Background Schools have a duty of care to prevent violence between students and play an important role in establishing prosocial norms and behaviours. However, a significant amount of dating-related violence (DRV) and gender-based violence (GBV) occurs in schools. DRV and GBV are important public health issues which are receiving increasing interest from educational practitioners and policy makers. Given that DRV and GBV share common risk factors and mechanisms, interventions that address both can meet a growing policy and practice need. Understanding the theories of change that underpin these interventions offers additional insights into the characteristics of effective interventions by uncovering underlying mechanisms.

Methods Systematic Review. A systematic review of school-based interventions for preventing DRV or GBV was carried out. As part of this review, we synthesised theories of change for the outcome evaluations, a subset of the total included studies. Data were extracted on basic study information (e.g. location, intervention description) and on theory (e.g. reference to existing theories, theoretical assumptions as to how intervention should work). Theoretical data were synthesised using meta-ethnographic methods to produce an overall theory of change for school-based interventions that prevent DRV and GBV.

Results 81 studies provided theoretical descriptions of 52 interventions. We used Markham & Aveyard’s theory of human functioning and school organisation to inform our understanding of the mechanisms of change, and two overarching themes, ‘strengthening relationships’ and ‘increasing belonging’ formed the core of our overarching theory of change. The first theme focused on strengthening relationships between the school and surrounding communities, and within the school, between teachers and students, and between students themselves. The second theme focused on increasing belonging by involving students in their own learning and in school-level decisions which in turn, promoted positive interactions and relations between students and teachers and among the students. Through these multiple and related mechanisms, interventions were theorised as intending to provide students with knowledge and skills to avoid and reduce engagement in DRV & GBV, within a context of interactive and supportive learning, that promoted pro-social attitudes and values to underpin behaviour change, and encouraged a sense of security and connection to the school.

Conclusion This theory synthesis offers an insight into the mechanisms of change through which the interventions are expected to work. Many interventions targeted change across multiple levels and we hypothesised that these interventions had the greatest potential of activating the widest complement of change mechanisms.