What works to prevent youth violence: evidence summary

Community coalitions

Classification: Promising

Background

In their report Preventing Violence, Promoting Peace, Bellis et al (2017:60-61) note that “community coalitions use local data to understand problems and inform preventative action” and encourage “partnership between young people, their families, schools, community organisations and public services”.  The authors of this policy toolkit for preventing interpersonal violence highlight that the implementation of community coalitions are associated with a reduction in homicide, violent crime, and violent behaviour.

Available Evidence

One example of an initiative that has taken this approach is Communities That Care. Here, community coalitions are implemented to collect data locally, which provides an opportunity to establish the extent to which young people are experiencing both risk and protective factors. The needs of youth are prioritised based on this assessment and individual, school-based, family-focussed, or community level interventions are implemented to fill any gaps in service and target the specific needs of this community (David-Ferdon, 2016). This intervention also sets out to strengthen community-level protective factors by reducing norms that encourage violent behaviours and enhancing young people’s attachment to the community. A randomized trial involving twenty-four communities in seven states (USA) demonstrated reductions in the incidence of self-reported violent behaviours one year following the implementation of Communities that Care^[31] (Hawkins et al., 2012). Further evidence regarding the effectiveness of community coalitions is needed. However, the complexity of communities, the wide range of contextual factors and the numerous challenges to implementing the programme with sufficient fidelity to the model, makes the evaluation of community based initiatives difficult.

The result of a five-year pilot of the Communities that Care scheme within three communities in the UK demonstrated that whilst the programme was supported and accepted by those involved, it was challenging to measure its impact on risk and protection. This is the case as it is designed to be a long-term community intervention and so a longer-term view needs to be taken when considering its potential role in preventing and reducing violence in the UK (Crow, 2004). This approach has been implemented within a Scottish context. However, it was deemed to be too early to examine the impact of the intervention at the stage the evaluation was conducted.

Moderating factors

As our understanding of the role of community coalitions in reducing and preventing violence is limited, it is not yet clear which factors enable or impede its effectiveness. However the Evaluation of three 'Communities that Care' demonstration projects (2004) conducted by the Joseph Rowntree Foundation in highlighted that the following factors played a key role in the successful implementation of the initiative:

• Presence of strong partnerships, active communities and good leadership prior to the start of the programme.
• Active and supportive 'champions' of the project among the senior executives of key local agencies.
• A wide range of professionals involved at operational and managerial level from the beginning - especially where specific programmes were to be implanted, such as in schools and social services.
• Structures and processes which allow for working between strategic and operational levels.
• Early and comprehensive inductions for new partners and staff
• Project co-ordinators who maintained momentum, increased the active involvement of partners, and kept projects focused on core objectives
• Moving from planning to delivery was the most difficult part of the process. Success was more likely where the programme was built on the consent and active involvement of all involved.
• Money and resources were critical. Staff and local people found it frustrating when time and effort spent devising a plan were not matched with the resources to implement the work.