Publication - Research and analysis

What works to prevent youth violence: evidence summary

This report draws together high-quality international evidence about what works to prevent youth violence, to inform policymakers and practitioners about the evidence base and effectiveness associated with different approaches and interventions.

70 page PDF

1.4 MB

70 page PDF

1.4 MB

What works to prevent youth violence: evidence summary
Executive Summary

70 page PDF

1.4 MB

Executive Summary

Research aims and overview

Youth violence, which occurs between individuals aged 10 to 29, can take many forms and has health, social, and economic consequences for individuals, families and communities (World Health Organisation, 2015). When considering figures relating to young people’s involvement in violence in Scotland alongside concerns that the indirect social and economic consequences of the COVID-19 pandemic may lead to an increase in youth violence (Irwin-Rogers, Muthoo, & Billingham, 2020), it is clear that there is a pressing need to better understand which strategies can be implemented to address violence in youth. This report was undertaken to draw together high-quality international evidence about what works to prevent youth violence and is intended to inform policymakers and practitioners about the extant evidence base and effectiveness associated with different approaches and interventions.

Key Findings

  • There is evidence to suggest that school and education-based approaches are effective in reducing youth violence. These include both bullying prevention programmes (e.g. Olweus Bullying Prevention Program, KiVa) and social and emotional learning programmes (e.g. PATHS).
  • Interventions that have been identified as promising include: school-based programmes which seek to prevent violence in dating and intimate partner relationships, parenting and family-focused approaches, mentoring programmes, and community-based coalitions.
  • There is mixed evidence about the effectiveness of out-of-school activities and early childhood home visitation programmes.
  • Deterrence and fear-based approaches have been identified as having no effect on youth violence outcomes and, at worse, are potentially harmful to young people.
  • Due to a limited body of evidence, it is not yet possible to draw reliable conclusions on the effectiveness of programmes that specifically aim to prevent gang involvement and subsequent gang violence. As a result, the evidence is inconclusive.

Moderating factors: Key Findings

Across this report, the importance of accounting for the moderating factors, potential facilitators, and potential barriers for prevention interventions for youth violence have been highlighted where evidence is available. Accounting for these factors can encourage effective implementation of these evidence-based interventions.

According to the Early Intervention Foundation the “key principles of effective programmes” for preventing youth violence include:

  • Strategies that seek to create positive changes in the lives of youth and/or their families, as well as reduce risk factors and prevent negative outcomes
  • The involvement of trained facilitators who are experienced in working with children and families
  • Working with young people in their natural setting (e.g. school or home)
  • Ensuring that programmes are delivered as originally designed, specified and intended (i.e. high implementation fidelity)
  • Regular and/or frequent contacts (e.g. regular weekly contact delivered over the school term or year)
  • Encouraging positive interactions between young people, families and teachers/schools (i.e. addressing violence at individual and relationship levels)
  • Regular and/or frequent contacts (e.g. regular weekly contact delivered over a school term, the school year or longer)
  • Delivery though interactive sessions that provide the opportunity for skills-based demonstrations and practice

In addition, it has been emphasised within the literature that programmes should be theory-driven (Nation et al., 2003; Kovalenko et al., 2020). That is to say that interventions should be based on an explicit theoretical model that describes and justifies how and why an intervention may lead to a change in violence-related outcomes.


The impact of the COVID-19 crisis has the potential to contribute to a rise in youth violence. Moreover, the direct and indirect consequences of violence are broad, extending beyond victims and perpetrators to families and communities. As such, the evidence presented within this report can contribute to decision-making in work to prevent youth violence. School and education-based approaches have been shown to be effective, and the factors that influence their effectiveness have been highlighted. It has been noted, however, that there is limited evidence regarding who is more likely to change (e.g. in relation to age, gender, and sociodemographic status) and when programmes should be implemented. Moreover, it remains to be seen whether interventions of this nature influence youth violence outcomes when delivered out-with education settings or within non-school-aged samples (e.g. those aged 19-29).

Overall, much of the available high-quality evidence on the effectiveness of interventions to prevent youth violence has come from high income countries (such as the USA). As such, it is important to account for cultural context when considering the application of interventions within a Scottish context (Annex B of the report outlines implementation fidelity and associated issues).

Some interventions have been identified as out of scope for this report (see Annex E for a full out of scope list). While these interventions have not been included within this report, this does not necessarily indicate that they do not work. Rather, they have been excluded due to limited available evidence (e.g. high-quality evaluations) or they are beyond the primary prevention focus of this report (e.g. topic out of scope).

Directions for future research

Based on the evidence presented within this report, the following areas for future research have been identified.

i. Further evaluations of interventions – both in Scotland and elsewhere – are necessary to understand ‘what works’. For example, for the interventions classified as ‘inconclusive’ additional evidence via high-quality longitudinal evaluations would be beneficial for understanding the impacts of these interventions on preventing youth violence over time. Embedding evaluation within the intervention programme approach will contribute to understanding the most effective approaches to preventing youth violence. Such evaluations should include both quantitative and qualitative approaches to better understand the impacts and effects of each intervention. There is still a need to grasp what strategies have sustained and long-lasting effects. Research of this nature will allow for progress to be made in this area.

ii. Evaluation research should incorporate a validated measure of behaviour change (e.g., victimisation and perpetration and bystander behaviour). Where possible, assessment of modifiable precursors of young people’s behaviour should be incorporated. This will provide the opportunity for researchers to elucidate the change mechanisms that underpin effective programmes and interventions.

iii. Education based approaches have been shown to be effective when delivered within school settings. Further research could investigate the impact of these interventions when delivered in alternative community-based settings and with those who do not attend school and when implemented in those who are not of school age (i.e. young adults).

iv. Evaluation research should continue to elucidate factors that moderate the effectiveness of youth violence interventions (Kovalenko et al., 2020). Improved evaluation efforts are necessary to better understand what works for whom and under what circumstances, as well as why certain approaches are effective, when others are not. Faggiano et al (2014) note that “not knowing why, how, and where prevention interventions work limits knowledge about generalizability and optimization of intervention”.

v. More broadly, very little research has focused on examining the effectiveness of these interventions in low and middle income countries. We cannot assume that findings from high income countries will transfer.

vi. Strengthening this evidence base will require longitudinal evaluations of effectiveness across different population groups and communities, using validated measures of violence related outcomes and relevant associated risk factors. It has also been highlighted that better reporting is needed if programmes are to be replicated elsewhere (Fagan & Catalano, 2013).