Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Relations of autonomy and relatedness to school functioning and psychological adjustment during adolescence Buote, Carol Anne 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-565157.pdf [ 10.61MB ]
Metadata
JSON: 831-1.0053902.json
JSON-LD: 831-1.0053902-ld.json
RDF/XML (Pretty): 831-1.0053902-rdf.xml
RDF/JSON: 831-1.0053902-rdf.json
Turtle: 831-1.0053902-turtle.txt
N-Triples: 831-1.0053902-rdf-ntriples.txt
Original Record: 831-1.0053902-source.json
Full Text
831-1.0053902-fulltext.txt
Citation
831-1.0053902.ris

Full Text

R E L A T I O N S O F A U T O N O M Y A N D R E L A T E D N E S S T O S C H O O L F U N C T I O N I N G A N D P S Y C H O L O G I C A L A D J U S T M E N T D U R I N G A D O L E S C E N C E by C A R O L A N N E B U O T E B . A . , The University of Western Ontario, 1973 M . E d . , The University of Western Ontario, 1980 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F E D U C A T I O N in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department of Educational and Counselling Psychology, and Special Education W e accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A October 2000 ® Carol Anne Buote 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and* study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of Vancouver, Canada Department of Date .fifrj. ID,£000 DE-6 (2/88) A B S T R A C T One criticism of previous work in the field of adolescent development has been the paucity of research examining the unique and combined contributions of different developmental contexts on adolescents' functioning. In an attempt to address this issue, the current study examined adolescents' perceptions of autonomy and relatedness within parent, peer, and school contexts in relation to school functioning and psychological adjustment. Adolescents ( N = 478) in Grades 8, 9, and 11 completed self-report questionnaires assessing feelings about their relationships with parents and peers, and perceptions of school. Teachers completed ratings of adolescents' strengths and competencies. Academic achievement was assessed using end of year school grades. Results revealed several significant gender and grade differences. Whereas girls reported greater deidealization of their parents and peers, and higher quality of attachment to peers than did boys, boys- reported being less dependent on their peers than did girls. Overall , adolescents in grade nine were more dependent on their peers and reported more trust and communication in their peer relationships than did adolescents in grade eight. Correlational results indicated that school functioning was positively associated with school autonomy, parental attachment, peer attachment and school belonging, and that problems in psychological adjustment were negatively associated with peer autonomy, school autonomy, parental attachment, peer attachment, school belonging, and positively associated with parental autonomy. Results of the multiple regression (ii) analyses indicated that autonomy and relatedness variables accounted for significant amounts of variance in G P A , teacher-rated school competencies, internalizing problems, and externalizing problems. Analyses also revealed variables which uniquely predicted areas of functioning across contexts and gender. This cross-sectional study provides new theoretical insights regarding relations of autonomy and relatedness to school functioning and psychological adjustment during adolescence across multiple contexts. The findings contribute to a more thorough understanding of the dimensions of autonomy and relatedness that may have important implications for educators and parents of adolescents for improving educational practice and for promoting school success and positive adjustment. (iii) T A B L E O F C O N T E N T S Page A B S T R A C T i i L I S T O F T A B L E S v i L I S T O F F I G U R E S v i i i A C K N O W L E D G E M E N T S ix C H A P T E R 1 Introduction 1 Definition of Terms 4 Significance of the Study 10 2 Review of the Literature : 11 Overview 11 Toward A Conceptual Framework For Research on Autonomy and Relatedness in Mult iple Contexts 11 Autonomy and Relatedness in the Parental Context 15 Autonomy and Relatedness in the Peer Context 27 Autonomy and Relatedness in the School Context 35 Statement of the Problem 47 Hypotheses.... 49 3 Method 56 Participants 56 Measures of Autonomy, Relatedness, and Psychological Adjustment.. 58 Measures of Adolescents' School Functioning 69 Procedures 71 4 Results < 75 Preliminary Analyses 75 Gender and Grade Differences in Adolescents' Perceptions of Autonomy and Relatedness 79 Relations of Autonomy and Relatedness to School Functioning and Psychological Adjustment 98 Regression Analyses Examining Autonomy and Relatedness in the Contexts of Parents, Peers, and School, as Predictors of School Functioning and Psychological Adjustment 116 (iv) 5 Discussion 132 Gender and Grade Differences in Adolescents' Perceptions of Autonomy and Relatedness 133 Relations of Autonomy and Relatedness to School Functioning and Psychological Adjustment 146 Implications and Importance of the Study to Education 162 Strengths and Limitations of the Study and Future Directions.. 166 Concluding Remarks ; 170 References 172 Appendixes 187 A . Emotional Autonomy Scale ( E A S ) . . . 187 B . Emotional Autonomy Scale - Peers (EASP) 190 C . Perceived Control at School Scale (PCSS) 192 D . Inventory of Parent and Peer Attachment (IPPA) 195 E . Psychological Sense of School Membership (PSSM) 201 F . Youth Self-Report ( Y S R ) . . . . 204 G . Comparison of Adolescents on Problem Behaviour Scores on the Youth Self-Report (YSR) 205 H . Teacher-Child Rating Scale (T-CRS) 207 I. Correlations Between Students' and Teachers' Ratings of Problem Behaviours for Boys and Girls 208 J . Student Recruitment Letter 210 K . Parent Permission Letter and Consent Form 212 L . Student Consent Form 215 M . Questionnaire Package Cover Page.. 217 - N . Student Identification Form 220 O. Intercorrelations Among Measures for Grades 8, 9, and 11 222 (v) L I S T O F T A B L E S Table Page 1 Means, Standard Deviations, Skewness, Kurtosis, and Range for al l Measured Variables 77 2 Gender and Grade Means and Standard Deviations for Parental Autonomy. 82 3 Gender and Grade Means and Standard Deviations for Peer Autonomy. 85 4 Gender and Grade Means and Standard Deviations for Parental Attachment 89 5 Gender and Grade Means and Standard Deviations for Peer Attachment 92 6 Gender and Grade Means and Standard Deviations for School Autonomy and School Belonging 95 7 Intercorrelations Among Measures for Boys and Girls 100 8 Correlations of Parental Autonomy to School Functioning and Psychological Adjustment for Boys and Girls 104 9 Correlations of Peer Autonomy to School Functioning and Psychological Adjustment for Boys and Girls 106 10 Correlations of Parental Attachment to School Functioning and Psychological Adjustment for Boys and Gir ls 108 11 Correlations of Peer Attachment to School Functioning and Psychological Adjustment for Boys and Gir ls 110 12 Correlations of School Autonomy and School Belonging to School Functioning and Psychological Adjustment for Boys and Girls 112 13 Summary of Hierarchical Mult iple Regression Analysis Predicting School Competencies for Boys 119 14 ^ Summary of Hierarchical Mult iple Regression Analysis Predicting School Competencies for Girls 121 (vi) 15 Summary of Hierarchical Mult iple Regression Analysis Predicting G P A for Boys 123 16 Summary of Hierarchical Mult iple Regression Analysis Predicting G P A for Gir ls . 124 17 Summary of Hierarchical Mult iple Regression Analysis Predicting Internalizing Problems for Boys 125 18 Summary of Hierarchical Mult iple Regression Analysis Predicting Internalizing Problems for Girls 127 19 Summary of Hierarchical Mult iple Regression Analysis Predicting Externalizing Problems for Boys 128 20 Summary of Hierarchical Mult iple Regression Analysis Predicting Externalizing Problems for Girls 130 (vii) L I S T O F F I G U R E S Page Figure 1. A conceptual framework and purpose of the study 17 ^ Figure 2. Measures of autonomy, relatedness, school functioning, and psychological adjustment 60 (viii) A C K N O W L E D G E M E N T S I gratefully acknowledge the contributions of so many people, whose invaluable guidance have assisted me in completing this research. I am indebted to the principals, teachers, and students who participated in this project. Teachers were more than generous of their time for allowing me to conduct this study in their classrooms at a particularly busy time of the school year. I thank al l the students who wil l ingly consented to participate. Their interest in being part of real research was greatly appreciated. I am deeply grateful to each of my committee members for their contribution and input. M y advisor, D r . Kimberly A . Schonert-Reichl, has directed and guided me with this dissertation to help make it the best it can be. I have gained much from her knowledge and scholarly contribution to the research procedure. The helpful comments of D r . Deborah Butler and D r . B i l l M c K e e on previous versions of this thesis and during the early phase of this work, have led to important improvements. I owe so much to those who have endured with me throughout this process. I would like to express my sincere thanks to Dr . Barbara Turnbull for her generous help and suggestions on the statistical aspects of this research. I would also like to thank my friend, T i l a MacDonald for her support and assistance with coding and checking data, and my friend Cathy Gifford for her help with data collection. I am deeply appreciative of the endless encouragement and assistance I received from my husband, John, throughout this endeavour, and whose support and confidence in 'me helped me realize my goal. A n d I thank Hoover and Scooter who have always been there to remind me when to stop and play. (ix) 1 C H A P T E R 1 Introduction Autonomy and relatedness have been identified as two salient dimensions of adolescent development that have figured extensively throughout much of the research literature on adolescence (Douvan & Adelson, 1966; H i l l , 1993; H i l l & Holmbeck, 1986; Hodgins, Koestner, & Duncan, 1996; Ryan & Powelson, 1991; Silverberg & Gondoli , 1996). In recent years, there has been a burgeoning research literature exploring the role of autonomy and relatedness in the academic and behavioural functioning of adolescents (Armsden & Greenberg, 1987; Barber, 1997; Barber & Olsen, 1997; Eccles et a l . , 1993; Eccles & Midgley , 1989; Goodenow, 1993a, 1993b, 1994; Paterson, Pryor, & Fie ld , 1995; Ryan & Powelson, 1991; Ryan, Stiller, & Lynch , 1994; Taylor & Adelman, 1990). In particular, Ryan and Powelson (1991) argue that,needs for autonomy and relatedness are fundamental to successful school functioning and "represent significant influences on the affective and cognitive outcomes of education" (p. 64). Moreover, the negotiation of autonomy and relatedness has been a main issue in theories regarding the individuation process and adolescent-parental relationships (Bios, 1967; Chen & Dornbusch, 1998; Coll ins, Laursen, Mortensen, Luebker, & Ferreira, 1997; Eccles, Early, Frasier, Belansky, & McCarthy, 1997; Grotevant & Cooper, 1986; Havighurst, 1952, 1972; H i l l & Holmbeck, 1986; Josselson, 1980). Accordingly, current theory and research suggest that autonomy and relatedness play critical roles in school competence as well as general psychological adjustment during adolescence (e.g., Chen & Dornbusch, 1998; Eccles & Midgley, 1989; Eccles et a l . , 1993; 2 Ryan & Grolnick, 1986). Research findings are in concert in suggesting that, during adolescence, autonomy and relatedness are important predictors of academic, social, and emotional adjustment (e.g., Eccles et a l . , 1997). For instance, the early adolescent period coincides with several age-related declines in self-esteem, academic and behavioural functioning associated with school transitions, and with needs for more independence and autonomy (e.g., Eccles, et a l . , 1993; Eccles & Midgley, 1989; Roeser, Midgley , & Urdan, 1996; Savin-Williams & Small , 1986; Turner, Irwin, Tschann, & Mil ls te in , 1994; Wenz-Gross, Siperstein, Untch, & Widaman, 1997). Moreover, those students who experience difficulties in school and may be "at risk" for dropping out of school, are l ikely to have poor interpersonal relationships with others, and weak attachments to family and school (Allen, Aber, Leadbeater, 1990; Kazdin, 1995; Parker & Asher, 1987). Indeed, researchers and theorists agree that adolescence is a crucial period in development in which biological, cognitive, emotional, and social contextual changes (e.g., times of school transitions, changes in the quality of relationships with parents and peers) pose challenges and stresses for adolescents that play an important role in an adolescent's adjustment (e.g., Berndt, 1979; Brown, 1990; Eccles et a l . , 1993; Fasick, 1984; Ful igini & Eccles, 1993; Isakson & Jarvis, 1999; Laible, Carlo & Rafaelli, 2000; Wentzel, 1996). There are three contexts, namely schools, families, and peers, that have been identified by researchers as the major social arenas in which adolescents invest time and commitment and develop attitudes and beliefs that shape maturation and that are critical to emotional health development and adjustment in general (e.g., Blum & Rinehart, 1997; Brown, 1990; Eccles et a l . , 1997; Greenberger, Steinberg, & Vaux, 1982; H i l l , 1993; 3 Isakson & Jarvis, 1999; Minuchin & Shapiro, 1983; Steinberg, 1993). The research on social contexts suggests that adolescents' experiences at school, with parents, and with peers, contribute to a developing sense of autonomy and relatedness that effect learning, motivation, psychological well-being, and achievement at school. Nevertheless, very few studies exist that have directly examined both the unique and combined effects of adolescents' experiences of autonomy and relatedness in relation to school functioning and psychological adjustment across multiple contexts during adolescence, despite the emergent literature highlighting the importance of autonomy and relatedness in education and development (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Roeser, 1998). Instead, most researchers have focused their attention on examining selected aspects of autonomy or relatedness within a specific context, such as the family (e.g., Al len & Hauser, 1996; Fuhrman & Holmbeck, 1995), or school (e.g., Cotterell, 1992; Chen & Dornbusch, 1998), or peers (e.g., Gavin & Furman, 1989) and have overlooked the concurrent relations of autonomy and relatedness in parent, peer, and school contexts. Given that adolescents who are at risk for school failure and serious maladjustment in adulthood are a growing concern of school professionals, parents, and the community (Jenson, Walker, Clark, & Kehle, 1991; Reid & Patterson, 1991; Saleh, 1991; Sprick & Nolet, 1991), researchers are now calling for studies that examine the concurrent relations of autonomy and relatedness on academic functioning and psychological adjustment during adolescence (Barber & Olsen, 1997; Eccles et a l . , 1997; Minuchin & Shapiro, 1983). The present study is a response to those calls and is designed to contribute to a better understanding of the relations of adolescents' experiences of autonomy and relatedness to 4 development across multiple contexts. In the current study I examine adolescents' perceptions of autonomy and relatedness systematically in parent, peer, and school contexts in order to determine the manner in which these two dimensions of adolescent development are associated with school functioning and psychological adjustment during adolescence. Definit ion of Terms The following definitions are put forward in order to provide the reader with an explanation of how specific terms are intended to be understood and used within the framework of the current study. Autonomy Autonomy is a term that has been used to indicate a variety of different ideas about freedom, independence-striving, self-direction, self-governance, and general control beliefs (Grolnick, Ryan & Deci , 1991; see H i l l & Holmbeck, 1986 for a review; Ryan & Powelson, 1991; Ryan et a l . , 1994; Silverberg & Gondoli , 1996). Some researchers have used the term autonomy to convey the notion of an individual's capabilities to take responsibility and to make decisions for themselves while maintaining relationships with significant others (e.g., Coll ins, Laursen, et a l . , 1997; Crittenden, 1990; H i l l & Holmbeck, 1986; Turner et a l . , 1994). In the present study, autonomy was uniquely defined within the contexts of parents, peers, and school. Fol lowing are the definitions of autonomy utilized in the present study. Parental autonomy. Autonomy in relationships with parents, or parental autonomy, is a term that has been broadly defined in the adolescent research as emotional separateness, emotional distancing, disengagement from parents, and detachment from parents (Frank, Pirsch, & Wright, 1990; Herman, Dornbusch, Herron, & Herting, 1997; Lamborn & 5 Steinberg, 1993; Papini & Roggman, 1992, Ryan & Lynch , 1989; Steinberg & Silverberg, 1986; Turner, Irwin, Tschann, & Mil ls tein, 1993). In the present study, parental autonomy refers to a sense of emotional separateness, and is based on Steinberg and Silverberg's (1986) conceptualization of Blos 's (1967) theoretical perspective on individuation and the development of a distinct sense of self. According to Bios ' (1967) theory of individuation-separation, differentiation is achieved when adolescents relinquish childish dependencies and conceptualizations of their parents and develop their individuality and a distinct sense of self or separateness while still maintaining their relationships with their parents. Thus, for the present investigation, parental autonomy was operationalized to indicate the degree to which an adolescent distinguishes or differentiates himself or herself from their parents (i.e., the extent which adolescents deidealize their parents, are less dependent on their parents, and feel individuated in their relationships with parents). In this study, parental autonomy was assessed using Steinberg and Silverberg's (1986) Emotional Autonomy Scale (EAS) . Peer autonomy. Autonomy in relationships with peers, or peer autonomy, is a term that researchers have used to represent concepts such as psychological control (e.g., Barber & Olsen, 1997), the extent of conformity to peers (e.g., Berndt, 1979), and resistance to peer pressure (e.g., Steinberg & Silverberg, 1986) rather than a direct assessment of emotional separateness. In the present study, peer autonomy was designed to parallel Steinberg and Silverberg's (1986) conceptualization of autonomy in relationships with parents. More specifically, the definition of peer autonomy in the present study is taken from Josselson's (1980) conceptualization of autonomy from peers. According to Josselson (1980), during middle adolescence, adolescents look for ways to distinguish themselves as 6 being different from their peers in order to achieve a sense of separateness from them. Thus, in the present study, peer autonomy was operationalized to indicate the degree to which an adolescent distinguishes or differentiates himself or herself from their peers (i.e., the extent to which adolescents deidealize their peers, are less dependent on their peers, and feel individuated in their peer relationships). For the current study, peer autonomy was assessed via the Emotional Autonomy Scale - Peers (EASP) . The E A S P was developed by adapting questions on Steinberg and Silverberg's (1986) Emotional Autonomy Scale to assess autonomy in relationships with peer rather than with parents.1 School autonomy. In the present study, I adopted Adelman and colleagues's (1986) conceptualization of school autonomy as perceptions of control over school outcomes (i.e., events, tasks, situations, rules) (Adelman, Smith, Nelson, Taylor, & Phares, 1986). Thus, school autonomy reflects beliefs about how much personal control students feel they have in school, such as being able to make choices and take part in making decisions (Adelman et a l . , 1986; Adelman & Taylor, 1990; Eccles, et a l . , 1993; Heavey, Adelman, Nelson, & Smith, 1989; Smith, Adelman, Nelson, Taylor, & Phares, 1987; Taylor & Adelman, 1990). It should be noted that, whereas peer autonomy and parental autonomy are intended to be parallel measures, school autonomy is uniquely operationalized for use in this study to indicate the degree to which adolescents feel they have some personal control or influence over school-related situations. School autonomy was assessed using the Perceived Control at School Scale (PCSS; Adelman et a l . , 1986). 'Details of the Emotional Autonomy Scale - Peers are provided in Chapter Three. 7 Relatedness Relatedness is a term commonly used throughout the research literature to refer to feelings of close interpersonal attachments and bonds with others that are thought to contribute to an individual's emotional well-being and positive adjustment (Ryan & Powelson, 1991). The concept of relatedness reflects the idea of secure and satisfying connections with others that are established through an individual's needs for contact, support, and validation (Goodenow, 1993a, 1994; O'Brien, 1989; Roeser et a l . , 1996; Ryan & Powelson, 1991). In the school context, relatedness refers to the "social bonding which connects the student to the school, the sense of attachment and commitment felt by students who believe that others in the school, both peers and adults, care about them, respect them, and are interested in their welfare" (Goodenow, 1994, p. 3). In the present study, relatedness was operationalized in parent and peer contexts to indicate the degree to which an adolescent feels a sense of emotional connection and feels accepted, supported, and cared for in his or her relationships with parents and peers. Relatedness was operationalized for use in the school context to indicate the degree to which an adolescent feels connected with, or attached to, his or her school and feels validated and respected within school. Relatedness was assessed in this study using two measures (1) Armsden and Greenberg's (1987) Inventory of Parent and Peer Attachment ( IPPA), and (2) Goodenow's (1993a) Psychological Sense of School Membership (PSSM) questionnaire. School Functioning School functioning is a general term that refers to students' skills of an academic, motivational, and interpersonal nature associated with his or her learning and behaviour at 8 school. Throughout the research literature, students' school functioning has been assessed through various indices connected with learning, ability to cope with failure, academic engagement, and achievement or grades (e.g., Eccles et a l . , 1997; Ryan et a l . , 1994; Wentzel, 1993). In this study, two dimensions of school functioning were assessed, namely school competencies and academic achievement. School competencies. In the present study, school competencies were operationalized to indicate the degree to which an adolescent exhibits abilities in school of a social, behavioural, and academic nature. School competencies were assessed using teacher-ratings of students' strengths in four areas (Frustration Tolerance, Social Skil ls , Task Orientation, and Peer Social Skills) obtained from the Teacher-Child Rating Scale (T -CRS; Hightower et a l . , 1986). Academic achievement. Academic achievement was operationalized in the present study by using academic grade point average ( G P A ) . G P A has been utilized by numerous researchers as a means for measuring a student's academic achievement in school (e.g., Berndt, Laychak, & Park, 1990; Foley & Epstein, 1992; Grolnick et a l . , 1991; Roeser & Eccles, 1998; Wentzel, 1993; Wentzel & Caldwell , 1997). In this study, a composite G P A score was calculated from students' end of the year report card grades that was based on the average grade obtained in four core academic classes (i.e., english, mathematics, science, and one of history, geography or senior social science course). The average grade was converted to a 13-point grading scale (e.g., Roeser et a l . , 1997; Wentzel, 1993; Wentzel & Caldwell , 1997). 9 Psychological Adjustment In the present study, I use the term psychological adjustment to refer to the quality of adolescents' social and emotional functioning. This definition is in accordance with the way in which other researchers in the field of adolescent development refer to psychological and social functioning (e.g., Barber & Olsen, 1997). In the research literature, psychological adjustment is typically represented through an individuals' socioemotional behaviours and has included a wide range of measures of problems associated with self-esteem, self-concept, feelings of depression, anger, internalizing and externalizing behaviours, self-restraint, anxiousness and loneliness in its assessment (e.g., Achenbach, 1991; Barber & Olsen, 1997; Conger, Conger, & Scaramella, 1997; Larose & Boiv in , 1998; Larson, 1997; Merre l l , 1989; Quay & Peterson, 1987; Roeser & Eccles, 1998; Wentzel & Feldman, 1996). In general, better psychological adjustment is indicated by fewer symptoms or low level manifestations of negative problems (e.g., Larson, 1997; Wenz-Gross et a l . , 1997), or the absence of psychological problems (Kazdin, 1993). In this study, the term psychological adjustment was operationalized to indicate the degree to which an adolescent exhibits internalizing and externalizing problem behaviours. More specifically, individuals with better psychological adjustment were those who reported fewer internalizing and externalizing problem behaviours. Psychological adjustment problems were assessed using self-reports of internalizing problem behaviours (anxiety, withdrawal, depression, and somatic complaints) and externalizing problem behaviours (delinquency and aggressiveness) on the Youth Self-Report ( Y S R ; Achenbach, 1991). 10 Significance of the Study In the present study, I seek to understand the extent to which adolescents' experiences of autonomy and relatedness in parent, peer, and school contexts are associated with educational and behavioural outcomes during adolescence. The present investigation w i l l contribute new information to the research literature which attempts to discern the way in which different social contexts of parents, peers, and school are associated with adolescents' academic and emotional functioning (e.g., Barber & Olsen, 1997; Feldman & Elliott , 1990). This cross-sectional study, including students in grades eight, nine, and eleven, has the potential to provide new theoretical insights regarding relations of autonomy and relatedness to school functioning and psychological adjustment during adolescence across multiple contexts. Moreover, it is hoped the findings of this investigation w i l l provide a more thorough understanding of autonomy and relatedness during adolescence, and thus provide future direction for educational practice for the purpose of improving school performance of adolescents. 11 C H A P T E R 2 Review of the Li tera ture Overview The review of the literature is organized into three main sections. First, I briefly introduce some of the research and theory on autonomy and relatedness during adolescence. In this section I also present the central themes associated with autonomy and relatedness across different developmental contexts. Next, I give a detailed review of several studies in which autonomy and relatedness have been investigated in the contexts of parents, peers, and school. In this section, I describe the major findings from studies in the area and discuss the specific limitations of that research. Finally, I present the problem statement and hypotheses that directed the current investigation. | Toward A Conceptual Framework For Research on Autonomy and Relatedness in Mult iple Contexts Amidst the concern of researchers for the emotional health and well-being of adolescents in relation to schooling, there have been increasingly more studies examining varying aspects of children's and adolescents' school experiences (i.e., opportunities for student decision making and choice; evaluation and teaching practices), and interpersonal relationships (i.e., feelings of belongingness, perceived social support from parents, teachers, and peers) in association with academic performance and social adjustment (e.g., Eccles, Lo rd , & Midgley, 1991; Eccles & Midgley, 1989; Goodenow, 1993a; Kasen, Johnson, & Cohen, 1990; Roeser etal. ,-1996; Roeser & Eccles, 1998; Wentzel, 1994, 1997, 1998). These researchers have focused on the links between adolescent development and the ways in 12 which school contexts and socialization experiences meet adolescents' psychological needs for autonomy and relatedness. For example, the longitudinal work of Eccles and her colleagues (Eccles, Lo rd , et a l . , 1991; Eccles et a l . , 1993; Eccles & Midgley, 1989; Eccles, Midgley & Adler , 1984) has provided considerable evidence to suggest that the fit or match between the developmental needs of adolescents and their environment is important to school adjustment. Their theoretical approach has been used to explain some of the difficulties adolescents experience in junior high or middle school environments. Specifically, Eccles and colleagues argue that a developmental stage-environment mismatch places adolescents at risk for negative motivational, behavioural, and psychological outcomes at school, such as poor school performance, greater school misconduct and negative relationships with teachers (Eccles, Lo rd , et a l . , 1991; Eccles et a l . , 1993; Eccles & Midgley, 1989; Roeser & Eccles, 1998). Based on longitudinal studies, these researchers claim that many of the negative outcomes (e.g., decreases in motivation, declining self-esteem) that predispose adolescents for declines in academic and behavioural functioning in school, result because adolescents' experiences in schools and families do not match their physiological, psychological, and cognitive developmental needs (e.g., Eccles et a l . , 1993). Moreover, the developmental needs of adolescents that have been suggested from that literature include needs for autonomy in relation to schools and families, and needs for emotionally supportive relationships with parents and teachers. It should be noted, however, that much of the research on developmental stage-environment fit has focused primarily on the developmental needs of early adolescents in family and school contexts and thus has failed to address other important 13 contexts of adolescents, such as peers. A second body of research complementary to that of Eccles and her colleagues has drawn attention to the influence of different social developmental contexts on individuals' experiences of autonomy and relatedness in association with learning and development. In a review of the research on autonomy and relatedness in relation to school-related functioning, Ryan and Powelson (1991) put forth the argument that school environments that provide for relatedness and support for autonomy are essential for positive school functioning and psychological well-being. That is, these researchers posit that psychological needs for autonomy and relatedness are important to educational processes (see Ryan & Powelson, 1991 for a review). Ryan, Stiller, and Lynch (1994), for instance, have extended the study of autonomy and relatedness by examining how early adolescents' relationships with parents, teachers, and peers predict school-related functioning and adjustment. Ryan and his colleagues found that early adolescents' "representations of relationships with teachers, parents, and peers have direct significance for adaptive functioning in school and for self-esteem in early adolescence" (Ryan et a l . , 1994, p. 243). Specifically, Ryan et al. 's research findings indicated that the quality of interpersonal relationships with parents and teachers was associated with educational outcomes and predictive of better school motivation and adjustment, whereas relatedness to friends was associated with greater self-esteem. Moreover, these researchers' findings suggest that relationships with parents, teachers, and peers are differentially related to school-related functioning and self-esteem during early adolescence. Researchers have recently begun to explore various dimensions of autonomy and 14 relatedness both within and across multiple social and developmental contexts, such as families, peers, siblings, neighbourhoods, and schools in relation to adolescent functioning (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Isakson & Jarvis, 1998). For example, Eccles et a l . , (1997) investigated adolescents' experiences of connection, behavioural regulation, and support for autonomy in association with academic performance, feelings of depression, and antisocial behaviour. These researchers found evidence indicating that the varying dimensions of autonomy and relatedness in different social contexts were related to adolescent functioning in important ways. More specifically, experiences of connection, regulation, and autonomy in the family, peer, and school contexts predicted better school performance and fewer problems. Thus, it can be surmised that a research approach that utilizes a multicontextual design is useful for disentangling the manner in which experiences of autonomy and relatedness in different contexts relate to school functioning and psychological adjustment during adolescence. There is, however, a paucity of research to date that has examined al l three contexts of adolescents collectively. Although it is well documented that there are changes in academic and psychological adjustment during early adolescence that occur during periods of transition from elementary (grades K-6) to middle school (grades 7-9) (Eccles et al , 1984; Eccles, Buchanan, Flanagan, Ful igni , Midgley, & Yee, 1991; Eccles et a l . , 1993; Eccles & Midgley, 1989; Roeser & Eccles, 1998; Simmons & Blyth, 1987), considerable less research has been conducted on school transitions from grade eight to grade nine (e.g., Isakson & Jarvis, 1999), or that has focused on a cross-section of grades in school. Therefore, the present study w i l l not only examine the effects of parent, peer, and school contexts in relation to adolescents' 15 experiences of autonomy and relatedness, but w i l l target specific grade levels in varying school contexts (i.e., elementary school, high school). This approach w i l l allow for the examination of differences across a wider range of grades arid school contexts. The conceptual framework for the current study integrates both theory and research on individuation and developmental tasks of adolescence that pertain to autonomy (e.g., Bios, 1967; H i l l & Holmbeck, 1986; Josselson, 1980; Steinberg, 1990), and relatedness (e.g., Armsden & Greenberg, 1987; H i l l & Holmbeck, 1986; Ryan & Powelson, 1991; Ryan et a l . , 1994) in family, peer, and school contexts (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Goodenow, 1993a, 1993b, 1994; Ryan e t a l . , 1994; Wentzel, 1996). Accordingly, studies that stem from the examination of normal developmental trajectories of adolescent adjustment and school functioning in relation to autonomy and relatedness are of theoretical relevance in the current study. Whereas research has typically emphasized either autonomy or relatedness in a specific context of adolescents, the current study provides the framework for assessing both autonomy and relatedness across a variety of salient contexts in relation to educational and psychological outcomes. Fol lowing, I present both theory and research on autonomy and relatedness in each of the contexts around which this study is organized. Figure 1 illustrates the conceptual framework for this study. Autonomy and Relatedness in the Parental Context Theoretical background. The development of autonomy is a critical developmental milestone of the teenage years (Collins, Gleason, & Sesma, 1997; Douvan & Adelson, 1966; H i l l & Holmbeck, 1986; Silverberg & Gondoli , 1996; Steinberg, 1990). Bios (1967) was one of the first theorists to highlight the importance of the development of autonomy in Figure Caption Figure 1. A conceptual framework and purpose of the study. A Conceptual Framework of the Study To examine the extent to which adolescents' perceptions of autonomy and relatedness in parent, peer, and school contexts are associated with educational and behaviourial outcomes. 18 adolescence. More specifically, he posited that it is during adolescence in which the "second individuation process" occurs. During this process, the adolescent "takes increasing responsibility for what he does and what he is, rather than depositing this responsibility on the shoulders of those under whose influence and tutelage he has grown up" (Bios, 1967, p. 168). Proponents of this perspective assert that growth towards autonomy involves psychological changes in the adolescent's relationships with his or her parents as the adolescent becomes increasingly differentiated from a past or present relational context and begins to see himself or herself as being psychologically separate and distinct from his or her parents (Bios, 1967; Josselson, 1980; Mazor & Enright, 1988). Theoretically, individuation results from the development of more sophisticated and complex cognitive abilities and represents a push towards independence as adolescents begin to assume more responsibility by taking part in decisions and processes affecting their lives (Bios, 1967). This change occurs, however, in contexts in which adolescents maintain significant attachments and connections with their parents and significant others ( H i l l & Holmbeck, 1986; Ful igni & Eccles, 1993; Grotevant & Cooper, 1986; Lamborn & Steinberg, 1993; Lapsley, 1990; Quintana & Lapsley, 1990; Ryan & Lynch , 1989; Steinberg, 1990; Steinberg & Silverberg, 1986). Indeed, some researchers assert that adolescents want a balance of independence from and connectedness with their parents (e.g., Ful igni & Eccles, 1993; Geuzaine, et a l . , 2000; Silverberg & Gondoli , 1996). Thus, within the parental context, although adolescents become more psychologically independent in their relationships with their parents, they do not sever attachments (H i l l & Holmbeck, 1986). 19 Studies relating parental autonomy and relatedness to school functioning and psychological adjustment. A number of studies exist examining autonomy and relatedness in regard to adolescent-parent relationships during early adolescence (e.g., Fuhrman & Holmbeck, 1995; Lamborn & Steinberg, 1993; Papini & Roggman, 1992; Ryan & Lynch , 1989; Steinberg & Silverberg, 1986). Overall, researchers have examined associations between parental autonomy and psychological adjustment and well-being (see H i l l , 1993 for a review). For example, Steinberg and Silverberg (1986) examined adolescents' emotional autonomy from parents in a large sample of 865 10 to 16 year olds from grades 5 through 9. Information was obtained from self-reports using the Emotional Autonomy Scale (a measure developed by the researchers) in conjunction with two other measures of autonomy (i.e., resistance to peer pressure and self-reliance). Higher scores on the emotional autonomy scale were indicative of greater autonomy from parents. Their findings indicated that older adolescents were more autonomous from their parents than were younger adolescents. Moreover, girls scored higher than boys on the overall measure of emotional autonomy from parents. Adolescents were then categorized as either high or low in autonomy from parents and peers, based on a median split on scores of emotional autonomy and peer resistance. Findings across the grade levels indicated differences in the number of adolescents who were parent-oriented (low autonomy from parents/high autonomy from peers) in grade 5 to the number of adolescents who were peer-oriented (high autonomy from parents/low autonomy from peers) in grade 9 for both girls and for boys, and thus, suggested a shift from parental to peer relationships as adolescents matured. Specifically, older adolescents were found to be more emotionally autonomous from their parents and less autonomous from their peers 20 than were adolescents in grade five. Ryan and Lynch (1989) investigated the relation between emotional autonomy from parents and adolescents' feelings of relatedness to their parents in a sample of youth from grades 7 ( N = 148), grades 9 through 12 ( N = 193), and undergraduate levels ( N = 104). A l l adolescents completed Steinberg and Silverberg's (1986) measure of emotional autonomy along with measures of parental attachment (Epstein, 1983; Greenberg, 1982). A n additional measure of self-concept was used in the older adolescent sample. Ryan and Lynch found that emotional autonomy was negatively associated with the quality of adolescents' attachment to their parents. That is, the investigators found that adolescents who reported high levels of emotional autonomy from parents also reported feeling more insecure and less connected to their parents. In contrast to Steinberg and Silverberg's (1986) findings, early adolescent boys scored higher than girls on emotional autonomy from parents. Lamborn and Steinberg (1993) found further evidence of the importance of autonomy and relatedness in association with measures of adjustment and academic competence. These researchers hypothesized that greater feelings of emotional autonomy from parents and parental support would predict better adjustment and academic competence. Lamborn and Steinberg's study was undertaken in a large sample of approximately 8,700 adolescents, ranging in age from 14 to 18 years, from grades 9 through 12. Adolescents completed self-report measures of emotional autonomy from parents, parental relationship support, adjustment and competence (e.g., self-esteem, school deviance, peer conformity, drug and alcohol use, antisocial behaviour, depression, academic self-competence, grade-point average). Adolescents were then categorized into four groups based on their scores on 21 measures of emotional autonomy from parents and relationship support. The researchers found that adolescent boys and girls who scored high on emotional autonomy from parents also reported more behaviour problems in school and internal distress than adolescents who scored low on emotional autonomy. Higher levels of relationship support also predicted fewer behaviour problems and better academic competence. Age and gender differences were found with respect to scores on emotional autonomy and relationship support. Specifically, emotional autonomy from parents increased with age whereas parental relationship support decreased. Girls reported higher levels of emotional autonomy and relationship support with parents than did boys. However, these researchers found that among those adolescents who reported high levels of both emotional autonomy and relationship support, also had higher scores on measures of academic competence, and reported more behaviour problems. Thus, it appears that the balance of autonomy with relationship support may have some beneficial as well as some deleterious effects on adolescent functioning. Other research findings seem to contradict the deleterious effects of high levels of emotional autonomy from parents on adolescent psychological and academic functioning, and suggest that the relation between autonomy and functioning is mediated by the familial context in which they occur. Fuhrman and Holmbeck (1995), for example, examined relations between emotional autonomy from parents, adjustment (i.e., competence, grade point'average, internalizing problems, externalizing problems), and family variables (i.e., parent-adolescent conflict, family cohesion, parental control, maternal warmth) in a sample of 96 adolescents, ranging in age from 10 to 18 years old. These researchers found that high 22 emotional autonomy from parents was associated with fewer teacher-reported externalizing problems, higher scores of competence and higher school grades in family situations characterized by more adolescent-parent conflict and low maternal warmth. In contrast, higher emotional autonomy from parents was associated with more problem behaviours in less stressful family situations characterized by less conflict and greater maternal warmth. These researchers suggested that greater emotional autonomy may be adaptive because of its association with positive adjustment in less supportive family situations. Other researchers examining factors associated with competence and well-being have found that maintaining close emotional connections and attachments to others is important for adolescent social and emotional adjustment (Papini, Roggman, & Anderson, 1991; Ryan & Powelson, 1991; Wentzel & Feldman, 1996). Papini and Roggman (1992), for example, explored relations of attachment to parents to dimensions of competence and emotional well-being in a longitudinal study involving 47 preadolescents. Students in grade six were followed over the transition period from elementary school to grade seven in junior high school. Data were collected at three points in time — at the end of the sixth grade, at the beginning of seventh grade, and at the end of the seventh grade. The researchers hypothesized that strong parental attachments, operationalized as psychological security, in terms of acceptance, trust, and communication (Armsden & Greenberg, 1987), would be associated with higher levels of self-competence and positive behavioural conduct, and lower levels of depression and anxiety. Students completed self-report questionnaires assessing self-worth (e.g., scholastic competence, behavioural conduct), emotional well-being (e.g., depression, anxiety), and parental 23 relationships (e.g., attachment, emotional autonomy). Support was found for the hypothesis that competence and well-being would be significant and positively related to higher levels of parental attachment. Moreover, emotional autonomy was found to be significantly and positively related to internalizing problems and significantly and negatively related to parental attachment. Taken together, these findings indicate that adolescents who report higher levels of behavioural and psychological problems also report lower levels of parental attachment. Thus, it appears that during adolescence, better parental attachments are associated with positive behaviours and academic competence. Papini and Roggman's findings are consistent with those of other researchers that suggest that close relationships with parents may help to buffer adolescents from negative behaviours and emotional problems, such as depression and anxiety (Papini et a l . , 1991; Wentzel & Feldman, 1996). Additional evidence for the positive relation between parental relatedness and psychological well-being was found in a cross-sectional study of attachment relations conducted by Greenberg, Siegel, and Leitch (1983). Specifically, Greenberg et al . examined the quality of parent and peer attachments in association with self-concept and life satisfaction in a sample of 213 adolescents, ranging in age from 12 to 19 years. The researchers hypothesized that (1) attachments to parents and peers would be positively associated with adolescent well-being, and (2) parental attachments would be a more powerful predictor of well-being than would peer attachments. Adolescents completed questionnaires designed to assess both positive and negative life events, self-concept, life satisfaction, and attachments to parents and peers. The quality of attachment to parents and peers was assessed using the Inventory of Adolescent Attachment ( I A A ; Greenberg, 1982). 24 Results revealed that adolescents' attachment to parents was positively and significantly related to well-being (i.e., self-esteem and life satisfaction). Thus, these results indicate that adolescents who perceive their relationships with their parents as positive, also report better adjustment. Although previous studies examining autonomy and relatedness have focused more often on psychological adjustment outcomes, rather than on school outcomes during adolescence, researchers are now beginning to examine the relation of adolescents' relationships with parents to school functioning. For example, a recent study by Eccles and her colleagues (1997) examined parental relatedness, in terms of experiences of connection and support, in association with school and psychological outcomes. The researchers followed 1,387 seventh grade adolescents over one year and hypothesized that experiences of connection in the family would predict positive school-related functioning and less involvement in problem behaviour. The findings from their study revealed that parent connection (i.e., emotional closeness and support from parents) was positively associated with academic performance and negatively associated with problem behaviour and depressive affect for both girls and boys. These results are consistent with those of Ryan and his colleagues (Ryan et a l . , 1994) in which relatedness to parents was found to be an important predictor of school functioning and academic engagement in school. Thus, the findings by Eccles et al . (1997) support the hypothesis that relatedness in the parental context is important to adolescent school and psychological functioning, especially during the early adolescent period of development. 25 Summary- The group of studies reviewed in the preceding section provide some empirical support for the view taken that both autonomy and relatedness play a role in adolescent development as adolescents seek to establish individuality and connectedness in their relationships with parents (Grotevant & Cooper, 1986; Silverberg & Gondoli , 1996). Although emotional autonomy provides adolescents with the opportunity to develop greater separateness and individuality (Bios, 1967; Steinberg, 1990), it also predicts less relatedness and insecurity in adolescent-parental relationships (Frank et a l . , 1990; Lamborn & Steinberg, 1993; Ryan & Lynch , 1989), and greater susceptibility to peer pressure (Steinberg & Silverberg, 1986). Moreover, when considered in contexts of supportive relationships with parents, high levels of emotional autonomy were found to have positive associations with problem behaviours (e.g., Fuhrman & Holmbeck, 1995; Lamborn & Steinberg, 1993). Steinberg and Silverberg's (1986) preliminary work examining emotional autonomy from parents during early adolescence demonstrates that significant changes occur in autonomy development in relationships with parents. Their research and subsequent findings have spurred considerable debate over the meaning and function of emotional autonomy in adolescent development. On the one hand, Steinberg and his colleagues (Lamborn & Steinberg, 1993; Steinberg & Silverberg, 1986) posit that adolescents can become emotionally autonomous from their parents without becoming detached from them. They suggest that emotional autonomy represents individuation and relinquishing of childhood conceptions of, and dependencies on parents that are associated with developing a more individuated sense of self. On the other hand, Ryan and Lynch (1989) interpret emotional autonomy to mean detachment or "the loss of developmentally appropriate attachments" (p. 26 354) because of its association with less positive outcomes in adolescence. However, recent advances in research on the nature of the adolescent developmental period reveal that adolescence is not the widespread tumultuous period that was once described in the literature and that adolescents and their parents have supportive relationships which involve both autonomy and relatedness (Allen & Hauser, 1996; Feldman & Ell iot , 1990; H i l l , 1993; H i l l & Holmbeck, 1986; Offer & Schonert-Reichl, 1992). Overall , considering the important changes that occur in the development of autonomy during adolescence, further examination of the links between emotional autonomy and relatedness in relationships with parents is clearly warranted. The parental context is an especially important arena in which to study age and gender differences in autonomy and relatedness because of normal developmental changes that occur in adolescent-parental relationships throughout this period (Bios, 1967; Coll ins , Laursen, et a l . , 1997; H i l l & Holmbeck, 1986; Josselson, 1980). A s seen throughout the research in this section, findings have been equivocal with respect to age and gender differences during adolescence. Steinberg and his colleagues (Lamborn & Steinberg, 1993; Steinberg & Silverberg, 1986) found that emotional autonomy increased with age, and girls scored higher than boys on emotional autonomy. In contrast to Steinberg and Silverberg's (1986) findings, Ryan and Lynch (1989) failed to find age differences, and boys in their sample scored higher than girls on emotional autonomy. Moreover, Lamborn and Steinberg (1993) found that emotional autonomy increased with age while parental relatedness decreased during middle adolescence, and girls reported higher relationship support than boys. But, no age and gender differences in relatedness were found among early and middle 27 adolescent age ranges by Papini and Roggman (1992) or Greenberg et a l . , (1983). Thus, it is not yet clear whether there are age-related and gender differences that extend across early and middle adolescence with respect to autonomy and relatedness with parents. Such findings indicate that further research is needed to examine gender and age separately in analyses. In summary, the results from the studies reviewed in this section are important because they shed light on the meaningful connection between autonomy and relatedness in the parental context. The findings add to the growing body of research that takes into account how different contexts uniquely and collectively contribute to adolescent development and predict successful school functioning (e.g., Eccles et a l . , 1997; Kenny, Lomax, Brabeck, & Fife , 1989; Wentzel, 1998). Studies of autonomy and relatedness in the parental context are best explicated through an examination of the association between these two dimensions of development in relation to adolescents' functioning. It should be noted, however, that few of the previous studies have included comprehensive measures of school-related functioning and adjustment. Indeed, scant research exists that has examined academic correlates directly in association with autonomy and relatedness. Thus, one goal of the present research is to expand the understanding of the associations among autonomy and relatedness, and adolescents' school functioning and adjustment by including a larger corpus of dimensions of school functioning and adjustment than included in previous research. Autonomy and Relatedness in the Peer Context Theoretical background. Researchers have documented the importance of the adolescent peer group as a context for development and adjustment (see Brown, 1990 for a 28 review), and because it provides a "bridge from childhood dependencies to a sense of autonomy and connectedness with the greater social network" (Gavin & Furman, 1989, p . 827). It has been wel l established that during adolescence, the salience of peer relationships increases, in part, as a result of adolescents' striving for autonomy from parents (Allen et a l . , 1990; Armsden & Greenberg, 1987; Brown, 1990; Coll ins, Gleason, et a l . , 1997; Gavin & Furman, 1989; H i l l , 1993). Thus, adolescents' relationships become less centred on their families as the "balance of influence on social development shifts progressively from parents to the peer group" ( H i l l , 1993, p. 75). Indeed, peer relationships expand to occupy a central role in the lives of adolescents (Brown, 1990; Fuligni & Eccles, 1993; Goodenow, 1993a; H i l l , 1993; Shulman, 1993; Steinberg & Silverberg, 1986). For instance, adolescents spend more time in socializing and leisure activities with peers than with parents, and become more dependent on peers for advice and support on personal matters (e.g., Ful igni & Eccles, 1993). Thus, in studies examining changes in parent-adolescent relationships, researchers have also examined the role that peers play in adolescent development (e.g., Bla in , Thompson, & Whiffen, 1993; Chen & Dornbusch, 1998; Gavazzi, Anderson, Sabatelli, 1993). The peer group provides a social context within which adolescents can develop a strong sense of self and explore their individuality as they establish psychological separateness and independence from parents (Allen, Aber, & Leadbeater, 1990; Eccles, Buchanan, et a l . , 1991; Ful igni & Eccles, 1993; Shulman, 1993). Peer relationships are considered to be fundamental to an individual's social and emotional development throughout childhood and adolescence and influence academic and social functioning in school (Berndt et 29 a l . , 1990; Oldenburg & Kerns, 1997; Wentzel & Caldwell , 1997). For example, those individuals who have poor peer relationships in childhood are more likely to exhibit problem behaviours in adolescence (Allen et a l . , 1990; Gillmore, Hawkins, Day, & Catalano, 1992; Oldenburg & Kerns, 1997; Parker & Asher, 1987; Savin-Williams & Berndt, 1990) and risk academic failure (Allen, Kuperminc, Philliber, & Herre, 1994; Wentzel, 1998). Consequently, the peer context becomes increasingly important as the adolescent becomes less dependent on parents and more susceptible to the influence of peers (Berndt, 1979; M c C o r d , 1990; Steinberg & Silverberg, 1986). Some theorists of adolescence suggest that the individuation process repeats itself with peers and that greater psychological differentiation from peers similar to that which occurs with parents, should take place in adolescence (e.g., Josselson, 1989). According to Josselson (1980) adolescents begin to differentiate themselves from their peers and relinquish dependencies on their peers. Moreover, according to individuation theory, peer dependencies that merely replace childhood dependencies on parents may result in unhealthy psychological adjustment and dependencies (Bios, 1967). A s can be surmised, less differentiation from peers may indicate failure of individuation and unhealthy adjustment that, in turn, could lead to greater susceptibility to peer pressure. Although some degree of psychological autonomy from peers appears to be necessary, researchers also acknowledge the importance of peer relationships in providing secure and close attachments that partly replace psychological dependence on parents (e.g., Collins, Gleason, et a l . , 1997). In recent reviews of the literature on socializing influences in education, researchers have suggested that peers have an influence on adolescents' school-related functioning, in 30 terms of learning, motivation, and behaviour in school (e.g., Minuchin & Shapiro, 1983; Wentzel, 1991) and contribute to the adolescents' sense of belonging or attachment to school (e.g., Goodenow, 1992). Moreover, researchers have shown that similarities, such as shared interests and common attitudes among friends, can exert pressure and influence behaviour of adolescents in antisocial situations (e.g., Berndt, 1979; Berndt et a l . , 1990; Brown, Clasen, & Eicher, 1986; Gil lmore et a l . , 1992; M c C o r d , 1990). Indeed, the nature and function of peer relationships have important implications for adolescents' functioning at school. Studies relating peer autonomy and relatedness to school functioning and psychological adjustment. Findings from studies examining the influence of peers, peer pressure and peer conformity among adolescents have provided empirical evidence demonstrating that peers influence adolescents' school adjustment (e.g., Berndt, 1979; Berndt, 1999; Berndt et a l . , 1990; Brown et a l . , 1986). For example, in an experimental study designed to examine the influence of peers on behaviour and achievement during early adolescence, Berndt et al . (1990) examined peer pressure in relation to students' motivation to achieve during early adolescence. These investigators used six hypothetical dilemmas in \ which students were making choices between either doing their school work or doing other activities that would interfere with that work. In a pretest, eighth grade students ( N = 118) responded individually to the six hypothetical dilemmas by choosing between two alternatives designed to assess the value they placed on their school work and behaviour in the classroom. Next, in pairs, participants first discussed the same dilemmas from the pretest and then reached a common decision. Following the discussions, students completed a posttest assessing their responses to the same six hypothetical dilemmas. Their responses to 31 the six dilemmas were rated on an 11-point scale (0 to 10) and averaged, with high scores representing high achievement motivation. In order to examine whether students' motivation to achieve (assessed v ia students' pretest scores) were associated with their school functioning, classroom teachers provided a rating of participating students' involvement (e.g., how often the student takes part in classroom discussions) and disruptive behaviour (e.g., how often the student disrupts the class by talking out). Teachers' ratings of classroom involvement and disruptive behaviour were significantly correlated with students' scores on the pretest, indicating a positive association between students' academic achievement motivation and their actual behaviour in school. The results also revealed that students' decisions were most similar to those of their partners' after having discussed the dilemmas in pairs, suggesting that classmates had influenced their choices. Peer influence was also associated with the quality of the interactions with their classmates, indicating that the more harmonious or compatible the discussions were, the greater the changes were in students' own decisions from pretest to posttest. In addition, students' decisions were associated negatively with their involvement in disruptive behaviour in school, thus indicating that better school-based decisions were associated with less disruptive behaviour in the classroom. The results of this study suggest that classmates influence students' academic effort and achievement and this influence can positively or negatively influence behaviours depending on the value the student places on academic work in school. Using a methodology similar to that of Berndt et al . (1990), Steinberg and Silverberg (1986) examined the influence of peers in a sample of early adolescents in grades five 32 through nine. In this study, peer autonomy was operationalized as resistance to peer pressure and was assessed using twenty hypothetical dilemmas that have been used previously to evaluate peer pressure in antisocial and neutral situations (Berndt, 1979). Adolescents had to solve the dilemmas by making a choice between two alternatives, one that asked what they should really do i f faced with the situation (i.e., autonomous decision making), and one that was suggested by their best friend (i.e., susceptibility to peer pressure). The researchers hypothesized that autonomy from peers would show a curvilinear pattern (i.e., susceptibility to peer pressure, dr peer conformity would be higher at early and middle adolescence than during preadolescence or later adolescence). In comparisons made across grades, results indicated that peer conformity was higher among eighth grade adolescents than among adolescents in grade nine and preadolescents in grade five, thus indicating a curvilinear trend. Gender differences were also found in that girls were found to be more autonomous from their peers and more likely to resist their influence, than were boys, especially in antisocial situations. Whereas some researchers have focused on autonomy in relationships with peers, others have examined the relevance of peer relationships in association with socioemotional functioning and well-being (e.g., Armsden & Greenberg, 1987; Blain et a l . , 1993; Patersoh et a l . , 1995), and problem behaviour (e.g., Al len et a l . , 1990; Oldenburg & Kerns, 1997). Researchers examining relatedness to peers in association with characteristics of psychological adjustment have found positive correlations between peer attachments and self-esteem measures, and negative correlations between peer attachments and problems such as depression and anxiety (e.g., Armsden & Greenberg, 1987; Eccles et a l . , 1997; Greenberg et 33 a l . , 1983; Oldenburg & Kerns, 1997). In a recent study conducted by Oldenburg and Kerns (1997), peer relationships were examined in association with symptoms of depression in a sample of fifth grade ( N = 1 6 6 ) and eighth grade ( N = 156) students. Two aspects of peer relationships, namely peer acceptance and friendship quality, were hypothesized to relate to symptoms of depression in girls and boys. Further it was hypothesized that friendship quality would be more salient among early adolescents than among preadolescents. Peer acceptance was assessed using a sociometric rating scale procedure that included both preference and popularity ratings that were summed to form a single score of popularity (Asher & Hymel , 1981). Participants were rated on how much they enjoyed spending time with their classmates (preference), and how much they were liked by their classmates (popularity). In addition, students completed self-report questionnaires assessing friendship quality (Parker & Asher, 1993) and depression (Birleson, 1981; Kovacks, 1981). Results revealed that peer acceptance and depressive symptoms were negatively related, and that peer acceptance was more significant for girls than for boys. Moreover, friendship quality was related negatively to symptoms of , depression more strongly among preadolescents than among early adolescents, which was contrary to the hypothesis put forth by the researchers. The relation between popularity and depressive symptoms differed for boys and girls. Popularity was related more to depressive symptoms for girls than for boys in both preadolescence and early adolescence, suggesting that girls placed more importance on peer popularity than boys. Other studies in which investigators have focused on adolescents with poor personal adjustment have suggested that the lack of positive peer relationships is related to problem 34 behaviour and involvement with deviant peers, academic failure, and less satisfactory relationships in general (e.g., Dishion, Patterson, Stoolmiller, & Skinner, 1991; Gillmore et al., 1992; Wentzel & Caldwell, 1997). Indeed, researchers have established that greater involvement with peers during adolescence without positive parent-adolescent relations appears to be one of the best predictors of problem behaviours (Dekovic & Meeus, 1997). Given the potential importance of peer relationships for adolescent development, determining the manner in which autonomy and relatedness in relationships with peers relates to adolescent functioning and adjustment in school is an important task for theory and research. Summary. Overall, the studies reviewed in this section indicate that the peer group is an essential context for positive development and adjustment. Clearly, there is a need to understand the role that peers play in meeting adolescents' developmental needs for autonomy and relatedness. Yet, in the majority of the studies, researchers have utilized samples composed mainly of early adolescents thereby limiting our understanding of peer relationships across other adolescent age levels and developmental contexts. Although findings from the empirical research reviewed suggest that peers influence adolescents' educational performance and affective states, these studies provide limited information on specific academic competencies and behavioural outcomes associated with school functioning and psychological adjustment across grades and gender with regard to peers. For example, Oldenburg and Kerns (1997) have suggested that stronger effects in the relation between popularity and psychological well-being might have been found in boys if externalizing, rather than internalizing problems had been used as the outcome measure. Therefore in order to learn whether gender and age differences are consistently found across 35 contexts, outcome measures that are sensitive to detecting adjustment problems (externalizing and internalizing problems, for example) in boys and girls should be included in studies. To address this issue, in the present study, I used a measure of psychological adjustment that assessed both dimensions of externalizing and internalizing problem behaviours. Previous research has shown that both peer involvement and conformity increase from childhood through middle adolescence (Brown et a l . , 1986; Coll ins, Gleason, et a l . , 1997). More specifically, the eighth and ninth grades are thought to represent the "developmental zenith of conformity to peer pressure" (Steinberg & Silverberg, 1986, p. 848). Measures used in studies of peer conformity have assessed autonomy in relationships with peers using self-reports of low levels of conformity to peers to indicate greater autonomy from peers (Kuperminc, A l l en , & Arthur, 1996). Nevertheless, individuation theory, which is characterized by both autonomy and relatedness, holds that the adolescent should begin to assume greater emotional separateness in their relationships with their peers for healthy psychological development (Josselson, 1980). Surprisingly, little research has specifically been directed toward examining autonomy in relationships with peers in terms of differentiation from peers. For this reason, in the present study, peer autonomy was examined by utilizing a measure of emotional autonomy which reflected the idea of individuation and an individual's perception of him or herself as being distinct and separate from peers. Autonomy and Relatedness in the School Context Theoretical background. Researchers examining school autonomy and belonging suggest that autonomy and relatedness are significant predictors of school functioning and V 36 adjustment (e.g., Adelman et a l . , 1986; Cotterell, 1992; Eccles & Midgley, 1989; Eccles et a l . , 1993, 1997; Goodenow, 1993a, 1993b, 1994; Ryan & Powelson, 1991; Ryan et a l . , 1994). Theorists acknowledge that students' perceptions of support and feelings of relatedness, as well as students' beliefs about how much control they feel they have in school, are two dimensions of school experiences that influence academic functioning and social behaviour in the classroom (e.g., Goodenow, 1993a, 1993b, 1994; Kasen et a l . , 1990; Wentzel, 1994, 1996, 1997). Moreover, a number of researchers have suggested that problem behaviour in the classroom often arises from unmet needs for autonomy and control (Glasser, 1986; Kohn , 1993; Taylor & Adelman, 1990). Studies of children's socialization processes in the classroom are particularly relevant to understanding how relatedness to teachers and peers at school are associated with students' academic and behavioural competence (e.g., Goodenow, 1994; Wentzel, 1994; 1997). Specifically, students are more likely to engage in classroom learning and behave responsibly i f they feel supported and valued by both teachers and peers. Indeed, research suggests that a sense of belonging and school membership contributes to motivation and academic engagement in school (e.g., Eccles et a l . , 1993; Goodenow, 1993a, 1993b, 1994; Roeser et al . 1996; Wentzel, 1998; Wentzel & Caldwell , 1997). Although several studies exist that have examined relatedness in association with academic achievement and social behaviour in early adolescence, few studies have examined students' experiences of autonomy and relatedness together in association with school functioning and adjustment across other developmental periods. Needs for autonomy and relatedness have been identified as important dimensions of development because they affect 37 general well-being as well as behaviour and academic achievement (Kuperminc et a l . , 1996; Ryan & Powelson, 1991). Thus, in the school context, autonomy and relatedness have been regarded by researchers as being critical components of school experiences because of their links to learning and social adjustment outcomes. Studies relating autonomy and relatedness to school functioning and psychological adjustment. One of the earliest studies that examined autonomy in relation to school functioning was conducted by Smith et al . (1987). In this study Smith and his colleagues conceptualized school autonomy as perceived control at school, that is, students' beliefs about how much influence they have in school-related situations (e.g., being included in decision making with regard to learning activities and classroom rules). These investigators hypothesized that students' perceptions of control at school would be associated positively with attitudes, affect, and behaviour. It was further hypothesized that lower levels of perceived control at school would be found among students experiencing problems in school, especially among students in public school special education resource room programs. The study involved three samples drawn from both regular and special education populations. The total sample included 188 students, ranging in ages from 9 to 19 years, from one public school regular education program ( N = 80), one public school special education resource room program ( N = 57), and one university experimental special education laboratory program ( N = 51) designed to enhance students' perceptions of control at school. Students completed self-report measures of perceived control at school (PCSS; Adelman et al . 1986), and several items pertaining to attitudes and feelings toward school (e.g., how he or she feels about wanting to go to school) and items relating to life 38 satisfaction and happiness (e.g., how satisfied he or she feels about life in general, and how happy he or she usually feels). Teachers completed ratings of students' behaviour at school. The results from the study showed that, within the regular education sample, higher perceived control at school was related positively to both teachers' ratings of appropriate behaviour and students' self-reports of positive adjustment or well-being (e.g., attitudes toward school, life satisfaction, happiness). Furthermore, the results indicated that, within the public school special education resource sample, low perceived control at school was associated with less positive behaviours and attitudes in school. The laboratory special education sample reported higher perceptions of control and more positive attitudes toward school than did either of the other two samples. In addition, the results did not reveal any significant gender or age differences in the total sample across groups of students ranging in age from 9 to 12 years (N = 73), 13 to 15 years ( N = 71), and 16 to 19 years (N = 44). In general, however, younger students tended to have lower perceived control scores than older students. Overall , the results confirmed the hypothesis that lower levels of perceived control at school would be associated with less positive adjustment among students who experienced problems in learning and behaviour at school. Additional empirical evidence regarding the role of autonomy in school can be found in the research conducted by Eccles and her colleagues on the developmental stage-environment fit model, as discussed earlier in this chapter (e.g., Eccles et a l . , 1993; Eccles, Buchanan, et a l . , 1991; Eccles et a l . , 1984; Eccles & Midgley, 1989 for reviews). Moreover, according to Eccles, Lord , et al . (1991), "in person-environment fit theory, behaviour, motivation, and mental health are influenced by the fit between the characteristics 39 • individuals bring to their social environments and the characteristics of these social environments" (p. 523). Adolescents' developing need for autonomy has been identified by Eccles and her colleagues as one example in which the school environment, through its teaching practices, produces a mismatch. Eccles et al . (1984) have argued that the declines in motivation, school grades, and school-related behaviour are a result of the mismatch between the developmental needs of early adolescents and the opportunities afforded them in their school environments. For example, in comparison to elementary grades, adolescents in middle school and high school classrooms perceive there are fewer opportunities to participate in decision making, choice, and self-management in the classroom (Eccles, et a l . , 1984). Person-environment fit theory suggests that "when the needs or goals of the individual are congruent with the' opportunities afforded by the environment, favourable effective, cognitive, and behavioral outcomes should result for that individual; conversely, when a discrepancy exits between the needs of the individual and opportunities available in that individual's environment, unfavourable outcomes should result" (Midgley & Feldlaufer, 1987, p. 237). Midgley and Feldlaufer (1987) investigated students' decision making opportunities before and after the transition from sixth grade elementary classrooms to seventh grade middle school classrooms in a longitudinal study involving 2210 early adolescents. The researchers compared students' and teachers' actual and preferred responses to five pairs of items measuring opportunities for classroom input from students concerning where to sit, amount of homework, what classwork to do, classroom rules, and what to do when work is completed. The findings showed that, although students expressed a desire for more decision 40 making in the seventh grade, both students and teachers reported that students experienced less decision making opportunities in the seventh grade than in the sixth grade. Thus, the study's findings indicated that older students perceived they had fewer opportunities for making classroom decisions after the transition. A number of recent studies on relatedness in school provide empirical support for the important role that belonging and support play in adolescents' school functioning and psychological adjustment (e.g., Goodenow, 1993a, 1993b, 1994; Wentzel, 1994, 1997). In one study, Goodenow (1993a) examined early adolescents' sense of belonging and support in school in relation to achievement, motivation, and effort. Goodenow hypothesized that belonging would be positively associated with motivation, and influence students' effort and achievement in the classroom. A sample of 353 students in grades 6, 7, and 8 completed self-reports assessing classroom belonging and support (using the Classroom Belonging and Support Scale; Goodenow, 1993a), and motivation (using the Student Opinion Questionnaire; Pintrich & DeGroot, 1990). Teachers completed effort and achievement ratings for each student. Goodenow (1993a) found that teacher support was positively related to students' motivation and effort, especially for girls. More specifically, girls placed greater importance on feelings of belonging in school than did boys. The correlational results of the study also showed that the strength of the associations between motivation and belonging/support dropped significantly from sixth to eight grade. Overall , these results suggest the importance of classroom belonging/support (i.e., relatedness) to academic motivation, effort and achievement among early adolescents. Wentzel (1994) examined the relation between perceived support from teachers and 41 peers and early adolescents' social adjustment in a sample of 475 middle school students. Wentzel hypothesized that the degree to which students felt connected to the social life of the classroom, operationalized in terms of students' perceptions of support from teachers and peers, would be critical factors that motivated students to pursue socially responsible behaviour. In this study, teachers rated students' prosocial and responsible classroom behaviour (helping other students learn, being considerate of others in class) and students' responsible/irresponsible classroom behaviour (rule compliance, acting out behaviours). The items used to assess perceived support from teachers and peers reflected notions of belonging and relatedness (e.g., concern and caring from teachers and peers) and were taken from the Classroom Life Measure (Johnson, Johnson, Buckman, & Richards, 1985). Wentzel found that teacher support was a significant and positive predictor of responsible behaviour or social adjustment. Social and academic support from teachers and peers were related significantly and positively to prosocial behaviour. ' Subsequent research conducted by Wentzel (1997) demonstrates the importance of students' relationships with teachers in predicting academic achievement motivation and classroom behaviour. Specifically, in a longitudinal study of eighth grade students, Wentzel examined students' perceptions of teacher caring (social and academic support from teachers) in relation to students' psychological distress, academic effort, and social behaviour. Wentzel hypothesized that psychological variables might "explain links between perceived support from teachers and students' effort and engagement in the classroom" (p. 412). Adolescents completed survey measures that assessed teacher caring/support (Johnson et a l . , 1985), psychological distress, control beliefs, several items assessing prosocial behaviour, 42 social responsibility, irresponsible behaviour, and academic effort. A composite psychological distress score was computed based on 12 items that assessed anxiety, depression, low self-esteem, and low well-being (Weinberger, Feldman, Ford, & Chastain, 1987). End of year grades provided a measure of academic achievement. Results indicated that psychological distress was related significantly and negatively to perceived teacher caring, responsible classroom behaviour and compliance, and student's academic effort. Results of this study showed that teacher caring was a significant predictor of behavioural competence (prosocial behaviour and social responsibility scores) and students' academic effort ("try" and "pay attention" scores). Moreover, gender analyses indicated that girls, more than boys, had significantly higher scores on teacher caring, distress, prosocial behaviour and social responsibility. The findings confirmed the study's hypothesis, that better psychological adjustment, indicated by lower scores on the items of distress, would be related positively to teacher caring, responsible classroom behaviour and academic effort. The results of Wentzel's (1994, 1997) studies suggest that a sense of relatedness or belonging is strongly associated with involvement in prosocial and responsible behaviour and may foster positive social and academic outcomes for adolescents. In one of the few studies examining relatedness in relationships with parents, peers, and teachers in the school context, Ryan and colleagues (1994) investigated associations between relationship perceptions and school-related functioning and adjustment in a study involving 606 middle school students in grades seven and eight. Relatedness was examined using self-report measures assessing adolescents' perceptions of emotional security and utilization of support in relationships with teachers, parents, and peers on the Inventory of 43 Adolescent Attachment (IAA; Greenberg, 1982), school utilization (e.g., sharing problems at school with others), and emulation of others (e.g., identification with parents, teachers, and peers as role models). School functioning was assessed using measures of students' school-related competencies (such as coping with academic failure, reasons for engaging in school tasks, motivational orientation, and perceptions of control over outcomes and grades). A global self-esteem measure, the Multidimensional Self-Esteem Inventory (O'Brien & Epstein, 1988) was used to assess adjustment. Ryan and his colleagues found that those adolescents who reported high levels of security and utilization of teachers and parents for emotional and school concerns also reported more positive attitudes and motivation in school. Relatedness to friends was associated more with self-esteem than with school-related outcomes. The findings showed that girls, more than boys, reported feeling secure in their relationships with teachers, and emulated teachers and friends. In contrast, boys scored lower than girls on emotional connections to school, and friends, and were more likely to report that they utilized no one for emotional and school concerns. Emulation of teachers and parents was found to be positively related to school adjustment and motivation whereas emulation of friends was negatively related to school outcomes. Analyses of grade differences indicated that utilization of support from others increased across the grades, with older adolescents more likely than younger adolescents reporting turning to teachers and friends for help with school and emotional concerns. The results of this study suggest that emotional security and utilization of others positively contributes to adolescent school functioning and self-esteem, thus indicating the 44 importance of attachments relationships with others in facilitating successful outcomes at school. Summary. The studies reviewed in this section provide empirical support for the notion that both autonomy and relatedness are necessary for successful outcomes in ? educational contexts. Researchers have established that adolescents' experiences of autonomy and relatedness are critical components of school functioning. Studies have consistently shown that as adolescents enter middle and secondary school environments they are afforded fewer opportunities for autonomy than children in elementary school environments (e.g., Eccles, Lord , et a l . , 1991; Eccles & Midgley, 1989; Eccles et a l . , 1984, 1993). A s a consequence, decreases in motivation along with corresponding increases in problematic behaviours result because students feel they do not have much choice or control over school processes that affect them. Smith et al. 's (1987) study of perceptions of control advances the theoretical argument that perceptions of control at school are important for explaining school and adjustment problems. Nevertheless, one problem with their study that may have affected the results that were presented earlier, was that gross age and gender comparisons were made between groups that were disproportionate in the ratio of boys to girls. Specifically, there were twice as many boys as there were girls in the regular education sample, twice as many girls than boys in the public school special education resource sample, and almost four times as many boys as there were girls in the laboratory school sample. The unequal proportions of boys and girls across samples raises the question as to whether the results were more representative of gender differences than of students experiencing problems in learning and 45 behaviour. Because research suggests that needs for autonomy become increasingly important and change as children enter adolescence (e.g., Savin-Williams & Small , 1986; Steinberg & Silverberg, 1986), there is a need for research that examines autonomy development across the adolescent time period. The studies of autonomy and relatedness reviewed in this section have focused primarily on early adolescent populations and thus have failed to embrace a wider sample of high school students and developmental contexts. Examining autonomy and relatedness across a wider span of age ranges during middle, as well as early adolescence, would extend the research by providing a comprehensive picture of these two dimensions. Researchers have found that many problems occur for adolescents as they make the transition from elementary to middle school, especially in terms of school performance and their relationships with teachers (Eccles & Midgley, 1989). Because teachers play an important role in facilitating school-relevant outcomes when they provide secure and supportive relationships in early adolescence (Ryan et a l . , 1994) the question that remains is what role teachers have regarding social development and school functioning during middle adolescence. Goodenow's (1993a, 1993b, 1994) studies on school belonging suggest "the possibility that a psychological sense of membership in school may affect school behaviour and academic achievement indirectly through its influence on motivation" (Goodenow, 1993b, p. 87). Although Goodenow did evaluate behavioural aspects of school functioning (e.g., students' absences, tardiness) in relation to school belonging among a sample of early adolescents, her examination of school belonging and membership focused on academic-46 related outcomes rather than on adjustment outcomes (Goodenow, 1993a, 1993b, 1994). In the present investigation, I attempted to supplement our understanding of autonomy and relatedness at school by assessing the relations between school belonging and student outcomes and examining these dimensions in relation to a more comprehensive range of behaviours than was investigated in previous research. Specifically, I examined the relations of autonomy and relatedness and adolescent outcomes including school-related functioning (i.e., competencies, grades), and psychological adjustment (i.e., internalizing and externalizing problem behaviours) across grade levels and contexts. For adolescents in high school, the whole school, rather than a particular classroom becomes the social context (Minuchin & Shapiro, 1983). In particular, Minuchin and Shapiro have noted the importance of examining age/grade groupings when examining the l ink between development and psychological and school adjustment. For example, schools represent "different social contexts at preschool, elementary, and secondary levels. They are organized differently, children perceive them differently, and different aspects of social behavior are expressed in school as a function of children's changing capacities and needs" (Minuchin & Shapiro, 1983, p. 199). That is, the school environment represents social contexts that change across elementary and secondary grades that reflect developmental changes adolescents experience at these different grade levels. In this regard, within the school context, microcontexts that are age/grade specific should be considered in research efforts that span multiple contexts. Previous research has been limited because of its focus on a specific academic class in school rather than the school in general. Thus, in the present investigation students were asked to respond to their experiences of autonomy and relatedness 47 in school, in a generalized manner. In summary, research examining the degree to which students' perceptions of autonomy and relatedness are associated with school functioning have critical implications for education. Indeed, research that attempts to disentangle the complex relations between adolescents and their school and behavioural functioning w i l l help clarify the role of autonomy and relatedness in the school context. Increased knowledge of autonomy and relatedness in relation to school functioning and psychological adjustment during adolescence may provide important information necessary for the design and implementation of successful intervention programs at school. Statement of the Problem Several researchers (e.g., Barber & Olsen, 1997; Chen & Dornbusch, 1998; Eccles & Midgley , 1989; Ryan & Powelson, 1991) contend that autonomy and relatedness needs play a meaningful role both in adolescent psychosocial development and academic achievement. For example, failure to establish autonomy and relatedness in the family has been linked to problem behaviours during adolescence (Allen, Hauser, Eickholt, Be l l , & O'Connor, 1994; Al l en , Moore , Kuperminc, & Be l l , 1998). Although the importance of experiences of autonomy and relatedness to children's and adolescents' adjustment has been supported by research in the family, we know very little about these experiences on adolescents' functioning in other contexts, such as with peers and at school during the adolescent years (Allen & Hauser, 1996; Armsden & Greenberg, 1987; Conger et a l . , 1997; Frank et a l . , 1990; Fuhrman & Holmbeck 1995; O'Brien, 1989). Unt i l recently, most researchers have focused their attention on examining children's and adolescents' experiences of autonomy and 48 relatedness in the family and have overlooked other important social developmental contexts, such as peer groups and school. Many motivation theorists, for instance, believe that problem behaviours at school (e.g., noncompliance, learned helplessness, school failure) may be reflective of adolescents' attempts to demonstrate their independence from parental and/or societal norms to achieve a sense of personal control at school (e.g., Adelman & Taylor, 1990; Glasser, 1986; Murtaugh & Zetlin, 1990; Taylor & Adelman, 1990; Taylor, Adelman, Nelson, Smith, & Phares, 1989). Other research suggests that adolescents are more likely to be successful in school and act responsibly i f they feel supported in their learning and perceive teachers and peers are concerned about them (e.g., Goodenow, 1994; Wentzel, 1994; 1997). Moreover, much research exists linking academic and social difficulties with poor parental and peer relationships (e.g., A l l en et a l . , 1998; Bardone, Moffitt, Caspi, Dickson, & Silva, 1996; Brier, 1995; Parker & Asher, 1987). In this regard, studies should be directed to better understanding the role that autonomy and relatedness play in adolescents' school functioning and psychological adjustment when one considers the detrimental consequences that problems in these areas have for the adolescent. The present investigation was directed toward that understanding. The purpose of the present research is to investigate two critical dimensions of adolescent development - autonomy and relatedness ~ in relation to school functioning and psychological adjustment. This study wi l l extend previous research examining autonomy and relatedness in families by including both peer and school contexts as wel l . Few investigators have either (a) studied these dimensions of adolescent development together across multiple 49 contexts, or (b) examined the interrelations of autonomy and relatedness in association with adolescent adjustment. Specifically the objective of the present study is to determine whether adolescents' perceptions of their experiences of autonomy and relatedness in parent, peer, and school contexts are associated with school functioning and psychological adjustment. Hypotheses Three sets of hypotheses have been developed to guide the present investigation. The first set of four hypotheses examines gender and grade differences in adolescents' perceptions of autonomy and relatedness in parent, peer, and school contexts. The second set of six hypotheses explores relations among autonomy, relatedness, school functioning, and psychological adjustment across contexts of parents, peers, and school. The third set of two hypotheses explores relations of autonomy and relatedness simultaneously as predictors of school functioning and psychological adjustment across contexts of parents, peers, and school. The hypotheses are numbered one through twelve. Gender and grade differences in adolescents' perceptions of autonomy and relatedness. Gender and grade level are important factors to be considered in relation to the developmental changes that occur as children move through adolescence (Conger et a l . , 1997; Eccles & Midgley, 1989; Eccles et a l . , 1993; H i l l & Holmbeck, 1986; Josselson, 1980). The parental context is an especially important context in which to study age and gender differences because of normal developmental changes that occur in adolescent-parental relationships throughout this period (Bios, 1967; Coll ins, Laursen, et a l . , 1997; Douvan & Adelson, 1966; H i l l & Holmbeck, 1986). Moreover, the upper elementary and lower high school years are times in which peer pressure and conformity to peers increases (Berndt, 50 1979; Brown, 1990; Steinberg & Silverberg, 1986) and the peer group becomes an increasingly important context for adolescents as they "rely more heavily than before on friendships and non-kin relationships for support and development" (Goodenow, 1993b, p. 81). Whi le girls at this age tend to place much more emphasis on peer attachments and relationships than boys, they are also at more at risk for depression and self-esteem problems (Armsden & Greenberg, 1987; Blain et a l . , 1993; Oldenburg & Kerns, 1997). W e know, for example, that the transition period from elementary to middle and secondary school is an especially problematic time for early adolescents in terms of a developmental stage-environment mismatch (Eccles et a l . , 1984; Eccles & Midgley, 1989; Eccles et a l . , 1993; Goodenow, 1993b; Midgley & Feldlaufer, 1987). The school environment is also an important context in which boys and girls receive gender-related messages that w i l l distinguish them and affect their self-conceptions and behaviour (Minuchin & Shapiro, 1983). When taken together, these research findings suggest that adolescents' experiences of autonomy and relatedness in relationships with parents, peers, and at school w i l l vary by gender and grade. In this study, four hypotheses regarding gender and grade differences have been developed based on findings suggested from previous research, as well as theoretical expectations. There are two hypotheses regarding gender (Hypotheses 1 and 2) and two hypotheses regarding grade (Hypotheses 3 and 4). Hypothesis 1. The first hypothesis is that when compared to boys, girls w i l l report higher levels of peer autonomy. This gender difference is expected given that previous research on peer conformity has indicated that girls are more autonomous from their peers ' 51 than boys and less susceptible to peer influences (e.g., Berndt, 1979; Brown et a l . , 1986; Steinberg & Silverberg, 1986). Hypothesis 2. The second hypothesis is that when compared to boys, girls w i l l have higher levels of parental attachment, peer attachment, and school belonging. Theory and research findings have suggested that girls place much more emphasis on feelings of belonging and attachment in their relationships with parents, peers, and at school than do boys (e.g., Douvan & Adelson, 1966; Eccles et a l . , 1997; Goodenow, 1993a, 1993b; Oldenburg & Kerns 1997; Ryan et a l . , 1994; Stern, 1990). Hypothesis 3. The third hypothesis is that the level of school autonomy w i l l show a decrease from grade eight to grade eleven. Previous research has shown that adolescents perceive less teacher support for student autonomy and decreases in opportunities for decision making and choice as they move from middle school to junior high school (e.g., Eccles et a l . , 1984, 1993; Eccles & Midgley, 1989) and into secondary school (e.g., Wil lower , E ide l l , & Hoy , 1973) that are a result of grade-related changes associated with school learning environments, school structure (McNe i l , 1986) or teacher control beliefs (e.g., Midgley, Feldlaufer, & Eccles, 1988; Wil lower et a l . , 1973). Hypothesis 4. The fourth hypothesis is that (a) adolescents in grade 11 w i l l show higher levels of parental autonomy and peer autonomy, and (b) lower levels of parental attachment and peer attachment than adolescents in grade eight or grade nine. Previous researchers have found that emotional autonomy from parents increases with age whereas relatedness to parents decreases (e.g., Blain et a l . , 1993; Coll ins, Gleason, et a l . , 1997; Lamborn & Steinberg, 1993; Papini et a l . , 1991; Steinberg & Silverberg, 1986). Moreover, 52 research examining peer groups suggests that peer conformity is higher in early adolescence than in preadolescence and late adolescence (e.g., Brown et al , 1986; Gavin & Furman, 1989; Steinberg, 1988; Steinberg & Silverberg, 1986), and there are declines in peer social support as a result of grade-related changes and changes in parent-adolescent relationships during this time (e.g., Berndt, 1989; Collins, Gleason, et a l . , 1997). Relations of autonomy and relatedness to school functioning and psychological adjustment. The second set of six hypotheses concerns the examination of relations of autonomy and relatedness to school functioning and psychological adjustment, across contexts of parents, peers, and school. In this study, six hypotheses have been developed based on findings suggested from previous research, as well as theoretical expectations. There are two hypotheses regarding school functioning (Hypothesis 5 and 6) and four hypotheses regarding psychological adjustment (Hypotheses 7 to 10). Hypothesis 5. The fifth hypothesis is that school autonomy w i l l be positively associated with school functioning. This hypothesis is based on research findings indicating the importance of school autonomy to motivation and engagement in school (e.g., Eccles et a l . , 1993; Kasen et a l . , 1990) and recent research findings indicating positive associations between student autonomy and school-related functioning (i.e., school grades, academic motivation) (e.g., Roeser & Eccles, 1998). Hypothesis 6. The sixth hypothesis is that parental attachment, peer attachment, and school belonging w i l l be positively associated with school functioning. Previous empirical research suggests that having positive parental and peer relationships are important to adolescents' school performance and well-being (e.g., Armsden & Greenberg, 1987; 53 Greenberg et a l . , 1983; Oldenburg & Kerns, 1997; Ryan et a l . , 1994). Research findings on school belonging have shown positive associations among school belonging, academic functioning, and appropriate behaviour (Goodenow, 1993a, 1993b, 1994). Hypothesis 7. The seventh hypothesis is that parental autonomy w i l l be positively associated with problems in psychological adjustment. This hypothesis is based in part on research findings indicating positive associations between problems of adjustment and emotional autonomy from parents (e.g., Fuhrman & Holmbeck, 1995; Greenberg et a l . , 1983; Lamborn & Steinberg, 1993; Ryan & Lynch, 1989; Turner et a l . , 1993) and partly on theory and research suggesting that adjustment problems are linked to insecure attachments and less differentiation and autonomy promoting behaviours in relationships with parents (e.g., see Al len et a l . , 1990 for a review; Al len & Hauser, 1996; Al len , Hauser, et a l . , 1997). Hypothesis 8. The eighth hypothesis is that peer autonomy w i l l be associated negatively with problems in psychological adjustment. Theory suggests that greater autonomy in relationships with peers should be associated with fewer psychological problems (e.g., Josselson, 1980). . Hypothesis 9. The ninth hypothesis is that school autonomy w i l l be associated negatively with problems in psychological adjustment. This hypothesis is based in part on the research findings of Smith et al . (1987) indicating negative associations between perceptions of control at school and behaviour problems and in part on research findings suggesting the importance of autonomy and decision-making opportunities to motivation and engagement in school (e.g., Eccles et a l . , 1993; Kasen et a l . , 1990). 54 Hypothesis 10. The tenth hypothesis is that parental attachment, peer attachment, and school belonging w i l l be associated negatively with problems in psychological adjustment. Research and theory suggest that the quality of attachments to parents, peers, and school are important for emotional well-being and adjustment (e.g., Armsden & Greenberg, 1987; Greenberg et a l . , 1983; Ryan & Powelson, 1991) and to motivation and engagement in school (e.g., Goodenow, 1993b; 1994). Research findings have shown negative associations between the quality of attachments to family, peers, and school with problem behaviours (e.g., depression, antisocial behaviours) (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Oldenburg & Kerns, 1997). Other research findings suggest that the lack of positive peer relationships during adolescence is related to increases in peer conformity, involvement with peers in antisocial activities, and school failure (Dishion et a l . , 1991; Wentzel & Caldwell , 1997) each of which have important implications for adolescents' functioning at school (Allen & Kuperminc, et a l . , 1994; Berndt et a l . , 1990; Gillmore et a l . , 1992; Savin-Williams & Berndt, 1990; Wentzel, 1998). Autonomy and relatedness in contexts of parents, peers, and school, as predictors of school functioning and psychological adjustment. The final set of two hypotheses further explores the relations of autonomy (i.e., parental autonomy, peer autonomy, school autonomy) and relatedness (i.e., parental attachment, peer attachment, school belonging) variables as predictors of school functioning and psychological adjustment. One hypothesis examines the extent to which the autonomy and relatedness variables explain variance in school functioning (Hypothesis 11). The second hypothesis examines the extent to which the autonomy and relatedness variables explain variance in problems in psychological adjustment (Hypothesis 12). 5 5 Hypothesis 11. The eleventh hypothesis is that teacher-rated school competencies and academic achievement (i.e., G P A ) wi l l each be positively associated with autonomy and relatedness in parent, peer, and school contexts. Thus, adolescents who perform better in school, in terms of these two dimensions of school functioning, would perceive themselves to be more autonomous in their relationships with their parents and peers, and at school, and also perceive themselves to be more attached to their parents, peers, and school. Hypothesis 12. The twelfth hypothesis is that self-reported internalizing and externalizing problems w i l l each be (a) associated positively with parental autonomy, and associated negatively with peer autonomy and school autonomy, and (b) associated negatively with parental attachment, peer attachment, and school belonging. Thus, those adolescents who experience more problems would report greater autonomy in relationships with parents, and report less peer autonomy and school autonomy, and poorer quality of attachments to parents, peers and school. The last two hypotheses are based on research findings indicating positive associations between problems of adjustment and emotional autonomy from parents (e.g., Lamborn & Steinberg, 1993); on findings suggesting that students who do not have supportive relationships with parents, teachers, and peers, or do not feel a sense of belonging in school are often at risk for academic and adjustment problems (e.g., Goodenow, 1993a, 1993b; Roeser et a l . , 1996; Wentzel, 1998); and on theory suggesting that meeting needs for autonomy and relatedness are fundamental to successful school functioning and adjustment in adolescence (e.g., Coll ins, Gleason, et a l . , 1997; H i l l & Holmbeck, 1986; Ryan & Powelson, 1991). 56 CHAPTER THREE Method Participants Participants for this study were 478 adolescents (n = 213 boys, n = 265 girls) who were enrolled in grades eight (n = 170), nine (n — 167), and 11 (n = 141). These adolescents were drawn from 27 classrooms and were attending one of seven schools in a publicly funded Catholic school district in a large Canadian city in Southwestern Ontario. It is important to note that in this school district, approximately one-third of the student population is not Catholic. In Ontario, both Catholic and Public schools follow a standardized provincial curriculum, except that Catholic schools offer a religious education course as a required course. Parental consent was obtained for 500 students (78% of the students in grades eight, nine, and eleven). O f those students, incomplete questionnaires were obtained from 22 students. Data from these participants were excluded from the analyses. There were no students who participated whose primary placement was in a full-time special education program, or who did not speak or read English. Ages of the participants ranged from 13 to 18 years old ( M = 14.71, S D = 1.32). The age categories employed in this study, representing early adolescence (ages 12 to 14 years) and middle adolescence (ages 15 to 17 years) are meaningful categories, both theoretically and empirically, that have been established in the literature and used by others in research on adolescence (e.g., Gavin & Furman, 1989; Schonert-Reichl, 1994; Steinberg, 1993; Wintre & Crowley, 1994). The early and middle adolescent age ranges were chosen for the present investigation because this time in the life span has been identified as a critical 57 period for individuation and autonomy development, and social development with parents and with friends, and at school (Bios, 1962, 1967; Coll ins, Gleason, et a l . , 1997; Grotevant & Cooper, 1986; H i l l & Holmbeck, 1986; Josselson, 1980; Minuchin & Shapiro, 1983; Silverberg & Gondoli , 1996), and thus would provide the most opportunity for revealing cross-sectional data within a developmental framework. Grade 8 elementary school students ( M = 13.48 years old, S D = 0.50) were recruited from regular education academic subject classrooms. The grade eight sample was a homogenous (not tracked by ability) population. Adolescents at this grade level represented the pretransition early adolescent. The ninth grade represents the transition year from elementary to secondary school. High school students in Grade 9 ( M = 14.47 years old, S D = 0.50) and Grade 11 ( M = 16.48 years old, SD = 0.53) were recruited from religious education classrooms in which students were not tracked by ability. Religious education is a required course for a l l students attending Catholic schools within the district. The high school students were recruited from all students enrolled in grade nine and grade eleven religious education classes in the second semester and therefore provided a representative sample of al l adolescents enrolled in the school. Although students were otherwise tracked by ability for other subject classes, the student sample represented a homogeneous (not tracked by ability) population. Students in grade nine were selected to participate because those adolescents are believed to be in "greater transition with respect to both personal characteristics associated with adolescent development and interpersonal characteristics dealing with normative social behaviour in this developmental period" (Ford, 1982, p. 337). Students in grade eleven were 58 selected to participate because at this grade level, older adolescents experiencing academic difficulties at school would likely be attending classes although they may be at risk for leaving school, and/or being considered for alternative educational programs. Students from grades eight, nine and eleven were chosen because these grade levels represented different age groupings and social contexts that reflected developmental changes adolescents experience in school (Minuchin & Shapiro, 1983). According to adolescents' reports of parents' occupations, the mean socioecononomic status (SES) of fathers and mothers, based on the Blishen, Carrol l , and Moore (1987) socioeconomic index for occupations in Canada, was 44.64 (SD = 12.87) and 41.98 (SD = 14.00) respectively. Scores ranged from 21 to 101, indicating the full range of social classes was represented. Some examples of occupations from this index are labourer in manufacturing industry (28.97), motor vehicle mechanics and repairers (39.19), secretary (41.82), dental lab technician (45.15), and c iv i l engineer (71.70). The ethnicity of the participants was 65.3% White (n = 312), 10.7% Black (n = 51), 10% Asian (n = 48), 4.2% East Indian (n = 20), 3.1% Hispanic (n = 15), and 6.7% Other (e.g., mixed ethnicity; n = 32). To obtain information regarding family composition, participants were asked to indicate to whom they were referring when completing the questionnaire items about parents. Eighty-one percent of the participants referred to both parents, 11% referred to mothers only, 1% referred to fathers only, 5% referred to mother and stepfather or father and stepmother, and 2% referred to other adults (e.g., grandparents, relative, foster parents). Measures of Autonomy. Relatedness. and Psychological Adjustment The measures used in this study are shown in Figure 2. Adolescents' measures \ Figure Caption Figure 2. Measures of autonomy, relatedness, school functioning, and psychological adjustment. Measurement 60 S c h o o l F u n c t i o n i n g P s y c h o l o g i c a l A d j u s t m e n t Grade point Average (GPA) Wentze l , 1993 Teacher - ra ted S c h o o l C o m p e t e n c i e s (T-CRS) Htghtower et a l . , 1986 Se l f - repor ted Internal iz ing P rob lem Behav iour <YSR) Se l f - repor ted Ex te rna l i z i ng P rob lem Behav iour (YSR) A c h e n b a c h , 1991 A c h e n b a c h , 1991 Emotional Autonomy S c a l e (EAS) Ste inberg & Si lverberg, 1986 Inventory of Parent and Peer At tachment (IPPA) A r m s d e n & Greenberg, 1987 Emot ional Autonomy S c a l e - Peers (EASP) adapted from Ste inberg & Si lverberg, 1986 Perce ived Contro l at Schoo l S c a l e (PCSS) Ade lman et a l . , 1986 S e n s e of S c h o o l (PSSM) Goodenow, 1993 61 included the following three sets of self-report questionnaires. Measures of autonomy. Autonomy in relationships with parents was assessed using the 14-item version of Steinberg and Silverberg's (1986) Emotional Autonomy Scale ( E A S ; see Appendix A ) . The E A S 2 (Steinberg & Silverberg, 1986) has been used to assess autonomy in the context of adolescent-parental relationships regarding parental deidealization (e.g., "Even when my parents and I disagree, my parents are always right"; reversed scored; five items, a = .63), nondependency on parents (e.g., "I go to my parents for help before trying to solve a problem myself"; reversed scored; four items, a = .51), and individuation (e.g., "There are some things about me that my parents don't know"; five items, a = .60). Students are asked to respond to each item on a four-point Likert scale (4 = "strongly agree"; 3 = "agree"; 2 = "disagree"; 1 = "strongly disagree"). A response of "strongly agree" reflects a high level of emotional autonomy in relationships with parents. Total scores range from 14 to 56 on the 14-item version. High scores on the E A S indicate greater differentiation from parents. Cronbach's alpha for the 14-item version of the E A S , as reported by Lamborn and Steinberg with their sample, was .82. The authors of the E A S reported that the scale demonstrates construct validity because the exploratory factor and internal consistency analyses "confirmed the theoretically generated components of emotional autonomy used to generate the initial pool of items" (Steinberg & Silverberg, 1986, p. 844). In addition, the E A S has been used in a number of studies relating emotional autonomy to parental support (Lamborn & Steinberg, 1993), parental fallibility (Frank et a l . , 1990) and 2 Cronbach alphas reported herein for each of the subscales of the E A S are from Steinberg and Silverberg's (1986) study. Lamborn & Steinberg (1993) reported the alpha for the entire 14-item scale only. 62 parental detachment and rejection (Ryan & Lynch , 1989) in adolescents' relationships with their parents. Cronbach alphas reported from these studies ranged from .71 to .82 for the total scale. In the present study, Cronbach's alpha computed for the total E A S scale was .78. In this sample, Cronbach alphas computed for each of the subscales for parental autonomy were as follows: Deidealization, .66, Nondependency, .48, and Individuation, .67. Autonomy in relationships with peers was assessed using a 14-item scale that parallels the E A S measure described in the previous section. The peer version, entitled the Emotional Autonomy Scale - Peers ( E A S P ; see Appendix B) was a modified version of Steinberg and Silverberg's (1986) E A S . However, rather than assessing autonomy in relationships with parents, items on the E A S were modified to assess autonomy in relationships with peers. Moreover, each item on the E A S P corresponds to an item on the E A S . The E A S P was designed to assess autonomy in the context of adolescent-peer relationships regarding deidealization of peers (e.g., " M y friends and I agree on everything"; five items al l reversed scored), nondependency on peers (e.g., "I go to my friends for help before trying to solve a problem myself"; reversed scored; four items), and individuation (e.g., "There are some things about me that my friends don't know"; five items). Adolescents are asked to respond to each item on a four-point Likert scale (4 = "strongly agree"; 3 = "agree"; 2 = "disagree"; 1 = "strongly disagree"). Total scores range from 14 to 56. A high score indicates greater differentiation in relationships with peers. In the present investigation, Cronbach's alpha computed for the total E A S P scale was .65. In this sample, Cronbach alphas computed for each of the subscales for peer autonomy were as follows: Deidealization, .48, Nondependency, .46, and Individuation, .62. 63 School autonomy was assessed using the Perceived Control at School Scale (PCSS; Adelman et a l . , 1986; see Appendix C) . The PCSS has been used in previous research to assess students' perceptions of their autonomy in school in a general way. This self-report scale consists of 16 items designed to tap students' perceptions regarding the degree of control or influence one has over school-related situations such as being able to make choices and take part in decisions (Heavey et a l . , 1989; Smith et a l . , 1987). Students respond to each statement on a 6-point Likert scale (1 = "never"; 2 = "not very often"; 3 = "slightly less than half the time"; 4 = "slightly more than half the time"; 5 = "very often"; 6 = "always"). Examples of items from the scale relating to students' involvement in making decisions and choices are "At school, I feel people want me take part in making decisions", and "At school, I feel I have a choice about what I am doing or learning." Total summary scores can range from 16 to 96, with high scores indicating greater perceived control at school. The P C S S demonstrates adequate reliability and has been validated in diverse school populations. For example, Adelman et al . (1986) reported Cronbach alphas ranging from .69 to .80 for the total scale. Test-retest reliabilities over a two week period ranged from .55 to .80. The P C S S demonstrates construct validity by its modest correlations with the Nowicki-Strickland Locus of Control scale (see Adelman et a l . , 1986). Correlations between the two measures are reported as .58 for students in grades three to six, and .35 for students in grades 7 to 12, suggesting that higher P C S S scores are associated with an internal locus of control orientation. The P C S S has the ability to discriminate between those students experiencing problems in learning and behaviour at school, and those who are not. In the 64 present study, the P C S S was found to have adequate internal consistency (Cronbach's alpha was .79). Measures of relatedness. Relatedness to parents and peers was assessed using the Inventory of Parent and Peer Attachment ( IPPA; Armsden & Greenberg, 1987; see Appendix D ) . This self-report inventory was used to examine the degree to which adolescents feel a sense of emotional connection and feel validated and cared for in their relationships with their parents and peers. This measure provides separate composite scores and subscale scores for parent attachment and peer attachment. The I P P A has been used in numerous studies of attachment with adolescents between the ages of 12 and 20 (e.g., Cotterell, 1992; Greenberg et a l . , 1983; Papini et a l . , 1991; Papini & Roggman, 1992; Paterson et a l . , 1995). There are 28 parent items and 25 peer items that make up the two attachment scales used to assess adolescents' relationships with parents and peers in a generalized manner. The use of global targets such as parents and peers "minimizes the need for defensiveness with respect to specific figures and gives a reasonable snapshot of the adolescents' general feelings concerning various types of relationships" (Ryan et a l . , 1994, p. 228). Students respond to the items using a 5-point Likert scale (1 = "almost never or never true", 2 = "not very often true", 3 = "sometimes true", 4 = "often true", 5 = "almost always or always true"). The items broadly assess dimensions of relatedness to parents and peers regarding mutual trust, quality of communication, and degree of anger and alienation. Total scores for the separate parent and peer attachment scales are computed by summing the scores for Trust and Communication, and reverse scoring al l Alienation items. Armsden and Greenberg (1987) report the inventory has "substantial reliability and / 65 good potential validity as a measure of perceived quality of close relationships" (p. 446). Moreover, Armsden and Greenberg (1987) report Cronbach alphas for the parent attachment subscales as follows: Trust (10 items; a = .91), Communication (10 items; a = .91), and Alienation (8 items; a = .86). Cronbach alphas reported for the peer attachment subscales are as follows: Trust (10 items; a = .91), Communication (8 items; a = .87), and Alienation (7 items; a = .72). Three week test-retest reliabilities using the composite measure of parent and peer attachment in Armsden and Greenberg's (1987) sample of 18 to 20 year olds were .93 for parent attachment, and .86 for peer attachment. Cotterell (1992) reported alphas similar to those found by Armsden and Greenberg (1987) for the parent attachment scale (a = .94), and peer attachment scale (cv = .87) in a sample of 14 to 17 year olds. Construct validity for the I P P A has been demonstrated through multi-trait multi-method by Armsden and Greenberg (1987). Moderate to high correlations were reported among the Family Environment Scale (FES; Moos , 1974), the Tennessee Self-Concept Scale (TSCS: Fitts, 1965) and the Family and Peer Utilization factors from the Inventory of Adolescent Attachment (Greenberg et a l . , 1983), suggesting convergent validity with these family measures (see Armsden & Greenberg, 1987). In the present investigation, internal consistencies were found to be adequate for both the parental attachment scale: Trust (a = .90), Communication (a = .82), Alienation (a = .84), and total parental attachment scale (ot = .93), as well as for the peer attachment scale: Trust (a = .92), Communication (a = .89), Alienation (a = .71), and total peer attachment scale (a = .92). A s a measure of relatedness in the school context, the Psychological Sense of School Membership ( P S S M ; Goodenow, 1993b; see Appendix E) was used. This 18-item self-report 66 scale was used to assess students' perceived belonging or psychological sense of membership in the school environment. Students respond to each item on a five-point Likert scale ranging from 1 ("not at al l true") to 5 ("completely true"). Examples of items from the scale are "I can really be myself at this school", "Most teachers at this school are interested in me", and "Other students here like me the way I am." The items are designed to tap school belonging by focusing on aspects of attachment that are believed to contribute to a student's sense of emotional connection in the school setting. L o w scores indicate alienation or an absence of perceived belonging or relatedness at school, while high scores indicate high perceived belonging and being part of the school in general. A total summary score is derived by reverse scoring the negative items and summing the scores of al l the items. The P S S M demonstrates acceptable internal consistency reliability and validity. Cronbach's alpha for the scale ranged from .77 to .88 across different samples (Goodenow, 1993b). Construct validity for the P S S M has been established by studying the results of contrasting groups (Downie & Starry, 1977). The instrument discriminates between groups and subgroups predicted to be different in terms of their sense of belonging or psychological membership in school (Goodenow, 1993b). For example, Goodenow (1993b) found that girls exhibited a higher sense of school membership or belonging than boys. In the current investigation, Cronbach's alpha for the P S S M was .89. Measures of Psychological Adjustment. The Youth Self-Report ( Y S R ; Achenbach, 1991; see Appendix F) is a standardized instrument designed to assess behavourial competencies and problems of youths who are 11 to 18 years of age. There are separate norms for boys and for girls to allow for comparisons between other populations of the same 67 age and gender. The Problem Scale provides a measure of internalizing, externalizing, and total problem behaviours derived from eight syndrome scales comprised of 103 specific problem items. Adolescents are asked to respond to each of the items using a 0-1-2 scoring system (0 = "not true"; 1 = "somewhat or sometimes true"; 3 = "very true or often true"). For the purposes of this investigation, adolescents' psychological adjustment problems were assessed using scores obtained from the Internalizing scale (withdrawal, somatic complaints, anxious/depressed), the Externalizing scale (delinquent, aggressive), and three other problem scales (social problems, thought problems, and attention problems) used to make up the Total Problem scale score. The problem scales of the Y S R have been used in research to discriminate between non-deviant and deviant populations based on T scores. Achenbach (1991) states that "scores in the normative samples ranging from about the 82nd to the 90th percentile were found to provide the most efficient discrimination for most sex/age groups. T scores of 60 to 63, which span these percentiles, were therefore chosen to demarcate the clinical range" (p. 45). For the total problem scores, T scores between 60 to 63 designate the borderline clinical range, whereas T scores above 63 represent more deviant problem scores than those scores which fall below 60. In order to determine whether the adolescents who participated in this study were different in the level of severity of problem behaviours from adolescents classified as clinically referred youths, means of the scores from the Internalizing, Externalizing and Total Problem scales on the Youth Self-Report in this sample were compared to means of the scores from normative samples provided in the manual. These data were taken from 68 Appendix B in the Manual for the Youth Self-Report and 1991 Profile (Achenbach, 1991). In the present study, the means of the Internalizing, Externalizing, and Total Problem scales were tested using one-sample directional t tests. The scale means for the participants in this sample differed significantly from the means of referred boys and girls on the Internalizing, Externalizing, and Total Problem scales of the Youth Self-Report (see Appendix G) . For the purposes of generalizing the results of the present study, these data support the contention that the participants in this study differed significantly from those adolescents in the referred group with respect to degree of problem behaviour. Extensive documentation for the reliability and validity of the Y S R measure has been provided in the manual (see Achenbach, 1991). Regarding reliability, for example, Cronbach alphas reported separately for the total problem scales in demographically matched referred and nonreferred male and female samples were both .95. Test-retest reliabilities over a 7-day period were reported as .70 in a sample of 11 to 14 year olds and .91 for 15 to 18 year olds. Over a period of 7 months, the reliability for the total problem scale was reported as .56 in a combined (male and female) sample of 11 to 14 year olds. Detailed validity information is described in the manual (see Achenbach, 1991). For example, as evidence of criterion-related validity, the Y S R T scores have the ability to discriminate between demographically matched clinically referred and nonreferred youths that indicate normal, borderline, and clinical ranges. The Y S R also demonstrates high concurrent validity with the pre-1991 counterpart Y S R scales, with correlations reported ranging from .80 to .90. In the present investigation, internal consistency assessed via Cronbach's alpha was as follows: Internalizing scale, .91, Externalizing scale, .86, and Total Problem scale, .94. 69 Measures of Adolescents' School Functioning Adolescents' school functioning was assessed in two ways: (1) teacher-ratings of adolescents' school competencies, and (2) academic achievement. School competencies. School competencies (SchComp) were assessed using teacher ratings on the Teacher-Child Rating Scale (T-CRS; Hightower et a l . , 1986; see Appendix H ) . The 38-item standardized checklist has been used to estimate students' strengths and competencies associated with social behaviour, learning, and academic abilities. Twenty items are designed to assess students' abilities and skills in accepting limitations (e.g., copes well with failure, accepts things not going his/her way), assertiveness (e.g., participates in class discussions, expresses ideas wil l ingly), task commitment (e.g., completes work, wel l organized, a self-starter) and peer relationships (e.g., is friendly toward peers, wel l liked by classmates). Teachers rate each item on how well they describe the student using a 5-point rating scale (1 = "not at a l l " ; 2 = "a little"; 3 = "moderately wel l " ; 4 = "wel l" ; 5 = "very well"). A total score is comprised of the sum of teacher ratings on four subscales relating to Frustration Tolerance, Assertive Social Skil ls , Task Orientation, and Peer Social Skil ls . Eighteen items describing school-related problems are designed to assess a variety of behaviours associated with externalizing problems (e.g., disruptive in class, disturbs others while they are working), internalizing problems (e.g, withdrawn, anxious, does not express feelings), and other difficulties associated with learning (e.g., poor work habits, poorly motivated to achieve, difficulty following directions). The items are rated by the teacher on a 5-point scale (1 = "not a problem"; 2 = "mild problem"; 3 = "moderate problem"; 4 = 70 "serious problem"; 5 = "very serious problem"). The total problem score (SchProbs) is the sum of the teacher ratings on the Acting-Out, Shy-Anxious, and Learning subscales. Scores on the T - C R S indicate the degree to which the student differs from a normative sample on the problems identified by their teachers on the subscales. Normative profiles developed by Hightower, Spinell, and Lotyczewski (1989) place those who score at or below the fifteenth percentile as having serious difficulties adjusting in the various domains. Teachers' reports of problem behaviour on the Acting-Out and Shy-Anxious subscales were used to assess the validity of adolescents' self-reports of problem behaviour on the Externalizing and Internalizing scales on the Y S R . Comparisons were made between students' and teachers' ratings of students' problem behaviours. Correlations between the Y S R and T - C R S subscales were significant and positive both for boys and for girls (see Appendix I). , Evidence of reliability and validity for the scale been described in the research literature on the development and validation of the T - C R S (see Hightower et a l . , 1986). H igh internal reliability coefficients in excess of .84 have been reported for the subscales by its authors (Hightower et a l . , 1989). In other research by Bear and Ryes (1994), alpha coefficients were reported to be as follows: .96 for the Total School Competency scale and .94 for the Acting-Out subscale. Moreover, in a sample of 138 ninth graders, Luthar (1995) reported Cronbach's alphas for the six subscales of the T - C R S ranged between .81 and .98 for boys, and .77 and .98 for girls. Test-retest coefficients ranged from .61 to .91 over 10 and 20-week periods (Hightower et a l . , 1986). The T - C R S was developed from the Classroom Adjustment Rating Scale ( C A R S ; Lor ion , Cowen, & Caldwell , 1975) and the 71 Health Resources Inventory (HRI; Gesten, 1976). Construct validity has been shown through correlations with these and other measures of school adjustment problems and competencies (see Hightower et a l . , 1986). The T - C R S has the ability to discriminate between children who are adjusting well and poorly in school. In the current investigation internal consistency, as measured by Cronbach's alpha, was found to be acceptable (Total School Competency scale, .94, Total Problem scale, .93, Acting-Out subscale, .91, and Shy-Anxious subscale, .88). Academic achievement. Academic achievement was assessed by computing a composite grade point average score. A t the end of the school year, report card grades were obtained from school records for each student participant. Grade point averages (GPAs) were calculated by averaging the grades obtained in four academic classes (i.e., english, mathematics, science, and social studies — either one history, geography, or senior social science course). Grades were converted to a 13-point scale ranging from 13 ( A + ) to 1 (R = extensive remediation) using the coding system of Roeser and colleagues (1996). Letter grades have been established by the Ontario Ministry of Education and Training for use by District School Boards. Procedures University ethics approval and School Board approval for conducting research in schools were obtained. Principals within the school system were informed of the approval for the study by the School Board. Principals and Vice-Principals were then contacted and solicited for research participation. A l l contacts were made by this investigator. Initially, information about the study and the requirements of students and teachers were given to / 72 classroom teachers at a grade level meeting or at a department meeting in early M a y . Subsequently, interested teachers were contacted in order to answer questions they might have about the study, and to schedule classroom visits in which to introduce the research project to the students, distribute letters and consent forms, and complete questionnaires. Students were informed of the study in a classroom visit scheduled by this investigator and were invited to participate (see Appendix J). A parent permission letter and consent form (see Appendix K ) and student informed consent form (see Appendix L ) were given to students to take home. A s an incentive for adolescents to return their signed consent forms, either indicating they would participate or not, students who returned signed consent forms before data collection began, were eligible for a classroom draw for one $5.00 gift certificate from a local video store. Students whose parents did not permit them to participate, or students who chose not to participate, worked on independent assignments given by the teacher during the classroom session designated for completing the questionnaires. These assignments were determined by the classroom teacher before the questionnaires were distributed. The date for completing the questionnaires was scheduled in consultation with the classroom teacher within one to two weeks of the initial classroom visits, in which the letters describing the study and permission forms were distributed. During the administration session, student participants were given a package of questionnaires in an envelope. Before removing the questionnaires from the envelope, three practice questions, including two negatively worded items, and the different types of rating scales were presented to ensure that the participants understood the questioning formats and scales that they would encounter. Students were then asked to take the questionnaires from 73 the envelope. The cover page was read aloud by this researcher while students followed along (see Appendix M ) . Students were encouraged by the researcher to answer questions thoughtfully and honestly and told their opinions mattered and were important to improving education. In order to protect students' identities, each student was given an identification number (ID) that was assigned at the onset of the study. Students completed a student identification form that was included in the package (see Appendix N ) . This form was used to match students' and teachers' questionnaires, and school grades. The ID forms were collected by the researcher and kept separate from the questionnaires. Names were not recorded on the questionnaire. The questionnaires identified students only by an ID number. The questionnaires were administered to students under group testing conditions during the regular classroom instructional period. Students took approximately 30 to 60 minutes to complete the measures. The sets of questionnaires in the package were counterbalanced to control for a possible order effect. The packages were randomly distributed among students in each classroom. The students' and teachers' measures were selected so that they could be completed efficiently in the classroom setting during one instructional period. To encourage honest responses, students were assured of confidentiality. Individual assistance was offered to al l participants by either the researcher and/or her trained assistant as required, in order to help students understand and follow directions, or read the question. Whi le students completed their questionnaires, teachers completed the T - C R S for each student participating from that teacher's classroom. After the questionnaires were collected, participants were debriefed in a follow-up discussion by this researcher, who provided information about how the questionnaires related 74 to the main ideas that were being explored. Students were given the opportunity to ask questions and comment about whether they thought the ideas being investigated in the study were relevant to their own lives. 75 C H A P T E R F O U R Results The results of this research are presented in four sections. The first section describes the preliminary analyses used to screen the data prior to conducting the main analyses. The next section provides results of analyses examining gender and grade differences in autonomy and relatedness across the contexts of parents, peers, and school. The third section presents the results of the correlational analyses examining relations of autonomy and relatedness to school functioning and psychological adjustment. Final ly, the last section presents results of the regression analyses examining autonomy and relatedness in the contexts of parents, peers, and school as predictors of school functioning and psychological adjustment. Because both theory and research suggest that gender differences exist in relation to schooling and mental health (e.g., Al len et a l , 1990; Conger et a l . , 1997; H i l l & Holmbeck, 1986; Roeser et a l . , 1998; Wentzel, 1998), in prevalence rates for internalizing and externalizing problem behaviours (e.g., Scaramella, Conger, & Simons, 1999), and in the quality of adolescent-parental and adolescent-peer relationships (e.g., Geuzaine et a l . , 2000; Laible et a l . , 2000; Stern, 1990), analyses were conducted separately for boys and for girls. Preliminary Analyses Prior to statistical analysis, the data were examined using the procedure outlined by Tabachnick and Fide l l (1989) for screening data. SPSS Frequencies were used to check on accuracy of data entry, missing values, and normality of distributions. The shape of the distributions were examined using SPSS Graphs. Specifically, the data were searched for minimum and maximum values, and means and standard deviations were inspected for 76 plausibility. Out of range values were corrected by checking the original data sheets for accuracy of input. Both visual inspection of the distributions and the coefficients for skewness and kurtosis were examined to detect departures from normality (Hopkins & Weeks, 1990; Stevens, 1992; Tabachnick & Fide l l , 1989). According to Tabachnick and Fide l l (1989), "with large samples the significance levels of skewness and kurtosis are not as important as their actual sizes (worse the farther from 0) and visual appearance of the distribution" (p. 74). Thus, one way to screen for nonnormal variables, is to examine distributions of the variables. Fol lowing this strategy, the skewness and kurtosis values were found to be close to zero, and thus, were considered acceptable in the present study. Table 1 includes the means, standard deviations, skewness, kurtosis, and range for a l l the measures used in the study. The data were also examined for univariate outliers. A z score was computed for al l of the measures utilized in the present study (i.e., parental autonomy, peer autonomy, parental attachment, peer attachment, school autonomy, school belonging, internalizing problems, externalizing problems, total problem scores, teacher-rated school competencies, teacher-rated school problems, G P A ) . Following the procedure outlined by Tabachnick and Fide l l (1989) a univariate outlier was indicated by a z score greater than or equal to 3.67, at p_ = .001 criterion. F ive cases (or 1 % of the sample) were identified as univariate outliers. A s noted by Stevens (1992) and others (e.g., Tabachnick & Fide l l , 1989), in samples with more than 100 cases, a few standardized scores in excess of +3.00 can be expected simply by chance. Thus, in the present study, the cases with outliers were retained for al l data analyses. There was no reason to believe that the outlier cases were not part of, or members Table 1 Means, Standard Deviations, Skewness, Kurtosis, and Range for a l l Measured Variables Variable M SD Skewness Kurtosis Range Parental Autonomy (EAS) 40.17 6.47 -.40 .25 20-56 Deidealization (Easde) 13.81 2.85 -.14 -.25 6-20 Nondependency (EASnon) 11.66 2.21 -.31 -.22 5-16 Individuation (EASind) 14.71 3.10 -.48 -.03 5-20 Parental Attachment (PA) 97.01 19.56 -.41 -.13 35-139 Trust (PATrust) 37.12 8.05 -.58 -.21 13-50 Communication (PACom) 32.47 7.16 -.31 .06 10-50 Alienation (PAAlien) 20.58 6.75 ,33 -.30 8-38 Peer Autonomy (EASP) 38:24 5.20 .00 , -.08 23-53 Deidealization (EASPde) 14.30 2.26 .04 .13 7-20 Nondependency (EASPnon) 11.02 • 2.24 -.10 -.24 5-16 Individuation (EASPind) 12.93 2.92 -.09 -.39 6-20 Peer Attachment (PeerA) 97.73 15.94 -.66 .28 38-123 Trust (PrTrust) 40.33 7.58 -1.10 1.23 10-50 Communication (PrCom) 27.64 6.87 -.41 -.58 8-40 • Alienation (PrAlien) 16.24 4.64 .42 -.23 7-31 School Autonomy (PCSS) 59.33 11.33 -.44 .13 23-86 School Belonging (PSSM) 62.41 12.68 -.42 .04 20-89 Teacher-rated School Competencies (SchComp) 73.69 16.00 -.12 -.70 32-100 Teacher-rated School Problems (SchProbs) 26.91 11.06 1.68 2.85 18-79 Grade Point Average (GPA) 8.28 3.07 -.47 -.64 1-13 Student-rated Internalizing Problems (YSRint) 16.21 10.22 .82 .50 0-53 Student-rated Externalizing Problems (YSRext) 15.25 7.95 .66 .18 0-45 78 of, the intended population for which the results of this study w i l l generalize. The summary statistics in Table 1 include the univariate outlier cases. The data were screened for multivariate outliers using the strategy in which a dummy dependent variable is used to screen for outliers among a set of independent variables (Tabachnick & Fide l l , 1989). To screen for multivariate outliers, SPSS Regression was used to calculate the Mahalanobis distance (D 2) for each case using the case identification number as the dummy dependent variable, with grade, parental deidealization, parental nondependency, parental individuation, parental attachment, peer deidealization, peer nondependency, peer individuation, peer attachment, school autonomy, and school belonging as the independent variables. A n outlier was indicated by a D 2 that was significant at p < .001 level. Mahalanobis distance was calculated as chi-square with degrees of freedom equal to the number of observed variables in the hypothesized model (Tabachnick & F ide l l , 1989). For the current study, which utilized eleven independent variables for the regression analysis, the chi-square critical value at g < .001 was 31.26. Thus, i f a D 2 for a single case exceeded 31.26 it was considered an outlier. Three cases (or less than 1% of the sample) were identified as multivariate outliers having D 2 values of 31.41, 31.65, and 32.50. Cohen and Cohen (1983) suggest that i f outliers represent less than one or two percent of the total number of cases, they can remain in the data set. Therefore, it was decided that the three multivariate outlier cases would be retained for all further data analyses. 3 Next, the data were screened graphically in order to examine the degree to 3 Regression analyses were conducted with and without the multivariate outlier cases. Excluding the outlier cases did not change the results that are reported. 79 which the data met the assumptions of normality, linearity, and homoscedasticity between predicted scores on the dependent variable and errors of prediction. Scatterplots were obtained from SPSS Graphs in which standardized residuals were plotted against the standardized predicted values (e.g., Tabachnick & Fide l l , 1989). Scatterplots were examined and appeared normal, thus indicating that assumptions of normality, linearity, and homoscedasticity of residuals had been met. Gender and Grade Differences in Adolescents' Perceptions of Autonomy and Relatedness The purpose of this section is to present the results of analyses of gender and grade differences in parental autonomy, peer autonomy, parental attachment, peer attachment, school autonomy, and school belonging. Each of the hypotheses w i l l be addressed in turn. It w i l l be recalled that one goal of this study was to determine whether gender and grade differences exist regarding the way in which adolescents report their experiences of autonomy and relatedness in parent, peer, and school contexts. Previous research on emotional autonomy and parent and peer attachment (using the same measures of autonomy and attachment) have used the total scale scores as wel l as subscales (e.g., Armsden & Greenberg, 1987; Chen & Dornbusch, 1998; Cotterell, 1992; Frank et a l . , 1990; Isakson & Jarvis, 1999; Lamborn & Steinberg, 1993; Steinberg & Silverberg, 1986; Steinberg, 1988). Both total scale scores and subscale scores were examined in the present study. Composite scores provide a more generalized measure of the variables of interest than would be obtained from individual subscales (Ghiselli , Campbell, & Zedeck, 1981). In the present investigation, the composite scales have demonstrated acceptable reliabilities greater than a = .60 to justify their use over individual subscales 80 (Murphy & Davidshofer, 1998). The results of subscale analyses have also been reported in this study because some researchers have suggested that subscales might reveal different aspects of adolescent functioning that may not otherwise be captured through the use of the composite measure (e.g., Chen & Dornbusch, 1998; Frank et a l . , 1990; Silverberg & Gondoli , 1996). Reliabilities for each of the subscales have been reported in the Methods section. In the present investigation, I first examined composite scale scores in the analyses. For these analyses, separate 2 (gender) x 3 (grade) analysis of variances ( A N O V A s ) were conducted using the total scale scores of the measures assessing parental autonomy, peer autonomy, parental attachment, peer attachment, school autonomy, and school belonging. Gender and grade were the independent variables or between-subject factors. Next, I explored the multidimensional characteristics of autonomy in relationships with parents and peers, and attachment to parents and peers using multivariate analyses of variance ( M A N O V A s ) . For these analyses, separate 2 (gender) x 3 (grade) M A N O V A s were conducted with the subscale scores for emotional autonomy from parents (i.e., deidealization, nondependency, individuation), attachment to parents (i.e., trust, communication, alienation), emotional autonomy from peers (i.e., deidealization, nondependency, individuation), and attachment to peers (i.e., trust, communication, alienation). To examine significant differences obtained among grade level means, the Tukey honestly significant difference (HSD) post hoc procedure was used to make al l pairwise comparisons at p. < .05 criterion. 81 Gender and grade differences in parental autonomy. It was hypothesized that adolescents in grade 11 would show higher levels of parental autonomy than would adolescents in grades eight or nine (Hypothesis 4a). That is, it was expected that older adolescents would be more emotionally autonomous from their parents than younger adolescents (e.g., Lamborn & Steinberg, 1993; Ryan & Lynch , 1989; Steinberg & Silverberg, 1986). Because previous research results have been equivocal with respect to differences between boys and girls on scores of emotional autonomy from parents (e.g., Lamborn & Steinberg, 1993; Steinberg & Silverberg, 1986; Ryan & Lynch , 1989), no hypothesis regarding gender was put forth. To examine differences in parental autonomy, a 2 (gender) x 3 (grade) A N O V A was performed. For this analysis, the total score on the Emotional Autonomy Scale (EAS) served as the dependent variable. Table 2 provides the means and standard deviations for the total scale and subscale scores for parental autonomy for gender and grade respectively. The A N O V A revealed a nonsignificant main effect for gender, F ( l , 472) = 2.81, ns; a nonsignificant main effect for grade, F(2, 472) = 3.44, p_ < .05; and a nonsignificant gender x grade interaction, F(2, 472) = 1.95, ns. Thus, the results of the A N O V A did not reveal significant gender or grade differences in parental autonomy when using the total E A S score. Boys and girls did not differ from one another in their perceptions of emotional autonomy from parents. In contrast to what was hypothesized, no grade differences were found. To further explore parental autonomy, a 2 (gender) x 3 (grade) M A N O V A was conducted with the three subscales from the E A S . For this analysis, scores for Table 2 82 Gender and Grade Means and Standard Deviations for Parental Autonomy Gender Boys Girls" Variable M S D M S D Deidealization 13.43 2.84 14.11 2.82 Nondependency 11.66 2.29 11.66 2.16 Individuation 14.53 3.23 14.85 2.99 Total Scale 39.62 6.53 40.61 6.41 Grade Grade 8C Grade 9d Grade I P M S D M S D M S D Deidealization 13.59 2.92 13.89 2.71 13.96 2.94 Nondependency 11.27 2.19 11.79 2.18 11.96 2.22 Individuation 14.38 3.08 14.95 3.12 14.82 3.09 Total Scale 39.24 6.22 } 40.63 6.39 40.74 6.78 an = 213. b n = 265. c n = 170. d n = 167. e n = 141. 83 Deidealization, Nondependency, and Individuation were the dependent variables, and gender and grade were the independent variables. The results of this analysis revealed a significant multivariate main effect for gender, F(3 , 470) = 2.68, p < .05, Wi lks ' s lambda = .983; a nonsignificant multivariate main effect for grade, F(6, 940) = 1.73, ns, Wi lks ' s lambda = .978; and a nonsignificant multivariate gender x grade interaction, F(6, 940) = 1.15, ns, Wi lks ' s lambda = .985. Fol lowing one of the recommended procedures typically used in multivariate research, significant multivariate effects were followed-up with univariate F tests in order to identify which of the subscales contributed to the overall main effect (Bray & Maxwel l , 1982; Stevens, 1992). Results of the follow-up univariate analyses revealed that gender was significant on the Deidealization subscale, F ( l , 472) = 6.72, p = .01. A s can be seen in Table 2, girls scored significantly higher than boys on this subscale. There were no differences between boys and girls on either the Nondependency subscale, F ( l , 472) = 0.00, ns, or the Individuation subscale, F ( l , 472) = 1.21, ns. The result from the M A N O V A indicates that girls, to a greater extent than did boys hold less idealized images of the parents and acknowledge that their parents may not always be right or share the same opinions. Gender and grade differences in peer autonomy. Next, it was of interest in this investigation to determine whether gender and grade differences exist in adolescents' perceptions of autonomy in relationships with peers. Specifically, it was hypothesized that when compared to boys, girls would report higher levels of peer autonomy (Hypothesis 1). That is, it was expected that girls would be more emotionally autonomous from their peers 84 because studies suggest that girls are less susceptible than boys to the influence of peers to conform with respect to antisocial pressure from peers (e.g., Berndt, 1979; Brown et a l . , 1986; Steinberg & Silverberg, 1986). Furthermore, it was hypothesized that adolescents in grade 11 would show higher levels of peer autonomy than would adolescents in grade eight or nine (Hypothesis 4a) because studies have shown that conformity to peer pressure decreases with age across early and middle adolescence (e.g., Berndt, 1979; Coll ins , Gleason et a l . , 1997; Gavin & Furman, 1989; Steinberg & Silverberg, 1986). Thus, peer autonomy was expected to increase across grade levels. To examine differences in peer autonomy, a 2 (gender) x 3 (grade) A N O V A was performed. For this analysis, the total score on the Emotional Autonomy Peer Scale (EASP) served as the dependent variable. Refer to Table 3 for the means and standard deviations for the total scale and subscale scores for peer autonomy for gender and grade respectively. The A N O V A yielded a nonsignificant main effect for gender, F ( l , 472) = 0.04, ns; a nonsignificant main effect for grade, F(2, 472) = 1.61, ns; and a nonsignificant gender x grade interaction, F(2 , 472) = 0.09, ns. N o differences were found between boys and girls, or across grade levels regarding peer autonomy. Thus, the hypotheses that gender and grade differences would exist in peer autonomy was not supported by the A N O V A when using the E A S P total score. To further explore autonomy from peers, a 2 (gender) x 3 (grade) M A N O V A was conducted with the three subscales from the E A S P . For this analysis, scores for Deidealization, Nondependency, and Individuation were the dependent variables, and gender and grade were the independent variables. The results of this analysis revealed a significant Table 3 85 Gender and Grade Means and Standard Deviations for Peer Autonomy Gender Boys" Gi r l s b Variable , M S D M S D Deidealization 14.09 2.22 14.46 2.28 Nondependency 11.40 2.24 10.72 2.20 Individuation 12.85 2.90 12.98 2.93 Total Scale 38.35 4.93 38.16 5.42 Grade Grade 8C Grade 9 d Grade l l e M S D M S D M S D Deidealization 13.95 a 2.33 14.37 2.25 14.63 b 2.14 Nondependency 11.28. 2.42 10.63 b 2.14 11.17 2.09 Individuation 13.30 3.07 12.63 2.84 12.82 2.78 Total Scale 38.52 5.68 37.64 5.07 38.62 4.70 Note. In each row, the means with different subscripts indicate significant differences using Tukey's H S D procedure, p < .05. a n = 213. b n = 265. c n = 170. d n = 167. e n = 141. 86 multivariate main effect for gender, F(3 , 470) = 6.62, p < .001, Wi lks ' s lambda = .959; a significant multivariate main effect for grade, F(6, 940) = 4.13, p < .001, W i l k s ' lambda = .949; and a nonsignificant multivariate gender x grade interaction, F(6, 940) = 0.23, ns, Wi lks ' s lambda = .997. Results of the follow-up univariate analyses revealed gender and grade differences on two of the three subscales. A s can be seen in Table 3, girls scored significantly higher than boys on the Deidealization subscale, F ( l , 472) = 3.87, p = .05. Boys, in turn, had significantly higher scores than girls on the Nondependency subscale, F ( l , 472) = 10.23, p = .001. N o difference was found between boys and girls on the Individuation subscale F ( l , 472) = 0.30, ns. A s can be seen in Table 3, significant grade effects were found for Deidealization, F(2 , 472) = 4.21, p < .05, and for Nondependency, F(2, 472) = 3.52, p < .05, but not for Individuation F(2, 472) = 2.23, ns. Tukey H S D post hoc tests for grade revealed that adolescents in grade eight scored significantly lower than adolescents in grade 11 on Deidealization. For Nondependency, adolescents in grade eight scored significantly higher than adolescents in grade nine. In summary, the results of the M A N O V A examining peer autonomy, showed that girls and boys differed with respect to their scores of deidealization and nondependency. Whereas girls had higher levels of deidealization than did boys, boys scored higher on nondependency than did girls. When compared to adolescents in grade eight, adolescents in grade nine had lower scores of nondependency on peers. When compared to adolescents in grade eight, adolescents in grade 11 had higher scores of peer deidealization. 87 The results of this M A N O V A provide partial support for the hypothesis that girls would have higher levels of peer autonomy than would boys. More specifically, higher scores on the Deidealization subscale indicate that girls, to a greater extent than do boys, hold less idealized images of their peers and acknowledge that their peers may not always be right or share the same opinions. However, contrary to the hypothesis, boys had higher scores than girls on the Nondependency subscale, indicating that boys depend less on their peers than do girls. Regarding grade differences, grade eight adolescents had lower scores on the Deidealization subscale than adolescents in grades nine and 11, which is consistent with the hypothesis that adolescents in grade 11 would have higher levels of peer autonomy than adolescents in grade eight or grade nine. That is, older adolescents would be more emotionally autonomous from their peers than younger adolescents and more likely to hold more mature perceptions of their peers. However, contrary to the previously stated hypothesis, grade eight adolescents reported greater nondependency in their relationships with peers than adolescents in grade nine, suggesting that adolescents in grade eight were less dependent on their peers. Gender and grade differences in parental attachment. Several differences were expected regarding adolescents' perceptions of their attachments to parents. Because research findings suggest that feelings of being connected and emotionally supported in relationships are especially important to girls (e.g., Eccles et a l . , 1997; Goodenow, 1993a; Laible et a l . , 2000; Stern, 1990), it was hypothesized that when compared to boys, girls would have higher levels of parental attachment (Hypothesis 2). Furthermore, it was hypothesized that adolescents in grade 11 would have lower levels of parental attachment 88 than adolescents in grade eight or grade nine (Hypothesis 4b). Studies have shown that parental attachments decrease as adolescents mature (e.g., Blain et a l . , 1993; Coll ins , Gleason et a l . , 1997; Lamborn & Steinberg, 1993; Papini et a l . , 1991). To examine differences in parental attachment, a 2 (gender) x 3 (grade) A N O V A was performed. For this analysis, the total score for the parental attachment scale (PA) of the Inventory of Parent and Peer Attachment (IPPA) served as the dependent variable. Table 4 provides the means and standard deviations for the total scale and subscale scores for parental attachment for gender and grade respectively. The A N O V A yielded a nonsignificant main effect for gender, F ( l , 472) = 0.77, ns; a nonsignificant main effect for grade, F(2, 471) = 1.78, ns; and a nonsignificant gender x grade interaction, F(2 , 472) = 0.97, ns. The results of this analysis failed to find significant differences between boys and girls, or differences across grade levels on scores for parental attachment. Thus, the A N O V A results did not support the hypotheses regarding gender and grade differences in parental attachment when using the total score from the I P P A . To further explore attachment to parents, a 2 (gender) x 3 (grade) M A N O V A was conducted with the three subscales from the I P P A associated with parental attachment. For this analysis, scores for parental Trust, Communication, and Alienation were examined as the dependent variables, and gender and grade were the independent variables. The M A N O V A revealed a significant multivariate main effect for gender, F(3 , 470) = 6.55, p < .001, Wi lks ' s lambda = .960; a nonsignificant multivariate main effect for grade, F(6, 940) = 0.93, ns, Wi lks ' s lambda = .988; and a nonsignificant multivariate gender x grade interaction, F(6, 940) = 0.90, ns, Wi lks ' s lambda = .989. Table 4 89 Gender and Grade Means and Standard Deviations for Parental Attachment Gender Boys" Gi r l s b Variable M S D M S D Trust 37.80 7.70 36.57 8.30 Communication 32.02 7.02 32.83 7.26 Alienation 20.01 6.48 21.03 6.94 Total Scale 97.81 18.65 96.37 20.27 Grade Grade 8C Grade 9 d Grade l l e M S D M S D M S D Trust 37.87 7.66 37.05 7.90 36.30 8.65 Communication 33.21 7.13 32.59 6.37 31.43 7.96 Alienation 20.38 6.95 20.49 6.80 20.91 6.48 Total Scale 98.69 19.40 97.15 18.87 94.82 20.45 a n = 213. b n = 265. c n = 170. d n = 167. e n = 141. 90 Results of the follow-up univariate analyses did not reveal any gender differences on the Trust subscale, F ( l , 472) = 3.07, p = .08, on the Communication subscale, F ( l , 472) = 1.31, p > .20, or on the Alienation subscale, F ( l , 472) = 2.80, p = .10. A s can be seen in Table 4, an inspection of the means of the subscales showed that relative to boys, girls had higher scores relating to parental communication and feelings of alienation, and lower scores of trust. Although the M A N O V A result was significant, the univariate results failed to reveal on which of the three subscales boys and girls were different. A significant M A N O V A finding does not necessarily mean that there w i l l be any significant A N O V A results (Huberty & Morr is , 1989). The univariate results do not take into account the correlations among the variables. Thus, univariate Fs may yield different results because each variable is considered separately and not in combination with others (Bray & Maxwel l , 1982; Stevens, 1992). The M A N O V A results examining parental attachment indicated that boys and girls differed significantly in their perceptions of their attachments to parents when scores on the Trust, Communication, and Alienation subscales of the I P P A are considered jointly. This result provides some evidence that boys and girls show differences in their attachments to their parents. Gender and grade differences in peer attachment. Adolescents were expected to differ in their perceptions of their attachments to peers. Because theory and research findings suggest that girls place more importance on security and support in peer relationships (e.g., Douvan & Adelson, 1966; Eccles et a l . , 1997; Goodenow, 1993a; Oldenburg & Kerns, 1997), it was hypothesized that girls would report higher levels of peer attachment than 91 would boys (Hypothesis 2). Furthermore, it was hypothesized that adolescents in grade 11 would show lower levels of peer attachment than adolescents in grade eight or grade nine (Hypothesis 4b). That is, it was expected that peer attachment would decrease across grade levels because studies have shown there are declines in peer conformity and social support by late adolescence (e.g., Coll ins, Gleason, et a l . , 1997). To examine differences in peer attachment, a 2 (gender) x 3 (grade) A N O V A was performed. In this analysis, the total score for the peer attachment scale (PeerA) of the Inventory of Parent and Peer Attachment (IPPA) served as the dependent variable. Table 5 provides the means and standard deviations for the total scale and subscale scores for peer attachment for gender and grade respectively. The A N O V A yielded a significant main effect for gender, F ( l , 472) = 21.54, p_ < .001; a significant main effect for grade, F(2, 471) = 4.12, p < .05; and a nonsignificant gender x grade interaction, F(2 , 472) = 0.04, ns. A s can be seen in Table 5, girls scored higher than did boys on peer attachment when using the I P P A total score. This result is consistent with the hypothesis that girls would report higher levels of peer attachment than would boys. Regarding grade differences, Tukey H S D post hoc tests for grade revealed that grade eight adolescents scored significantly lower than grade nine adolescents on peer attachment (see Table 5). This result is contrary to the expectation that scores for peer attachment would show a decrease across grades. To further explore attachment to peers, a 2 (gender) x 3 (grade) M A N O V A was conducted with the three peer subscales of the Inventory of Parent and Peer Attachment Table 5 92 Gender and Grade Means and Standard Deviations for Peer Attachment Gender Boys" Gi r l s b Variable M S D M S D Trust 39.38 7.42 41.09 7.64 Communication 24.49 6.85 30.17 5.75 Alienation 15.81 4.41 16.59 4.79 Total Scale 90.07 15.26 96.67 15.90 Grade Grade 8C Grade 9 d Grade I P M S D M S D M S D Trust 39.08 a 8.48 41.27 b 6.97 40.71 6.95 Communication 26.51 a 7.13 28.51 b 6.63 27.98 6.67 Alienation 16.45 4.72 15.89 4.45 16.40 4.75 Total Scale 91.14 a 17.18 95.88 b 14.70 94.30 20.45 Note. In each row, the means with different subscripts indicate significant differences using Tukey's H S D procedure, p < .05. "n = 213. b n = 265. c n = 170. d n = 167. c n = 141. 93 ( IPPA) . For this analysis, scores for peer Trust, Communication, and Alienation were the dependent variables, and gender and grade were the independent variables. The results of this analysis revealed a significant multivariate main effect for gender, F(3 , 470) = 49.50, 2 < .001, Wi lks ' s lambda = .760; a significant multivariate main effect for grade, F(6, 940) = 2.27, p < .05, Wi lks ' s lambda = .972; and a nonsignificant multivariate gender x grade interaction, F(6, 940) = 0.17, ns, Wi lks ' s lambda = ,998. Results of the follow-up univariate analyses revealed significant gender and grade differences on two of the three subscales. A s can be seen in Table 5, a significant gender effect was obtained for the Trust subscale, F ( l , 472) = 6.27, p < .05, the Communication subscale, F ( l , 472) = 100.30, p < .001, but not for the Alienation subscale, F ( l , 472) = 3.69, p = .06. Specifically, girls had significantly higher scores than boys on the Trust and the Communication subscales. A s can be seen in Table 5, significant grade differences were also found for Trust, F(2 , 472) = 3.93, p < .05, and Communication, F(2 , 472) = 5.55. p < .01, but not for Alienation, F(2, 472) = 0.81, ns. Tukey H S D post hoc tests for grade revealed that grade eight adolescents had significantly lower scores than grade nine adolescents on both these subscales at the .05 criteria. In summary, the results of the M A N O V A showed that boys and girls differed regarding their attachments to peers. As expected, girls reported a significantly higher quality of relationship with peers in perceived mutual trust and communication than did boys. One unexpected result was that adolescents in grade nine reported a higher degree of trust and communication in their relationships with peers than adolescents in grade eight. It was expected that peer attachment would decrease from grade eight to grade nine. 94 Gender and grade differences in school autonomy. Grade level differences were expected regarding adolescents' perceptions of school autonomy. Because previous findings suggest increases in teacher control and,corresponding decreases in opportunities for decision-making and choice across grade levels (e.g., Eccles, Buchanan, et a l . , 1991; Eccles, Lo rd , et a l . , 1991; Eccles et a l . , 1984; Eccles & Midgley, 1989; Midgley & Feldlaufer, 1987; Midgley et a l . , 1988; Wil lower et a l . , 1973), it was hypothesized that the level of school autonomy would show a decrease from grade eight to grade 11 (Hypothesis 3). N o hypothesis was put forth regarding differences between boys and girls in their perceptions of school autonomy. To examine differences in school autonomy, a 2 (gender) x 3 (grade) A N O V A was performed. For this analysis, the total score on the Perceived Control at School Scale (PCSS) was used as the dependent variable. Table 6 provides the means and standard deviations for school autonomy for gender and grade respectively. The A r ^ O V A yielded a nonsignificant main effect for gender, F ( l , 472) = 1.91, ns; a nonsignificant main effect for grade, F(2, 472) = 1.31, ns; and a nonsignificant gender x grade interaction, F(2 , 472) = 1.62, ns. The results revealed an absence of gender and grade differences on scores of perceived control at school. In contrast to what was hypothesized, adolescents' perceptions of school autonomy did not show a decrease across grades eight, nine, and eleven. Gender and grade differences in school belonging. Next, it was of interest to determine whether boys and girls differed in their perceptions of school membership or belonging at school. Because previous research has documented gender differences in school belonging, with girls reporting greater belonging than boys (e.g., Goodenow, 1993a; / Table 6 95 Gender and Grade Means and Standard Deviations for School Autonomy and School Belonging Variable Gender Boys" M S D Gi r l s b M S D School Autonomy School Belonging 58.43 11.83 61.78 13.16 60.05 10.88 62.92 12.29 Grade Grade 8C Grade 9 d Grade l l e M S D M S D M S D School Autonomy School Belonging 59.05 12.23 63.66 14.03 60.34 9.58 62.56 10.71 58.46 12.08 60.74 13.03 "n = 213. b n = 265. °n = 170. d n = 167. e n = 141. 96 1993b), it was hypothesized that girls would report higher levels of school belonging than would boys (Hypothesis 2). N o hypothesis was put forth regarding grade differences. To examine differences in school belonging, a 2 (gender) x 3 (grade) A N O V A was performed. For this analysis, the total score on the Psychological Sense of School Membership (PSSM) scale was used as the dependent variable. Table 6 provides the means and standard deviations for school belonging for gender and grade respectively. The A N O V A revealed a nonsignificant main effect for gender, F ( l , 472) = 0.57, ns; a nonsignificant main effect for grade, F(2, 472) = 1.74, ns; and a nonsignificant gender x grade interaction, F(2, 472) = 1.76, ns. Thus, the hypothesis that girls would report higher levels of school belonging than would boys was not supported. Summary. One purpose of the present study was to examine differences in adolescents' perceptions of their experiences of autonomy and relatedness in parental, peer, and school contexts. Four hypotheses were posited in which differences between boys and girls, and differences across grades eight, nine, and eleven were examined. The first hypothesis was that girls would report higher levels of peer autonomy than boys (Hypothesis 1). The results of the total scale analyses showed that girls did not differ from boys in their perceptions of peer autonomy, however, gender differences were found in the subscale analyses. Specifically, when compared to boys, girls deidealized their peers to a greater extent, thus acknowledging that their peers may not always be right or that they share the same opinions. Contrary to the hypothesis, results regarding nondependency indicated that when compared to girls, boys were less dependent on their peers. The second hypothesis was that, when compared to boys, girls would report higher 97 levels of attachments to parents and peers, and belonging at school (Hypothesis 2). The results of the total scale analyses supported the hypothesis that girls had higher levels of peer attachment. From the subscale analyses, it was found that girls also reported more trust and communication in their relationships with their peers. However, the analyses did not show that girls had higher levels of parental attachment or higher levels of school belonging in comparison to boys. The third hypothesis was that the level of school autonomy would show a decrease from grade eight to grade eleven (Hypothesis 3). No differences across grade levels were found to support this claim. Moreover, there were no gender differences found. - The fourth hypothesis was that adolescents in grade eleven would show higher levels of parental autonomy and peer autonomy (Hypothesis 4a), and lower levels of parental attachment and peer attachment than adolescents in grade eight or grade nine (Hypothesis 4b). Results of the total scale analyses did not support the hypotheses. Regarding autonomy, one finding emerged from the subscale analyses that supports the study's hypotheses with respect to grade level differences in peer autonomy. Specifically, when compared to adolescents in grade eight, adolescents in grade 11 deidealized their peers to a greater extent. This result provides some support for the hypothesis that adolescents in grade 11 would be more autonomous from their peers than adolescents in grade eight. One finding emerged from the subscale analyses which was contrary to the hypothesis with respect to grade level differences in autonomy. Specifically, adolescents in grade nine were found to be more dependent on their peers than adolescents in grade eight, whereas less 98 dependency was expected. Regarding attachment, one rinding from the total scale analyses was contrary to the hypothesis with respect to peer attachment. Specifically, when examining the total score for peer attachment, grade nine adolescents reported being more attached to their peers than adolescents in grade eight. Furthermore, the subscale analyses revealed that adolescents in grade nine had higher levels of mutual trust and communication in their peer relationships in comparison to adolescents in grade eight. Thus, contrary to expectations, both the total scale analyses and subscale analyses showed that adolescents in grade nine reported better quality of peer relationships in perceived mutual trust and communication than adolescents in grade eight, whereas lower levels were expected. Relations of Autonomy and Relatedness to School Functioning and Psychological Adjustment The purpose of this section is to present the results of the correlational analyses. It w i l l be recalled that the second major goal of-this study was to determine the associations among measures of autonomy, relatedness, school functioning, and psychological adjustment across different contexts. For these analyses, Pearson product-moment correlations were calculated to examine the relations between parental, peer, and school autonomy, relatedness to parents, peers, and school, G P A , teacher-rated school competencies, and internalizing and externalizing problems. For the correlational analyses, the significance level was set at p < .01 due to the large number of correlations being calculated (cf., Brown et a l . , 1986; Foley & Epstein, 1992). Intercorrelations among measures for boys and for girls are reported in Table 7 4 . The correlations were inspected for consistency in the way boys and girls reported their experiences of autonomy and relatedness within the contexts of parents, peers, and school. A s can be seen in Table 7, the patterns for boys and girls were similar. Fisher's Z transformation was performed on the correlations between boys and girls and evaluated for significance at p = .01 against the critical value of z = 2.58 (Downie & Starry, 1977). Results of the Fisher Z transformations revealed that al l the correlations between boys and girls were not statistically different. Table 7 was also inspected for the intercorrelations among the subscales and total scores associated with measures of parental autonomy, parental attachment, peer autonomy, and peer attachment. Each of the subscales was significantly and strongly related to the composite scale and the intercorrelations among the subscales varied, with generally stronger correlations among the attachment subscales and low to moderate correlations among the autonomy subscales. Moreover, given the suggestion made in the literature that autonomy and attachment are negatively associated (e.g., Ryan & Lynch, 1989), the correlations were inspected for evidence of this relationship. With the exception of alienation, which was positively related to individuation, dimensions of autonomy (i.e., deidealization, nondependency) and 4 The correlations reported in Table 7 were computed with grade levels combined. Intercorrelations between measures at each grade level can be found in Appendix 0. Results of the Fisher Z transformations revealed that only one correlation was significantly different between grades, which was less than would have been expected by chance. The correlation between perceived control at school (PCSS) and teacher-rated school competencies (SchComp) was found to be significantly different between grade eight and grade eleven. 1/1 • r t o T3 1/3 >, O t/3 s c o 6 < vi C _o * r t 13 fc o o a c CN es o V O eS v to -H Tt "-i © es Tt Tt JS - H r-o a© to ov to vt to >/-) in to tO ITS Tt O Tt 00 Tt 00 w-i Tt rt i-l Vi to to tO ts VO V O o ts © 1™H O 00 o o oc ts irt rt O »H t~- in Tt CO rt O N ts oc vi vs CO to Tt Tt VS Tt r-o O N ts r- © O rt ON Tt in oo m O i—I rt rt i—I o Tt O i—I rt i—I O S O es o es o Tt o © o es 00 ON rt Tt es O es O Tt O ^ i—i i—i i—i o o 00 rt o © V O © N O © in © ON © ' m I O © rt i—I VO ts © i-H O N f l fS Vl co o co O N Tt CO Tt V O Tt ON CO ts Tt X co Tt rt O N tS rt ON r- © CO rt rt © © tN © O N ts CO Tt Vl to Tt ts 00 IT, oo r-tO rt V O ts © Tt vi Tt NO r-ts vt ts Tt 00 vo 00 to to to ON to © rt Vl t— to V O NO Vi ts ts ts V O © V O © © 1—I r- i—H © © es © Tt es NO —i 00 ^ rt —I Tt ts 00 to ON to vt 00 00 00 ts O N ts © V O © r- vo © rt Tt rt to to vi ts 1/5 rt to vi ts ts to ON Tt N O O N Ov vo Tt ts Tt to Tt ts © 1—I © m N O vo i-c © rt Tt to Ov. ts Vi ts 00 to to Vi 00 ts ts to to Vt to ON Tt 00 © rt V O Tt Tt O N © N O © © Tt , © V O ts t- ov N O tS rt rt 00 to O N ts N O ts 00 Tt Tt ts NO Tt 00 ts ON ts es © ts ts 00 © NO ts to ts ts ts © ts Vi to i—i to rt r- ^ es Tt © )£, S ^ o © i—i IT; ts © 00 © in © m rt © © vo © ts 00 rt © rt ,— ts ts ts <n © to Vl rt es 00 ts rt © © r-© m © o in © rt Vi vo ts VO ts r-ts Vl t-Tt ON ts r-ts Vi ts to to to NO V O rt oo r-Tt V O © r-© . Vi ts VO ts ts Vi O N Vl o Ov ON r-Vi g vo cn © irv ts to ts to ts w-> rt © Tt es es to to ts Tt V O Vi -H to V O © © Vi vo V O Tt O N Tt NO NO •o Tt r- rt o ^ ^ ^-i Tt ts 00 © © -rtI to to r- rt Tt rt o Vi s r- ts Ov 00 00 to 00 r- Tt rt © rt <s Vl es N O ts Tt tO rt ts Vl ' to ts ts Vl rt ts VI 00 © Vl © to rt Vl ts Tt es to Ov ts 00 © V O Tt tO rt 100 Ov to Vl es to ts ts vp m © 1—I VO © tS rt to © tn © rt tS © vo rt r~ © NO Tt rtl »H V©. © to ts © r~ r~ rt to to V O X rt © Tt Tt i« m i - ~ r t N « f - l f S l / 1 X T t T t O N © ' ~ l ' H 9 0 e S > / 1 f O 00 Tt Tt ' Tt Tt ^ 2 rt tS tS © © © tS tS tO © © tS tO . . . I* I* I* * " I I • I I I ? 5 . ^ 5 Tt K tS O rt 3 S S © © P O rt rt © © rt SS . ? g U a ? 8 8 3 8 S 8 . 2 s a ; 8 8 0 . 2 rt • • • • • • * • - I l l 1 1 1 . g 5 So vo 3 3 8 2 s a s s s s s s s s s s s *s r • i i i I I i l § i ^ s l l « l § l » * I < i l S S S 2 2 2 2 2 i s a a s < 5 : Q . o . Q . o . o . w o > - > -o ^ - c M t n T r m c o r - S S S S R ! » - C M e o r t i n < D h - o o c » T - T - ' ^ ' « - ' ^ T - ^ - » - ' « - f - « - " « ^ > C/3 u C3 o o 5 CU c o q <u X 3 .3 13 e o U 13 3 o SP •o s •8 0 0 1-4 .o e o • i-H •4—> o o a 3 o SP T3 <U •4—» o T3 x> CA >, O X ) i-i c .2 rt 1 o U o • i-H 3 to • 1-H • 1-H X5 0) ~ S3 ^ o CA 3 o Vi >> o X i l-l o PU "8 % fe X 3 oo X3 CD CU xs i i O PH I O q OJ X 5 o • i-H 3 6 0 •1-H fe X3 Wt cn c o 13 fc o o CU q <u -4—1 rt o • i - H 3 6 0 •I—I O i S3 fe X 3 6 0 • 1-H X 3 0O S © 1 ) X S rt • i-H 3 CU fe X 3 6 0 x: •a vo CA e o 1 o o <U q X 3 rt 3 rt o • i—t 3 6 0 <U X 5 6 0 X 5 T3 CA c o o • ^ H C 6 0 rt 13 rt I O o CD o CA OH fc -s o o CA 5b <D 13 o 00 >> s o c 2 3 < c o '4—> o S pq 00 pq <D O q rt o <4H '8 6J0 CD S3 fe xs &o X 5 o cs CA c o ' I fc o o < PH 13 o CA X i 3 CA a o 5 CA '4^ F? 4-> nl rt 3 •a -a c 00 < pq •a .S 00 < iT 13 o CA X i 3 CA >% O § •a § & 3 o Z 00 < pq c o a 00 < W cu 13 o CA JO 3 CA e _o '4—» rt N 1 "<U Q 00 < pq CU -o 00 < pq cu 13 o CA JO 3 CA 3 o 1 4-» rt o • H a 3 6 6 o U § S3 OH B o O < CH hi 13 o Vi JO 3 CA 4-> CA 5 • § S3 OH 2 H < PH i f 13 o CA e X 5 O IS CU S3 OH < CH PH PH pq c u T3 PH < pq i r 13 o OH £ c u 13 o >> B o a o 3 < 13 c o *4-» o 6 pq PH 00 < pq j u 13 o CA JO 3 CA c *4—» rt C c u S 3 s OH PH PH < PH i r 13 o CA X 3 3 CA a 0 "4—> rt 3 - 3 ^ 3 1 £ PH 00 < pq T 3 C PH 00 < pq iT 13 o CA X i 3 CA O a CD § & T3 C o PH 00 < pq c o c PH 00 < pq <u 13 o CA JO 3 CA c .2 *4—> rt N 3 .2 rt § 8 § • i - H < t-H PH X 3 3 CA c 0 '4^ rt O • i-H 3 3 6 £ o U V-l OH 6 o U I-I PH ii 13 o CA X i 3 CA 101 < < PH cu <u OH CA 3 cu 13 o CA a CU S x: o 4-» rt OH 6 o u X 3 o 00 CD JO B CU I-H o o X ! o 00 o c u CA 3 c u 00 13 o "5b "o X 3 O > > CA PH 00 O O PH <S 13 o 00 5 -s •5 1 S 8 -a Q oo xs o 00 4-J rt s 4-> 3 O U "8 > o l-H CU PH 00 00 U PH t u 13 o CA X i 3 CA 8 PH 00 hi 13 0 CA JO 3 CA OJ) 3 • ^ H N •T—I 13 e 5J 3 O OH ST Oi CU 00 3 O 3 3 00 hi 2 (U c • i-H O CU 6 0 < PH a c/T c u • i - H o I CU OH 6 o CJ "o O X 3 O CA 3 rt c u X S o 5 c u 13 o CA X i 3 CA (SO 3 • i - H N • i - H 13 s <D tJ O OH w (U 00 x: 4—> 3 O >< vo CS J 3 1 cn CS II . 3 1 102 relatedness (i.e., trust, communication) were related negatively both for boys and for girls, suggesting there is some disparity between autonomy and relatedness in both contexts of parents and peers. Also of interest was the correlation between school autonomy and school belonging. Research suggests that student autonomy and influence at school is positively linked with a sense of school community and belonging (e.g., Battistich, Solomon, Watson, & Schaps, 1997). A s can be seen in Table 7, perceived control at school was positively correlated with school belonging. Moreover, for autonomy, school autonomy was negatively related to dimensions of parental autonomy and peer autonomy. Whereas positive correlations were found between parental autonomy (i.e., deidealization, individuation) and peer autonomy (i.e., deidealization, individuation), peer nondependency was negatively related to parental nondependency, indicating that less dependency on parents is associated with more dependency on peers. Additionally, the correlations among the four criterion variables (i.e., teacher-rated school competencies, G P A , internalizing problems, externalizing problems) were examined. Moderate and positive correlations were found between school competencies and G P A and between internalizing and externalizing problems. School competencies and G P A were negatively related to problems in psychological adjustment. Relations of parental autonomy to school functioning and psychological adjustment. It was hypothesized that parental autonomy would be associated positively with psychological 103 adjustment problems (Hypothesis 7). Table 8 5 presents the correlations of parental autonomy to school functioning and problems in psychological adjustment for boys and for girls. Results from the total scale analyses revealed that parental autonomy was significantly and positively related to internalizing and externalizing problem scores both for boys and for girls. A s can be seen in Table 8, several significant relations emerged with regard to the subscale analyses pertaining to parental deidealization, nondependency on parents, and individuation. Parental deidealization was significantly and positively associated with externalizing problems for girls, but not for boys. That is, girls who perceived that their parents may not always be right or shared the same opinions, also reported more externalizing problems. Nondependency was found to be significantly and positively related to externalizing problems both for boys and for girls. Thus, adolescents who reported being less dependent on their parents also reported more externalizing problems. Moreover, individuation from parents was found to be significantly and positively related to internalizing and externalizing problems for both genders. The Individuation subscale reflects adolescents' perceptions that parents do not fully understand or know them well (e.g., Chen & Dornbusch, 1998; Ryan & Lynch , 1989). This result indicated that those adolescents who report higher levels of individuation from parents, also report more problems. To further explore relations between autonomy from parents and school functioning, the relation of emotional autonomy from parents to teacher-ratings of adolescents' school competencies, and G P A were examined. As can be seen in Table 8, results revealed no 5 Table 8 is presented here for convenience and because it provides a concise summary of the observed relations. ( 104 c o cl • 1—I .E5 -a c I-H a O P Q oo o o IT) cs r t T t cn cn a -a 1» s CA ••5' < 13 o • rt o O O >, cn 1/5 >, 6 o c o < C Oh O S •a § & •o c o c,. o • 1—< t3 N • rt •s Q cw a c/5 O P Q CN O o oo o CN o I-l • rt a Vi O P Q ON O T t o r-o CN cn oo in cn e • rt c .2 o c P H 8 o 00 13 o 00 o H a o P Q o T t o o 00 CN OO 00 T t o cn B o e o rt < I PH c/5 c o U 1 Oh S o U JG O 00 Vi U • rt o c B <U Oh 6 o U 13 o JG o 00 PH a c 3 • b Vi B CU 1 I-l PH 00 C ' N • •—I 73 e d> rt c 8 P< 00 b co B 3 8 PH 00 c • rt N 13 £ & X W >*n NO CN Cl 2 q CN VI o o VI _C1 Dl 105 significant relations for boys or for girls. Relations of peer autonomy to school functioning and psychological adjustment. Next, it was of interest to examine relations of peer autonomy to school functioning and psychological adjustment. It was hypothesized that peer autonomy would be associated negatively with psychological adjustment problems (Hypothesis 8). Table 96 presents the correlations of peer autonomy to school functioning and problems in psychological adjustment for boys and for girls. There were no significant findings in the total scale, but there were two significant relations found in the subscales. Specifically, nondependency on peers was significantly and negatively related to self-reports of externalizing problems for boys and for girls. This result supports the hypothesis and suggests that those adolescents who are less dependent on their peers report fewer externalizing problems. One other significant finding revealed that individuation from peers was significantly and positively related to self-reports of internalizing problems both for boys and for girls. Although no specific a priori hypothesis was made regarding individuation from peers, this finding suggests that those adolescents who feel their peers do not really know or understand them, report more internalizing problems. Next, the relation between peer autonomy and school functioning was examined. A s can be seen in Table 9, there was one significant finding in the total scale that was supported by each of the subscales. Specifically, for boys only, G P A was significantly and positively related to peer autonomy. That is, boys who reported higher levels of autonomy in their 6 Table 9 is presented here for convenience and because it provides a concise summary of the observed relations. o T3 -CA CO < s 1 • i"H c o o 3 3 PH "3 O J 3 o 00 >, 6 o 3 o H—> 3 < 8 PH <4h o CA 3 _o 'H-H o U CH 00 •— s o 3 2 3 • < 3 o • i-H 3 '> •i-H 3 o 3 <D •a § •a 3 o 3 o 1 • i-H 1 12 Q CA TH •r-1 o CA >> o PQ a CA o PQ O o PQ cu 13 o 00 o o CQ > VO o o O Y O C S C S 00 C S Os C S OS o cn o i-H O i™H m CS 00 o m 00 o cs o o 00 o p-o cs vo cn f o U - 3 o 00 ^—-" CA <D "3 I CD OH 6 o U *o o . 3 o 00 < PH O 3 • ^ H P< 00 >* e o «H bib 3 N e 3 cs 0 cs vo o cn o 1—1 >—1 o cs x (D P< 0 0 e p 0 0 3 ' N •i-H •a I -4—* X pq 106 >n vo cs i 3 1 o q VI oi cn i-H C S 2 9 31 VI .01 107 relationships with peers (i.e., deidealization, nondependency, and individuation) had higher G P A s . Relations of parental attachment to school functioning and psychological adjustment. Next, it was hypothesized that parental attachment would be positively associated with school functioning (Hypothesis 6) and negatively associated with psychological adjustment problems (Hypothesis 10). Table 10 7 presents the correlations of parental attachment to school functioning and problems in psychological adjustment for boys and for girls. First, the association between parental attachment and school functioning was examined by investigating the relation of parental attachment to teacher-ratings of adolescents' school competencies and G P A . There were no significant relations found in the total scale, but there was one significant relation found in the subscales. Specifically, greater mutual trust in relationships with parents was significantly and positively associated with G P A for girls, but not for boys. This result provides some evidence in support of the hypothesis that positive relationships with parents are associated with better school functioning. Next, the relation of parental attachment to psychological adjustment was examined. A s can be seen in Table 10, there were significant relations found in the total scale and in each of the subscales. Both for boys and for girls, significant and negative correlations were found between parental attachment and adolescents' self-reports of internalizing and externalizing problems. Thus, as expected, the correlational results support the hypothesis that perceived quality of mutual trust and communication and the extent of alienation in 7 Table 10 is presented here for convenience and because it provides a concise summary of the observed relations. 108 <ji Th1 • < a •a 1A rH -*-» c 0) Vi 3 < o '3 o o Vi PH T3 bl C C o 3 3 PH 3 O -a o 00 c <u e I 3 PH CA C o cd 1 o U PH § e X5 O 3 te PH o *4—» § a o o 3 S S o U a CA o PQ o vo r-o VO o CN OO VO rt r-H [•» VO Tt rH a CA >•> o PQ m vo o oo CA s a CA o CQ Tt cn CD 13 o 00 o E-i a CA >< o CQ CN CN in vo o cn oo Tt u rt D H s o U X3 o 00 CA CD • rH o $ CD D H Ou o X 5 o 00 3 3 00 PH o o bO 3 • rH N ' - i i e > 3 O Tt VO CN Tt cn cn CD r^ ! CD 00 p bO 3 • rH N 3 e X pq a o o CA CD 3 o CA CD xs H-> bfl 3 3 t O o s r3 i O o CA "8 CA fe £ WH <D CD 3 o CA rO 3 CA 3 O "Z2 rt 3 CD CD rC H—» 3 O CA CD Vh p O OO CD H—> P ZI vo CN II 31 cn CN II PI O q VI VI .Dl 109 relationships with parents are associated with fewer problems in psychological adjustment. Relations of peer attachment to school functioning and psychological adjustment. In examining relations of peer attachment to school functioning and psychological adjustment, it was hypothesized that peer attachment would be positively associated with school functioning (Hypothesis 6) and negatively associated with psychological adjustment problems (Hypothesis 10). Table l l 8 presents the correlations of peer attachment to school functioning and problems in psychological adjustment for boys and for girls. First, the association between peer attachment and school functioning was examined by investigating relations of peer attachment to teacher-ratings of adolescents' school competencies and G P A . There were no significant findings that emerged from the total scale or subscale analyses for either teacher-rated school competencies or G P A . Next, the association between peer attachment and psychological adjustment was examined by investigating relations of peer attachment to adolescents' self-reports of internalizing and externalizing problems. With regard to internalizing problems, as can be seen in Table 11, there was a significant finding from the total scale that was supported by the trust and alienation subscales. Specifically, for boys and for girls, peer trust was related significantly and negatively to adolescents' self-reports of internalizing problems, whereas less alienation in relationships with peers was related to fewer internalizing problems. That is, adolescents who reported greater mutual trust and less alienation in their relationships with peers also reported fewer internalizing problems. 8 Table 11 is presented here for convenience and because it provides a concise summary of the observed relations. To >, O PQ bli o 8 8 PH c cu 6 J C CJ <3 PH ' r t § c o o 'c 3 e e o u CO e cu o 0 0 o CU •s > CO >•> O PQ CO >» O PQ CO O PQ o PQ o o o oo C -O v O o T t o V O O VO O o CN o o o CN vo o O N O CN o vo o C N £ o U X i O w co CU •i-H O c CU 4—» CU CH . 6 o U 13 o o 0 0 c s 0 0 PH o p 0 0 c N •r-< 13 e CU rt c ON C N m - H cn vo o vo i " 0 0 co a X) o l H PH 0 0 c • rt N CU a o o co 1) 13 o co <U X i rt 0 0 c • rt rt 3 CX S o o c CU X ! 1 o o co "8 CO IH CU $ l H so CU 13 o CO X> 3 co C _o ' r t C u 13 cu c o co o o oo CU o o o m £ VI II .Pi ci XI co" C N II P VI .a Ill With regard to externalizing problems, there was a significant finding from the total scale that was supported by the trust and alienation subscales. Specifically, for girls only, peer trust was related significantly and negatively to self-reports of externalizing problems, whereas less alienation in relationships with peers was related to fewer externalizing problems. That is, adolescent girls who reported greater mutual trust and less alienation in their relationships with peers also reported fewer externalizing problems. The results support the hypothesis that peer attachment would be negatively associated with psychological adjustment problems. Whereas for girls, peer attachment was significantly and negatively related to both self-reports of internalizing and externalizing problems, for boys, peer attachment was significantly and negatively related only to self-reports of internalizing problems. Relations of school autonomy to school functioning and psychological adjustment. In examining relations of school autonomy to school functioning and psychological adjustment, it was hypothesized that school autonomy would be positively associated with school functioning (Hypothesis 5) and negatively associated with psychological adjustment problems (Hypothesis 9). Table 12 8 presents the correlations of school autonomy to school functioning and problems in psychological adjustment for boys and girls. To test these hypotheses, the association between school autonomy and school functioning was first examined by investigating the relation of perceived control at school to teacher-ratings of adolescents' school competencies and G P A . There was only one significant relation found 8 Table 12 is presented here for convenience and because it provides a concise summary of the observed relations. Table 12 Correlations of School Autonomy and School Belonging to School Functioning and Psychological Adjustment for Boys 8 and Gi r l s b 112 Variable School Autonomy School Belonging (PCSS) (PSSM) Boys Girls Boys Girls School Competencies (SchComp) .22*" .15 .24*** .15 G P A .06 .12 .10 . 18" Internalizing Problems (YSRint) -.42*** -.43*** -.53*** -.46*** Externalizing Problems (YSRext) -.17 -.36*** -.28"* -.45*" a n = 213. b n = 265. " p < .01. "*p < .001. 113 between school autonomy and school functioning. Specifically, for boys only, perceived control at school was significantly and positively related to teacher-ratings of school competencies. Moreover, for both genders, no significant relation emerged between perceived control at school and G P A . Thus, the results provide partial support for the hypothesis that school autonomy would be positively related to school functioning. Next, the association between school autonomy and psychological adjustment was examined by investigating the relation of perceived control at school to internalizing and externalizing problems. Perceived control at school was related significantly and negatively to internalizing problems for both genders, and negatively to externalizing problems for girls only. Thus, the results support the hypothesis that school autonomy would be negatively related to psychological adjustment problems. That is, those adolescents who reported higher levels of perceived control at school, reported fewer problems. Relations of school belonging to school functioning and psychological adjustment. In examining relations of school belonging to school functioning and psychological adjustment, it was hypothesized that school belonging would be positively associated with school functioning (Hypothesis 6) and negatively associated with psychological adjustment problems (Hypothesis 10). Table 12 presents the correlations of school belonging to school functioning and problems in psychological adjustment for boys and girls. To test this hypothesis, the association between school belonging and school functioning was first examined by investigating the relation of school belonging to teacher-ratings of adolescents' school competencies and G P A . For boys, school belonging was significantly and positively associated with teacher-114 ratings of school competencies. For girls, school belonging was related significantly and positively to G P A . Taken together, these results provide support for the hypothesis that school belonging would be positively associated with school functioning. Next, the association between school belonging and psychological adjustment was examined by investigating the relation of school belonging to adolescents' self-reports of internalizing and externalizing problems. For both genders, school belonging was related significantly and negatively to both internalizing and externalizing problems. A s hypothesized, these results indicate that those adolescents who perceived positive connections to school, reported fewer internalizing and externalizing problems. Summary. The second major goal of the present investigation was to examine the relations of autonomy and relatedness in parent, peer, and school contexts to school functioning and psychological adjustment. Six hypotheses were posited, two with regard to school functioning (i.e., teacher-rated school competencies, G P A ) and four with regard to psychological adjustment (i.e., internalizing problems, externalizing problems). First, school functioning was examined. It was hypothesized that school functioning would be positively associated with school autonomy (Hypothesis 5) and also positively associated with parental attachment, peer attachment, and school belonging (Hypothesis 6). Wi th regard to Hypothesis 5, results showed that only teacher-rated school competencies and not G P A were positively associated with school autonomy, and only for boys. Wi th regard to Hypothesis 6, results are summarized within the parent, peer, and school contexts, first for teacher-rated school competencies and next for G P A . Teacher-rated school competencies were associated as follows: findings for parental attachment and peer 115 attachment did not reveal any significant associations for either gender; and school belonging was positively associated only for boys. G P A was associated as follows: greater mutual trust in relationships with parents was positively associated for girls only; findings for peer attachment did not reveal any significant association for either gender; and school belonging was positively related for girls only. Next, psychological adjustment was examined. It was hypothesized that problems in psychological adjustment would be positively associated with parental autonomy (Hypothesis 7), negatively associated with both peer autonomy (Hypothesis 8) and school autonomy (Hypothesis 9). It was further hypothesized that problems in psychological adjustment would be negatively associated with parental attachment, peer attachment, and school belonging (Hypothesis 10). Results are summarized first for internalizing problems and next for externalizing problems. Wi th regard to Hypothesis 7, internalizing problems were positively associated with parental autonomy (total scale) and individuation from parents for boys and for girls. For both genders, externalizing problems were positively associated with parental autonomy (total scale), nondependency on parents, and individuation from parents. Moreover, externalizing problems were positively associated with deidealization of parents only for girls. Wi th regard to Hypothesis 8, internalizing problems were positively related to individuation from peers. Externalizing problems were negatively associated with nondependency on peers. Wi th regard to Hypothesis 9, internalizing problems and externalizing problems were each negatively associated with school autonomy. 116 Final ly , with regard to Hypothesis 10, results are summarized within the parent, peer, and school contexts, first for internalizing problems and next for externalizing problems. Internalizing problems were associated as follows: parental attachment (including trust, communication, and alienation scores) was negatively associated; overall peer attachment and peer trust were negatively associated, and alienation was positively associated; and school belonging was negatively associated. Externalizing problems were associated as follows: parental attachment (including trust, communication, and alienation scores) was negatively associated; overall peer attachment and peer trust were negatively associated, and alienation was positively associated for girls only; and school belonging was negatively associated. Regression Analyses Examining Autonomy and Relatedness in the Contexts of Parents. Peers, and School, as Predictors of School Functioning and Psychological Adjustment The purpose of this section is to present the results of the regression analyses. It w i l l be recalled that the third major goal of the present investigation was to determine the extent to which autonomy and relatedness predict school functioning and psychological adjustment problems. To further explore relations of autonomy and relatedness to school functioning and psychological adjustment, four separate hierarchical multiple regressions were computed for each gender in which teacher-rated school competencies, G P A , internalizing problems and externalizing problems served as the dependent variables. Parental deidealization, parental nondependency, parental individuation, parental attachment, peer deidealization, peer nondependency, peer individuation, peer attachment, school autonomy, and school belonging and grade level were the independent variables. Grade level served as a control variable in these analyses for two reasons: (a) grade in school was part of 117 the cross-sectional research design and used as an index of developmental status (cf., Larson et a l . , 1996), and (b) significant grade differences were found on dimensions of peer autonomy and peer attachment in the M A N O V A analyses examining gender and grade differences. It was hypothesized that autonomy and relatedness in parent, peer, and school contexts would be significant predictors of school functioning and psychological adjustment problems. First it was hypothesized that teacher-rated school competencies and academic achievement would each be positively associated with autonomy and relatedness in parent, peer, and school contexts (Hypothesis 11). Next, it was hypothesized that internalizing and externalizing problems would each be positively associated with parental autonomy, and negatively associated with peer autonomy and school autonomy (Hypothesis 12a). It was further hypothesized that internalizing and externalizing problems would be negatively associated with parental attachment, peer attachment, and school belonging (Hypothesis 12b). For each hierarchical regression analyses, the predictors were entered into the analyses in the following way. In Step 1, grade level in school was entered first. The hierarchical approach was used to statistically control for grade in assessing the relations of autonomy and relatedness variables to the dependent variable beyond that of grade level (Cohen & Cohen, 1983). Theoretically, different grade levels would reflect age-related differences in adolescent characteristics during a critical period for individuation and autonomy development with parents, peers, and at school (e.g., H i l l & Holmbeck, 1986; Josselson, 1980; Minuchin & Shapiro, 1983). Thus, it was of interest to determine the unique contribution of dimensions of autonomy, and relatedness after controlling for variance 118 associated with adolescents' grade level. Next, parental deidealization, parental nondependency, parental individuation, parental attachment, peer deidealization, peer nondependency, peer individuation, peer attachment, school autonomy, and school belonging variables were entered as a block in Step 2. To avoid any potential problems with multicollinearity between dimensions of trust and communication in relationships with parents or with peers (see correlations in Table 7), the composite scores for parental attachment and peer attachment were used. The hierarchical multiple regression procedure was chosen because it allows one to control entry of variables and to determine the unique variance contributions of the independent variables after accounting for the proportion of variance attributable to one or more independent variables already in the equation (Tabachnick & Fide l l , 1989). A s noted by Pedhazur (1982), interrelations among the independent variables (i.e., multicollinearity) may pose difficulties in regression analyses. Thus, to check for multicollinearity in the present investigation I examined tolerance levels within the regression analysis (Gavazzi et a l . , 1993; Tabachnick & Fide l l , 1989). Tolerance levels range from 0 to 1, and i f values are less than 0.1, the potential for multicollinearity increases (Norusis, 1997). For each of the regression analyses, tolerance values for the independent variables were examined and did not indicate problems with multicollinearity (i.e., tolerance values ranged from .34 to .81). Mult iple regression analyses predicting teacher-rated school competencies. Hierarchical multiple regression analyses on teacher-rated school competencies were performed separately for boys and for girls. A s can be seen in Table 13, for boys, results from the hierarchical regression analysis predicting teacher-rated school competencies Table 13 Summary of Hierarchical Mult iple Regression Analysis Predicting School Competencies for Boys 8 119 Variable B S E B 0 R R 2 £ i-change Step 1 Grade 2.69 0.87 . 2 1 " .21 .04 9.57* Step 2 ..41 .17 3.08* Parental Deidealization -0.75 0.52 -.13 Parental Nondependency 1.04 0.55 .14 Parental Individuation -0.06 0.43 -.02 Parental Attachment -0.01 0.09 -.01 Peer Deidealization 0.97 0.53 .13 Peer Nondependency 1.02 0.54 .14 Peer Individuation 0.62 0.45 .11 Peer Attachment 0.07 0.10 .07 School Autonomy 0.14 0.12 .10 School Belonging 0.23 0.12 .18 a n = 213. * E < .01. ***p = .001. 120 indicated that, after controlling for variance associated with grade level, the autonomy and relatedness variables contributed significantly to school competencies. Specifically, in the first step in the analysis, grade level accounted for only a small part of the variance (4%) in school competencies. In the second step, inclusion of parental autonomy, peer autonomy, parental attachment, peer attachment, school autonomy and school belonging variables produced a significant change in R 2 , accounting for 13% of the variance in teacher-rated school competencies. Inspection of the standardized beta weights shown in Table 13 revealed that the variables jointly predicted teacher-rated school competencies for boys. Overall , the total model accounted for 17% of the variance in boys' school competencies, F ( l l , 201) = 3.76, p = .001. A s can be seen in Table 14, for girls, results from the hierarchical regression analysis predicting teacher-rated school competencies indicated that in the first step in the analysis, grade level accounted for very little variance (1 %) in school competencies. In the second step, the autonomy and relatedness variables did not contribute significantly to an increase in R 2 , after controlling for variance associated with grade. Inspection of the standardized beta weights shown in Table 14 indicated that the variables, when considered together, predicted teacher-rated school competencies for girls. Overall, the total model accounted for 8% of the variance in girls ' school competencies, F ( l l , 253) = 1.87, p < .05. Table 14 121 Summary of Hierarchical Mult iple Regression Analysis Predicting School Competencies for Gir l s 8 Variable B S E B 0 R R 2 F, harige Step 1 Grade 1.44 0.77 .12 .12 .01 3.56 Step 2 .27 .08 1.67 Parental Deidealization 0.29 0.43 .06 Parental Nondependency 0.52 0.50 .08 Parental Individuation -0.48 0.44 -.10 Parental Attachment 0.07 0.08 .09 Peer Deidealization 0.40 0.49 .06 Peer Nondependency -0.62 0.50 -.09 Peer Individuation 0.73 0.38 .15 Peer Attachment -0.27 0.08 -.03 School Autonomy 0.11 0.12 .08 School Belonging 0.15 0.12 .13 Note. A l l p values are nonsignificant, "n = 265. 122 Mult iple regression analyses predicting G P A . Hierarchical multiple regression analyses on G P A were performed separately for boys and for girls. The results of the regression analysis predicting G P A for boys are presented in Table 15. A s can be seen in Table 15, in the first step, grade level did not account for variance in G P A . In the second step, the autonomy and relatedness variables produced a significant change in R 2 , accounting for 19% of the variance in G P A . Inspection of the standardized beta weights shown in Table 15 revealed that peer deidealization, peer nondependency, and peer individuation were significant and positive predictors of G P A for boys. Overall , the total model accounted for 19% of the variance in G P A , £ ( 1 1 , 201) = 4.36, p < .001. A s can be seen in Table 16, for girls, results from the hierarchical regression predicting G P A revealed that in the first step, grade level did not account for variance in G P A . In the second step, the autonomy and relatedness variables produced a significant change in R 2 , accounting for 8% of the variance in G P A . Inspection of the standardized beta weights shown in Table 16 revealed that school belonging was a significant and positive predictor of G P A for girls. Overall , the total model accounted for 8% of the variance in G P A , F ( l l , 253) = 1.94, p < .05. Mult iple regression analyses predicting internalizing problem behaviours. Hierarchical multiple regression analyses for self-reported internalizing problems were performed separately for boys and for girls. A s can be seen in Table 17 for boys, in the first step, grade level did not account for variance in self-reported internalizing problems. In the second step, the autonomy and relatedness variables produced a significant change in R 2 , accounting for 44% of the variance in boys' self-reported internalizing problems. Inspection Table 15 123 Summary of Hierarchical Mult iple Regression Analysis Predicting G P A for Boys" Variable B S E B B R R 2 £ hange Step 1 Grade -0.12 0.17 -.05 .05 .00 0.49 Step 2 .44 .19 4.74* Parental Deidealization -0.02 0.10 -.02 Parental Nondependency 0.19 0.10 .14 Parental Individuation 0.02 0.08 .03 Parental Attachment 0.00 0.02 .01 Peer Deidealization 0.28 0.10 .21** Peer Nondependency 0.32 0.10 .24" Peer Individuation 0.27 0.08 .25*** Peer Attachment 0.02 0.02 .10 School Autonomy 0.01 0.02 .05 School Belonging 0.03 0.02 .14 "n = 213. " p < .01. "*p < .001. Table 16 124 Summary of Hierarchical Mult iple Regression Analysis Predicting G P A for Girls" Variable 2 S E B • 0 R R 2 F , hange Step 1 Grade -0.07 0.16 -.03 Step 2 Parental Deidealization 0.04 0.09 .04 Parental Nondependency -0.00 0.10 -.00 Parental Individuation -0.10 0.09 -.10 Parental Attachment 0.01 0.02 .04 Peer Deidealization 0.10 0.10 .08 Peer Nondependency 0.07 0.10 .05 Peer Individuation 0.13 0.08 .13 Peer Attachment 0.00 0.02 .01 School Autonomy -0.01 0.02 -.03 School Belonging 0.05 0.02 .21* .28 .08 2.12* } *p < .05. a n = 265. Table 17 125 Summary of Hierarchical Mult iple Regression Analysis Predicting Internalizing Problems for Boys" Variable B S E B B R R 2 F, :hange Step 1 Grade 0.42 0.50 .06 Step 2 Parental Deidealization -0.84 0.24 -.26*" Parental Nondependency -0.56 0.25 -.14* Parental Individuation -0.03 0.20 -.01 Parental Attachment -0.22 0.04 -.44*** Peer Deidealization 0.19 0.25 .05 Peer Nondependency -0.55 0.25 -.13* Peer Individuation 0.49 0.21 .15* Peer Attachment -0.01 0.04 -.02 School Autonomy -0.10 0.06 -.13 School Belonging -0.18 0.06 -.26"* .67 .44 15.95* a n = 213. *p < .05. ***p < .001. 126 of the standardized beta weights shown in Table 17 revealed that parental deidealization, parental nondependency, parental attachment, peer nondependency, and school belonging were significant and negative predictors of self-reported internalizing problems for boys. Peer individuation emerged as a significant and positive predictor of boys' self-reported internalizing problems. Overall , the total model accounted for 44% of the variance in boys' self-reported internalizing problem behaviours, F ( l l , 201) = 14.61, p < .001. A s can be seen in Table 18, for girls, results from the hierarchical regression predicting self-reported internalizing problem behaviours revealed that the autonomy and relatedness variables made a significant contribution to predicting internalizing problems, after controlling for grade level. In the first step, grade level did not account for variance in self-reported internalizing problems. In the second step, the variables produced a significant change in R 2 , accounting for 40% of the variance in internalizing problems. Examination of the standardized beta weights indicated that parental attachment and peer attachment were significant and negative predictors of self-reported internalizing problems for girls. Overall , the total model accounted for 40% of the variance in girls ' self-reported internalizing problem behaviours, F ( l l , 253) = 15.32, p < .001. Mult iple regression analyses predicting externalizing problem behaviours. Mult iple regression analyses for externalizing problems were performed separately for boys and for girls. A s can be seen in Table 19, in the first step, grade level did not account for variance in self-reported externalizing problems for boys. In the second step, the variables produced a significant change in R 2 , accounting for 24% of the variance in externalizing problems. Inspection of the standardized beta weights shown in Table 19 revealed that parental Table 18 127 Summary of Hierarchical Mult iple Regression Analysis Predicting Internalizing Problems for Gi r l s 8 Variable B S E B j8 R R 2 F, change Step 1 Grade 0.19 0.55 .02 Step 2 Parental Deidealization -0.42 0.25 -.11 Parental Nondependency -0.54 0.29 -.11 Parental Individuation 0.07 0.25 .02 Parental Attachment -0.23 0.04 - . 4 5 " Peer Deidealization -0.27 0.28 -.06 Peer Nondependency -0.15 0.29 -.03 Peer Individuation 0.19 0.22 .05 Peer Attachment -0.10 0.05 -.16* School Autonomy -0.11 0.07 -.11 School Belonging -0.11 0.07 -.13 .63 .40 16.84* "n = 265. *p < .05. *"p < .001. Table 19 128 Summary of Hierarchical Mult iple Regression Analysis Predicting Externalizing Problems for Boys 8 Variable B S E B B Step 1 Grade 0.40 0.43 .06 Step 2 Parental Deidealization -0.36 0.24 -.13 Parental Nondependency 0.04 0.25 .01 Parental Individuation 0.39 0.20 .16* Parental Attachment -0.13 0.04 -.31* Peer Deidealization -0.12 0.24 -.03 Peer Nondependency -0.51 0.25 -.15* Peer Individuation -0.11 0.21 -.04 Peer Attachment 0.03 0.04 .06 School Autonomy -0.02 0.06 -.03 School Belonging -0.08 0.06 -.13 TJ i-change .49 .24 6.30* a n = 213. *p < .05. **p < .01. ***p < .001. 129 individuation was a significant and positive predictor of self-reported externalizing problems, whereas parental attachment and peer nondependency were significant and negative predictors of boys' self-reported externalizing problems. Overall , the total model accounted for 24% of the variance in boys' self-reported externalizing problem behaviours, F(11, 201) = 5.82, p < .001. A s can be seen in Table 20, for girls, results from the hierarchical regression analysis predicting externalizing problem behaviours revealed that the autonomy and relatedness variables made a significant contribution to predicting externalizing problems, after controlling for grade level. In Step 1, grade level did not account for variance in girls ' self-reported externalizing problems. In Step 2, the variables produced a significant change in R 2 , accounting for 37% of the variance in externalizing problems. Examination of the standardized beta weights revealed that parental attachment and school belonging were significant and negative predictors of self-reported externalizing problems for girls.Overall , the total model accounted for 37% of the variance in girls ' self-reported externalizing problem behaviours, F ( l l , 253) = 13.25, p < .001. Summary. The third major goal of the present investigation was to determine the extent to which autonomy and relatedness in the contexts of parents, peers, and school account for variance in school functioning and problems in psychological adjustment. Overall , results of the hierarchical regression analyses on measures of school functioning showed that autonomy and relatedness variables accounted for significant proportions of variance in teacher-rated school competencies and G P A , after controlling for grade level. For boys, peer deidealization, peer nondependency, and peer individuation were Table 20 Summary of Hierarchical Mult iple Regression Analysis Predicting Externalizing Problems for Gi r l s 8 130 Variable B S E B 8 R R 2 F, bange Step 1 Grade 0.31 0.42 .05 .05 .00 0.56 Step 2 .61 .37 14.49* Parental Deidealization 0.18 0.19 .06 Parental Nondependency 0.25 0.23 .07 Parental Individuation 0.21 0.20 .08 Parental Attachment -0.10 0.03 -.27** Peer Deidealization 0.06 0.22 .02 Peer Nondependency -0.31 0.22 -.09 Peer Individuation -0.05 0.17 -.02 Peer Attachment 0.05 0.04 .09 School Autonomy -0.03 0.05 -.04 School Belonging -0.19 0.05 -.29*** 8 n = 265. " p < .01. ***p < .001. 131 positive and significant predictors of G P A . For girls, school belonging emerged as a significant and positive predictor of G P A . However, hone of the independent variables emerged as unique predictors of teacher-rated school competencies for boys or for girls. Results of the hierarchical regression analyses on measures of psychological adjustment showed that the autonomy and relatedness variables accounted for significant proportions of variance in adolescents' self-reports of internalizing and externalizing problem behaviours after controlling for grade level, which was not significantly related to problems in psychological adjustment. Findings for boys showed that parental deidealization, parental nondependency, parental attachment, peer nondependency, and school belonging were significant and negative predictors of self-reported internalizing problems, whereas peer individuation was a significant and positive predictor of boys' self-reported internalizing problems. Moreover, parental attachment and peer nondependency were significant and negative predictors of self-reported externalizing problems for boys, whereas parental individuation was a positive and significant predictor of boys' self-reported externalizing problems. For girls, results showed that parental attachment and peer attachment were significant and negative predictors of self-reported internalizing problems. Moreover, parental attachment and school belonging were significant and negative predictors of girls ' self-reported externalizing problems. Results of the regression analyses provide further evidence to suggest that experiences of autonomy and relatedness within the contexts of parents, peers, and school make independent contributions to the prediction of adolescent outcomes. Thus, the results support the notion that experiences of autonomy and relatedness are important for understanding adolescents' school functioning and problems in psychological adjustment. 132 C H A P T E R F I V E Discussion This study explored two critical dimensions of adolescent development - autonomy and relatedness - in relation to school functioning and psychological adjustment. Specifically, adolescents' perceived experiences of autonomy and relatedness were examined in the contexts of parents, peers, and school in order to better understand their relations to academic and behavioural functioning during adolescence. These three contexts have been identified by researchers as the major areas of socialization in which adolescents invest time and commitment and develop attitudes and beliefs that are critical to schooling and emotional health development and adjustment in general (e.g., Blum & Rinehart, 1997; Brown, 1990; Eccles et a l . , 1997; Greenberger et a l , 1982; Minuchin & Shapiro, 1983; Roeser, 1998). The present study contributes to the literature by adding to the recent growth in research examining socialization experiences and tasks of adolescence (e.g., social support, connection, regulation, autonomy, coping) simultaneously across multiple contexts (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Isakson & Jarvis, 1999; Wenz-Gross et a l . , 1997) and was aimed at identifying the way in which parent, peer, and school contexts relate to specific dimensions of academic and psychological adjustment. The present research is unique because the findings presented in this study illustrate the complexity of the interrelatedness of the dimensions of autonomy and relatedness in different contexts. This chapter is organized into three sections. In the first section, I discuss the findings as they relate to the hypotheses, as presented in Chapter Two, around which this study is organized: (a) gender and grade differences in adolescents' perceptions of autonomy 133 and relatedness across the contexts of parents, peers, and school, (b) relations of autonomy and relatedness to school functioning and problems in psychological adjustment, and (c) autonomy and relatedness in the contexts of parents, peers, and school as predictors of school functioning and problems in psychological adjustment. In the second section, I consider the implications and importance of this study to theory, research, and education. Final ly , in the third section, I present the strengths and limitations of the present study and provide recommendations for future research. Gender and Grade Differences in Adolescents' Perceptions of Autonomy and Relatedness The present study compared adolescent boys' and girls ' perceptions of parental autonomy, peer autonomy, school autonomy, and relatedness to parents, peers, and school at three different grade levels (grade 8, grade 9, and grade 11). Significant findings of hypothesized gender and grade differences in autonomy and relatedness clarify prior research and extend knowledge about autonomy and relatedness in important ways. Gender differences in autonomy. Although there were no hypotheses concerning gender differences in parental autonomy, findings from the analyses examining emotional autonomy from parents revealed that girls scored higher than boys on parental deidealization, indicating that girls held more realistic conceptions of their parents and were less likely to idealize their parents. This finding is in accord with findings from previous investigations indicating that girls score higher than boys on parental deidealization (e.g., Steinberg & Silverberg, 1986). Wi th regard to gender differences in perceived autonomy from peers, it was hypothesized that girls would report higher levels of peer autonomy than would boys 134 (Hypothesis 1). Findings from the analyses examining dimensions of peer autonomy (i.e., deidealization, nondependency, individuation) indicated that girls were less likely than boys to hold idealized images of their peers. This finding provides some support for the hypothesis and suggests that girls may have a more realistic view of their peers and acknowledge that their peers may not always be right or share their opinions. Emotional autonomy from peers has been associated theoretically with greater psychological differentiation and expressions of distinctiveness of the self from others (e.g., Josselson, 1980), and empirically associated with less conformity to peer pressure in girls (e.g., Steinberg & Silverberg, 1986). Theoretically, it could be that the concept of deidealization may shed light on the development of emotional autonomy from peers in that adolescents may have to replace idealized images of their peers with more realistic ones, similar to that which is hypothesized to occur in the development of emotional autonomy from parents (e.g., Bios, 1967). Because studies have also shown that girls are less susceptible than boys to peer pressure and conformity, especially in antisocial situations (e.g., Berndt, 1979; Brown et a l . , 1986; Steinberg & Silverberg, 1986), perhaps peer deidealization is the mechanism by which girls may be less susceptible to peers' influences. The present findings also indicated that, when compared to girls, boys had higher scores of nondependency on peers, suggesting that boys are less dependent on their peers than are girls. Perhaps this finding is not surprising given research indicating that when compared to boys, girls place more importance on friendship quality and report more affection, commitment and intimacy in their relationships with peers (e.g., Brendgen, Markiewicz, Doyle, & Bukowski, 2000; Goldbaum & Crawford, 2000; Stern, 1990). The 135 results of the present study provide further understanding of extant notions of peer autonomy because peer autonomy was assessed in a way different than in previous investigations. Researchers have used hypothetical situations, or reports of peer conformity and perceived peer pressures to assess susceptibility to peers as a way in which to measure autonomy from peers (e.g., Berndt, 1979; Brown et a l . , 1986; Gavazzi et a l . , 1993; Kuperminc et a l . , 1996; Oldenburg & Kerns, 1997; Steinberg & Silverberg, 1986). In contrast, the present study operationalized peer autonomy to indicate the degree to which adolescents distinguished or differentiated themselves from their peers. Thus, the present investigation advances research on autonomy because peer autonomy, as uniquely defined in this study, has not been empirically investigated to date, and as such, provides new information regarding the unique ways in which boys and girls perceive their peer relations. Clearly, research on emotional autonomy from peers warrants further investigation in order to better understand its potential significance in autonomy development during adolescence. Gender differences in relatedness. It w i l l be recalled that relatedness was operationalized in the present study in terms of one's perceptions of attachment to parents, peers, and school. With regard to gender differences in relatedness, it was hypothesized that girls would report higher levels of parental attachment, peer attachment, and school belonging than would boys (Hypothesis 2). Previous research has suggested that girls, in comparison to boys, place much more emphasis on feelings of connection and belonging (e.g., Douvan & Adelson, 1966; Eccles et a l . , 1997; Goodenow, 1993a; Laible et a l . , 2000; Stern, 1990), Wi th regard to parental attachment, findings from the multivariate analysis examining 136 parental trust, communication, and alienation scores provide some evidence to suggest that boys and girls differ in their attachments to their parents. Relative to boys, girls had higher communication and alienation scores, and lower scores of trust. In the present study, parental attachment was assessed via a generalized attachment to parents. Specifically, adolescents were asked to respond to the questionnaire items on parental relationships for the parent or persons(s) who had acted as his or her parents most of the time. Adolescents predominantly responded with both parents in mind (81%) without distinguishing between mother or father. There is some suggestion made in the research literature indicating that the gender of the parent may influence adolescents' feelings of attachment such that adolescent-father or adolescent-mother attachments may be different for boys and girls (e.g., Armsden & Greenberg, 1987; Paterson et a l . , 1995). It has been suggested that boys and girls have a stronger emotional bond with their mother than with their father (see Geuzaine et a l . , 2000), or that attachments may vary because of different expectations adolescents have of their mothers and fathers (e.g., Paterson et a l . , 1995). Although few gender differences have been found in studies that have distinguished between adolescent-father and adolescent-mother relations, Papini et al . (1991) found some evidence to suggest that across pubertal status, early adolescent girls perceived less attachment to their mothers and fathers, whereas boys exhibited a slight (though nonsignificant) increase in perceived attachment to their mothers and less attachment to their fathers. These researchers also found that girls, in comparison to boys, reported significantly greater attachment to fathers. Thus, future studies that examine boys' and girls ' attachments to mothers and fathers separately, might help to 137 clarify and extend our understanding of adolescent-parental attachments, especially at a time in development when adolescents begin to integrate parental and peer attachments. Wi th regard to peer attachment, findings from the analyses examining peer trust, communication and alienation scores revealed that, when compared to boys, girls had higher levels of mutual trust and communication in their relationships with their peers. These results are consistent with the hypothesis of this study and are congruent with research findings indicating that girls have a better quality of relationship with peers than do boys (e.g., Armsden & Greenberg, 1987; Oldenburg & Kerns, 1997). Wi th regard to school belonging, girls did not report higher levels of school belonging than did boys. This finding was inconsistent with the hypothesis of this study as well as previous research suggesting the existence of gender differences. Indeed, past research indicates that girls place much more emphasis on feelings of belonging and connection at school than do boys (e.g., Goodenow, 1993a; Eccles et a l . , 1997). One explanation for the nonsignificant gender differences in the present investigation may be due to the difference in measures used to assess school belonging across studies. Whereas some researchers have used only a few items to assess school connection, such as teacher availability (e.g., Barber & Olsen, 1997) or school l iking (e.g., Eccles et a l . , 1997), the current study utilized a more comprehensive measure of school belonging (Goodenow's, 1993b, Psychological Sense of School Membership scale) than used in most other studies. Goodenow's measure focuses on students' attitudes about their place in the school as a whole and reflects their sense of attachment to the entire school. A s noted by Goodenow (1993b) school practices such as cooperative learning tasks, inclusive group projects and activities that call on the 138 collaborative efforts of students and teachers working together are likely to contribute to students' sense of belonging at school. For the boys and girls who took part in this study, cooperative learning activities are common classroom and school practices in the elementary schools as well as in the regular instructional program in the high school. The Catholic school system is characterized by communal activities including retreats, masses, liturgies that contribute to a shared sense of community. It may be that the adolescents in this sample represented a cohesive group of individuals who had relatively stable feelings of school membership possibly due to a pre-existing sense of identification with their schools (Isakson & Jarvis, 1999) that could result in there being no difference between boys and girls in school belonging. Alternatively, it may be that the particular classroom or school from which the adolescent participants were drawn were places in which both boys and girls were made to feel an equal sense of belonging or lack thereof. Grade differences in autonomy. With regard to grade differences in parental autonomy, previous research findings have been somewhat equivocal. For instance, some researchers have found that emotional autonomy from parents increases with age (e.g., Lamborn & Steinberg, 1993; Steinberg & Silverberg, 1986). In contrast, Ryan and Lynch (1989) found no grade level differences in adolescents' emotional autonomy from parents in their sample of 212 adolescents in grades nine through 12. Because research has documented age-related increases in emotional autonomy from parents during adolescence, it was hypothesized that adolescents in grade 11 would report higher levels of parental autonomy than would adolescents in grades eight or nine (Hypothesis 4a). Results showed that adolescents in grade 11 did not report higher levels of parental 139 autonomy than adolescents in grades eight or grade nine. Thus, these results are in accord with those of Ryan and Lynch (1989). Perhaps the results of the present study are not in concert with those of other researchers (e.g., Steinberg & Silverberg, 1986) because of differences in age ranges (maturational level) examined in this study and those examined in previous research. For example, Steinberg and Silverberg (1986) found that emotional autonomy increased linearly in their sample of 865 early adolescents in grades five through nine. Whereas Steinberg and Silverberg's grade levels spanned preadolescence to early adolescence, the present study spanned early to middle adolescence. Most notably, although Steinberg and Silverberg found that emotional autonomy was significantly related to grade, the significant differences were those between the fifth and sixth grade, and between the sixth and eighth grade. Moreover, no significant differences were found between grades eight and nine. Although researchers suggest that development of emotional autonomy during adolescence is a lengthy process that extends beyond late adolescence into adulthood (Frank et a l . , 1990; H i l l & Holmbeck, 1986; Steinberg & Silverberg, 1986), it may be that overall levels of parental autonomy are relatively stable during the middle adolescent years. Support for this contention is'found in the findings of the current study as well as in those of Ryan and Lynch (1989). Accordingly, longitudinal studies are called for that could better determine the developmental trajectory of emotional autonomy from parents during adolescence that would help to further our understanding of emotional autonomy development during early and middle adolescence. With regard to grade differences in peer autonomy, it was hypothesized that adolescents in grade 11 would report higher levels of peer autonomy than would adolescents 140 in grades eight or nine (Hypothesis 4a). Studies have shown that conformity and peer pressure increases from childhood through middle adolescence and peaks around grade 8 and grade 9, then declines (Berndt, 1979; Coll ins, Gleason, et a l . , 1997; Gavin & Furman, 1989; Steinberg & Silverberg, 1986). In the present study, findings from the analyses examining dimensions of emotional autonomy from peers indicated that adolescents in grade 11 reported higher levels of deidealization from peers than adolescents in grade eight. Recall that higher deidealization scores signify that adolescents hold less idealized images of their peers. Thus, in the present study, older adolescents were more likely to have more mature conceptualizations of their peers than were younger adolescents. This result supports the hypothesis and extends findings from previous research on peer conformity, as previously mentioned, by suggesting the possible relevance of peer deidealization in relation to declines in peer conformity. Further studies might examine more specifically deidealization of peers in relation to peer conformity as an important next step to understanding the connection and distinction between peer autonomy and peer conformity. Results of the present investigation regarding nondependency on peers indicated that grade eight adolescents, who were in elementary school, reported being less dependent on their peers than adolescents in grade nine, who were in high school. This finding is contrary to the hypothesis that older adolescents would be more autonomous from their peers than younger adolescents. One plausible explanation for this finding can be found in the literature on transitions from elementary school to high school (e.g., Gavin & Furman, 1989; Isakson & Jarvis, 1999). For instance, Gavin and Furman (1989) found that early adolescents in grades seven and eight who were enrolled in middle schools, had a strong sense of unity and 141 superiority within their peer group and displayed more negative behaviours toward those outside the group. For the students in the present study, grade nine is a transition year characterized by new curriculum, a different organization for the school day, and new peers. Such changes give rise to a loss in feelings of unity and superiority. Thus, the school transition from grade eight to grade nine is a stressful time in which adolescents experience large peer groups, new challenges, and academic and social pressures in the new school situation (e.g., Gavin & Furman, 1989, Isakson & Jarvis, 1999; Simmons & Blyth , 1987). It has been suggested that because adolescents in high school do not have a regular classroom of peers, other than homeroom with whom to identify and associate with, they may feel a greater need to depend on the peer group to feel more secure and feel they belong in the new school environment (Gavin & Furman, 1989). The present findings suggest that adolescents might adjust to the transition from grade eight to grade nine by depending more on their peers in grade nine than they did in grade eight, which would account for the finding in this study regarding the difference in nondependency on peers between grade eight and grade nine. Clearly, more research is needed to examine the effects of school transitions from grade eight to nine in relation to hypotheses about emotional autonomy from peers. Given the findings of the present study, further research that focuses on dimensions of peer autonomy is warranted. In particular studies that more broadly examine emotional autonomy from peers during school transitions may be especially helpful. Wi th regard to grade differences in school autonomy, school transition studies among early adolescent samples have shown that adolescents perceive there is less support for 142 student autonomy and fewer opportunities for decision making and choice as they move through the elementary and middle school grades (e.g., Eccles, Buchanan, et al., 1991; Eccles et al., 1984, 1993; Eccles & Midgley, 1989; Midgley & Feldlaufer, 1987; Roeser & Eccles, 1998). In contrast, Smith and colleagues (e.g., Adelman et al., 1986; Heavey et al., 1989; Smith et al., 1987) have found no differences in students' perceptions of control across ages ranging from nine years to 19 years. Whereas the most recent research suggests a decline in school autonomy across early adolescence, it was hypothesized that school autonomy would decrease from grade eight to grade eleven (Hypothesis 3). Contrary to the hypothesis, school autonomy did not decrease across grades. One possible explanation for the lack of significant grade differences in school autonomy may be due to the manner in which school autonomy was assessed. Studies that have found a decrease in school autonomy (e.g., see Eccles et al., 1984, 1993 for a review, Midgley & Feldlaufer, 1987), have used a brief measure of school autonomy designed to assess students' perceptions of opportunities at the classroom level, such as making decisions about classroom rules, homework, where to sit, choosing partners, and sharing ideas in classroom discussions. In the present study, school autonomy was broadly assessed via students' perceptions about "decision making regarding school socialization processes, reactions of significant others at school to student's efforts to act autonomously, availability of options and choices, fairness of the rationale offered for limits imposed, and ability of the student to counter the control efforts of others at school" (Adelman et al., 1986, p. 1007). Thus, the measure of school autonomy used in the present investigation focused on adolescents' perceptions of autonomy in the school as a whole and not, per se, characteristics of 143 individual classrooms. It appears that an important constraint limiting research on school autonomy may be due to the multitude of ways in which autonomy is conceptualized and operationalized across the research literature to date. Thus, a clearer definition of the construct and its assessment in future research on school autonomy is needed. Grade differences in relatedness. It w i l l be recalled that relatedness was operationalized in terms of adolescents' perceptions of attachment. Wi th regard to grade differences in parental attachment, given past research findings indicating a decline in parental relationship support across adolescence, as well as theory suggesting that feelings of acceptance and closeness in families are lower in middle adolescence than in childhood (e.g., Coll ins , Gleason, et a l . , 1997; Steinberg, 1990), it was hypothesized that adolescents in grade 11 would report lower levels of parental attachment than would adolescents in grades eight or nine (Hypothesis 4b). In the present study, no grade differences were found in parental attachment on dimensions of mutual trust, quality of communication, or extent of alienation. Thus, the present findings contrast with those of other researchers who found age-related declines. For instance, Lamborn and Steinberg (1993) found that parental relationship support decreased between grades nine and eleven. Nevertheless, other researchers have found no age differences among early and middle adolescents in the quality of adolescent-parental relationships (e.g., Greenberg et a l . , 1983; Papini & Roggman, 1992). Although the results of the present investigation are not in accord with those of several others, the results herein are in concert with some of the previous research in this area. One possible explanation for 144 ' these equivocal findings may be due to differences in the measures used to assess parental attachment. For example, Lamborn and Steinberg (1993), researchers who have found a decrease in parental support over time, used a self-designed measure of parental relationship support. In contrast, Papini and Roggman (1992) and Greenberg et al . (1983), researchers who found no age differences, used measures of parental attachment that were used in the present study (i.e., Inventory of Parent and Peer Attachment; Armsden & Greenberg, 1987). Whereas Lamborn and Steinberg's measure assesses "adolescents' confidence that parents are supportive, are available to help when needed, and spend time with them" (p. 497), the Inventory of Parent and Peer Attachment assesses three broad dimensions of attachments to parents (i.e., mutual trust, communication, alienation). With regard to grade differences in peer attachment, it was hypothesized that adolescents in grade eleven would report lower levels of peer attachment than either adolescents in grade eight or grade nine (Hypothesis 4b). Research and literature on peer relations suggest there are age-related declines in conformity behaviour and social support from peers as autonomy from peers increases during adolescence (e.g., Coll ins , Gleason, et a l . , 1997). The grade eight and grade nine findings from the analyses examining dimensions of peer attachment indicated that, grade nine adolescents reported significantly higher levels of mutual trust and communication in their peer relationships than did adolescents in grade eight. That is, grade nine adolescents reported feeling more secure and supported in their peer relationships, and thus were more attached to their peers than were grade eight adolescents. Although these results were contrary to the hypothesis, they are consistent with 145 findings from a recent longitudinal study on school transitions conducted by Isakson & Jarvis (1999). Isakson and Jarvis hypothesized that support from peers would decrease over the transition from grade eight to grade nine as a response to increased pressures to fit in and belong in the new school situation. Contrary to their hypothesis, these researchers found that social support from peers actually increased from grade eight to grade nine. Although the sample in their study was small ( N = 48), their findings provide some evidence that adolescents' attachments to peers increase during school transitions from grade eight to grade nine. One plausible explanation for the increase in peer attachments from grade eight to grade nine may be due to adolescent coping patterns. It may be that during or after the initial disruptions of changing schools and adapting to a new school environment, adolescents seek out and expand their relationships with peers in order to feel less anonymous and more secure. Researchers have suggested that peer relationships are increasingly important to adolescents' feelings of affiliation and acceptance during adolescence (Berndt, 1979; Gavazzi et a l . , 1993; Goodenow, 1994; Oldenburg & Kerns, 1997; Steinberg & Silverberg, 1986), and help them manage stress (e.g., Brown, 1990; Isakson & Jarvis, 1999). It may be recalled that the finding in the present study regarding nondependency on. peers revealed that grade nine adolescents reported depending more on their peers, as evidenced by lower scores of nondependency, than did adolescents in grade eight (see Hypothesis 4a). Thus, when the findings regarding peer attachment and peer autonomy are considered together, they suggest that adolescents in grade nine are more dependent on'their peers (i.e., less autonomous from peers) and feel more supported in their peer relationships 146 (i.e., more attached to peers) than adolescents in grade eight. These findings extend previous research on school transitions and peer attachment by suggesting the importance of peer attachments in the context of the transition to high school. Relations of Autonomy and Relatedness to School Functioning and Psychological Adjustment Relations of autonomy and relatedness to school functioning and psychological adjustment were examined both correlationally and in a series of hierarchical regression analyses. In the following section I begin with a discussion of the findings from the correlational analyses and follow with a discussion of the regression analyses. Relations of autonomy to school functioning. Wi th regard to the associations between autonomy and school functioning, it was hypothesized that school autonomy would be positively associated with school functioning (Hypothesis 5). It may be recalled that school functioning was assessed in two ways: (1) teacher-ratings of adolescents' school competencies, and (2) academic achievement ( G P A ) . School autonomy was operationalized in terms of the adolescent's perceptions of control at school. Findings indicated that, for boys only, school autonomy was positively associated with teacher-rated school competencies. This finding is in accord with research and theory suggesting the importance of school autonomy to academic motivation and school-related functioning (e.g., Eccles, Buchanan, et a l . , 1991; Eccles et a l . , 1984; Roeser & Eccles, 1998). Contrary to what was hypothesized, no significant relation emerged between school autonomy (i.e., perceived control at school) and G P A , for boys or for girls. Although some researchers have found no significant correlation between school autonomy and G P A (e.g., Eccles et a l . , 1997), related research has indicated positive associations between perceived 147 control, motivation, and academic achievement (Adelman et a l . , 1986; Dec i & Ryan, 1987; Dec i , Vallerand, Pelletier, & Ryan, 1991; Grolnick et a l . , 1991; Smith et a l . , 1987; see Stipek & Weisz, 1981 for a review). Moreover, recent research conducted by Roeser and Eccles (1998) examining associations between adolescents' school perceptions and academic functioning, found significant and positive links between school autonomy (assessed as student autonomy) and G P A . These researchers examined behavioural and school-related perceptions of grade eight adolescents, and found that school autonomy was associated positively with "academic grade point average in a sample of 1,046 eighth grade adolescents. Similar results have been found in research conducted by Barber and Olsen (1997) in which school autonomy and G P A were positively related for eighth grade girls, but not for boys. It may be that differences between the results of the present study and those of prior research are due to the manner in which autonomy in school has been conceptualized and measured. Differences may also be due to disparate sample sizes and/or grade levels (for example being restricted to middle grades). For instance, most of the research on school autonomy has been confined to elementary or middle schools (e.g., Eccles et a l . , 1984; Roeser & Eccles, 1998). Most research on school autonomy has been conducted examining school autonomy in the form of opportunities in the classroom for decision-making, choice and self-management (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Eccles et a l . , 1993; Kasen et a l . , 1990; Midgley & Feldlaufer, 1987) using measures that were comprised of only a few items which were limited in scope, such as deciding where to sit in the classroom, choosing partners, having a say in deciding rules, and being able to express/share their own ideas. Because recent research findings on caring school communities have suggested that 148 opportunities for students' participation and influence in school might be more important at the secondary level than in elementary school (e.g., Battistich et a l . , 1997), more research examining the nature and function of school autonomy in secondary school is clearly warranted. Although there was no hypothesis regarding the relation between peer autonomy and school functioning, it is important to note the significant and positive relation that was found between peer autonomy and G P A for boys only. Boys who perceived themselves as more individuated and less dependent on their peers, had higher end of year G P A s . One plausible explanation for this finding can be found in the research of Ryan et al . (1994) exploring early adolescents' relationships with parents, peers, and teachers in relation to school functioning. These researchers found that peer emulation (i,e., idealizing or strongly identifying with peers) was negatively related to school adjustment and motivation, in terms of school functioning. These researchers speculated that adolescents who emulate their peers may be more focused on their peers than on school. It is plausible that the construct of peer emulation is conceptually similar to that of peer autonomy as defined in this study. The findings of the present study are also in concert with those of others' research suggesting that better academic achievement for boys is associated with greater differentiation from peers (i.e., less emulation of peers) (e.g., Ryan et a l . , 1994). Such findings suggest that i f adolescent boys can distinguish themselves from their peers and achieve a sense of separateness from them, they may be more focused on school. This interpretation should be approached cautiously, however, given the cross-sectional nature of the data in the present study. Indeed, no firm conclusions can be made regarding the directionality of this result. 149 Relations of attachment to school functioning. With regard to associations between attachment and school functioning, it was hypothesized that parent attachment, peer attachment, and school belonging would each be positively associated with school functioning (Hypothesis 6). Findings revealed that, for girls only, parental trust was significantly associated with higher G P A . This finding is congruent with results found in other studies (e.g., Cotterell, 1992; Eccles et a l . , 1997; Wentzel, 1998). For example, Cotterell (1992) found positive, although weak relations, between attachment to parents and academic adjustment (operationalized as academic self-concept and academic plans) in a sample of 57 adolescents, ranging in ages from 14 to 17 years. Also using a measure that reflected emotional connections with parents, Eccles and her colleagues (Eccles et a l . , 1997) found small, but positive associations between parental attachment and G P A for both genders in a large sample of approximately 1400 early adolescents. Although speculative at this point, it may be that for girls in the present study, mutual trust in relationships with parents may be reflected in a generalized trust in other adults with parental type roles (e.g., teachers) and may be associated with positive academic outcomes (e.g., Wentzel, 1991). There is some evidence in relational theory (e.g., Gil l igan, Rogers, & Brown, 1990) that girls value intimacy more than do boys. It may be that girls who have a positive parental trust relationship are more likely to adopt parental values and attitudes regarding academic achievement and/or are more likely to feel confident and positive about themselves which may also be associated with positive academic outcomes. The results from the present study regarding the association between peer attachment and school functioning did not support the hypothesis. N o significant relations between peer 150 attachment (i.e., mutual trust, communication, alienation) and school functioning (i.e., teacher-rated school competencies, G P A ) were evidenced either for boys or for girls. Despite the plethora of literature and research that suggests the importance and influence of friends on educational outcomes (e.g., Berndt et al . 1990; Cotterell, 1992; Eccles et a l . , 1997; Goodenow, 1992; Oldenburg & Kerns, 1997; Wentzel & Caldwell , 1997; Wenz-Gross et a l . , 1997), previous research, for the most part, has focused broadly on educational or achievement-related goals and psychological adjustment outcomes rather than on specific academic competencies or achievement. For example, Cotterell (1992) also examined relations of peer attachment and peer support to academic adjustment (assessed as academic self-concept and academic plans) and well-being (i.e., general self-concept, depressed mood, self-esteem). Although Cotterell found significant and negative associations between academic adjustment and support from friends and positive associations between attachment to friends and academic self-concept for boys and for girls, peer attachment and support were found to be more closely linked to psychological well-being. Further studies are needed to examine more specifically the associations between peer attachment relationships and academic competencies and achievement. A s Goodenow (1992) has suggested, it is "necessary in research on education to disaggregate the influence of close friends, reference-group peers, and classmates" (p. 185). A s hypothesized, school belonging was related positively to school functioning, although these relations differed by gender. More specifically, for boys only, school belonging was positively associated with teacher-ratings of school competencies, whereas, for girls only, school belonging was positively related to G P A . These results are in accord with 151 previous studies indicating that feelings of belonging and connection at school are associated positively with academic achievement and school-related performance (Eccles et a l . , 1997; Goodenow, 1993a, 1993b, 1994; Roeser et a l . , 1996; Wentzel, 1998; Wentzel & Caldwell , 1997). When taken together, these correlational findings provide some evidence to suggest that adolescents who feel a sense of connection and support at school have better school functioning. It should also be noted that with the exception of parental trust for girls, few relations between dimensions of attachment to parents or attachments to peers were found to be significantly associated with school-related functioning. Mult iple regression analyses predicting school functioning. Regression analyses were conducted to further examine relations of autonomy and relatedness as predictors of school functioning. The results supported the hypothesis that autonomy (i.e., parental deidealization, parental nondependency, parental individuation, peer deidealization, peer nondependency, peer individuation, school autonomy) and relatedness (i.e., parental attachment, peer attachment, school belonging) would explain a significant proportion of variance in both teacher-rated school competencies and G P A for boys and for girls (Hypothesis 11). For boys, dimensions of peer autonomy (i.e., deidealization, nondependency, individuation) uniquely predicted higher G P A s . This finding extends the correlational results by suggesting that the relations are still significant after taking into account a l l the other variables. This finding is in concert with research suggesting that emulation of peers (i.e., idealizing peers) may discourage academic engagement (e.g., Ryan et a l . , 1994). Thus, the finding in the present investigation contributes to that literature by 152 providing new evidence to suggest that, for boys only, greater differentiation from peers may allow boys to concentrate on their academics and skills necessary1 for better academic achievement in school. The only other single variable that reached significance in predicting school functioning was school belonging for girls only. School belonging was related to higher ratings of G P A for girls. This finding is consistent with previous research that has shown that adolescents' feelings of belonging and membership at school are important to motivation, academic success and engagement in school (Eccles et a l . , 1993; Goodenow, 1993a, 1993b, 1994). Gil l igan et al . (1990) conducted extensive interviews with adolescent girls from six to eighteen years of age, and reported that girls have an intense interest in intimacy not shared by boys. Gil l igan et al. 's finding may provide an explanation for the gender specific finding in the present study that school belonging for girls uniquely predicted G P A for girls. It may be that school belonging provides girls with opportunities for intimacy which is more highly valued by girls. Thus, peer autonomy and school belonging made independent contributions to the prediction of G P A , although in different ways for boys and for girls. It was surprising to find that the school context variables (i.e., school autonomy, school belonging) were not more influential in predicting school functioning, especially for boys. In the correlational data, positive and significant correlations of teacher-rated school competencies to school autonomy and school belonging were found for boys. Although one might expect that school autonomy and school belonging might uniquely predict school functioning, it may be that when combined with al l the other variables, they become less influential, resulting in reduced relations in the regression equation (c.f., Roeser & Eccles, 153 1998). Relations of autonomy to psychological adjustment. Wi th regard to the association between autonomy and psychological adjustment, it was hypothesized that parental autonomy would be positively associated with problems in psychological adjustment (Hypothesis 7). It may be recalled that psychological adjustment was assessed in two ways: (1) self-reported internalizing problem behaviours, and (2) self-reported externalizing problem behaviours. In the present investigation the findings showed that overall, parental autonomy was found to be positively associated with both internalizing and externalizing problem behaviours for boys and for girls. That is, greater differentiation in relationships with parents was positively associated with internalizing and externalizing problems in similar ways for girls and boys. More specifically, those adolescents who felt more individuated from parents, indicating that his or her parents did not really understand them or know them, reported more internalizing and externalizing problems. For boys, nondependency on parents was associated with more externalizing problems. In contrast, for girls, nondependency on parents and greater parental deidealization were associated with more externalizing problems. These findings are consistent with previous research that has linked perceptions of autonomy in the parent-adolescent relationship to adolescent adjustment. For example, Lamborn and Steinberg (1993) found positive associations between emotional autonomy from parents and problem behaviours (i.e., antisocial behaviour and internal distress) among early and middle adolescents. Other researchers have reported similar results, suggesting that parental insecurity and detachment in relationships with parents may play an important role in psychological adjustment and health risk behaviours (e.g., Chen & Dornbusch, 1998; 154 Fuhrman & Holmbeck, 1995; Turner et a l . , 1993, 1994). The present result supports previous research by suggesting that adolescents who reported greater feelings of emotional autonomy in relationships with parents may be more at risk for internalizing and externalizing problems, although, given the cross-sectional nature of the data, a causal relationship can not be established by the findings. There is a current debate in the literature regarding the adaptive function of emotional autonomy from parents. For example, Lamborn and Steinberg (1993) found that in family situations in which parental support was lacking, greater emotional autonomy from parents was associated with more problem behaviours and low academic competence. Moreover, in situations in which parental support was high, these researchers found that greater emotional autonomy was associated with more problem behaviours, but better psychosocial adjustment and better academic competence. In contrast, Fuhrman and Holmbeck (1995) found that greater emotional autonomy was associated with fewer behaviour problems and higher academic competence in family situations characterized by more adolescent-parent conflict and low maternal warmth. These researchers suggested that higher emotional autonomy from parents may be adaptive for the adolescent in more stressful family situations. The results of the present study support the importance of positive attachments to parents by showing positive relations of parental attachment to school functioning and negative associations with problems in psychological adjustment. However, no consistent pattern of relations were established by the data with respect to dimensions of autonomy in relationships with parents to overall functioning. A s noted by Silverberg and Gondoli (1996), "for most teens, these affective and conceptual changes, sometimes referred to as 155 features of emotional autonomy, neither demand nor signify radical detachment from parents" (p. 17). Indeed, most theorists of adolescence agree that emotional independence from parents is achieved in the contexts of close relationships with parents (e.g., Coll ins , Gleason, et a l . , 1997; Grotevant & Cooper, 1986; H i l l & Holmbeck, 1986). Certainly, more research is needed to clarify the nature and function of emotional autonomy from parents in adolescence. For instance, it would be desirable to examine more narrowly defined groups of adolescents (e.g., those who report both high emotional autonomy and high parental attachment vs. those who report low emotional autonomy and high parental attachment). Such studies would be able to better determine whether emotional autonomy in relationships with parents and attachments to parents could be adaptive. It may also be useful to examine what constitutes stressful family situations in order to better understand how those situations affect emotional autonomy and psychological adjustment. Wi th regard to the associations between peer autonomy and psychological adjustment, it was hypothesized that peer autonomy would be negatively associated with problems in psychological adjustment (Hypothesis 8). N o a priori hypotheses were made regarding relations of specific dimensions of peer autonomy (i.e., deidealization, nondependency, individuation) to problems. Findings from the present study examining deidealization, nondependency, and individuation in relationships with peers, indicated that adolescent boys and girls who reported being less dependent on their peers had fewer externalizing problems. This finding provides support for the hypothesized relation; In contrast, adolescent boys and girls who reported feeling more individuated from their peers (indicating that their peers did not really know them well) reported higher levels of internalizing problems. These results 1 5 6 suggest that nondependency and individuation are differentially associated with adolescent problem behaviours (i.e., depression/anxiety, involvement in delinquent and aggressive behaviours). The finding regarding peer nondependency is consistent with findings from research examining relations of peer autonomy to psychological adjustment problems (e.g., Barber & Olsen, 1997). These researchers found that high peer autonomy (i.e., less psychological control) was negatively associated with feelings of depression and antisocial behaviour, suggesting that adolescents who report less autonomy from their peers, have more problems. Far less is known about peer individuation as operationalized in the present study. In this study, a positive association was found between peer individuation and internalizing problems. This study also found positive associations among peer alienation, peer individuation, and internalizing problems for boys and for girls (see Table 7). Thus, adolescents who reported more internalizing problems also reported feeling more individuated and more alienated in their relationships with peers. It may be that the construct of individuation is conceptually similar to that of alienation in relationships with peers. The results of the present study support theory and extend research on peer autonomy. Peer autonomy was uniquely defined in this study as an individual's perception of him or herself as being differentiated from his or her peers, and as such, represents a different construct of peer autonomy than those used in studies that have assessed peer autonomy as peer pressure, or psychological control (e.g., Barber & Olsen, 1997; Berndt et. al , 1990; Steinberg & Silverberg, 1986). Clearly, additional research is needed to investigate and clarify the relations between different dimensions of peer autonomy (i.e., 157 deidealization, nondependency, individuation) and psychological adjustment. Because the results of the present investigation were correlational in nature, future research efforts should investigate peer autonomy longitudinally in order to discern any causal links between autonomy in peer relations and problems in psychological adjustment. Wi th regard to the association between school autonomy and psychological adjustment, support was found for the hypothesis that school autonomy would be negatively associated with problems in psychological adjustment (Hypothesis 9). A s expected, in the present study, both for boys and for girls, the findings showed that school autonomy was negatively associated with both internalizing and externalizing problems. That is, adolescents who felt they had some personal control or influence in school also reported fewer problem behaviours. This finding is congruent with those of related research. For example, Kasen et al . (1990) found negative associations between student autonomy (i.e., the degree of responsibility and independence that students are given) and emotional and behavioural problems (i.e., conduct problems, depression, anxiety). The results of the present study support the hypothesis and are consistent with other research suggesting that adolescents' perceptions of their experiences of autonomy in school have important implications for motivation and behaviour in school (e.g., Fxcles & Midgley, 1989; Eccles et a l . , 1993; Eccles et a l . , 1997; Heavey et a l . , 1989; Smith et a l . , 1987; Taylor et a l . , 1989). Thus, the present findings contribute to research that examines relations of school autonomy to school-related functioning by applying a measure of school autonomy (i.e., perceived control at school) that includes a wider array of autonomous behaviours than previously measured. 158 Relations of attachment to psychological adjustment. Wi th regard to associations between attachment and psychological adjustment, it was hypothesized that parental attachment, peer attachment, and school belonging would be negatively associated with problems in psychological adjustment (Hypothesis 10). Overall , findings in the present study indicated, as expected, that parental attachment, peer attachment, and school belonging were each associated negatively with internalizing and externalizing problems both for boys and for girls. That is, those adolescents who reported feeling less attached to their parents, peers, and school had more internalizing and externalizing problems. These results are in accord with theory and research linking adolescents' adjustment to experiences of attachment with parents, peers, and school (e.g., A l l e n , Hauser et a l . , 1994; Eccles et a l . , 1997; Wentzel, 1998). For example, research conducted by Eccles and colleagues (Eccles et a l . , 1997) examining associations between emotional attachment and psychological and behavioural functioning among seventh grade boys and girls, found that depressive affect and involvement in problem behaviours were linked negatively to emotional connections in the family, with peers, and to school. Research has also shown that adolescent-parental relationships are potential sources of influence that are strongly related to adolescents' social and emotional functioning and internalizing and externalizing behaviours (e.g., Conger et a l . , 1997; Wentzel & Feldman, 1996). Furthermore, research on peer relationships has documented negative associations between attachment to friends and depressed affect (e.g., Oldenburg & Kerns, 1997), and deviant behaviours (e.g., Gil lmore et a l . , 1992). The results of research examining adolescents' sense of attachment and support at school suggest the theoretical importance of a school belonging for positive social 159 functioning and engagement in school (e.g., Goodenow, 1993a; Wentzel, 1994, 1997). Overall , the results of the present investigation suggest that affective attachments in adolescents' relationships across contexts are strongly related to better psychological adjustment. Indeed, the present study uniquely contributes to the existing body of research by providing evidence that close, caring relationships are important to adolescents' social and emotional adjustment. Mult iple regression analyses predicting psychological adjustment. Regression analyses were conducted to further examine relations of autonomy and relatedness as predictors of psychological adjustment. The results supported the hypothesis that experiences of autonomy and relatedness in parent, peer, and school contexts would explain a significant proportion of variance in both adolescents' self-reported internalizing problems and externalizing problems (Hypothesis 12). The results of the regression analyses regarding internalizing and externalizing problems indicated specialized associations within the parent, peer, and school contexts to problems in psychological adjustment that were different for boys and for girls. Focusing first on internalizing problems, with the exception of parental deidealization and parental nondependency, which were significant and negatively (instead of positively) associated with internalizing problems, parental attachment, peer nondependency, peer individuation, and school belonging uniquely predicted internalizing problems for boys. For girls, attachments to parents and peers were significant and negative predictors of self-reported internalizing problems. The negative relations of parental deidealization and parental nondependency to internalizing problems contrasts with the significant and positive 160 associations found in the correlational data. The difference in the pattern of correlation coefficients for parental deidealization and parental nondependency suggests the presence of a suppressor variable. The detection of a suppressor variable can be done by systematically omitting each variable from the regression analysis until the sign of the coefficient changes to match the correlation (Tabachnick & Fide l l , 1989). Fol lowing this strategy, the parental attachment variable was identified to be the suppressor variable. In a suppression situation, redundant variance in a variable is suppressed, enhancing the importance of that variable in the prediction of the dependent variable (Cohen & Cohen, 1983; Tabachnick & Fide l l , 1989; Tzelgov & Henik, 1991). Suppression frequently occurs when predictors are correlated. In the present study, there were significant negative correlations between dimensions of parental autonomy and parental f attachment. Thus, the parental attachment variable was suppressing the positive relation between parental autonomy and internalizing problems. The findings regarding externalizing problems also indicated unique relations of autonomy and relatedness to self-reported externalizing problems. For boys, parental individuation, parental attachment, and peer nondependency were the only significant predictors of self-reported externalizing problems. For girls, parental attachment and school belonging were uniquely predictive of self-reported externalizing problems. Whereas both dimensions of autonomy and relatedness appear to play a role in the psychological functioning for boys, experiences of attachment are more important to predicting problems in psychological adjustment for girls. The multiple regression analyses revealed that autonomy and relatedness experienced 161 in parent, peer, and school contexts uniquely predicted different areas of functioning for boys and for girls. It was evident that the importance of adolescents' experiences of autonomy and relatedness to predicting problems in psychological adjustment varied by context (i.e., parent, peer, school) and gender. Summary The results of the present investigation are generally consistent with research and literature documenting the relevance of parent, peer, and school experiences of autonomy and relatedness to adolescent development and adjustment (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Roeser & Eccles, 1998; Wenz-Gross et a l . , 1997). Overall , the results of the correlational analyses indicated that autonomy and relatedness in parent, peer, and school contexts were significantly associated with adolescents' school functioning and psychological adjustment in different ways. Several of the correlational findings uniquely contribute to the literature that examines the role that autonomy and relatedness play in adolescent development (e.g., H i l l & Holmbeck, 1986; Grotevant & Cooper, 1986; Ryan & Powelson, 1991; Silverberg & Gondoli , 1996), and in relation to academic and behavioural functioning during adolescence (e.g., Armsden & Greenberg, 1987; Eccles et a l . , 1993; Eccles & Midgley, 1989; Goodenow, 1993a, 1993b, 1994; Paterson et a l . , 1995; Ryan et a l . , 1994; Taylor & Adelman, 1990). Specifically, the results of the present study build on that literature and extend recent research on schooling and mental health (e.g., Roeser, Eccles, & Strobel, 1998) by exploring more specifically dimensions of parental and peer autonomy (i.e., deidealization, nondependency, individuation) and dimensions of parental and peer attachment (i.e., trust, communication, 162 alienation) and by including both academic and behavioural adjustment outcomes. Regression analyses were conducted to further examine the unique predictive effects of adolescents' perceptions of autonomy and relatedness on different areas of functioning (teacher-rated school competencies, GPA, internalizing problems, externalizing problems). The multiple regression procedure allowed for the examination of the simultaneous or combined influence of the individual variables that have been examined previously in the correlational analyses and to identify the unique variables, or combination of variables that best predict adolescents' functioning. The findings of the present study suggest the importance of assessing autonomy and relatedness in more than one context in order to understand the unique contributions to explaining the variations in adolescents' school functioning and psychological adjustment. Implications and Importance of the Study to Education The present research extends our understanding of the adolescent age-period through the examination of adolescents' experiences of autonomy and relatedness in different areas of functioning. The results indicate that unique dimensions of autonomy and relatedness are associated with school or behavioural difficulties. Such associations may be relevant when considering interventions and strategies aimed at providing assistance to adolescents experiencing school or behavioural difficulties. Results from this study can be used to inform intervention research at the school, family, and clinical levels to help adolescents function better in school. It is important for teachers, parents, and other professionals who work with adolescents to realize that adolescents have strong needs to belong and develop independence and that these needs are central developmental tasks of the adolescent period. 163 The knowledge that adolescence is a period of diversity and challenge serves to highlight the potential risk factors associated with adolescent functioning and suggest implications for educational practice, such as targeting adolescents who may be at-risk for school failure or maladjustment (Isakson & Jarvis, 1999; Kazdin, 1993; Wexler, 1991). The results showed that fewer self-reported internalizing problems for boys were uniquely predicted by higher levels of parental attachment, peer nondependency, school belonging, and lower levels of peer individuation. For girls, fewer self-reported internalizing problems were uniquely predicted by higher levels of parental and peer attachments. The results also showed that fewer self-reported externalizing problems for boys were uniquely predicted by higher levels of parental attachment, peer nondependency, and lower levels of parental individuation. For girls, fewer self-reported externalizing problems were uniquely predicted by higher levels of parental attachment and school belonging. These results suggest that fewer internalizing and externalizing problems may result from strategies developed to improve these unique relations. A decreased sense of school membership and autonomy at school may contribute to problem behaviour, poor motivation, and school failure (Eccles & Midgley , 1989; Goodenow, 1993a; Smith et a l . , 1987). Thus, approaches to intervention at the school level aimed at improving education for adolescents might include (a) collaborative communication between teachers and parents, and between teachers and students with input and involvement from other counselling professionals for academic and socioemotional difficulties, (b) teacher support for student autonomy and increased emphasis on empowering students to make decisions and participate in learning that w i l l facilitate school-related processes and encourage 164 student motivation and positive attitudes and behaviours, and (c) the provision of cooperative learning situations that are designed to increase feelings of belonging and membership at school. Parents and peers also play a role in fostering positive school functioning and feelings of belonging at school. The results of the present study indicate that greater perceived attachment to parents and peers is positively associated with school belonging and negatively associated with problem behaviours. Positive attachments in relationships with parents and peers provide support and feelings of security for adolescents, which may be especially critical during school transitions and times of increased academic stress (e.g., Isakson & Jarvis, 1999). The present study contributes to our theoretical understanding of the importance of peer relationships at the transition to high school. It may be recalled that grade nine adolescents were found to be more dependent on their peers and more attached to their peers than either adolescents in grade eight or grade 11. The results also showed that for boys only, more autonomy from peers was associated with better grades, suggesting that academic achievement might be improved for boys when strategies are developed to encourage more autonomy from peers. The results also showed that, for girls only, higher levels of parental trust was associated with higher G P A s , suggesting that academic achievement might be improved for girls when strategies are developed to encourage more parental trust, or perhaps more trust in other adults with parental type roles (e.g., teachers, coaches). Regular communication between parents and teachers about student progress, involvement with peers, and parental relations may assist in identifying and implementing strategies to help ensure successful performance in school. 165 The results showed that, for boys only, higher levels of school autonomy was associated with teacher-rated school competencies, suggesting that school performance might be improved for boys when strategies are developed to encourage them to have more influence over school-related situations. Strategies designed to motivate adolescents to exercise self-management, or maximize participation in educational and treatment programs should focus on themes of autonomy, empowerment, and personal control (e.g., Kaser-Boyd, Adelman, & Taylor, 1985; Taylor & Adelman, 1990; Taylor, Adelman, & Kaser-Boyd, 1984; Wexler, 1991). Educational interventions such as Lawrence Kohlberg's Just Community approach in high school (Higgins, 1991), for example, endorse ideas of participatory democracy. Just Community interventions emphasize student collaboration and participation in decision-making and rule-making as ways to decrease problem behaviours. A s wel l , university laboratory programs designed to enhance perceptions of control at school also serve to illustrate the beneficial aspects of perceived control in relation to attitudes and behaviour (Smith et a l . , 1987; Taylor et a l . , 1989). Ryan and Powelson (1991) posit that in "educational contexts and tasks where students experience support for their autonomy, where they feel connected to and supported by significant others they are likely to be highly motivated. B y contrast, in contexts that are perceived as controlling (vs. autonomy supportive) and where persons feel disconnected or unrelated to significant others, alienation and disengagement are the likely outcomes" (p. 53). Clearly, a better understanding of the influences of parents, peers, and school regarding experiences of autonomy and relatedness is necessary in order to improve education and foster positive adjustment outcomes during adolescence. 166 Strengths and Limitations of the Study and Future Directions The results from the present study extend research and literature on autonomy and relatedness in the following ways. First, two dimensions of adolescent development were investigated in three different contexts (i.e., parent, peer, and school) which permits analyses of differences as a function of grade-related developmental levels. The results showed how adolescents' school functioning and psychological adjustment related to experiences of autonomy and relatedness with parents, with peers, and at school. The present study suggests that different contexts may contribute to adolescents' experiences of autonomy and relatedness in important ways that are associated with better emotional and behavioural functioning and academic outcomes during adolescence (Blum & Rinehart, 1997; Brown, 1990; Eccles et a l . , 1997; Isakson & Jarvis, 1999; Kasen et a l . , 1990; Roeser, 1998). Moreover, adolescents' needs for autonomy and relatedness should not be overlooked when one is investigating schooling and adjustment outcomes. Second, consistent measures were used concurrently across three grade levels and three, con texts. The lack of parallel measures across multiple contexts has been a notable drawback of past research (e.g., Barber & Olsen, 1997; Eccles et a l . , 1997; Ryan & Lynch , 1989). Therefore, findings of the present research should not be attributed to different measures being used. Final ly , the present research utilized multiple sources of information on adolescent functioning obtained from self-reports, teachers, and school records concerning a broad range of outcome measures assessing both academic (i.e., school competencies and school grades) and psychological functioning (i.e., internalizing and externalizing problems). This allowed 167 for the attainment of a cogent and comprehensive portrayal of adolescent functioning across multiple contexts v ia multiple perspectives. There are several limitations of the present study that are important to note. First, the correlational data do not lend themselves to causal interpretation because the relations are based on cross-sectional data. Thus, firm conclusions about the effects of relations of autonomy and relatedness on school and adjustment outcomes cannot be drawn. Longitudinal research is clearly needed in order to make causal inferences concerning the relations found in the present investigation. Indeed, longitudinal studies are needed for examining developmental changes in autonomy and relatedness in relation to adolescent functioning over time. Such studies would serve to strengthen the findings and implications of the results of the present investigation and for better understanding the role that autonomy and relatedness play in adolescent development and functioning. Second, the findings of the present study are exploratory and the nonrandom selection of schools and classrooms in this sample limits generalizability. More specifically, although this study was based on a large sample of adolescents intended to be representative of a population in the early and middle adolescent age ranges, there are several caveats that should be presented. A l l students who participated in the present study attended schools in the same Catholic school district. Approximately one-third of the students were non-Catholic. Thus, the sample may be biased because it includes only Catholic school students, making it different from a randomly selected non-Catholic school sample. There were no students who participated whose primary placement was in a full-time special education program or who did not speak or read English, and thus bias due to under 168 representation may have been introduced. The sample was ethnically diverse comprising, for the most part, of white middle-class adolescents from two-parent families. Because analyses were not conducted by ethnic groups, the generalizability of these findings to adolescents from other racial/ethnic backgrounds and/or family compositions and social classes is limited. There is evidence to indicate that there are patterns of ethnic differences in school performance and academic achievement (e.g., Steinberg, Dornbusch, & Brown, 1992; Sue & Okazaki, 1990). Steinberg et al . (1992) in their study of approximately 15,000 students at nine high schools, comprising of 66% White and nearly equal proportions of Black, Hispanic, and Asian-American argued that "ethnic differences in school performance can be explained more persuasively by examining the interplay between the major contexts in which youngsters develop - the family, the peer group, and the school - than by examining any one of these contexts alone" (p. 724). Their findings suggest that (1) parenting practices affect academic performance and behaviour, (2) the effect of parenting practices is moderated to a large extent by the peer group, and (3) peer support for academic achievement varies by ethnic group. Third , the data for this research relied primarily on adolescents' self-reports. Thus, the findings of this study should be understood in terms of adolescents' perceptions of themselves and their experiences that do not necessarily reflect actual experiences and behaviours (Savin-Williams & Small , 1986). Whi le self-reports can provide considerable information about perceived experiences and unobserved attitudes and behaviours, they are subject to both underreporting and overreporting and are strongly influenced by the individual's interpretation and understanding of the questions, and thus introduce bias in this 169 regard (Achenbach, 1991; M c C o r d , 1990; Schwarz, 1999). However, self-reports emphasize the importance of adolescents' beliefs about themselves and their experiences across situations and contexts which was the focus of this study. Final ly , there are a number of measurement issues that need to be taken into account when considering the results of this study. First, the observed relations among the variables in this study may be inflated due to common method variance with respect to adolescents' self-reports, and classroom teachers' ratings of students' school competencies. Teacher-ratings of school competencies were provided by one classroom teacher, which at the high school level, represented information about school functioning in a single subject area. To improve the assessment of adolescents' school-related functioning, perhaps future research might include a composite of teacher-ratings of school competencies in the four academic classes included in the composite G P A . Furthermore, obtaining integrative reports of problem behaviours from teachers and parents, in addition to those obtained from adolescents, would provide a more complete picture of psychological functioning that is based on more than one informant. Another important measurement issue that should be considered concerns the use of specific subscales of the measures assessing parental and peer autonomy. Because the parental nondependency subscale, and the peer deidealization and nondependency subscales did not have reliabilities as high as might be desired, results of analyses examining differences on the subscales must be interpreted with this in mind. The low reliabilities on these subscales may be a result of sample characteristics and error. It may be recalled, however, that much of the past research was conducted using composite scale scores rather 170 than subscales. Subscale scores allow for a more thorough and complete analysis of the variables of interest and thus may help to shed light on the function of these constructs. A s suggested from the results of the present study, it is likely that individual subscales might reveal important information about emotional autonomy in relationships with peers. Thus, further research that investigates hypotheses relating to different dimensions of emotional autonomy from peers and adolescent adjustment is warranted. Studies of peer autonomy that include both measures Of behaviour and emotional aspects of peer relationships would be a valuable addition to research examining the development of peer autonomy during adolescence. Moreover, the assessment of emotional autonomy from peers, as a construct, requires further study as a valid measure of autonomy from peers that is conceptually distinct from measures of peer conformity or peer pressure. For example, an important task for further research is to determine the way in which adolescents conceptualize peer autonomy. Methodologies designed to examine the qualitative features of peer autonomy would serve to strengthen and refine the meaning of the construct and its interpretation. Notwithstanding the limitations inherent in this study, the results of this research provide impetus for future speculation and studies of autonomy and relatedness on adjustment during adolescence. Concluding Remarks The present research is part of the growing number of studies that focus on promoting school success and mental health during adolescence. One criticism of previous work in the field of adolescent development has been the paucity of research examining the unique and 171 combined contribution of various dimensions of development within different social contexts on adolescents' functioning. Both autonomy and relatedness have been afforded significance in the developmental literature as being two dimensions of development that are important for understanding adolescent functioning. Researchers have identified parents, peers, and school as the major social contexts in which adolescents experience autonomy and relatedness. This study provides new empirical data and new perspectives on the way in which autonomy and relatedness are associated with school functioning and psychological adjustment within the three major social arenas of adolescents, namely parents, peers, and school. > The results of this research indicate that adolescents' experiences of autonomy and relatedness are associated with school-related functioning and problems in psychological adjustment. The findings have important implications for educators and parents of adolescents for improving educational practice and for promoting school success and positive adjustment outcomes during adolescence. 172 References Achenbach, T . M . (1991). Manual for the Youth Self-Report and 1991 Profile. Burlington, V T : University of Vermont Department of Psychiatry. Adelman, H . S., Smith, D . C , Nelson, P . , Taylor, L . , & Phares, V . (1986). A n instrument to assess students' perceived control at school. Educational and Psychological Measurement.46. 1005-1017. Adelman, H . S., & Taylor, L . (1990). Intrinsic motivation and school misbehavior: Some intervention implications. Journal of Learning Disabilities. 25(9). 541-550. A l l e n , J . P . , Aber, J . L . , & Leadbeater, B . J . (1990). Adolescent problem behaviors: The influence of attachment and autonomy. Psychiatric Clinics of North America. 13(3). 455-467. Al l en , J . P . , & Hauser, S. T . (1996). Autonomy and relatedness in adolescent-family interactions as predictors of young adults' states of mind regarding attachment. Development and Psychopathology. 8. 793-809. A l l e n , J . P . , Hauser, S. T . , Eickholt, C , Be l l , K . L . , & O'Connor, T . G . (1994). Autonomy and relatedness in family interactions as predictors of expressions of negative adolescent affect. Journal of Research on Adolescence. 4(4). 535-552. A l l e n , J . P . , Kuperminc, G . , Philliber, S., & Herre, K . (1994). Programmatic prevention of adolescent problem behaviors: The role of autonomy, relatedness, and volunteer service in the Teen Outreach Program. American Journal of Community Psychology. 22(5). 617-638. Al l en , J . P . , Moore , C , Kuperminc, G . , & Bel l , K . (1998). Attachment and psychosocial functioning. Chi ld Development. 69(5). 1406-1419. American Psychological Association (1995). Publication manual of the American Psychological Association (4th ed.). Washington, D C : American Psychological Association. Armsden, G . C , & Greenberg, M . T. (1987). The Inventory of Parent and Peer Attachment: Individual differences and their relationship to psychological well-being in adolescence. Journal of Youth and Adolescence. 16(5). 427-454. Asher, S;. R . , & Hymel , S. (1981). Children's social competence in peer relations: Sociometric and behavioral assessment. In J . D . Wine & M . D . Smye (Eds.), Social competence (pp. 125-157). New York: Guilford. 173 Barber, B . K . (1997). Introduction: Adolescent socialization in context - connection, regulation, and autonomy in multiple contexts. Journal of Adolescent Research. 12(2). 173-177. Barber, B . K . , & Olsen, J . A . (1997). Socialization in context: connection, regulation, and autonomy in the family, school, and neighborhood, and with peers. Journal of Adolescent Research. 12(2). 287-315. Battistich, V . , Solomon, D . , Watson, M . , & Schaps, E . (1997). Caring school communities. Educational Psychologist. 32(3). 137-151. Bardone, A . M . , Moffitt, T . E . , Caspi, A . , Dickson, N . , & Silva, P . A . (1996). Adult mental health and social outcomes of adolescent girls with depression and conduct disorder. Development and Psychopathology. 8. 811-829. Bear, G . G . , & Ryes, G . S. (1994). Mora l reasoning, classroom behavior, and sociometric status among elementary school children. Developmental Psychology. 30(5). 633-638. Berndt, T . J . (1979). Developmental changes in conformity to peers and parents. Developmental Psychology. 15(6). 608-616. Berndt, T . J . (1989). Obtaining support from friends in childhood and adolescence. In D . Belle (Ed.), Children's social networks and social supports (pp. 308-331). New York : Wi ley . Berndt, T . J . (1999). Friends' influence on students' adjustment to school. Educational Psychologist. 34(1). 15-28. Berndt, T . J . , Laychak, A . G . , & Park, K . (1990). Friends' influence on adolescents' academic achievement and motivation: A n experimental study. Journal of Educational Psychology. 82(4). 664-670. Birleson, P . (1981). The validity of depressive disorder in childhood and the development of a self-rating scale: A research report. Journal of Psychology and Psychiatry. 22, 73-88. Bla in , M . D . , Thompson, J . M . , & Whiffen, V . E . (1993). Attachment and perceived social support in late adolescence: The interaction between working models of self and others. Journal of Adolescent Research. 9(2). 226-241. 174 Blishen, B . R . , Carrol l , W . K . , & Moore, C . (1987). The 1981 socioeconomic index for occupations in Canada. Canadian Review of Sociology and Anthropology. 24(4). 465-488. Bios , P . (1962). On adolescence. New York: The Free Press of Glencoe. Bios, P . (1967). The second individuation process of adolescence. In R . S. Eissler (Ed.), Psychoanalytic study of the child. ( V o l . 22, pp. 162-186). New York : International Universities Press. B lum, R . W . , & Rinehart, P . M . (1997). Reducing the risk: Connections that make a difference in the lives of youth. Minneapolis, M N : Divis ion of General Pediatrics and Adolescent Health, University of Minnesota. Bray, J . H . , & Maxwel l , S. E . (1982). Analyzing and interpreting significant M A N O V A s . Review of Educational Research. 52(3). 340-367. Brier, N . (1995) Predicting antisocial behavior in youngsters displaying poor academic achievement: A review of the risk factors. Developmental and Behavioral Pediatrics. 16(4). 271-276. Brown, B . (1990). Peer groups and peer cultures. In S. S. Feldman & G . R . Elliott (Eds.), A t the threshold: The developing adolescent (pp. 171-196). Cambridge, M A : Harvard University Press. Brown, B . B . , Clasen, D . R . , & Eicher, S. A . (1986). Perceptions of peer pressure, peer conformity dispositions, and self-reported behavior among adolescents. Developmental Psychology.22(4). 521-530). Chen, Z . , & Dornbusch, S. M . (1998). Relating aspects of adolescent emotional autonomy to academic achievement and deviant behavior. Journal of Adolescent Research. 13(3), 293-319. Cohen, J . , & Cohen, P . (1983). Applied multiple regression/correlation analysis for the behavioural sciences (2nd ed.). Hillsdale, N J : Erlbaum. Coll ins, W . A . , Gleason, T . , & Sesma, A . (1997). Internalization, autonomy, and relationships: Development during adolescence. In J . E . Grusec and L . Kuczynski (Eds.), Parenting and children's internalization of values (pp. 78-79). New York : John Wi ley & Sons. Coll ins , W . A . , Laursen, B . , Mortensen, N . , Luebker, C , & Ferreira, M . (1997). Conflict processes and transitions in parent and peer relationships: Implications for autonomy and regulation. Journal of Adolescent Research. 12(2). 178-198. 175 Conger, K . J . , Conger, R . D . , & Scaramella, L . V . (1997). Parents, siblings, psychological control, and adolescent adjustment. Journal of Adolescent Research. 12(1). 113-138. Cotterell, J . L . (1992). The relation of attachments and supports to adolescent wel l -being and school adjustment. Journal of Adolescent Research. 7. 28-42. Crittenden, P . M . (1990). Toward a concept of autonomy in adolescents with a disability. Children's Health Care. 19. 162-168. Dec i , E . L . , & Ryan, R . M . (1987). The support of autonomy and control of behavior. Journal of Personality and Social Psychology. 53(6). 1024-1037. Dec i , E . L . , Vallerand, R . J . , Pelletier, L . G . , & Ryan, R . M . (1991). Motivation and education: The self-determination perspective. Educational Psychologist. 26(3). 325-346. Dekovic, M . , & Meeus, W . (1997). Peer relations in adolescence: Effects of parenting and adolescents' self-concept. Journal of Adolescence. 20. 163-176). Dishion, T . J . , Patterson, G ; . R . , Stoolmiller, M . , & Skinner, M . L . (1991). Family, school, and behavioral antecedents to early adolescent involvement with antisocial peers. Developmental Psychology. 27(1). 172-180). Douvan, E . , & Adelson, J . (1966). The adolescent experience. New York : John Wi ley & Sons. Downie, N . M . , & Starry, A . R . (1977). Descriptive and inferential statistics. New York : Harper & Row. Eccles, J . S., Buchanan, C . M . , Flanagan, C , Ful igni , A . , Midgley , C , & Yee, D . (1991). Control versus autonomy during early adolescence. Journal of Social Issues. 47(4). 53-68. Eccles, J . S., Early, D . , Frasier, K . , Belansky, E . , & McCarthy, K . (1997). The relation of connection, regulation, and support for autonomy to adolescents' functioning. Journal of Adolescent Research. 12(2). 263-286). Eccles, J . S., Lord , S., & Midgley, C . (1991). What are we doing to early adolescents? The impact of educational contexts on early adolescents. American Journal of Education. 99(4). 521-542. Eccles, J . S., & Midgley, C . (1989). Stage-environment fit: Developmentally appropriate classrooms for young adolescents. In R. Ames & C . Ames (Eds.), Research in motivation in education ( V o l . 3, pp. 139-186). Orlando, F L : Academic Press. 176 Eccles, J . S., Midgley, C , & Adler, T . F . (1984). Grade-related changes in the school environment: Effects on achievement motivation. In J . G . Nicholls (Ed.), Advances in motivation and achievement (pp. 283-331). Greenwich, C T : J A I Press. Eccles, J . S., Midgley, C , Wigfield, A . , Buchanan, C , Reuman, D . , Flanagan, C , & M a c Iver, D . (1993). Development during adolescence: The impact of stage-environment fit on young adolescents' experiences in schools and in families. American Psychologist. 48(2), 90-101. Epstein, S. (1983). Scoring and interpretation of mother-father-peer scale. Unpublished manuscript, University of Massachusetts. Fasick, F . A . (1984). Parents, peers, youth culture and autonomy in adolescence. Adolescence. 19(73). 143-157. Feldman, S. S., & Elliott, G . (1990). Progress and promise of research on adolescence. In S. S. Feldman & G . R . Elliott (Eds.), A t the threshold: The developing adolescent (pp. 479-505). Cambridge, M A : Harvard University Press. Fitts, W . H . (1965). Tennessee Self-Concept Scale Manual . Nashville, T N : Counselor Recordings and Tests. Foley, R . M . , & Epstein, M . H . (1992). Correlates of academic achievement of adolescents with behavioral disorders. Behavior Disorders. 18(1). 9-17. Ford , M . E . (1982). Social cognition and social competence in adolescence. Developmental Psychology. 18. 323-340. Frank, S. J . , Pirsch, L . A . , & Wright, V . C . (1990). Late adolescents' perceptions of their relationships with their parents: Relationships among deidealization, autonomy, relatedness, and insecurity and implications for adolescent adjustment and ego identity status. Journal of Youth and Adolescence. 19(6). 571-588. Fuhrman, T . , & Holmbeck, G . (1995). A contextual-moderator analysis of emotional autonomy and adjustment in adolescence. Chi ld Development. 66. 793-811. Ful igni , A . J . , & Eccles, J . S. (1993). Perceived parent-child relationships and early adolescents' orientation toward peers. Developmental Psychology. 20(4). 622-632. Gavazzi, S. M . , Anderson, S. A . , & Sabatelli, R . M . (1993). Family differentiation, peer differentiation, and adolescent adjustment in a clinical sample. Journal of Adolescent Researcher. 8(2). 205-225. 177 Gavin, L . A . , & Furman, W . (1989). Age differences in adolescents' perceptions of their peer groups. Developmental Psychology. 25(51. 827-834. Gesten, E . L . (1976). A Health Resources Inventory: The development of a measure of the personal and social competence of primary grade children. Journal of Consulting and Cl in ica l Psychology.44. 775-786. Geuzaine, C , Debry, M . , & Liesens, V . (2000). Separation from parents in late adolescence: The same for boys and girls? Journal of Youth and Adolescence. 29(1). 79-91. Ghisel l i , E . E . , Campbell, J . P . , Zedeck, S. (1981). Measurement theory for the behavioural sciences. New York : W . H . Freeman & C o . . Gil l igan, C , Rogers, A . , & Brown, L . M . (1990). Soundings into development. In C . Gil l igan, N . P . Lyons, & T. J . Hanmer (Eds.), The relational worlds of adolescent girls at the Emma Wil la rd School. Cambridge, M A : Harvard University Press. Gil lmore, M . R . , Hawkins, J . D . , Day, L . E . , & Catalano, R . F . (1992). Friendship and deviance: New evidence on an old controversy. Journal of Early Adolescence. 12(1). 80-95. Glasser, W . (1986). Control theory in the classroom. New York : Harper & Row. Goldbaum, S., & Crawford, A . M . (Apr i l 2000). Family and peer relationships of early adolescents with internalizing problems. Paper presented at the Eighth Biennial Meeting of the Society for Research on Adolescence, Chicago, I L . Goodenow, C . (1992). Strengthening the links between educational psychology and the study of social contexts. Educational Psychologist. 27(2). 177-196). Goodenow, C . (1993a). Classroom belonging among early adolescent students: Relationships to motivation and achievement. Journal of Early Adolescence. 12(1). 21-43. Goodenow, C . (1993b). The psychological sense of school membership among adolescents: Scale development and educational correlates. Psychology in the Schools. 30. 79-90. Goodenow, C . (1994, February). Conceptualizing and measuring classroom belonging and support among adolescents. Paper presented at the Biennial Meeting of the Society for Research on Adolescence, San Diego, C A . Greenberg, M . T. (1982). Reliability and validity of the Inventory of Adolescent Attachments. Unpublished manuscript, University of Washington. 178 Greenberg, M . T . , Siegel, J . M . , & Leitch, C J . (1983). The nature and importance of attachment relationships to parents and peers during adolescence. Journal of Youth and Adolescence. 12(5). 373-386. Greenberger, E . , Steinberg, L . , & Vaux, A . (1982). Person-environment congruence as a predictor of adolescent health and behavioral problems. American Journal of Community Psychology. 10(5). 511-526. Grolnick, W . S., Ryan, R . M . , & Deci , E . L . (1991). Inner resources for school achievement: Motivational mediators of children's perceptions of their parents. Journal of Educational Psychology. 83(4). 508-517. Grotevant, H . D . , & Cooper, C . R . (1986). Individuation in family relationships: A perspective on individual differences in the development of identity and role-taking skills in adolescence. Human Development. 29. 82-100. Havighurst, R . J . (1952). Developmental tasks and education (2nd ed.). New York : David M c K a y . Havighurst, R . J . (1972). Developmental tasks and education (3rd ed.). New York : David M c K a y . Heavey, C . L . , Adelman, H . S., Nelson, P . , & Smith, D . C . (1989). Learning problems, anger, perceived control, and misbehavior. Journal of Learning Disabilities. 22(1). 46-59. Herman, M . R . , Dornbusch, S. M . , Herron, M . C , & Herting, J . R . (1997). The influence of family regulation, connection, and psychological autonomy on six measures of adolescent functioning. Journal of Adolescent Research. 12(1). 34-67. Higgins, A . (1991). The just community approach to moral education: Evolution of the idea and recent findings. In W . M . Kurtines & J , L . Gewirtz (Eds.), Handbook of moral behavior and development: V o l . 3. Application (pp. 111-141). Hightower, A . D . , Spinell, A . , & Lotyczewski, B . S. (1989). Teacher-Child Rating Scale (T-CRS) guidelines (Rev. August 1990). Primary Mental Health Project. Hightower, A . D . , Work, W . C , Cohen, E . L . , Lotyczewski, B . S., Spinell , A . P . , Guare, J . C , & Rohrbeck, C . A . (1986). The Teacher-Child Rating Scale: A brief objective measure of elementary children's school problem behaviors and competencies. School Psychology Review. 15. 393-409. H i l l , P . (1993). Recent advances in selected aspects of adolescent development. Journal of Chi ld Psychology and Psychiatry. 34(1). 69-99. 179 H i l l , J . P . , & Holmbeck, G . N . (1986). Attachment and autonomy during adolescence. Annals of Chi ld Development. 3. 145-189. Hodgins, H . S., Koestner, R . , & Duncan, N . (1996). On the compatibility of autonomy and relatedness. Personality and Social Psychology Bulletin. 22(3). 227-237. Hopkins, K . D . , & Weeks, D . L . (1990). Tests for normality and measures of skewness and kurtosis: Their place in research reporting. Educational and Psychological Measurement. 50. 717-729. Huberty, C . J . , & Morr is , J . D . (1989). Multivariate analysis versus multiple univariate analyses. Psychological Bulletin. 105(2). 302-308). Isakson, K . , & Jarvis, P . (1999). The adjustment of adolescents during the transition into high school: A short-term longitudinal study. Journal of Youth and Adolescence. 28(1). 1-26. Jenson, W . R . , Walker, H . M . , Clark, E . , & Kehle, T . (1991). Behavior disorders: Training needs for school psychologists. In G . Stoner, M . R . Shinn, & H . M . Walker (Eds.). Interventions for achievement and behavior problems (pp. 763-786). National Association of School Psychologists. Johnson, D . W . , Johnson, R . T . , Buckman, L . A . , & Richards, P . S. (1985). The effect of prolonged implementation of cooperative learning on social support within the classroom. The Journal of Psychology. 119. 405-411. Josselson, R . (1980). Ego development in adolescence. In J . Adelson (Ed.), Handbook of adolescent psychology (pp. 188-210). New York : John Wi ley . Kasen, S., Johnson, J . , & Cohen, P . (1990). The impact of school emotional climate on student psychopathology. Journal of Abnormal Chi ld Psychology. 18(2). 165-177. Kaser-Boyd, N . , Adelman, H . S., & Taylor, L . (1985). Minor ' s ability to identify risks and benefits of therapy. Professional Psychology: Research and Practice. 16(3). 411-417. Kazdin , A . E . (1993). Adolescent mental health: Prevention and treatment programs. American Psychologist. 48. 127-141. Kazdin , A . E . (1995). Conduct disorders in childhood and adolescence (2nd ed., V o l . 9). Thousand Oaks, C A : Sage. 180 Kenny, M . E . , Lomax, R . , Brabeck, M . , & Fife, J . (1998). Longitudinal pathways linking adolescent reports of maternal and paternal attachments to psychological well-being. Journal of Early Adolescence. 18(3). 221-243. Kohn , A . (1993, September). Choices for children: Why and how to let students decide. Ph i Delta Kappan. 9-20. Kovacs, M . (1981). Rating scales to assess depression in school-aged children. Acta Paedopsychiatrica. 46. 305-315. Kuperminc, G . P . , Al l en , J . P . , & Arthur, M . W . (1996). Autonomy, relatedness, and male adolescent delinquency: Toward a multidimensional view of social competence. Journal of Adolescent Research. 11(4). 397-420. Laible, D . J . , Carlo, G . , & Raffaelli, M . (2000). The differential relations of parent and peer attachment to adolescent adjustment. Journal of Youth and Adolescence. 29(1). 45-59. Lamborn, S. D . , & Steinberg, L . (1993). Emotional autonomy redux: Revisiting Ryan and Lynch . Chi ld Development. 64. 483-499. Lapsley, D . K . (1990). Continuity and discontinuity in adolescent social cognitive development. In R . Montemayor, G . R . Adams, & T. P . Gullotta (Eds.), F rom childhood to adolescence: A transitional period? (pp. 183-204). Newbury Park, C A : Sage. Larose, S., & Boiv in , M . (1998). Attachment to parents, social support expectations, and socioemotional adjustment during the high school-college transition. Journal of Research on Adolescence. 8(1). 1-27. Larson, R . W . (1997). The emergence of solitude as a constructive domain of experience in early adolescents. Chi ld Development. 68(1). 80-93. Lor ion , R . P . , Cowen, E . L . , & Caldwell , R. A . (1975). Normative and parametric analyses of school maladjustment. American Journal of Community Psychology. 3. 293-301. Luthar, S. S. (1995). Social competence in the school setting: Prospective cross-domain associations among inner-city teens. Chi ld Development. 66. 416-429. Mazor , A . , & Enright, R . D . (1988). The development of the individuation process from a social cognitive perspective. Journal of Adolescence. 11. 29-47. M c C o r d , J . (1990). Problem behaviors. In S. S. Geldman & G . R . Elliott (Eds.), A t the threshold: The developing adolescent (pp. 414-430). 181 M c N e i l , L . M . (1986). Contradictions of control: School structure and school knowledge. New York : Routledge. Merre l l , K . W . (1989). Concurrent relationships between two behavioral rating scales for teachers: A n examination of self-control, social competence, and school behavioral adjustment. Psychology in the Schools. 26. 267-271. Midgley, C . & Feldlaufer, H . (1987). Students' and teachers' decision-making fit before and after the transition to junior high school. Journal of Early Adolescence. 7(2). 225-241. Midgley , C , Feldlaufer, H . , & Eccles, J . S. (1988). The transition to junior high school: Beliefs of pre- and posttransition teachers. Journal of Youth and Adolescence. 17(61. 543-562. Minuchin , P . P . , & Shapiro, E . K . (1983). The school as a context for social development. In P . H . Mussen (Ed.), Handbook of child psychology: Formerly Carmichael's manual of child psychology (4th ed.) (pp. 197-274). New York : Wi ley . Moos , R . H . (1974). Family Environment Scale. Palo A l to , C A : Consulting Psychologists Press. Murphy, K . & Davidshofer, C . (1998). Psychological testing: Principles and applications (4th ed.). Upper Saddle, N J : Prentice H a l l . Murtaugh, M . , & Zetlin, A . G . (1990). The development of autonomy among learning handicapped and nonhandicapped adolescents: A longitudinal perspective. Journal of Youth and Adolescence. 19(31. 245-255. ' Norusis, M . J . (1997). SPSS 7.5 guide to data analysis. Upper Saddle River, N J : Prentice-Hall. O 'Br ien , R . . W . (1989, Apr i l ) . Family influences on the development of individuation during adolescence. Paper presented at the Biennial Meeting of the Society for Research in Chi ld Development, Kansas Ci ty , M O . O 'Br ien , E . , & Epstein, S. (1988). The Multidimensional Self-Esteem Inventory: Professional manual. Odessa, F L : Psychological Assessment Resources. Offer, D . , & Schonert-Reichl, K . A . (1992). Debunking the myths of adolescence: Findings from recent research. Journal of the American Academy of Chi ld and Adolescent Psychiatry. 31(6). 1003-1014. 182 Oldenburg, C . M . , & Kerns, K . A . (1997). Associations between peer relationships and depressive symptoms: Testing moderator effects of gender and age. Journal of Early Adolescence. 17(3). 319-337. Papini, D . R . & Roggman, L . A . (1992). Adolescent perceived attachment to parents in relation to competence, depression, and anxiety: A longitudinal study. Journal of Early Adolescence. 12(4). 420-440. Papini, D . R . , Roggman, L . A . , & Anderson, J . (1991). Early-adolescent perceptions of attachment to mother and father: A test of the emotional-distancing and buffering hypotheses. Journal of Early Adolescence. 11(2). 258-275. Parker, J . G . , & Asher, S. R . (1987). Peer acceptance and later personal adjustment: Are low-accepted children "at-risk"? Psychological Bulletin. 101. 357-359. Parker, J . G . , & Asher, S. R . (1993). Friendship and friendship quality in middle childhood; Links with peer group acceptance and feelings of loneliness and social dissatisfaction. Developmental Psychology. 19. 611 -621. Paterson, J . , Pryor, J . , & Fie ld , J . (1995). Adolescent attachment to parents and friends in relation to aspects of self-esteem. Journal of Youth and Adolescence. 24(3). 365-376. Pedhazur, E . J . (1982). Mult iple regression in behavioral research (2nd ed.) New York : Holt , Rinehart & Winston. Pintrich, P . , & DeGroot, E . (1990). Motivational and self-regulated learning components of classroom academic performance. Journal of Educational Psychology. 32. 33-40. Quay, H . C . , & Peterson, D . R . (1987). Manual for the Revised Behavior Problem Checklist. M i a m i , F L : Herbert C . Quay & Donald R . Peterson. Quintana, S. M . , & Lapsley, D . K . (1990). Rapprochement in late adolescent separation-individuation: a structural equations approach. Journal of Adolescence. 13. 371-385. Reid, J . B . , & Patterson, G . R . (1991). Early prevention and intervention with conduct problems: A social interactional model for the integration of research and practice. In G . Stoner, M . R . Shinn, & H . M . Walker, H . M . (Eds.), Interventions for achievement and behavior problems (pp. 763-786). Silver Spring, M D : National Association of School Psychologists. 183 Roeser, R . W . (1998). On schooling and mental health: Introduction to the special issue. Educational Psychologist. 33(4). 129-133. Roeser, R . W . , & Eccles, (1998). Adolescents' perceptions of middle school: Relation to longitudinal changes in academic and psychological adjustment. Journal of Research on Adolescence. 8(1). 123-158. Roeser, R . W . , Eccles, J . S., & Strobel, K . R . (1998). L inking the study of schooling and mental health: Selected issues and empirical illustrations at the level of the individual. Educational Psychologist. 33(4). 153-176. Roeser, R . W . , Midgley, C , & Urdan, T. C . (1996). Perceptions of the school psychological environment and early adolescents' psychological and behavioral functioning in school: The mediating role of goals and belonging. Journal of Educational Psychology. 88(3), 408-422. Ryan, R . M . , & Grolnick, W . (1986). Origins and pawns in the classroom: Self-report and projective assessments of individual differences in children's perceptions. Journal of Personality and Social Psychology. 50(3). 550-558. Ryan, R . M . , & Lynch , J . H . (1989). Emotional autonomy versus detachment: Revisiting the vicissitudes of adolescence and young adulthood. Ch i ld Development. 60. 340-356. Ryan, R . M . , & Powelson, C . L . (1991). Autonomy and relatedness as fundamental to motivation and education. Journal of Experimental Education. 60(1). 49-66. Ryan, R . M . , Stiller, J . D . , & Lynch , J . H . (1994). Representations of relationships to teachers, parents, and friends as predictors of academic motivation and self-esteem. Journal of Early Adolescence. 14(2). 226-249. Saleh, W . (1991). Behavior problems in our schools: A lesson in patience, tolerance and understanding. Mississauga, O N : Dufferin-Peel Roman Catholic Separate School Board. Savin-Williams, R . C , & Berndt, T . (1990). Friendship and peer relations. In S. S. Feldman & G . R . Elliott (Eds.), A t the threshold: The developing adolescent (pp. 277-307). Cambridge, M A : Harvard University Press. Savin-Williams, R . C , & Small , S. A . (1986). The timing of puberty and its relationship to adolescent and parent perceptions of family interactions. Developmental Psychology. 22(3). 342-347. 184 Scaramella, L . V . , Conger, R . D . , & Simons, R . L . (1999). Parental protective influences and gender-specific increases in adolescent internalizing and externalizing problems. Journal of Research on Adolescence. 9(21. 111-141. Schonert-Reichl, K . A . (1994). The interrelations of moral reasoning, empathy, and age among adolescents with behavioural disorders. B . C . Journal of Special Education. 18(3). 248-262. Schwarz, N . (1999). Self-reports: H o w the questions shape the answers. American Psychologist. 93-105. Shulman, S. (1993). Close relationships and coping behavior in adolescence. Journal of Adolescence. 16. 267-283. Silverberg, S. B . & Gondoli , D . M . (1996). Autonomy in adolescence: A contextualized perspective. In G . R. Adams, R . Montemayor, & T . P . Gullotta (Eds.), Psychosocial development during adolescence (pp. 12-61). Thousand Oaks, C A : Sage. Simmons, R . G . , & Blyth, D . A . (1987). Moving into adolescence: The impact of pubertal change and school context. New York: Aldine DeGruyter. Smith, D . C , Adelman, H . S., Nelson, P . , Taylor, L . , & Phares, V . (1987). Students' perception of control at school and problem behavior and attitudes. Journal of School Psychology. 25. 167-176. Sprick, R . S., & Nolet, V . (1991). Prevention and management of secondary-level behavior problems. In G . Stoner, M . R . Shinn, & H . M . Walker (Eds.), Interventions for achievement and behavior problems (pp. 519-538). Silver Spring, M D : National Association of School Psychologists. Steinberg, L . (1988). Reciprocal relation between parent-child distance and pubertal maturation. Developmental Psychology. 24(1). 122-128. Steinberg, L . (1990). Autonomy, conflict, and harmony in the family relationship. In S. S. Feldman & G . R . Elliott (Eds.), A t the threshold: The developing adolescent (pp. 255-276). Cambridge, M A : Harvard University Press. Steinberg, L . (1993). Adolescence (3rd ed.). New York : M c G r a w - H i l l , Inc.. Steinberg, L . , Dornbusch, S. M . , & Brown, B . B . (1992). Ethnic differences in adolescent achievement; A n ecological perspective. American Psychologist.47(6). 723-729. Steinberg, L . , & Silverberg, S. (1986). The vicissitudes of autonomy in early adolescence. Ch i ld Development. 57. 841-851. 185 Stern, L . (1990). Conceptions of separation and connection in female adolescents. In C . Gil l igan, N . P . Lyons, & T. J . Hanmer (Eds.), Making connections: The relational worlds of adolescent girls at Emma Wil lard School (pp. 73-87). Cambridge, M A : Harvard University Press. Stevens, J . (1992). Applied multivariate statistics for the social sciences (2nd ed.). Hillsdale, N J : Lawrence Erlbaum. Stipek, D . J . , & Weisz, J . R . (1981). Perceived personal control and academic achievement. Review of Educational Research. 51(11. 101-137. Sue, S., & Okazaki, S. (1990). Asian-American educational achievement: A phenomenon in search of an explanation. American Psychologist. 45. 913-920. Tabachnick, B ; . G . , & Fide l l , L . S. (1989). Using multivariate statistics (2nd ed.). New York : Harper-Collins. Taylor, L . , Adelman, H . S., & Kaser-Boyd, N . (1984). Attitudes toward involving minors in decisions. Professional Psychology: Research and Practice. 15(31. 436-449. Taylor, L . , & Adelman, H . S. (1990). School avoidance behavior: Motivational bases and implications for intervention. Chi ld Psychiatry and Human Development. 205(4). 219-233. Taylor, L . , Adelman, H . S., Nelson, P . , Smith, D . C , & Phares, V . (1989). Perceptions of control at school among students in special education programs. Journal of Learning Disabilities. 22(7). 439-443. Turner, R . A . , Irwin, C . E . , Tschann, J . M . , & Mil ls te in , S. G . (1993). Autonomy, relatedness and the initiation of health risk behaviors in early adolescence. Health Psychology. 12(3). 200-208. Turner, R . A . , Irwin, C . E . , Tschann, J . M . , & Mil ls te in , S. G . (1994, March). Autonomous-Relatedness and Changes in Self-esteem During Early Adolescence. Paper presented at the biennial meeting of the Society for Research on Adolescence, San Diego. Tzelgov, J . , & Henik, A . (1991). Suppression situations in psychological research: Definitions, implications, and applications. Psychological Bulletin. 109(3). 524-536. Weinberger, D . A . , Feldman, S. S., Ford, M . E . , & Chastain, R . L . (1987). Construct validation of the Weinberger Adjustment Inventory. Unpublished manuscript, Stanford University. 186 Wentzel, K . R . (1991). Social competence at school: Relation between social responsibility and academic achievement. Review of Educational Research. 62(1). 1,-24. Wentzel, K . R . (1993). Does being good make the grade? Social behavior and academic competence in middle school. Journal of Educational Psychology. 85(2). 357-364. Wentzel, K . R . (1994). Relations of social goal pursuit to social acceptance, classroom behavior, and perceived social support. Journal of Educational Psychology. 86(2). 173-182. Wentzel, K . R . (1996). Social goals and social relationships as motivators of school adjustment. In J . Juvonen & K . R . Wentzel (Eds), Social motivation: Understanding children's school adjustment (pp. 226-247). Cambridge, M A : Cambridge University Press. Wentzel, K . R . (1997). Student motivation in middle school: The role of perceived pedagogical caring. Journal of Educational Psychology. 89(3). 411-419. Wentzel, K . R . (1998). Social relationships and motivation in middle school: The role of parents, teachers, and peers. Journal of Educational Psychology. 90(2). 202-209. Wentzel, K . R . , & Caldwell , K . (1997). Friendships, peer acceptance, and group membership: Relations to academic achievement in middle school. Chi ld Development. 68(6), 1198-1209). Wentzel, K . R . , & Feldman, S. S. (1996). Relations of cohesion and power in family dyads to social and emotional adjustment during early adolescence. Journal of Research on Adolescence.6(2). 225-244. Wenz-Gross, M . , Siperstein, G . N . , Untch, A . S., & Widaman, K . F . (1997). Stress, social support, and adjustment of adolescents in middle school. Journal of Early Adolescence. 17(2). 129-151. Wexler, D . B . (1991). The adolescent self. New York : W . W . Norton & Company, Inc.. Wil lower , D . J . , E ide l l , t. L . , & Hoy , W . K . (1973). The school and pupil control ideology (Pennsylvania State Studies Monograph N o . 24). University Park, P A : Pennsylvania State University. Wintre, M . G . , & Crowley, J . M . (1994, February). A Developmental Assessment of the Perception of Parental Reciprocity Scale (POPRS) . Paper presented at the fifth biennial meeting of the Society for Research on Adolescence. Appendix A Emotional Autonomy Scale ( E A S : Steinberg & Silverberg. 1986) o < § H u < B a P 1^ CU Ol •S ^ ,08 © 2 © a os a w fc- ° fc- 5 8 ft a> Ui a) t-08 e fc- 08 « fc-08 <V e ~ DC *2 . fc-| & ^ s a a 08 -d fc- £ % & J S3 cu fc- t; OX) o 5 Q C/3 S I 5 CN 00 c -4-> >. l-l <u I c o 00 c cu S3 &, cn CN <U > bO cu l-l cu. X5 cu ty b a cu 00 D i M rt cn CN I >> B £ (SO rt 1 OH >-> a ^ XJ C e CU > CO >» rt =5 cn CN X2 T3 <U co <u . X i no •S .2 x: x: P a ~ a o o 00 <*> © S *-> p •g o ^ "5! o t^  « * lH fe <S CO S3 OH CO r-H CN cn 189 cn cn cn cn CO CO CO C N CN CN CN CN CN CN CN CN CN a c o e D O -<u T3 CU H H B Oi) o c <r t o wr no OX) CO C ci x <U 6 C o Si CO XI CU 00 e o 2 -o +3 CO CU O 4—» h H CO C rt c <u U 13 OH o 5 a o •a <o 6 3 O •8 CO OX) C <u e o CO CO S3 % <u X H cu 6 H-» 3 O •s o s I OH cu XI on C5 o rt3 1 OH S S3 OH B CO CCj CO c o • rt c ' E l o CU 6 CXj CO CU X rt CU X 3 T3 3 2 X i © O r. rt co ^, *g s 55 m > H ? * no g rt C J o <D CD a o u CD X l l CD B xx O IH CX I-l rt D I i rt X oo rt £ fa rt OV a CD X & CD •rt 1 rt X o o -4-» •8 & 3 CO CO X B CD 5 co t i rt *^ c CD I"1 'S rt O CX c >-> c o rt X o S CD ti 6 §•"8 O CU CD o «J S>0 X i—i aj ^ Ul CX >> S e CD CX « CD S o o rt >> rt • ° CD ~ B S to X 4 ) £ 5 CD O s s , o CD I-H ,<U rt 13 I-H ra CD £ s i—i o o •a CD i £ CO r H M § .S -5 6 * 2 | CN CD •a CD •a-s x CO rt § t-H rt CX >> s CO 13 CD O 1 1 rt co VH CD 3 3 O i CD t-H rt CX >> a X CO 190 Appendix B Emotional Autonomy Scale - Peers (EASP) r. 191 : W CJ s i , CM Tt- T j " T t T t T t 3 co co co co co co CS cs cs cs cs cs cs cs cs cs cs cs cs cs fl° 53 cD-fl o cu <u U l bi) co 1 •a CD I o u, CX CO CD "o co bi) C 49 CU XI CX X T3 C CD •c S -a 13 CD 13 T3 s O bfl S cu cu U l §P CO CO •a 0) IT § X s > CD X U l '53 o s CD S3 CX . •fl bi) CD o S bo o <=> 2 •>-> fl co O 3 CD i2 .2 « x; O ^3 <rt CO U l U l CD o co CD tt * 6 c o -§ CD CX " 3 CD 6 fl cB O rt > o bO co e bp o 5 c o T3 CO T3 fl CD ' e g B ca CD B rt fl o CD a rt o •8 o o CD U l CD •3 e CD S3 CO C o OH o « 3 a fl lings ver] me i o S3 lings CD CO S.SP o S me th aiow the s o CD „. CO CO CO > & — > o #. rt ^ CO are iend CO X O c CD chl rt CD C U l CD * r t >> x 2 >> u rt 3 CO •3 2 o & HJ /• 3 CD c^  CX 3^ >> O c -a - I CD ^ C bO o c 5 3 •r o a CD CD u X<2 O CD OH « CO CO 13 •r1 ,I-I > C r t CO ^ X a CO B rt & fa rt cU CO "I o 5 fl a « S3 *§ CO rt H a .2 rt 43 A S i o CO CD 2 . A°a bO a £ 1 I - 2 •Tj 2 fl •S .2 fl cfc •O fl fl rt 2 co * -S CD >-> | & rt CD C co CD co X CD -3 T3 C CD a o o •e. I—l 4 ) to S3 Xj OH CO « bo cu •S 6 « 8 CD g +-• a CD •a a U l CD c3 x CO •g CD CO CD O 1 53 T3 C 3 2 o CO T3 s ,a cs co i n oo Ov CS CO Appendix C Perceived Control at School Scale (PCSS: Adelman et a l . . 1986) 193 8 I < C/3 fc H CO Si «u #j > © c -a 53 S-d-5 vo vo vo vo vo vo • vo vo in m in m «n m m m T t T j - T t T f T t T t T t T r fa o tc If a-s „ a « a g t » .2 J= '5 cu c ^ t r ^ c o c o c n c n c o c o CU • a H-> cu u cu • cu fc C N C N C N C N C N C N C N C N o Js3 B s S H o o cu a -° &o O Co CO <U co > CU cd —1 CU i-i JS J s 0 a '8 <U S o •a ^ 5 8 s rt cu _ £ —< o <u rt <8-o - 3 ,—TM-H O co CO 6 < JD C N bO C • rH TD . 'S'-S C rt i i rt J=> cu 1 — 1 rt^ J S <U HH g 2 2 <H-H >—I CH cu — T DH O D H O rt si J= co a <W J S rt CU o a cu p a ° ai - 6 M-H HH bO C T 'rH 1 8-I f -4—* < . 2 CU JD cu rt D H rt 8 D H CU 2 § j2jt5 O CU co So < B m <U C3 l H-> > »-i cu »—i • 1 8 +-> 8 a 3H O © -rt & 8 bO — a JS a o CO H-> < , D H VO 00 194 P O (A vo vo vo vo vo vo vo vo -S3; H 5ft ti I? | cu >j 0 "Si cu cu ~ ill i n ^ S a cn cn i n cn m <n in m in Tt ^ ^ Tt- rt cn cn cn cn cn ti o fc? cu — cu 1 1 C N C S C S C S C S cs cs cs o; fc. cu > cu fc 5S ti ti o g JO CH "3 00 © ^ 1/3 i—I 00 - c X ! o a-a «-> © < s O N o •a 2 § 1 * rt CD 00 8 i-n ,8 c .. o 2 w X3 jo O "o M s *-> PH < H H CD 1 — 1 ^ >H 5 o 00 C3 © O > -a a* s £ rt J S K H H H -S J S CD o o <. o CD 00 J=l o CD +-> ^ «• rt ^ rt *-I - H C ^ H O 00 P. c . . J S & O CD " 6 < 8 cs CD • a 6 CD J S .8 CD B o 52 ° n < T3 cn op <D I 6 WH O H2 -o « S oo - -c CD .8 ' J S p o J S rt Vi JS CD B rt o © 00 P H _ C *© "3 © 2 JS rv O ^ « p 3H •—I . V " 1 f 1 H-> © < B vo 195 Appendix D Inventory of Parent and Peer Attachment ( IPPA: Armsden & Greenberg. 1987) ) o n us £3 «> V3 >> ee l l I S ! m m m m m m m m >n 196 0 & o £ J aa -ft Tt Tt Tt Tt co co co co co co co co co o cu ** • fl •ta* CU cu o Ui CU > fl-u • o cu CS CS CN CN CS CS CS CN CN P o 9 g U SO ings. 2 CO CO rt am. y feel ce: HH y feel sue Sa PH CO cd cu 6 e H-» a -*—> S3 e -4-> 0 O , CO CO rt C iffe cep (U o CH „ T-j cd parents jl my p arents. sh I ha parents >> PH HH CO Cd CU X to CO rti CN CO CU co ^ o 0) rt S -• cU 3 X OH HH Cd m cu > C 3 o x> PH Cd a "8 cu c H CU ci o PH C el CU HH M CO O <~ C 9 ^ i HH O CO c 3 .8 6 0 0 C B rtH CO CO =3 O C .8 o X co 3 O "8 rt CU CO OH 3 S X CU CO § CO OH 0 0 c CU >> s rti § oo § •a OH '-rt co rt 6 ° 1 1 « cd O H X ^ CO B ° I 0 0 " 3 CO -a I H B ON P Cfl »>. ces Si "3 CO ee ee 1J5 ISJ m i n m 1 9 7 m ti 1 OO ti " 5 o cu a s s g cn cn cn cn cn cn cn cn cn 3 cu cu ** • a +J cu fc © CN CS CS CS CS cs cs cs cs cs u cu > 8 o I CU .2 ti ti • tf w fc ti 00 s CU a a p X5 CJ 3 p o 8 CU I ex rt CO s CO OH 3 CU bO cu a p X3 • B O <u a § Xi O rt - I •S cu HH CM a ex 6 W) > .S ^ J= p 3 2 CJ ex T3 (U S S3 -o S-3 £ 8 a cu a bO T 3 3 •'—j >-. a I ex o CO B cu 3 * g ji ex a | 6 5 -5 •a •* u -5 S cu rt i_ xj s co • £ a © CU Xi ex a ^ p CU c 3 CJ a CU . X5 V-CU co +5 4-> » a cu 5 x i rt it! ex cu . CO a <-» Xi 3 3 O p •s * <2 a" a § §, a e x m e g -3 ex CU ^ a (U a P X5 CO +-> +± rt ^ a o rt • •—< xi rs *-> a bO .8 a <u rt X3 o I *H CU bJO a p o ^ cs cn m oo 198 8 (X) Si? I S ! m m m m m m m m m ca « cu S 2 T t T f T t T t T t T t T t T t T t cu © 2 C D ro co co co co co co co I m cu cu a. — CU CN CN CN CN CN CN CN CN CN SI cu CU a o H IM cu > CU a u © cu a u • 93 m CU (U tt 50 3 cc) p <0 3 o o .8 c=J XJ Cd S3 CO ' »H CU *o o 3 I T3 § •a 6 o X co >> cd •a H_» CO CU S — i o •a S ^ 50 w> - .S cu- S £ . CO 3 -o -X) cu cd cd 3 8 CX 3 1 ) X 6 cd X co fro _ CO rt 5? cu cd -a *o 5 cu 3 CO cu -§ f o co 3 OH 50 . c - r t S 50 1 — 1 c0 3 XI * S § ° cx.S >*& a co § g 3 3 O o co 50 o t-H 3 CU B CO CO T3 3 3 CO 3 O O c rt cd X ! .8 CO rt •rt 3 f l CO 3 i l i & =2 <o o S T - 8 l-H X) o CN CN CN CN CO CN Tt CN m CN CN CN OO CN 199 >n <n m m m m m m m Tt Tt cn cn cn cn cn cn cn cn cn cn cn CN CN CS CS CS CS CS CS CS CS cs HH > © C O ex P § XJ c c © -p co XJ 1 I I ° tr*. 4-> ^ 3 fe. O Vi T3 C © it-1 fe. 6 bO > XJ O C O cx-fe. 6 cs © co XJ C cn C O •a c u fe. . 1 - 1 C O •c '-3 % & Vi VH s ° J I p £ exxs . C O ft ,8 © <4M bO K C C O S 3 -H £ 3 O •s 3 C O 2 frien me. me 1 -a © c 5 bO © "S cd VH VH . © 3 iff de CO e 3 © cd C O C O XJ •o T3 c • G I—1 © u XJ ;c :c C O H - H M - H fe. "fe. *— •—1 \% \% in vd C O © 3 O fe. £ fe. £ XJ G XJ cd o C O 4-> * CJ CD £ 8 . o C O G .© O £ c © CU XJ jH 4—» C O fe.,8 . s OO OS cd XJ <3 oi 5- S? -a ^3 c 3 © co G ~ co P c 2 © fj cfc bO . G \% bO fe. £ XJ 4-» "fe !> £ • cd l-H G © XJ apa VH o © G o 13 C O •a c u :c 200 CA o I o » S H CU >H & 0 o ^ CA •» ee j * "« cu o CU g m m m «n «n i n io <n Tt ^ Tt Tt 2l l CU a CU CU > a 1 I < oo *rt I I Vi H H & O JE a [-! w m £ z &5 u Si a w a O | < ui cu £ S3 cu u CU • rt u © CU CO CO CO co co co co co co co co co co CO CN CN CN CN CN CN CN CN CN CN CN CN CN CN cd CO •8 CO g -& cu I 8 fl CO rt 3 co C •rt CO co cS fl >~> . B 3 -- 00 .S 3 5 t i * <- e S CO p CO CO 6 to -a a p B *-» P o •9 _ CO CO 3 fr CO &J>X> S c3 _g cu B co CO o c o i i •rt CO <rt rt *2 Si CN co Tt in CO • O fl co S ^ co • - 3 op C fl .2 '53 £ -° <rt i rt?"l> c CO . X co > -a > ^ C co o "C fl E? <rt CO g c5 <rt fl " r t - o & B w> * § 2 w rt >i rt CO rt e c to 3 n § 8 ° I § & * ° o CO T3 c co 6 CO c • 1—I 1 Crt >> B •a 8. <~ CO 6 § CO CO l-i O B CO cd « j rt p CO 5. „ xi § < i d - £ CO o rtfl c CO Lo S3 S CO -fl C CO • r t »_ <rt ,o ^ <rt >» CU s a ^ X rt CO •rt cd I -8 3 - ' 6 x> S O H > > 6 rt P o • a CO -o 5 • IS J 6 o 3 "° p W)8 •S ^ X CO - B i - S . >> o <u co cO cfa .g s _ T - i S ) — i X) OO ON O rt4 CN CN CN CN CO CN Tt CN m CN Appendix E Psychological Sense of School Membership ( P S S M ; Goodenow. 1993b) J5> cu o CJ Wi <n <n <n <n 202 H3 cu 8 g ee •8 a l l o cn cn cn cn cn cn cn cn cn cn CU cu 1 B ee cs cs cs cs cs cs cs cs cs cs Q ee cu ee g © o o •s CO Vi • r H X S P H 1 rt J3. • rH oa c « ' 6 o t3 T3 O O Of) § X ! =S 0> o o c a C U X ! JD SH 8 CH CU X5 P H o rt cu X) C U B f P H VH 1 xs VI CS P fe. a <u o o X ! o Vi Vi X J CO e CJ co 12 fe § O v\ CU a c • r H (U VH CU Vi X5 CU WI C . CU c •4—» c o •rH TJ X i la H—» "cs o o p •o ch H H Vi M-H fe. CO a rt H—> r-H rt 8 CO VH <U r—1 xs CO CU g CU H—» UH• r H -*-» CU Vi o •a o CO • r H xs 1 a 0 3 CU VH *S cu -g S -*-» VH O P H VH rt ° ti VH > a> rt X S cj rt i—i <u o ® § rt ° 0) p fe j2 H 8 cu a s g CJ la *o o xs o Vi Vi X S *H -4-» rt JD "OH CJ a J3 • rH f P H C CJ o c CU <U X S co VH CU xs s P P X S o co co • r H X S -*-» 4—> rt CO CU o rt JO e • rH "S j2 "o c • rH a rt g 3 VH CU X S co rt co CJ CU CD •43 P H CO a X S o 3 a CO rt 0 a rt r - H CS cn i n vo oo o 0 \ r - H CU - H <v) IS o in m 203 B U SO s £ a ~ oi D S i ee _ey cu a g cn cs cn cs cn cs cn cs cn cs cn cs CO ti ee S3 o w ti I tt r>3 cu <u XJ g -a 3 cu X J O a a & H-1 c a ,<o <u > 3' o o Xi o CO co © s a (U co >. e CU X ) 1 cu X J CU X J O CU X J H o. o o O D o cu X J g <u *"ri PH o <u CH a • rH CU i* o o X J o co O O X J O co * H O CS % .1) a s 00 C .rH c o X i 3 O XJ a rt rt i i <u X J CU 6 J3 CU X J 3 H H O VH !U X J cs cn m vo oo Appendix F Youth Self-Report (YSR: Achenbach. 1991) is available from University of Toronto Press, Inc., 5201 Dufferin Street, North York, Ontario, Canada M3H 5T8 2 0 5 Appendix G Comparison of Adolescents on Problem Behaviour Scores on the Youth Self-Report Appendix G 206 Comparison of Adolescents on Problem Behaviour Scores on the Youth Self-Report Participants Referred Boys 8 Gi r l s b Boys c G i r l s d M S D M S D M S D M S D Problem Behaviour Score Internalizing 13.59 9.30 Externalizing 15.99 7.92 Total Problem 50.69 23.57 Note. Participants = adolescents who participated in the present study; Referred = sample of adolescents with scores in the clinical range on the Youth Self-Report; Internalizing = YSRin t raw scale scores; Externalizing = YSRext raw scale scores; Total Problems = Y S R raw scale scores. A l l one-sample t tests were significant, p < .05, one-tailed. a n = 213. b n = 265. c n = 536. d n = 518. 18.31 10.45 14.65 7.93 54.86 24.91 16.1 9.90 17.3 9.60 54.9 26.60 21.5 11.10 17.7 9.50 63.4 28.00 Appendix H Teacher-Child Rating Scale (T -CRS: Hightower et a l . . 1986) is available from P M H P , Primary Mental Health Project 575 M t . Hope Avenue, Rochester, New York , 14620 208 Appendix I Correlations Between Students' and Teachers' Ratings of Problem Behaviours for Boys and Girls Appendix I 209 Correlations Between Students' and Teachers' Ratings of Problem Behaviours for Boys" and Gi r l s b Teacher-rated Problem Behaviours Variable Acting-Out Shy/Anxious Boys Girls Boys Gir ls Student-rated Problem Behaviours Internalizing -.08 .07 . 1 9 " .13* Externalizing .24*" .23*** -.06 -.02 Note. Acting-Out = Teacher-Child Rating Scale acting-out subscale scores; Shy/Anxious = Teacher-Child Rating Scale shy/anxious subscale scores; Internalizing = Youth Self-Report Internalizing raw scores; Externalizing = Youth Self-Report Externalizing raw scores. "n = 213. b n = 265. *p <.05. " p < . 01. ***p < .001. 2 1 0 Appendix J Student Recruitment Letter Appendix K Parent Permission Letter and Consent Form PARENT CONSENT FORM 2 1 4 "Promoting School Success in Adolescence" Kimberly A. Schonert-Reichl, Ph.D. Associate Professor Department of Educational Psychology and Special Education University of British Columbia 2125 Main Mall Vancouver, B. C. V6T 1Z4 Carol Anne Buote, M.Ed. Graduate Student Department of Educational Psychology and Special Education University of British Columbia ( K E E P THIS PORTION FOR Y O U R R E C O R D S ) I have read and understand the attached letter regarding the study entitled, "Promoting School Success in Adolescence". I have also kept copies of both the letter describing the study and this consent form. Yes, my son/daughter has my permission to participate. No, my son/daughter does not have my permission to participate. Parent's Signature: ', Son or Daughter's Name: Date: (DETACH HERE AND RETURN THE BOTTOM PORTION TO THE SCHOOL) I have read and understand the attached letter regarding the study entitled, "Promoting School Success in Adolescence". I have also kept copies of both the letter describing the study and this consent form. Yes, my son/daughter has my permission to participate. No, my son/daughter does not have my permission to participate. Parent's Signature: . Son or Daughter's Name: Date: Page 2 of 2 Title of Study: Researchers: Appendix L Student Consent Form 216 T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A Department of Educational Psychology and Special Education Faculty of Education 212S Main Mall Vancouver, B.C. Canada V6T 1Z4 Tel: (604) 822-8229 Fax: (604) 822-3302 Student Consent F o r m The purpose of this form is to give you the information you need in order to decide whether or not you want to participate in a research study entitled, "Promot ing S c h o o l S u c c e s s in A d o l e s c e n c e " . You may choose not to participate in this study now or at any time during the study, and there will be no penalty. If you do not choose to participate, that choice will not effect your class standing in any way. Students who do not participate will be given other work, as determined by the classroom teacher, that is related to regular classroom instruction. The purpose of this study is to investigate your opinions about yourself, your friends, your family, and your experiences in school. This study is being organized by Mrs. Buote and her advisor from the University of British Columbia, Dr. Kim Schonert-Reichl. The information collected will be for Mrs. Buote's graduate thesis. It is hoped that the results of this study will help teachers and parents better understand students and therefore be able to improve education for all. In order accomplish this purpose, you will be asked to fill out four sets of questionnaires during one class period. One set of questionnaires asks you questions about your background and feelings about yourself. A second set of questionnaires asks you about your relationships with your parents. A third set of questionnaires asks you about your relationships with your peers. Finally, the fourth set of questionnaires ask you questions about how you feel about your school and situations that sometimes happen in school. The research study is not a test. There are no right or wrong answers - just what you think. Please answer all the questions if you can. Do your best to answer truthfully and honestly. Your name will not be kept with your answers so no one but the researchers will know who answered the questions. All answers are completely confidential. No one at your school or in your community (not even your parents) will see your answers, so please answer honestly. I will be happy to answer any questions you might have before you sign or later. Please indicate that you have read this form by signing your name on the lines below. Y o u may keep a copy of this consent form for your records. Thank you for your help. Date: : : Name (please print): Signature: Appendix M Questionnaire Package Cover Page 218 u K m u CO! 3 < co CO u CO O: o Hi U co U o o ea " ° .SP 8 § • 3 7 3 1) -_ CO O • — T J r; *-* 2 * to fe> ca > CO > to w • —' CO 3 to _ to T J !- C _co ca T 3 —' JS to 3 eO X ^ o H •25 tD U ca co to 3 a Q. — <u X ! (U o p CO o to CD ca a to <u co to a <o -o c <u ca •< ii CO o tO o •a ca 3 ° •a 5 3 ea - — 4—> o 2 p to C to 43 X5 |_ ~ 3 C O — >-» CO -.9- P O .JJ ' •2 ' L-i U. <& ca ,. >> ca o. CO *5 to CO to P £ <a O CL, p .y o P a , o 55 O PQ 60 - >> 'as T J 5 <D P « to ca CO CO eo ca 3 ^ >\ tU ca c CO ca 3 O c _ to ca co ••5 ca o c/i ca ca ca - - U o >•> rW g T3 U o • S -2 2 to ~ o _ v- 3 3 O o x> ca ca •"8 o a . T3 P CO rt •g >>"o l a « > 60 eu io 3 XJ P 2 W ca fj g xi c u H .2 3 o a e 5 cr 5 • : « > a j-j «. CO w CO O e XS co co to CO « s E o w w o 3 J2 is o 2. « ca C G B! O CO ~< co - g u ^ o 5 C r—? CJ ca o co 3 jo cu o ca < cj Q . = O to CU ? V] S CO CO Z * u 3 O CO *-• O 5 — 5 i- cu ^ c u to O C cu p > » «u > « = E «* c <! *! ss . w * to 3 o a >» co O T3 S I d 6 co CO CU co c: »-3 O « C J : a co ca ca ~ •£ c ca e Ii 0 ca 1 -3 ^ O co > . CO to ca rt .2 . to e a o CO o - S CO o to tU _co "Q. s o CJ Er « E3 ca ca J= O O ca g 2 - 6 ^ - T3 « S -s 5 o to % « P = •2"gj o « , a a » "^ •S a CO ca a -2 -c ca Q J2 Z 6 * I J* t o CO u > 3 C cr co o c eo cu JD CU o >> u cS 3 O >> c « H o w ^ - . ^ 2 o ca _ ^ «*-cu 3 Q • o 0 / 5 >> s o •-CO ta 52 . 3 co •<£ u o p _o ta o 3 -o W M _ ca - « 3 5 CM CJ O 'CO O '-3 5 « m CO O M-4 n > o o cu fj a oo ca J2 . ' o CO < o > co * 3 . to P Appendix, N Student Identification Form 221 ID# PROMOTING SCHOOL SUCCESS IN ADOLESCENCE THIS SHEET ALLOWS US TO KEEP YOUR NAME APART FROM YOUR ANSWERS (SO NO ONE WILL KNOW WHAT YOU SAY), AND THEN TO PUT YOUR QUESTIONNAIRES TOGETHER WHEN YOU ARE FINISHED. Please put your name on this sheet. NAME (please print clearly) What grade are you in? (CIRCLE ONE) 7 8 9 10 11 12 Appendix 0 Intercorrelations Among Measures for Grades 8. 9 and 11 223 CN O tN v© fO r o o r o o tN tN O tN tN —< r- 1-1 t o o tN —H 00 o m tN C N ro o to ae to o 05 CO CO © 00 tN CO i—l i - l C N to to i«o Tt m in - 4 to t- " O rt tN rt rt o O rt o r o O O N tS ^ CO tO Tt r— c n O rt to ve tN O O N o O 00 o r o tN r o o NO o rt o o o T t •© ON o tN o o o ts tN O 00 o tN o 00 tN »S O N rt i — i ve rt tN O rt r o tN O ON rt O O N o r- o © rt r o © r o rt o rt © © tN © T t © tN rt tN tN Tt tN t o O N ve o Tt rt tN r o © 00 00 rt rt ts IT) 00 rt to in rt r-oo to 00 ve </-> t N 00 © rt tN 00 to to to r-tO rt CN © © tN t N m 00 rt ve Tt ID Tt Tt ve. ON to 00 CN r> © IT) to Tt C N rt Tt fO to m ve vt r~-CO rt rt r~ oo rt -r r- r- to Tt Vl to Vl t N O N CN CN r- rt _< © rt 0 r o © oo oo tN Vl to ON Tt ve to ve oo ON T t tN rt t N © © Vt © 00 vo i—I rt • © i—l <N r - rt rt fO •o CN 00 C N © tN vt CN t-CN © to r- rt ID to ve Vi tN to ON Tt ve Tt O IT) © Vl r-©\ CO ON Vl ve tN 00 Vl Tt ON Vl ON to to Tt CN ON to ve to © O N tO rt T t tN to ve ON to vt tN to Vi tN Vl 00 rt tN Vi VO t N NO to © tN CN © 00 © VO rt i — i tN to CN Vi 00 Vi rt to r -CN Tt Vi VO © C N © tN © Vt tN CN r o © T t © VO V) © rt CN © O N © Vt © T t © 00 rt © 00 © rt tO CN ve oo o m ve r o © 00 Vi 00 © Vl to t o 00 CN r o © © r-ts to r--CN © NO 00 CN r-tN Vl VI Vt tN to to V O Vi VI © Vt tN NO © © T t CN r - <o © rt tN rt 00 CN Vt tN r o © 9 in to ON CN NO CN vo Tt r-Vl rt tO Tt V O 00 VO to to © O N to ON m m oo r-to ON v© r-oc oo ve Vi ON ve oo r o tN © r - T t r o © rt rt rt i n © oo rt **> rt r o © T t © CN © tN © Vt CN 00 CN rt " O tN NO f N 9 to NO CN © OS © r o tN tN to 00 CN 9 to Tt 9 in oo CN 00 fN © VO CN tN t N © in ve to fN ve © © rt tO rt tO Tt to fN r o CN rt Tt rt 00 to to fN r o tO rt O N CN 00 Vl vo 0 \ rt to to Vi O N r--oo ve Tt CO Vi Vi Tt ve S 0 0 Vl rt O N © ON r*- r o CN rt „ 00 © vt r -rt © ON © r o tN © rt Tt „ rt Tt . Tt rt r o CN r o oo oo io r o NO VI rt rt rt rt rt rt CO IN (O 00 Vl 00 m 00 © © CN T t © T t CN m 00 00 r o to to 1 CN l © a. E Vl m in ON to •—i NO fN rt 00 ve CN Tt NO Tt rt in to VO to Vt fN Vt © 00 V O V O to 00 Tt Vi Tt to r o CN O N fN tN r o © rt 00 T t CN rt rt T t r o © CN rt CN rt 00 © Tt to O N Tt to Tt 00 Tt to in 9 to s rt CN rt fN 00 CN © rt rt 9 to co ii 3 C • rt rt c o o 1 Si CO Si CO Si < Q. I— o < < CL Q_ CO Si Q_ CO Si Si Si CL Q> — B 5 o — y < Q. CL-I O CO CO CO O 0. CO CO a. 8 CO < Q. O —• — —• — CM cc a: CO CO — CM 2 2 5 CN O CS V O CO t n t n Ov tn Ov © © V O I-H © ts © V © tn ts VO tN Vi © tN X T t 9 T t Ov IT) IT) VO 9 rt' x ve ts Vi Vl T t ts X x tn rt 9 rt in ve vo 9 in tN in >n • © cc © © 00 t S - H T t CS VO © tN tn ts tn T t T t © cs m tn m tn 9 m tn tn tN -H V© tN © — © © tn cv tn © ts VO m tN tn tn r- r o ts —< — - i vo © —( CS CS T t r-ts —< 00 © VO © T t t N ON i / - ) t N © rt rt OV © T t © CO ts rt O —< t S ve ts © ts X ts tN tn 9 tn rt x Vi T t tN tN ve ov tN T t T t X © Vi T t 00 CO VO OV T t rt © rt rt rt rt _| © CS ov m T t ts VO tn CM tn rt in tn tn x ts T t CS © © r--ts r o r o tn Vi tn ts m tn tn ts VO T t r-tn ts VO tN VO r> rt c s © ts ts m m in T t 9 T t m v© X x r-t S rt tn V© tn tn t o © Ov © X tN r- x tN tN tN tN ts Tt Vl ts in ts ov tn tn x T t r-ts © VO ts tn tn Ov t s rt X © rt V O © t— i n tn rt T t CO tn rt r o c s Vi ts ts m tn m V O © r- m rt tN rt V© m T t ov m T t T t rt tN CS rt i ' r v© T t rt IS in in in tn 9 X v© tn cs in ts rt 00 rt m rt r~ v© ts 9 tn in ts r o c s © c s T t tN T t tn v© T t tn Vl T t © V © o\ tn OA rt m r> tn T t X X ts tn rt ve x tn rt ov T t CN tn X © x T t T t tn Ov m ts VI © © r o © CN rt © © tO rt 00 tN © rt tn tn vo © 00 © © T t © V O rt © rt © V O © CN © V, tN © —( T t ve T t cs Vl cs r- CN r-rtI CN rt W-l 00 T t rt o ^ ^ <s r o ts m m m rt ts r- T t 9 T t Vl tn © cs ve tn rt VO tn rt ts T t 9 ve ve v© x ve x x 9 Ov tn r-c\ tn m Vi Ov T t VO VO X VO T t r-CN CN X T t Ov Ov T t r-m 9 T t Vi Vl Ov vs Vl VC r o r o rt tN CN O v © tN T t CS x tn Vl © Ov © VO 00 © fN fN CN VO CS ts tn f S T t rt fN 9 rt r- rt fN oo r o rt CN rt rt o r-© 00 © T t © Vl ts in tn V ) © © tN rt Ov X V© r o X rt © t s OV OV T t X rt C> ts T t V © tn x T t m m tn Ov T t r o X CS rt CS ts r o X © x m rt o © © VO © f S © I—I . rt T t " ° tn rt tn ts V O © © - H T t © VO cs Ov © VO © T t V) rt © © t s OV © rt T t r- vi © —I rt rt >/-> ts T t © tn tn © ro vo © X CN r o fN rt X tN r o vo ^-1 ^ ^ ^ ^ ^ V © tn CN m © T t rt © \ rt © T t tn x tn ts rt Vi tn © © Ov T t V, f S © t o V O 9 rt T t m rt CS rt Ov rt C N " © tN © m ts rt © O r o T t ts vi r-m rt V O rt © ts X rt tn © VI © r o © f S tn 9 m r-© x ts T t tn 2 3 vo © © CN CO U 3 _C +3 C o c j CU 1 CO Si CO Si CO Si 2 2 E o o 2 a. co Si CO Si CO Si CO Si I .2 co <1 < 8 < cc cc CL CO CO CD >- >-2 2 7 cn o CD CO TH1 • r H o T 3 co >, O PQ CD s o CO C o B C O c .2 1 o 6 o C N NO CO CO o CN T t O 1/1 CN CN o o CN o CO o NO </•> CN - H O r H VO If) X r H ON ON CN 00 Tt T t i n i n i n ON - H CN © - H T t OS O r H r H CN r o © <N CO CN if) CN ON If) r H if) co T t CO ve if) CO CN CN if) CN CN ON CO 0 0 CN o ve CO O o © © VO T t CN r H ve ro ON © CN VO CO r H ON © 00 T t 00 ve C N © CO C N © © r H r H r-CN 0 0 © CO T f T t T t r H r H © r H © T t © VO r H © CN NO © CN ve en ro T t © r o NO r H © T t CO CN r H cn If) T t f N T t t " ~ CO T t ^ 0 0 © 00 r H wi ve if) CO CO If) X T t ON CO r H © T t CN ON cn X if) T t IT) 00 If) r H If) © T t © CN T t If) T t 0 0 C N ON T t T t © NO CO r H CN T t r H r H 0 0 0 © C-- 0 0 * H NO r H r H r H r H r H © © © T t <f) NO CN C N r H © If) r-© r o CN r-T t If) T t ON T t 00 ON NO r-ON CO © © r -CO r H © © t - r H © co CO CO NO CO ve CO © . CN CN NO © t> ON CN ^ 0 0 © ve CO o If) 0 0 CN T t CO r H IT; NO ON CN r- r H 0 0 © r H © CN NO © r H 0 0 © T t © If) CO o If) co <o CN r H If) If) If) 00 CN CN T t VO CO © © CN T f CN CN <n <n C N C N r H r H CO CO ON r- — r- r H T t r H © H r i' i' 0 0 CN T t CN ON CN r H CN CN NO T t CO T t r H 0 0 t " - r H r H © r H CN © ON © r H t - ~ ve C N 00 ve ON © T t If) © C N T t co ve r H CO T t If) T t © CO NO © r H CN r H r H CO NO © r-© oo © «/-> © T t CO T t CN ON - H - H O l r o ON >n CN CN r H r o r~-r H © T t © CN © © CN r- co ve if) 00 r-CN if) CN T t ON T t © T t CN NO CO CN r H 0 0 CO r H T t © 0 0 CN 00 h 08 If) T t CO 00 If) CO 00 © r- if) © r o CN r H r H CN T t CN If) CO CN CO CN « N CO If) CO T t NO C - r H r H © o If) If) If) T t If) ve ve if) ve o ON T t ON ON r-cs ON f N VO r- ve r H CN ON OS r o T t vo CN CN © ON ro oo © — O NO ON co VO CN r o CN ro o CO 0 0 CO CO CO If) CO 00 CO T t T t ON CN © CN 00 CO T t If) r- oo CO - H ve T t CO ve NO T t © NO © ON ON ON 00 If) If) 00 00 r- T t CO T t T t f N oo r-~ CO ve if) if) x r-if) T t CO x T t If) T t o If) T t If) T t If) r o CN CN T t if) r H fN CN CN VO ON ^ r~i <^ CN CN r n ON r o r H © r-CN CO T t NO CN r H C N © © ON r H - H CN CN © NO © ON © V© v« If) ve o co if) if) if) © © CN CO 00 ON ON r H fN r H CN CN T t CN NO T t r H r H fN © ve ve CO T t C N r H rs © T t CO T t CN © NO © ve if) oo CN © 2 r H CO CN © If) T t CO CO CN co T t CN T t vo o CO If) T t © CO VO © X CN X CN CO CO CO CO 2 55 55 Si r - CM CO 2 2 B 5 o = o. a. r» oo TJ C= .5 a_ a. a. a. J/j CO CO CO jjj 55 2 55 Oi T— T— *— .— E O '— Q. Q. Q . E o o o a. 2 r - T- T- - r - CM ON CN Tt ro C N NO IT) r H T t r H ve If) ON - H r H © T t © CN © T t r H CN © ON © © r H • n © o T t © ON CO r H CN CO CO X C N NO CO OL OC CO CO >- > CO CD 3 C c o o CD § rt rt & co 3 cd I o U 13 3 o —) •3 > o •s OA co 3 o 1 o o i3 s o X . 4—> o 0> > > O x co 3 o 1 o U 4—» o CU Ii in q cu X rt o is a 0 0 s 0 0 x •a co CN co 3 o 1 o o o I-l o 2 X 1 I-l CU cm CU X ! O IS • f—I 3 6 0 o PH CU O q cu xj CJ i g '8 •l-H XJ X —» 0 0 cn C O a o •X3 rt 1 O CJ CU X ! IS 4 — cc) o Ig '2 6 0 • i—1 C/5 I t-i CU JS op Js —) o C O C O c O • i-H o o CU _ X i rt CJ •I-H C OA • — i co CU X ! 6 0 C N C O co a o 1 o o CU I m q cu X ! O rt o i g "S 0 0 s _ X ! op 'Js —I T t C N co 3 o 1 o o CU •a 0 0 < W i f 13 C J C O < OH PH 228 0H CU 13 o co X 3 co 3 o i -c2 _ cu 73 CJ >> 2 o s 2 3 < 13 c > o G W < W *a3 c5 o q cu X ! rt o i g '8 .2? 'co l-H CU X ! op Js —I oo C O <4-l o co C O • i—i 4—* al 13 fc o o •a c 0 0 < PH —I C • i-H 0 0 < W <U 13 o C O X 3 co >N O S —I a CU a, cu —i s o 0 0 < PH C o a Vi < PH cu 13 o C O X 3 co C o • .-H a •i-H 1 "2 "<3 Q oo < PH S •i-H OH CU 13 o C O X 3 co 3 o » O •i-H 3 3 2 2 o u I 2 o U < OH CU 13 o co X 3 co 3 c II s OH b" 13 o C O 3 CU 2 X o <-rt rt 3 o rt N 1 — J • l-H cU Q OH 0 0 < PH CU —» OH 0 0 < PH i f 13 o C O C O I-l _ l l c2 _CU 13 O 0 0 2 o c o 4-> 3 13 3 " r t o 2 PH OH 0 0 < PH b~ 13 o C O X 3 C O Ci o § 8 CX < OH OH 8 OH if 13 C J C O X 3 co Ci _o *H—» rt 3 —t •a 3 OH 0 0 < PH —) 3 •«-H OH 0 0 < P4 o _cu 3 3 § 13 cx C X rt o C O X 3 co O Ci CU —I c CU C X cu -a c o OH 0 0 < PH 3 O 3 OH 0 0 < PJ af 13 o C O X 3 co a _o *a rt § i - i C X § •fH < I-I OH af 13 o co X 3 co 3 O *4-» rt o "2 3 2 e o U I-H S3 C X 2 o U rt c j C O X 3 C O rt C O 3 C O 2 H OH j u " 13 o co 3 0) 2 x o XS rt C X 2 o U X c j OO X co a> x 2 a> 2 c o o X o 0 0 <4-4 o 0) co 3 0) 0 0 13 o • —4 0 0 J D "o X o co OH 0 0 oo OH rt o 0 0 13 o X o 0 0 i_ ^ 8 "o cx c II § u a> C J >-i OH 0 0 0 0 U OH of 13 o co X 3 5 04 oo •< of 13 o C O X 3 C O 6 0 3 • i-H N • i-H 13 I c "8 « o CX a) OH 0) 0 0 3 O 0< 0 0 of 0 0 rt l-H CU o C X -a 0 0 < OH a co" <U 'o c cU C X 2 o c j o X o C O _ - 4 — » e i a> x o u 13 o C O X 3 C O 0 0 3 • rt N • rt 13 I P1 t-4 o C X CU OH a) oo 3 O C3I m 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0053902/manifest

Comment

Related Items