This study is one of the first to provide first-hand evidence of whether such programmes are effective. It focuses on ten 'Mentoring Plus' programmes. The report reviews both the process by which the programmes were implemented and the outcomes they achieved.
It examines how vulnerable young people get involved with mentoring. As well as reviewing young people's experiences, it also looks at those of staff and volunteers. The authors identify the medium-term impact of mentoring on those young people involved, focusing specifically on: education, training and work; levels of offending; drug use; and general psychological well-being.
The study concludes with a review of the successes and weaknesses of the programme and considers what policy-makers, funders and participants can do to improve the effectiveness of such schemes.
Although mentoring has become a very popular response to youth disaffection there is little evidence as to its efficacy. This research - by a team from the LSE - sought to evaluate the role of mentoring by looking at the experiences of 10 'Mentoring Plus' programmes run by Crime Concern and Breaking Barriers. These programmes targeted disaffected young people and offered a one-to-one mentoring service (with a volunteer mentor recruited from the local community), a programme of education and training and a series of social activities (the Plus element). The key findings were:
- The programme successfully recruited a group of highly disaffected young people who were, or were at serious risk of becoming, socially excluded.
- More than half of the young people recruited onto the programme engaged with it on a monthly basis or more often. The programme appeared to be particularly successful in engaging those young people who were most at risk of social exclusion.
- Mentoring disaffected young people was found to be a delicate, cyclical and reactive process. Many relationships did not progress beyond basic 'mundane' social interaction on to specific problem-focused or goal-oriented activity.
- The young people were generally positive about both mentoring and the Plus element of the programme. Most felt that both components had been helpful to them.
- Evidence of impact was most marked in relation to engagement in education, training and work. Participation in the programme was associated with a heightened rate of engagement in these areas. This change was most marked in projects that were well implemented.
- There was no clear evidence of the programme having an impact in relation to offending, family relationships, substance use and self-esteem.
- The researchers conclude that more realistic expectations and models of mentoring are required.
In recent years mentoring has emerged as the latest in a long line of interventions aimed at reducing youth disaffection. Although this approach has become very popular there is remarkably little empirical evidence as to its efficacy. The research described here sought to address this gap by evaluating a mentoring initiative for disaffected young people.
The study focused on 10 programmes, known as 'Mentoring Plus', run by Crime Concern and Breaking Barriers. The research team surveyed a group of programme participants at the beginning and end of the programme and, again, six months later. They also surveyed a comparison group of young people who were referred to the programme but who did not participate in it.
In addition to the survey, researchers undertook qualitative interviews with young people, mentors, project workers and referral agents and observed numerous programme sessions.
The nature of the programme
The programmes specifically targeted disaffected young people and were based on the model provided by the award-winning Dalston Youth Project. Each wave of the programme ran for 10-12 months, starting with a residential weekend. Volunteer mentors were recruited from the local community and matched with the young people. Alongside this component, regular education and training sessions were provided and frequent social activities were organised (the Plus element).
The young people on the programme
Almost 400 young people between the ages of 12 and 19 were recruited onto the programme. Many had been referred to Mentoring Plus by Youth Offending Teams, Educational Welfare Services and schools. More than half (57 per cent) of those recruited went on to engage with the programme on a monthly basis or more often.
There is little doubt that the young people recruited onto the programme were at significant risk of social exclusion. Many had experienced substantial disruption in their schooling and family lives; truancy and disengagement were widespread; many of those who had left school had done so without any qualifications; and levels of offending, illicit drug use and contact with the criminal justice system were much higher than in the general youthful population.
The programme appeared to be particularly successful in engaging those young people who were at most risk of social exclusion, including large numbers of Black African/Caribbean young people.
The nature of mentoring
Existing models of mentoring are poorly theorised. As a consequence, how such interventions are expected to work remains unclear. It is often assumed that mentoring will proceed steadily through a series of stages, before coming to focus
on the needs of the young people and on devising strategies to mitigate problems and overcome difficulties. Such thinking over-simplifies the
nature of mentoring and tends to overstate the centrality of goal-oriented, instrumental activities in such work.
This study found mentoring to be a delicate process based largely on 'ordinary' social interaction which often has little obvious connection with responding to challenging behaviour or the causes or consequences of social exclusion. Typically it is cyclical and reactive. The research team identified three potential cycles:
- the basic cycle: contact-meeting-doing;
- the problem-solving cycle: contact-meeting-doing-fire-fighting;
- the action-oriented cycle: contact-meeting-doing-fire-fighting-action.
The activities in the basic cycle were generally fairly mundane - having tea/coffee, playing pool, shopping, bowling, or perhaps going to the cinema - and many relationships did not progress much beyond this. Proactive planning and action were relatively rare. Where relationships did progress beyond the basic cycle, they often did so in response to a problem or crisis (e.g. homelessness, family breakdown, specific forms of offending, substance misuse, violent behaviour). Some relationships did progress beyond the reactive fire-fighting stage to become genuinely action-oriented and closer to what often appears to be the typical conception of ideal mentoring. Where relationships progressed, they tended to do so fairly late on, once trust had developed and key issues were identified.
"The first half of the year was definitely just more like going out every week and chatting, her getting to know as much about me as possible because I thought that would be quite important and then I'd slowly try to bring a little bit of structure and we did our action plans." (Mentor)
Young people's assessments
The vast majority of young people recruited on to the programme considered the mentors and the Plus element to have been helpful: 37 per cent felt the mentors had been very helpful and 33 per cent felt they had been fairly helpful, while 45 per cent felt the Plus element had been very helpful and 33 per cent felt it had been fairly helpful. More of the young people felt that the programme had helped them in relation to education and work than in relation to tackling offending.
"[Mentors are] genuinely interested in what is going on in your life, and they are not just there because it is a job, because as far as I'm aware they are like volunteers so they are not getting paid, so they can't really pretend because there is nothing in it for them." (Young person)
Risk and risk management
Although the young people's assessments of the programme were generally positive, much of the work carried out involved an underlying sense of instability and unpredictability. On some occasions control broke down and the research team observed instances of violence and anti-social behaviour (particularly during the residential weekends). This suggests that the realities of working with highly disadvantaged young people are such that notions of risk and risk management should be at the forefront of programme design and implementation.
"I would go again. It was just mad though, chaos, a weekend where there was just no sleep, and they [the project workers] say they are the sex police, and the women patrol the girls' landing and the men patrol the boys' landing ... there is no sex, but it is just jumping out of each others' bedroom, you know, and keeping each other apart. But some people want to sleep, so they put the mentors in the bedrooms with us, because the lads crept in, but we weren't doing nothing. We were just sat on the beds talking and then my mate nearly had a fight with one of the lads over something - a packet of crisps or something - so the lads just kept bursting into our room and they broke the door ... [The mentors] didn't like it." (Young person).
The study assessed the impact of the programme in relation to engagement with education, training and work, family relationships, offending, substance use and self-esteem.
Evidence of impact was strongest in relation to engagement with education, training and work. The proportion of programme participants who were engaged in such activities increased substantially between the start and end of the programme (from 49 per cent to 63 per cent); the proportion of non-participants engaged in such activities did not change during this period.
"[My mentor] just helped me feel more organised, basically, I've told her what I want to do, and she just helped me to do it, and she's been there to push me the few steps I've needed to be pushed... If she weren't there I wouldn't be going back to school. I used to go and hang around the streets and do crime and stuff but now I can't be bothered to do that because I want to get a job, just want to go to college." (Young person)
There were, moreover, good reasons for thinking that the programme played an important part in these changes. The vast majority of participants who moved from a position of disengagement to engagement indicated that the programme had helped them in relation to education and/or work (most indicated that both the mentors and the Plus component had helped them in this way). In addition, increased levels of engagement were most marked in projects that were well implemented. In projects which ran into difficulty the overall balance between engagement and disengagement remained unchanged.
There was no clear evidence that the programme had an impact in any of the other areas considered. Although levels of offending did fall markedly among programme-participants, similar rates of decline were evident among non-participants.
The researchers conclude that the achievements of Mentoring Plus are particularly impressive when set in context. Interventions with disaffected young people are inherently difficult to implement and these difficulties are exacerbated by insecure funding, fixed-term employment for project workers and high staff turn-over. While many of the Mentoring Plus projects ran into operational difficulties, those that were well implemented had the greatest impact in terms of encouraging young people into education and work. This suggests, that in a more secure environment, the overall impact of the programme would have been greater. The researchers recommend that policy-makers and funders think carefully about how they support such programmes in the future. Potentially positive work with young people could be undermined by circumstances which serve to undermine, or at least limit, the implementation of programmes in this area.
In terms of future development, however, mentoring does not simply require more secure and longer-term funding. There is also a need to think through the details of the process more fully. It is perhaps unsurprising that Mentoring Plus appeared to have greatest impact on engagement in education, training and work as it is here that the structured activities related most clearly to the aims of the programme. By contrast, while reducing offending was a general aim of the programme, it was not a specific goal of any of its structured elements. This is not a criticism of mentoring per se, but it does highlight the need for more fully theorised models and for more realistic expectations.
When set out in these terms such findings may seem unsurprising. And, yet, it is all too often the case that insufficient attention is given to the aims of particular programmes and to how the aims relate to the activities. This lack of clarity is then compounded by the absence of any explicit model of change. Why is it that a particular intervention might be thought to work; under what circumstances might it be expected to work; and with whom might it be expected to work? These are the central questions which should provide the focus for future practice and research in this area.
About the project
The research was conducted by Michael Shiner, Tim Newburn, Tara Young and Sylvie Groben of the Manheim Centre for Criminology at the London School of Economics. The fieldwork for the study was carried out between November 2000 and April 2003.
The Mentoring Plus cohort was made up of 378 young people, of whom 188 responded to the first follow-up survey and 102 responded to the second follow-up survey. The comparison group was made up of 172 young people, of whom 56 responded to the first follow-up (the comparison group was not included in the second follow-up).
Depth interviews were conducted with 25 project workers, 20 referral agents, 40 mentors and 36 young people. In addition, the research team observed almost 150 project sessions and events, including recruitment events, programme activities and staff meetings.