[A comparison of self- and peer-reports of relational aggression in elementary school children]

A number of different methods for assessing relational aggression have been utilized in previous research. It has been suggested that scores on self-report measures of relational aggression especially in children and adolescents are not associated with scores on other types of measures, suggesting that peer-report measures are more accurate. Nevertheless, few prior studies have directly compared self-and peer-reports. The purpose of the current study was to investigate which type of report is superior. We first developed a new questionnaire to measure relational aggression by self and peers. Elementary school children (grades fourth to sixth; n=243) completed this questionnaire. In addition, their teachers nominated high and low relational-aggressive children in their classes. The two reports were different in terms of correlational analyses, but were almost the same with regards to group-related analyses using analyses of variance (ANOVAs). Moreover, girls' peer-reports were more similar to the teacher nominations than they were to their self-reports, and boys also showed a similar trend. We discuss how self- and peer-reports should be used in future studies.