An examination of teacher and peer perceptions of subtypes of antisocial behavior and their relation to peer status

The purpose of the study was to determine if empirically-derived dimensions would emerge from teacher and peer ratings of subtypes of childhood aggression. In addition, the relationship between these subtypes and peer status was examined. Drawing upon research on subtypes of childhood aggression and antisocial behavior, a 28-item rating scale was developed assessing proactive aggression (characterized by a planned, purposeful quality), reactive aggression (characterized by an overreactive, defensive quality), covert antisocial behavior (behavior not involving direct confrontation, characterized by a hidden, clandestine quality), and prosocial behavior. Teacher and peer ratings were completed on 135 third- through fifth-grade boys. Responses on the aggression items were entered into a principal components analysis which indicated that teachers discriminated between reactive aggression and proactive aggression. These factors demonstrated adequate levels of internal consistency and independence. One factor emerged from the peer data, suggesting that the students were unable to differentiate among the subtypes of antisocial behavior. Factor structure relations analysis indicated that the proactive aggression factor was similar to the peer factor, while the teacher reactive factor was not. Regression analysis indicated that the three empirically-derived dimensions combined to account for 22% of the variance in predicting the number of negative nominations received by peers. When entered alone into a regression analysis, the two teacher aggression dimensions combined to account for 15% of the total variance in the prediction of negative nominations. Limitations of principal components analysis, and advantages of common factor analysis are discussed. Due to the selectivity of the sample, generalizations were limited to school districts comprised of white, low-middle income families. Results of this study are compared to those of previous research and the need to replicate the study\u27s findings is discussed