeater, Departmental Member (Department of Psychology)
Dr. Marion F. Ehrenberg, Departmental Member (Department of Psychology)
ABSTRACT
The relations between mismatched parent-adolescent autonomy beliefs, and psychological adjustment and parenting self-efficacy were investigated among 89 immigrant Chinese Canadian families with early adolescents. Hierarchical regression analyses assessed whether interactions between parents’ autonomy beliefs and adolescents’ autonomy beliefs predicted psychological adjustment and parenting self-efficacy. Parent-adolescent conflict intensity was evaluated as a mediator of these relations. Parental warmth was evaluated as a moderator of the relations between autonomy beliefs and adolescent psychological adjustment. When adolescents held early expectations for autonomy, mothers who held later expectations for autonomy reported lower parenting self-efficacy. When fathers perceived high decision making independence, adolescents who perceived low decision making independence reported fewer depressive
symptoms. Conflict intensity did not mediate these relations. Although parental warmth did not moderate in the predicted manner, fathers' warmth moderated the relation between fathers’ expectations for autonomy and adolescent self-esteem. Implications for healthy adolescent autonomy among immigrant Chinese Canadian families are discussed.
Table of Contents
Supervisory Page ... ii
Abstract ... iii
Table of Contents ... iv
List of Tables ... viii
List of Figures ... ix
Acknowledgements ... x
Dedication ... xi
The Acculturation Process ... 2
Parent-Adolescent Acculturation Gaps and Psychological Adjustment ... 4
The Current Study ... 5
Culture and Autonomy ... 8
Goodness-of-Fit During Early Adolescence ... 9
Mismatched Autonomy Beliefs and Psychological Adjustment in Immigrant Families ... 10
Depressive Symptoms ... 11
Adolescent Self-Esteem ... 12
The Moderating Role of Parental Warmth ... 13
The Mediating Role of Parent-Adolescent Conflict Intensity ... 15
Hypotheses ... 17
Method ... 19
Participants ... 19
Procedure ... 20
Measures ... 21
Expectations for autonomy. ... 21
Decision making independence. ... 23
Conflict intensity. ... 24 Parental warmth. ... 24 Depressive symptoms. ... 25 Self-esteem. ... 25 Parenting self-efficacy. ... 26 Results ... 26 Preliminary Analyses ... 26 Demographic Variables ... 32
Autonomy Beliefs, Psychological Adjustment, and Parenting Self-Efficacy ... 33
Expectations for autonomy. ... 33
Decision Making Independence. ... 38
Direct relations. ... 41
Parental Warmth as a Moderator for Adolescent Psychological Adjustment ... 42
Parent-Adolescent Conflict Intensity as a Mediator ... 47
Discussion ... 53
Expectations for Autonomy ... 54
Decision Making Independence ... 59
Direct relations between autonomy beliefs and parent outcomes ... 62
Parental Warmth as a Moderator and Conflict Intensity as a Mediator ... 64
Healthy Adolescent Behavioural Autonomy among Immigrant Chinese Canadian Families... 66
Strengths and Limitations ... 72
Future Directions ... 74
References ... 77
Appendix A: Teen Timetable (Adolescent Report) ... 84
Appendix C: Decision Making Questionnaire (Adolescent Report) ... 86
Appendix D: Decision Making Questionnaire (Parent Report) ... 88
Appendix E: Conflict Intensity (Adolescent Report) ... 90
Appendix G: Parental Warmth ... 94
Appendix H: Centre for Epidemiological Studies Depression Scale... 95
Appendix I: Rosenberg Self-Esteem Scale ... 96
Appendix J: Parenting Self-Agency Measure ... 97
Appendix K: Exploratory Factor Analysis of Teen Timetable ... 98
Appendix L: Exploratory Factor Analysis of Decision Making Questionnaire ... 99
List of Tables
Table 1: Descriptive statistics of main study variables ...29 Table 2: Intercorrelations between main study variables ...31 Table 3: Hierarchical regression model for relation between expectations for autonomy and
depressive symptoms, parenting self-efficacy, and adolescent self-esteem ...35 Table 4: Hierarchical regression model for relations between adolescent decision making
independence and depressive symptoms, parenting efficacy, and adolescent self-esteem.. ...39 Table 5: Hierarchical regression model for three-way interactions between fathers' autonomy
beliefs, adolescents’ autonomy beliefs, and father warmth, and its relation to adolescent psychological adjustment ...44 Table 6: Hierarchical regression model for three-way interaction between mothers’ autonomy
beliefs, adolescents’ autonomy beliefs, and mother warmth, and its relation to adolescent psychological adjustment ...45 Table 7: Hierarchical regression model for mother-adolescent conflict intensity as a mediator in
the moderated relation between mothers' expectations for autonomy, adolescents'
expectations for autonomy, and maternal parenting self-efficacy ...51 Table 8: Hierarchical regression model for father-adolescent conflict intensity as a mediator in
the moderated relation between fathers' perceptions of decision making independence, adolescents' perceptions of decision making independence, and adolescent depressive symptoms ...52
List of Figures
Figure 1: Adolescent expectations for autonomy moderate the relation between mothers’
expectations for autonomy and mothers’ parenting self-efficacy ...37
Figure 2: Adolescents’ perceptions of decision making independence moderate the relation between fathers’ perceptions of decision making independence and adolescent depressive symptoms. ...41
Figure 3: Fathers' warmth moderates the relation between fathers’ expectations for autonomy and adolescent self-esteem ...46
Figure 4: Conceptual model of mediated moderation. ...48
Figure 5: Significant relations between fathers’ autonomy beliefs, adolescents’ perceptions of decision making independence and fathers' warmth , and adolescent psychological
adjustment. ...68
Acknowledgements
I would like to take this opportunity to acknowledge the faculty members at the
University of Victoria who supervised the writing of this thesis, the families who participated in the Intercultural Family Study, and the organizations that provided the financial support that made it possible for me to complete my Master’s degree. Without these contributions, this thesis would not have been possible.
I would like to thank my graduate supervisor Dr. Catherine Costigan for her mentorship and dedication from the inception of this project, as well as for her continued involvement while on leave. I would also like to thank Dr. Bonnie Leadbeater for taking me on as her student this past year, and her excellent supervision throughout the latter half of my thesis. I’m also grateful to Dr. Marion Ehrenberg for her feedback and contributions to the development of this project as a member of my committee, and to Dr. Stuart MacDonald and Dr. Michael Hunter, who
provided statistical consultation. Finally, I would like to thank Dr. Karen Kobayashi for acting as the external examiner for my defence.
I would also like thank the families who participated in the Intercultural Family Study. Their willingness to share their experiences as immigrant families in Canada, and to take the time to participate in our lab’s research, is greatly appreciated.
Finally, I would like to acknowledge the Social Sciences and Humanities Research Council of Canada and the University of Victoria for their generous financial support throughout my Master’s degree.
Dedication
This thesis is dedicated to my family – Mom, Dad, Evan, Grandma and Poppa. Thank you so much for your love, support, and encouragement!
Immigrant families face a host of challenges as they create a new life in a new country. Both adults and children may encounter stress as they adjust to a different culture, learn an unfamiliar language, and build a support system far away from their homeland. This transition may be especially difficult when children raised in immigrant families reach adolescence, a period of significant developmental change. Within the family, it is particularly challenging when children adapt to Canadian culture faster than their parents, while parents retain more of the culture of origin (Berry, 2003). Such differing rates of acculturation may translate into mismatched parent-adolescent beliefs about adolescent autonomy, and may be related to poor outcomes in the domains of psychological adjustment and parenting self-efficacy.
The granting of adolescent autonomy is a key arena in which differing levels of acculturation may manifest. Broadly defined as a state of independence and self-governance (Beyers, Goosens, Vansant & Moors, 2003; Daddis & Smetana, 2005; Spear & Kulbok, 2004), the construct of autonomy is multifaceted (Smetana, 2002). Autonomy can be achieved in behavioural, emotional and cognitive domains (Goossens, 2006). The current study focuses on behavioural autonomy, which can be differentiated from emotional autonomy (children’s decreased emotional dependence on parents as they proceed through adolescence; Smetana, 2002) and cognitive autonomy (the development of one’s own system of values; Goossens, 2006). Behavioural autonomy refers to the ability to make independent decisions and regulate one’s behaviour (Goossens, 2006; Smetana, Campione-Barr, & Daddis, 2004). While the achievement of behavioural autonomy is considered to be a developmental milestone of the adolescent period in Canadian culture, less emphasis is placed on adolescent autonomy in Chinese culture. Although parent-adolescent differences in beliefs about adolescent autonomy
and subsequent conflicts are common during this developmental stage, immigrant families face greater challenges than non-immigrant families, as they must attempt to bridge cultural gaps when negotiating the granting of adolescent autonomy, in addition to the generational gap that all families confront (Szapocznik & Kurtines, 1993). The current study investigates the relation between mismatched parent-adolescent autonomy beliefs, and psychological adjustment and parenting self-efficacy among parents and adolescents in immigrant Chinese Canadian families. Two additional variables believed to influence these relations are also explored: parental warmth as a moderator in the relations between autonomy beliefs and adolescent psychological
adjustment, and parent-adolescent conflict intensity as a mediator in the relations between
autonomy beliefs and parent psychological adjustment, adolescent psychological adjustment, and parenting self-efficacy.
The Acculturation Process
The acculturation process that immigrants undergo following their arrival in a new country has been studied extensively. In the past, acculturation was considered a unidimensional
process, in which becoming oriented towards the culture of one’s adopted country necessitated a loss in orientation towards one’s ethnic culture (Berry, 2003). However, it is now generally recognized that acculturation is a bidimensional process, in which becoming oriented towards the culture of the new country occurs independently of the maintenance of one’s ethnic culture (Berry, 2003; van de Vijver & Phalet, 2004). In other words, increased orientation towards the adopted culture does not imply the loss of one’s heritage culture. Aspects of cultural orientation that are studied as part of acculturation include both behaviours visible in the public domain, such as language use, and internalized values and identity, which constitute the private domain (Costigan & Dokis, 2006b; van de Vijver & Phalet, 2004).
Acculturation does not unfold in a uniform pattern across individuals. Some of the diversity in the degree to which immigrants adopt aspects of the new culture depends on the amount of contact that individuals have with the culture of the new country (Berry, 2003). Because immigrant children are immersed in the culture of the adopted country on a daily basis through their school attendance, they tend to become oriented towards the new culture faster than their parents, since both mothers and fathers often spend less time in contact with the culture of the new country (Berry, 2003). Likewise, diversity in the degree to which immigrants maintain their ethnic culture is related to the extent of contact with members of their ethnic community. This may further differentiate immigrant parents from their children, if children also spend proportionally less time engaged in activities with members of their own ethnic culture than parents. These parent-child discrepancies are commonly referred to as acculturation gaps. Parent-child acculturation gaps may occur in the context of ethnic or host cultural orientations. Costigan and Dokis (2006b) found that Chinese Canadian children raised in immigrant families engaged in more Canadian behavioural practices and endorsed Canadian values to a greater degree than either mothers or fathers, and engaged in fewer Chinese behavioural practices than either parent. Other studies have found acculturation gaps in the endorsement of values that are more characteristic of collectivistic cultures than individualistic cultures. For example, Phinney, Ong, and Madden (2000) compared levels of endorsement of family obligation values among members of Armenian, Mexican, and Vietnamese immigrant families living in the United States. They found that immigrant parents endorsed family obligation values to a greater degree than their adolescent children.
Acculturation gaps between parents and children may be particularly apparent in
compared to families in which children were born in the country of origin (the first generation; Szapocznik & Kurtines, 1993). For example, Phinney, Ong & Madden (2000) found that parent-child differences in the endorsement of family obligation values were significantly greater among immigrant families with American-born children compared to immigrant families with foreign-born children. Thus, while immigrant parents endorsed family obligation values to a similar degree regardless of length of residence in the United States, there was a generational effect for adolescents, such that the second generation adolescents of Armenian and Vietnamese heritage endorsed family obligation values less strongly than foreign-born adolescents of the same ethnic background.
Parent-Adolescent Acculturation Gaps and Psychological Adjustment
Acculturation gaps present challenges to parents and adolescents alike, as all family members must negotiate differences in cultural orientation, in addition to the other challenges of immigration. For example, Costigan and Dokis (2006a) found that parent-adolescent differences with regards to the ethnic Chinese culture were related to psychological adjustment. Mother-adolescent differences in preference for Chinese behavioural practices and father-child
differences in endorsement of Chinese values were related to more symptoms of depression for the adolescent. Similarly, Costigan, Chia, and Dokis (2002) found that greater parent-adolescent differences in Chinese orientation were related to higher levels of adolescent somatisation symptoms. Parent-child acculturation gaps with regards to the host culture may also have an impact on the psychological adjustment of adolescents in immigrant families. Larger parent-adolescent discrepancies in orientation to Canadian and American culture are related to higher levels of adolescent depression in immigrant families of Asian descent (Crane, Ngai, Larsen & Hafen, 2005; Ying & Han, 2007). Much less is known about how the presence of parent-child
acculturation gaps impacts the psychological adjustment of immigrant parents, but the available literature suggests that these gaps may function as a risk factor for parents as well. For example, Costigan and Dokis (2004) examined the relations between acculturation discrepancies and parent self-esteem and parenting self-efficacy, and found that greater differences in orientation towards Canadian culture predicted poorer parent psychological adjustment. Specifically, parent-child gaps in Canadian behavioural involvement and the endorsement of Canadian values were associated with parents’ lower self-esteem and lower feelings of parenting self-efficacy.
The Current Study
Overall, the pattern that emerges from the literature clearly shows that parent-adolescent acculturation gaps are linked with poor psychological adjustment in children and adults, as well as lower parenting self-efficacy. A next step in this research is to explore how these gaps manifest themselves in the everyday lives of immigrant families. That is, acculturation gaps between parents and their adolescent children may translate into more specific differences in aspects of family functioning, which in turn impact the psychological adjustment and parenting self-efficacy of parents and adolescents. For example, parent-adolescent gaps in cultural values may influence beliefs regarding appropriate family interactions, such as whether or not it is appropriate to challenge parental authority (Costigan & Dokis, 2006a). Parent-adolescent gaps in behavioural practices, such as language use, may impact the ability of parents and children to communicate effectively and resolve conflicts (Costigan & Dokis, 2006a).
Beliefs about the granting of adolescent behavioural autonomy may be one consequence of broader parent-child acculturation gaps. Because of differential rates of acculturation towards Canadian culture, children may develop autonomy beliefs more congruent with Canadian
culture, while their parents continue to hold later autonomy beliefs more consistent with Chinese culture. Some empirical evidence exists to support the relation between acculturation gaps and different beliefs about autonomy granting. For example, Costigan and Dokis (2004) found that greater parent-adolescent gaps in the endorsement of cultural values were related to greater differences in what parents and their children considered to be appropriate ages for gaining behavioural autonomy. Thus, beliefs about autonomy may be a more proximal predictor of psychological adjustment and parenting self-efficacy within immigrant families than parent-adolescent acculturation gaps.
The current study investigates parents’ and adolescents’ beliefs about behavioural autonomy, and how parent-adolescent differences in these beliefs relate to psychological
adjustment and parenting self-efficacy in the context of immigrant families of Chinese ethnicity living in Canada. Autonomy beliefs are defined in two ways: as age expectations for the achievement of adolescent autonomy and as perceptions of adolescent decision making
independence. By examining the degree to which mismatched autonomy beliefs are associated with outcomes such as depressive symptoms, adolescent self-esteem, and parenting self-efficacy, one can determine whether discrepant beliefs are related to adjustment among immigrant parents and their adolescents.
The current study also evaluates parental warmth as a moderator of the relations between mismatched parent-adolescent autonomy beliefs and adolescent psychological adjustment. The positive influence of parental warmth on the parent-adolescent relationship may buffer the stress associated with discrepant autonomy beliefs, such that mismatched parent-adolescent autonomy beliefs will be less detrimental to adolescent psychological adjustment in the context of high parental warmth. Finally, parent-adolescent conflict intensity is evaluated as a mediator of the
relations between mismatched parent-adolescent autonomy beliefs and adolescent psychological adjustment, parent psychological adjustment, and parenting self-efficacy. Divergent opinions about acceptable levels of autonomy could lead to emotionally charged arguments between parents and adolescents. Given the importance of family cohesion for psychological adjustment, it is possible that conflict intensity functions as a mechanism through which discrepant autonomy beliefs influence the psychological adjustment of parents and adolescents. Intense disagreements between parents and adolescents that result from differing views about the amount of decision making independence that adolescents are granted, and the age at which such independence should be granted, may also lead to lower parenting self-efficacy, as parents may doubt their ability to handle current and future challenges.
Understanding the link between parent-adolescent differences in autonomy beliefs and psychological adjustment and parenting self-efficacy among immigrant Chinese Canadian families is extremely relevant given the large influx of immigrants from such regions as the People’s Republic of China, Taiwan, and Hong Kong in recent years. The largest proportion of new immigrants that arrived in Canada between 1997 and 2006 emigrated from the People’s Republic of China, and Hong Kong was the main sending region from 1987 until 1996 (Chui, Tran & Maheux, 2007). A specific focus on how parent-adolescent differences in autonomy beliefs relate to psychological adjustment and parenting self-efficacy among one particular ethnic group is also important because diversity exists with regards to how patterns of family processes unfold within immigrant groups. For example, Phinney, Ong, and Madden (2000) found significant cultural differences in the size of parent-child discrepancies in views of family obligations, despite the fact that the cultural groups studied shared a collectivistic orientation. These findings emphasize the need to examine family processes within specific cultural groups.
Culture and Autonomy
The adolescent development literature suggests that the achievement of behavioural autonomy is an important developmental task to be accomplished by the time an adolescent makes the transition to adulthood (Beyers et al., 2003; Hektner, 2001; Noom, Dekovic & Meeus, 2001; Spear & Kulbok, 2004). However, this view is primarily based on research conducted with adolescents raised in individualistic cultures where autonomy is highly valued, and does not take into account cultural variation (Smetana, Campione-Barr, & Daddis, 2004). Cross-cultural research suggests that the social norms of collectivistic cultures (e.g., Chinese culture) encourage later achievement of autonomy compared to individualistic cultures (e.g., Canadian culture). Both adolescents and adults from collectivistic cultures tend to hold significantly later age expectations for the granting of behavioural autonomy compared to their counterparts raised in individualistic cultures (Feldman & Quatman, 1988; Fuligni, 1998). As long as parents and adolescents hold similar beliefs about adolescent autonomy, the delay of autonomy granting relative to adolescents from more individualistic cultures should not cause distress among adolescents and parents, or cause parents to question their ability to manage the challenges of parenting.
However, although parents and adolescents in immigrant families both tend to endorse later autonomy granting than their non-immigrant counterparts, the differences in beliefs about adolescent autonomy between immigrant parents and children tend to be larger than in non-immigrant families. For example, Kwak and Berry (2001) found that although adolescents and parents of Vietnamese, Korean, and East Indian descent endorsed parental authority more than their European Canadian counterparts, the gap between Asian Canadian adolescents and their parents in the endorsement of parental authority was greater than the gap seen in European
Canadian families. Since adolescents from immigrant families tend to become oriented to the new culture more quickly than parents, their autonomy beliefs may come to resemble those of their non-immigrant peers, whereas their parents retain beliefs about adolescent autonomy more congruent with those of their ethnic culture. This is particularly true when families immigrate to countries where the cultural norms regarding family interactions and adolescent behaviour are quite different, as is the case when families emigrate from the People’s Republic of China, Taiwan, or Hong Kong to Canada.
Goodness-of-Fit During Early Adolescence
The goodness-of-fit between an individual’s characteristics and the characteristics of the contexts in which they develop are thought to be crucial for optimal development across the lifespan (Lerner, 1982). This is especially true during early adolescence, a period which is marked by multiple changes both within the individual and their context (Lerner, 1982). Behavioural autonomy represents a domain in which changes are expected to occur in early adolescence. As children make the transition to adolescence, they begin to desire increasing autonomy to make independent decisions about various issues. They may also develop
expectations about the age at which they believe it will be appropriate to make decisions in the future, later in adolescence. However, early adolescents must also contend with limits on autonomy established by authority figures in their social context, such as their parents and their teachers.
The degree of fit between adolescents’ autonomy beliefs and their parents’ autonomy beliefs is an example of a dimension on which person-context mismatch may be related to poorer outcomes. Adolescents growing up in immigrant Chinese Canadian families must negotiate
cultural differences in autonomy beliefs, since cultural norms in Canada support earlier behavioural autonomy compared to Chinese cultural norms. If adolescents adopt autonomy beliefs consistent with Canadian culture, while their parents maintain autonomy beliefs consistent with Chinese culture, these families may be more likely to experience difficulty compared to families in which adolescents and their parents hold similar beliefs about adolescent autonomy. Such differences in autonomy beliefs between adolescents and their parents
constitute a specific type of person-context mismatch. The concept of goodness-of-fit is proposed to be important across the lifespan (Lerner, 1982). Therefore, such mismatches are likely to affect parents’ well-being as well.
Mismatched Autonomy Beliefs and Psychological Adjustment in Immigrant Families
Parent-adolescent differences in autonomy beliefs are expected to be related to poor psychological adjustment among immigrant parents and their adolescent children, as well as lower feelings of parenting self-efficacy among immigrant parents. To date, few studies have investigated mismatches in such beliefs in relation to psychological adjustment, although preliminary evidence is supportive of such a link. For example, Juang, Lerner, McKinney, and von Eye (1999) found that the degree of fit between Asian American adolescents’ preferred autonomy timetables and involvement in decision making and the actual level of autonomy allowed by their parents was more predictive of adolescent psychological adjustment and quality of family relationships than the absolute level of autonomy granted to adolescents. Specifically, mismatches in autonomy granting were related to poorer adolescent psychological adjustment and poorer quality of family relationships, compared to families in which preferred and actual autonomy timetables and decision-making involvement matched.
The current study assesses expectations for autonomy and decision making independence from the perspectives of adolescents, mothers and fathers. In the available literature, reports of autonomy beliefs are often based on the perspective of only one family member (i.e., adolescents only) and it is rare that independent self-reports from multiple sources within a family are obtained. By obtaining reports from each family member, one can gain valuable information from the perspective of adolescents and their parents. In addition to obtaining reports of autonomy beliefs from each family member, various outcomes can be assessed for each family member as well. Depressive symptoms are evaluated as an indicator of distress for both parents and adolescents. Self-esteem is evaluated for adolescents only and parenting self-efficacy is assessed based on parents’ self-reports of their feelings of agency and competence in their role as a parent. Few existing studies in this area have focused on the relation between
parent-adolescent differences in autonomy beliefs and parent-adolescent well-being, and it is even rarer that the correlates of such differences are examined among immigrant parents. From a family systems perspective, the psychological well-being of each individual family member impacts all other members (Crane et al., 2005). Furthermore, parent well-being is related to quality of parenting, and is therefore also relevant to the understanding of adolescent well-being (Hughes & Gullone, 2008).
Depressive Symptoms
Tensions that arise from mismatched parent-adolescent autonomy beliefs in immigrant Chinese Canadian families may function as a source of stress within the family and be related to higher levels of depressive symptoms for both adolescents and their parents. Stress related to gaps in autonomy beliefs may compound other stressors associated with the process of immigration. Across cultures, the occurrence of stressful life events has been linked to
depressive symptoms, including Chinese adolescents living in the People’s Republic of China, and Asian American adolescents growing up in immigrant families (Greenberger, Chen, Tally & Dong, 2000). Furthermore, internalizing problems and depressive symptoms are a common mode of expression of distress in individuals of Chinese ethnicity (Hwang, Myers, Abe-Kim & Ting, 2008). Younger adolescents are particularly vulnerable, since the emergence of depressive symptoms most frequently occurs in early to middle adolescence (Hammen & Rudolph, 2003). Therefore, depressive symptoms are an effective measure of distress among immigrant Chinese Canadian families.
Adolescent Self-Esteem
Adolescent self-esteem may also be negatively related to mismatched parent-adolescent autonomy beliefs. Immigrant Chinese Canadian adolescents must continuously balance
conflicting cultural values in their everyday lives. Trying to reconcile the values of an ethnic culture that emphasizes family obligation, with the often conflicting values of the more dominant Canadian culture is a potential source of stress, and greater numbers of stressful life events have also been found to predict lower self-esteem in adolescents (Baldwin & Hoffmann, 2002). Also, when parents hold more restrictive ideas about autonomy than adolescents, it is likely that adolescents will not be granted the permission to join their peers in various activities. Such limitations imposed by parents could compound adolescents’ struggle to fit in, making them feel left out, and increasing their sense of isolation, all of which could influence adolescents’ self-esteem.
Parenting Self-Efficacy
Parenting self-efficacy refers to parental perceptions of their ability to provide effective caregiving to their children, and their feelings of competence in their role a parent (Dumka, Stoerzinger, Jackson, & Roosa, 1996; Jones & Prinz, 2005). Past research has shown that challenges within the family (e.g., the presence of child psychopathology), as well as raising children in challenging environments (e.g., socioeconomic disadvantage) have been shown to predict lower parenting self-efficacy (Jones & Prinz, 2005). For immigrant families, conflicting cultural norms about adolescent autonomy between one’s ethnic culture and the culture of one’s adopted country may represent another type of challenge that may undermine parents’ sense of self-efficacy. If parents hold autonomy beliefs that differ markedly from the autonomy beliefs of their adolescent children, parents may doubt their ability to handle the disagreements that arise as a result, and question their ability to raise well-adjusted adolescents when faced with the
pressures of differing cultural norms about adolescent autonomy in their adopted country. This is particularly salient during early adolescence, since parents have several years of potential struggles over adolescent autonomy ahead of them. The construct of parenting self-efficacy has rarely been studied in the context of immigrant families, and is even less frequently studied in the context of immigrant families of Chinese ethnicity. It is important that more is understood about how these processes unfold within this specific population.
The Moderating Role of Parental Warmth
In addition to examining the direct relations between mismatched parent-adolescent autonomy beliefs and psychological adjustment, the current study also evaluates whether
such as reliance on parents for emotional (as opposed to material) support, and is an important component of closeness in the parent-adolescent relationship (Kagitcibasi, 2005). The link between parental warmth and positive developmental outcomes for children and adolescents has been established (Laursen & Collins, 2004). Research on the development of adolescent autonomy has recently incorporated the simultaneous consideration of parental warmth. That is, rather than focusing on the process of autonomy development in isolation, recent research has expanded its focus to understand how adolescents balance their increasing autonomy with maintaining close, emotionally supportive relationships with their parents (Beyers, 2003).
The importance of maintaining family relatedness while promoting adolescent autonomy is increasingly recognized in studies conducted within individualistic cultures (Allen, Hauser, Bell & O’Connor, 1994; Soenens & Vansteenkiste, 2005; Smetana, 2002). However, little research has been conducted outside the context of non-immigrant adolescents of European descent living in Western nations. Less is known about how autonomy and relatedness interact within the context of immigrant families, although the subject has been addressed on a
theoretical level (Kagitcibasi, 2005; Szapocnik & Kurtines, 1993). Rather than being mutually exclusive, a combination of autonomy and relatedness is thought to be ideal for healthy
development, as a balance between these two constructs fosters optimal levels of connectedness and independence in the parent-adolescent relationship. For example, Szapocnik and Kurtines (1993) suggested that parent-child acculturation gaps are less likely to be associated with negative outcomes when the quality of the parent-child relationship is high. Studies of
immigrant families often focus exclusively on the development of adolescent autonomy, without understanding how warmth and relatedness may buffer the relation between mismatched
In the current study, it is expected that parental warmth will function as a moderator of the relation between mismatched parent-adolescent autonomy beliefs and adolescent
psychological adjustment. More specifically, it is hypothesized that mismatched parent-adolescent autonomy beliefs are related to more depressive symptoms and lower self-esteem when parental warmth is low. When levels of parental warmth are high, it is hypothesized that parent-adolescent differences in autonomy beliefs are not related to psychological adjustment. High levels of warmth may buffer the relation between autonomy beliefs and psychological adjustment in a number of ways. If adolescents perceive that their parents are warm, they may be more likely to interpret more traditional parental views on autonomy as being rooted in genuine concern, even if adolescents do not agree with their parents’ opinions. Discussions between parents and adolescents about the different autonomy beliefs they hold may also be more positive in the context of warm parent-adolescent relationships. It has also been suggested that warmth in the parent-child relationship functions as a quality that helps children internalize their parents’ choices as their own (Bao & Lam, 2008). In the context of the current study, it is possible that adolescents may come to view their parents’ autonomy beliefs as more acceptable when the parent-adolescent relationship is characterized by greater parental warmth.
The Mediating Role of Parent-Adolescent Conflict Intensity
The intensity of parent-adolescent conflict is examined as a mediator of the link between parent-child differences in autonomy beliefs and parent psychological adjustment, adolescent psychological adjustment, and parenting self-efficacy. Moderate levels of parent-child conflict are a normative part of adolescence (Smetana, Campione-Barr, & Metzger, 2006), and can serve an adaptive function (Greenberger & Chen, 1996). Rather than focusing on the amount of conflict as an indicator of relational distress, the emotional intensity of parent-child conflict is
assessed, since it is intensity that has been linked to maladaptive psychological outcomes (Laursen & Collins, 1994).
Based on clinical work with Cuban refugee families in the United States, Szapocznik and Kurtines (1993) suggested that many of the difficulties encountered by immigrant families are related to adolescents striving for greater autonomy than is considered developmentally appropriate by the parents. Within immigrant families, adolescents’ desire for autonomy is thought to clash with parents’ desire for family connectedness, leading to conflict within the family and a loss of intergenerational social support. In the context of the current study, Chinese Canadian children in immigrant families may acculturate towards Canadian culture faster than parents, while their parents may continue to hold beliefs more consistent with Chinese culture. Differences in beliefs about appropriate autonomy granting may create disagreements between adolescents and their parents about independent activities and peer relationships. That is, adolescents may desire the same level of autonomy as their non-immigrant peers, and worry about fitting in if they are more limited by their parents’ beliefs. However, as adolescents strive to gain the autonomy that their peers are allowed, their more traditional parents may resist, creating conflict within the family, which in turn could have a negative impact on the psychological functioning of family members.
High levels of intense disagreement may be a sign of compromised family cohesion. Family cohesion has been found to be important for psychological functioning across diverse cultures, especially when children are entering adolescence (Greenberger & Chen, 1996; Greenberger, Chen, Tally & Dong, 2000). It may be especially important for families in the current study, given the age of the adolescents, who range in age from 10 to 14 years. It is hypothesized that greater differences in age expectations for the granting of behavioural
autonomy are related to more intense disagreements. Furthermore, it is hypothesized that if parents perceive that they allow their children more input in the decision making process than their children perceive, more intense conflicts may result since parents may feel that they are already giving their children an acceptable amount of decision-making independence. In Chinese culture, where family harmony is emphasized and overt conflict is less acceptable, intense parent-child conflict resulting from differing autonomy beliefs may be particularly distressing, and be related to parents’ confidence in their role as the parent. It is hypothesized that parent-adolescent conflict intensity functions as a mechanism through which differences in autonomy beliefs are linked to psychological adjustment and parenting self-efficacy.
Hypotheses
It is hypothesized that relations between adolescents' autonomy beliefs and adolescent psychological adjustment depend on parents' autonomy beliefs, and that relations between parents' autonomy beliefs and parents' psychological adjustment and parenting self-efficacy depend on adolescents' autonomy beliefs. Two aspects of autonomy beliefs are measured: age expectations for the granting of adolescent autonomy and perceptions of adolescents’ decision making independence. When parents hold late expectations for autonomy, earlier adolescent expectations for autonomy are indicative of mismatched autonomy beliefs, and are expected to be related to poorer adolescent psychological adjustment (i.e., more depressive symptoms and lower self-esteem), whereas later adolescent expectations for autonomy are indicative of similar autonomy beliefs, and are expected to be related to fewer adolescent depressive symptoms and higher self-esteem. Similarly, when adolescents hold early expectations for autonomy, later parent expectations for autonomy are expected to be related to poorer parent psychological adjustment (i.e., more depressive symptoms) and lower levels of parenting self-efficacy, whereas
earlier parent expectations for autonomy are expected to be related to fewer parent depressive symptoms and higher levels of parenting self-efficacy.
It is also hypothesized that adolescent psychological adjustment, parent psychological adjustment, and parenting self-efficacy are adversely affected if parents and adolescents hold mismatched perceptions of adolescent decision making independence. When parents perceive that their adolescents are highly involved in making independent decisions, adolescent
perceptions of less decision making independence are indicative of mismatched autonomy beliefs, and are expected to be related to more adolescent depressive symptoms and lower
adolescent self-esteem. When adolescents perceive low levels of decision making independence, higher parent perceptions of decision making independence are expected to be related to more parent depressive symptoms and lower parenting self-efficacy.
Parental warmth is expected to moderate the relation between mismatched parent-adolescent autonomy beliefs and parent-adolescent psychological adjustment. It is hypothesized that mismatched expectations for autonomy and perceptions of decision making independence are only be related to adolescent depressive symptoms and adolescent self-esteem when adolescents perceive low levels of parental warmth. It is also hypothesized that the emotional intensity of adolescent conflict functions as a mediator in the relations between mismatched parent-adolescent autonomy beliefs and parent psychological adjustment, parent-adolescent psychological adjustment, and parenting self-efficacy, such that mismatched autonomy beliefs lead to intense parent-adolescent conflict, which in turn contributes to depressive symptoms, self-esteem, and parenting self-efficacy.
Method
Participants
The current study is part of the larger Intercultural Family Study (IFS), which focuses on the adaptation of immigrant families of Chinese ethnicity in Canada (Costigan & Dokis, 2006). Families (N = 95) were recruited from a mid-sized city and a large metropolitan area in Western Canada. Families were eligible if they identified their ethnicity as Chinese, if both parents were foreign-born, and if they had a child between 10 and 14 years. For the current study, six families in which one or both parents immigrated before the age of 18 were excluded. Of the remaining 89 families, two fathers declined participation. At the time of the study, mothers were on average 42.01 years old (SD = 4.25), fathers were on average 44.85 years old (SD = 4.64), and children were on average 11.89 years old (SD = 1.80). All of the participating families were two-parent families, and parents had been married an average of 17.09 years (SD = 4.20). The mean length of residence in Canada was 6.64 years (SD = 5.25) for mothers and 7.39 years (SD = 6.40) for fathers. In the majority of families, both parents emigrated from either the People’s Republic of China (42%) or Taiwan (45%). A small number of families emigrated from Hong Kong (7%), and in the remaining families (7%), each parent emigrated from a different sending region, including regions other than the three main regions of origin (e.g., Singapore). Slightly more than half of the children (55%) were foreign-born and immigrated at the age of 6 or older, while the remaining 45% of children were either Canadian born, or immigrated prior to the age of 6. The sample was approximately equal in terms of gender, with 55% girls and 45% boys. In terms of education level, 19% of parents completed elementary, junior high or high school, 31% completed vocational school or college, 33% completed a 4-year university degree, and 17% completed graduate or professional school.
Procedure
Families were recruited to the IFS primarily through a Chinese intercultural agency. Employees at the agency approached all potentially eligible families in order to explain the purpose of the study and the procedures should a family choose to participate. Contact information for interested families was then passed to the research team. A research assistant contacted each interested family, answered further questions, and scheduled an appointment for data collection. Additional families were recruited through Chinese religious services and referrals from participating families.
Most participating families completed the study in their own homes, although one family chose to come to the university. Two research assistants were always present during the study, at least one of which spoke the families’ native language. Each family member (father, mother, and target child) completed a package of self-report measures independently. Family members had the option of completing the measures in English or Chinese script. Most fathers (89.7%) and mothers (92.1%) completed the measures in Chinese script, whereas all participating children completed the measures in English. All English measures were translated into Chinese by a team of bilingual individuals from China, Taiwan, and Hong Kong. The Chinese versions were then back-translated by another team of bilingual individuals. In order to ensure accuracy, the back-translated English measures were compared to the original English measures, and the few discrepancies were resolved by discussion. Families received small monetary compensation ($30.00) for their time. The project received approval from the Human Research Ethics
Measures
Expectations for autonomy. Age expectations for the achievement of adolescent behavioural autonomy were assessed using the Teen Timetable (Feldman & Quatman, 1988). Adolescents were asked to indicate, “At what age do you expect to be able to do the following:” with regard to a list of 23 behaviours (‘Choose alone what clothes to buy,’ ‘Be able to come home as late as you want,’ ‘Smoke cigarettes’) on a five point scale (1 = before age 12, 2 = 12-14 years, 3 = 15-17 years, 4 = 18 years or older, 5 = not at all). Parents independently indicated the age at which they believed it was appropriate for their child to engage in the same list of behaviours (‘At what age do you expect your child to be able to do the following:’). Higher scores indicate later expectations for autonomy. The Teen Timetable was originally developed with European American and Asian American early adolescents (Feldman & Quatman, 1988). Subsequent studies consistently show that this measure has good reliability in samples of Asian adolescents (Fuligni, 1998; Juang, Lerner, von Eye, & McKinney, 1999). The Teen Timetable showed excellent reliability in the current sample (α =.85 for fathers, α =.85 for mothers, and α =.89 for children).
Differences in expectations for autonomy are evaluated using multiple regression. Although parent-adolescent differences are often assessed using simple difference scores (i.e., subtracting values representing adolescents’ autonomy expectations from values representing parent’s autonomy expectations), there are several problems associated with this method. When using this method to obtain difference scores, a value of zero is an indicator of similarity. However, information about the nature of the difference is lost (i.e., scores close to zero may reflect parent-child dyads with similar late expectations or similar early expectations).
range from 4 (early adolescent/late parent autonomy expectations) to -4 (late adolescent/early parent autonomy expectations). Such scales can be problematic for interpretation, since numerical values that comprise ranges are usually thought of as going from lowest to highest. Creating absolute difference scores avoids the issue of interpreting quadratic distributions, but means that information about the direction of the difference is also lost (i.e., parents may have later autonomy expectations than their children, or children may have later autonomy
expectations than their parents). Although the latter situation is counterintuitive, there is evidence that it occurs in a small proportion of immigrant families. For example, Juang and colleagues (1999) found that a small subsample of Asian American adolescents held later autonomy expectations than their immigrant parents. The direction of parent-adolescent discrepancies in autonomy expectations may have different implications for psychological adjustment. For example, if adolescents seek later autonomy than their parents expect, it would likely result in less strain on the parent-adolescent relationship, since parental authority is not being challenged, compared to the more common scenario in which children desire earlier autonomy than their parents.
An alternative to the difference score method is the use of hierarchical multiple regression analyses to determine interactions between parents’ and adolescents’ reports of expectations for autonomy. Using this technique, mean levels of expectations are taken into account: parents and adolescents are compared based on whether their expectations for
autonomy are early or late relative to the rest of the sample. Interaction terms are used to assess whether the relations between parents’ expectations for autonomy and psychological adjustment and parenting self-efficacy depend on adolescents’ expectations for autonomy. Although the specific within-family dyad differences are lost by using this variable-centred method, the use of
multiple regression is preferred because this technique preserves the continuous nature of
variables while allowing for the exploration of different combinations of early and late autonomy expectations.
Decision making independence. Perceptions of adolescent decision making
independence were assessed using the Decision Making Questionnaire (Dornbusch et al., 1985). Parents and adolescents were independently asked “Who decides about the following issues in your house?” with regards to 26 issues (“When your child/I can begin dating,’ ‘How late your child/I can stay out on weeknights.’) Possible responses were (1) adolescent alone, (2) adolescent and parent together, and (3) parents alone. Thus, perceptions of involvement in decision-making can be determined by obtaining the percentage of decisions made by adolescents only, parents only, and parents and adolescents together, from the perspective of each family member. The Decision Making Questionnaire is a widely used measure of the decision-making aspect of behavioural autonomy and displayed good reliability in the original sample of early and middle adolescents (Dornbusch et al., 1985). It has also shown good reliability in a sample of African American late adolescents (Smetana & Gettman, 2006). The Decision Making Questionnaire showed excellent reliability in the current sample (α =.87 for fathers, α =.87 for mothers, and α =.86 for children).
The current study focuses on parent’s and adolescents’ perceptions of the proportion of decisions made independently by adolescents (as opposed to decisions made jointly with parents, or made by parents alone). Higher scores indicate greater adolescent decision making
independence. In order to examine the correlates of adolescent and parent differences in perceptions of decision making independence, hierarchical multiple regression is used in the same manner as the analyses outlined for the Teen Timetable. Interaction terms are used to
assess whether the relations between parents’ perceptions of decision making independence and depressive symptoms and parenting self-efficacy depend on adolescents’ perceptions, and whether the relations between adolescents’ perceptions of decision making independence and adolescent psychological adjustment depend on parents’ perceptions.
Conflict intensity. The intensity of parent-adolescent conflict was assessed with the Issues Checklist (Robin & Foster, 1989). The measure consists of a list of 25 items that describe topics that could potentially cause parent-adolescent conflict, such as ‘Doing homework,’ ‘Going on dates,’ and ‘How money is spent.’ Parents and children indicated whether or not each issue had been a source of conflict in the past four weeks. For each issue of conflict that had arisen, participants were asked to indicate the level of anger intensity that typically accompanied the discussion of each topic on a scale of 1 (very calm) to 5 (very angry). Higher scores indicate greater conflict intensity. The Issues Checklist has been used in populations of diverse cultural and ethnic backgrounds, and has shown good reliability in populations of Chinese adolescents (Fuligni, 1998). The Issues Checklist showed excellent reliability in the current sample for children’s reports (α =.91 when reporting on conflict with fathers, α =.88 when reporting on conflict with mothers), fathers’ reports (α =.91) and mothers’ reports (α =.88).
Parental warmth. Adolescents’ perceptions of parental warmth were assessed using a 7-item measure developed for the larger IFS. Children were asked to indicate how frequently each parent displayed various acts of warmth (“When you have troubles, does your mom/dad comfort you?), on a scale of 1 (never) to 7 (always). Higher scores indicated higher levels of perceived parental warmth. The scale showed excellent reliability in the current sample for children’s reports of both fathers’ warmth (α = .85) and mothers’ warmth (α = .86). Previously published research using IFS data has analyzed parental warmth using this measure, and found that when
parental warmth is high, adolescents report stronger ethnic identity than their parents, and when parental warmth is low, parent-adolescent differences in Asian cultural values are greater (Costigan & Dokis, 2006a).
Depressive symptoms. Parents’ and adolescents’ depressive symptomatology was
assessed with the Centre for Epidemiological Studies Depression Scale (CES-D; Radloff, 1977). The CES-D is a 20-item self-report measure that was developed for use in community samples (Radloff, 1977). Participants were asked to rate the frequency with which they had experienced various feelings or displayed various behaviours in the past week on a 4-point scale, with a rating of 0 indicating rarely or none of the time and a rating of 3 indicating most or all of the time. A sample item reads “I was bothered by things that usually don’t bother me.” A higher overall score on the CES-D indicates more frequent and numerous depressive symptoms (Radloff, 1991). The CES-D shows good reliability and validity in populations of adolescents and adults (Radloff, 1977; Radloff, 1991). This measure has been widely used in cross-cultural samples, and has shown good reliability in samples of early Chinese American adolescents (Greenberger & Chen, 1996; Juang, Syed & Takagi, 2007), as well as late Asian American adolescents of various ethnicities, including Chinese American and Taiwanese American adolescents (Juang, Lerner, von Eye, & McKinney, 1999). The CES-D showed good reliability in the current sample (α = .88 for children, α = .88 for fathers, and α = .83 for mothers).
Self-esteem. Adolescent self-esteem was assessed with the Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1979). The measure consists of a list of 10 items, and respondents are asked if they strongly agree, agree, disagree, or strongly disagree with each item. A sample item reads “I am able to do things as well as most people.” Higher overall scores indicate higher
self-esteem. The RSE has been found to have good reliability in adolescents living in China (Shek, 2008), and showed excellent reliability in the current sample (α = .88).
Parenting self-efficacy. Parenting self-efficacy was assessed with the Parenting Self-Agency Measure (PSAM; Dumka et al., 1996 ). The measure consists of 10 items, and a sample item reads “I feel sure of myself as a mother/father.” Five of the items are positively worded and the remaining five are negatively worded, and reverse scored. Respondents are asked to answer on a 7-point scale, with 1 corresponding to ‘never’ and 7 corresponding to ‘always.’ Higher scores indicate greater parenting self-efficacy. The PSAM was developed on a sample of European American mothers and immigrant Mexican American mothers. It has not, however, been studied among parents of Chinese ethnicity. This measure showed good reliability for fathers in the current sample (α = .81), but mediocre reliability for mothers (α = .67).
Results
Preliminary Analyses
Exploratory factor analysis was conducted in order to ensure that the Teen Timetable, the Decision Making Questionnaire, and the Issues Checklist each assessed a singular construct in the current sample. The factor structures of each of these particular measures were of interest due to the wide variety of issues sampled by items (i.e., personal issues such as choice of clothing, monitoring issues such as level of supervision when spending time with friends, prudential issues such as substance use). Consistent with previous studies that have used these scales, evidence was found for a unitary factor structure. Items were dropped if factor loadings were less than .30 for two out of three family members. This method prevented the loss of too much data due to slight variations in factor loadings between family members, and maintained
the general structure of the questionnaires. According to these criteria, one item was dropped from the Teen Timetable (what is the appropriate age for a child to not tell their parents where he/she is going), and one item was dropped from the Decision Making Questionnaire (who decides what kind of job or career the child will have). No items were dropped from the Issues Checklist. The factor loadings, eigenvalues, and percentage of variance explained are
summarized for each family member in Appendix K (Teen Timetable), Appendix L (Decision Making Questionnaire) and Appendix M (Issues Checklist). These tables display factor loadings only for the factor accounting for the most variance. Bolded values indicate items with factor loadings of less than .30.
The means, standard deviations, ranges, and values of skewness and kurtosis of the main study variables are reported in Table 1. Transformations were applied to variables that showed significantly non-normal distributions. For the purpose of consistency of interpretation,
transformations were applied to each family member's score (father, mother, and adolescent) if one of the three scores on a given variable showed significant skewness and/or kurtosis. The degree of skewness and kurtosis for each variable was evaluated by dividing skewness and kurtosis values by their respective standard errors (i.e., comparing values to the z distribution). Values obtained through this comparison were considered to be problematic if they surpassed a critical value of z = 3.27, p <.001, a value which is sufficient in small to moderately sized samples (Tabachnick & Fidell, 2007). The method of transformation (e.g., square root, log, or inverse) for each variable was selected based on which transformation created the distribution closest to normality. A square root transformation was applied to the Decision Making Questionnaire, Issues Checklist, and Parental Warmth scale, and a log transformation was applied to the Centre for Epidemiological Studies Depression Scale. For the Rosenberg
Self-Esteem Scale, all transformations applied created a distribution even further from normality than the original distribution, and this variable was therefore left in its original metric.
Table 1
Descriptive statistics of main study variables.
M SD Range Skewness Skewness
/SESkewness Kurtosis Kurtosis/ SEKurtosis Teen Timetable Father 3.62 .41 2.60-5.00 .18 .71 .57 1.12 Mother 3.57 .43 2.60-4.50 -.11 -.41 -.20 -.39 Teen 3.34 .58 2.10-4.75 .22 .87 -.19 -.37 Decision Making Questionnaire Father 17.20 18.06 0.00-100.00 1.74 6.73* 4.82 9.43* Mother 17.89 16.64 0.00-80.00 1.31 5.13* 1.85 3.65* Teen 25.84 18.49 0.00-64.00 .06 .22 -1.29 -2.54 Conflict Intensity Father 1.44 .55 0.00-3.84 1.56 6.05* 6.47 12.66* Mother 1.52 .51 0.00-2.69 .45 1.75 .262 .52 Teen (Father) 1.46 .79 0.00-4.24 .91 3.56* 1.66 3.29* Teen (Mother) 1.56 .64 0.00-3.84 1.25 4.90* 2.10 4.14* Parental Warmth Teen (Father) 5.70 .97 3.29-7.00 -.73 -2.86 -.18 -.35 Teen (Mother) 5.89 .95 2.71-7.00 -.99 -4.39* .55 1.09 Depressive Symptoms Father 9.87 6.06 0.00-25.00 .58 2.23 -.27 -.54 Mother 10.64 6.98 0.00-36.00 1.38 5.40* 2.70 5.34* Teen 11.52 8.41 0.00-41.00 1.25 4.90* 1.73 3.43* Parenting Self-Efficacy Father 5.21 .69 3.50-6.90 .03 .12 -.10 -.20 Mother 5.35 .58 3.90-6.70 -.24 -.93 -.43 -.85 Self-Esteem Teen 3.17 .53 1.00-4.00 -.61 -2.40 1.78 3.51*
Note. SEskewness = .26, SEkurtosis= .51 for all family members.
Intercorrelations among main study variables are presented in Table 2. The intercorrelations involving the Decision Making Questionnaire, Issues Checklist, Parental Warmth, and CES-D scales were calculated using the transformed variables. An alpha level of .05 was used for all statistical tests Consistent with expectations, expectations for autonomy were significantly negatively associated with decision making independence for both mothers (r = -.40) and adolescents (r = -.52): later expectations for autonomy were associated with less decision making independence. The two measures of autonomy beliefs were not significantly related for fathers (r = -.06, p = .59). For the measure of expectations for autonomy, mothers’ reports were significantly correlated with fathers’ reports (r = .29) and adolescents’ reports (r = .34). Fathers’ expectations for autonomy and adolescents’ expectations for autonomy were not significantly correlated (r = .07, p = .51). Reports of decision making independence were significantly correlated between all family members (r = .49 for the father-mother dyad, r = .36 for the father-adolescent dyad, r = .35 for the mother-adolescent dyad). With regards to the outcome measures of interest, depressive symptoms were negatively correlated with parenting self-efficacy for both fathers (r = -.37) and mothers (r = -.46). Adolescent depressive symptoms were also negatively correlated with adolescent self-esteem (r = -.56).
Table 2
Intercorrelations between main study variables.
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. TTT 1. Father — 2. Mother .29** — 3. Teen .07 .34** — DMQ 4. Father -.06 -.11 -.09 — 5. Mother -.03 -.40*** -.22* .49*** — 6. Teen -.07 -.23* -.52*** .36** .35** — Conflict Intensity 7. Father -.01 .02 .01 -.02 -.01 -.00 — 8. Mother .09 .02 -.04 -.03 .00 .05 .40*** — 9. Teen (Father) -.12 .03 -.07 -.13 -.05 -.17 .11 .05 — 10. Teen (Mother) .03 -.15 -.16 .00 .07 -.06 -.03 .05 .80*** — Parental Warmth 11. Teen (Father) .04 .09 .25* -.16 -.18 -.16 .12 -.21* -.07 -.23* — 12. Teen (Mother) .11 .11 .35** -.20 -.22* -.13 .05 -.20 -.13 -.25* .73*** — Depressive Symptoms 13. Father .05 -.03 .05 .09 -.01 -.10 .01 -.01 -.10 -.08 -.08 .04 — 14. Mother .28** .21* .04 -.08 -.09 -.06 -.14 .28** -.07 -.04 -.08 -.16 .25* — 15. Teen -.15 -.09 -.12 -.07 .01 .04 -.10 .17 .30** .39*** -.37*** -.39*** .07 .04 — Parenting Self-Efficacy 16. Father -.22* -.01 .05 -.31** -.09 -.15 -.15 -.18 .05 .05 .14 .10 -.37** -.15 .04 — 17. Mother -.32** -.15 -.05 .03 -.11 -.08 -.17 -.44** .02 .04 .05 .10 .01 -.46*** .00 .22* — Self-Esteem 18. Teen .15 .06 -.08 .01 -.12 -.01 -.05 -.16 -.21* -.26* .36** .30** -.09 .06 -.56*** .05 .01 —
Note. TTT = Teen Timetable, DMQ = Decision Making Questionnaire * p < .05, **p < .01, ***p < .001
In general, correlations among measures of autonomy beliefs and measures of
psychological adjustment and parenting self-efficacy for individual family members were not significant, with the exception of associations between fathers’ expectations for autonomy and fathers’ parenting self-efficacy (r =-.22), fathers’ reports of decision making independence and fathers’ parenting self-efficacy (r =-.31), and mothers’ expectations for autonomy and mothers’ depressive symptoms (r = .21). This pattern suggests parents who expect or perceive greater adolescent autonomy report poorer psychological adjustment and lower feelings of parenting self-efficacy, compared to parents who expect or perceive less adolescent autonomy. The direct relations between adolescents’ autonomy beliefs and adolescent psychological adjustment were not significant in the current sample.
Demographic Variables
The relations between the dependent variables and a series of demographic variables were examined prior to the main analyses, in order to assess control variables. Relations between dependent variables and categorical demographic variables (parent education,
adolescent gender, child birth order, and adolescent generational status) were examined using a series of one-way ANOVA analyses, and relations between dependent variables and continuous demographic variables (parent length of residence, adolescent age) were examined with Pearson product-moment correlations. Adolescent gender and maternal parenting self-efficacy were significantly related, F(1,87) = 4.81, such that mothers with adolescent daughters reported higher levels of parenting self-efficacy (M = 5.47, SD = .57) compared to mothers with adolescent sons (M = 5.21, SD = .57). No other significant relations between demographic variables and the
dependent variables were found. Therefore, adolescent gender was included as a control variable in all analyses.
Autonomy Beliefs, Psychological Adjustment, and Parenting Self-Efficacy
The first set of regression analyses examined whether adolescents’ autonomy beliefs interacted with parents' autonomy beliefs to predict psychological adjustment and parenting self-efficacy. All predictor variables were mean centered prior to creating interaction terms, as recommended by Aiken and West (1991). The data were analyzed using hierarchical
regression analyses. Adolescent gender was entered as a control variable in Step 1, followed by the main effects of each family member's autonomy beliefs in Step 2, with the interaction terms representing the interaction of autonomy beliefs within each parent-adolescent dyad entered in Step 3. A significant interaction term indicates that the relation between adolescents' autonomy beliefs and adolescent psychological adjustment is moderated by parents' autonomy beliefs, and that the relation between parents' autonomy beliefs and parent psychological adjustment and parenting self-efficacy is moderated by adolescents' autonomy beliefs. Separate regression analyses were conducted for each predictor (expectations for autonomy, decision making independence), since each assessed different aspects of autonomy beliefs. Expectations for autonomy assessed both current and future age expectations for gaining autonomy, whereas decision making independence assessed perceptions of current adolescent decision making independence.
Expectations for autonomy. First, the relations between expectations for autonomy, as measured by the Teen Timetable, and psychological adjustment and parenting self-efficacy were examined. The standardized regression coefficients, R2values, and F values are reported in
Table 3. The interactions between parents’ expectations for autonomy and adolescents’ expectations for autonomy did not predict fathers’, mothers’ or adolescents’ depressive symptoms, or adolescent self-esteem. The interaction between fathers’ expectations for autonomy and adolescents’ expectations for autonomy did not predict fathers’ parenting self-efficacy.
Table 3
Hierarchical regression model for relation between expectations for autonomy and depressive symptoms, parenting self-efficacy, and adolescent self-esteem.
Father Mother Adolescent
Variables Depressive Symptoms n = 87 Parenting Self-Efficacy n = 87 Depressive Symptoms n = 87 Parenting Self-Efficacy n = 87 Depressive Symptoms n = 87 Self-Esteem n = 87 1.Control variable (β) Adolescent gender .10 .03 -.18 .24* .03 .17 2.Main effects (β) Father .08 -.21 .23* -.31** -.15 .17 Mother -.06 .03 .22 -.08 .01 .01 Adolescent .01 .06 -.10 -.10 -.16 -.04 3.Interactions (β) Father x Adolescent .04 .07 -.01 -.17 -.07 .11 Mother x Adolescent .12 -.00 .13 .24* .16 -.12 R2 .03 .06 .15 .23 .06 .08 F .47 .81 2.28* 4.07** .78 1.08
Note. Adolescent gender was coded as 0 (male) and 1 (female).
