Peer ratings of childhood aggression: Relation to social skills, problem behaviors, and peer acceptance

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)


Teaching and Learning

First Committee Member

Sharon Vaughn - Committee Chair


This study examined the aggressive behaviors of children through peer ratings. The relations between peer ratings of aggression to teacher ratings of problem behaviors and social skills and peer ratings of friendship were examined. Subjects were 310 students in first and second grade. A peer rating scale of aggressive behavior items (CAPERS), derived from teacher ratings and peer nominations (Dodge & Coie, 1987), was developed then evaluated for reliability, item intercorrelation, and factor structure. All items highly intercorrelated. Results of principal axis analyses indicated items converging as a single factor structure for aggression.The CAPERS was examined in relation to other measures: peer ratings of friendship (Singleton & Asher, 1977); teacher ratings of problem behavior on conduct disorders and anxiety withdrawal (Revised Behavior Problem Checklist: Quay, 1987); and, teacher ratings of social skills (Revised Social Skills Rating Scale for Teachers; Vaughn & Hogan, 1990). Significant positive correlations were found between peer ratings of aggression and least liked friendships. Peer aggression ratings and teacher ratings of conduct disorders were significantly positively correlated. No significant correlations were found between peer ratings of aggression and teacher ratings of anxiety withdrawal. The relation between peer aggression ratings and teacher ratings of social skills showed significant negative correlations. The hypothesis that sex differences would consistently exist in the CAPERS factor structure and in relation to other measures was not confirmed.Aggressive behaviors as rated by peers in first and second grade children appear stable across at least a one year span. Aggressive peer ratings are also related to teacher measures of conduct disorders and peer ratings of children who are least likely to be chosen as friends. Although it was expected that different dimensions of childhood aggression would be distinguished in children's peer groups, this finding was not supported in the results of this study. Limitations of this study and implications for future research are addressed.


Education, Elementary; Education, Educational Psychology; Education, Special

Link to Full Text


Link to Full Text