STEER

What does this project involve?

Salford Foundation’s STEER programme is a mentoring intervention that aims to reduce children’s involvement in violence. Delivered by youth workers (mentors), the programme provides four weeks of initial interactions and assessment, followed by 24 weeks of weekly, face-to-face, one-to-one, one-hour mentoring sessions, and one-hour per week of wraparound casework. Casework may include phone calls, online support, advocacy with other agencies and participation in multi-agency meetings.   

The face-to-face mentoring sessions follow the STEER toolkit of mandatory and optional topics. Mandatory topics include safety planning, relationship mapping, understanding healthy relationships, exploitation, weapon carrying, attitudes, behaviours and goal setting.  Sessions take place where children feel comfortable including home, school, Salford Foundation offices or in the community. All parents and carers are also offered 14 hours of support from a Family Support Worker over six months. This support aims to improve parenting skills, provide strategies to manage boundaries, and improve parent communication (with both children and professionals).  

Children are referred into STEER by partners including multi-agency panels, education settings, complex safeguarding teams and the police. Children are eligible if they are aged 10 to 17 and identified as at risk of involvement in violent crime because they have an association with peers or family members involved in serious violence, organised crime or gangs. 

Delivery in this project took place in six Greater Manchester boroughs: Trafford, Wigan, Manchester, Salford, Tameside and Bolton. 

Why did YEF fund this project?

As the YEF’s Toolkit strand on mentoring explains, mentoring children is associated with a moderate impact on reducing violent crime. On average, mentoring reduces violence by 21%, all offending by 14%, and reoffending by 19%. We also know that mentoring programmes have tended to have larger impacts when they work with children at higher risk of involvement in crime. However, there are a very limited number of robust evaluations undertaken in an English and Welsh context, and the impact of mentoring interventions varies.  

YEF, therefore,  funded a randomised controlled trial of STEER.  The evaluation aimed to establish whether STEER reduced offending behaviour (as measured by the Self-Reported Delinquency Scale [SRDS]). It also measured the impact on children’s pro-social behaviours, emotional symptoms, relationships with peers and adults, and conduct problems (all measured using the Strengths and Difficulties Questionnaire). The quality of the mentor–mentee relationship was also assessed using the Social Support and Rejection Scale. Children were randomly allocated to a treatment group (which received STEER) or a control group (which received signposting to business-as-usual services). 689 children were randomised – 337 to receive STEER and 352 to the signposting group.  

An implementation and process evaluation was also undertaken to assess how effectively STEER was implemented, explore the key factors that influenced delivery and describe children’s experiences of support. The evaluators conducted interviews with 15 children, 11 STEER project staff and 10 wider stakeholders (including local councils, police and youth justice services). Delivery data was also collected. Data collection took place between May 2022 and March 2025. 80% of the children in the trial were White, 10% were of mixed or multiple ethnicity, 4% were Asian, 3% were Black, and 3% were from an other ethnic group. 

Key conclusions

STEER demonstrated a small negative impact on children’s offending behaviour. Children who received STEER were slightly more likely to report offending behaviour than children who did not receive STEER. There is uncertainty surrounding the estimate. This result has a moderate security rating.

STEER showed a moderate impact on reducing emotional symptoms and a small impact on reducing peer difficulties and on improving pro-social behaviour. It demonstrated a small negative impact on conduct problems. These are secondary outcomes and should be treated with more caution.

STEER may not have been delivered at the intended intensity. Children were recorded as having received, on average, 14 one-to-one mentoring sessions versus the intended 24. Only 50% of families were recorded as having received family support, with an average of three sessions (compared to the expected 14). The reduced dosage may have played some role in preventing the programme’s intended impact.

STEER staff and wider stakeholders identified a range of factors that supported effective delivery. The Salford Foundation’s strong local presence, trusted relationships and involvement in multi-agency panels were viewed as enabling smooth referral and engagement processes. When expanding into new areas where relationships were still developing, referrals took longer to build. Stakeholders reported high demand for STEER.

Children who received STEER reported stronger relationships with their STEER mentors than children in the control group reported with identified trusted adults. The more mentoring sessions children attended, the stronger these relationships were. Interviews with children also suggested that mentor–mentee relationships were a consistent strength of STEER.

How secure is the evidence? 

These findings have a moderate security rating (3/5 magnifying glasses). The trial was a well-designed efficacy randomised controlled trial. It was large enough to detect meaningful impacts and larger than most previous evaluations of mentoring. 25% of children who started the trial were not included in the final analysis. We do not know whether the effect found for STEER would be the same if the children missing from the final analysis were included.

Should you fund or deliver STEER?

STEER has not evidenced a desirable impact on offending behaviours in a robust causal evaluation. While we know from the international evidence that mentoring can protect children from involvement in violence, there may be mentoring programmes that may have larger, and more positive, impacts.  If you are delivering STEER or a similar mentoring programme, it is important to carefully monitor how well it is meeting your local aims and to ensure delivery follows best practice. Drawing on the international evidence base, providers may want to consider the following: 

• Ensure your mentoring programme offers and delivers weekly, hour-long mentoring sessions for at least six months. 
• Provide mentors with appropriate training, including, for example, safeguarding; setting expectations (rules, goals and expectations); relationship development; listening skills; understanding local services; dealing with difficult situations; and advice on where mentors can go to seek support. 
• Ensure mentors receive regular supervision to reflect, share challenges and receive guidance.  
• Inform parents and carers about mentoring, as their awareness can help sustain engagement. 
• Communicate clearly with children about how mentoring will end to avoid abrupt endings and support a positive close to the relationship. 

Additional learning from the STEER evaluation: 

• Build and maintain strong multi-agency relationships and clear referral pathways into your mentoring programme.  Consider the local service landscape – including existing referral routes and specialist or culturally specific services – to establish clear, appropriate referral pathways and enable effective partnership working. 
• Carefully monitor attendance and engagement and aim to implement strategies that reduce barriers to it. Many young people experience instability or involvement with multiple services, which can negatively impact engagement. 
• Identify and address barriers to family engagement early (for mentoring programmes with a family support element), especially where families are already working with several professionals. 

What will YEF do next?

YEF has no further plans to invest in STEER at this stage, but we remain very interested in how to effectively implement high quality mentoring to children at risk of involvement in violence.  

Download the report