Teacher factors contributing to implementation of the KiVa anti-bullying program

Date
2016
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
As concern regarding bullying and victimization in schools has grown, the need for efficacious anti-bullying programs has become of paramount importance. The KiVa Anti-Bullying Program (KiVa) seeks to meet this need through a school-based, teacher-led intervention for elementary-school children. Despite the important role of teachers in this program, little is known about how teacher implementation is related to program outcomes. In addition, questions exist regarding teacher factors that predict implementation. This study aims to examine how teacher implementation impacts KiVa outcomes and how teacher factors influence implementation in the context of KiVa. Participants included 74 teachers and 1409 students in nine elementary schools participating in the KiVa Program. Teachers and students completed data collection at the beginning and end of the school year, including measures of bullying and victimization, correlates of victimization (depression, anxiety, somatization, peer rejection, withdrawal, and school avoidance), hypothesized intervention mediators (anti-bullying attitudes, empathy toward victims), bystander behaviors, and teacher factors thought to relate to implementation (self-efficacy for teaching, professional burnout, perceived principal support, and perceived intervention acceptability). In addition, the dosage of KiVa delivered to classrooms was measured throughout the school year. Results highlight dosage as an important predictor of change in bullying, victimization, socioemotional correlates, bystander behavior, and hypothesized intervention mediators. Of the teacher factors, professional burnout uniquely predicted intervention dosage. Moreover, a comprehensive structural equation model including professional burnout, dosage, and child-level outcomes demonstrated good fit, suggesting that the theoretical model linking these constructs fits the data well. Implications for intervention design and implementation are discussed.
Description
Keywords
Citation