The purpose of this longitudinal study was to identify subgroups of rejected boys with different developmental pathways of aggression and prosocial behavior across the elementary school years. Peer, teacher, and parent reports and behavior observations yielded composite scores for aggression and prosocial behavior at 3 measurement waves. A cluster analysis with these composites on 87 initially rejected boys identified 4 subgroups with different developmental pathways of prosocial behavior and aggression that were associated with different patterns of sociometric acceptance and rejection over time and with social emotional adjustment in the last measurement wave. Changes in acceptance and rejection tend to precede changes in aggression and prosocial behavior. Cluster differences on social emotional adjustment indicators converged into I moderately discriminating factor, Social Maladaptation in Peer-Oriented Behavior.

The importance of relationships to human health and happiness has been recognized for a very long time. Nevertheless, scientific efforts to describe, assess, and establish the significance of these entities have emerged slowly over the last 100 years. Deep-seated and widely held prejudices having to do with the impropriety of studying something so intimate and private as close relationshipsincluding relationships between children and their parents-still constrain progress. Research methods in this area also remain relatively unsophisticated, tempering support for scientific studies both within the behavioral sciences and the natural sciences more generally (Berscheid & Peplau, 1983; Hinde, 1997). Considerable progress, however, has been made since the mid-1960s.

This chapter presents the case study of Dan, an eleven-year-old fifth-grader in a Boston public elementary school, who is Identified by his teachers and classmates as the biggest bully in the school. Both the external and internal data were subjected to two kinds of analyses, a qualitative examination of themes and a quantitative analysis of developmental levels in the components of the friendship framework. The interview data indicate that Dan does not identify himself as a bully, despite his quite sophisticated understanding of the role of fighting in fifth-graders' interpersonal relationships. Jimmy, Dan's partner in pair counseling, discussed Dan's fighting in the school context several times in the course of his fighting interview. In terms of developmental analysis, Dan showed an uneven pattern across the three psychosocial domains of interpersonal understanding, skills, and personal meaning, but these patterns …

The nature and significance of peer conflict is discussed within a social relational framework derived from theories of social exchange and emotional investment. Overall, the rewards of exchanges with close friends must exceed the costs. Because relationships with friends and romantic partners are voluntary, close peers must be especially wary about conflict and handle interactions in a manner to preserve rewards in future exchanges. When conflict occurs, close peers are less likely than parents, siblings, and classmates to display coercive, angry tactics that harm relationships and diminish rewards. Evidence indicates that close peers' conflict management styles help to avert the threat that disagreement poses to their relationships. In other relations, conflict management styles reflect power differentials and personal gain.

Considerable evidence tells us that ¿being liked¿ and ¿being disliked¿ are related to social competence, but evidence concerning friendships and their developmental significance is relatively weak. The argument is advanced that the developmental implications of these relationships cannot be specified without distinguishing between having friends, the identity of one's friends, and friendship quality. Most commonly, children are differentiated from one another in diagnosis and research only according to whether or not they have friends. The evidence shows that friends provide one another with cognitive and social scaffolding that differs from what nonfriends provide, and having friends supports good outcomes across normative transitions. But predicting developmental outcome also requires knowing about the behavioral characteristics and attitudes of children's friends as well as qualitative features of these relationships.

Disagreements between school-aged children were examined as a function of friendship status. 66 same-sex dyads were selected, including equal numbers of "best friends" and nonfriends, who were then observed while playing a board game (a closed-field situation). Conflicts occurred more frequently among friends than among nonfriends and lasted longer. Friends did not talk more during their conflicts than nonfriends, but assertions were used selectively according to friendship and sex: With friends, girls used assertions accompanied by rationales more frequently than boys whereas boys used assertions without rationales more frequently than girls. These sex differences were not evident during conflicts between nonfriends. Results are discussed in relation to the social constraints intrinsic to closed-field competitive conditions as they apply to friendship relations in middle childhood.

Many studies examining the effects of treatments for socially isolated/withdrawn children have used behavioral measures to assess children's peer relations. In an attempt to examine the concurrent validity of these measures, we observed 258 preschool children during free play and coded their interactions into the categories of positive, negative, and neutral behaviors. We also interviewed these children individually, using a sociometric nomination procedure, and asked them to indicate liked and disliked peers. Our examination of the results revealed that, consistent with other research, the correlation between the sociometric measures and the children's total rate of interaction with peers was low. In addition, measures of positive, negative, and neutral behaviors were also only weakly correlated with the children's sociometric scores. We agree with others in concluding that multiple methods of assessment are needed to properly assess the adequacy of children's peer relations.

BROWNELL, CELIA A., and HARTUP, WILLARD W. Indeterminate and Sequential Goal Structures in Relation to Task Performance in Children's Small Groups. CHILD DEVELOPMENT, 1981, 52, 651-659. Task performance was studied in triads of third-grade children exposed to 4 instructional conditions, each consisting of 3 phases. During the first and third phases all groups received promotive instructions (equal rewards given to each participant). During the second phase triads received 1 of 4 types of instructions: individualistic (rewards allocated proportionate to individual performance), contrient (winner take all), no reward (all rewards suspended), and undesignated (participants told that rewards would be allocated, but not to whom). Performance under the no reward condition resembled performance under standard promotive instructions, while performance under the undesignated condition resembled that under standard individualistic instructions. This differentiation is discussed in relation to the immediate experimental context (cooperatively oriented) and the larger context of the school (individualistically oriented). There was no carry-over of second-phase task strategies into third-phase performance.

24 socially withdrawn preschool children were located through classroom observations and assigned to 3 conditions: (a) socialization with a younger child during 10 play sessions, (b) socialization with an age mate during a similar series of sessions, and (c) no treatment. The socialization sessions, particularly those with a younger partner, were found to increase the sociability of the withdrawn children in their classrooms. Significant increases occurred mainly in the rate with which positive social reinforcement was emitted. Generally, the results support a leadership deficit theory of social isolation. Possible mechanisms responsible for the observed changes are discussed.

NEWCOMB, ANDREW F.; BRADY, JUDITH E.; and HARTUP, WILLARD W. Friendship and Incentive Condition as Determinants of Children's Task-oriented Social Behavior. CHILD DEVELOPMENT, 1979, 50, 878-881. Firstand third-grade children, together with either a friend or a nonfriend (N = 88), were exposed to cooperative and competitive goal structures to assess the influence of friendship and incentive condition on task performance and social interaction. Children's performance was rewarded sequentially as follows: phase 1-shared rewards; phase 2-"winner take all" or proportional rewards; and phase 3-shared rewards. Friendship facilitated the expressive and reciprocal components of social interaction regardless of superordinate goal structure. Friends and nonfriends did not differ in terms of performance outcome.

LOUGEE, MICHAEL D.; GRUENEICH, ROYAL; and HARTUP, WILLARD W. Social Interaction in Sameand Mixed-Age Dyads of Preschool Children. CHILD DEVELOPMENT, 1977, 48, 1353-1361. Social interaction was studied in pairs of preschool children in which the individuals differed in chronological age by either 2 months (same-age dyads) or 16 months (mixed-age dyads). Videotape records were made of the children's behavior during 2 10-min play periods. Social interaction and verbal communication was least frequent in younger same-age dyads, intermediate in mixed-age dyads, and most frequent in older same-age dyads. The appropriateness of the verbal behavior varied as a function of age mixture in this same manner. Mean lengths of utterance units, however, were similar in mixedand same-age older dyads. Individual accommodation to the developmental status of the companion was indicated, but this accommodation was more consistent in relation to the appropriateness of the communication than to the amount of social interaction. The results indicate that normative data derived from observations of same-age interaction may not be generalized to the mixed-age situation and that the two types of social experience may serve different functions in children's development.

GRAZIANO, WILLIAM; FRENCH, DORAN; BROWNELL, CELIA A.; and H.ARTUP, WILLARD W. Peer Interaction in Sameand Mixed-Age Triads in Relation to Chronological Age and Incentive Condition. CHILD DEVELOPMENT, 1976, 47, 707-714. Task performance was compared in sameand mixed-age triads of firstand third-grade children. Half the triads worked for individual rewards, while the remainder worked for group rewards. Considered as units, the mixed-age triads performed as well as the same-age triads, although there was more conversation in the latter. Considered in terms of individual performance, mixed-age conditions produced different effects at the 2 grade levels. Mixed-age conditions stimulated greater task activity than same-age conditions among the third graders, but not among the first graders. Performance was better with group than with individual rewards, but the effects of age mixture did not depend on incentive condition.