Abstract
This study evaluated the structure and validity of the Problem Behavior Frequency Scale-Teacher Report Form (PBFS-TR) for assessing students' frequency of specific forms of aggression and victimization, and positive behavior. Analyses were conducted on two waves of data from 727 students from two urban middle schools (Sample 1) who were rated by their teachers on the PBFS-TR and the Social Skills Improvement System (SSIS), and on data collected from 1,740 students from three urban middle schools (Sample 2) for whom data on both the teacher and student report version of the PBFS were obtained. Confirmatory factor analyses supported first-order factors representing 3 forms of aggression (physical, verbal, and relational), 3 forms of victimization (physical, verbal and relational), and 2 forms of positive behavior (prosocial behavior and effective nonviolent behavior), and higher-order factors representing aggression, victimization, and positive behavior. Strong measurement invariance was established over gender, grade, intervention condition, and time. Support for convergent validity was found based on correlations between corresponding scales on the PBFS-TR and teacher ratings on the SSIS in Sample 1. Significant correlations were also found between teacher ratings on the PBFS-TR and student ratings of their behavior on the Problem Behavior Frequency Scale-Adolescent Report (PBFS-AR) and a measure of nonviolent behavioral intentions in Sample 2. Overall the findings provided support for the PBFS-TR and suggested that teachers can provide useful data on students' aggressive and prosocial behavior and victimization experiences within the school setting. (PsycINFO Database Record (c) 2018 APA, all rights reserved).