Parental Aggression and Adolescent Peer Relationships

Although child maltreatment studies have demonstrated a link between parental aggression and peer relationship functioning in children, few studies have investigated this relationship in adolescents. The current investigation was approached with two main goals. The first goal of this study was to investigate the relationship between parental aggression and adolescent peer relationship intimacy and social skill competence. The second goal was to investigate trust, self-efficacy and attitudes about aggression as potential mediators in the relationship between parental aggression and peer relationship functioning. This study used adolescent reports of their parents’ verbal and physical aggression to predict peer reports of the target adolescent’s peer relationship intimacy and social skill competence. This study examined data for 121 adolescents ages 10-14 at Time 1 and 16-22 at Time 2. Mothers’ verbal aggression predicted lower peer relationship intimacy in mid-adolescence and late adolescence and also predicted lower social skill competence in late adolescence. Fathers’ physical aggression predicted lower peer relationship intimacy in late adolescence. Trust, self-efficacy and attitudes about aggression were not found to mediate the relationship between parental aggression and peer relationship quality. Future research regarding the relationship between parental aggression and peer relationships should include investigation of the context within which parental aggression occurs. Parental Aggression and Peer Relationships 4 Parental Aggression and Adolescent Peer Relationships Adolescence is marked by an increased orientation toward peers, with the task of this stage focused on the development of supportive relationships marked by intimacy, loyalty and trustworthiness (Berndt, et al., 1986; Buhrmester & Furman, 1987; Hartup, 1983; Youniss & Smollar, 1985). Peer relationship quality at all ages has been found to be associated with both current and future adjustment outcomes (Garmezy & Streitman, 1974; Cowen, Pederson, Babigian, Izzo and Trost, 1973), and there is evidence to suggest that child and adolescent peer relationships may provide contributions to social and emotional development that is different from that provided by parents (Price, 1996). Although recognized as an important aspect of development, little is known about the influence of parenting practices on child and adolescent peer relationships. The focus on the development of supportive relationships in adolescence makes this stage particularly salient in the investigation of the relationship between parenting and peer relationship quality. One potential major influence on social competence with peers is parental aggression. Understanding the specific areas of social competence associated with parental aggression and the process by which parental aggression may influence social functioning, advances our understanding of social development through the exploration of the potential negative consequences of parental aggression. In addition, this area of research informs clinical practice with regard to intervention strategies for children with poor peer relations who have experienced aggression from a parent. The theoretical and clinical ramifications of understanding the process by which parental aggression influences peer relationships are extensive both in informing theory and clinical practice. Empirical research in the child maltreatment literature has provided the Parental Aggression and Peer Relationships 5 most extensive investigation of the relationship between parental aggression and peer relationship development. Studies of maltreated children have investigated their peer relationships in an effort to describe the influence of parental aggression at the extreme negative end of the continuum of parenting. Understanding differences between maltreated and nonmaltreated children’s peer relationships, provides information regarding the influence parental aggression may have on social functioning, as well as offering clues regarding potential mediating factors present in the relationship between parent-child interactions and children’s peer relationships (Cicchetti, 1984). This “experiment in nature” (Bronfenbrenner, 1979) enables researchers to investigate the impact of severe family relationship circumstances on peer relationships in order to inform theories of socioemotional development. To date, there are few studies investigating the impact of parental aggression on adolescent peer relationships. A review of the literature on maltreated children’s peer relationships provides information regarding the specific areas of social dysfunction associated with parental aggression that may also be important in investigating adolescent peer relationships. However, research on child maltreatment has historically been wrought with methodological problems and constraints. One problem is the use of parent report measures regarding maltreated children. Studies have indicated that abusive and neglectful parents have unrealistic expectations and distorted perceptions of child behavior (Azar, Robinson, Hekimian & Twentyman, 1984; Bauer & Twentyman, 1985; Rosenberg & Reppucci, 1983). Another consideration in this area of research is the co-occurrence of risk factors in maltreating families, such as lower socioeconomic status. When comparing maltreated and non-maltreated children, it is important to control for additional risk factors and demographic variables in order to Parental Aggression and Peer Relationships 6 increase the likelihood of finding differences related to maltreatment. Keeping these methodological issues in mind, the following review of research on maltreatment outcomes includes studies designed with well-matched samples of maltreated and non-maltreated children as well as those using measures of child behavior other than parent report. Understanding the impact of parental aggression on maltreated children’s peer relationships provides the basic foundation for investigating the impact of parental aggression in adolescent peer relationships. Child Maltreatment and Peer Relationships Observational Studies Studies using systematic observational coding systems have examined young maltreated children’s peer relationships and have found physically abused children to exhibit more disturbed peer relationships and interactions than non-maltreated children (Cicchetti, Lynch, Shonk, Manly, 1992; Conaway & Hansen, 1989; Mueller & Silverman, 1989 for reviews). The main findings reveal maltreated toddlers and preschoolers to be more aggressive (Fagot, Hagan, Youngblade & Potter, 1989; George & Main, 1979; Haskett & Kistner, 1991; Herrenkohl & Herrenkohl, 1981; Hoffman-Plotkin & Twentyman, 1984) more withdrawn and avoidant (Fagot, et. al., 1989; George & Main, 1979; Hoffman-Plotkin & Twentyman, 1984; Jacobson & Straker, 1982) and less prosocial and competent (Alessandri, 1991; Fagot, et. al., 1989; Hoffman-Plotkin & Twentyman, 1984; Howes & Espinosa, 1985; Lewis & Schaeffer, 1981) in their interactions with peers than non-maltreated children. In addition to the methodological strength of examining actual behavior, observational studies of maltreated children have the benefit of allowing for fine-grained analyses of specific responses of maltreated children to specific behaviors of their peers. Physically abused children Parental Aggression and Peer Relationships 7 have been found to have higher negative reactions and lower positive reactions to behaviors in general (Fagot, et. al., 1989) and have been found to display higher levels of aggressive responses to peer aggression, peer distress and friendly overtures by peers. Physically abused and neglected toddler and preschool age children have been found to respond to aggression from other children with higher levels of aggressive behavior compared to non-maltreated children (Fagot, et. al., 1989; Howes & Eldredge, 1985). Maltreated toddler and preschool age children have also been found to demonstrate aggressive and withdrawn behavior in response to peer distress, while their non-maltreated counterparts were noted to exhibit prosocial responses to peer distress (George & Main, 1979; Howes & Eldredge, 1985; Klimes-Dougan & Kistner, 1990). Taken together, these descriptive studies of maltreated toddler and preschool children’s peer relationships indicate that young children who have experienced physical maltreatment demonstrate significant maladaptive interactions with and responses to the negative, neutral and affiliative behaviors of their peers. These interaction patterns of maltreated children suggest initial difficulties with basic social skills that are likely to leave them vulnerable to less intimate and less supportive relationships with peers. Maladaptive responses to peers of maltreated children put them at risk for peer rejection. The basic aggressive response of maltreated children to their peers, especially in response to affiliative and friendly overtures, leaves these children vulnerable to less closeness in their peer relationships. Although the above mentioned studies do not examine the quality of peer relationships of physically abused and harshly disciplined children, these observational studies indicate that children receiving parental Parental Aggression and Peer Relationships 8 aggression are at risk for difficulty in establishing close relationships with peers. Teacher Report Studies Studies investigating adult observations of maltreated children’s peer relationships suggest that maltreated children differ from non-maltreated children in their social competence. Cicchetti, Lynch, Shonk & Manly (1992) report findings from several of their studies in which they found camp counselor and teacher ratings of social competence to differentiate between maltreated children and non-maltreated children between the ages 5-12. Maltreated children were found to demonstrate poorer peer relations measured by camp counselor and teacher ratings of social competence using a Q-sort procedure. Maltreated children, which included physical abuse for 39% of the sample, matched with non-maltreated children on race, education, socioeconomic status, and household composition demonstrated lower self esteem, lower prosocial ratings and higher withdrawn scores than non-maltreated children as assessed by camp counselors blind to maltreatment status (Kaufman & Cicchetti, 1989). The authors suggest that maltreated children’s impaired self-esteem may lead to difficulties with initiating relationships with peers. These findings suggest that adults working closely with children experiencing parental aggression view their interactions with peers as maladaptive compared to non-maltreated children. This maladaptive social interaction has negative implications for the establishment of close, supportive, intimate peer relationships. Peer Report Studies Cicchetti, et. al., (1992) suggest that peers may be able to identify qualities about their interactions with maltreated children that are unique from those provided by adult ratings. Research based on peer sociometric ratings and peer ratings of behavior, find maltreated Parental Aggression and Peer Relationships 9 children to be rated less favorably than non-maltreated children in a variety of areas. Studies using peer sociometric ratings have found physically abused children to be rated less favorably than nonabused children with whom they have been individually matched on gender, age and race (Haskett & Kistner, 1991; Kaufman & Cicchetti, 1989; Manly & Cicchetti, 1989; Salzinger, Feldman, Hammer & Rosario, 1993). When reciprocal nominations were used as a measure of friendship or social support, maltreated children received less positive nominations and more negative nominations from those they nominated as best friends (Salzinger, et. al, 1993). This suggests that maltreated children’s perceptions of their relationships with other children may be somewhat distorted. Peer sociometric rating studies suggest that physically abused children are vulnerable to peer rejection, lower support and other negative outcomes resulting from the unfulfilled expectations associated with their inaccurate perceptions of others as best friends. Peer ratings of maltreated children’s behavior are found to be less positive than ratings of non-maltreated children. Peers rate maltreated children as less likely to be leaders and more likely to start fights (as reported in Cicchetti, et. al., 1992; Salzinger, et al., 1993) suggesting deficits in assertiveness as well as control of aggression. Physically abused children were also rated lower on sharing and higher on meanness and attention seeking (Salzinger, et al, 1993). These studies suggest that physically abused children are not only perceived by their peers with an increased propensity toward aggression, but also as demonstrating more negative intentions. Additionally, these studies suggest that peers perceive physically abused children as less likely to exhibit prosocial behavior. Similar to teachers and researchers, physically abused children’s peers view them more negatively and less positively than they view non-abused children. Parental Aggression and Peer Relationships 10 The child maltreatment literature suggests that maltreated children demonstrate poorer social skills. Specifically, these children are perceived as more aggressive and less prosocial by researchers, teachers and their own peers. Deficits in maltreated children’s prosocial behaviors and maladaptive responses to peer affiliative overtures, likely result in lower positive interactions with peers. This literature as a whole suggests that maltreated children are at risk for more conflictual, less intimate, and less socially supportive peer relationships. Studies examining parental aggression that does not reach the level of abuse have found similar findings to the child maltreatment research. An observational study found children receiving harsh punishment to exhibit aggressive behavior toward peers. After controlling for socioeconomic status, Strassburg, Dodge, Pettit & Bates (1994) found parental use of physical punishment of kindergarten children, assessed by a modified version of the Conflict Tactics Scale, to be positively correlated with observations of children’s subsequent aggressive behavior toward peers. Peers rated children receiving harsh punishment as aggressive twice as often as they did control children (Dodge, Bates, & Pettit, 1990). Mediating factors in the relationship between parental aggression and peer relationships Although the presence of an association between parental aggression and peer relationship difficulties in maltreated children has been well documented and described, few studies actually examine mediating models of this relationship. Theories of socio-emotional development provide suggestions for mediating factors of the relationship between parental aggression and maladaptive social behavior. Attachment theory and cognitive social learning theories suggest possible mediating influences of parental aggression and suggest similar areas of investigation in this relationship. The behaviors of children experiencing parental aggression also Parental Aggression and Peer Relationships 11 suggest that potential mediators of aggressive, less assertive, and less prosocial behavior would be important in the relationship between parental aggression and peer relationships. Cognitive social learning theories suggest that direct and indirect parental influences affect children’s expectations regarding interactions and relationships. These theories suggest that parent’s direct attempts to teach and advise children regarding relationships combined with indirect influences such as the quality of the parent child relationship, disciplinary styles, and modeling result in the child’s development of social cognitions of self, others and relationships. These mental representations or schemas influence the child’s social information processing, which influences the child’s perception of other’s intentions and the child’s choices regarding responses. Research on parenting and discipline styles have suggested cognitive factors as important mediators of the relationship between parenting and peer relationships. In one study, interpersonal problem solving skills appeared to mediate the relationship between parents’ disciplinary styles and peer acceptance in children (Pettit, Dodge & Brown, 1988). In another of their studies Pettit, Harris, Bates & Dodge (1991) found social cognitions to mediate the relationship between coercive and intrusive family interactions and children’s social competence. Hart, Ladd & Burleson (1990) found outcome expectations to mediate the relationship between parenting style and peer acceptance in kindergarten and fourth grade children so that children of power-assertive mothers demonstrated instrumentally focused outcome expectations and tended to be less accepted by peers. Social information processing has also been suggested to mediate the relationship between parental physical discipline and children’s peer relationships. Studies have demonstrated harsh discipline (Weiss, Dodge, et al, 1992) and abuse (Dodge et. Parental Aggression and Peer Relationships 12 al., 1990) to influence social information processing resulting in these children demonstrating hypersensitivity to hostile and aggressive aspects of interpersonal situations. These studies suggest that parenting behaviors may influence peer relationships through children’s cognitions, perceptions and expectations regarding interpersonal relationships. Viewing relationship expectations as mediators of the relationship between parenting and peer relationships is in keeping with the assumptions of attachment theory that postulates that children develop “internal working models” of self, others and relationships from interactions with parents. These internal working models generalize to expectations of other relationships (Bowlby, 1969/1982). Consistent care giving is thought to influence the development of positive working models of self (i.e. potent, competent and worthy) and others (i.e.trustworthy), while inconsistent care giving is thought to influence the development of negative working models of self (i.e. impotent, incompetent and unworthy) and others (i.e. untrustworthy) (Ainsworth, Blehar, Waters, & Wall, 1978). Insecure attachment classifications, assessed from patterns of behavior of a child observed throughout a series of separations and reunions with a parent, have been associated with maltreatment in a number of studies. The strong association between child maltreatment and insecure attachment styles (Cicchetti & Braunwald, 1984; Egeland & Sroufe, 1981a, 1981b; Lyons-Ruth, Connell, Zoll & Stahl, 1987), evidence linking attachment styles with peer relationship quality (Lieberman, 1977; Waters, Wippman, & Sroufe, 1979) and the theorized importance of internal working models in the development of attachment styles, suggest parental aggression may influence children’s internal working models of themselves, others and relationships that may, in turn, influence the children’s expectations regarding peer relationships. Parental Aggression and Peer Relationships 13 Attachment theory suggests that parental aggression may influence internal working models of self as incompetent or impotent, others as untrustworthy and relationships as aggressive indicating that these areas may be important mediators of the relationship between parental aggression and the children’s subsequent peer relationships. One study that suggests parental aggression influences children’s expectations of interpersonal relationships found that maltreated children told fewer stories involving kind acts of reciprocity than non-maltreated children. Additionally, physical aggression was the most common form of peer unkindness in maltreated children's stories (Dean, Malik, Richards & Stringer, 1986). This study suggests maltreated children may not expect their own kind acts to be reciprocated and may expect physical aggression to occur in relationships. This further suggests that parental aggression may influence children to view peer interactions as less reciprocal resulting in those experiencing parental aggression having different expectations regarding interpersonal relationships. Although there has been some support described above for relationship expectations to mediate the relationship between parental aggression and peer relationship competence, this support has only been found in studies with children. It is expected, however, that perceptions and expectations may also mediate this relationship for adolescents. In fact, it has been suggested that adolescents’ expectations and values, may play an important role in adolescents’ models of themselves in social interactions. These models, may in turn, mediate the relationship between problematic family relationships in childhood and maladaptive behavior in adolescence (Allen, Leadbeater & Aber, 1990). Parental Aggression and Peer Relationships 14 Parental Aggression, Relationship Expectations and Peer Relationships The power assertive and controlling nature of parental aggression may influence adolescents to be less trusting of others, exhibit lower self-efficacy and endorse attitudes supporting aggression. There is empirical evidence which suggests trust, self-efficacy and attitudes about aggression may be important in the relationship between parental aggression and children’s and adolescents’ peer relationships. Empirical investigations demonstrate relationships between each of these relationship expectation variables and parental aggression. Additionally, research has demonstrated relationship expectations to be associated with peer relationships suggesting a potential mediating influence of relationship expectations. Trust. The attitude or expectation that others are trustworthy, fair, and helpful may be important in an adolescent’s willingness or motivation to develop intimate or close friendships with others. Expectations of others as untrustworthy may result in an adolescent’s avoidance of confiding in their peers resulting in a lack of mutual feelings of intimacy. There is some evidence that trust may be related to pre-adolescents’ acceptance by peers. One study of 9-12 year old children found those who rated care givers as trustworthy, responsive and physically and emotionally available received higher ratings of acceptance and fondness by peers (as discussed in Kerns, 1996). This study suggests that for pre-adolescents, viewing caregivers as trustworthy may be an important component in the development of positive peer relationships. Studies of maltreated children indicate that parental aggression may influence the level of trust children have in others. A study comparing maltreated, non-maltreated children receiving public assistance, and non-maltreated middle class children found maltreated children to score significantly lower than either of the other two groups on a factor measuring readiness to explore Parental Aggression and Peer Relationships 15 and learn in the presence of novel adults. (Aber & Allen, 1987). The authors state that lower scores on this factor suggest that maltreatment disrupts the balance between children establishing safe, secure relationships and feeling free to explore. Differences in trust of others between maltreated children and non-maltreated children have also been demonstrated in response to hypothetical vignettes regarding keeping promises and secrets. Maltreated children were found to respond with no trust or a great deal of trust, while non-maltreated children exhibited more variability in their responses (Bernath, Feshbach & Gralinski, 1993). Another projective measure of trust revealed maltreated children to demonstrate less trust than nonmaltreated children in a sample of children aged 5-12 matched on birth order, family structure, residence, and socioeconomic status (Kinard, 1980). The above studies suggest a relationship between parental aggression and a child’s sense of trust. Additionally, trust is suggested to be associated with acceptance by peers. Taken together, these studies suggest that trust may mediate the relationship between parental aggression and children’s peer relationship. Self-efficacy. The expectation that one can successfully perform competent behaviors to produce desired outcomes, influences the amount of effort and motivation one will exert to achieve a desired outcome (Bandura, 1977). In adolescents, the relationship of self-efficacy and close friendship has been demonstrated in studies that found low self-efficacy expectations of socially competent behaviors to be associated with low prosocial and higher maladaptive strategies in social situations (Allen, Leadbeater, & Aber, 1994; Allen, Leadbeater, & Aber, 1990). The construct of “locus of control” (Rotter, 1954) relates to perceptions of the self as more competent (internal locus of control) or less competent (external locus of control) in affecting outcomes, and has been associated with parental disciplinary styles. Parental hostility Parental Aggression and Peer Relationships 16 and less inductive discipline styles have been associated with external locus of control in children and adolescents (Davis & Phares, 1969; Krampen, 1989b; Tolor & Jalowiec, 1968, Whitbeck, 1987). Putallaz (1987) found an association between parenting that was high in control and less assertiveness for children in peer interactions, suggesting that these children may have felt less confident in their abilities or opinions. Additionally, Cicchetti, Lynch, Shonk & Manly (1992) report that, on average, maltreated children scored lower than non-maltreated children on teacher reported social self-efficacy in non conflictual situations with peers. It seems likely that the power-assertive or controlling nature of aggressive disciplinary styles may serve to limit a child’s sense of his or her own power in interpersonal situations resulting in avoidance of or emotional withdrawal from interpersonal situations or handling conflict non-competently. Aggressive Attitudes. Parental aggressive behavior may influence adolescents to hold attitudes that support aggression. Attitudes that support aggressive responses to conflict or view aggression as promoting a positive self image may in turn interfere with the development of close friendships. Although studies have not focused on the relationship between attitudes about aggression and friendships, research has found these attitudes to be associated with aggression in children and adolescents (Bentley & Li, 1995; Guerra & Slaby, 1990; Guerra, Huesmann, & Hanish, 1995; Guerra, Huesmann, Tolan, Van Acker, & Eron, 1995; Huesmann, Guerra, Zelli, & Miller, 1992; Parke & Slaby, 1983; Perry, Perry & Rasmussen, 1986; Slaby & Guerra, 1988). Furthermore, selfefficacy and attitudes about aggression were found to demonstrate a significant relationship with close friendship in data being used for the current study (Land & Allen, in preparation) suggesting that these attitudes or expectancies play an important role in adolescent close friendships. Parental Aggression and Peer Relationships 17 Few studies have investigated a mediating model with regard to parental aggression and peer relationships. One study that compared a mediation and a compensatory model of social cognitions in the relationship between maltreatment and peer rejection, did not find support for the mediation model, but did find support for the compensatory model (Downey & Walker, 1989). The use of maternal reports of peer rejection renders the results of this study difficult to interpret since using maltreating parents as reporters of their children’s behavior may have resulted in a biased assessment of the child. Summary Studies relying on peer and teacher reports as well as observational assessments of peer interaction have found maltreated children to exhibit maladaptive social interaction with their peers. Specifically, observational studies of toddlers and preschool age children indicate that children who have experienced physical maltreatment demonstrate negative interactions with their peers and significantly maladaptive responses to negative, neutral and affiliative behaviors of their peers. Peers rate maltreated children less favorably compared to non-maltreated children in sociometric ratings, even by those the maltreated children consider best friends. Peers rate maltreated children as higher in meanness, fighting, attention getting and lower on leadership when compared to their non-abused peers. Teachers and camp counselors view maltreated children as more hostile and less socially competent in their peer interactions. Empirical support for a significant relationship between parental maltreatment and attachment style and research demonstrating a significant relationship between attachment style and peer relationships, suggests that attachment style may play a mediating role between parental aggression and peer relationships. The focus of attachment theory on “internal working models” Parental Aggression and Peer Relationships 18 suggests that internal working models may further mediate this relationship. The few studies that have investigated mediating factors in the relationship between parenting and peer relationships, suggest that perceptions and expectations may be important mediators of the relationship between parents and peers. A small number of empirical studies have demonstrated a relationship between parental aggression and expectancies regarding relationships (trust, self efficacy and attitudes about aggression). Additionally, research also supports the presence of a link between expectancies and peer relationships. It is suggested here that the power assertive and controlling nature of parental aggression influence adolescent expectancies regarding self and other in relationships in the areas of trust, self-efficacy and attitudes about aggression. These expectancies, in turn, are thought to influence the intimacy and social skills in adolescent peer relationships. Overview of the Current Study The current study uses data from the Virginia Study of Teens and Families (VSTF) to explore parental aggression and it’s relationship to different aspects of adolescent peer relationships functioning. This study also examines the mediating role of expectations of self and other in relationships by investigating the potential mediating role of trust, self-efficacy and attitudes about aggression in the relationship between parental aggression and close peer relationships. Additionally, the influence that high-risk environment may have on this relationship is also examined. Crittenden (1998) has suggested that abuse of children may be related to the motivation to promote the safety of children against what parents may see as a dangerous environment. The response of anger, which may become dangerously harsh, to behaviors that may endanger a child, is proposed to be responsible for violence against children Parental Aggression and Peer Relationships 19 in some cases. By the same token, parents may use aggression to protect their children in highrisk environments. Although previous studies have provided evidence regarding maladjustment in the peer relationships of children experiencing parental aggression, these studies have focused on preschool and school aged maltreated children and have been primarily descriptive in nature. The current study builds on previous studies by investigating the relationship between parental aggression and peer relationship quality during adolescence within a non-abusive population. Studying this relationship during adolescence provides the opportunity to better understand the relationship between parental aggression and intimacy since adolescence is an important developmental period for acquiring intimacy. The investigation of this relationship in at risk nonabusive families is important because it assists in understanding the influence of parental aggression that does not reach the level of abuse. In addition, this study attempts to investigate both the relationship between parental aggression and friendship quality and the process by which aggression from parents may influence children’s peer relationships by examining the potential mediating role of relationship expectations in the link between parental aggression and peer relationships. One strength of the current study is the multiple measure construction of the variables to be examined. Additionally, the use of different reporters for predictor and outcome variables avoids method variance confounds. That is, adolescent reports were used to measure parental aggression and each of the relationship expectation variables (self-efficacy, attitudes about aggression and trust) while peer reports were used to measure friendship quality. Hypotheses It was generally hypothesized that experiencing parental aggression from parents during Parental Aggression and Peer Relationships 20 mid-adolescence would be related to less closeness in peer relationships in mid-adolescence and in late adolescence. It was expected that this relationship would be mediated by adolescent expectations regarding interpersonal relationships. The experience of aggression from parents was thought to influence adolescent expectations regarding trust, aggression and self-efficacy in interpersonal relationships with peers. Negative expectations of interpersonal relationships were in turn thought to influence adolescent intimacy with peers. Parental aggression and peer relationship quality Hypothesis 1. Higher levels of parental verbal and physical aggression through midadolescence were each expected to be associated with less intimacy in mid-adolescent peer relationships. This relationship was expected after accounting for potential differences in intimacy for relevant demographic variables. Hypothesis 2. Higher levels of parental verbal and physical aggression in late adolescence were each expected to be associated with less intimacy in late adolescent peer relationships. This relationship was expected after accounting for potential differences in intimacy for relevant demographic variables. Hypothesis 3. Higher levels of parental verbal and physical aggression in late adolescence were each expected to be associated with less social skill competence in late adolescent peer relationships. This relationship was expected after accounting for potential differences in intimacy for relevant demographic variables. Hypothesis 4. Higher levels of parental verbal and physical aggression until midadolescence were each expected to be associated with less intimacy in peer relationships in late adolescence. This relationship was expected after accounting for potential differences in peer Parental Aggression and Peer Relationships 21 relationship intimacy for relevant demographic variables. Hypothesis 5. Higher levels of parental verbal and physical aggression until midadolescence were expected to be associated with lower social skill competence in peer relationships in late adolescence. This relationship was expected after accounting for potential differences in peer relationship social skill competence for relevant demographic variables. Hypothesis 6. Higher levels of parental verbal and physical aggression in midadolescence were each expected to be associated with a decrease in intimacy in peer relationships from mid-adolescence to late adolescence. This relationship was expected after accounting for potential differences in peer relationship intimacy for relevant demographic variables. Parental Aggression, Relationship Expectations and Peer Relationship Quality Hypothesis 7. Higher levels of parental verbal and physical aggression were each expected to be associated with pessimistic expectations regarding relationships specifically defined as lower trust of others, a lower sense of self efficacy and higher aggressive attitudes. These pessimistic expectations regarding relationships were expected to mediate the relationship between parental aggression and adolescent intimacy and between parental aggression and social skill competence in peer relationships. Aggression and Level of Risk. Hypothesis 8. It was expected that parents living in high-risk environments would exhibit more verbal and physical aggression toward their adolescents than those living in low risk environments. Parental Aggression and Peer Relationships 22 Method Sample This study used data from the Virginia Study of Teens and Families (VSTF), a longitudinal project involving 177 adolescents and their parents. Interview, self report and observational data regarding family relationships and adolescent socio-emotional functioning were collected when the target adolescent was approximately 16 (Time 1) and again when the target adolescent was approximately 18 (Time 2). Attrition for the full sample was quite low with 146 of the 177 families who participated in the first wave of data collection returning for the second wave of data collection. In order to obtain adolescent participants from the “Forgotten Half” of high school students, school records were used to select adolescents with one or more risk factors including low grades, one or more failed classes, multiple absences, suspensions and grade retention or attendance at an alternative school (W.T. Grant Commission, 1988). These adolescents were likely not to attend college and were considered at risk for poverty and unemployment in the future. This moderately at risk sample provides a wide range of psychosocial functioning including a substantial number of adolescents who are functioning adequately while also obtaining a substantial number of adolescents who are experiencing serious difficulties. Separate samples were created for analyzing mother and father variables of interest resulting in a significantly smaller sample for father analyses. The difference in mother and father sample sizes is the product of the overrepresentation of single parent families in this sample resulting in a large amount of missing data for adolescent report of their fathers. Mother and father analyses for this study were conducted separately because of the difference sample size, Parental Aggression and Peer Relationships 23 but also because of the possible differences in interaction effects for these two samples. Each of these samples will be described separately below. Mother Sample. Of the 146 families who participated in both waves of data collection, 121 families completed measures used in the analyses including mothers. Family income for the sample ranged from $2,500 to $70,000 with a median income of $25,000. The gender breakdown for the adolescents in the study was 47% male and 53% female. At the first data collection time point (Time 1), adolescents ranged in age from 14 to 18.75 with a mean age of 16. At the second data collection time point, 2 years later, (Time 2) adolescents ranged in age from 16 to 22 with a mean age of 18. The self-identified racial/ethnicity breakdown of the adolescents consisted of 60% Caucasian, 46%African-American, 1% Native American and 1% other. The sample varied with respect to family structure with 55% single parent families, 26% intact two-parent families and 18% stepfamilies. Mothers’ education level ranged from “some high school” to “graduate degree” with the median level of education being “some college”. All of the families in the current sample who participated in the first wave of data collection returned for the second wave of data collection. Father Sample. Of the 146 families who participated in both waves of data collection, 80 families completed measures used in the analyses including fathers. For this sample, family income ranged from $2,500 to $70,000 with a median income of $35,000. The self-identified racial/ethnicity breakdown of the sample consisted of 73% Caucasian, 26%African-American, 1% other. The sample varied with respect to family structure with 39% single parent families, 34% intact two-parent families and 28% stepfamilies. Fathers’ education levels ranged from “some high school” to “graduate degree” with the median level of education being “some Parental Aggression and Peer Relationships 24 college”. All of the families participating in the first wave of data (Time 1) returned for the second wave of data (Time 2). Peer Sample. During the first wave of data collection, 203 peers participated in the study. The average age of the peers at Time 1 was 16.29 and the peer sample was 52.5% female and 60% Caucasian. Peers reported that they had known the targeted teens an average of 4.9 years (SD= 3.28, range= 0-17 years). During the second wave of data collection, 207 peers participated in the study. The average age of the peers at Time 2 was 17.9 ( SD=1.8) and the peer sample was 53% female and 58% Caucasian. At Time 2, peers reported that they had known the teens in the study an average of 5.5 years (SD= 4.3, range= 0.8-18.8 years). In addition, 57% of the peers believed they knew the target adolescent "very well"; 38% described themselves as the target adolescent's best friend. Procedure Adolescents and their families were recruited through two local high schools with approximately half of the families contacted participating in the study. All participation was conducted at the University of Virginia in 2 separate three-hour sessions. Individual face-toface interviews were held with each family member in separate rooms where information regarding adolescent behaviors and family and peer relationships was gathered. Family members were later brought together for videotaped dyadic interaction tasks providing observational data regarding family interaction patterns. Particular effort was made to assure participants of the confidentiality of their responses. Confidentiality and limits of confidentiality with regard to suicidal or homicidal intentions and child abuse reporting requirements were explained during an introduction provided to each Parental Aggression and Peer Relationships 25 family at which time informed consent forms were reviewed and signed by participants. Participants were also reminded of confidentiality when introducing measures requesting particularly sensitive information. Adolescents and their parents were reassured that no one including parents, school personnel or the court would be provided or allowed access to any of the information collected. Participants were also assured that their names would not be associated with their data and only identification numbers would be linked with their responses. Adolescents also named several friends who knew them well, and two of these peers were recruited to participate in the study. Two years later, adolescents and their parents completed similar batteries of measures. After obtaining parental consent, peers came to the University of Virginia for a one-hour session to complete measures about themselves and the adolescent study participant. Families were paid a total of $105 for their participation At both waves of data collection, adolescents were asked to list the names and phone numbers of five friends whom they felt knew them well, and whom they would feel comfortable having participate in the study. Researchers explained that peers would fill out measures to describe themselves and the target adolescent, and that information supplied by themselves and peers would be kept confidential. Researchers also emphasized to teens that they did not have to supply names of friends if they did not feel comfortable doing so. Once adolescents had given consent for their friends to be contacted, two peers of each adolescent were selected randomly and contacted by telephone. After the study was explained, peers who were interested in participating were scheduled to attend a 60-minute interview session at the University of Virginia, during which they completed measures about the adolescent and about themselves. Parental consent was required to participate for peers under 18 years old. Parental Aggression and Peer Relationships 26 Participants were instructed that all information was kept completely confidential. Transportation for the peers was provided if necessary and each peer was paid for participating. Measures Demographic information. Parents, adolescents, and peers provided basic demographic information including gender, age, and race/ethnicity. In addition to this, mothers provided information regarding their education level, family composition, marital status, household income, and number of persons supported by that income. Adolescents and peers reported on which local high school they attended, and peers reported how long they had known the adolescent in the study. Risk Variable. In order to examine differences in parental aggression for families living in low and high-risk environments, this sample was divided into low-risk and high-risk groups (Boykin & Allen, under review). Location of residence and family income was used to determine level of risk. Families who lived within the city boundaries and had a family income that placed them at 200% of the Federal poverty line or below were classified as living in a high-risk environment. These risk factors are used in order to account for the families’ exposure to criminal activity. Crime rates indicate that index offenses and drug-related arrests are at least 2 times higher in the city versus the surrounding county (Virginia Department of State Police, 1995) Parental Verbal and Physical Aggression. Verbal and physical aggression were measured using a modified version of the Conflict Tactics Scale (CTS) (Straus, 1988). The CTS has been used in over 100 papers as a measure of child maltreatment and shows good Parental Aggression and Peer Relationships 27 reliability and validity (Straus & Hamby, 1997). This modified version of the CTS included adolescent ratings of each parent separately regarding how often the target parent employed certain behaviors during a conflict. At Time 1, adolescents were asked to report on lifetime experience of conflict related behaviors, whereas at Time 2, adolescents were asked to report on these parental behaviors for the past year. Two subscales, designed to measure verbal aggression (6 items) and physical aggression (11 items) were derived from adolescent reports on a 4-point scale (1=never, 2=once or twice, 3=several times, and 4=many times). The physical aggression scale items measured the frequency of parental physically aggressive behaviors ranging in severity from “pushed, grabbed or shoved” the respondent to “used a knife or a gun” against the respondent. The verbal aggression scale items measured the frequency of parental non-physically aggressive behaviors including “insulted and swore” at the respondent and “did or said something to spite” the respondent. The physical aggression score was created for each parent by weighting the frequency of each behavior on the physical aggression subscale by a rating of the seriousness of that behavior and then summing across behaviors. (Strauss, 1988). This technique was used in order to give higher weights to more aggressive and traumatic behaviors (e.g. being threatened with a gun or intentionally burned). The physical aggression and verbal aggression scales on this version of the CTS demonstrated adequate internal consistency for adolescent reports of parental aggression at both data collection time points. At time 1, Cronbach alphas were .78 and .70 for mothers’ verbal and physical aggression respectively and were .77 and .80 for fathers’ verbal and physical aggression respectively. At time 2, Cronbach alphas were .78 and .50 for mothers’ verbal and physical aggression respectively and were .82 and .48 for fathers’ verbal and Parental Aggression and Peer Relationships 28 physical aggression respectively. This modified version of the CTS has been used to measure harsh parental punishment (measured from the weighted physical aggression scale), which was demonstrated to contribute to delinquency by inhibiting parent-teen autonomy and relatedness. Father’s harsh punishment was also associated with aggression but only for boys (Davis, 1996). Copies of the Time 1 CTS measure used to obtain adolescents’ reports of each parent’s aggression is attached as Appendix A. Copies of the Time 2 CTS measure used to obtain adolescents reports of mothers’ and fathers’ aggression are attached as Appendix B and C respectively. Relationship Expectation Variables Trust. Adolescent trust in others was measured with adolescents rating their level of agreement to 6 items on a 5-point Likert scale ranging from strongly disagree to strongly agree on The Trust in People Scale. Higher scores on this scale indicating higher trust in others. Examples of items on this scale include “Most people can be trusted” and “Most people would try to take advantage of you if they got the chance”. This scale was adapted from Rosenberg’s (1957) Faith in People Scale which contained 2 forced choice and 3 agree/disagree items designed to assess one’s degree of confidence in others’ trustworthiness, honesty, generosity etc. In addition, to the face validity, the original scale has shown internal consistency of .92. Another adaptation of this scale demonstrated internal consistency ranging from .48 .54 when tested on a national cross section of 1,450 people and was found to be correlated with trust in government, self-worth, life satisfaction, education and religion (Survey Research Center, 1969). The current version of this scale demonstrated adequate internal consistency with a Cronbach alpha of .68. Parental Aggression and Peer Relationships 29 Attitudes about Aggression. Adolescent attitudes about aggression were measured using adolescent ratings of their level of agreement to 15 items on a 5-point Likert scale ranging from strongly disagree to strongly agree on The Beliefs Supporting Aggression. The scale is comprised of four subscales including aggression is legitimate (6 items), aggression increases self-esteem (3 items), aggression helps avoid a negative image (3 items), and victims of aggression deserve that aggression (3 items). Higher scores indicate higher agreement with attitudes supporting aggression. Guerra (1986) and Slaby & Guerra (1988) developed this scale and Slaby and Guerra (1988) reported test-retest reliability with a sample of 66 institutionalized delinquent adolescents of .86 for this scale and alpha coefficients of .53-.72 for the 4 subscales. Highly aggressive teens have scored higher on the Beliefs Supporting Aggression subscale than other adolescents (Slaby & Guerra, 1988). The current version of this measure demonstrated adequate internal consistency with a Cronbach alpha of .74. A copy of the Adolescent Attitudes Questionnaire which contains the Trust in people scale and the Beliefs Supporting Aggression Scale is attached as Appendix D. Self-Efficacy Expectations. In order to assess adolescent expectations of their performance of competent behaviors, a subset of nine vignettes adapted from the Adolescent Problem Inventory (API) for boys (Freedman, Rosenthal, Donahoe, Schlundt & McFall, 1978) and the Problems Inventory for Girls (PIAG) (Gaffney & McFall, 1981) were administered in interview format. The items selected for use from these measures were highly similar in order to maximize comparability of the two forms (Allen, Leadbeater, et al, 1990; Allen, Leadbeater, & Aber, 1994). Adolescents listened to nine different audio taped descriptions of situations Parental Aggression and Peer Relationships 30 involving problems with peers, teachers, parents and other adult authorities. After each vignette, the adolescent was asked to state what they would do in that situation. After reporting what they would do in these hypothetical situations, adolescents were asked to listen to each scenario again. This time each scenario was followed by a competent response as determined by the manual. After each of these competent responses, the adolescent was asked to rate on a 10point Likert scale the extent to which they believed they could or could not do the competent behavior. Response choices ranged from 1 = “Definitely could not do it” to 10 “Definitely could do it.” Adolescent responses to this question were averaged across the nine vignettes and higher scores reflect greater self-efficacy expectations. Research has shown the API and PIAG to be reliable and valid (Freedman, et al., 1978; Gaffney & McFall, 1981; Hunter & Kelley, 1986; Leadbeater, Hellner, Allen & Aber, 1989; Ward & McFall, 1986). Self-efficacy expectations have been found to be predictive of adolescent problem behaviors (Allen, Leadbeater et al., 1994; Kuperminc, 1994). Cronbach alphas for the self-efficacy scale were .66 for girls and .78 for boys demonstrating adequate internal consistency. Copies of the API and PIAG are attached as Appendix E and F respectively. Adolescent peer relationship functioning. Peer reports were used to assess the target adolescents’ relationship functioning using a modified version of the Self-Perception Profile (Harter, 1985) and the Adolescent Interpersonal Competence Questionnaire (ICQ) (Buhrmester, 1990, 1993). Each of these measures was originally designed to obtain self report data and was modified in order to obtain ratings about the target adolescent from two peers nominated by the teen as people who new him/her well. For each of the scales described below, the peer reports were averaged and the resulting single score was used as an assessment Parental Aggression and Peer Relationships 31 of relationship functioning for that scale. Self-Perception Profile. The close friendship scale from the Self-Perception Profile for Adolescents (Harter, 1988) was used in this study. This subscale, consisting of 5 items, designed to assess the adolescents’ ability to make close friends with whom they can share personal thoughts and secrets (Harter, 1988). Each item contains two contrasting statements regarding close friendship competence. One statement characterizes a more competent adolescent with regard to quality of close friendships and the other provides a statement characterizing a less competent adolescent (i.e. Some people are able to make really close friends BUT Other people find it hard to make really close friends). After choosing one of these statements, the peer is asked to choose whether the statement is “sort of true” or “really true” about the target adolescent and this choice is then translated into a score ranging from 1 to 4 with higher scores indicating higher competence. Harter (1988) reports Cronbach’s alphas of .79-.85 for the friendship subscale. Harter (1988) also reports correlations between the close friendship subscale and the self worth and acceptance subscales ranging from .34 to .56. Cronbach alphas for the Close Friendship scale in this study were .81 (Time 1) and .79 (Time 2), demonstrating good internal consistency. A copy of the Self Perception Profile for Adolescents is attached as Appendix G. Adolescent Interpersonal Competence Questionnaire (AICQ). The peer version of the Adolescent Interpersonal Competence Questionnaire (AICQ) Buhrmester, 1990, 1993) contains 40 items which assessed five domains of interpersonal competence: asserting influence (i.e. “How good is this person at taking charge”) initiating relationships (i.e. “How good is this person at going out of his or her way to start up new relationships”) conflict resolution (i.e. Parental Aggression and Peer Relationships 32 “How good is this person at resolving disagreements in ways that make things better instead of worse”); self-disclosure (i.e. “How good is this person at sharing personal thoughts and feelings with others?”), and providing emotional support (i.e. “How good is this person at being able to make others feel like their problems are understood?”). Participants responded to each of the 40 items (8 per domain) on a 5-point rating scale ranging from “1=poor at this; would be so uncomfortable and unable to handle this situation that it would be avoided if possible” to “5=extremely good at this; would feel very comfortable and could handle this situation well.” Higher scores indicate greater interpersonal competence. Buhrmester (1990) reported Cronbach alpha coefficients for mean scores across all 40 items as .92 for preadolescents and .92 for adolescents. Furthermore, adolescent interpersonal competence, as measured by the Adolescent ICQ, was related to higher sociability, greater self-esteem, less hostility and lower anxiety/depression. This measure demonstrated good internal consistency with Cronbach alphas ranging from .85 to .92 for the scales described above. A copy of the AICQ is attached at Appendix H. In order to gain a more complete assessment regarding the target adolescent’s social skills and closeness in peer relationships, intimacy and social skill competence indices were created by collapsing scales from the above described relationship functioning measures which were expected to best tap these constructs of peer relationship functioning. Intercorrelations among the subscales used to create these indices are presented in Table 1. Social Skills. A single index of social skill competence was created by collapsing peer reports of the target adolescent’s competence on the Initiating Relationships, Asserting Influence and Conflict Resolution subscales of the Interpersonal Competence Questionnaire (Buhrmester, Parental Aggression and Peer Relationships 33 1990). The rationale for collapsing these scales was based on the previous research that indicates that parental aggression may influence these particular areas of social skill functioning (Cicchetti, Lynch, Shonk & Manly, 1992; Fagot, Hagan, Youngblade & Potter, 1989; George & Main, 1979; Haskett & Kistner, 1991; Herrenkohl & Herrenkohl, 1981; Hoffman-Plotkin & Twentyman, 1984). Correlations among these scales ranged from .12 to .62. Scores for each subscale were standardized and averaged to create a single measure of adolescent social skill competence as reported by the target adolescent’s peers. Intimacy. A single index of intimacy was created by collapsing peer reports of the target adolescent’s competence on the Close Friendship subscale from the Self-Perception Profile for Adolescents (Harter, 1988), and the Self-Disclosure and Providing Emotional Support subscales of the Interpersonal Competence Questionnaire (Buhrmester, 1990). The rationale for collapsing these scales was based on the observation that these scales each provided slightly different aspects of intimacy and would provide a more complete measure of relationship intimacy. Correlations among these scales ranged from .34 to .63. Scores for each subscale were standardized and averaged to create a single measure of adolescent peer relationship Intimacy as reported by the target adolescent’s peers. Table 2 provides a list of the measures and subscales used to assess each variable of interest in this study. Table 3 provides a summary of the Cronbach alphas for each of the scales discussed above. Results Preliminary Analyses Preliminary analyses were first conducted to test for assumptions of normality and Parental Aggression and Peer Relationships 34 appropriate transformations were performed to reduce skewness, kurtosis, and outliers. Analyses were also conducted to assess the extent to which the target adolescents experienced parental aggression and to assess the change in parental aggression over time. Correlations among all variables were examined and are reported below. Predictor, mediator and outcome variables were examined for demographic variable main and interactive effects for possible inclusion in regression models for primary analyses. Data Screening Descriptive statistics are presented in Table 4. Box plots of each variable were used to detect the presence of extreme values or outliers. A data point was considered an outlier if it exceeded 3 standard deviations from the sample mean of the variable in question. Outliers were detected in the measures of adolescent reported mothers’ verbal and physical aggression and fathers’ physical aggression at Time 1 and mothers’ physical aggression and verbal aggression and fathers’ physical and verbal aggression at Time 2. Outliers were also detected in adolescent self-report of attitudes about aggression and self-efficacy at Time 2 as well as peer reports regarding the target adolescent’s assertion of influence. Outliers were rescored to a value that was 3 standard deviations from the sample mean. This strategy allows extreme scores on measures to remain in the distribution without having undue influence on results (Tabachnick & Fidell, 1989). After rescoring outliers and computing new sample means no new outliers were detected. The weighted physical aggression scale on the Conflict Tactics Scale remained negatively skewed (indicating less physical aggression in the sample) after the outlier transformations and was subjected to logarithmic transformations so they would more closely approximate normality. This transformed scale was used in all analyses regarding Parental Aggression and Peer Relationships 35 parental physical aggression. Mean difference in study variables across demographic variables. Gender, environmental risk, minority status, age and family type mean differences were examined for all study variables and are presented in tables 5 through 9. These differences were examined in order to better understand demographic differences in aggression, relationship expectations and peer relationship quality. In addition, these differences were examined to determine demographic variables to be included in the regression models. Few demographic variable differences were found in these analyses. Mothers’ verbal and physical aggression were each found to differ by age, with older adolescents reporting higher mean levels of verbal and physical aggression than younger adolescents. Mothers’ verbal aggression was also found to differ for environmental risk with those in low-risk environments reporting higher levels of verbal aggression from their mothers than those in high-risk environments. Intimacy in peer relationships was found to differ by gender and minority status with females and Caucasian adolescents reporting higher mean levels of peer relationship intimacy. A possible mean difference in social skills in peer relationships by family type was indicated with a trend finding in this analysis indicating the possibility of slightly lower peer relationship social skills for adolescents in nondivorced families. Adolescent attitudes about aggression demonstrated a gender difference with males reporting higher aggressive attitudes than females Correlational Analyses For descriptive purposes, correlational analyses are presented in Tables 10 through 14 and described below as a preliminary analysis to the presentation of the primary analyses in this Parental Aggression and Peer Relationships 36 study. Parental Aggression Variables. Intercorrelations among the parental aggression variables presented in Table 10 reveal significant positive correlations between parental verbal and physical aggression. Parents’s verbal aggression was moderately correlated with their own physical aggression at both time points. At time 1, mothers and fathers demonstrated little similarity in verbal and physical aggression with the exception of mothers’ verbal and fathers’ physical aggression being weakly correlated. At time 2 mothers and fathers demonstrated moderate correlations for the same category of aggression while correlations between mothers’ and fathers’ different types of aggression were slightly lower. Mothers’ and fathers’ verbal aggression at Time 1 demonstrated a moderate correlation with their own verbal aggression at Time 2, but physical aggression for each parent showed no correlation between Times 1 and 2. Parental aggression and relationship expectation variables. Correlations between the parental aggression and relationship expectations variables for Time 1 and Time 2 are depicted in Tables11 and 12 respectively. These tables reveal little to no relationship between each parent’s aggression and the relationship expectation variables. Table 2 reveals no significant relationship between Time 1 parental aggression variables and Time 2 relationship expectation variables. Table 3 reveals correlations between Time 2 parental aggression and time 2 relationship expectation variables in the expected directions although overall, these correlations reveal weak relationships between these variables. The only significant relationship found in this analysis was that of fathers’ verbal and physical aggression demonstrating a moderately negative relationship with adolescent self-efficacy. These results Parental Aggression and Peer Relationships 37 suggest self-efficacy to warrant further investigation as a mediator in the relationship between parental aggression and peer relationship quality. Parental aggression and adolescent peer relationship quality. Correlations between parental aggression and adolescent peer relationship variables are presented in Table 13 indicating that parental aggression demonstrated no relationship to peer relationship quality at Time 1. At Time 2 correlations were found to be in the expected direction and mother’s physical aggression demonstrated a weak relationship to both peer relationship intimacy and social skills and fathers’ physical aggression demonstrated a weak association with peer relationship intimacy. Relationship expectations and peer relationship quality. Correlations between relationship expectations and peer relationship quality are presented in Table 14 and demonstrate significant yet weak relationships between these variables with the exception of trust demonstrating no relationship to social skills. . The above correlations demonstrate the complexity of the relationships among the constructs of parental aggression, relationship expectations and peer relationship quality. These constructs were further examined in the primary analyses described below. Primary Analyses Hierarchical regression analyses were used to examine the relationship between parental aggression and peer relationship functioning. Results for demographic mean differences in each of the outcome variables, discussed above, were examined to determine the demographic covariates to be included in the hierarchical regression analyses. Demographic variables demonstrating significant (p<.05) or trend level (p<.10) mean differences were then included in Parental Aggression and Peer Relationships 38 the subsequent regression analyses. For all regression analyses predicting peer relationship intimacy, gender and minority status main effects were accounted for in the first step of the model. For all analyses predicting peer relationship social skills, gender and family type were accounted for in the first step of the model. The remaining steps of the regression models for both outcome variables included parental aggression main effects added in the second step and interaction terms added in the third and final step of the analyses. Interaction terms were created by standardizing the parental aggression variables and multiplying them by each of the demographic variables included in the model. For all of the regression analyses described below, gender was found to be a significant predictor of peer relationship intimacy suggesting that girls demonstrated more intimacy in their peer relationships than boys. In analyses including minority status, minority status was not found to predict peer relationship intimacy. Neither gender nor family type was found to significantly predict peer relationship social skill competence. None of the interaction terms were found to be significant in any of the following analyses. Cross Sectional Analyses Regarding Parental Aggression and Peer Relationship Closeness in Mid-Adolescence (Time 1). It was hypothesized that parental verbal and physical aggression up through midadolescence would each be associated with less closeness in mid-adolescent peer relationships. This relationship was expected after accounting for potential differences in closeness for boys and girls and minority and non-minority adolescents. Mother Analyses. Hierarchical regression results for mothers’ verbal aggression and mothers’ physical aggression up through mid-adolescence predicting mid-adolescent close Parental Aggression and Peer Relationships 39 friendship are presented in Tables 15 and 16 respectively. Consistent with the hypothesis, mother’s verbal aggression up through mid-adolescence predicted less intimacy in midadolescent peer relationships after controlling for gender and minority status. This means that adolescents experiencing higher levels of verbal aggression from their mothers up through midadolescence, were reported by their peers to demonstrate less closeness in their mid-adolescent peer relationships. However, contrary to the hypothesis, mothers’ physical aggression up through mid-adolescence was not associated with mid-adolescent peer relationship closeness after controlling for gender and minority status. This means that adolescents experiencing higher physical aggression from their mothers up through mid-adolescence, were not reported by their peers to demonstrate differences in closeness in their mid-adolescent peer relationships. Father Analyses. Hierarchical regression results for fathers’ verbal aggression and fathers’ physical aggression up through mid-adolescence predicting mid-adolescent close friendship are presented in Tables 17 and 18 respectively. Contrary to the hypothesis, fathers’ verbal aggression up through mid-adolescence was not associated with mid-adolescent peer relationship closeness after controlling for gender and minority status. This means that adolescents experiencing higher verbal aggression from their fathers up through midadolescence were not reported by their peers to demonstrate differences closeness in their midadolescent peer relationships. Also contrary to the hypothesis, fathers’ physical aggression was not associated with mid-adolescent peer relationship closeness after controlling for gender and minority status effects. This means that adolescents experiencing higher levels of physical aggression from their fathers up through mid-adolescence were not reported by their peers to demonstrate differences in closeness in their mid-adolescent peer relationships. Parental Aggression and Peer Relationships 40 Cross Sectional Analyses Regarding Parental Aggression and Peer Relationship Intimacy in Late Adolescence (Time 2). It was hypothesized that parental verbal and physical aggression in late adolescence would each be associated with less intimacy in late adolescent peer relationships. This relationship was expected after accounting for potential differences in intimacy for boys and girls and minority and non-minority adolescents. Mother Analyses. Hierarchical regression results for mothers’ verbal aggression and mothers’ physical aggression in late adolescence predicting late adolescent peer relationship intimacy are presented in Tables 19 and 20 respectively. Consistent with the hypothesis, mother’s verbal aggression in late adolescence predicted less intimacy in late adolescent peer relationships after controlling for gender and minority status. This means that adolescents experiencing higher levels of verbal aggression from their mothers in late adolescence were reported by their peers to demonstrate less closeness, self-disclosure and to provide less emotional support in their late adolescent peer relationships. However, contrary to the hypothesis, mothers’ physical aggression in late adolescence was not associated with peer relationship intimacy after controlling for gender and minority status. This means that adolescents experiencing higher physical aggression from their mothers in late adolescence were not reported by their peers to demonstrate differences in their closeness, self-disclosure and provision of emotional support in their late adolescent peer relationships. Father Analyses. Hierarchical regression results for fathers’ verbal aggression and fathers’ physical aggression in late adolescence predicting late adolescent peer relationship intimacy are presented in Tables 21 and 22 respectively. Contrary to the hypothesis, fathers’ Parental Aggression and Peer Relationships 41 verbal aggression in late adolescence was not associated with late adolescent peer relationship intimacy after controlling for gender and minority status. This means that adolescents experiencing higher verbal aggression from their fathers in late adolescence were not reported by their peers to demonstrate differences in their closeness, self-disclosure and provision of emotional support in their late adolescent peer relationships. However, consistent with the hypothesis, fathers’ physical aggression in late adolescence predicted less intimacy in late adolescent peer relationships after controlling for gender and minority status effects. This means that adolescents experiencing higher levels of physical aggression from their fathers in late adolescence were reported by their peers to demonstrate less closeness, self-disclosure and to provide less emotional support in their late adolescent peer relationships. Cross Sectional Analyses Regarding Parental Aggression and Peer Relationship Social Skill Competence in Late Adolescence (Time 2). It was hypothesized that parental verbal and physical aggression in late adolescence would each be associated with less social skill competence in late adolescent peer relationships. This relationship was expected after accounting for potential differences in social skills for boys and girls and adolescents from divorced and non-divorced families. Mother Analyses. Hierarchical regression results for mothers’ verbal aggression and mothers’ physical aggression in late adolescence predicting late adolescent peer relationship social skill competence are presented in Tables 23 and 24 respectively. Consistent with the hypothesis, mother’s verbal aggression in late adolescence predicted lower social skill competence in late adolescent peer relationships after controlling for gender and family type differences in social skill competence in peer relationships. This means that adolescents Parental Aggression and Peer Relationships 42 experiencing higher levels of verbal aggression from their mothers in late adolescence were reported by their peers to demonstrate less competence in asserting influence, resolving conflicts and initiating relationships with their peers in late adolescence. Also consistent with the hypothesis, mothers’ physical aggression in late adolescence predicted lower social skill competence in late adolescent peer relationships. This means that adolescents experiencing higher physical aggression from their mothers in late adolescence demonstrated less competence in asserting influence, resolving conflicts and initiating relationships with their peers in late adolescence. Father Analyses. Hierarchical regression results for fathers’ verbal aggression and fathers’ physical aggression in late adolescence predicting late adolescent peer relationship social skill competence are presented in Tables 25 and 26 respectively. Contrary to the hypothesis, father’s verbal aggression in late adolescence was not associated with social skill competence in late adolescent peer relationships after controlling for gender and family type differences in social skill competence in peer relationships. This means that adolescents experiencing higher levels of verbal aggression from their fathers in late adolescence were not reported by their peers to demonstrate differences in competence in asserting influence, resolving conflicts and initiating relationships with their peers in late adolescence. Also, contrary to the hypothesis, fathers’ physical aggression in late adolescence was not associated with social skill competence in late adolescent peer relationships. This means that adolescents experiencing higher physical aggression from their fathers in late adolescence were not reported by their peers to demonstrate differences in their competence in asserting influence, resolving conflicts and initiating relationships with their peers in late adolescence. Parental Aggression and Peer Relationships 43 Longitudinal Analyses of Parental Aggression and Peer Relationship Quality. Longitudinal analyses were conducted in order to investigate the association between parental aggression up through mid-adolescence and peer relationship intimacy in late adolescence. It was hypothesized that higher levels of mothers’ verbal and physical aggression until mid adolescence would each be associated with a less intimacy in peer relationships in late adolescence. This relationship was expected after accounting for potential differences in intimacy for boys and girls and minority and non-minority adolescents. Mother Analyses. Hierarchical regression results for mothers’ verbal and physical aggression up through mid-adolescence predicting late adolescent peer relationship intimacy are presented in Tables 27 and 28 respectively. Contrary to the hypothesis, mothers’ verbal aggression up through mid-adolescence did not predict late adolescent peer relationship intimacy after controlling for gender and minority status. This means that adolescents experiencing higher verbal aggression from their mothers up through mid-adolescence were not reported by their peers to demonstrate differences in closeness, self disclosure and provision of emotional support in their peer relationships in late adolescence. Also, contrary to the hypothesis, mothers’ physical aggression up through mid-adolescence did not predict late adolescent peer relationship intimacy after controlling for gender and minority status. This means that adolescents experiencing higher physical aggression from their mothers up through mid-adolescence were not reported by their peers to demonstrate differences in the closeness, self-disclosure and provision of emotional support in their peer relationships in late adolescence. Father Analyses. Hierarchical regression results for fathers’ verbal and physical Parental Aggression and Peer Relationships 44 aggression up through mid-adolescence predicting late adolescent peer relationship intimacy are presented in Tables 29 and 30 respectively. Contrary to the hypothesis, fathers’ verbal aggression up through mid-adolescence did not predict late adolescent peer relationship intimacy after controlling for gender and minority status. This means that adolescents experiencing higher verbal aggression from their fathers up through mid-adolescence did not demonstrate any difference in the closeness, self-disclosure and provision of emotional support in their peer relationships in late adolescence. Also contrary to the hypothesis, fathers’ physical aggression did not predict late adolescent peer relationship intimacy after controlling for gender and minority status. This means that adolescents experiencing higher physical aggression from their fathers up through mid-adolescence did not demonstrate any differences in the closeness, self disclosure and provision of emotional support in their peer relationships in late adolescence. Longitudinal Analyses of Parental Aggression and Peer Relationship Social Skills. Longitudinal analyses were conducted in order to investigate the association between parental aggression up through mid-adolescence and peer relationship social skill competence in late adolescence. It was hypothesized that higher levels of mothers’ verbal and physical aggression until mid adolescence would each be associated with a less social skill competence in peer relationships in late adolescence. This relationship was expected after accounting for potential differences in social skill competence for boys and girls and adolescents from divorced and non-divorced families. Mother Analyses. Hierarchical regression results for mothers’ verbal and physical aggression up through mid-adolescence predicting late adolescent peer relationship social skill Parental Aggression and Peer Relationships 45 competence are presented in Tables 31 and 32 respectively. Contrary to the hypothesis, mothers’ verbal aggression up through mid-adolescence did not predict late adolescent peer relationship social skill competence after controlling for gender and minority status. This means that adolescents experiencing higher verbal aggression from their mothers up through midadolescence were not reported by their peers to demonstrate differences in competence in asserting influence, resolving conflicts and initiating relationships in their peer relationships in late adolescence. Also, contrary to the hypothesis, mothers’ physical aggression up through midadolescence did not predict late adolescent peer relationship social skill competence after controlling for gender and minority status. This means that adolescents experiencing higher physical aggression from their mothers up through mid-adolescence were not reported by their peers to demonstrate differences in the competence in asserting influence, resolving conflicts and initiating relationships in their peer relationships in late adolescence. Father Analyses. Hierarchical regression results for fathers’ verbal and physical aggression up through mid-adolescence predicting late adolescent peer relationship social skill competence are presented in Tables 33 and 34 respectively. Contrary to the hypothesis, fathers’ verbal aggression up through mid-adolescence did not predict late adolescent peer relationship social skill competence after controlling for gender and family type. This means that adolescents experiencing higher verbal aggression from their fathers up through midadolescence did not demonstrate any difference in the competence in asserting influence, resolving conflicts and initiating relationships in their peer relationships in late adolescence. Also contrary to the hypothesis, fathers’ physical aggression did not predict late adolescent peer relationship social skill competence after controlling for gender and family type. This means that Parental Aggression and Peer Relationships 46 adolescents experiencing higher physical aggression from their fathers up through midadolescence did not demonstrate any differences in the competence in asserting influence, resolving conflicts and initiating relationships in their peer relationships in late adolescence. Change Over Time Analyses of Parental Aggression (Time 1) and Peer Relationship Intimacy (Time 2) Controlling for Close Friendship at Time 1. Longitudinal analyses were conducted in order to investigate the association between parental aggression up through mid-adolescence and peer relationship intimacy in late adolescence after controlling for close friendship quality at Time 1. It was hypothesized that higher levels of mothers’ verbal and physical aggression until mid adolescence would each be associated with a decrease in intimacy in peer relationships in late adolescence. This relationship was expected after accounting for potential differences in intimacy for boys and girls and minority and non-minority adolescents. Mother Analyses. Hierarchical regression results for mothers’ verbal and physical aggression up through mid-adolescence predicting late adolescent peer relationship intimacy after controlling for mid-adolescent close friendship are presented in Tables 35 and 36 respectively. Contrary to the hypothesis, mothers’ verbal aggression up through midadolescence did not predict a decrease in peer relationship intimacy in late adolescence after controlling for close friendship in mid-adolescence, gender and minority status. This means that adolescents experiencing higher verbal aggression from their mothers up through midadolescence did not demonstrate a decrease in their peer relationship closeness, self-disclosure and provision of emotional support from mid-adolescence to late adolescence. Also, contrary Parental Aggression and Peer Relationships 47 to the hypothesis, mothers’ physical aggression did not predict a decrease in peer relationship closeness after controlling for mid-adolescent close friendship, gender and minority status. This means that adolescents experiencing higher physical aggression from their mothers did not demonstrate a decrease in peer relationship closeness from mid-adolescence to late adolescence. Father Analyses. Hierarchical regression results for fathers’ verbal and physical aggression up through mid-adolescence predicting late adolescent peer relationship intimacy after controlling for mid-adolescent close friendship are presented in Tables 37 and 38 respectively. Contrary to the hypothesis, fathers’ verbal aggression up through midadolescence did not predict a decrease in peer relationship intimacy after controlling for close friendship in mid-adolescence, gender and minority status. This means that adolescents experiencing higher verbal aggression from their fathers up through mid-adolescence were not reported by their peer to demonstrate a decrease in their peer relationship closeness, selfdisclosure and provision of emotional support from mid-adolescence to late adolescence. Also, contrary to the hypothesis, fathers’ physical aggression did not predict a decrease in peer relationship closeness after controlling for mid-adolescent close friendship, gender and minority status. This means that adolescents experiencing higher physical aggression from their fathers up through mid-adolescence were not reported by their peers to demonstrate a decrease in peer relationship closeness from mid-adolescence to late adolescence. Parental Aggression, Relationship Expectations and Peer Relationship Quality Mediating models were investigated in order to explore the utility of relationship expectation variables as potential mediators in the relationship between parental aggression and Parental Aggression and Peer Relationships 48 peer relationship quality. Baron & Kenny (1986) outline the requirements necessary for testing mediating models in regression analyses. They suggest that in order to test a mediating model, the mediating variable must demonstrate a relationship to the predictor variable and the predictor variable must demonstrate a relationship with the outcome variable. Entering the mediator variables prior to entering the predictor variable then tests the mediating relationship. If the previously significant relationship between the predictor and the outcome variable is no longer significant, a mediating influence is indicated. In the current study, mediating analyses were conducted only when a relationship was found between parental aggression (predictor) and expectations about relationships (mediator) and a relationship was also found between parental aggression and peer relationship functioning variables. It was hypothesized that higher levels of parental verbal and physical aggression were each expected to be associated with pessimistic expectations regarding relationships specifically defined as lower trust of others, a lower sense of self efficacy and higher aggressive attitudes. These pessimistic expectations regarding relationships were expected to mediate the relationship between parental aggression in adolescent closeness and social skill competence in peer relationships. As can be seen in Tables 11 and 12, mother aggression variables demonstrate no relationship with the relationship expectation variables suggesting that mothers’ aggression is not related to adolescent trust, self-efficacy and attitudes about aggression. These results do not support further investigation of a mediating model between mothers’ aggression and adolescent peer relationship quality. On the other hand, the correlations between fathers’ aggression and relationship expectation suggest that the investigation of a mediating model is warranted. Parental Aggression and Peer Relationships 49 Mediating model of the Relationship between Fathers’ Physical Aggression and Peer Relationship Intimacy. Table 39 depicts the results of the hierarchical regression analysis testing the mediating influence of relationship expectations on the relationship between fathers’ physical aggression and peer relationship intimacy. This analysis did not suggest a mediating relationship to be present. As can be seen in Table 39, fathers’ physical aggression remains significant after adolescent trust, self-efficacy and attitudes about aggression have been entered into the model. These results do not suggest trust, self-efficacy and attitudes about aggression to mediate the relationship between fathers’ physical aggression and late adolescent peer relationship intimacy. Aggression and Level of Risk. It was hypothesized that parents living in high-risk environments would exhibit more verbal and physical aggression toward their adolescents than those living in low risk environments. This hypothesis was not supported for mothers’ verbal or physical aggression or fathers’ verbal or physical aggression. Although mean differences were demonstrated between high and low risk families with regard to mothers’ verbal aggression in mid-adolescence, they were not in the expected directions. Adolescents in high-risk environments reported less verbal aggression in mid-adolescence than adolescents in high-risk environments. Discussion This study was undertaken in order to investigate the relationship between parental aggression and peer relationship quality during the developmental period of adolescence in a non-abusive population. Although earlier studies of parental aggression and peer relationships have demonstrated an association between parental aggression and more disturbed peer Parental Aggression and Peer Relationships 50 relationships, these studies have focused primarily on young children experiencing physical abuse. The primary purpose of the current study was to explore the nature of this relationship in mid-adolescence and late adolescence in a non-abusive at-risk sample. A second aim of this study was to compare the influence of parental aggression in the areas of intimacy and social skill competence in peer relationships. Intimacy is thought to be important to study for this age group since the development of intimacy is thought to be a primary task of adolescent development (Berndt, et al., 1986; Buhrmester & Furman, 1987; Hartup, 1983; Youniss & Smollar, 1985). Social skills were chosen based on results of previous research that suggest less competence in conflict resolution, initiating relationships and assertiveness in children experiencing parental aggression (i.e. Cicchetti, Lunch, Shonk & Manly, 1992; Fagot, Hagan, Youngblade & Potter, 1989; Kaufman & Cicchetti, 1989; Salzinger, et. al, 1993). An additional aim of the current study was to explore the process by which parental aggression may influence offspring peer relationships by investigating possible mediating factors in the relationship between parental aggression and peer relationship quality. One strength of this study was the investigation of the relationship between parental aggression and peer relationship quality using adolescent perceptions of parental aggression and peer perceptions of the relationship with the adolescents. This methodology was useful for two reasons. First, it was important that this investigation take into account the adolescents’ perception of the frequency of parents’ aggression since that perception would likely be important with regard to the influence it would have on the adolescent. Second, independent reporters of the constructs in this relationship allow a potentially more valid test of the relationship since the biasing influence of confounds is avoided. Additionally, understanding Parental Aggression and Peer Relationships 51 how the adolescent’s peers view the relationship provides information regarding how others perceive the adolescent, which is a potentially valuable aspect of functioning in peer relationships. Overall, the results of this study focusing on understanding the relationship between parental aggression and peer relationship functioning demonstrate mixed findings. The results of this study indicate that in late adolescence, parental aggression and peer relationship functioning demonstrate a weak yet significant relationship. Adolescent trust, self-efficacy and attitudes about aggression were not found to mediate this relationship. In addition, no relationship was found longitudinally between the adolescents’ report of parental aggression up through midadolescence and late adolescence peer relationships. The results of each of the hypotheses addressed in this study are discussed in detail below. Parental aggression and Peer relationship intimacy in late adolescence. It was hypothesized that parental verbal and physical aggression in late adolescence would each be associated with less intimacy in late adolescent peer relationships over and above differences in peer relationships for boys and girls and minority and non-minority adolescents. As expected, adolescents reporting higher levels of verbal aggression from their mothers in late adolescence were reported by their peers to demonstrate less closeness, selfdisclosure and to provide less emotional support in their late adolescent peer relationships. However, contrary to expectations, adolescents experiencing higher physical aggression from their mothers in late adolescents did not demonstrate differences in their closeness, selfdisclosure and provision of emotional support in their peer relationships. Also contrary to expectations, adolescents experiencing higher verbal aggression from their fathers in late Parental Aggression and Peer Relationships 52 adolescence did not demonstrate differences in their closeness, self-disclosure and provision of emotional support in their late adolescent peer relationships. However, consistent with expectations, adolescents experiencing higher levels of physical aggression from their fathers in late adolescence were reported by their peers to demonstrate less closeness, self-disclosure and to provide less emotional support in their peer relationships. The above findings regarding the association between fathers’ physical aggression and adolescent peer relationship intimacy complement previous studies that find maltreated children to demonstrate lower positive reactions to affiliative behaviors of their peers (Fagot, Hagan, Youngblade & Potter, 1989) to demonstrate higher social withdrawal from peers in general (Hoffman-Plotkin & Twentyman, 1984; Kaufman & Cicchetti, 1989) as well as in response to peer distress (George & Main,1979; Howes & Eldredge, 1985; Klimes-Dougan & Kistner, 1990). These behaviors are likely associated with less closeness, self-disclosure and less emotional support in peer relations. One explanation for the relationship between mothers’ verbal aggression and lower peer relationship intimacy and fathers’ physical aggression and lower peer relationship intimacy is that mothers may increase their expression of verbal aggression while fathers’ may increase their expression of physical aggression toward their adolescent children who are experiencing difficulties that simultaneously affect the ability to form close friendships. Alternatively, adolescents who experience verbal aggression from their mothers or physical aggression from their fathers may also expect to experience aggression from others including their peers. The fear of aggression may result in an unwillingness to engage more openly in peer relationships resulting in less intimacy. This explanation would suggest mothers’ physical aggression and Parental Aggression and Peer Relationships 53 fathers’ verbal aggression to be related to lower peer relationship intimacy as well, however this relationship was not found. It is possible that the small sample used in this study resulted in a lack of power to reveal these findings. Related to this, is the possibility that mothers use less physical aggression than verbal aggression, while fathers use more physical aggression and less verbal aggression resulting in less variance for mothers’ physical aggression and fathers’ verbal aggression in this study. Parental aggression and Peer relationship social skills in late adolescence. Similar to the hypotheses above for peer relationship intimacy, it was expected that parental verbal and physical aggression would each be associated with less social skill competence in late adolescent peer relationships over and above differences in peer relationships for boys and girls and divorced and non-divorced adolescents. As expected, adolescents experiencing higher levels of verbal aggression or physical aggression in from their mothers in late adolescence were reported by their peers to demonstrate less competence in asserting influence, resolving conflicts and initiating relationships with their peers. Contrary to expectations, adolescents experiencing higher levels of verbal or physical aggression from their fathers in late adolescence were reported by their peers to demonstrate no difference in their competence in asserting influence, resolving conflicts and initiating relationships with their peers than those experiencing lower levels of aggression. These results suggest that mothers’ aggression is related to late adolescent social skills while fathers’ aggression is not. The small sample size in the analyses of fathers’ aggression must be considered as an alternative explanation to the lack of findings for fathers’ aggression. However, these findings could also reflect real differences in the association of peer relationship Parental Aggression and Peer Relationships 54 social skills and mother aggression versus father aggression. Mothers may be more influential with regard to prosocial functioning of their children while fathers may be less influential in this area. Adolescents have expressed views of parents as demonstrating differences in involvement, describing fathers as emotionally distant and mothers as more communicative and emotionally connected (Youniss & Smollar, 1985). More communication between mothers and adolescent children may serve to increase mother influence with regard to prosocial functioning, whereas, higher emotional distance of fathers may result in less influence in this area. If mothers demonstrate more influence in adolescent prosocial functioning, aggression from mothers may demonstrate more influence on adolescent social skills while aggression from fathers may not influence this area of peer relationship functioning. Consistent with these findings for adolescents, differences in social skill competence have been found in children experiencing parental aggression in maltreated samples (Cicchetti, Lunch, Shonk & Manly, 1992; Fagot, Hagan, Youngblade & Potter, 1989; Kaufman & Cicchetti, 1989; Salzinger, et. al, 1993) suggesting less competence in conflict resolution, initiating relationships and assertiveness in children experiencing parental aggression Summary of findings regarding mothers and fathers aggression and peer relationship functioning in late adolescence For the cross-sectional analyses in late adolescence, mothers’ and fathers’ aggression differ in their associations with peer relationship functioning. All of the models investigating mothers’ aggression in late adolescence found either a significant or trend level relationship between mothers’ aggression and peer relationship functioning, whereas, only the model investigating the relationship between fathers’ physical aggression and peer relationship intimacy Parental Aggression and Peer Relationships 55 demonstrated a significant relationship. These findings suggest that mothers’ aggression may play a more important role in peer relationship functioning and may demonstrate a more pervasive relationship with young adult peer relationships. That is, mothers’ aggression may influence many aspects of peer relationship functioning whereas fathers’ aggression may demonstrate a more specific influence. However, it is important to remember that the relationship between parental aggression and peer relationship functioning in this study has only been demonstrated to be correlational. Mothers’ aggression may be related to late adolescent peer relationship functioning because mothers’ aggression may be activated in situations where her young adult offspring is experiencing other difficulty that impacts peer relationships. Additionally, the differences in results by parental gender must also be considered in light of the differences in sample size for mothers and fathers. The limited sample size in the analyses of father aggression lessens the statistical power for finding the relationship between fathers’ aggression and peer relationship functioning. It is quite possible that fathers’ aggression would demonstrate the same pattern for statistically reliable associations with a larger sample size. Longitudinal and change over time analyses regarding Parental aggression in mid-adolescence and Peer relationship intimacy and social skills in late adolescence. It was hypothesized that parental verbal and physical aggression up through midadolescence would each be associated with less intimacy in late adolescent peer relationships as well as a decrease in adolescent intimacy from mid-adolescence to late adolescence. Contrary to expectations, adolescents experiencing higher verbal aggression or physical aggression from their mothers up through mid-adolescence did not demonstrate less intimacy in late adolescence or a decrease in intimacy from mid-adolescence to late adolescence. Also contrary to Parental Aggression and Peer Relationships 56 expectations, adolescents experiencing higher verbal aggression or physical aggression from their fathers up through mid-adolescence did not demonstrate less intimacy in late adolescent peer relationships or a decrease in intimacy from mid-adolescence to late adolescence. Similar to the above, it was expected that paternal verbal and physical aggression up through mid-adolescence would each be associated with lower social skill competence in late adolescent peer relationships as well as a decrease in adolescent social skill competence from mid-adolescence to late adolescence. Contrary to expectations, adolescents experiencing higher verbal aggression or physical aggression from their mothers up through mid-adolescence were not reported by their peers to demonstrate less social skill competence in late adolescent peer relationships or a decrease in social skill competence from mid-adolescence to late adolescence. Also contrary to expectations, adolescents experiencing higher verbal aggression or physical aggression from their fathers up through mid-adolescence were not reported by their peers to demonstrate less social skill competence in late adolescence or a decrease in social skill competence from mid-adolescence to late adolescence. One explanation for the difference in cross sectional and longitudinal findings, may be related to the fact that at the first data collection time point, adolescents reported lifetime parental aggression, whereas at the second time point, they reported on parental aggression occurring in the past year. In addition to the likelihood that past year reports may be more accurate than lifetime reports, the absence of a discrete time period, resulted in treating all reports of parental aggression identically regardless of the developmental period within which the aggression may have occurred. Parental aggression during different developmental periods may influence different aspects of peer relationships or may influence peer relationships in Parental Aggression and Peer Relationships 57 different ways. For instance, parental aggression experienced by young children is thought to affect the child’s attachment to the aggressive parent (Cichetti & Braunwald, 1984; Egeland & Sroufe, 1981a, 1981b; Lyons-Ruth, Connell, Zoll & Stahl, 1987) and thus may influence the child’s subsequent peer relationships as a result of the disruption in the establishment of a secure attachment. On the other hand, parental aggression at age 9 may influence the child’s peer relationships through the influence it has on the child’s own aggressive behavior. These different pathways are likely to influence different aspects of peer relationship functioning. The use of a lifetime measure of parental aggression to predict later peer relationship functioning may have resulted in a less specific measurement of parental aggression resulting in the lack of association between these two variables. An alternative explanation is that parental aggression alone demonstrates a short-term influence on peer relationship quality. It may be that parental aggression that does not reach the level of abuse, is only really salient to the adolescent when it has been experienced recently. In other words, for peer relationship intimacy and social skill competence, aggression that occurred earlier in the adolescent’s life may not demonstrate much influence. Parental Aggression, Relationship Expectations and Peer Relationship Quality It was hypothesized that higher levels of parental verbal and physical aggression were each expected to be associated with pessimistic expectations regarding relationships specifically defined as lower trust of others, a lower sense of self efficacy and higher aggressive attitudes. These pessimistic expectations regarding relationships were expected to mediate the relationship between parental aggression in adolescent closeness and social skill competence in peer relationships. Contrary to the hypothesis, trust, self-efficacy and attitudes about aggression Parental Aggression and Peer Relationships 58 were not found to mediate the relationship between parental aggression and peer relationship intimacy. One explanation for the lack of findings regarding a mediating relationship in the relationship between parental aggression and peer relationship quality is that parental aggression may not influence peer relationships in isolation. In other words, aggression in the parentadolescent relationship may be less important in isolation, but demonstrate more influence in the context of other aspects of the quality of the parent-adolescent relationship. The constructs of trust, self-efficacy and attitudes about aggression may be influenced by the interaction of aggression combined with warmth. For instance, high levels of parental aggression combined with low levels of parental warmth may be more important in predicting adolescent trust, selfefficacy and attitudes about aggression. Aggression and Level of Risk. It was hypothesized that parents living in high-risk environments would exhibit more verbal and physical aggression toward their adolescents than those living in low-risk environments. Contrary to expectations, adolescents in high-risk environments reported similar levels of parental aggression with the exception of mothers’ verbal aggression. However, contrary to expectations the differences in mothers’ verbal aggression were in the opposite direction than hypothesized. Adolescents in high-risk environments reported less verbal aggression from their mothers in mid-adolescence than adolescents in low risk environments. It is difficult to interpret this finding particularly in the absence of any other differences in parental aggression for high and low risk groups and may be a chance finding given the number of mean difference analyses conducted. Parental Aggression and Peer Relationships 59 The findings of this study suggest that peer relationship intimacy and social skills are potentially important areas of intervention for adolescents experiencing parental aggression. These results indicate that those experiencing recent or current maternal aggression or paternal physical aggression may be at higher risk for difficulties in these areas of peer relationship functioning than those experiencing verbal aggression from fathers. Social skills training for adolescents experiencing maternal aggression may be helpful in the specific areas of asserting influence, conflict resolution and initiating relationships. Additionally, clinician awareness of this relationship may provide an avenue for intervention in family therapy. Educating parents regarding the potential influence that parental aggression may have in the areas of peer relationship functioning may provide a concrete representation of the impact of parental aggression. It is important to note that this study focused on parental aggression and not abuse. In fact, our self-report measures of physical and verbal aggression indicated that none of the participants in the current sample experienced aggression that reached the level of abuse. These findings generalize to adolescents experiencing moderate levels of aggression, but do not necessarily apply to those experiencing physical or verbal abuse. The relationship between parental aggression and peer relationship quality is likely quite different for abused and nonabused offspring. The presence of a weak relationship between parental aggression and peer relationship quality and the absence of this relationship in many of the analyses is likely related to the small sample size and the difficulties associated with measuring aggression in general. The small sample size particularly in the investigation of father aggression may have limited the power Parental Aggression and Peer Relationships 60 available to discover findings of lesser magnitude and larger sample sizes are needed to address these questions in the future. The difficulty in obtaining accurate reports of parental aggression likely contributed to the weak and null findings in this study. The influence of social desirability likely impacts the accuracy of the reports regarding parental aggression. Additionally, families were informed that information regarding parental aggression was subject to limits with regard to confidentiality. Families were informed at the beginning of the study and when providing information regarding parental aggression toward children that the investigators were required by law to report knowledge of child abuse to social service agencies. Concerns regarding confidentiality, may have resulted in less accurate reporting with a tendency to underreport parental aggression. In addition the use of a lifetime aggression measure in the investigation of peer relationship quality may provide a more generalized view of parental aggression. It seems important given the many developmental stages over an adolescents’ lifetime that parental aggression be measured contemporaneously in specific time periods in order to increase accuracy as well as specificity with regard to the developmental period that may have been affected. Larger sample sizes and greater specificity in aggression measures with regard to time periods are needed to further investigate the relationship between parental aggression and peer relationship quality. Another important direction for future research in this area is the investigation of the context within which parental aggression occurs. It is important to understand to what extent parental aggression is occurring in the context of discipline, the level of anger associated with the aggression and the intentions of the parent dispensing the aggression. Additionally, focusing on parental aggression in isolation provides a very limited view into the parent-child relationship. It Parental Aggression and Peer Relationships 61 is important to understand to what extent the presence or absence of other aspects of the relationship may contribute to offspring functioning. For example, investigating both aggression and warmth in the parent-child relationship would provide a fuller picture of the context within which the aggression is occurring. For example, offspring who experience high levels of aggression from their parents while also experiencing high levels of warmth may function better in peer relationships than offspring who experience high levels of aggression in the absence of warmth. Parental Aggression and Peer Relationships 62 ReferencesAber, J.L. & Allen, J.P. (1987). Effects of maltreatment on young children’ssocioemotional development: An attachment theory perspective. Developmental Psychology,23, 406-414.Ainsworth, M.D.S., Bleher, M.C., Waters, E., & Wall, S. (1978). Patterns ofattachment. Hillsdale, NJ: Erlbaum.Allen, J.P., Leadbeater, B.J., & Aber, J.L. (1990). The relationship of adolescents’expectations and values to delinquency, hard drug use, and unprotected sexual intercourse.Development and Psychopathology, 2, 85-98.Allen, J.P., Leadbeater, B.J., & Aber, J.L. (1994). The development of problembehavior syndromes in at-risk adolescents. Development and Psychopathology, 6, 323-342.Alessandri, S.M. (1991). Play and social behavior in maltreated preschoolers.Development and Psychopathology, 3, 191-205Ammerman, R. T., Cassisi, J. E., Hersen, M., & Van Hasselt, V. B (1986).Consequences of physical abuse and neglect in children. Clinical Psychology Review, 6, 291-310.Armsden, G.C. & Greenberg, M.T. (1987). The inventory of parent and peerattachment: Individual differences and their relationship to psychological well-being inadolescence. Journal of Youth and Adolescence, 16, 427-454.Armsden, G.C. & Greenberg, M.T. (1989). Inventory of parent and peer attachment:Revised manual. Unpublished revised version. University of Wahsington, Seattle, Washington.Azar, S.T., Robinson, D.R., Hekimian, E., & Twentyman, C.T. (1984). Unrealistic Parental Aggression and Peer Relationships 63 expectations and problem-solving ability in maltreating and comparison mothers. Journal ofConsulting and Clinical Psychology, 52, 687-691.Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.Psychological Review, 84, 191-215.Baron, R.M. & Kenny, D.A. (1986). The moderator-mediator variable distinction insocial psychological research: Conceptual, strategic, and statistical considerations. Journal ofPersonality and Social Psychology, 51, 1173-1182.Bauer, W.D. & Twentyman, C.T. (1985). Mother-child interactional style in abuse,neglect, and control groups: Naturalistic observations in the home. Journal of AbnormalPsychology, 93, 106-114.Bentley, K.M., & Li, A.K.F. (1995). Bully and victim problems in elementary schoolsand students’ beliefs about aggression. Canadian Journal of School Psycholgy, 11, 153-165.Bernath, M.S., Feshbach, N.D., & Gralinski, H.J. (1993, March). Physicalmaltreatment and trust in peers: Feelings, reasons, and behavioral intentions. Paper presented atthe biennial meeting of the Society for research in Child Development, New Orleans.Berndt, T.J., Hawkins, J.A., & Hoyle, S.G. (1986). Changes in friendship during aschool year: Effects of children’s and adolescents’ impressions of friendship and sharing withfriends. Child Development, 57, 1284-1297Bowlby, J. (1969/1982). Attachment and loss: Vol 1. Attachment. (2 ed.). NewYork: Basic Books. (Original work published 1969).Boykin, K.A. & Allen, J.P. (1997). Autonomy and adolescent social functioning: Themoderating effect of risk. Under review. Parental Aggression and Peer Relationships 64 Bronfenbrenner, U. (1979). The ecology of human development: Experiments bynature and design. Cambridge, MA: Harvard University Press.Buhrmester, D. (1990). Intimacy of friendships, interpersonal competence, andadjustment during preadolescence and adolescence. Child Development, 61, 1101-1111.Buhrmester, D., & Furman, W. (1987). The development of companionship andintimacy. Child Development, 58, 1101-1113.Cicchetti, D. (1984). The emergence of developmental psychopathology. ChildDevelopment, 55, 1-7.Cicchetti, D. & Braunwald, K.G. (1984). An organizational approach to the study ofemotional development in maltreated infants. Infant Mental Health Journal, 5, 172-183.Cicchetti, D., Lynch, M. Shonk, S, & Manly, J.T. (1992). An organizationalperspective on peer relations in maltreated children. In R.D. Parke & G.W. Ladd (Eds.),Family-peer Relationships: Modes of Linkages. (pp. 345-383).Conaway, L. P. & Hansen, D. J. (1989). Social behavior of physically abused andneglected children: A critical review. Clinical Psychology Review, 9, 627-652.Cowen, E., Pederson, A., Babigian, H., Izzo, L., & Trost, M. (1973). Long-termfollow-up of early detected children. Journal of Consulting and Clinical Psychology, 41, 438-446.Crittenden, P.M. (1998). Dangerous behavior and dangerous contexts: A 35 yearperspective on research on the developmental effects of child physical abuse. In P.K. Trickett& C.J. Schellenbach (Eds.) Violence against children in the family and the community.Washington, DC: American Psychological Association Parental Aggression and Peer Relationships 65 Davis, S. (1996). The relevance of autonomy and relatedness ot he relationshipbetween harsh punishment and juvenile delinquency. Unpublished Dissertation. University ofVirginia.Davis, W.L., & Phares, E.J. (1969). Parental antecedents of internal-external control ofreinforcement. Psychological Reports, 24, 427-436.Dean, A.L., Malik, M.M., Richards, W., & Stringer, A. (1986). Effects of parentalmaltreatment on children’s conceptions of interpersonal relationships. DevelopmentalPsychology, 22, 617-626.Dodge, K.A., Bates, J.E. & Petit, G.S. (1990). Mechanisms in the cycle of violence.Science, 21, 1678-1683.Downey, G. & Walker, E. (1989). Social cognition and adjustment in children at riskfor psychopathology. Developmental Psychology, 25, 835-845.Egeland, B. & Sroufe, L.A. (1981a). Attachment and early maltreatment. ChildDevelopment, 52, 44-52.Egeland, B. & Sroufe, L.A. (1981b). Developmental Sequalae of maltreatmentininfancy. In R. Rizley and D. Cicchettti (Eds.), Developmental perspectives on childmaltreatment. San Francisco: Jossey-Bass.Fagot, B.I., Hagan, R., Youngblade, L.M., & Potter, L. (1989). A comparison of theplay behaviors of sexually abused, physically abused, and nonabused preschool children. Topicsin Early Childhood Special Education, 9(2), 88-100.Freedman, B.J., Rosenthal, L., Donahoe, C.P., Schlundt, G.G., & McFall, R. M.(1978). A social-behavioral analysis of skill deficits in delinquent and non-delinquent boys. Parental Aggression and Peer Relationships 66 Journal of Consulting and Clinical Psychology, 48, 1448-1462.Gaffney, L.R., & McFall, R.M. (1981). A comparison of social skills in delinquent andnon-delinquent adolescent girls using a behavioral role-playing inventory. Journal of Consultingand Clinical Psychology, 49, 959-967.Garmezy, N. & Streitman, S. (1974). Children at risk: The search for the antecedentsof Schizophrenia: I. Conceptual models and Research Methods. Schizophrenia Bulletin, 8, 14-90.George, C. & Main, M. (1979). Social interactions of young abused children:Approach, avoidance, and aggression. Child Development, 35, 306-318.Guerra, N.G. (1986). The assessment and training of interpersonal problem-solvingskills in aggressive and delinquent youth. Unpublished doctoral dissertation, Harvard University,Boston.Guerra, N.G. , Huesmann, L.R. & Hanish, L. (1995). The role of normative beliefs inchildren’s social behavior. In N. Eisenberg (Ed.), Social development (pp. 140-158). ThousandOaks, CA: Sage.Guerra, N.G., Huesmann, L.R., Tolan, P.H., Van Acker, R., & Eron, L.D. (1995).Stressful events and individual beliefs as correlated of economic disadvantage and aggressionamong urban children. Journal of Consulting and Clinical Psychology, 63, 518-528.Guerra, N.G. & Slaby, R.G. (1990). Cognitive mediators of aggression in adolescentoffenders: Intervention. Developmental Psychology, 26, 269-277.Hart, C. H., Ladd, G.W., & Burleson, B.R. (1990). Children’s expectations of theoutcomes of social strategies: Relations with sociometric status and maternal disciplinary styles. Parental Aggression and Peer Relationships 67 Child Development, 61, 127-137.Harter, S. (1988). Manual for the Self-Perception Profile for Adolescents. Unpublishedmanuscript, University of Denver, CO.Hartup, W.W. (1983). Peer relations. In E.M. Hetherington (Series Ed.) and P.H.Mussen (Vol. Ed.), Handbook of Child Psychology, Vol. 4: Socialization, Personality andSocial Development (pp. 103-196). New York: Wiley.Hartup, W. W. (1992). Peer relations in early and middle childhood. In V. B. VanHasselt & M. Hersen (Eds.), Handbook of Social Development: A Lifespan Perspective (pp.257-281). New York: Plenum Press.Haskett, M.E. & Kistner, J.A. (1991). Social interactions and peer perceptions ofyoung physically abused children. Child Development, 62, 979-990.Herrenkohl, R.C., & Herrenkohl, E.C. (1981). Some antecedents and developmentalconsequences of child maltreatment. In R. Rizley and D. Cicchetti (Eds.), Developmentalperspectives on child maltreatment. San Francisco: Jossey-Bass.Hoffman-Plotkin, D., & Twentyman, C. (1984). A multimodal assessment of behavioraland cognitive deficits in abused and neglected preschoolers. Child Development, 55, 794-802.Howes, C. & Eldredge, R. (1985). Responses of abused, neglected, and non-maltreated children to the behaviors of their peers. Journal of Applied DevelopmentalPsychology, 6, 261-270.Howes, C. & Espinosa, M.P. (1985). The consequences of child abuse for theformation of relationships with peers. Child Abuse & Neglect, 9, 397-404.Huesmann, L.R., Guerra, N.G., Zelli, A., & Miller, L. (1992). Differing normative Parental Aggression and Peer Relationships 68 beliefs aobut aggression for boys and girls. In K. Bjorkqvist & P. Niemela (Eds.), Of mice andwomen: Aspects of female aggression (pp. 77-87). San Diego, CA: Academic Press.Hunter, N. & Kelley, C.K. (1986). Examination of the validity of the adolescentproblem inventory among incarcerated juvenile delinquents. Journal of Consulting and ClinicalPsychology, 54, 301-302.Jacobson, R.S. & Straker, G. (1982). Peer group interaction of physically abusedchildren. Child Abuse & Neglect, 6, 321-327.Kaufman, J. & Cichetti, D. (1989). Effects of maltreatment on school-age children’ssocioemotional development: Assessments in a day-camp setting. Developmental Psychology,23, 516-524.Kerns, K. (1996). Individual differences in friendship quality: Links to child-motherattachment. In W.M. Bukowski, A.F. Newcomb & W.W. Hartup (Eds.), The company theykeep: Friendship in childhood and adolescence. Cambridge: University Press.Kinard, E.M. (1980). Emotional development in physically abused children. AmericanJournal of Orthopsychiatry, 50, 686-696.Kinard, E.M. (1982). Experiencing child abuse: Effects on emotional adjustment.American Journal of Orthopsychiatry, 52, 82-91.Klimes-Dougan, B. & Kistner, J. (1990). Physically abused preschoolers’ responses topeers’ distress. Developmental Psychology, 26, 599-602.Krampen, G. (1989). Perceived childrearing practices and the development of locus ofcontrol in early adolescence. International Journal of Behavioral Development, 12, 177-193.Kuperminc, G.P. (1994). Social orientation, social competence and the prevention of Parental Aggression and Peer Relationships 69 adolescent problem behavior. Unpublished doctoral dissertation. University of Virginia.Lamphear, V. S. (1985). The impact of maltreatment on children’s psychosocialadjustment: A review of the research. Child Abuse & Neglect, 9, 251-263.Land, D. & Allen, J.P. (In preparation). Aggression, close friendship competence andunderlying social schemata in adolescence. University of Virginia.Leadbeater, B.J., Hellner, I., Allen, J.P., & Aber, J.L. (1989). Assessment ofinterpersonal negotiation strategies in youth engaged in problem behaviors. DevelopmentalPsychology, 25, 465-472.Lewis, M.L. & Schaeffer, S. (1981). Peer behavior and mother-infnat interaction. InM.L. Lewis and S. Schaeffer (Eds.), The uncommon child. New York: Plenum.Lieberman, A.F. (1977). Preschoolers’ competence with a peer: Relations withattachment and peer experience. Child Development, 48, 1277-1287.Lyons-Ruth, K., Connell, D.B., Zoll, D., & Stahl, J. (1987). Infants at social risk:Relations among infant maltreatment, maternal behavior, and infant attachment behavior.Developmental Psychology, 23, 223-232.Manly, J., & Cichetti, D. (1989). Peer relationships and social competence ofmaltreated children: Ratings in a summer daycamp setting. Unpublished manuscript, Universityof Rochester.Mueller, E. & Silverman, N. (1989). Peer relations in maltreated children. In D.Cicchetti & V. Carlson (Eds.), Child Maltreatment: Theory and Research on the Causes andConsequences of Child Abuse and Neglect, 529-578.Parke, R.D., & Slaby, R.G. (1983). The development of aggression. In P.H. Mussen Parental Aggression and Peer Relationships 70 (Series Ed.) and E.M. Hetherington (Vol. Ed.), Handbook of child psychology. Vol.4Socialization Personality, and Social Development, (pp. 547-641). New York: Wiley.Perry, D.G., Perry, L.C., & Rasmussen, P. (1986). Cognitive social learning mediatorsof aggression. Child Development, 57, 700-711.Pettit, G.S., Dodge, K.A., & Brown, M. (1988). Early family experience, socialproblem-solving patterns, and children’s social competence. Child Development, 59, 107-120.Pettit, G.S., Harrist, A.W., Ates, J.E. & Dodge, K.A. (1991). Family interaction, socialcognition, and children’s subsequent relations with peers at kindergarten. Journal of Social andPersonal Relationships, 8, 383-402.Price, J. (1996). In W.M. Bukowski, A.F. Newcomb & W.W. Hartup (Eds.), Thecompany they keep: Friendship in childhood and adolescence. Cambridge: University Press.Putallaz, M. (1987). Maternal behavior and children’s sociometric status. ChildDevelopment, 58, 324-340.Rosenberg, M. (1957). Occupations and values. (pp. 25-35). Glencoe, IL. Free Press.Rosenberg, M.S., & Reppucci, N.D. (1983). Abusive mothers: Perceptions of theirown and their children’s behavior. Journal of Clinical Child Psychology, 51, 674-682.Rotter, J.B. (1954). Social learning and clinical psychology. Englewood Cliffs, NJ:Prentice-Hall.Salzinger, S., Feldman, R.S., Hammer, M., & Rosario, M. (1993). The effects ofphysical abuse on children’s social relationships. Child Development, 64, 169-187.Slaby, R.G., & Guerra, N.G. (1988). Cognitive mediators of aggression in adolescentoffenders: Assessment. Developmental Psychology, 24, 580-588. Parental Aggression and Peer Relationships 71 Strassburg, Z., Dodge, K.A., Pettit, G.S. & Bates, J.E. (1994). Spanking in the homeand children’s subsequent aggression toward kindergarten peers. Development &Psychopathology, 6, 445-461.Straus, M.A. (1988). Measuring Psychological and Physical Abuse of Children with theConflicts Tactics Scale. Family Research Lab, New Hampshire University, Durham.Straus, M.A. & Hamby, S.L. (1997). Measuring physical psychological maltreatment ofchildren with the Conflict Tactics Scales. Out of darkness: Contemporary perspectives on familyviolence. (pp. 119-135). Thousand Oaks, CA: SageSurvey Research Center, (1969). 1964 Election Study. Ann Arbor, Michigan:International University Consortium for Political Research. University of Michigan.Tolor, A., & Jalowiec, J.E. (1968). Body boundary, parental attitudes, and internal-external expectancy. Journal of Consulting and Clinical Psychology, 32, 206-209.Virginia Department of State Police, Univorm Crime Reporting Section (1995). Crimesin Virginia. Richmond, VA:Author.Ward, C.I., McFall, R.M. (1986). Further validation of the problem inventory foradolescent girls: Comparing Caucasian and black delinquents and non-delinquents. Journal ofConsulting and Clinical Psychology, 54, 732-733.Waters, E., Wippman, J., & Sroufe, L.A., (1979). Attachment, positive affect, andcompetence in peer group: Two studies in construct validation. Child Development, 50, 821-829.Weiss, B., Dodge, K., Bates, J. & Pettit, G. (1992). Some consequences of early harshdiscipline: Child aggression and a maladaptive social information processing style. Child Parental Aggression and Peer Relationships 72 Development, 63, 1321-1335.Whitbeck, L.B. (1987). Modelling efficacy: The effect of perceived parental efficacy onthe self-efficacy of early adolescents. Journal of Early Adolescence, 7, 175-177.William T. Grant Commission on Work, Family, and Citizenship (1988). The forgottenhalf: Pathways to success for America’s youth and young families. Washington, D.C.: Youthand American’s Future: The William T. Grant Commission on Work, Family, and Citizenship.Youniss, J. & Smollar, J. (1988). Adolescent relations with mothers, fathers andfriends. Chicago: University of Chicago Press.