FRIENDLY, caring adults who become mentors to young people with difficulties are often convinced about the value of their relationship and the positive effect they bring.
But British studies are inconclusive in contrast to the upbeat message from the United States, according to the Scottish Council for Research in Education Centre at Glasgow University.
Most mentors tend to be white, middle-class women. Few studies actually record what young people feel and mismatches can ruin any chance of success.
John Hall, the SCRE researcher, notes: "If the mentor comes to be seen as acting on behalf of authority rather than on behalf of the young person, then the mentoring relationship is endangered, and probably destroyed."
The literature review was commissioned by the Scottish Executive after the follow-up to the Beattie report on the post-school inclusion of 16-24s raised the prospect of using mentors to ease their transition to education, training and employment.
Many of the young people involved have mental health problems, learning difficulties or low attainment. Drug and alcohol misuse or homelessness may be other characteristics.
Dr Hall finds that mentoring is fraught with problems and that much existing research must be treated with caution. Some studies question whether mentoring is a valid approach if it is trying to sell reluctant and disaffected young people a different set of values.
In the US, where mentoring has a stronger tradition, studies confirm that it can be useful on a number of measures but the impact may not be large.
"The best US evidence is that mentoring may have some impact on problem or high-risk behaviours, academiceducational outcomes and careeremployment outcomes. There is a very poor evidence base in the UK. Claims are made for the impact of mentoring but there is as yet little evidence to substantiate them," Dr Hall states.
One study in England, for example, compared a group of at risk primary pupils who were mentored with a similar group who were not, in terms of behaviour, attendance, exclusion and academic performance. After a year, the mentored group made gains in confidence, self-control, social awareness and relationships.
"Unfortunately (for the researchers) they found that similar gains occurred in the comparison group and both groups continued to show serious problems," Dr Hall comments.
"Mentoring and Young People: A Literature Review", by John Hall of the SCRE Centre at Glasgow University.