Abstract

Bystander-based violence prevention interventions have shown efficacy to reduce dating violence and sexual violence acceptance at the individual level yet no large randomized controlled trial (RCT) has evaluated this effect at the high-school level and over time. This rigorous cluster-randomized controlled trial addresses this gap by evaluating intervention effectiveness at both school and individual levels. Kentucky high schools were randomized to intervention or control conditions. In intervention schools educators provided school-wide 'Green Dot' presentations and bystander training with student popular opinion leaders. Each spring from 2010 to 2014; 73,044 students completed anonymous surveys with no missing data on relevant outcomes. Dating violence and sexual violence acceptance were the primary outcomes for this analysis. At the school level, slopes from linear mixed models using averaged school-level dating violence acceptance (condition-time, p < 0.001) and sexual violence acceptance (condition-time interaction, p < 0.001) differed indicating a significant reduction in the violence acceptance in the intervention relative to control schools over time and specifically in years 3 and 4 when 'Green Dot' was fully implemented. Analyses based on student's self-reported receipt of 'Green Dot' training by condition confirmed the school level finding of significant reductions in both dating violence and sexual violence acceptance in years 3 and 4 for both males and females. In this RCT we find evidence that the bystander-based violence prevention intervention 'Green Dot' works, as hypothesized and as implemented, to reduce acceptance of dating violence and sexual violence at the school and individual levels.

Original languageEnglish
Pages (from-to)153-164
Number of pages12
JournalJournal of Family Violence
Volume34
Issue number3
DOIs
StatePublished - 2019

Fingerprint

Dive into the research topics of 'Bystander Program Effectiveness to Reduce Violence Acceptance: RCT in High Schools'. Together they form a unique fingerprint.

Cite this