Social Competence: The Role of Quality Relationships and Good Social Skills, Lecture notes of Literature

Download Social Competence: The Role of Quality Relationships and Good Social Skills and more Lecture notes Literature in PDF only on Docsity! Background for Community-Level Work on Social Competency in Adolescence: Reviewing the Literature on Contributing Factors By Elizabeth C. Hair, Justin Jager, and Sarah Garrett Principal Investigator: Kristin A. Moore Project Director: Jonathan F. Zaff December 2001 Prepared for the John S. and James L. Knight Foundation 2 Social Competency in Adolescence Table of Contents Acknowledgement .......................................................................................................................... 5 Structure of this Report................................................................................................................... 6 Methodological Guidelines for this Report..................................................................................... 7 What are the Antecedents of Social Relationships? ....................................................................... 8 Family Relationships .................................................................................................................. 8 Parents..................................................................................................................................... 8 Introduction......................................................................................................................... 8 Stability ............................................................................................................................... 9 Antecedents....................................................................................................................... 11 Individual ...................................................................................................................... 11 Parent ............................................................................................................................ 12 Family ........................................................................................................................... 13 Neighborhood ............................................................................................................... 15 Programs ........................................................................................................................... 15 Summary ........................................................................................................................... 17 Siblings ................................................................................................................................. 18 Introduction....................................................................................................................... 18 Stability ............................................................................................................................. 19 Antecedents....................................................................................................................... 19 Individual ...................................................................................................................... 19 Parents........................................................................................................................... 21 Siblings ......................................................................................................................... 22 Programs ........................................................................................................................... 22 Summary ........................................................................................................................... 23 Grandparents and Other Family members ............................................................................ 23 Introduction....................................................................................................................... 23 Stability ............................................................................................................................. 24 Antecedents....................................................................................................................... 25 Individual ...................................................................................................................... 25 Family ........................................................................................................................... 26 Societal.......................................................................................................................... 27 Programs ........................................................................................................................... 28 Summary ........................................................................................................................... 28 Non-Family Relationships ........................................................................................................ 29 Non-familial Adults .............................................................................................................. 29 Introduction....................................................................................................................... 29 Stability ............................................................................................................................. 30 Antecedents....................................................................................................................... 30 Individual ...................................................................................................................... 30 Family ........................................................................................................................... 31 Neighborhood ............................................................................................................... 32 Programs ........................................................................................................................... 32 5 ACKNOWLEDGEMENT We would like to express our appreciation to all of the people that helped with the creation of this report. We thank our external reviewer, William Bukowski of Concordia University’s Department of Psychology, for his thoughtful comments. We also thank Kristin Moore and Jonathan Zaff for their insightful comments and critiques that contributed to making this a better report. We are also grateful for Thomson Ling's help with literature searches and technical assistance. Finally, we want to express our gratitude to the John S. and James L. Knight Foundation for their support of this project and, in particular, John Bare for his integral role in shaping and guiding this series of reviews. 6 INTRODUCTION The transition to adolescence is characterized as a time of dramatic change for youth (Larson & Richards, 1994). During this stage of the life cycle, youth experience puberty (Steinberg, 1993), expand their cognitive abilities (Keating, 1990; Lapsley, 1990), develop a sense of self and identity (Hair, 1999; Harter, 1999; Keating, 1990; Zaff & Hair, in press), and may alter expectations from school and for academic achievement (Eccles & Midgley, 1990; M. A. Hoffman, Levy-Shiff, Ushpiz, & Schlatter, 1993; Simmons & Blyth, 1987). Their relationships with their parents (Cox, in press; Hair, Jager, & Cochran, 2001; Holmbeck, Paikoff, & Brooks-Gunn, 1995) and peers change, as well (Bukowski, in press; Csikszentmihaly & Larson, 1984; Savin-Williams & Berndt, 1990). As they develop and change, youth must develop the skills or the competence to maintain quality relationships. Social competence is defined as “the ability to achieve personal goals in social interaction while simultaneously maintaining positive relationships with others over time and across situations” (K. H. Rubin & Rose-Krasnor, 1992). Inherent in this definition of social competence are two related, but separate constructs: good social skills and quality social relationships. An adolescent’s level of social competence is associated with a variety of desired outcomes. It is related positively to peer acceptance (Buhrmester, 1990; Coyne & DeLongis, 1986; Kohlberg, Ricks, & Snarey, 1984), to his or her social values (Allen, Weissberg, & Hawkins, 1989), to his or her level of self-efficacy (Connolly, 1989), and to his or her level of self-esteem (Buhrmester, 1990). Deficits in social competencies have likewise been linked to negative adolescent outcomes, including mental health problems, behavior problems, delinquency, substance abuse, sexual offending, loneliness, high-risk sexual behavior, and academic and vocational difficulties (Hansen, Giacoletti, & Nangle, 1995; Kupersmidt & Coie, 1990; R. Miller, 1990; Parker & Asher, 1987). For instance, youth with poor social competence advance more rapidly in their alcohol use and have an accelerated rate of decline in the their self- esteem (Scheier, Botvin, Griffin, & Diaz, 2000). Social competence deficits are often present among adolescents who exhibit disruptive, externalizing behavior problems such as delinquency and conduct disorder (e.g., Dishion, Loeber, Stouthamer-Loeber, & Patterson, 1984; Freedman, Donahue, Rosenthal, Schlundt, & McFall, 1978; Hansen, St. Lawrence, & Christoff, 1988), as well as internalizing disorders such as depression and anxiety (Christoff et al., 1985; Sarason & Sarason, 1984). Since social competency is an important element for adolescents’ healthy development, a central question for those interested in promoting adolescent well-being is what can be done to help adolescents achieve and maintain social competency. STRUCTURE OF THIS REPORT This chapter addresses two key aspects of social competency: quality social relationships and good social skills. As will be seen in this chapter, the association between quality social relationships and good social skills is not clear. We do know that quality social relationships promote good social skills and that good social skills enhance quality relationships. It is quite likely that these two aspects of social competency build on each other; quality relationships promote good social skills, which in turn, enhance quality relationships. Since social relationships, such as the parent-child relationship, are likely to be present in a child’s life before 7 the social skills have a chance to develop1, the paper presents the antecedents of quality social relationships first, and then presents the antecedents of good social skills. For each social relationship and social skill, we present research evidence from individual-, family-, peer-, and community-level factors that have been shown to relate to the development of quality social relationships and/or good social skills. In addition, we present intervention programs that demonstrate improvements in adolescents’ social relationships and/or skills. The social relationships that this chapter will focus on fall into two domains: family and non-family. Family relationships include those with parents, siblings, grandparents, and other family members. Non-family relationships include other adults and peers. The social skills described in this chapter fall into two domains, as well. The interpersonal skills domain includes social skills such as conflict resolution, intimacy, and prosocial behaviors. The individual attributes domain includes skills such as self-control, social confidence, and empathy/sympathy. METHODOLOGICAL GUIDELINES FOR THIS REPORT We emphasize (1) studies that are rigorously implemented experimental evaluations of interventions, in which aspects of the environment are manipulated and social competencies are examined; and (2) studies that are longitudinal, that involve the examination of aspects of the environment as predictors of social relationships and/or social skills and that use multivariate analyses taking background characteristics of the youth into account.2 We have emphasized these types of studies for several reasons. An experimental/control group study is the only research design that permits causal conclusions (as long as they are well- implemented and there is not extensive attrition in the sample over time). It should be noted, though, that experimental designs run the risk of not being generalizable to populations other than those which were studied. Therefore, we highlight studies that have been replicated with similar results across different populations and geographic regions, because successfully replicated interventions have a better chance of being reproduced in additional locations than do studies that have been carried out in a single place, at one single point in time. In addition, longitudinal studies that control for background characteristics can address change over time and address predictive validity better than studies that collect data from one time period (i.e., a cross- sectional design). When little or no information exists for a certain topic, we have included cross-sectional studies with strong theory and rigorous, multivariate; these studies are identified, however, so readers will not place undue confidence in their findings. Since the focus of the paper is on adolescents, we have also restricted studies to those that assess outcomes during adolescence. Therefore, the studies that have outcome data only for childhood are not considered. However, we include longitudinal studies that began in childhood and continued into adolescence or adulthood. In many cases, we discuss characteristics of the youth, family, neighborhood, or society that a program may not be able to change. For instance, programs designed to promote quality 1 We recognize that some researchers would argue that social skills are, in fact, present at the earliest of ages in the form of temperament. The child’s temperament interacts with the parent’s personality/social skills to promote (or degrade) the development of positive relationship with between the parent and the child. 2 These criteria for study inclusion in the present report were culled from a review of school readiness written for the John L. and James S. Knight Foundation (Halle, Zaff, Calkins, and Margie, 2000). 10 influential figures in adolescents’ lives for advice on major life decisions (Holmbeck et al., 1995). Certain longitudinal studies describe a positive change in the parent-child relationship over time. Kim et al. (2001) conducted a nine-year study of parents and adolescents from 451 Caucasian families in rural Iowa. The investigation found that “negative affect” in the parent- child relationship increased through childhood, then declined slightly just prior to the end of the his or her high school career. Similarly, Rice and Mulkeen (1995) found that from adolescence through young adulthood there is a steady, moderate increase in closeness between parent and child. This study tracked reported levels of parent-child intimacy in a sample of 109 predominately Caucasian, middle- to upper-middle class adolescents from the Midwest; data was collected in eighth grade, twelfth grade, and four years subsequent to 12th grade. Finally, literature suggests that, relative to fathers, mothers may be particularly able to maintain a functional parent-adolescent relationship by tailoring the relationship to accommodate the youth’s changing socio-developmental needs (Youniss, 1994). Research has also, however, found a negative change in certain qualities of the parent- child relationship. Larson & Richards (1991) conducted a cross-sectional study of the daily time us of 493 9- to 15-year-old suburban youth; the sample was almost exclusively Caucasian American, and was socio-economically mixed. Participants were paged at random times over a one-week period, to prompt self-reports of whom they were spending time with, their location, their activities, and their mood. Data showed a dramatic decrease in time spent with the family as a whole, the quantity halving from fifth to ninth grade. Affect for the family was reported at lower levels in sixth to eighth grade, but had regained the fifth grade level by ninth grade. Time with just the mother or just the father figures, and time spent conversing with either parent, was not found to change over time. Reviews of the current literature also emphasize that adolescence is a time of increasing influence from peers as opposed to parents (Brown & Theobald, 1999), of increased conflict and negative affect between parents and children (Cox, in press), and of diminished orientation towards the parent-child relationship (K. H. Rubin, Bukowski, & Parker, 1998). Some studies have concluded, however, that different elements of the parent-child relationship change or remain the same depending on particular events in, and characteristics of, the family in question. Aquilino (1997) employed longitudinal data from the National Survey of Families and Households (NSFH). Respondents consisted of 1,507 parents, mostly biological or adoptive (91%), and mostly mothers, who had at least one child between 12 and 18 years of age at Time 1. Analysis of the data suggested that the elements of emotional closeness, control, and conflict in the parent-child relationships remain largely continuous into adulthood. However, certain life transitions appeared to clearly change the relationship in a positive direction. The study found that events such as leaving home, marriage, full-time employment, and enrollment in college all appeared to encourage better-quality parent-child relationships. However, a decrease in emotional closeness was reported by parents subsequent to the child’s home-leaving, and the child’s transition to parenthood was associated with reports of lower relationships quality . In sum, there are mixed findings on the stability of the parent-child relationship. Longitudinal studies have found that certain qualities of the union remain the same, while others change in positive and negative directions. 11 Antecedents Individual Given that parent-child relationships can affect such positive outcomes, such as social accomplishment and positive mental health, it is valuable to understand the factors and contexts that influence their likelihood and success. There is a moderate amount of literature on the individual antecedents of parent-child relationships. Longitudinal results are discussed below, but cross-sectional investigations are also included. Though they are not necessarily conclusive, these studies provide a larger set of studies with which to consider these relationships. As noted above, cross-sectional studies are always identified. The adolescent’s rapport with parental figures appears to predict divergent outcomes. Hightower (1990) conducted a longitudinal investigation of 141 subjects from Berkeley, California. Measures of interpersonal relations, psychological health, and personality were administered when the subjects were age 13 and again at age 50. Results suggest that the degree of respect in a parent-adolescent relationship predicted the adolescent’s focus on and involvement in relationships with peers and parents. A cross-sectional study by Rice, Cunningham, and Young (1997) found that the gender of the parent appears to predict the effects of parent-child relationships. The researchers sampled 630 Southern college students, of which 249 were African American and the rest were Caucasian American. The degree of a child’s attachment to the father was found to be a more accurate predictor of social competence than his or her attachment to the mother. In this case, social competence is understood as the ability to develop relationships with peers, to access social support when needed, and to be comfortable in social situations, for example. Interestingly, in Rice and Mulkeen’s (1995) longitudinal investigation findings showed that boys have closer relationships with their father than girls do. Furthermore, cross-sectional research suggests that individual personality characteristics may influence parent-child conflict. Data from a subsample of the NSFH indicated that parent reports of characteristics such as unhappiness, anxiety, bullying, or a quick temper in the adolescent, for example, were positively associated with higher reports of conflict in the parent- child relationship (Barber, 1994). The sample was composed of 1828 parents of at least one child between 12- and 18-years-old. Participants were Caucasian (73%), African American (21%), and Hispanic (6%). Adolescents’ dating experiences also seem to play a key role in determining the character of parent-child relationships. Dowdy and Kliewer (1998) investigated the dating behavior and reports of parent-child relationships in a cross-sectional study of 859 high school students. The sample was from the southeast, and was diverse in regard to ethnicity, socioeconomic background, and family structure, though gender was split unequally (71% female). The investigation found that adolescents who dated, especially females and short-term daters, were more likely to have “intense conflict” with their parents than were non-daters. This association is moderate, though significant (Dowdy & Kliewer, 1998). These results support findings by Quatman et al. (2001). A cross-sectional study of 380 adolescents in eighth, tenth, and 12th grade was recently conducted in Northern California. The researchers found that frequent dating, described as dating more than “once or twice a month,” is associated with poorer familial relationships than those in families of adolescents who date infrequently (Quatman, Sampson, Robinson, & Watson, 2001). 12 The individual characteristics of the adolescent appear to exert a potentially significant influence on the character of the parent-child relationship. Variables, such as gender, unique personality traits, levels of attachment and respect, and even dating history, can influence the trajectory of this union. Parent Individual qualities of the parent appear to influence the parent-child relationship, as well. Longitudinal data collected over three years by Paley et al. (2000) revealed that higher levels of negative affect towards the adolescent predicted adolescent reports of parents as untrustworthy and unsupportive. Affect was measured by the ratio of parental hostility versus warmth, as expressed during parent-child interaction. The sample consisted of 337 Caucasian adolescents between the ages of 12 and 14 at time who lived with both biological parents and at least one sibling within four years of their age. The participants were of mostly middle- to lower-middle-class families in the rural Midwest. A cross-sectional study by Gavin and Furman (1996) investigated the individual characteristics of mothers and daughters, and daughters and best friends, in harmonious and disharmonious relationships. The sample consisted of 60 adolescent girls between 15 and 18 years old, and was almost entirely from a Caucasian American, middle class background. The participants were selected according to their membership in very harmonious or very disharmonious relationships. Certain characteristics in both mother and daughter, such as higher levels of socioemotional support, displays of affection, appropriate power-sharing (featuring increasing autonomy for the adolescent), and similar interests and emotional needs, appeared to promote a harmonious relationship. Joint usage of “cooperative social skills” and problem-solving ability were also predictive of harmonious relationships. Individual characteristics may also influence the father-child relationship. Data from a subset of the National Survey of Adolescent Health (Add Health)--- Wave 1 participants who reported living with a father or father figure (n=13,308)--- indicate that the father’s income level and employment status are positively associated with his level of closeness to the child, as is religiosity. This closeness is negatively affected by the father’s coresidence with a women “with whom the adolescent does not have affective feelings” (Harris, Heard, & King, 2000, p. 25). The mother’s lifestyle may also affect this relationship. Data from a longitudinal study of 1,158 10- to 14-year-olds indicate that problems in the mother’s relationship with a spouse or partner, and poor quality of parent or parent-figure employment, predict lower reports of maternal warmth in the mother-child relationship (Menaghan, Kowaleski-Jones, & Mott, 1997). Participants were children of women in the nationally representative NLSY79; however, this sample does not represent 10- to 14-year-olds in general, as they were all born to mothers of lower-than-average childbearing age (over a quarter were under age 18, all were under age 23). A mother or father’s parenting style also predicts the child’s inter- and extra-familial relationships in later life. Hightower’s (1990) longitudinal findings show that children who grew up with authoritative parents--- those who balance the enforcement of rules with responsive and encouraging parenting--- had more positive reports of mental health in later life. Mental health has been identified as a quality which facilitates establishing and maintaining social relationships (Hightower, 1990). Also, cross-sectional research based on NSFH data has found that “negative parenting”, measured in levels of spanking, slapping, or yelling at the adolescent, appears to predict conflict in the parent-child relationship (Barber, 1994). Cross-sectional analysis cannot clarify, however, whether negative parenting may have provoked this conflict, or may have developed as reaction to it. Finally, it is possible that certain parental practices can 15 significantly more negative than those of participants from intact families. Conversely, reports of custodial father-child relationships from nonintact families reveal a significantly higher quality relationship than in the case of noncustodial fathers or even resident fathers from intact families. Finally, data collected by Harris, Heard & King (2000) from a subsample of Add Health participants (n=13,308), described above, indicate that the makeup of the family may affect the parent-child relationship. The presence of siblings under the age of six, for example, appears to have a negative affect on father-child closeness, though it does not affect the pair’s participation in joint activities. Residing with an adult relative, such as a grandparent, appears to have a negative affect on the quantity of time fathers spend with their children. The family unit itself appears to greatly influence parent-child relationships. Findings are mixed in concerning the effects of family structure, but there is strong support that family discord, in general, has negative effects on this union. A parent’s non-residential status also appears to compromise the parent-child relationship. Neighborhood The neighborhood in which families reside also appears to affect the parent-child relationship. Over three years, Leventhal and Brooks-Gunn (2001) conducted a quasi- experimental evaluation of the Moving to Opportunities project, a program which helps families to move from public housing to better neighborhoods. The study suggests that living in less disadvantaged neighborhoods, as defined by the Census or, to a lesser extent, as perceived by the family, is associated with a decrease in children arguing with or disobeying their parents; it was also correlated to mothers being less harsh in their parenting. Participating families were largely African American or Hispanic (n=293), living in public housing. Roughly a third of the families chose to participate in one of the two randomly assigned treatment housing options; this group was more disadvantaged than the study group in general. Similarly, cross-sectional data from the NLSY79 indicate a significant association between living in areas with high levels of high school dropouts--- indicative of insufficient educational resources in the community--- and lower levels of maternal warmth and responsiveness towards her child (Kowaleski-Jones, 1996). The sample consisted of the 860 children between the ages of 14 and 18 (in 1994) of women in the NLSY79. These participants are not representative of 14- to 18-year-old adolescents in general, however, as they were all born to mothers of unusually young childbearing age. Programs Positive parent-child relationships can be promoted in a variety of different ways. There are numerous programs offered by both public and private social services agencies. Unfortunately, the vast majority of these have not been evaluated with experimental methods. A handful have been the focus of small, non-representative studies, however, and the available evidence suggests that programs can foster positive changes in the parent-child relationship. One program which has shown promising results in promoting positive parent-child relationships does so through the instruction of various communication, problem-solving, and perspective-taking skills. The goals of this program, called the Iowa Strengthening Families Program (ISFP), include reducing substance use among adolescents and improving the parent- child relationship, among others (Molgaard & Spoth). An experimental evaluation by Project 16 Family, at the Institute for Social and Behavioral research at Iowa State University, investigated 446 families from low-income areas in Iowa. The sample was randomly assigned to either a control or treatment group, which underwent two-hour teaching sessions over seven weeks. The first hour of the session consisted of separate parent and adolescent training. Among other issues, parents were taught limit-setting, communication, encouraging good behavior, and using community resources; adolescents received training on goal-setting, appreciating parents, dealing with stress, and how to deal with peer pressure. A subsequent hour of joint training focused on appreciating others, understanding family values, conflict resolution, and various communication skills. Data, which was collected for four years after the initial treatment, showed that parents experienced more sophisticated parenting skills, had increased “positive feelings” towards their child, and strong parent-child relationships continued to develop over time. This program has been tailored to work with specific ethnic populations, as well; informal measures suggest similarly successful outcomes, but longitudinal studies are still underway. Improvement in the parent-child union can also be accomplished through a reduction of conflict in the relationship. Programs such as ASSET (Adolescent Social Skills Effectiveness Training) seek to reduce such conflict with social skills training for both parents and children. A quasi-experimental evaluation by Noble, Adams and Openshaw (1989), and later by Openshaw, Mills, Adams and Durso (1992), investigated the impact of ASSET on 25 Mormon, dual-parent, middle-class parent-child dyads. The dyads were recruited through advertisements, and self- selected themselves into the control and treatment groups according to their ability to accommodate the training schedule (Openshaw et al., 1992). Pre- and post-tests were performed. One study found that the parent-child pairs in the treatment group improved communication and problem-solving skills at a rate of two- to three- times that of the comparison group (Noble et al., 1989). The other study reported “only modest evidence… that social skills enhancement was able to significantly improve interpersonal relationships,” but found improvements in social skills for both groups; post-test, adolescents demonstrated improved problem-solving and negotiability, parents perceived changes, in the expected directions, of warmth and hostility in the relationships, and both parties reported increases in their ability to give and receive negative feedback (Openshaw et al., 1992). Furthermore, an intervention program focused on parenting appears to ultimately encourage better quality parent-child relationships. The program, geared to educate adolescents on the responsibilities and consequences of parenthood, was part of Save the Children’s Positing Parenting Project in a rural area of Scotland (Cutting & Tammi, 1999). Program exercises included generating ideas of skills and qualities participants considered important in a parent, employing a parent’s perspective to evaluate problems, conflicts, and responsibilities, and discussing the lifestyle changes that would occur if participants had children. In a retrospective evaluation, 27 respondents, 18 males and 9 females between the ages of 13 and 14, reported an increased understanding of their parents’ decisions, motivations, and sacrifices. Respondents also reported being more empathetic towards their parents and more understanding of the demands parents placed on them. However, no females and only a quarter of the males perceived a positive change in their parent-child relationship. Improvements in this union may also come from intervention programs which seek to accomplish their goals through changes in the parent-youth relationship. A program in Southern California, for example, sought to prevent tobacco and alcohol use through the informative bilingual lessons, social skills training, and communication development between parents and children. Six hundred sixty migrant Hispanic families with adolescents participated in the eight- 17 session program, including three parent-child sessions, and a homework component with which parents were encouraged to help. Experimental evaluation found that Hispanic adolescents from families with fewer children experienced notable improvements in parent-child communication (Litrownik et al., 2000). Mentoring and mentor-like relationships between adults and adolescents may also improve the parent-child relationship. This positive change may result by virtue of the adolescent’s participation in a “successful” relationship with a mentor, in which he or she can develop fundamental elements of social interaction, such as trust in others and the productive expression of emotions (Tierney, Grossman, & Resch, 1995). This phenomenon was explored by Tierney, Grossman and Resch (1995), who experimentally evaluated the effects of the Big Brothers/Big Sisters (BB/BS) program. Data were collected from 959 youth, split nearly evenly into the control and treatment groups, at the time of entry into the study, at the time the child- mentor match was made, and 18 months after entry into the study. The majority of the sample was from a minority background (55%), male (60%), and between 11 and 13 years old (69%); all of the sample was between the ages of 10 and 16. Over 25% these youth had been subject to physical, emotional, or sexual abuse, and many (over 40%) were participating in public assistance programs. The researchers found that the quality of parent-child relationships, measured in changes in trust, communication, anger/alienation, was positively associated to participation in the program. Participants also reported lying to their parents less frequency than youth in the control groups (Tierney et al., 1995). Rhodes, Grossman, and Resch (2000) also found that “mentoring led to statistically significant improvements in… youth’s relationships with their parents”. Overall, it appears that various kinds of intervention programs have the potential to positively affect the parent-child relationship. The diverse programs appear to be achieve these improvements through the development of general communication, conflict management, and perspective-taking skills. Summary In sum, adolescents’ quality relationships with their parent(s) have implications for the many outcomes related to the healthy development of adolescents. For instance, the parent- child relationship is associated with development of social skills such as conflict resolution and intimacy. In addition, the parent-child relationship appears to influence the development of other social relationships, such as romantic relationships and other friendships. This relationship also influences the psychological and psychosocial development of youth. There are mixed findings on the stability of the parent-child relationship; for example, qualities such as conflict and closeness are found not to change in certain studies, while they are both found to increase in other studies. However, data from representative samples are not available, and without such data definitive statements are not possible. Individual characteristics, such as the degree of respect in the relationship, the gender of the parent, the gender of the youth, the youth’s personality, and the youth’s dating experience are all related to the quality of the parent-child relationship. Similarly, individual parents’ behavior, level of affect, and parenting styles, as well as characteristics of the family as a whole, influence the quality of the parent-child relationship. There is evidence that intervention programs may positively influence the quality of the parent-youth relationship, though most have not been evaluated experimentally or on representative populations. For instance, there is some evidence that participation in social skills 20 early adolescent sibling relational quality (Stocker, Dunn, & Plomin, 1990). Those personality traits that had a positive effect on relationship quality were a shy older sibling, a sociable younger sibling, a low frequency of upset in both the older and younger sibling, and a short duration of upset for the younger sibling. Those personality traits that had a negative effect on relationship quality were a high activity level of the younger sibling, a younger sibling prone to fits of extreme anger, and a younger sibling with an overall high level of anger. Cross-sectional research on 40 same-sex sibling dyads with the oldest sibling nearing adolescence found that children high in emotionality and activity and low in persistence tend to direct more negative behavior towards their sibling (Brody, Stoneman, & Burke, 1987)s. Beyond a sibling’s temperament, research suggests that specific sibling behaviors can affect the relationship as well. Cross-sectional research carried found that among the 826 mostly white adolescents and adults ranging from age 18 to 25 sampled, that youth with a heavy drinking sibling report weaker relationships with their heavy drinking sibling than do those youth with a non-heavy drinking sibling (Stevenson & Lee, 2001). Authors assert that because heavy drinking is associated with behaviors that are considered disruptive by family members, and these behaviors complicate family dynamics, relationships with heavy-drinking siblings tend to be strained. Somewhat counter-intuitively, some research suggests that delinquent acts can serve as a common interest between siblings and thereby foster sibling relationships. Slomkowski et al. (2001) found that common levels of delinquency are positively related to sibling relationship quality. Utilizing a nationally representative sample, Rowe et al. (1992) found a similar relationship between shared levels of delinquency and positive sibling relationships. Research has shown that demographic characteristics such as age and gender are also associated with the quality of sibling relationships. Dunn et al. (1994) found that during early adolescence boys tend to report less warmth and intimacy with their younger sisters than do early adolescent girls. It should be noted, however, that one recent study found no gender effects when researching sibling relationships (Dunn, Deater-Deckard, Pickering, Golding, & the ALSPAC Study Team, 1999). With respect to the influence of age, both longitudinal research conducted by Dunn et al. (1999) in the United Kingdom on an ethnically diverse sample of over 3,500 early-adolescent sibling dyads, and cross-sectional research mentioned earlier conducted by Stocker et al. (1990) found that older adolescents are less likely to report conflict with their siblings than are younger adolescents and children. Beyond environmental influences, research also suggests that genetic influences may play a role in determining the quality of sibling relationships. Researchers believe that personality development is equally influenced by environmental and genetic factors (Plomin, 1991; Turkheimer, 2000). The influence of genetics on personality development influences social relationships through the effect one’s personality has on how others relate to him or her. Research by Rende, Slomkowski, Stocker, Fulker, and Plomin (1992) suggest that the relationship between genetics, personality, and relationship quality pertains to sibling relationships (Rende, 1992). In sum, individual factors ranging from temperament, levels of alcohol consumption, age, gender, and genetic influences on personality can affect the quality of adolescent sibling relationships. 21 Parents A significant portion of research on sibling relationships has focused on the family and both its positive and negative influence on sibling relationship quality. The family affects sibling relationships in a variety of ways. The family’s level of functioning and make-up have unique effects on sibling relationship quality. Much research documents the negative effects of parental hostility on sibling relationships. Attachment theorists suspect that based on early relational patterns with parents, children develop their own interaction style (Stocker & Dunn, 1994). If parents largely relate to each other and other family members in a hostile manner, then that hostility is likely to manifest itself in their children’s own relational style. Ultimately, the child’s maladaptive relational style has a negative effect on the sibling relationship. Cross-sectional research conducted by K. J. Conger, Conger, and Elder (1994) on 221 seventh-graders with older siblings found that maternal and parental hostility, both towards each other as well as other family members, have a negative effect on sibling relationships. Specifically, this research found that parents’ hostility had a strong negative effect on warm and supportive feelings between siblings. Similar to Stocker and Dunn (1994), Conger et al. (1994) suggest that siblings may emulate their parents’ hostile relational styles during sibling interactions. Furthermore, Dunn et al.’s (1999) longitudinal research on sibling relationships found that both low levels of affection and high levels of hostility between a mother and her partner (usually the father and more often than not a married-partner) were related to later negative behavior directed to the younger sibling by the older sibling, while high levels of reported marital affection was positively linked to the older siblings rating of the younger sibling four years later. In a study of 73 same-sex sibling dyads from two-parent upper- to middle-class families found that negative aspects of the parents’ marital relationship were linked to an older siblings’ negative behavior towards his or her younger sibling (Erel, Margolin, & John, 1998). Finally, according to cross-sectional research, those mothers out of the 64 families sampled who described their marriages as low in affection were more likely to have children who reported higher levels of conflict and competition between one another (Stocker, Ahmed, & Stall, 1997). A perceived discrepancy in parental treatment has a negative effect on sibling relationships. Observational research by Stocker et al. (1990) found that excessive maternal control directed toward older sibling, as rated by an independent coder, leads to a sibling relationship rated as both competitive and controlling. Their study also showed similar results for excessive maternal control directed toward the younger sibling. For those mother-sibling- sibling triads that were rated as having a mother who directed attention toward the younger sibling, the sibling relationship was also rated as competitive and controlling while being witnessed during a video-taped session. According to the results of research by Brody et al. (1987), families in which mothers are differentially controlling, responsive, or affectionate toward their children are more likely to have siblings who report relationships high in conflict and low in friendliness. Finally, while researching mostly white, middle-class families, McHale and Pawletko (1992) found that, among the 62 adolescent sampled (half of which had a younger sibling who was physically disabled), perceived differential treatment of siblings was negatively related to sibling relationship quality. Finally, cross-sectional research found that among the 64 predominantly white, middle- class tenth-grade adolescents sampled, that males who have a mother who works are more likely to have arguments with their siblings and the arguments are likely to last longer when compared 22 to males who have a mother that does not work (Montemayor, 1984). No relationship was found with female siblings. Montemayor (1984) concludes that male adolescents with mothers who work full-time exhibit higher levels of conflict towards their siblings because they often have more responsibilities to tend to at home when compared to adolescent children of mothers who do not work full-time, and they spend more time at home together outside the presence of an adult, who, if home, could serve as a mediator during sibling conflict. In sum, the manner in which the parents of adolescents treat both each other and their adolescent children, a perceived discrepancy in parental treatment of siblings, and a full-time working mother all affect sibling relationship quality. Siblings With respect to the effects of sibling dyad structure on sibling relationship quality, the research is somewhat mixed. Although limited research suggests that there is no relationship between sibling structure variables (Dunn, 1988), a majority of the research suggests otherwise. As mentioned earlier, large-sample longitudinal research conducted by Dunn et al. (1999) found that among the older siblings, there was a decrease with age in both positive and negative behavior towards younger siblings. Cross-sectional research conducted by Buhrmester and Furman (1990) found, among the 363 middle- to-upper class early to late adolescents sampled, that the older the siblings were in the sibling dyad, the less conflict the younger sibling reported with the older sibling. Their findings suggest that sibling relationships become more egalitarian with age. Similarly, cross-sectional research conducted by Stocker et al. (1990) found that the larger the difference in ages between two siblings the more likely they were to be rated as having a positive relationship while being viewed during a video-taped session. Additionally, Stocker et al. (1990) found that same gender sibling dyads were rated as less controlling and less competitive while being viewed during a videotaped session. Finally, according to correlational research of a large sample of adolescent-sibling dyads, same-sex sibling dyads were more likely to enjoy and join in the same behaviors (Rowe et al., 1992). Programs Overall, few programs have been developed to reduce sibling conflict and thereby foster sibling relationships (Furman & McQuaid, 1992). Furman and McQuaid (1992) suggest that because sibling conflict is often considered inevitable, and because the actual point at which sibling conflict crosses over from typical to dysfunctional levels is not well defined, sibling conflict is rarely thought to require clinical intervention. Subsequently, few clinical interventions have been developed. Despite this complacent approach to sibling conflict, practitioners have developed a couple of programs or interventions. An experimental study conducted by Vickerman, Reed, and Roberts (1997) on 26 sibling dyads in late childhood or early adolescence found that a parent training program on dealing with sibling conflict can lower levels of sibling conflict and foster sibling relationships. After a baseline measurement for conflict was attained for the entire sample, mother-sibling-sibling triads were randomly assigned to the control or treatment group. Those triads placed in the group that included training on reprimanding children during and after conflict showed a significant decrease in sibling conflict when compared to those triads that were placed in the control group that who received no training. Other researchers have shown through cross- sectional research that, similar to the study of training mentioned above, a method of reinforcing children for cooperative play and sending them to time-outs for fighting reduces conflicts among 25 profoundly different from those two generations ago, the surrounding community does not itself provide a “common ground” for the grandchild and the grandparent to relate. Subsequently, families must build a relational bridge that will accommodate different stages of life, generational positions, and different historical perspectives. For example, Hagestad (1983) found that both grandchildren and grandparents reported avoiding certain topics of discussion in order to avoid conflict. Typically, those mutually avoided topics involved areas of life in which sociocultural change had occurred. Not only must there be an effort to bridge the generations, but that effort must be reciprocal (Baranowski, 1982; Hagestad, 1985). Hagestad (1983) points out that both the grandparent and the grandchild must initiate and actively negotiate the evolving relationship in order to insure its maintenance. Finally, differences between the roles that grandmothers and grandfathers are comfortable playing affect the strength of the bond between the grandchild and the grandparent as the grandchild passes though adolescence (Hagestad, 1985). As their grandchildren age, grandfathers tend to show a higher regard for grandsons, and they consider topics such as education and life responsibilities as important domains. Grandmothers on the other hand, distinguish less between granddaughters and grandsons as they age, and are more comfortable talking about family concerns, family relationships, and other social topics. Antecedents Individual Researchers studying adolescent relationships with non-parent familial adults have focused on adolescent demographic characteristics as predictors of relationship quality. Specifically, research has shown that both adolescents’ gender and age are predictive of both the quantity and quality of their relationships with non-parent familial adults. Several researchers have documented the effects of gender on adolescent non-parent familial adult relationships. Longitudinal research conducted by Dubas (2001) revealed, among the 335 predominantly white, middle class adolescents sampled, that males tend to report greater closeness to their grandfathers than do females. Interestingly, this finding is consistent with Hagestad’s (1983) contention that grandfathers tend to show higher regard for grandsons. The author attributes her findings to cultural norms on gender relationships that make it easier for grandfathers to relate to late-adolescent grandsons than to late-adolescent granddaughters (Dubas, 2001). Similarly, Creasey (1993) found that regardless of family status, granddaughters reported closer relationships with their grandmothers than grandsons reported with their grandfathers. To summarize the effects of gender (in this case both grandchild and grandparent gender) on the grandchild-grandparent relationships, Dubas (2001) found, based on grandchild reports, that granddaughters were closest to grandmothers, grandsons reported the next level of closeness to grandfathers, grandsons reported the next level of closeness with grandmothers, and granddaughters reported the lowest level of closeness with their grandfathers. According to cross-sectional research conducted by Blyth, Hill, and Theil (1982) on over 2,800 early to mid- adolescents, girls tend to have somewhat larger numbers of familial adult relationships than do boys and have higher frequencies of contact with non-parent familial adults such as grandparents and aunts and uncles. Cross-sectional research by Coates (1987), which consisted of a sample of 390 African American adolescents, found similar results to those of Blyth et al. (1982). Beyond a higher quantity of familial adult relationships, girls also report higher levels of intimacy with familial adults. Cross-sectional research conducted by Benson (1993) found, among the 46,000 26 adolescents sampled, that females reported higher levels of intimacy with non-parent familial adults than did males. Research on adolescent relationships with non-parent familial adults also suggests that the age of the adolescent predicts relationship quality. According to cross-sectional research conducted on 269 adolescent females, early adolescence marks the height of emotional distance between a granddaughter and her grandparents (E. Hoffman, 1980). Some research suggests that as grandchildren pass through adolescence their bond with their grandparents tends to increase. Cross-sectional research on a large sample group found that the older children or adolescents were more likely to use non-parent familial adults as a source of support and guidance than were the younger children (Benson, 1993). Conversely, research conducted by Blyth et al. (1982), which focuses on just grandparents as opposed to all non-parent familial adults, found that adolescent reports of closeness to their grandparents did not increase with age. Family A significant portion of the research on adolescent relationships with non-parent familial adults focuses on parental relationships with the other adults as well as family type as predictors of relationship quality. Parents largely mediate the grandchild-grandparent relationship by serving as a “lineage bridge” between the first and third generations (Baranowski, 1982; Hill, Foote, Aldous, Carlson, & MacDonald, 1970). Several studies document the effect of parents as the intergenerational link between grandchildren and their grandparents. According to longitudinal research based on 400 white rural families, grandchildren with parents who have a positive relationship with the child’s grandparents are more likely to have a strong relationship with their grandparents than are those grandchildren whose parents report a negative relationship with the child’s grandparents (V. King & Elder, 1995). Cross-sectional research on a small sample of adolescent-grandparent dyads found that two-thirds of the 86 young-adult grandchildren sampled indicated that their parents set the pace or standard for their relationship with their grandparents (Robertson, 1976). A different cross-sectional study, on 398 two-parent, middle-class families, found that parents who reported negative relationships with their parents were more likely to have adolescents who reported lower levels of closeness and frequency of contact with their grandparents (Whitbeck, Hoyt, & Huck, 1993). Finally, cross-sectional research of 208 late adolescents found that young-adult grandchildren tend to look to their parents’ relationship with their own parents as a model for how to act towards their grandparents (Hodgson, 1992). Hodgson (1992) suggests two possible explanations for parents’ influence on the grandchild-grandparent relationship. One, parents spending time with their own parents may establish family norms that foster cross-generational relationships. Second, parents who interact with their own parents often bring their own children, which in turn foster grandparent- grandchild interaction. Research also suggests that parental divorce has mixed effects on grandparent-grandchild relationships. Depending on which parent is awarded custody, the grandchild’s relationship with his or her grandparents can be either positively or negatively affected. Results of cross-sectional research by Creasey (1993), using a sample comprised of mostly (85%) custodial mothers, suggest that adolescent grandchildren from divorced families have less satisfactory relationships with their paternal grandparents when compared to adolescent grandchildren from intact families. Fortunately, amount of contact -- both physical and phone -- helped to mitigate the negative effects of divorce on grandchild-paternal grandparent relationships (Creasey, 1993). Similarly, cross-sectional research on 30 single-mother and 30 single-father middle-class 27 families found that divorce leads to a lower frequency of contact between the grandchild and paternal grandparents when the mother was awarded custody, and a lower frequency of contact between the grandchild and maternal grandparents with the father was awarded custody (Hilton & Macari, 1997). Numerous studies report similar results for divorce on grandchild-grandparent relationships (Cherlin & Furstenberg, 1986; Gladstone, 1991; Johnson, 1988; Kivett, 1991; Myers & Perrin, 1993). In short, following a divorce, the relationship is likely to be strained between the grandchildren and the grandparents of the adult child who did not gain custody. Additionally, research on the effects of divorce on grandchild-grandparent relationships suggests that grandchildren of divorced, single-parent families report higher levels of closeness with the parents of their custodial parents and are more likely to turn to them for support than are grandchildren of two-parent families. Cross-sectional research on 391 late-adolescents revealed that adolescents from single-parent families reported higher levels of closeness and support from their grandparents on the custodial parent side when compared to adolescents from intact, two- parent families (Kennedy & Kennedy, 1993). Additional cross-sectional research using a nationally representative sample of grandparents found that in times of family distress, such as divorce, grandparents report extending higher levels of emotional support as well as parent-like behavior toward their grandchildren (Cherlin & Furstenberg, 1986). The grandparents on the custodial parent’s side also reported seeing their grandchildren more after a divorce. Finally, longitudinal research, based on 186 early adolescents from white, middle-class families, found that children in single-parent families were more likely to report closer relationships with their maternal grandparents than were children from two-parent families (Clingempeel, Colyar, Brand, & Hetherington, 1992). Despite numerous studies supporting the relationship between divorce and grandchild- grandparent relationships mentioned above, however, one study using the National Survey of Families and Households (NSFH) found that children’s closeness and amount of contact with both paternal and maternal grandparents was not affected by separation or divorce (Gravenish & Thomson, 2001). Utilizing a national sample of over 13,000 families measure over time, Gravenish and Thomson (2001) were able to include data on children’s contact with grandparents as reported by both parents and the child. It should be noted, however, that Gravenish and Thomson (2001) studied the effects of divorce and separation whereas the studies mentioned above focused only on divorce. In all, around one-third of the families studied were not actually divorced, but merely separated. In sum, parents influence grandchild-grandparent relationships in two key ways. First, they, potentially, serve as intergenerational mediators that foster contact and interaction through the time and respect they provide or do not provide the grandparents. Second, parental divorce, according to a majority of the research, has a negative effect on the grandchild-grandparent relationship of the non-custodial parent. Societal Children raised in African American families tend to have higher levels of non-parent familial adult relationships. Results of cross-sectional research on nearly 400 African American adolescents suggest that African American youth are especially accustomed to turning to non- parent familial adults for support (Coates, 1987). Family type also predicts the quantity of relationships with non-parent familial adults. Cross-sectional research on 125 African-American adolescents showed that those respondents from a single-parent household had an average of 10 kin members who provided social support whereas those respondents from a two-parent 30 with lower levels of depression and anxiety, the development of social support system usage, the ability to deal with relationship problems (Rhodes, Contreras, & Mangelsdorf, 1994), better communication skills, pro-social behaviors, improved parent-child relationships, and adolescent self-esteem (Rhodes et al., 2000). Furthermore, youth in mentoring or mentor-like relationships have been found more likely to develop into competent and autonomous young adults, overall (Rhodes et al., 1992). Teachers are other adults that also have the potential, though less influential than that of extended family members or mentors, to promote positive social outcomes in adolescents (Scales & Gibbons, 1996). Quality teacher-child relationships appear to motivate investment in academic pursuits (Wentzel & Caldwell, 1997), to foster youths’ individual differences (Campbell, Lamb, & Hwang, 2000), to encourage prosocial behavior (Zahn-Waxler & Smith, 1992), and to predict the youth’s later behavioral outcomes (Hamre & Pianta, 2001). Stability As there are few studies on ‘other adult’-child relationships, very little is known about their stability. In a cross-sectional study, Rhodes et al. (1994) interviewed 54 Latino adolescent mothers of low socio-economic status about the effect the presence or absence of a “natural mentor” had on their lives. Natural mentors were defined as an older, more experienced adult who has taken a special interest in the adolescent’s decisions and well-being. Data from participants described particularly stable relationships; almost half of the sample had known their natural mentor for at least 15 years, and over three-quarters expected to maintain the relationship indefinitely. Youth relationships with natural mentors, however, may be significantly more stable than relationships with assigned mentors who are not part of the youth’s social circle and who often originate from a different socio-cultural milieu (Rhodes & Davis, 1996). Some evidence suggests that relationships with pseudo-relatives also appear to be stable. Chatters et al. (1994) employed the nationally representative National Survey of Black Americans dataset to investigate the occurrence of “fictive relatives” (e.g., non-related individuals designated “aunts” and “uncles” by a family) (n=2,107). This study suggests that relationships with fictive kin experience some relational stability, as they represent a level of intimacy with families that is greater than that of informal friends. Literature in the field shows that variations of fictive kin relationships exist in nearly all other ethnic groups, as well, though seemingly less frequently (Chatters, Taylor, & Jayakody, 1994; MacPhee, Fritz, & Miller-Heyl, 1996; Rhodes & Davis, 1996). Unlike mentoring and fictive kin relationships, close teacher-student relationships are most likely limited to one academic year as students advance through each grade. The preceding studies suggest, preliminarily, that promoting natural mentoring relationships when children are young could have long-term beneficial effects. There is inconclusive information on the stability of various other adult-adolescent relationships. The pattern suggests, however, that relationships that occur naturally, as opposed to assigned relationships, may be maintained for longer periods of time. Antecedents Individual There are certain individual characteristics that may encourage or facilitate positive relationships of other adults with adolescents. Gender and race appear to be associated with the likelihood and the effects of such relationships, though the limited research is inconclusive and 31 only functions to suggest dissimilarity along demographic lines. In Hamre and Pianta’s (2001) longitudinal study on teacher-child relationships, the benefits of the types of relationship a teacher established with a student differed by gender. This investigation was based on data from 179 children from a small city school district; the sample differed from the entire cohort class (who were not able to participate in the study through its conclusion in eighth grade) only in that it had a higher percentage of African American subjects (40%, to 60% Caucasian). It was found that boys experienced better long-term social and academic outcomes if low levels of conflict and dependency characterized their relationship with kindergarten teachers. Girls, however, evidenced better outcomes when their teacher-student relationship had a “close” quality (Hamre & Pianta, 2001). Gender also predicted the number of relationships adolescents had with unrelated adults in Blyth et al.’s (1982) exploratory study. When asked to list all of those individuals significant in the subject’s life, males listed an average of 1.89 other adults; females listed 2.31. In both of these cases, females appear to participate in closer and more social relationships with nonparental adults than do males. This is a pattern that has been discovered and discussed by other researchers as well (Rhodes & Davis, 1996). Race may have a similar association with adolescents’ participation in and responses to social relationships. Rhodes (1994) noted that such unions occur more often among African American participants than among Hispanic American participants. Similarly, a discussion of existing literature about fictive kinships concludes that Caucasian Americans families, and thus their children, are less likely to engage in pseudo-family relationships than other groups, particularly African Americans and Hispanic Americans (Chatters et al., 1994; Rhodes & Davis, 1996). Overall, it appears that females are more likely to establish and maintain relationships with other adults, as are non-Caucasian adolescents, especially African American youth. Family Family characteristics also appear to predict the quality and quantity of ‘other adult’ relationships with youth. The few studies that currently exist on this subject suggest that close, positive parent-child relationships early in life encourage later social relationships with other adults. Unfortunately, the cross-sectional nature of these studies provide inconclusive findings. In the comparative study by Rhodes et al. (1994), participant reports suggest a positive association between accepting mother-daughter relationships early in life and the likelihood of having a natural mentor relationship during adolescence. Similarly, many researchers concur that a close parent-child relationship may, by way of more effective socialization, enable the adolescent to establish and maintain successful relationships with non-related adults (Rhodes & Davis, 1996; Wills & Cleary, 1996). Reviews of existing literature suggest that the character of parent-child relationships often predicts the character of the child’s relationship with his or her teachers (Lynch & Cicchetti, 1992; Pianta & Steinberg, 1992). This model may not hold true in cases of maltreated and otherwise insecurely attached children, however, who may seek close relationships – absent in his or her family – with a teacher or other non familial adult (Lynch & Cicchetti, 1992). This study compared parent- and teacher-child relationships of 215 7- to 13- year-old urban children of low socioeconomic status; the sample was split between families with a history of child maltreatment and a demographically similar selection of families with no known history of abuse. Lynch and Cicchetti (1992) found that those children from maltreated families were more likely to report wanting increased psychological closeness with their teachers than were the non-maltreated sample. 32 Finally, it has been found that fictive kin relationships in African American families are more prevalent in female- than male-headed households, and in families of higher versus lower socioeconomic status (Chatters et al., 1994). A study based on NLSMW data also supports the finding that unmarried women are significantly more likely to share a household with nonnuclear kin In sum, it appears that positive parent-child relationships may encourage the adolescent to develop social relations with non-familial adults later in life. Also, female-headed households may be more receptive to non-familial involvement than those headed by males. Neighborhood In a cross-sectional study by Blyth (1982), participant responses suggest that neighborhood and community characteristics influence the likelihood of relationships with ‘other’ adults. This investigation showed that the non-familial adults nominated as “significant” in the adolescent’s life were most often residents in his or her same neighborhood or metropolitan area. Furthermore, the average non-family adult nominated was seen more often and in more different contexts, such as school, activities, and especially in their or the youth’s residence. It is possible that geographic accessibility predicts the non-family candidates with whom the youth may establish relationships. Finally, regional location and tradition may play a role in the participation of ‘other adults’ in the everyday lives of youth. Chatters et al. (1994) found that families from the South were more likely than families in the Northeast to participate in fictive kin relationships. In conclusion, an adolescent’s neighborhood may affect an influence on the youth’s relationships with other adults, given that these relationships occur most often with those who are geographically accessible to them. Programs There are myriad programs in communities across America that match non-related adults and youth; most of these matches would relate to the adolescent as mentors and role-models, as opposed to fictive kin, teachers, or other less formal roles. Unfortunately, as these programs view the establishment of a mentor-youth relationship as an instrument to accomplish the goals of the program (e.g., drug prevention, violence prevention, or school retention) and not as an end in itself, there are few studies on the development and quality of the relationship. Some studies have, however, suggested certain characteristics, which may optimize non-related adult, mentoring relationships. For example, these unions are most successful when mentors establish set meeting times, have access to reliable transportation, are given supervision and training before and after the initiation of the relationship, and are responsive to the mentee’s input and changing needs. Furthermore, participation in social activities together and mentor-mentee matching based on similar interests appear to encourage closer and more supportive relationships. More detailed information about these studies can be found in Jekielek, Moore, & Hair (2002). Overall, there are several promising program conditions and personal characteristics which appear to encourage successful development of adolescents-other adults relationships. Summary In conclusion, relationships with non-familial adults--- such as teachers, mentors, neighbors, and fictive aunts and uncles--- have the potential to positively change an adolescent's social development. Researchers agree that respected "other adults" can transmit social skills to 35 on perception of social support or on levels of pro- or anti-social behavior (Bender & Loesel, 1997). Similarly, the degree of intimacy in a peer friendship appears to be unrelated to psychosocial well-being (Field, Lang, Yando, & Bendell, 1995). Research has found romantic peer relationships to be beneficial to the individual's social development by fostering autonomy (Dowdy & Kliewer, 1998), by enhancing peer group status, allowing for experimentation in heterosocial interactions (Hansen et al., 1992), and by increasing self-perceived popularity and levels of comfort in their peer group (O'Connor, Allen, Bell, & Hauser, 2002; Quatman et al., 2001). These relationships may be especially beneficial for high- risk youth; aggressive girls have been found to engage in less antisocial behavior when dating (Bender & Loesel, 1997). However, research has found that dating, when participated in more than once or twice a month, is associated with lower levels of academic achievement, academic motivation, and investment in teachers, and with higher levels of depressive symptoms (Quatman et al., 2001). Stability Given that peer relationships generally can produce such positive outcomes in the adolescent, their development and maintenance is of particular interest. Unfortunately, there is a paucity of longitudinal studies, which would function best to describe the course of peer relationships. Peers in general During adolescence, many researchers have found that relationships with peers often become more important to the individual than relationships with family members (O'Koon, 1997). The outcome of these relationships years later, however, is not as well-known. During school years, it is suggested that the stability of general peer friendships is largely determined by whether a friend stays in an individual’s class from year to year (Cairns & Cairns, 1994). This data comes from a cohort (n=100) of the Carolina Longitudinal Study (CLS), followed from fourth to twelfth grade. The CLS sample, as a whole, was representative in terms of race, socioeconomic status, and parental occupations of the makeup of the midatlantic communities from which it originated. Tokuno (1986) conducted a cross-sectional study of 34 college students, ages 20 to 28, from diverse ethnic backgrounds and majors (Tokuno, 1986). Overall, Tokuno found that friends remain active influences in many parts of young adults’ lives, even through their twenties. Best Friend There is little research on the stability of the quality of adolescent best frienships. Analysis of a subset of Cairns & Cairns’ (1994) longitudinal study (n=100) showed that that the stability of best friendships increases each year from 10 to 18. However, the likelihood of keeping a particular best friend from year to year remained rather low throughout. The trajectory of certain other relationship characteristics may further suggest the stability of this type of relationship. Rice and Mulkeen (1995) conducted a longitudinal study on adolescent intimacy with parents and peers. The sample of 109 young adults, interviewed at the average ages of 13, 17, and 21, was almost entirely Caucasian, middle- or upper-middle class, and from intact families. The study found that females reported consistently high levels of intimacy with best friends through high school, and that boys’ reports of intimacy increased through high school (at a more significant rate than that of their female peers) (Rice & Mulkeen, 1995). It is reasonable 36 to believe that the level of intimacy is related to the health, functionality, and/or quality of the relationship. Dating Partners There is a paucity of studies on the stability of adolescent romantic relationships, so there are no specific characteristics to report. However, it is widely believed that these unions are relatively unstable, intense, and short-lived (Dowdy & Kliewer, 1998; Feiring, 1996; Quatman et al., 2001). Overall, there are few conclusive findings about the stability of adolescent peer relationships. However, most of the existing literature suggests that these relationships, platonic and romantic, are largely unstable throughout adolescence. Antecedents Individual It appears that there are several individual characteristics that predict the success and the likelihood of relationships with best friends, romantic partners, and peers. Many of these variations differ according to gender and family attachment and experience. Peers in general and best friends In general, it appears that females have more and closer social relationships through childhood and adolescence. In an exploratory study by Blyth, Hill and Thiel (1982), over 2800 seventh- through tenth-grade subjects were asked to list all people they considered significant in their lives. The sample was largely from a middle class background, and from mostly college- educated, intact families. According to the data, female subjects listed peers (“nonrelated young people”) as significant more than males (7.24 versus 5.52) (Blyth et al., 1982). Similarly, O’Koon (1997) found that, compared to males, the female high school students in his cross- sectional study (n=167) had significantly higher levels of attachment to peers. This sample was predominantly Caucasian and from middle- or upper middle-class families. Data from the longitudinal study by Rice and Mulkeen (1995) indicate that until young adulthood, girls are significantly closer to best friends than boys. In young adulthood, the level of intimacy reported by each gender is finally comparable, with the females’ reports nevertheless remaining slightly higher. Interestingly, findings from a cross-sectional study by La Greca and Lopez (1998) seem to contradict this model. The population was an ethnically mixed sample of 250 high school students (3:2 female-male ratio), who came from a predominantly middle-class background. Females rated higher on scores of social anxiety, which may have the potential to interfere with peer relationships. The study showed that higher levels of social anxiety were associated with fewer and less intimate best friend relationships and could lead to fewer opportunities for socialization experiences or impaired social functioning (La Greca & Lopez, 1998). Certain learned individual behaviors and attitudes also seem to influence peer relationships. Gavin & Furman’s (1996) cross-sectional study of adolescent girls and their best female friends identified the following characteristics as those which appear to promote harmonious relationships: socioemotional support, displays of affection, appropriate power- sharing, and similar emotional needs (Gavin & Furman, 1996). In a longitudinal study of 21 boys and 19 girls from infancy into adulthood, Englund, Levy, Hyson, and Sroufe (2000) found that children who were socially isolated during middle childhood had less self-confidence and 37 were less socially competent with their peers in adolescence than those children who interacted with peers in middle childhood. In addition, children who were included socially in middle childhood were more likely to be self-confident and to hold leadership roles in adolescence than children with social difficulties in middle childhood (Englund, Levy, Hyson, & Sroufe, 2000). Finally, relationships with peers may affect young adults who participate in antisocial or deviant behavior differently than non-deviant young adults. Bender and Loesel’s (1997) employed a two-year longitudinal study to investigate the effect of peer relations on antisocial behavior. The sample consisted of 100 high-risk adolescents from several different residential care facilities in north-west Germany. The findings suggest that in antisocial participants, the lack of membership in a peer group can have protective effects. As individuals often befriend those like themselves, participation in a social group would most likely not encourage, and would even discourage, the rehabilitation of antisocial behavior. Dating Partners In a review of some of the social skills necessary for the establishment and maintenance of social and romantic relationships, Hansen et al. (1992) cited the “exposure to appropriate social skill models,” reaction to the individual’s past behavior (such as reinforcement or punishment), and experience in peer social activities. An individual with such skills would be more likely to have, and therefore experience the effects of, romantic and social relationships. Another antecedent to the likelihood of romantic relationships is physical attractiveness. Studies by various researchers (e.g., Galassi and Galassi (1979); Kelly (1982)) have found a positive correlation between physical attractiveness and frequency of dating, popularity, and imagined “stereotyped… pleasant personalities” (Hansen et al., 1992). However, overall, there is little research on individual-level antecedents of romantic relationships. In sum, various individual characteristics appear to predict the character of peer relationships. Certain personality types, higher levels of social skills, and even physical attractiveness appear to particularly encourage the development of these friendships. Family Peers in general and best friends It appears that characteristics of a family, structure and interpersonal rapport, for example, may predict certain characteristics of peer relationships. In the three-year longitudinal study by Paley et al. (2000), parental hostility appeared to be replicated in the adolescent’s own social interactions, which seemed to negatively affect the adolescent’s peer acceptance, as measured through sibling reports. Mounts and Steinberg (1995) conducted a one-year longitudinal study of 500 ninth- through eleventh-graders to investigate the association between parenting styles and peer influence. The findings from this study suggest that an authoritative parenting style (e.g., warm, communicative rapport, appropriate discipline, et cetera) encourages the internalization of parental norms and their subsequent referencing in the adolescent’s social interactions. The authors conclude that youth raised in this parenting style are more influenced by peers’ positive behaviors, which are presumably behaviors valued by parents, and less influenced by peers’ negative behaviors (Mounts & Steinberg, 1995). Conversely, Hightower’s (1990) longitudinal study found a significant association between the adolescent’s decline in respect for his or her parents (due to the interpersonal character of the family), and an increased orientation towards and involvement in peer relationships. 40 representative of 14- to 18-year-olds in general, as they were born disproportionately to young mothers (Kowaleski-Jones, 1996). Programs There are very few programs in existence that have the singular goal of developing and maintaining peer relationships. At this time, those that do exist appear to be too small or limited in scope to have warranted scientific evaluations. However, cross-sectional research has been conducted on programs that discourage peer violence and on programs that may otherwise promote conditions that facilitate the development of positive peer relationships. Peers in general Some of the larger mentoring programs, for example, may improve the individual’s relationship with peers by encouraging the development of necessary social skills. Participants in the Big Brothers/Big Sisters (BB/BS) program, for example, appear to experience this phenomenon. Tierney, Grossman and Resch (1995), described above, studied 959 adolescent mentees in the BB/BS program. After 18 months, mentees were experiencing better quality relationships with, and more emotional support from, their peers than the control group. Programs that use education-driven discourage adolescent violence also appear to promote more positive peer relationships. Anti-violence and –bullying programs may also promote higher quality peer relationships. The Expect Respect Elementary School Project is an anti-bullying, -sexual harassment, and -gender violence program that employs a “whole school” approach to discourage these behaviors (Sanchez et al., in press). The program helped school staff and bus drivers to establish a universal understanding of, and consistent, effective responses to, bullying/sexual harassment; trained of counselors and organized special counseling sessions for victims of bullying/sexual harassment; developed a 12-session curricula to increase student awareness of, prevention of, and responsiveness to bullying/sexual harassment; designed education seminars for parents to learn how to prevent, recognize, and rehabilitate bullying/sexual harassment behavior; and disseminated information on community resources able to help in cases of violence and victimization. Participating schools also had partnerships with local universities and the lead agency of the project, SafePlace, from which they could resource expertise and information. A quasi-experimental evaluation Expect Respect in Texas, based on data from 747 fifth-graders split between six control and six treatment schools, yielded promising results: increases in treatment participants’ ability to identify sexual harassment, knowledge of and awareness of bullying, and proactive reactions to bullying situations by intervening or telling an adult. . The may not be representative of fifth-graders in general, however, as data from participants who spoke only Spanish, did not complete surveys at all three collection times, or changed schools during the study were not included in the evaluation. Dating Partners Currently, there appear to be few programs with the goal of facilitating and/or enhancing heterosocial relationships. Instructional dating programs, in fact, seem to have reached their peak in the 1970’s. Several social scientists conducted experimental and quasi-experimental 41 studies on college-age youth to measure the effectiveness of the instruction of heterosocial skills through various combinations of practice dating, counseling, self-reinforcement, behavioral rehearsal, “heterosexual group discussion”, participant modeling, self-observation by videotape, desensitization, sensitivity training, behavior-skills-training, sexual education, and “cognitive modification” (James P. Curran, 1977; Hansen et al., 1992). Many of these techniques were found to produce positive outcomes, such as increased dating frequency and decreased anxiety in heterosocial situations. For example, 14 female and 21 male college students were selected--- according to their measured hetero-social anxiety and self-reported infrequency of dating--- to participate in an experimental study by Curran and Gilbert (1975). The subjects were randomly assigned to “either the replication skills-training program, a systematic desensitization training program, or a waiting list control”. The researchers found that both treatment groups demonstrated significant decreases in anxiety and increases in dating frequency post-test and at later follow-ups (J. P. Curran & Gilbert, 1975). Intervention programs which discourage interpersonal violence also appear to encourage more positive heterosocial relationships. The Safe Dates project was designed to prevent violence in adolescent dating couples. The program consisted of several components: a role- playing performance; a curriculum focused on violence, gender stereotyping and conflict management taught over ten 45-minute sessions; a poster contest; and a community component including services for victims of dating violence and training for service providers (Foshee, 1998). An experimental evaluation was conducted at 14 schools in a rural North Carolina county. According to matching enrollment size, the institutions were split into treatment (school and community activities) versus control (community activities) schools. Participants were in the eighth or ninth grade, three-quarters Caucasian, half male and female; roughly 37% of females and 39% of males reported having ever experienced partner violence. The sample numbered 1,965 at Time 1, 1,909 a month after the completion of the program, and 1,892 the following year. At Time 2, the treatment group evidenced significant intervention effects; these participants were less accepting of dating violence, had improved communication and anger management skills, were more knowledgeable about victim services, and reported committing 60% less violence against their romantic partner than the control group (Foshee, 1998). Overall, it appears that there are several programs that may, directly or indirectly, encourage the development of positive adolescent peer relationships. For friendships, the development of social, conflict management, and communication skills appears to improve peer relationships. In the case of heterosocial unions, anxiety-reduction, anti-violence and gender- stereotyping components appear to promote positive results. Summary Social relationships between peers--- romantic or platonic--- have the potential to promote social skills in adolescents. Longitudinal and cross-sectional research has found associations between peer adolescent relationships and the development of emotionally positive interpersonal skills, autonomy, mental health, self-confidence, satisfaction with social support, joint decision-making, empathy, and more sophisticated perspective-taking and reasoning skills. These relationships appear to discourage aggression, emotional distress, and antisociality. Peers can also provide models of successful social relationships. One longitudinal study found no association between adolescent peer relationships and some of the preceding prosocial qualities, but the researchers admit that their measurement instruments may not have been valid. Certain 42 studies on romantic relationships have associated frequent dating with poor academic performance and depressive symptoms, but not with the decrease of any particular social skills. There are several different factors, such as characteristics from the individual, family, and neighborhood antecedent levels that appear to be important predictors of an adolescent's relationship with peers. In addition, there is reason to believe that certain programs may encourage, and enhance the quality of, these relationships. On the individual level, females appear to participate in more, and closer relationships with their peers than males do, though this trend becomes much less pronounced as adolescents age into adulthood. Certain other individual behaviors, such as socioemotional support and displays of affection, for example, also appear to promote successful peer relationships. Adolescents with who have been exposed to models of successful social interactions, and who are physically attractive, appear most likely to develop romantic peer relationships. Deviant adolescents appear to experience the effects of peer relationships differently than non-deviant adolescents; if they pursue relationships based on similar characteristics, often found in other deviant adolescents, they are likely to maintain or increase antisocial behavior. This outcome is much less likely to occur in relationships between non-deviant adolescents. The family also appears to influence greatly the quality of adolescent peer relationships. Various studies suggest that a warm, communicative, and strongly connected relationship with parents is associated with more positive exchanges and closer relationships between the youth and his or her peers. Communicative relationships with mothers, in particular, seem to be related to higher quality relationships with dating partners. Conversely, negative qualities in the parent- youth relationship, such as hostility, may be replicated in peer relationships and, subsequently, lead to their deterioration. Ironically, youth with insecure relationships with parents seem to invest in, and rely a great deal on, their romantic relationships with peers. Even the location of the adolescent's residence may influence the likelihood of such relationships. Youth appear to engage in relationships with peers who reside in their neighborhood, go to their school, or who are otherwise geographically accessible. There are very few programs that endeavor to establish and develop relationships between adolescents. There are certain programs, however, which teach various social skills that are necessary for the development and maintenance of peer friendships. Some of these programs have been experimentally evaluated, and appear to improve, indirectly, relationships between adolescents. There is also a paucity of programs that encourage the development of heterosocial relationships between adolescents, the majority of relevant literature having been written in the 1970's. At that time, skills replication, practice dating, desensitization, and other techniques appeared to aid in the initiation and development of peer romantic relationships. However, there has recently been a surge in anti-violence dating programs. These programs appear to change perceptions about gender roles and the perpetration of violence which, in turn, improves heterosocial relationships. While we have identified a number of antecedents for quality relationships with peers, there is still a need to expand the research in this area. For instance, the majority of the research on peers has focused on children in school. There is little to no research on the quality of peer relationships or even the composition of peer networks for older adolescents. In addition, the majority of the studies used cross-sectional and occasionally longitudinal designs to address their research question. As noted before, the causal nature of the associations for peer relationships and antecedents can not be determined with these types of study designs. However, from the 45 interpersonal skills domain includes conflict resolution, intimacy, and prosocial behaviors. The individual attributes domain includes self-control, social confidence, and empathy/sympathy. As mentioned previously, quality social relationships and good social skills are inter- related. In order for social skills both to develop and be perfected, youth must be able to connect, relate, and interact with others. If youth cannot connect, relate, or interact with others, then their sources of support and guidance are cut off, and their ability to progress successfully through adolescence is significantly compromised. Therefore, it is possible that quality social relationships may be listed as antecedents for good social skills. Likewise, social skills may be antecedents of social relationships. At this point it is not clear which comes first, whether they came first or if they develop in parallel or whether they develop in phases or sequences, each influencing the other in turn. Interpersonal Skills Conflict Resolution Skills Introduction The ability to solve conflicts fairly and consistently is fundamental to the maintenance and the growth of friendships. Results of a meta-analysis conducted by Newcomb and Bagwell (1995) reveal that adolescents instigate approximately the same amount of conflict with friends as they do with non-friends; however, it is primarily with friends that adolescents show a concern for resolving that conflict. Friendships may serve to hone conflict resolution strategies, but at the same time require the ability, at least at some level, to solve conflicts. If there is no ability to solve conflicts, then the friendship is likely to deteriorate (Newcomb & Bagwell, 1995). Both the extent to which the adolescent is able to communicate, and the manner in which he or she chooses to communicate with peers, teachers, parents, and other adults, affects his or her level of social success and development. For instance, an adolescent’s positive interpersonal communication skills are predictive of positive peer relationships and acceptance (Allen et al., 1989; Kurdek & Krile, 1982), positive relationships with adults, as well as the adolescent’s academic performance (Allen et al., 1989). Antecedents Individual A youth’s ability to successfully negotiate conflict situations may be influenced by their individual characteristics, personality, or behavior. Specifically, some of the research studying adolescent conflict resolutions strategies has focused on adolescent gender and personality as predictors of the youth’s choice of conflict resolution strategies. An adolescent’s gender may be associated to their choice of conflict resolution strategies. In a cross-sectional study of 65 white seventh-and eighth-graders (28 girls, 37 boys) examining the relationship between an adolescents’ social competence and communication strategies, Allen, Weissburg, and Hawkins (1989) found that girls tended to favor direct positive communication during social interactions more often than boys. Similarly, a two-week follow-up of 142 ethnically diverse students in grades four to six found that girls (n=62) tended to favor more 46 prosocial or passive conflict resolution strategies and that boys (n=80) tended to select more hostile/coercive strategies for dealing with conflict (T.-Y. Chung & Asher, 1996). An agreeable disposition may be associated with positive conflict resolution strategies. For example, adolescents who had rated themselves as warm, cooperative and trusting (high agreeable) at the beginning of one three-month study were more likely to engage in negotiation- and problem-solving behaviors during conflict than were adolescents who had rated themselves as cold, inconsiderate, and rude (low agreeable) (Graziano, Jensen-Campbell, & Hair, 1996). The sample consisted of 124 ethnically diverse, same-sex adolescent dyads, ages 17 –19. In addition, a longitudinal study of 167 African American, Mexican American, and Caucasian American adolescents in sixth through eighth grade, found that endorsement of power-assertion strategies during conflict was more likely in adolescents with lower levels of agreeableness than in adolescents with higher levels of agreeableness (Jensen-Campbell, Graziano, & Hair, 1996). In analyses of the same longitudinal study, Jensen-Campbell and Graziano (2001) found that adolescents with an agreeable personality style were more likely to compromise during conflicts with peers, less likely to walk away without addressing the problem, less likely to use physical force, and less likely to engage in esteem threatening remarks. Ultimately, these approaches to conflict resulted in higher rates of conflict resolution and friendship stability (Jensen-Campbell & Graziano, 2001). Family Characteristics of the family may be important antecedents for the development of conflict management skills. For instance, factors such as the parents’ use of conflict resolution strategies and the role that siblings play in the development of these skills may be important. Parents’ use of conflict resolution skills may influence the development of children’s conflict resolution skills. Katz and Gottman (1993) conducted a three-year longitudinal study of 56 primarily Caucasian American families (32 boys and 24 girls). When exposed to marital conflict, children acquired, through a process of observational learning, their parents' negative patterns of negotiating conflict and then displayed similar conflict-management strategies. For example, when husbands are angry and emotionally distant when resolving marital conflict, their children showed signs of anxiety and social withdrawal three years later (Katz & Gottman, 1993). Conflict resolution strategies formed in sibling relationships during childhood may be associated with conflict resolution strategies in adolescence. In a cross-sectional study of siblings, Rinaldi and Howe (1998) examined 34 fifth- and sixth-graders with a younger sibling. Results revealed that siblings tended to respond to each other in similar fashions (e.g., warm behaviors to warm behaviors, hostile behaviors to hostile behaviors). In addition, siblings who used constructive conflict resolution tactics had a positive resolution to the disagreement and a warm and intimate relationship with their sibling. The authors suggest that the long-term implication of the process of resolving the disagreement may be the development of a healthy and constructive set of conflict resolution strategies that could be used outside of the home with peers, romantic partners, teachers, and other adults. Conversely, when siblings used destructive conflict resolution tactics, the immediate outcome of the disagreement was a heightened level of animosity towards their sibling. The authors suggest that siblings with destructive conflict resolution strategies may assume that disengagement, non-resolution, and manipulative strategies are common and acceptable approaches to conflict (Rinaldi & Howe, 1998). 47 There is little research on the effects of poverty and homelessness on conflict management, though some studies suggest that homeless youth may be especially at risk for not developing the skills needed to manage conflict successfully (Gewirtzman & Fodor, 1987; Horowitz, Boardman, & Redlener, 1994). For example, in a cross-sectional study of 176 mother-adolescents dyads living New York City welfare hotels, homeless adolescents typically asked for more help from their mothers or teachers when dealing with peer conflict, and used more assertion during conflict situations than did adolescents from middle class neighborhoods (Horowitz et al., 1994). Peers Several cross-sectional studies have found that popular children are viewed as more conciliatory during conflict than rejected children, and that rejected children are seen as more coercive during conflict than other classmates (Bryant, 1992; T.-Y. Chung & Asher, 1996; French & Waas, 1987). In a cross-sectional study of 165 students in grades 4 through 6, children who were popular were more likely to resolve conflicts with a calm discussion and less likely to use tactics such as anger retaliation or avoidance. Rejected peers were more likely to use anger retaliation during conflict than other students (Bryant, 1992). Similarly, during playground observations of 26 third- and fourth-grade boys, French and Waas (1987) found that the rejected boys used more aggressive and less effective strategies during playground conflicts than other boys. Similarily, Chung and Asher (1996) found that youth who were “oriented toward having good relationships with peers” tended to select strategies that involved “giving in.” The authors suggest that this may be a conflict resolution strategy that will maintain or improve peer relationships. Since all of the studies examining the role of peers in the development of adolescent conflict management skills are cross-sectional, the direction of the association is not clear. It is possible that youth who use positive conflict resolution strategies with their peers (i.e., negotiation, calm discussion) may be viewed as popular. Conversely, youth who use negative conflict resolution strategies with their peers may be more likely to be rejected by their peers. Additional longitudinal research on the association between peers and conflict resolution strategies could help to clarify the direction of this association. Also, effective strategies may differ across social and cultural groups, a topic that has also received little attention. Societal There is evidence that cultural expectations and values may affect how children learn to manage conflict (Verbeek, Hartup, & Collins, 2000). For instance, a cross-sectional study of 526 Mexican children (4 – 12 years old) from three distinct cultures found differences in the children’s preferences for conflict resolution strategies (Kagan, Knight, Martinez, & Santana, 1981). Children from a small commercial town used more competitive strategies during conflict than children from an industrialized city or a small agricultural rural town. Children in the latter two communities used strategies similar to those used by Western middle class children. A cross-sectional study in southern Mexico of 48 3- to 8-year-olds provides additional support for these findings (Fry, 1988). Children from a competitive Zapotec community tended to be more aggressive during conflict than youth from a neighboring community that was considered generally peaceful. 50 designed to encourage and enhance the use of the social skills in the family environment. According to the evaluation, both parents and adolescents acquired improved conflict resolution, negotiation, and social skills due to ASSET training. Modest evidencesuggests that social skills enhancement is associated with significantly better interpersonal relationships in the family. Though, it should be noted that the participants in the treatment group may have been more motivated than the comparison to reduce conflict since they were willing to commit to the 10- week program. These program evaluations provide relatively strong evidence that the skills necessary to successfully negotiate conflict can be taught to youth. Programs that emphasize conflict resolution skills as one of their primary goals appear to be particularly able to affect change. Summary Although the majority of the research on conflict resolution skills is either cross-sectional or longitudinal, there is some evidence that development of these skills is key to an adolescent’s social success and development. The ability to communicate successfully and to resolve conflicts has been linked to peer acceptance and the development of friendships. The predictors of conflict resolution skills extend beyond individual characteristics to include family-, peer-, and societal-level factors. The development of conflict resolution skills may differ by an adolescent’s individual characteristics, such as gender and disposition. For instance, youth who perceive themselves as warm, considerate, and trusting use positive conflict resolution strategies, such as negotiation. Furthermore, development of conflict resolution skills may be associated with family- level characteristics, such as parents’ conflict resolution tactics during marital disputes and conflicts with siblings. Conflicts with siblings allow youth to “practice” techniques that can be used with a broader group. Conflict resolution skills have been linked to both peer social status and peer dominance hierarchies, though all of this work is cross-sectional and should be interpreted with caution. Youth who are perceived as popular are more likely to use constructive conflict resolution strategies, such as negotiation. There is some evidence that conflict resolution skills may be related to societal-level characteristics, such as cultural differences. Youth from more individualistic cultures, such as that of the United States, tend to use negative tactics such as power assertion more often than youth from other cultures. Research has revealed a number of antecedents related to the development of conflict resolution skills. These findings should be interpreted with some prudence, however, as the studies which generated them were mostly cross-sectional or longitudinal in nature. As stated before, the only kind of study that can determine an actual casual relationship is that of experimental design. However, certain experimental evaluations of the skills-training programs provide strong evidence on this topic: adolescents who lack the skills necessary to negotiate conflict can attain them through intervention programs that are specifically focused on conflict resolution training. 51 Intimacy Skills Introduction An adolescent’s ability to be emotionally intimate is associated with academic and socio- emotional adjustment. In a cross-sectional study of 255 adolescents (14 – 19 years old), it was found that an adolescent’s ability to be emotionally intimate with others was positively associated with interest in school, quality attachment with parents, high self-esteem, and family responsibility taking, and negatively associated with depression and risk-taking (Field et al., 1995). In a longitudinal study of 213 sixth-graders, friendships and peer acceptance were significantly associated with academic performance and social-emotional adjustment of youth in the eighth grade (Wentzel & Caldwell, 1997). The ability to be emotionally intimate with another person fosters one’s ability to connect and maintain relations with others. Results of a cross-sectional study of 207 adolescents suggests that adolescents with moderate to high levels of intimacy skills were more likely to pursue peer relations than were those adolescents with low levels of intimacy skills (Romig & Bakken, 1992). Field et al. (1995) found that levels of intimacy were positively associated with the number of same-sex adolescent friends. Intimacy skills have been associated with personal characteristics such as mental health. In a longitudinal study of 88 children, Rubin and colleagues found that children who were socially withdrawn in kindergarten were more likely to feel depressed, lonely, and have social difficulties such as poor peer acceptance, social isolation, and perceptions of social incompetence in the fifth grade. The teachers also rated these students as more anxious in social settings than other children (Hymel, Rubin, Rowden, & LeMare, 1990; K.H. Rubin & Mills, 1988). Further analyses of these children indicate that youth who were socially withdrawn in second grade are more likely to report loneliness, depression, lower levels of social-competence, and peer group isolation, and are more likely to have parental reports of internalizing behavior problems in the ninth grade (K. H. Rubin, Chen, McDougall, Bowker, & McKinnon, 1995). Antecedents Family Young people learn from interactions with their parents how to initiate and maintain satisfying, warm friendships. Engels et al. (2001) conducted a longitudinal study of 412 adolescents to explore whether social skills are the pathways through which parental attachment is associated with adolescent emotional adjustment. They found that higher quality parental relationships are associated with adolescent social skills, which, in turn, influenced the competence of older adolescents in friendships and romantic relationships (Engels et al., 2001). Similarly, in a longitudinal study of 67 families, a positive affective parental relationship with the child during preschool (age 4 –5) predicted higher levels of intimacy skills at age 12 (Estrada, Arsenio, Hess, & Holloway, 1987). A longitudinal study of 121 adolescents found that a mother’s verbal aggression toward her child in early- to middle-adolescence (ages 10 – 14) predicted lower intimacy in the child’s peer relationships in later adolescence (ages 16 –22) (Schlatter, 2001). Furthermore, in a 5 ½ year longitudinal study of 40 adolescents from intact families (21 females, 29 males), higher 52 levels of maternal firm control during adolescence were associated with less secure early adult romantic attachment (Jones, Forehand, & Beach, 2000). Mothers and fathers may serve different, but complementary, roles in socializing adolescents for peer relationships. Findings from a cross-sectional study of 386 13 – 20 year olds suggest that mothers engage more in teaching dependability and displaying feelings or intimacy, while fathers engage more in the elements of social play (Bigelow, Tesson, & Lewko, 1999). In a longitudinal study of 110 white adolescents, researchers found that living in a distressed family (i.e., high conflict between parents, parental divorce) lead to diminished self- esteem; this, in turn, predicted poorer quality adult relationships and diminished adult interpersonal competence (Armistead, Forehand, Beach, & Brody, 1995). Conversely, a different longitudinal study found that adolescents who experienced democratic, warm, and supportive parental relationships at ages 12 through 18 appeared to be better-adjusted adults five years later (i.e., higher self-esteem, higher satisfaction with friendships and love life, lower levels of irritability) (Aquilino & Supple, 2001). This investigation used data from the National Survey of Families and Households (NSFH) to study 1,066 adolescents over a 5-year period. In addition, a 15-year longitudinal study of 160 families found that good interactions in the family during early development foster social competence and intimacy skills in adolescence (age 14 – 15) (Aronen & Kurkela, 1998). Research has shown that sibling relationships may influence adolescent intimacy skills. Depending on the quality of the relationship, siblings can serve as a stepping stone for their brothers and sisters into the peer social world. In a cross-sectional study of 386 13- to 20-year- olds, Bigelow, Tesson, and Lewko (1999) found that when an adolescent has a sibling who is relatively close in age, their typically peer-like interaction style aids the development of social rule usage with peers, friends, and dating partners. Adolescents with siblings are more likely to use rules of compliance and social facilitation, to exhibit loyalty, and to use information management with close friends; they are more likely to use rules of compliance, to exhibit loyalty, and to engage in the management of feelings with dating partners. In short, an adolescent’s sibling serves as a testing ground in which the rules of interaction largely mirror the rules of peer, friendship, and romantic relationships. With a sibling, adolescents have the opportunity to practice and hone their social skills. In light of these benefits, it is excellent news that intimacy-related activities with a sibling (i.e., telling each other about a bad day; sharing private things) has recently been found to increase with age (Cole & Kerns, 2001). The study was cross-sectional in design, and based on a sample of 170 primarily white fourth-, sixth-, and eighth-graders. Peers Research has found an association between peer rejection in middle childhood and degree of adolescent adjustment at a later time. In a longitudinal study of 128 7- to 12-year-olds, youth who were socially withdrawn in middle childhood were more likely to be rated lonely one year later (Renshaw & Brown, 1993). Furthermore, in a seven-year longitudinal study of 207 third- to sixth-graders, youth with a positive peer reputation in middle childhood were rated has having a more active social life, and as being more socially accepted, more competent at dating, closer to friends, and more competent with peers (Morison & Masten, 1991). 55 Gibbs, 1996). In addition, a cross-sectional study of 58 third-graders found that parents who value prosocial behaviors tend to have children who also behave prosocially (Eisenberg et al., 1992). Neighborhood/Community/School The school environment, as well, may influence the development of prosocial behaviors. An observational study of 65 Israeli third-graders found that children in classes structured to promote prosocial behavior, by emphasizing cooperation, for example, helped their peers more than children in traditional classes (Hertz-Lazartowtiz, Fuchs, Sharabany, & Eisenberg, 1989). In addition, an observational study of 99 children in kindergarten through sixth-grade found that warm and supportive interactions with teachers were associated with prosocial behaviors and positive interactions with peers (Serow & Solomon, 1979). Societal Children from rural and semi-agricultural communities are more cooperative than children from urban or westernized cultures (see Eisenberg & Mussen, 1989, for a complete review). A cross-sectional study of 433 youth (5-22 years old) from Mexican villages and small towns, youth were more cooperative, helpful and more likely to avoid conflict than middle class Mexican (n=60), Mexican-American (n=155), or Anglo-American (n=215) children (Kagan, Knight, & Martinez-Romero, 1982). Similarly, a cross-sectional study of 120 Mexican American and Anglo American fourth- through sixth-graders found that the Mexican American children were more cooperative and helpful than the Anglo American children (Knight, Kagan, & Buriel, 1982). In addition, in a cross-sectional study of 20 groups of 9 – 11 year old children, children from kibbutzim in Israel were more helpful than their urban counterparts (Shapira & Lomranz, 1972). Eisenberg and Mussen (1989) argue that children from urban areas tend to be more competitive and less prosocial than children from other regions. Programs Though there are few evaluations of programs aimed at improving adolescent prosocial orientation, certain key findings have been produced. There is evidence, for example, that intervention programs can influence prosocial behaviors. The Child Development Project is a seven-year longitudinal study designed to promote prosocial behaviors (Battistich, Solomon, Watson, & Schaps, 1997; Battistich, Watson, Solomon, Schaps, & Solomon, 1991). The program provided opportunities for children to 1) engage in cooperative learning activities; 2) practice important social skills (i.e., understanding others’ thoughts and feelings; and 3) provide help to others. Teachers were trained to provide the main components of the intervention. Subjects include 2438 fifth- and sixth-graders assigned to the treatment group and 2321 students assigned to the comparison group. Children in the program experienced increases in prosocial behavior and prosocial moral reasoning. Summary Developing prosocial behaviors has been linked to developing other positive socio- emotional outcomes. Specifically, prosocial children are viewed as considerate, good social problem solvers, and low in aggression. 56 Antecedents from the individual, family, neighborhood, and societal levels are predictive of youth prosocial behaviors. For instance, personality characteristics of youth are associated with them. Youth who are resilient, warm, considerate, sociable, assertive, and not easily distracted are more likely to behave prosocially. In addition, family-level antecedents, such as a warm and supportive parenting style, positive discipline styles, and parental prosocial values are all associated with youths’ prosocial behavior. There is some evidence that the structure of the classroom may promote prosocial behaviors, as well. For instance, children in classrooms that emphasize cooperation tend to be more prosocial. Several cross-sectional studies indicate that cultural background may also influence the development of prosocial behaviors; youth from urban areas, for example, may be less cooperative and less prosocial than youth from rural areas or cultures. The review of the research on the prosocial development of adolescents has produced a number of antecedents. However, most of the studies were cross-sectional, therefore only allowing us to make tentative statements about the association between the antecedents and the development of prosocial behaviors. There is a large collection of research on the prosocial behaviors of younger children (see Eisenberg & Fabes, 1998), but a dearth of research on the prosocial behaviors of adolescents. Longitudinal research on adolescent prosocial development would greatly enhance this area. There is also a lack of experimental program evaluations for this topic. However, the one study described above provides some initial evidence that programs have the capacity to influence youths’ prosocial behaviors. Individual Attributes Interpersonal skills, alone, are not fully representative of adolescent social skills. Individual attributes, such as self-control/behavior regulation, empathy or sympathy, and social confidence, are also associated with adolescent social competence. While interpersonal skills are skills that foster social connection, individual attributes are personal characteristics adolescents typically value in others. Subsequently, adolescents with these valued attributes are, themselves, valued and appreciated by their peers and family. These individual attributes foster relationships through their positive effects on others’ perceptions of adolescents, and, therefore, are forms of social competence. Self-Control/Behavior Regulation Introduction The ability to regulate behavior and emotions at a level appropriate for any given interaction, whether with peers or adults, is highly predictive of relationship success. For example, a cross-sectional study of over 440 ethnically diverse, Midwestern seventh-graders found that friendship attainment strategies used by early adolescents that centered on physical and psychological aggression, fighting, and yelling were significantly less likely to result in peer acceptance (Wentzel & Erdley, 1993). These strategies are all behaviors indicative of poor impulse control and regulatory abilities. Another cross-sectional study revealed that, out of the 120 predominantly white, sixth-grade males sampled, those adolescents rated by teachers as hyperactive and aggressive were more likely to be rejected by peers (Pope, Bierman, & Mumma, 1989). Similarly, cross-sectional research on 79 white, fourth- through fifth-graders suggested that those who scored higher in emotional regulation were more likely to be viewed positively by 57 peers (Melnick & Hinshaw, 2000). Finally, longitudinal research of 145 white, low- to middle- income, early adolescent males showed that youth who exhibited impulsive, hyperactive behavior were more likely to have peer difficulties concurrently and later in adolescence (Pope & Bierman, 1999). Developmentally appropriate regulatory abilities are also positively associated with parent-child relations. In a longitudinal study, Murphy, Shepard, Eisenberg, Fabes, and Guthrie (1999) found that, out of the 64 predominantly white adolescents of varied socio-economic class sampled, youth rated as high in regulatory abilities by adults were more likely to be rated as sympathetic than were youth rated as low in regulatory ability by adults. The authors assert that levels of sympathy and regulatory abilities relate to each other and that they both promote interpersonal relations. They suggest that adolescents high in regulatory abilities are less likely to be overwhelmed by potentially stressful situations, which allows them to focus on another’s distress rather than their own. This sympathetic reaction fosters an intimate connection with others, which, in turn, fosters the maintenance of that relationship (Murphy, Shepard, Eisenberg, Fabes, & Guthrie, 1999). Antecedents Individual The majority of research in this domain focuses on the influence of others, such as parents or peers, on a given focal child’s self-control and behavior regulation. There is some evidence, however, that particular characteristics of the child, him- or herself, may also influence the development of these qualities. Successes in school and level of intelligence have been linked to adolescent regulatory abilities. Research on 411 urban males from the United Kingdom, ages 8 to 32, found that both lower levels of non-verbal intelligence and limited academic success during junior high and high school are predictive of regulation and self-control problems during adolescence (Farrington, 1989). Additionally, Wentzel et al. (1993) found that a young adolescent’s level of intelligence is negatively associated with poor regulatory abilities. Beyond intelligence and academic performance, personality traits have also been linked to adolescent regulatory abilities. Research by Wentzel et al. (1993) found that prosocial behavior and appropriate friendship-making strategies are negatively associated with anti-social behavior and poor regulatory abilities. In addition, Pope and Beirman (1999) found, in a sample of 145 boys, that youth with an irritable-inattentive personality were more likely to exhibit poor regulatory abilities (both emotional and behavioral) during adolescence than their peers. Finally, it should be noted that poor regulatory abilities, similar to those exhibited by children diagnosed with attention deficit hyperactivity disorder (ADHD), can be the result of a hormonal imbalance (Biederman et al., 1995; Conners, 1997). Often, it is ineffective to simply discipline a child for his or her aggressive, overly active behavior. Experimental research conducted by The National Institute of Mental Health found that the combination of medication and behavioral treatments is more effective in improving these behaviors than just medication or just behavior treatment (National Institute of Mental Health, 1999). Further, the study found that in certain cases, where medication or behavior therapy alone yielded little to no improvement in behavior, the combination of behavioral therapy and medication lead to behavioral improvements. 60 as having higher levels of self-control than members of control group. The sample consisted of 137 Native-American adolescents (Schinke, Orlandi, Botvin, Gilchrist, & et al., 1988). Additionally, a cognitive-behavioral training program--- focused on the reduction of aggression and the improvement of social-control--- proved very effective for behavior- disordered adolescents. The program consisted of 12 half-hour sessions over three weeks, during which participants received cognitive-behavioral training and positive reinforcement for learning and using key skills. Sessions were designed to have clear objectives and structured lectures. In post-test data from the program’s experimental evaluation, program participants were rated as having significantly more self-control and observed as having significantly fewer aggressive behaviors than youth in the control group. The study sample consisted of 30 behavior-disordered adolescents, equally divided between the treatment and control groups (Etscheidt, 1991). Lochman, Coie, Underwood and Terry (1993) developed and empirically tested a social relations intervention program that addressed social problem-solving, positive play training, group-entry skill training, and dealing with strong negative feelings. The program was conducted over 26 individual half-hour sessions and 8 half-hour sessions of group-play. In a sample of 52 African-American early-adolescent children, youth who participated in the training group exhibited decreased levels of aggression towards their peers when compared to youth from the control group (Lochman, Coie, Underwood, & Terry, 1993). The Linking the Interests of Parent and Teachers (LIFT) program, mentioned earlier, successfully lowered levels of adolescent aggression during peer interaction. An experimental study of LIFT evaluated the impact of the program on delinquent behaviors of 600 first- and fifth-graders from high juvenile crime neighborhoods. The program targets certain youth social behaviors, such as opposition, deviance, and social ineptitude, and parenting practices, such as disciplining and monitoring (Eddy et al., 2000). Results indicate that participants in the LIFT program were rated by teachers as less aggressive towards peers during play and social interaction than those in the control group. Empirical research suggests that Project Northland, a 3-year training program consisting of weekly training sessions or group activities, effectively lowers levels of anti-social behavior, and fosters the ability to regulate behavior and resist peer pressure. Through joint parent-child training sessions and community organizing, Project Northland aims to increase participants’ bonding, self-efficacy, prosocial involvement, and social, emotional, and behavioral competencies. An experimental study found that students who participated in the program were more likely to report increased communication with parents, a heightened ability to resist peer pressure, increased levels of self-efficacy, and lower levels of cigarette, alcohol, and marijuana use. The sample consisted of 1900 junior high students (Perry et al., 1996). Summary Beyond the influence of genetics or hormonal imbalances, mentioned above, self-control and behavioral regulation are related to the success of relationships with peers, adults and parents. For example, youth who can regulate their behaviors and emotions are more likely to be viewed positively by peers and adults and less like to have difficulties with their social relationships. However, the research literature on the development self-control and behavior regulation suggests that a developmentally appropriate level of self-control is dependent on, or reflects, the quality of relationships in which an adolescent participates. Positive relationships with parents and peers lead to appropriate levels of self-control, whereas negative relationships with parents 61 and peers lead to a lack of self-control. Furthermore, it appears that the ability to self-regulate is not solely dependent on present relationships; it may be, largely, the consequence of past relationships with parents and peers during early and middle childhood. Most adolescents, through adaptive relationships, develop the ability to self-regulate their behavior. However, it is important to recognize that the association between self-control and quality relationships may be recursive; that is, self-control may promote quality relationships just as quality relationships may foster self-control in adolescents. In addition, neighborhood characteristics such as poverty, unemployment, and public assistance may be antecedents for the development of self-control. It is important to note, again, that the majority of the studies on self-control were correlational and/or cross-sectional. These study designs do not allow for causality to be determined. As described above, it is possible for antecedents, such as quality relationships, to not only predict self-control but also to be influenced by self-control. However, for adolescents without a positive source of influence for their development of self-regulation and control, certain training programs may successfully fill the void. Programs that teach social problem- solving skills and coping and monitoring strategies, and that train youth to consider the consequences of behavior, appear to succeed, at least partially, at imparting methods of self- control and behavior regulation. Social Confidence: Assertiveness/Social Initiative/Social Self-Efficacy Introduction The positive effects of three social constructs--- specifically, social assertiveness7, social self-efficacy8, and social initiative9--- have been found to largely overlap with one another. While each construct has distinct qualities, they are similar in that they are each an outgrowth of one’s confidence in his or her ability to connect to, and be liked by, others. According to longitudinal research based on a sample of 205 Midwestern, lower- to middle-class junior high students varying in race/ethnicity, social assertiveness is associated both concurrently and over time with higher levels of peer acceptance and friendship attainment (Shiner, 2000). Cross-sectional research on 243 predominantly white, middle-class early adolescents has shown that youth who reported lower levels of social assertiveness also reported higher levels of loneliness and introversion (Young & Bradley, 1998). A longitudinal study on the social self-efficacy of 793 suburban Canadian tenth-graders revealed that adolescents with higher levels of social self-efficacy were more likely to feel socially accepted than were adolescents with lower levels of social self-efficacy (McFarlane, Bellissimo, & Norman, 1995). A cross-sectional study found that adolescents who scored low on social self-efficacy also reported higher levels of loneliness, social dissatisfaction, and social discomfort. The data was based on a sample of 238 Greek early adolescents of middle-class background (Galanaki & 7 Defined as assertiveness as measured during social activities or situations Jaminon, C., & Desmette, D. (1998). Path to integration: Interaction between training schemes and individual variables. Unpublished manuscript, Scotland.. 8 Defined as the expectation of success in different social tasks that contribute to success in social relationships McFarlane, A. H., Bellissimo, A., & Norman, G. R. (1995). The role of family and peers in social self-efficacy: Links to depression in adolescence. American Journal of Orthopsychiatry, 65(3), 402-410.. 9 Defined as a particular form of adolescent competence that indexes the degree to which adolescent initiate social contacts outside the home (Barber and Erikson, 2001) 62 Kalantzi-Azizi, 1999). According to longitudinal research of 900 Northwestern early adolescents of varied race/ethnicity, levels of social initiative and social self-efficacy are positively associated with levels of peer acceptance and harmonious parent child relations (Barber & Erickson, 2001). Finally, a longitudinal investigation revealed that, of 236 early adolescents sampled, those with low levels of social self-efficacy and social assertiveness were more likely to report high levels of loneliness later in adolescence (Moore & Schultz, 1983). Antecedents Individual Individual characteristics associated with global self-confidence have been found to be associated with social confidence. Cross-sectional research on 131 Northeastern, white college- age adolescents of varyied socio-economic background (Filsinger & Anderson, 1982), 242 suburban high-schoolers of varied race/ethnicity (Connolly, 1989), and participants in Barber and Erikson’s (2001) longitudinal study, described above, shows that social assertiveness, social self-efficacy, and social initiative are linked to levels of self-esteem. Gender is also associated with social self-efficacy, as females tend to report higher levels than males (Connolly, 1989). Finally, levels of aggression and delinquent behavior are negatively associated with both social assertiveness and social initiative. Longitudinal research by Barber and Erikson (2001) found that anti-social, aggressive youth exhibit lower levels of social initiative and social assertiveness. One theme present in the research on social confidence is that of the negative influence of adolescent attributional style. Individuals who consistently attribute social failures to internal, stable factors, and attribute social success to unstable, external factors, tend to score lower on levels of social self-efficacy, social initiative, and social assertiveness. Cross-sectional research has revealed that, in 93 Australian high-school students sampled, adolescents who reported low levels of social self-efficacy and social assertiveness also tended to have the maladaptive attributional styles mentioned above (Innes & Thomas, 1989). Similarly, a more recent study, based on data from 220 college-age adolescents, found that lonely adolescents who were more likely to report lower levels of social initiative and social assertiveness were also more likely to disparage themselves after social failures than were non-lonely persons (Christensen & Kashy, 1998). Due to the cross-sectional nature of the studies discussed above, it is difficult to determine whether one’s attributional style influences his or her level of social confidence, one’s level of social confidence affects his or her attributional style, or whether there is a common underlying antecedent. Family The quality of interaction between adolescents and their families has a direct association with levels of social assertiveness, social self-efficacy, and social initiative. The amount of social support received from both parents and siblings is positively associated with both higher levels of social self-efficacy during early and late adolescence (McFarlane et al., 1995), and with higher levels of social assertiveness and social initiative during late adolescence (Barber & Erickson, 2001). Barber and Erikson (2001) concluded that ongoing, quality relationships with parents and/or siblings promote the development of a healthy and confident relational/interpersonal style in adolescents. Similarly, McFarlane et al. (1995) suggest that familial social support provides a secure environment in which adolescents can develop interpersonal confidence, a necessary quality for successful adolescent peer interaction. Cross- 65 communication and problem-solving with both peers and parents (e.g., LIFT, SIS, and SLT) also seem to foster social assertiveness, social self-efficacy, and social initiative. Empathy/Sympathy Empathy is defined as an “affective response that stems from the apprehension or comprehension of another’s emotional state or condition, and which is identical or very similar to what the other person is feeling or would be expected to feel” (Eisenberg, in press). In contrast, sympathy is the affective response that can stem from empathy, and consists of feelings such as sorrow and concern for the distressed or needy (Eisenberg, in press). Numerous studies explicate the positive effects of empathic skills on relationships with others. Longitudinal research conducted by Adams, Schvaneveldt, and Jenson (1979) found, among the 80 Western, rural seventh- and eighth-graders sampled, that empathic skills in the form of social sensitivity are precursors to effective social interaction. With respect to peer relationships, one longitudinal study found that empathic adolescents are more likely to be well- liked by their peers (Murphy et al., 1999), and cross-sectional research has found association between social knowledge about emotional states and peer popularity (Adams, 1983). Similarly, cross-sectional research has found that sensitive and empathetic adolescents may be more likely to have supportive, satisfying, lower-conflict friendships (Phillipsen, 1999). Finally, longitudinal research conducted by Davis and Oathout (1987) found that the level of adolescent empathy among 264 heterosexual couples sampled predicted increased satisfaction in intimate relationships. In order to appreciate how a certain social context facilitates empathy development, empathy, itself, and the manner in which researchers define it should be addressed. The ability to respond empathically requires both the level of intellect necessary to put one’s self in another’s shoes, and the emotional maturity necessary to correctly deduce the emotional state of another (Pecukonis, 1990) Empathy then is the outgrowth of both cognitive and affective processes (Adams, Schvaneveldt, & Jenson, 1979; Henry, Sager, & Plunkett, 1996). Antecedents Individual As outlined above, researchers believe that empathic responses are rooted in both cognitive and affective processes. There are numerous individual characteristics that both foster and hinder the development of empathy. Experimental research, on 24 lower- to middle-class aggressive adolescent females who participated in an empathy training program, revealed that treatment group participants who had healthy ego10 development were more likely to benefit from the program than were treatment group participants who had poor ego development (Pecukonis, 1990). Results of a cross-sectional study examining the effects of religiosity on empathy development found that levels of religiosity, especially among late adolescents, were positively associated with levels of adolescent empathy. The sample consisted of 569 Western, rural high-schoolers from a predominately middle class background (Francis & Pearson, 1987). Research by Adams et al. (1979) revealed that adolescents high in social cooperativeness were more likely to exhibit empathetic behavior. Similarly, cross-sectional research by Henry, Sager 10 A healthy ego is a personality trait characterized as a person’s ability to monitor his/her own needs and wants against the needs of others in order to maximize relationship success (Pecukonis, 1990). 66 and Plunkett (1996) found that, among 149 Southwestern adolescents of varied race/ethnicity, communicative responsiveness was positively associated with empathic response. Finally, according to correlational research on 198 early adolescents, youth who reported reading more had higher scores on empathy and sympathy (Van der Bolt & Tellegen, 1994). Though limited, research also suggests that adolescents high in emotional regulatory ability also exhibit higher levels of empathy over time. Longitudinal research conducted by Murphy et al. (1999) found, among the 65 fifth- and sixth-graders sampled, that those youth rated as high in regulatory abilities were more likely to exhibit or use empathic responses with peers, as measured six years later. Murphy et al. (1999) concluded that, throughout development, regulatory abilities most likely prevent adolescents from becoming overwhelmed when they experience another’s distress; this allows them to focus on the other’s distress rather than their own. Researchers have also found qualities that relate inversely to adolescent empathy development. In a cross-sectional study, Ritter (1979) found that egocentrism was negatively associated with perspective taking or empathy. The sample consisted of 40 white, suburban high-schoolers. Henry et al. (1996) found that high levels of self-esteem were associated with low levels of empathic response. Finally, a cross-sectional study of 172 late adolescents revealed that adolescent aggression and empathy appear to be inversely related (Mehrabian & Epstein, 1972). With respect to gender, a plethora of studies show that adolescent girls exhibit higher levels of empathy than do adolescent boys (Adams et al., 1979; Henry et al., 1996; R. Miller, 1990). Henry et al. (1996) found that girls scored higher in both the cognitive and emotional aspects of empathy, and concluded that both societal and familial socialization tended to encourage girls to be more empathic than boys. Miller (1990), while conducting cross-sectional research on a sample of 112 urban, white, middle-class high-schoolers, attributed girls’ superior empathic skills to their ability to use empathic response with both same-sex and opposite-sex peers. Boys, on the other hand, generally use empathy with opposite-sex peers but refrain from using it with same-sex peers. As outlined above, a host of individual characteristics impact adolescent empathy development. Adolescents with a healthy ego, high levels of religiosity, a sense of social cooperativeness, and high regulatory abilities are more likely to exhibit empathy. Conversely, adolescents who are egocentric and aggressive are less likely to exhibit empathy. Finally, gender is also related to empathy response; both sexes exhibit higher levels of empathy towards the opposite sex, and girls tend to exhibit higher levels of empathetic response in general. Family As the family is a key source of influence for the developing adolescent, it should come as no surprise that there is abundant research on various family dynamics and their influence on adolescent empathy development. A variety of research studies suggest that cohesive, supportive families facilitate adolescent empathy development. For example, research by Henry et al. (1996) found that adolescents who described their family as high in cohesiveness were more likely to report regard and sympathy for the feelings of others; adolescents who saw their parents as supportive also perceived themselves as higher in empathic concern. Henry et al. (1996) conclude that families that are generally supportive, loving, and/or close, foster adolescent empathy development in multiple ways. Not only do these families provide an excellent training 67 ground for empathy skills, but they also establish a precedent for the use of empathic response, which the adolescent will refer to during later interaction with peers and friends. Cross-sectional research on male adolescent empathy development found that sympathetic mothers served as empathic models, which promoted both their son’s identification with them and their son’s development of empathetic responses. The sample consisted of 72 white mother-late adolescent dyads from upper to middle class backgrounds (Eisenberg-Berg & Mussen, 1978). Simply having an older sibling has also been linked to adolescent empathy development. Longitudinal research conducted by Tucker, McHale and Crouter (2001) of 197 families with one or more adolescent found that siblings influence adolescent empathy development in numerous ways. Overall, the authors concluded that adolescents with older siblings tend to exhibit higher levels of empathy when compared to those adolescents without an older sibling. Further, beyond the mere presence of an older sibling, the quality of the relationship between the older and younger sibling, as well as the older sibling's gender, can affect the younger sibling's empathy development. Specifically, Tucker et al. (2001) found that younger sisters had greater empathy when their older siblings were more empathic, that younger sisters of older brothers were more empathic than were younger sisters of older sisters, and that younger sisters reported the highest levels of empathy when older brothers displayed more positive behaviors and when their sibling relationship could be characterized as highly positive. Similarly, even in female adolescent-older sister relationships that were characterized as more negative than positive, the younger sister reported greater levels of empathy than did adolescent females without an older sibling. With respect to younger brothers, the authors found that both having a more empathic older brother and having a positive sibling relationship were linked with development of empathy; in fact, simply having an older sibling was positively linked to greater empathy development for males. Unlike younger sisters, younger brothers tended to focus only on older brothers as role models (Tucker, McHale, & Crouter, 2001). (Tucker et al., 2001). Research also suggests that children who are exposed to models of logical reasoning by their parents are more likely to exhibit empathy at younger ages. Henry et al. (1996) found that adolescents who witness their parents using logical reasoning to solve problems are more likely to report higher levels of perspective taking, and a greater ability to understand the situation of another person. The authors conclude that adolescents of parents who engage in logical reasoning become socialized to use cognitive processes when trying to understand another individual’s feelings or emotional state (Henry et al., 1996). In sum, adolescents with cohesive, supportive families, an older sibling, and parents who engage in logical reasoning tend to exhibit higher levels of empathic response. Peers Research on the influence of peer relationships on empathy development is limited. According to available information, the association between peers and empathy seems to be reciprocal. As noted above, adolescents higher in empathy are at an advantage with respect to peer relations (Davis & Oathout, 1987; Hay, 1994). The converse is also true. That is, there is research to suggest that levels of peer interaction foster the development of empathy. Research by Ritter (1979) found that an adolescent’s level of empathy in a given social exchange is dependent on whether or not the other person is a friend or just a peer. Adolescents are more likely to engage in empathic responses with their friends than with peers. Thus, according to the author, relationships with friends foster the use and development of empathic responses (Ritter, 1979). Research conducted by Miller (1990) found that social interaction with the opposite sex 70 and empathy. Similarly, having parents and siblings who use constructive conflict resolution strategies is related to the development of conflict management skills in adolescents. Peer-level antecedents Peer acceptance was documented to be an important antecedent for several social skills. For instance, being accepted by peers is associated with conflict management skills, intimacy skills, self-control, and social confidence. However, since the vast majority of the studies in the peer antecedent level are cross-sectional, it is equally plausible that the social skills could be antecedents of peer acceptance and peer relationships. Neighborhood/Community-level antecedents Neighborhood-level antecedents were only documented for three social skills: prosocial behaviors, self-control, and self-confidence. For prosocial behaviors, classrooms that promote prosocial behaviors, as compared to traditional classrooms, foster prosocial development in the students. Several neighborhood characteristics, such as residential instability and poverty, are associated with behaviors (i.e., delinquency and anti-social behaviors) that are indicative of low self-control. In addition, frequent contact with community members promotes social assertiveness and social initiative. Societal-level antecedents Research on the influence of societal- or cultural-level characteristics on the development of social skills is scarce. However, research on conflict-resolution skills and prosocial behaviors suggests that societal-level antecedents play a role in the development of these skills. For instance, youth from individualistic cultures (i.e. the United States) use negative conflict resolution tactics, such as power assertion, more often than youth from other cultures. In addition, several studies indicate that a youth’s cultural background may influence the development of prosocial skills. For instance, youth who live in urban areas tend to be less cooperative and less prosocial than youth from rural areas. Programs There are a variety of programs that have been experimentally evaluated that have attempted to improve adolescents’ social skills. Strong evidence exists that youth programs aimed at increasing youths’ conflict management skills, self-control/behavioral regulation, self- confidence, and empathy appear to have been successful. However, little evidence exists of successful programs that attempted to impact adolescents’ intimacy skills or prosocial behaviors. Additional research needs to be conducted on these two important social skills. Research Quality Throughout the section on the antecedents of good social skills, we have discussed an array of possible antecedents. It is crucial to understand that the majority of the findings on the development of good social skills are based on cross-sectional and longitudinal studies. As discussed throughout this report, these study designs do not allow research to assess the direction of the association between two variables. Especially in cross-sectional designs, it equally likely 71 that the social skill predicts the antecedent, as it is that the antecedent predicts the social skill. However, experimental studies suggest that when specific social skills (i.e., conflict resolution strategies) are targeted by a program, improvements in these skills can be accomplished. 72 CONCLUSIONS AND RECOMMENDATIONS Quality social relationships and good social skills are important aspects of people’s lives. For instance, adolescents who are socially competent are likely to have good psychological well- being, have good academic outcomes, maintain successful marriages in adulthood, and have positive relationships with their own children. In contrast, those adolescents who have deficits in their social competency are likely to engage in problem behaviors, delinquency, substance use, and high-risk sexual behaviors, as well as have poor psychological well-being and academic outcomes. During childhood and adolescence, youth should develop the skills or competence to maintain quality relationships. To gain a better understanding of how youth gain the skills necessary to engage in and maintain relationships, we examined the antecedents of, and intervention strategies for, quality social relationships and good social skills. Summary for the Antecedents of Quality Social Relationships Based on experimental, quasi-experimental, and multivariate longitudinal studies, individual-, familial-, and neighborhood-level variables were documented as antecedents of quality social relationships. For instance, individual characteristics of the youth, such as gender, age and personality, were found to be important for predicting the quality of relationships, including parent, sibling, non-familial adults, and peers. In addition, several family-level characteristics are important for predicting quality relationships. A positive relationship between the parent and youth, parenting style, and low family discord/parental divorce are important antecedents for quality social relationships. Neighborhood characteristics such as accessibility and proximity to non-familial adults and peers are important factors for quality relationships. Experimentally-evaluated programs that have attempted to increase the quality of adolescents’ social relationships are scarce. However, some programs, such as mentoring, have been able to promote quality social relationships (i.e., with parents, mentors, or peers) for adolescents. In addition, education and skills training programs appear to be effective at increasing adolescents’ skills when a deficit has been detected (i.e., conflict management skills with siblings). Summary for the Antecedents of Good Social Skills As mentioned above, antecedents from the individual-, family-, peer-, and neighborhood- level were found to be important predictors for good social skills. Specifically, youth who are warm, considerate, and friendly are likely to use more positive conflict-resolution strategies and to behave in a prosocial manner. Other individual characteristics (sociability, non-verbal intelligence, self-esteem) were found to be important for specific social skills. The most consistent family-level characteristic for predicting good social skills was warm and responsive parenting. In addition, youth with siblings typically have better social skills. Peer acceptance was documented as an important predictor of many social skills; however, since most of the research is cross-sectional, it is equally likely that the social skills could be antecedents to the quality of the peer relationships. There is a solid set of literature for experimentally-evaluated programs that have attempted to increase adolescents’ social skills. When a program has targeted a specific skill (i.e., conflict management skills, self-control/behavioral regulation, or self-confidence) that an 75 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental FAMILY RELATIONSHIPS Parent-Child Relationship Programs: Mentoring Big Brothers Big Sisters (BBBS): one-on-one mentoring program for youth. (t=480, c=479) -Treatment sample as a whole experienced an increase, significant at the .05 level, in quality of the parent-child relationship. -White male participants experienced particularly significant increases in comparison to the white males in the control group: Parent-child relationship quality: 5% Reported level of trust: 7% Individual-level Significant degree of respect in parent-child relationships. Attachment to parents, but particularly the father, during childhood. Discouragement of qualities such as anxiety, bullying, or a quick temper in the adolescent. Limited participation in dating and heterosocial relationships. Peer-level In cases of divorce or single parenthood, parental coresidence with a partner to whom they are married, for whom the adolescent has affective feelings, and with whom the parent has a positive relationship. Parental and parent-figure employment, particularly in a high-quality, satisfying job. Parental religiosity. 76 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental Parents, especially mothers, who offer socioemotional support, displays of affection, and appropriate power-sharing; who share similar interests and emotional needs with their child; and who employ cooperative social skills and problem-solving skills. Continuously adjusting the parent-child relationship to accommodate adolescents’ changing socio-developmental needs (such as more responsibility, autonomy, and a more peer-like parent-child rapport.) Authoritative parenting style (warm, communicative, responsive, firm and consistent with discipline). Not employing “negative parenting” behavior, such as spanking ,slapping, or yelling at the adolescent. Parenting which encourages independent problem-solving and the acquisition of new and challenging social skills. 77 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental Family-level Parent-child attachment beginning early in the child’s life. Avoiding parental divorce. Minimal family arguments, stress, and general conflict; promotion of love, fun, teamwork, and family cohesion. In the case of divorced or separated families, shared child custody, or particular attention to maintaining the child’s relationship with the non- custodial parent. Particular attention to parent- child relationship if there are young siblings or adult relatives residing with the family. Neighborhood-level Living in neighborhoods perceived to be of good quality; neighborhoods with sufficient educational resources. 80 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental Family-level Non-hostile parents or parenting styles. High levels of parent- child and inter-parent affect. Consistent, non-differential parenting of both/all siblings. Employed mothers, particularly in the case of male siblings. Programs: Education/Training for Parents Positive reinforcement of cooperative play; use of time-outs in response to fighting. Reduction of conflict in the family environment through modification of parenting style. Grandparent and Other Family Member Relationships Programs: Education/ Training for Grandparents Education of grandparents on perspectives and experiences of other generations in family. Encouragement for grand- parents to become more influential in lives of grand- children. Individual-level Same-gender grandparent-grand- child relationships; particular attention to opposite-gender relationships. Particular attention to boys’ relationships with family members, as they often have fewer family relationships than girls do. 81 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental Family-level Active negotiation and development of the grandparent- grandchild relationships by both parties. In most cases, avoidance of parental divorce. Positive parent-grandparent relationships. Particular attention to non-parent family member relationships for non-African American youth. Programs: Education/Training for Grandparents Filial/Family Play Therapy (FFPT). Training to create an accepting, non-judgmental environment in custodial grandparent’s home, and to become a source of positive influence. Multimodal Home Based Intervention for Custodial Parents Program. Provides social work, nursing, and legal aid to promote psychological stress, improve physical and mental health, and 82 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental increase the custodial grandparent’s level of social support and resources. NON-FAMILY RELATIONSHIPS Non-Familial Adult Relationships Individual Particular attention to boys’ development of non-family adult relationships, as they often have fewer than girls. Particular attention to non-familal relationships for non-African American youth. Particular attention to developmental needs according to gender: Optimal teacher-child relationships are characterized by low levels of conflict and dependency for boys, and by closeness for girls. Mentor-like relationships that occur naturally between a child and an adult in his or her life, especially if established at an early age. 85 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental Programs: Anti-violence Training The Expect Respect anti-bullying, anti-sexual harassment, anti- gender violence program. It employs a “whole school” approach, as well as parent involvement, to establish a universal understanding of, and response to, this kind of violence. Peer Relationships: Romantic Relationships Programs: Social Skills Training Anxiety-reduction through desensitization training and/or replication skills training. Programs: Anti-Dating Violence Safe Dates Project: Intervention consisted of role- playing, a poster contest, and a curriculum on violence, gender stereotyping and conflict management. Development of victim services available in the community. (N=1,892) A month after the end of the program, participants in the Individual-level Physical attractiveness. Previous exposure to successful heterosocial exchanges and functional heterosocial relationships. Family-level Communication with parents, particularly mothers, about romantic relationships. Parents who have remained married or who otherwise model functional romantic relationships. Programs: Social Skills Training Practice dating, counseling, self- reinforcement, behavioral rehearsal, “heterosexual group discussion”, participant modeling, 86 Table 1. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Relationships Social Relationships What Works ------------------------------------ What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" Non-experimental treatment group reported committing 60% less violence against their romantic partner than the those control group. . self-observation by videotape, desensitization, sensitivity training, behavior-skills-training, sexual education, and “cognitive modification”. Program Symbols ASSET BBBS FFPT ISFP Adolescent Social Skills Effectiveness Training Big Brothers Big Sisters Filial/Family Play Therapy Iowa Strengthening Families Program 87 Table 2. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Skills Relational Skill Areas What Works ------------------------------- What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" (Non-experimental) INTERPERSONAL SKILLS Conflict Resolution Skills Programs Linking the Interests of Families and Teachers (LIFT): intervention involving parent training and child- behavior modification program. (t=300, c=300) Adolescent Transitions Program (ATP): program that focuses on both improving parent management skills and developing the adolescent’s goals/limit setting ability, peer supports, and problem solving ability. (t=105, c=38) Anger Coping Program: a program designed to increase adolescent perspective taking, social problem solving and social skills for managing conflict situations. (t=39, c=13) Individual-level Agreeable disposition. Family-level Siblings with constructive conflict strategies. Parents with constructive strategies to resolve marital conflicts. Peer-level Peer acceptance. Ranking high in classroom social hierarchies. Programs Reciprocal Social Skills training for delinquent adolescents and their parent(s). Adolescent Social Skills Effectiveness Training (ASSET). Social skills training program aimed at reducing parent-child conflict. 90 Table 2. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Skills Relational Skill Areas What Works ------------------------------- What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" (Non-experimental) Warm and supportive interactions with teachers. Societal-level Growing up in a rural community or small, non-urban town. INDIVIDUAL ATTRIBUTES Self-Control/behavior regulation Programs BB/BS: mentoring (see earlier description) The Bicultural Competence Skills Program (BCSP) is a program that promotes competence and positive identity in the bicultural adolescent through skills training. (N=137) Cognitive/behavioral training program that provides skills training to promote competence. (t=15, c=15) Social relations intervention program that addresses social problem solving, positive- Individual-level Non-verbal intelligence and academic success. A combination of medication and behavioral treatments for youth with a hormonal imbalance or other health-related behavior problems. Family-level Authoritative parenting style. Absence or a minimum of corporal punishment and conflict in the parent-child relationship. Peer-level Peer acceptance. 91 Table 2. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Skills Relational Skill Areas What Works ------------------------------- What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" (Non-experimental) play training, group entry skill training, and dealing with negative feelings. (t=26, c=26) LIFT (see earlier description) Project Northland: Program that use youth skill and parent competence training. (t=1450, c=1450) Friendships with non-delinquent peers. Neighborhood/Community Adaptive neighborhood dynamics such as residential stability, structural advantage, and low concentration of recent immigrants. Social Confidence: Social Assertiveness, Social Self- Efficacy, and Social Initiative. Programs LIFT program (see earlier description) Say it Straight (SIS): Social skills training. (t=1564, c=1295) Structured Learning Training (SLT): Social reinforcement/ behavioral modeling program aimed at fostering assertion skills. (t=45, c=45) Bicultural Competence Skills Program: A program that provides skills training to promote competence and positive identity. Participants Individual-level Development of self-esteem, rehabilitation of certain manifestations of low self- esteem. Minimal aggression and antisociality in behavior. Family-level High quality relationships and a high level of social support from parents and/or siblings. Peer-level Peer acceptance and peer support. Friend with high social status. 92 Table 2. Review of the the Research Literature and Implications for Targeted Activities to Improve Adolescent Social Skills Relational Skill Areas What Works ------------------------------- What Doesn't Work -----------Experimental------------ Mixed Reviews ------------------------------- "Best Bets" (Non-experimental) were rated higher on assertiveness than control group youth. Project Northland (see earlier description) Participation in school-related activities. Interpersonal peer interaction. Neighborhood/Community Frequent contact with community members, neighbors, and church leaders. Empathy/Sympathy Individual-level Healthy ego development, especially for special populations, such as aggressive teens. Programs Cognitive/Affective Empathy Training (CAET ): a progressive training program aimed at improving adolescent empathetic response. (t=12, c=12) Individual-level Religiosity. The development of certain characteristics, such as social cooperativeness, communicative responsiveness, and the ability to regulate one’s emotions. Minimal egocentrism and aggression. Family-level A cohesive, supportive family. Parents who express sympathy for others. The presence of an older sibling. 95 Battistich, V., Watson, M., Solomon, D., Schaps, E., & Solomon, J. (1991). The Child Development Project: A comprehensive program for the development of prosocial character. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development: Vol 3. Application (pp. 1-34). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Bearman, P., & Bruckner, H. (1999). Peer effects on adolescent sexual debut and pregnancy: An analysis of a national survey of adolescent girls. In S. Philliber (Ed.), Peer Potential: Making the Most of How Teens Influence Each Other (pp. 7-27). Washington, DC: The National Campaign to Prevent Teen Pregnancy. Bender, D., & Loesel, F. (1997). Protective and risk effects of peer relations and social support on antisocial behaviour in adolescents from multi-problem milieus. Journal of Adolescence, 20(6), 661-678. Benson, P. L. (1993). The troubled journey: A portrait of 6th-12th graders. Minneapolis, MN: Search Institute. Berner, M. L., Fee, V. E., & Turner, A. D. (2001). A multi-component social skills training program for pre-adolescent girls with few friends. Child & Family Behavior Therapy, 23(2), 1-18. Biederman, J., Faraone, S. V., Mick, E., Spencer, T., Wilens, T., Kiely, K., et al. (1995). High risk for attention deficit hyperactivity disorder among children of parents with childhood onset of the disorder: A pilot study. American Journal of Psychiatry, 152(3), 431-435. Bigelow, B. J., Tesson, G., & Lewko, J. H. (1999). The contextual influences of sibling and dating relations on adolescents' personal relations with their close friends, dating partners, and parents: The Sullivan-Piaget-Hartup hypothesis considered. New Directions for Child and Adolescent Development, 84, 71-86. Blyth, D. A., Hill, J. P., & Thiel, K. S. (1982). Early adolescents' significant others: Grade and gender differences in perceived relationships with familial and nonfamilial adults and young people. Journal of Youth & Adolescence, 11(6), 425-450. Boykin, A. W. (1986). The triple quandary and the schooling of Afro-American children. In U. Neiser (Ed.), The school achievement of minority children. Hillsdale, NJ: Erlbaum Publishers. Bratton, S., Ray, D., & Moffit, K. (1998). Filial/family play therapy: An intervention for custodial grandparents and their grandchildren. Educational Gerontology, 24(4), 391- 406. Brody, G. H., Stoneman, Z., & Burke, M. (1987). Child temperaments, maternal differential behavior, and sibling relationships. Developmental Psychology, 23(3), 354-362. Brown, B. B., & Theobald, W. (1999). How peers matter: A research synthesis of peer influences on adolescent pregnancy. In S. Philliber (Ed.), Peer Potential: Making the Most of How Teens Influence Each Other (pp. 27-80). Washington, DC: The National Campaign to Prevent Teen Pregnancy. Bryant, B. K. (1992). Conflict resolution strategies in relation to children's peer relations. Journal of Applied Developmental Psychology, 13, 35-50. Buhrmester, D. (1990). Intimacy of friendship, interpersonal competence, and adjustment during preadolescence and adolescence. Child Development, 61(4), 1101-1111. Buhrmester, D., & Furman, W. (1990). Perceptions of sibling relationships during middle childhood and adolescence. Child Development, 61(5), 1387-1398. 96 Bukowski, W. (in press). Peer relations and well-being. In M. H. Bornstein, L. Davidson, C. L. M. Keyes, K. A. Moore & The Center for Child Well-being (Eds.), Well-being: Positive development across the life course. Mahwah, NJ: Erlbaum. Burks, V. S., Dodge, K. A., & Price, J. M. (1995). Models of internalizing outcomes of early rejection. Development & Psychopathology, 7(4), 683-695. Burton, C. B., & Krantz, M. (1990). Predicting adjustment in middle childhood from early peer status. Early Child Development & Care, 60, 89-100. Cairns, R. B., & Cairns, B. (1994). Lifelines and risks: Pathways of youth in our time. New York: Cambridge University Press. Campbell, J. J., Lamb, M. E., & Hwang, C. P. (2000). Early child care experiences and children's social competence between 1.5 and 15 Years of Age. Applied Developmental Science, 4(3), 166-175. Caplan, M., Weissberg, R. P., Grober, J. S., Sivo, P. J., Grady, K., & Jacoby, C. (1992). Social competence promotion with inner-city and suburban young adolescents: Effects on social adjustment and alcohol use. Journal of Consulting and Clinical Psychology, 60, 56-63. Carnegie Council on Adolescent Development. (1989). Turning points: Preparing American youth for the 21st century. Washington, DC: Author. Carnegie Council on Adolescent Development. (1992). A matter of time: Risk and opportunity in the non-school hours. Washington, DC: Author. Chatters, L. M., Taylor, R. J., & Jayakody, R. (1994). Fictive kinship relations in black extended families. Journal of Comparative Family Studies, 25(3), 297-312. Cherlin, A., & Furstenberg, F. F. (1986). Grandparents and family crisis. Generations, 10(4), 26- 28. Christensen, P. N., & Kashy, D. A. (1998). Perceptions of and by lonely people in initial social interaction. Personality & Social Psychology Bulletin, 24(3), 322-329. Christoff, K. A., Scott, W. O., Kelley, M. L., Schlundt, D., Baer, G., & Kelly, J. A. (1985). Social skills and social problem-solving training for shy young adolescents. Behavior Therapy, 16(5), 468-477. Chung, H., & Elias, M. (1996). Patterns of adolescent involvement in problem behaviors: Relationship to self-efficacy, social competence, and life events. American Journal of Community Psychology, 24(6), 771-784. Chung, T.-Y., & Asher, S. R. (1996). Children's goals and strategies in peer conflict situations. Merrill-Palmer Quarterly, 42(1), 125-147. Clingempeel, W. G., Colyar, J. J., Brand, E., & Hetherington, E. M. (1992). Children's relationships with maternal grandparents: A longitudinal study of family structure and pubertal status effects. Child Development, 63(6), 1404-1422. Coates, D. L. (1987). Gender differences in the structure and support characteristics of black adolescents' social networks. Sex Roles, 17(11-12), 667-687. Cole, A., & Kerns, K. A. (2001). Perceptions of sibling qualities and activities of early adolescents. Journal of Early Adolescence, 21(2), 204-226. Collins, W. A., Laursen, B., Mortensen, N., Luebker, C., & Ferreira, M. (1997). Conflict processes and transitions in parent and peer relationships: Implications for autonomy and regulation. Journal of Adolescent Research, 12(2), 178-198. Collinsworth, P., Strom, R., Strom, S., & Young, D. (1991). The Grandparent Strengths and Needs Inventory: Development and factorial validation. Educational & Psychological Measurement, 51(3), 785-792. 97 Conger, K. J., Conger, R. D., & Elder, G. H., Jr. (1994). Sibling relations during hard times. In R. D. Conger & G. H. Elder, Jr. (Eds.), Families in troubled times: Adapting to change in rural America (pp. 235-252). New York: Hawthorne. Conger, K. J., Conger, R. D., & Scaramella, L. V. (1997). Parents, siblings, psychological control, and adolescent adjustment. Journal of Adolescent Research, 12(1), 113-138. Conger, R. D., & Rueter, M. A. (1996). Siblings, parents, and peers: A longitudinal study of social influences in adolescent risk for alcohol use and abuse. In G. H. Brody (Ed.), Sibling relationships: Their causes and consequences (pp. 1-30). Stamford, CT: Ablex Publishing Corp. Conners, C. K. (1997). Is ADHD a disease? Journal of Attention Disorders, 2(1), 3-17. Connolly, J. (1989). Social self-efficacy in adolescence: Relations with self-concept, social adjustment, and mental health. Canadian Journal of Behavioural Science, 21(3), 258- 269. Cox, M. (in press). Parent-child relationships. In M. H. Bornstein, L. Davidson, C. L. M. Keyes, K. A. Moore & The Center for Child Well-being (Eds.), Well-being: Positive development across the life course. Mahwah, NJ: Erlbaum. Coyne, J. C., & DeLongis, A. (1986). Going beyond social support: The role of social relationships in adaptation. Journal of Consulting & Clinical Psychology, 54(4), 454-460. Creasey, G. L. (1993). The association between divorce and late adolescent grandchildren's relations with grandparents. Journal of Youth & Adolescence, 22(5), 513-529. Crockett, L., & Losoff, M. (1984). Perceptions of the peer group and friendship in early adolescence. Journal of Early Adolescence, 4(2), 155-181. Csikszentmihaly, M., & Larson, R. (1984). Being Adolescent. New York: Erlbaum. Curran, J. P. (1977). Skills training as an approach to the treatment of heterosexual-social anxiety: A review. Psychological Bulletin, 84(1), 140-157. Curran, J. P., & Gilbert, F. S. (1975). A test of the relative effectiveness of a systematic desensitization program and an interpersonal skills training program with date anxious subjects. Behaviour Therapy, 6, 510-521. Cutting, E., & Tammi, L. (1999). Understanding parents, understanding parenthood: An education for parenthood course piloted at Monifeith High School, Angus. Scotland, Edinburgh: Save the Children. Darling, N., & Steinberg, L. (1993). Parenting style as context: An integrative model. Psychological Bulletin, 113(3), 487-496. Davis, M. H., & Oathout, H. A. (1987). Maintenance of satisfaction in romantic relationships: Empathy and relational competence. Journal of Personality & Social Psychology, 53(2), 397-410. Dekovic, M., & Janssens, J. M. A. M. (1992). Parents' child-rearing style and child's sociometric status. Developmental Psychology, 28, 925-932. Dishion, T. J., & Andrews, D. W. (1995). Preventing escalation in problem behaviors with high- risk young adolescents: Immediate and 1-year outcomes. Journal of Consulting & Clinical Psychology, 63(4), 538-548. Dishion, T. J., Loeber, R., Stouthamer-Loeber, M., & Patterson, G. R. (1984). Skill deficits and male adolescent delinquency. Journal of Abnormal Child Psychology, 12(1), 37-53. Dowdy, B. B., & Kliewer, W. (1998). Dating, parent-adolescent conflict, and behavioral autonomy. Journal of Youth & Adolescence, 27(4), 473-492.