The effectiveness of longitudinal intervention programs for adolescent cyberbullying and its influencing factors: A three-level meta-analysis

Cyberbullying poses a significant threat to the mental health of adolescents; however, the effectiveness of longitudinal intervention programs and their influencing factors have not been systematically evaluated. This study, based on the General Aggression Model (GAM), employed a three-level meta-analysis to examine 33 longitudinal cyberbullying intervention programs (105 effect sizes, N = 20,050). The results indicated that current interventions targeting cyberbullying among adolescents (aged 10-19) had a significant yet small effect size (g = -0.169). Interventions were most effective at the behavioral level, showing no significant improvements in cognition or psychological states. Additionally, the effectiveness of interventions was moderated by factors such as the proportion of females, cultural background, publication type, experimental design, program type, intervention strategy, intervention method, and intervention duration. Specifically, interventions were more effective in Western cultural contexts, unpublished studies, non-RCT designs, evolutionary programs, group counseling strategies, offline intervention methods, and when the intervention duration was >=4 months. Furthermore, the number of female participants negatively predicted intervention outcomes. This study clarifies the factors influencing the effectiveness of longitudinal interventions on the mental health of various actors in cyberbullying, offering theoretical and methodological foundations for future research.