Australian Government Attorneys General's Department - Crime Prevention and Community Safety

Early Intervention Youth Mentoring Programmes

An overview of mentoring programmes for young people at risk of offending

Study Team:

Dr Ania Wilczynski, Senior Social Researcher, Urbis Keys Young
Clare Culvenor, Social Researcher, Urbis Keys Young
Associate Professor Chris Cunneen, Director, Institute of Criminology, University of Sydney Law School
John Schwartzkoff, Director, Urbis Keys Young
Kerry Reed-Gilbert, Principal Consultant, Kuracca Consultancy

To order any National Crime Prevention publications please contact:

Crime Prevention Branch, Australian Government Attorney-General's Department
Robert Garran Offices, National Circuit, BARTON ACT 2600

Ph: +61 2 6250 6711

Fax: +61 2 6273 0913

Publications are also available at www.crimeprevention.gov.au

Early Intervention: Youth Mentoring Programmes - An overview of mentoring programmes for young people at risk of offending

Australian Government Attorney-General's Department, Canberra

ISBN 0 642 21098 5

The views expressed in this publication are those of the author and do not necessarily represent the view of the Commonwealth of Australia. Whilst all reasonable care has been taken in the preparation of this publication, no liability is assumed for any errors or omissions.

See also: ISBN 0 642 21090 X Early Intervention: Diversion and Youth Conferencing - A national profile and review of current approaches to diverting juveniles from the criminal justice system

Design: Design Direction

Print: Union Offset

Publisher: Australian Government Attorney-General's Department

Foreword

Jurisdictions across Australia are developing alternative processes to reduce the number of young people entering the criminal justice system.

Overseas research suggests that mentoring, as an early intervention strategy, has promising benefits with young offenders and young people at risk of offending.

This report, commissioned under the Australian Government's National Crime Prevention Programme, provides a national profile of mentoring programmes for young people at risk of offending, identifies differing approaches and models for mentoring programmes and highlights good practice in the field.

In particular, this research analyses the crime prevention outcomes from youth mentoring, and the value of strengthening links between families and communities.

SENATOR CHRIS ELLISON
Minister for Justice and Customs
Senator for Western Australia

Abbreviations

Executive Summary

Introduction
What is mentoring?
Background to the research
Objectives of the project
Structure of this report

Methodology
Definition of mentoring
Key components of the methodology
Fieldwork
Some caveats

Organisational context of mentoring programmes
Location of mentoring projects
Auspicing organisation
Funding
Formal versus naturalistic mentoring programmes
Programme objectives
Stand-alone versus integrated projects
Staff
Establishment and administration
Initial meetings
Length of contact
Frequency of contact
Parental involvement
Activities undertaken by the pair
Nature of the relationship between the mentor and mentee
Ending the mentoring relationship
Monitoring and supervising mentors
Training and orienting mentors

Mentors
Recruiting mentors
Characteristics of mentors
Screening procedures
Excluding mentors with criminal records
Matching
Paid versus volunteer mentors

Mentees
Voluntary versus mandated clients
Recruitment processes
Screening/eligibility of mentees
Characteristics of mentees
Specific issues for mentoring programmes servicing young offenders

Evaluation
The importance of evaluation
Lack of evaluations of mentoring programs
Reasons for lack of evaluation
Increasing the emphasis on evaluation

The effectiveness of mentoring programmes
Mentoring is a promising but unproven strategy
Mentoring can achieve positive benefits for some young people
Mentoring is only suitable for some young people
The impact of demographic characteristics of mentees

Conclusion
Models and organisational structure: good practice
The mentoring process: good practice
Finding and employing mentors: good practice
Selecting mentees and developing a programme: good practice
Evaluation: an essential part of good practice
Determining the effectiveness of mentoring
Indigenous programmes
Recommendations for stage 2

Bibliography and select references

Appendix A: Letter inviting nomination of mentoring projects

Appendix B: Postal survey to mentoring projects

Appendix C: Questionnaire guides for fieldwork

Abbreviations

ACT	...	Australian Capital Territory
BSBB	...	Big Sisters Big Brothers, NSW
BBBS	...	Big Brothers Big Sisters, USA
CA	...	Conference Administrator, NSW
CPB	...	Crime Prevention Branch, Commonwealth Attorney-General's Department
DHS	...	Department of Human Services, SA
DJJ	...	Department of Juvenile Justice, NSW
EEO	...	Equal Employment Opportunity
FAYS	...	Family and Youth Services, SA
JUMP	...	Juvenile Mentoring Programme, USA
MAYT	...	Metropolitan Aboriginal Youth Team, Adelaide
NSW	...	New South Wales
NT	...	Northern Territory
NZ	...	New Zealand
SA	...	South Australia
TAFE	...	Technical and Further Education
UK	...	United Kingdom
WA	...	Western Australia
YLO	...	Youth Liaison Officer, NSW
YJC	...	Youth Justice Conference
YWCA	...	Young Women's Christian Association, NSW

Executive Summary

The Crime Prevention Branch (CPB) of the Commonwealth Attorney-General's Department commissioned social research consultants Urbis Keys Young to conduct a project to profile and review mentoring programmes for young offenders, particularly those in the 10 to 13 year age group.

This is the Final Report on the project.

Objectives

The project involved a national audit and review of mentoring programmes around Australia in order to:

provide a national profile of mentoring programmes for young offenders
identify models and good practice
identify key crime prevention outcomes from youth mentoring
make recommendations for Stage 2 of the project.

The objectives of Stage 2 are yet to be determined, but key activities might include, for instance, the development and production of resource materials for a range of audiences, evaluations of specific mentoring programmes, or the trial of a mentoring programme with a specified target group.

Methodology

The key components of the methodology were as follows:

A review of Australian and overseas literature. This primarily covered material published between 1997-2001.
Initial scoping consultations to identify mentoring projects within the scope of the study. This involved distribution of a standard letter (mostly via email) to key agencies around Australia which may have been aware of mentoring programmes. Initiatives were selected for inclusion in the review where the primary aim of the project was mentoring, the target group was young offenders or young people at risk of offending and the mentors were community-based individuals rather than people acting in a professional or semi-professional capacity as youth workers.
postal survey of the 24 projects identified nationally.
In-depth face-to-face consultations on field visits with seven mentoring projects nationally. Consultations were conducted with field staff for all projects, and in some cases external stakeholders, mentors and young people.
Telephone and face-to-face interviews with selected key informants to supplement the material gathered on the field visits.
A face-to-face consultation with Crime Prevention Officers from around Australia at one of their regular meetings.
Data analysis and reporting, including entry of the quantitative data from the survey into a database.

Overview of mentoring programmes

This audit and review of mentoring programmes was based on 21 projects nationally, from the initial group of 24 which were identified in the scoping process. This report provides a 'snapshot' of mentoring in early 2002. Since then new projects have started and others have ceased operation. From the snapshot the following profile emerges:

62 per cent are run by community organisations.
Government is the primary funding source for most projects.
More than half those surveyed cited insufficient funding as a significant weakness for the programme.
Almost half the projects are pilot programmes.
Most programmes have organisational autonomy from government.

Models and organisational structure: good practice

Both the Australian and international literature, as well as the consultations which took place as part of this project, point to a number of features of mentoring programmes which can be seen as good practice.

Programme objectives

The literature and consultations show that it is important to develop well defined programme objectives and operating principles (involving designation of tasks, accounting principles etc). Ideally, these should be developed in consultation with potential participants and stakeholders. Programme plans should be realistic and attainable, and maintain a degree of flexibility as the programme develops.

Some key principles that a service should establish prior to operation include:

A tightly defined target group.
Sufficient trained staff to run the programme and provide support to participants.
Clear guidelines relating to the recruitment and screening of mentors.
Well developed links with a range of local agencies who can support the work of the programme.

Naturalistic versus formal

Mentoring programmes can be divided between 'formal' and 'naturalistic' in their approach, with formal programmes being more regimented. There is no evidence from the literature or the consultations concerning whether formal or naturalistic programmes are more effective, other than it appears important for formal programmes to avoid simply replicating a 'youth worker' function.

A mentoring project should establish a clear conceptual philosophy regarding the nature of the programme and whether it will take a formal or naturalistic approach to mentoring.

Stand-alone versus integrated projects

A key Australian external evaluation report (of One2One in NSW) indicates that stand-alone mentoring projects may have greater difficulties 'getting off the ground' than integrated projects, and there was general agreement amongst the stakeholders consulted for this project that integrated projects are more likely to be effective. The literature also suggests that services other than mentoring should be available to the clients of mentoring programmes, either via that or other organisations.

Mentoring projects which are integrated into a range of other services appear more likely to be effective than those which are stand-alone.

Staff

Both the literature and the consultations indicate the importance of effective project staff. The need for strong overall coordination of the mentoring programme is essential to its success.

Staff running mentoring projects should be appropriately skilled and have clearly written job descriptions.

Establishment and administration

Establishing mentoring programmes involves effective scoping to determine need. It also requires effective consultation with specific groups, in particular Indigenous people and their organisations.

Strong organisational administration and infrastructure are essential for a mentoring programme to operate effectively.
Accountable and appropriate policies, practices and codes of conduct dealing with a range of issues need to be developed and put in place.

The mentoring process: good practice

The literature and consultations show that specific policies and processes around the mentoring relationship are likely to improve positive outcomes.

Initial meetings

The initial meeting between mentors and mentees should be used to establish a number of elements regarding the relationship, for example the procedure if someone cannot attend a meeting, when and where meetings will occur and so on.

A project should decide how much information the mentor will have in regards to the mentee prior to an initial meeting, taking into consideration the aims of the project and the safety of all participants.

Length of contact

To be fully effective, the mentoring relationship should be sustained over a long period of time. It is recommended that relationships last for a minimum of six months, with an ideal length of 12 months. Continuity and length are significant factors in the success of a mentoring relationship.

Frequency of contact

The frequency of contact will largely be determined by the nature of the project and the activities undertaken by the mentor and mentee. Contact must be 'regular and consistent.' The ideal frequency of contact suggested is once per week.

Parental involvement

Mentoring projects should engage parents or guardians of the mentee in the project where possible, including obtaining their permission and informing them as to the nature and extent of the mentoring relationship.

Activities undertaken by the mentor and mentee

Activities undertaken by the matched pair will be largely dependent on the nature of the programme. However where activities are not designated, it is recommended that the young person play a significant role in determining the activities that the pair engage in, and that they be fun, developmentally useful and low-cost. There should also be recognition of the need for differences in activities for young men and young women.

Ending the mentoring relationship

Programmes should have a clear policy on how and when to end mentoring relationships. This should include the mentor meeting with the mentee and discussing the close of the relationship. It should be kept in mind that while a relationship may have an 'official' end, the relationship may continue on a casual or indefinite basis.

Monitoring and supervising mentors

Provision of supervision, feedback and support by project staff for mentors that is consistent, timely and regular (although not too burdensome) is essential for any programme. Support can be in the form of group meetings, telephone contact, or face-to-face interviews. The frequency can be determined by each individual project, but an average of once per month (at least in the early stages of the relationship) is recommended.

Training and orienting mentors

Mentors should be required to undergo at least 20 hours of training prior to being matched with a young person. Training should be provided by people with appropriate qualifications, and include an overview of the programme, clarification of commitment requirements, boundaries and limitations, crisis management and problem solving, communication skills and a summary of policies and guidelines.

Finding and employing mentors: good practice

The literature and consultations show that there are a number of practices which should be put in place when selecting and employing mentors.

Recruiting mentors

When recruiting mentors, a project must provide a clear and realistic idea of the benefits and expectations of mentoring, including the level of commitment required.

There is no one 'good practice' in recruiting - the mode of recruiting will depend largely on the types of people who are desired as mentors (such as ethnicity, interests, age, etc). It should be noted that personality is often considered to be more important than physical, social or racial demographics.

Characteristics of mentors

There are a number of characteristics which should be focused on when recruiting mentors, with the most important elements being the ability to listen, a non-judgemental attitude, flexibility, respect for and ability to relate to young people and reliability/consistency.

Screening procedures

Mentors should be rigorously screened prior to being matched with a young person. At a minimum, screening processes for mentors should include at least one personal interview with project staff, a criminal record check, a reference check and an application form.

Excluding mentors with criminal records

A criminal record should not necessarily preclude someone from being a mentor; indeed, some programmes even seek out people with a criminal record to act as mentors. However, any person who has been convicted of sexual offences, any offences against children, any violence offence, or any serious offence within the past five years should not be accepted as a mentor.

Matching

The matching process should be based on a clear and consistent policy. While sex, race, ethnicity and cultural background may be considerations, the primary factors in matching should be the interests, needs and goals of the young person.

Any mentoring programme should take care to consider the cultural circumstances of both their mentors and their mentees when developing programme and relationship structures. This issue is likely to be particularly significant for Indigenous people as both mentors and mentees.

Paid versus volunteer mentors

There is no evidence to indicate whether it is preferable to employ paid or volunteer mentors. However, payment may be particularly important where there is an attempt to include specific groups of people as mentors (such as Indigenous people) who are likely to fall within a lower socio-economic bracket.

It is good practice to reimburse mentors for agreed costs and out of pocket expenses relating to the relationship (such as fares, mileage, entry fees etc). Receipts and/or proof of purchase should be required for reimbursement.

Selecting mentees and developing a programme: good practice

Like the selection of mentors, the literature and consultations show that there are a number of issues which need to be considered when selecting mentees and developing an appropriate mentoring relationship and programme. Consideration of these issues will assist in developing good practice.

Recruiting

For mentoring projects to operate they need a steady supply of suitable clients. The importance of strong referral networks is demonstrated by the experience of several Australian mentoring projects outlined in this report.

Projects which have had the greatest success in recruiting young people have well-established and effective networks to provide a regular and adequate source of appropriate referrals.

Screening and assessing mentees

There needs to be put in place a screening process to assess suitable mentees. It is particularly important to identify a number of issues in this process.

Voluntary versus mandated clients. The research and consultations show that there are potential problems with taking young people as mentees when they are mandated to attend the programme. It may be preferable to only accept young people voluntarily into a mentoring project, irrespective of whether they are subject to a court-imposed supervisory order.
Violence and the potential for harm. It is generally regarded as good practice not to accept any mentees who may pose a threat to their mentor.
The need for professional assistance or intervention in place of mentoring. Some potential mentees may require far more professional support than can be provided by a mentor (eg for mental health problems or serious drug problems).

The programme activities

There are no specific 'rules' about what works in relation to the activities undertaken during the mentoring process. However, several points emerge from the literature and the consultations.

There is a need to proactively 'sell' the programme to young people.
There needs to be a focus on activities which are appealing to young people.
High profile mentors or coordinators can be particularly appealing to young people.

Mentoring programmes for young offenders

In the course of the scoping consultations for this study, it appeared that there were numerous mentoring projects operating around Australia for 'at risk' young people, but relatively few covering the specific target group of young offenders or young people at risk of offending.

The evidence shows that mentoring with young offenders may be more intensive or complex for young offenders than generally 'at risk' young people (especially in relation to mentoring frequency).
Therefore, programmes need to carefully consider the implications of the target group and develop programme elements and strategies accordingly, especially:
- the greater resources needed for young offenders
- the greater demands on mentors.

Evaluation: an essential part of good practice

This report has found that in practice most mentoring programmes both overseas and in Australia have not been subject to a formal evaluation, particularly of an external nature. Monitoring and evaluation is essential for any mentoring programme to determine its effectiveness.

Projects should engage in thorough data collection and be subject to regular external evaluation. This is particularly critical for projects in the earlier stages of development.
There is a need to build evaluation processes and mechanisms into any mentoring project from its very inception.
Information and data should be collected and retained throughout the programme, including personal data for all participants, administrative information, programme data, output measures and outcome measures.
External evaluations are preferable to internal evaluations.

Determining the effectiveness of mentoring

The general lack of evaluation of mentoring programmes significantly limits the conclusions that can be drawn about the effectiveness of mentoring in reducing offending and problematic behaviour or in improving self-esteem and social skills.

From the overseas and Australian literature the following conclusions can be drawn:

Mentoring is a promising but unproven strategy.
Mentoring can achieve positive outcomes for some young people.
Mentoring is only suitable for some young people.
Where positive outcomes have been reported from mentoring programmes in both Australia and overseas, they have been short-term. Overall there is a dearth of evidence of long-term impacts of mentoring programmes.

Despite the significant limitations on research and evaluation, some tentative points can be drawn on the basis of some successful mentoring projects.

Reduced offending

Some mentoring programmes both overseas and in Australia report a reduction in offending by young people who have participated in mentoring programmes.

Reduced substance misuse, and other risky behaviours

Some evaluations of mentoring programmes also report reductions in substance misuse (drugs and/or alcohol) or other risk-taking behaviours.

Increased participation/performance in education, training and employment

Another finding reported by some mentoring programmes is either greater participation in or better performance in education, training and/or employment.

Improved self-esteem, social/communication skills and personal relationships

A number of mentoring programmes also report improvements in more general qualities or skills in the young person such as enhanced self-esteem and social/communication skills and improved relationships with others such as family and peers.

Mentoring is only suitable for some young people

The literature and consultations show that mentoring will only be suitable for some young people. There are two aspects to this point:

Not all young people will be suitable to refer into mentoring programmes. In other words, some young people will be screened out as unsuitable for mentoring.
Mentoring will not work for all young people referred into programmes. There will also be some young people who will not successfully engage in a mentoring relationship.

There is some limited evidence that the effectiveness of mentoring may vary according to the demographic characteristics of mentees.

Age. Generally younger people (in their earlier teens) are more receptive to mentoring approaches.
Sex. There is some limited UK evidence which suggests that mentoring may reduce offending more effectively in girls than in boys.
Race/ethnicity. Based on the available US information, the benefits of mentoring programmes for Indigenous and ethnic minority youth are unclear. However, in the Australian consultations some argued that mentoring programmes are likely to be particularly effective for Indigenous young people.

Indigenous programmes

Information was gathered on a number of Australian Indigenous specific programmes through both the consultation process and the literature review. These programmes included the Aboriginal Family Supervision Programme (AFSP, Perth), the Metropolitan Aboriginal Youth Team (MAYT) Mentor Programme (Adelaide), Panyappi (Adelaide), Bush Breakaway (Ceduna, SA) and the Finding Yourself Programme (Echuca, Victoria). The data indentified a number of mentoring issues which are specific to Indigenous programmes:

There needs to be sensitivity about the location of the programme - particularly if it is seen as 'too close' to government.
There may be an added need for flexibility in programme delivery if Aboriginal young people are in remote or isolated communities.
Adequate scoping and consultation is particularly important for Indigenous projects, to ensure that they will be acceptable to and engage with those communities.
Indigenous projects need to continue to engage with the Indigenous community once they are in operation. This may involve special measures to ensure participation in steering committees.
Projects need to be specifically promoted within Aboriginal communities using measures which are likely to reach those communities.
Some Indigenous projects reported that because the mentoring positions were only part-time it was more difficult to recruit mentors. There are also stronger arguments for the payment of Indigenous mentors given the lower socio-economic status of Indigenous people.
While it should not be assumed that it is always appropriate to match an Indigenous young person with an Indigenous mentor, most Indigenous projects found that it was often beneficial to match Aboriginal mentors and mentees.
Aboriginal family and kinship networks may place significant cultural requirements on how mentoring is organised.

In terms of the specific external evaluation of the AFSP in Perth, some significant results were:

All nine caregivers interviewed reported positive changes in their children as a result of the programme, and most identified positive effects on the family.
All of the 20 participants interviewed 'overwhelmingly endorsed the programme' and would recommend it to others like themselves. Of the nine caregivers interviewed, eight also thought that the programme was 'good'.

As previously noted, there is a relatively widespread view that mentoring is likely to be particularly effective for Indigenous young people. This may in part derive from the opportunity mentoring provides for Aboriginal organisations and Aboriginal people to actively participate in the process of working with Aboriginal young offenders.

Recommendations for Stage 2

A requirement of this research is to make recommendations to the CPB for Stage 2 of the project. Arising from the review of the Australian and international literature and the consultations which were undertaken, there are a number of activities which could be carried out in the future.

National workshop

Option 1: Organise a national workshop/conference for all mentoring projects and other key stakeholders (eg agencies which fund mentoring projects).

It was clear from the consultations that many mentoring projects have little contact with, or knowledge about, other mentoring projects which are operating within Australia. It could be highly productive to organise a national workshop/conference for all mentoring projects (and funders of such projects) from around Australia to come together to discuss key issues including: good practice, evaluation techniques, key research findings etc. Such a meeting would provide the opportunity to develop links between mentoring projects. A report from the national workshop could be prepared for public distribution.

An example of a similar exercise is a prior project conducted by Urbis Keys Young for the CPB. This involved organising, facilitating and reporting on a national roundtable conference on the design and management of public space, particularly focusing on youth issues (Keys Young 2000).

A good practice manual

Option 2: Develop a good practice manual.

The development of a manual which outlines good practice would be a direct and practical way to assist existing mentoring projects, as well as new projects which are in the process of development.

A good practice manual would provide hands-on advice about and examples of how to set up a programme, policy and practice issues, recruitment, training, targeting, evaluation etc.

Both Option 1 and Option 2 provide the opportunity for practical assistance to mentoring projects, particularly when the consultations revealed relatively limited knowledge of or contact between mentoring programmes. Options 1 and Option 2 could be undertaken jointly.

Evaluation

Option 3: Fund evaluations of existing mentoring programmes.

It was apparent from the consultations that most mentoring projects remain unevaluated, and that where evaluation does take place it is often internal and not rigorous in its evaluative standards.

CPB could fund one or more evaluations of existing mentoring programmes which have not been previously externally evaluated. Such evaluations would need to be longitudinal and include measures related to re-offending.

Funding existing programmes

Option 4: Provide funding to existing mentoring programmes.

A further option is to provide funding to existing mentoring programmes, preferably on an ongoing rather than pilot basis. The consultations revealed that many projects are uncertain about their funding and exist with a 'pilot' status. Such a situation is unsatisfactory in terms of developing and maintaining professional and effective service delivery. In providing funding to existing mentoring programmes, preference should be given to projects without any ongoing government funding.

Introduction

This is the Final Report on the project.

What is mentoring?

Mentoring has been defined as:

A mutually beneficial relationship which involves a more experienced person helping a less experienced person. A mentor programme for young people at risk aims to match the young people with appropriate mentors, and support the relationship over time. (ARTD 2001b, p.iv)

Background to the research

The CPB has received $8 million in funding for early intervention initiatives with young people and their families under the Youth Crime and Families Strategy. There has also been an emphasis on early intervention with young people and their families across the Commonwealth government generally, through the Stronger Families and Communities Strategy.

The CPB was aware that:

There had been some promising overseas research reporting the benefits of mentoring with young offenders and young people at risk of offending.
Various mentoring projects concerning this target group had been or were being conducted around Australia, but there was little information about them.

The Branch therefore wanted to research how mentoring was being used in these programmes across Australia, and which strategies appeared to be effective.

Objectives of the project

The project involved a national audit and review of mentoring programmes around Australia in order to:

provide a national profile of mentoring programmes for young offenders
identify models and good practice
identify key crime prevention outcomes from youth mentoring
make recommendations for Stage 2 of the project.

Structure of this report

The report is divided into the following sections:

Methodology
Organisational context of mentoring programmes
Mentors
Mentees
Evaluation
The effectiveness of mentoring programmes
Conclusion

The research for this report draws upon the following sources:

the Australian and international research collected for the literature review. Since only a small amount of Australian literature was available, most of the discussion refers to international studies. However, particular emphasis is given to the findings of the small number of Australian evaluation reports, particularly those which are external in nature. (This inevitably means that greater emphasis is placed in this report on these externally evaluated projects.)
the consultations for this project, including a postal survey, field work and interviews. The findings of the qualitative consultations were largely consistent with the findings of the overseas literature. Therefore more detailed discussion of these qualitative findings has generally been limited to:
- any areas where this was not the case
- useful examples/lessons which appear to have more general applicability
- topics of particular relevance to the Australian context eg specific issues concerning Indigenous mentoring projects.

Some issues discussed in the literature were not specifically covered in the Australian consultations. In these instances discussion is limited to the findings from the literature.

Methodology

Definition of mentoring

One of the most challenging aspects of the project was developing an appropriate and precise definition of mentoring for the purpose of the study. Projects were included only where the following five criteria were satisfied:

Mentoring is a significant or primary component of the project, and this has been defined and documented by the project itself (eg in the aims and objectives).
A significant or primary target group of the project is young people who (i) have previously committed criminal offences or (ii) have been identified as being at risk of committing criminal offences.
The target group is young people, including (but not necessarily exclusively) young people aged 18 years or below.
The mentors used are community members rather than people acting in a professional capacity such as youth workers.
The programme is currently operational.

Quite a few projects were excluded from the study because they did not meet this definition. For instance, a number were aimed at young people at risk generally, rather than those who had been specifically identified as at risk of offending.

Key components of the methodology

The project comprised the following key components:

literature review
initial scoping consultations to identify mentoring projects
postal survey
fieldwork
supplementary telephone and face-to-face interviews
meeting with crime prevention officers from around Australia
data analysis and reporting.

These are discussed in turn below.

Literature review

A review was conducted of Australian and selected overseas literature on mentoring for young offenders and young people identified as at risk of offending. Generally, material published between 1997 and 2001 was included. Literature databases were only searched for material published in this timeframe. However, a small amount of material published before and after this timeframe was identified through other methods (eg internet searches and the consultation process).

Searches were conducted in January 2002 by RAPID Services at the University of NSW (a specialist search service) and Urbis Keys Young researchers. The following Australian databases were searched:

AFPD (Australian Federal Police Digest)
AGIS (Attorney-General's Information Service)
APAIS (Australian Public Affairs Information Service)
ATSIC (ATSIC Library Catalogue)
ATSIHEALTH (Aboriginal and Torres Strait Islander Health)
CINCH (Australian criminology)
FAMILY (Australian Family and Society)
Indigenous Studies Bibliography
KINETICA (The National Bibliographic Database - holdings of major Australian libraries)
MAIS (Multicultural Australian and Immigration Studies)
Australian Domestic and Family Violence Clearinghouse (good practice database)
Australian Clearinghouse for Youth Studies (abstracts database).

The following international databases were searched:

Current Contents
Family and Society Abstracts
Index New Zealand
Index to Foreign Legal Periodicals
Index to Legal Periodicals
NCJRS (National Criminal Justice Reference Service)
Social Sciences Citation Index
Social Services Abstracts
Social Work Abstracts
Sociological Abstracts.

An internet search for relevant material was also conducted.

In addition, some literature was identified and obtained during the project consultations. This was the case with most of the Australian evaluations cited in this report, which were often either not formally published or not widely circulated.

Initial scoping consultations to identify mentoring projects

An initial scoping exercise was conducted to identify mentoring projects within the scope of the definition described earlier. This involved distribution of a standard letter in January/February 2002 (in most cases via email) to key agencies around Australia which may have been aware of mentoring programmes. (A copy of this letter is attached at Appendix A.) The letter was also distributed via various email discussion lists. These included, for instance, crimnet (criminology) and yarn (youth issues).

This consultation process was effective. Use of email meant that people were able to easily forward on the information to others. A number of emails and phonecalls were received from people suggesting projects which might potentially fall within the scope of the study. This included a number of smaller and/or newer projects which may not have been identified otherwise.

During this consultation process, informal telephone consultations were conducted with most of the projects ultimately included in the project. This material has been drawn on in the report.

The scoping process ultimately identified 24 projects nationally. 1

Postal survey

A postal survey was conducted of the 24 identified mentoring projects. The survey was distributed in March 2002, in most cases by email. Responses from 21 projects were received.2 A copy of the survey is attached at Appendix B.

Fieldwork

Face-to-face consultations

More in-depth face-to-face consultations were conducted with seven mentoring projects nationally between February and June 2002. The projects were:

Whitelion, Melbourne
Great Mates, Perth
the Aboriginal Family Supervision Programme (AFSP), Perth
Metropolitan Aboriginal Youth Team (MAYT), Adelaide
the NSW Department of Juvenile Justice (DJJ) Mentoring Scheme
One2One, Parramatta and Coffs Harbour/Clarence,3 NSW
the Bush Law Action Plan Partners project, Port Pirie, SA.

Consultations were conducted with a range of informants, including:

field staff
key external stakeholders who could comment on the project
mentors
young people who had participated in mentoring projects.

In most instances the consultations were conducted in groups.

There were some constraints concerning who could be consulted for the project. These constraints included, for instance, lengthy ethics clearance processes and the fact that some key people were unavailable at the time the consultations took place. For some projects access to young people would not have been permitted in any circumstances.

Ultimately, consultations were conducted with:

field staff for all seven projects
external stakeholders for six projects
mentors for five projects
young people for two projects. A small financial incentive ($25) was provided to the young people to encourage their participation.

Copies of the question guides used for consultations with project staff, external stakeholders, mentors and young people are provided at Appendix C.

Supplementary telephone and face-to-face interviews

Telephone and face-to-face interviews were conducted with selected key informants to supplement the data from the field visits. The question guide for stakeholders (see Appendix C) was used for these consultations.

Meeting with crime prevention officers

Crime prevention officers from around Australia were consulted face-to-face at one of their regular meetings in Sydney in February 2002.

Data analysis and reporting

The final stage of the project involved data analysis and reporting. The data analysis included:

entry of the quantitative data from the surveys into a database, and processing of the results
analysis of the qualitative data from the survey, field work and telephone interviews.

Some caveats

Three caveats should be noted about the project:

Three projects did not provide a completed survey. Therefore data from these projects is not included in this report, other than in relation to the location of mentoring projects (unless information about the project was available from another source).
This report provides a 'snapshot' of mentoring projects identified in early 2002. However, there appears to be a high turnover of mentoring projects in Australia, with both new projects being established and existing ones closing. For instance, the consultants are aware that at the time of writing this report three of the projects included in the study are no longer operational,4 and two other new mentoring projects have been established.5 The 'snapshot' provided could therefore potentially be rather different in say 12 months. However, given that Australian mentoring projects were found to have many similarities in relation to their core features, the general findings of this study are likely to hold true.
Only a small amount of Australian literature exists on mentoring for young offenders. Most of the projects included in the study have not been subject to formal evaluation (particularly external evaluation). Of those which have been evaluated, not all of the reports could be obtained.6 The consultations conducted for this project cannot be regarded as 'evaluation' of the projects. The lack of evaluation limits the extent to which definitive conclusions can be drawn about the most effective way to conduct mentoring projects in this country. The discussion of good practice in this report should therefore be understood in this context.

Notes

1 This included the One2One project which was operating in two different locations. One2One was treated as two separate projects for the purpose of analysis because of differences between the two programmes.
2 See Footnote 1 above.
3 The Coffs Harbour site was not visited, but stakeholders in Sydney discussed both sites of the programme.

4 The Parramatta site of One2One in NSW, the Bushlaw Action Plan Partners project in Port Pirie, SA, and the Brayton Mentoring Initiative in Shepparton, Victoria.
5 As at January 2003 these are: the Education Queensland Pilot Mentoring Projects funded by the Department of Education, Science and Training, Queensland (which operate in seven schools around the State, including in two schools in youth detention centres ie Brisbane Youth Detention Centre and Cleveland Youth Detention Centre in Townsville); and the Mates and Legends Programme run by the Programmes Branch of the Young Men's Christian Association in Alice Springs for young offenders and young people at risk of offending (although no clients have been seen to date).
6 Examples of reports which could not be obtained include the full report on the external evaluation of the NSW DJJ Mentoring Scheme, which has not been publicly released. Only the Executive Summary of this report was made available, and is referred to in this report. Internal evaluations of Great Mates in Perth, and the Family and Youth Services (FAYS) Mentor Programme within the Department of Human Services (DHS) in South Australia (SA), were also unavailable.

Organisational context of mentoring programmes

Location of mentoring projects

A total of 24 mentoring projects were identified around Australia. The number of projects in each state and territory varied considerably:

Almost two-thirds of the projects were located in two states, Victoria (nine projects) and South Australia (six projects each).
Several other jurisdictions had two or three projects each: Western Australia (three); New South Wales ( three, including two sites for one programme); the Northern Territory (two); and the Australian Capital Territory (one).

There were no projects identified in Tasmania or Queensland (although it is known that two projects are currently being established in Queensland).

There are a number of possible reasons for the marked variation in the number of mentoring projects across states. These include:

differences in the perceived 'fit' between mentoring programmes and the overall policy emphasis and priorities in particular states (particularly by key funding agencies). This was apparent from the meeting with crime prevention officers from around Australia for this project. Some informants also suggested that one of the reasons for the large number of mentoring projects in Victoria was the strong policy emphasis on early intervention for young offenders in that state.
the perception that mentoring may be particularly appropriate for certain sub-groups of that state's population. For instance, three Indigenous-specific mentoring projects are located in SA, all of which have government funding (Bush Breakaway in Ceduna, and Panyappi and the Metropolitan Aboriginal Youth Team (MAYT) Mentor Programme in Adelaide).
previous experiences with mentoring projects in that state. For instance one government funder reported that it would have some reservations about funding another mentoring project specifically for young offenders given its past experiences with funding projects of this nature.
the interest in mentoring developed by key individuals. For instance, consultations indicated that some mentoring projects - particularly smaller, community-based projects - have been set up on the initiative of one key person.

The remainder of this report refers only to the 21 mentoring projects from which survey responses were received (except where information on the project was available from another source).

Auspicing organisation

The majority of mentoring projects are run by community organisations (62 per cent).
Government agencies are the sole or primary funding source for most mentoring projects (67 per cent). No projects were funded primarily by non-government agencies. Initiatives that did not receive funding from the government stated that their funding came from a mix of sources. One service noted that its mentoring project was provided on a fee-for-service basis.
Government agencies that currently fund mentoring programmes cover a variety of different sectors. For instance they include organisations concerned with justice/juvenile justice, crime prevention, child and family welfare and health issues. This reflects the range of different programme areas that mentoring is seen as relating to. (This is commonly the case with early intervention programmes.)
Forty-eight per cent of projects reported that their initiative was a pilot or trial programme, 29 per cent stated that their project was running for a fixed time (some projects nominated both these options), and 43 per cent stated that their project was an ongoing one.

The literature does not provide guidance on the implications of different auspicing organisations and funding sources, other than noting that many of the young people that mentoring projects of this nature seek to reach as mentees are from low socio-economic backgrounds. A fee-for-service mode of provision where the young person or their family is expected to meet the costs may reduce or preclude access for some young people in need. However, a different, and less problematic, fee-for-service model is where government contracts non-government organisations to provide individual mentoring services on a fee basis. This is a common model used for Australian mentoring programmes.

The literature also notes that to be successful, mentoring programmes require organisational autonomy (Jurik et al 2000, p.315). Many of the Australian programmes appear to have organisational autonomy, although there are some mentoring projects that are run under the auspices of various state government departments. Some issues were raised during the consultations about the potential complications this could raise, particularly where the agency has statutory or legal authority over the young person. These include:

potential dilemmas for mentors as they are mandatory notifiers of child abuse in some programmes:

This is very difficult for mentors, since they're working very closely with the young people. They're saying 'I'm your friend but I can tell on you'. We've really had to work hard with our mentors about what their statutory obligations are. There's also a cultural element - there's often mistrust of the welfare sector [in the Aboriginal community]. (Indigenous Mentoring Programme)
whether participation by the young person can be regarded as truly voluntary when this is pursuant to a formal court order (see below).

However, auspicing by government agencies does not seem to have caused any significant problems for Australian mentoring programmes. All the government-sponsored programmes in the study received higher levels of funding than those run by community organisations and reported similar project strengths and issues compared to non-government programmes. It was also clear that - perhaps not surprisingly - the programmes that have been in operation the longest time tend to have ongoing government funding.

Nonetheless, from the consultations it appeared that it is preferable for a mentoring programme to at least have the feel of being a community-based, youth-friendly organisation, and be perceived as such by young people and others. This is regardless of whether the funding source is government or otherwise.

More successful mentoring projects seem to be viewed by young people and others as being 'separate' and 'different' from the 'standard' agencies which might be working with young people such as juvenile justice and child welfare. This makes the project more welcoming to young people, and differentiates it from those other agencies. From the consultations with mentees it was clear that this is particularly critical given that they have often had contact with multiple workers from numerous agencies over considerable periods of time. It is important to counteract the view that this will be just another agency who's going to tell them what to do.

This 'community-based' feel of a mentoring agency might extend, for example, to the nature of the premises selected. As noted in the literature, it is important for mentoring programmes to be established in premises where young people feel comfortable and welcome, which are accessible by public transport, and where there is sufficient meeting and activity space for events and meetings. It is also important to consider any associations or inferences that may be drawn from selection of premises, as young people may not feel comfortable in premises located near certain organisations, such as the police (Benioff 1997, p.20).

An example of an innovative youth-friendly choice of premises is the Bushlaw Action Plan Partners project based in Port Pirie, SA. This initiative has premises that include a converted old train, which also houses a restaurant used for employment training of young offenders. The MAYT Mentor Programme in Adelaide also reports that it is housed in a kid friendly building, without any security arrangements required to enter.

The attitude and approach of staff and the organisation generally can also promote the 'community-based' feel of a mentoring organisation and again differentiate it from 'standard' agencies. For instance the MAYT Mentor Programme in Adelaide reported that it provides a very welcoming and informal environment to young people through means such as:

adopting a general philosophy that kids are first, process is second
always ensuring that when they attend the premises someone here will always find something to do with them
providing showers and toiletries for the young people to use, since, for example, there may be no power on at their home
providing food at lunchtimes.

Through approaches such as these, MAYT gives the message to say we care, and lets it be known that young people are welcome to drop in at any time. The project also observed that their clients don't see us as part of the criminal justice system.

It may also be beneficial for mentoring projects funded by government to 'downplay' this fact. For instance, the Panyappi project in Adelaide is auspiced by MAYT. MAYT is part of DHS but has experience working at the community level. Mentors with Panyappi have reported that association of the project with the state government is:

sensitive with the [Aboriginal] community, and that when introducing the project, emphasis is placed on the project and its co-location with MAYT rather than the government connection. (Crime Prevention Unit, SA Attorney-General's Department 2002, p.10)

Funding

Under-funding is a major issue for Australian mentoring projects, with more than half of those surveyed citing insufficient funding or a lack of resources as a significant weakness in their programme, and a serious challenge to the programme's operation.

Sherman et al (1998) estimate the cost of mentoring (if mentors are volunteers) to be $US1,000 (around $A1,800) per match, including staffing, infrastructure and procedures (Grossman & Garry 1997, p.5). On the other hand, Peel and Calvert estimate the cost to be £2,000 (around $A5,600) per young person (Peel & Calvert n.d.).

Funding levels in Australia for current mentoring projects range from $20,000 - $500,000 annually. The average amount of annual funding is $125,962. When projects were asked if they were confident about the continuation of their funding over the next two years, 71 per cent reported that they were confident, while 29 per cent stated that they were not.

Funding insecurity has clearly impacted on the development of mentoring programmes in Australia. For instance, there is a tendency for mentoring programmes to be established and then vanish, to concentrate on their most immediate programme goals, and to place little emphasis on evaluation. It also appears that mentoring programmes can take a comparatively long time to effectively establish. This may mean that newly-established programmes can be in danger of not getting fully 'off the ground' before their funding ceases.

Formal versus naturalistic mentoring programmes

The literature refers to mentoring projects as being either 'formal' or 'naturalistic':

Formal programmes tend to be more regimented in nature and involve the mentor and young person completing designated tasks together. Some formal programmes are more focused on the juvenile justice or education systems, and involve activities such as the completion of homework, job applications or the requirements of juvenile justice orders, attending appointments and so on. Australian examples include the mentoring programmes run by the WA Ministry of Justice and the NSW DJJ, which both focus on completion of specific tasks in case plans. Another example is the Big hART programme in Melbourne, which solely comprises activities based around the production of artistic materials.
Naturalistic programmes take a more informal approach to the development of the relationship and activities that the mentor and young person undertake together. For example, they might go for a coffee together, do sporting activities, or just 'hang out' together. Examples in Australia include Great Mates in WA and One2One in NSW.

The conceptual philosophy of the initiative (ie formal versus naturalistic) will usually develop from the nature of the programme and its objectives, and may also be influenced by the auspicing or funding organisation (Ave et al 1999, p.79). Formal programmes will tend to be more regimented and structured in nature.

Some Australian programmes have both formal and naturalistic elements. An example is the Bushlaw Action Plan Partners project in Port Pirie, SA, which is funded under the Commonwealth JPET (Job Placement, Employment and Training) initiative. Activities include employment-related activities such as completion of job applications (formal), as well as general 'hanging out', meeting for coffees etc (naturalistic).

There is no evidence from the literature or the consultations concerning whether formal or naturalistic programmes are more effective, other than it appears important that formal programmes avoid simply replicating a 'youth worker' or conventional worker role. This can be a temptation for mentoring programmes, particularly those run by agencies with some statutory or other formal responsibility for the young person, such as juvenile justice.

For instance the external evaluation of the mentoring programme run by the NSW DJJ observes that this programme, as well as the AFSP in Perth and the programme run by FAYS in SA:

have significant differences from conventional mentor programmes, as they set up short term relationships within the period of the order, focus on tasks in the case plan, and employ para-professional 'mentors' rather than use volunteers. (ARTD 2001b, p.v)

The report notes that in retrospect the aims and objectives of the DJJ scheme encompassed two different functions. These were the provision of both:

a traditional mentoring relationship
para-professional support (of a practical, emotional and cultural nature) to clients within their case plans.

One of the key conclusions of the report was that the programme had become primarily directed towards the latter rather than the former function, ie providing case support rather than mentoring (ARTD 2001b, p.v).

Programme objectives

From the outset, a mentoring project should have well defined objectives, a clearly stated mission, and established operating principles (involving designation of tasks, accounting principles etc). Ideally, these should be developed in consultation with potential participants and stakeholders. Programme plans should be realistic and attainable, and maintain a degree of flexibility as the programme develops.

A mentoring project should establish a clear conceptual philosophy regarding the nature of the programme, and whether it will take a formal or naturalistic approach to mentoring.

The overseas literature states that it is important for mentoring programmes to have clear and realistic objectives. It is important that objectives for any mentoring programme be developed at the outset of the programme, so that the programme may be tailored to accommodate them (Youth Justice Board n.d.; Mentoring Australia 2000). These objectives or statements of purpose should ideally be developed in conjunction with stakeholders and participants, and will differ depending on the nature and goals of the programme (Mentoring Australia 2000).

Both overseas and in Australia, mentoring programmes for young offenders and young people at risk of offending typically have aims such as:

reducing crime and other anti-social behaviour by young people
reducing other problematic behaviours, such as drug and alcohol use
improving self-esteem, social skills etc.

Programmes also typically aim - either implicitly or explicitly - to achieve long-term and sustainable changes in these areas.

From the objectives provided by survey respondents, it appears that the majority of Australian mentoring projects have recognised the need for clear programme objectives. The external evaluation of the NSW DJJ Mentoring Scheme highlights some problems experienced by the programme in relation to this issue. The programme aimed to provide both a conventional mentoring relationship and a case support function. This created some 'confusion' in how the programme was implemented in practice (ARTD 2001b, p.v).

Among the Australian mentoring programmes surveyed, some had simple, forthright statements as objectives such as:

To reduce a young person's need to offend, to facilitate the use of social skills in the community, and to ensure young persons can access and utilise resources available in the community. (WA Department of Justice's Juvenile Justice Mentoring Programme)

On the other hand, some programmes had more complex and ambitious objectives such as those for the Panyappi programme in Adelaide:

to promote self-discovery and self-determination by young people participating in the project and family and community.

While many project objectives were general in nature and focused on the need to reduce reoffending or anti-social behavior by young people, some were more specific and directly based on individual programme activities. An example is the objectives of the Big hART programme, run from the Parkville Detention Centre in Victoria, which include:

Mentoring young women in Parkville to produce profound artworks in film, music and new media that can represent them and their issues to local, state and national forums.

Along with a well-defined mission statement, established operating principles are central to a programme's success (California Mentor Initiative 2000, p.1; Mentoring Australia 2000). These principles should ideally cover a broad range of issues and provide clear guidance for staff, project managers and stakeholders on key aspects of the project's operations (California Mentor Initiative 2000, p.2). The Youth Justice Board in the UK suggests that some of the key principles that a service should establish prior to operation include:

a tightly defined target group
sufficient trained staff to run the programme and provide support to participants
clear guidelines relating to the recruitment and screening of mentors
well-developed links with a range of local agencies which can support the work of the programme (Youth Justice Board n.d.).

Flexibility and adaptability are crucial for any mentoring programme. Programmes need to be imaginative and thoughtful in overcoming barriers and altering the programme to best serve the target group (OJJDP 1998, pp.18-22). A number of the mentoring projects examined for this study reported that they had changed or modified different aspects of their work or strategies as it became apparent that certain approaches worked better than others. The external evaluation of the NSW DJJ Mentoring Scheme concluded that one of the strengths of the programme was that it had enabled:

the Department to support clients around the state with very diverse needs and in very different circumstances, including a large number of Aboriginal clients. It has been a flexible service achieved by an open programme definition, and the creative responses of staff to meeting the needs of clients. (ARTD 2001b, p.vi)

Any programme plan must be realistic and easily attainable, with full descriptions of the role of each participant, an assessment of need, and clear goals, objectives and timelines for all aspects of the programme (Mentoring Australia 2000). Programmes should also ensure they have written policies and procedures covering issues such as rights, responsibilities, confidentiality, legal issues, insurance, duty of care, Equal Employment Opportunity (EEO), sexual harassment, grievance issues, and ethical issues (Mentoring Australia 2000).

Stand-alone versus integrated projects

Mentoring projects which are integrated into a range of other services are more likely to be effective than those which are stand-alone.

Mentoring projects can be either stand-alone initiatives, or integrated into a range of services offered by the auspicing organisation.

Many programmes, both in Australia and overseas, are integrated into other service offerings. Mentoring is therefore a component which supplements or enhances other project elements. Common examples of additional elements, both in Australia and overseas, include:

drug education programmes
life skills programmes
career or job-seeking techniques
educational/academic assistance.

Mentoring projects that are linked to juvenile justice agencies/facilities tend to feature more of these additional elements, such as education, psycho-social assessment, violence reduction programmes, parent education sessions, supervised recreational activities and skill-building activities (Mathieson 1997, pp.110-115; Howitt et al 1998, p.39). Again this is a feature of both Australian and overseas programmes.

In projects where mentoring is linked with other services, the mentors will often specifically concentrate on building on these other programme elements through tutoring, or positive reinforcement (Ware & Lucas n.d., pp.11-12).

In the United States for example, the Juvenile Mentoring Programme (JUMP) is tied in with a range of other programmes under the SafeFutures Initiative. These other programmes include after school, weekend and summer programmes, mental health services, family strengthening and support services, delinquency prevention programmes, tutoring, and the provision of drug treatment centres (Morley et al 2000, pp.10-11).

In Australia, many of the mentoring programmes surveyed incorporated casework, groupwork and counselling. Some also included activities such as:

employment or skills training eg the Bushlaw Action Plan Partners project, Port Pirie, SA; the Whitelion Mentoring Programme in Melbourne; the NSW DJJ Mentoring Scheme; and One2One in NSW
residential components, wilderness activities and bush camps eg the Bush Breakaway Youth Action Programme in Ceduna, SA; the New Trax Programme in Melbourne; the NSW DJJ Mentoring Scheme; and the Bushlaw Action Plan Partners project in Port Pirie, SA.

For some programmes, mentoring takes place on a more informal basis, rather than the formal basis used by traditional mentoring projects. For instance, as discussed below, this is the case with the New Trax Programme in Melbourne.

Another example is the Whitelion programme based in three juvenile justice institutions in Melbourne, which has developed a role-modelling programme (Lemmon and Bowen 2001). This programme involves visits by high profile personalities from the sports, arts and business communities, working with inmates on a group basis. However, informal one-to-one mentoring relationships between the 'role models' and the inmates have on occasion developed as a result of this experience. It was reported that such relationships have an advantage in that they evolve naturally. Having seen the benefits resulting from informal mentoring, Whitelion has recently established a formal one-to-one mentoring programme.

The external evaluation of One2One in NSW strongly suggests that stand-alone mentoring projects may experience greater difficulty getting off the ground than integrated projects. The project had trouble recruiting adequate numbers of young people (particularly in Parramatta), and lack of effective networks with key referral agencies was felt to be the most significant reason for this. It is of note that the project was run by the Young Women's Christian Association (YWCA), which had experience working with at-risk young people generally (under its Big Sisters Big Brothers - BSBB - programme),7 but not with the specific sub-group of young offenders.

The literature also suggests that services other than mentoring should be available to the clients of mentoring programmes, either via that or other organisations. Additional community support services, or interagency support is noted as an important factor in a responsible mentoring programme (California Mentor Initiative 2000, p.1).

There was clear agreement among the informants for this project that integrated mentoring projects are also more likely to be effective than those which are stand-alone. This is particularly critical given that many mentoring programmes appear to take a while to establish, and typically only have one or a small number of staff. Informants felt that it is more cost-effective, quicker and easier to establish a mentoring programme if there are already established infrastructure, administrative and professional support and networks with key agencies and potential clients. Therefore:

less establishment time is required
agency networks are already in place. This includes agencies both to act as referral sources and as referral points for clients requiring specialist services once they enter the mentoring programme
potential clients may be more likely to agree to participate in the mentoring programme.

For instance, the MAYT Mentor Programme in Adelaide estimated that around half of its mentoring clients have come to the project via various other youth programmes run by the organisation. Since those clients knew and trusted the staff, they were more receptive to the suggestion that they participate in the mentoring project than if approached 'cold' by another agency.

Another factor that was stressed in the consultations was the need for mentoring programmes to establish effective referral networks eg drug and alcohol services, mental health services, and educational institutions.

Knowing the key person in each of those agencies is crucial, so you can cut through stuff and eg avoid waiting lists. (Mentoring Programme)

Staff

Staff running mentoring projects should be appropriately skilled and have clearly written job descriptions.

Strong overall coordination of a mentoring programme is essential to its success.

Although mentors provide the most obvious labour undertaken for mentoring programmes, the role of the programme staff is critical. These staff require effective organisational and people skills in order to:

select appropriate mentors
provide ongoing monitoring of mentoring relationships once established
develop and monitor an effective network
provide strong structure and coordination of the programme.

The overall coordination role can be a quite labour-intensive task to undertake effectively and must be considered when developing and funding mentoring projects.

The literature states that staff running mentoring projects should be appropriately skilled, with clearly written job descriptions. In hiring, programmes should adhere to EEO principles (National Mentoring Working Group 1991, p.1; California Mentor Initiative 2000, p.1). Cultural sensitivity and understanding should also be primary considerations in staff selection (Rogers & Taylor 1997, p.131; California Mentor Initiative 2000, p.1).

For continuity, it is generally recommended in the literature that full-time rather than part-time staff run the programme, as full-time staff are more accessible and will have an increased awareness of current issues (Jones-Brown & Henriques 1997, p.224). However, this is more likely to exclude female applicants with family responsibilities.

Benioff outlines some key positions that no project should be without, including a project director, an administrative assistant, a mentoring coordinator, and an educational coordinator (Benioff 1997, p.19).

Nonetheless in Australia, this list of essential key positions seems somewhat of a luxury. Staffing is a significant issue for Australian mentoring projects, with many only having an average of one full-time staff member and an average of five part-time staff members. On average, each project only had one full-time or equivalent staff member, who tended to be the project coordinator.

Many projects commented that a lack of resources prevented them from hiring more staff, or that a lack of staff resulted in inadequate levels of contact between staff and participants. One project with only part-time staff commented that this placed significant limitations on its operation, while another commented that it felt the project would be more effective if two full-time workers could be employed.

This suggests that Australian mentoring programmes, particularly those on a larger scale, tend to be under-staffed. Some of the mentoring projects consulted for this review argued that this was the case. This was also a finding of the external evaluation of the Brayton Mentoring Initiative in Shepparton, Victoria (ARTD 2001a, p.A24).

Given that in Australia there is often only one staff member running the whole mentoring programme, it is perhaps not surprising that consultations indicated that the skills and capacity (or otherwise) of the project coordinator can make or break a mentoring programme. In other words, a programme is much more likely to be successful if it has a coordinator who is skilled, enthusiastic, develops effective networks, works well with mentors and young people, and provides effective overall coordination.

Establishment and administration

Establishing mentoring programmes involves effective scoping to determine need. It also requires effective consultation with specific groups, in particular Indigenous people and their organisations.

Strong organisational administration and infrastructure are essential for a mentoring programme to operate effectively. Accountable and appropriate policies, practices and codes of conduct dealing with a range of issues need to be developed and put in place.

Initial scoping

According to the literature, it is important that mentoring programmes are not hastily designed or implemented (Jones-Brown & Henriques 1997, p.228). Programmes should be adequately scoped and researched in advance.

This is clearly demonstrated by the experience of One2One in NSW. The external evaluation report concluded that inadequate scoping and poor planning of the project had been a key reason behind the slow development of the project following implementation. For instance:

Both sites were trying to cover geographical areas that were too large.
The Coffs Harbour/Clarence area had a high proportion of low income and unemployed families, making it difficult to recruit volunteers.
The Parramatta area had a large commuter population and much smaller residential population, making it more difficult to find mentors who lived in the area (ARTD 2002, pp.62-63).

Scoping and consultation for Indigenous projects

Consultations indicated that adequate scoping and consultation is particularly important in relation to Indigenous projects, to ensure that they will be acceptable to and engage effectively with those communities (see Crime Prevention Unit, SA Attorney-General's Department 2002, p.10). For example, it was reported that there was extensive consultation with the Aboriginal community before implementation of the NSW DJJ Mentoring Scheme. Originally Indigenous young people constituted the sole target group for the project, and although this is no longer the case, they still comprise some 40-60 per cent of the programme's clients (ARTD 2001, p. vi). As noted above, the external evaluation of the scheme found that the programme had improved services to Aboriginal clients.

The Panyappi project in Adelaide is another project targeting young Indigenous people. Although the project has experienced a number of problems which have hampered its operations, one of the positive aspects of the programme has been its strong support in the Aboriginal community (Crime Prevention Unit, SA Attorney-General's Department 2002, p.20). As with the NSW DJJ scheme, there was extensive community consultation undertaken during the development of the project through its Advisory Group (which included various key Aboriginal community groups and organisations) and more informal mechanisms.

The Advisory Group was also given as much decision-making authority as possible aside from funding and budgetary considerations (p.9). The input of the group was important in shaping the operationalisation of the project, through a variety of measures such as:

redrafting the project objectives to more realistically reflect community needs and achievable outcomes
changing the name of the project from the Indigenous Youth Mentoring Project to Panyappi (Kaurna language for younger brother or sister), to provide the project with a more positive identity for the project staff and its branding in the community (p.10).

It is also important for Indigenous projects to continue to engage with the Indigenous community once they are in operation. This can provide some challenges in practice. For instance, the external evaluation of the AFSP in Perth found that there was little awareness of the programme among both Aboriginal Elders and the Aboriginal Affairs Department. There had also been difficulty getting both involved in management of the project, and attendance at Steering Committee meetings had been poor. In addition, most members of the metropolitan Commission of Elders had no knowledge of the programme (Social Systems and Evaluation n.d., p.17). The evaluation concluded that the Aboriginal community had not been satisfactorily consulted or fully informed about the project.

It was reported that the Steering Committee for the AFSP was struggling to find the most constructive way of involving the Aboriginal community in the programme. This was felt to be important both in itself and as a means to facilitate the recruitment of mentors (Social Systems and Evaluation n.d., pp.iii-iv). The report notes that one approach which might be considered is to have Elders only attend one or two special Steering Committee meetings annually at which they can be briefed on progress and asked for advice on particular issues, rather than be expected to attend all Committee meetings when many of the issues covered are administrative matters of no great interest to them (p.18).

Strong support structure

According to the literature, a strong and well coordinated support infrastructure is also necessary for effective mentoring programmes (Singh & White 2000, p.29; Foster 2001, p.10). For example, the inconsistency of mentor record-keeping has been identified as a problem that can reduce the effectiveness of a programme (Singh & White 2000, p.29).

The available literature indicates that an inadequate management and support structure is one of the most common difficulties experienced by mentoring programmes in this country. For instance, one of the key problems identified by the external evaluation of the NSW DJJ Mentoring Scheme was substantive administrative difficulties, which have gradually been overcome in more recent years. These problems in part stemmed from inadequate state-wide management of the programme. This had led to marked differences in implementation by different offices and individual case managers, limited performance measurement and insufficient integration of the programme into the case management framework (ARTD 2002, pp. vi-vii).

Likewise with other Australian mentoring programmes:

Problems with inadequate case management, recording and accountability were identified by the external evaluation of the AFSP in Perth (Social Systems and Evaluation n.d., pp.20-21).
Deficits in management were noted in the external evaluation of One2One in NSW, given that it was being managed by the Sydney-based YWCA (ARTD 2002, p.63).

There are significant risks inherent to any mentoring programme. These include, for instance, risk of abuse, parental hostility, or potential feelings of rejection if a mentor fails to meet commitments. It is therefore crucial to establish accountable and appropriate policies, practices and codes of conduct on a wide range of issues (Ave et al 1999, p. i; National Mentoring Working Group 1991). These policies and procedures should respect the preferences of the young people and their families (Morley et al 2000, p.21; Grossman & Garry 1997, p.4).

Initial meetings

The initial meeting between mentors and mentees should be used to establish a number of elements regarding the relationship eg the procedure if someone cannot attend a meeting, when and where meetings will occur, and so on.

A project must decide how much information the mentor will have concerning the mentee prior to an initial meeting, taking into consideration the aims of the project and the safety of all participants.

The overseas literature recommends that the first meeting between the mentor and young person should be used as an opportunity to set some ground rules about the relationship. These tend to be basic issues such as reaching agreement on how often meetings will occur, keeping conversations confidential, calling if someone cannot make a meeting, and other issues that the young person may wish to raise (Benioff 1997, pp.44-45).

The consultations indicated that this is consistent with Australian practice also. As well as establishing ground rules, the initial meetings between a mentor and young person will typically involve getting to know you activities and conversations, making the young person feel comfortable, developing trust and so on, before addressing more sensitive or complex issues at subsequent meetings.

It was reported, however, that there can be significant differences between how young men and women respond as mentees during these initial meetings. As discussed in more detail below, it was felt that young women can be more receptive to opening up and talking about their feelings and issues at an earlier stage of the mentoring relationship than young men.

The literature notes that some mentoring relationships will initially encounter difficulties, and this was also apparent from the consultations. For instance, the mentee or their parents may be hostile, the mentee may behave badly, or the pair may not get along well. In these circumstances the literature states that it is important that staff members be supportive of both the mentee and mentor, and work towards a solution that is appropriate for everybody. This may require extra support or even encouragement for the mentor, additional meetings, or a rematch (Benioff 1997, pp.45-47).

There is some contention in the literature concerning whether or not mentors should be made aware of their mentee's complete criminal or behavioural history. Some authors argue that mentors should always be informed or made aware of the nature and type of their mentee's offending history (if any) (Ave et al 1999). Others such as the Mentoring Knowledge Base suggest that this is not always advisable, and it may be better to start with a clean slate, only informing the mentor of any risk issues associated with their young person (Mentoring Knowledge Base n.d.).

Length of contact

To be fully effective, the mentoring relationship should be sustained over a long period of time, and it is recommended that relationships last for a minimum of six months, with an ideal length of 12 months. Continuity and length are significant factors in the success of a relationship.

The literature and consultations indicated that typically overseas and Australian mentoring projects stress that mentors must be willing to commit for a reasonable period of time (regardless of the length of the intended relationship), and will not accept mentors if they are unable to commit to the designated timeframe required (Crowley & McIntyre 1999, p.14).

Mentoring projects both overseas and in Australia usually intend that the relationship will last for a set period of time. However, the literature and the consultations indicated that there are differing views concerning the ideal length of a mentoring relationship. Aside from the general statement that contact should be regular and consistent (California Mentor Initiative 2000, p.1), there is no clear evidence concerning this issue in the literature.

However, it is assumed by many overseas authors that long-term commitment is more likely to produce better outcomes for young people (Barron-McKeagney et al 2001, p.135). A year is recommended by many overseas programmes (Rogers & Taylor 1997, p.129), and the Youth Justice Board in the UK also recommends mentoring relationships be supported over a sustained period of time (Youth Justice Board n.d.).

There is some evidence in the overseas literature that young people in longer-term relationships are perceived to receive greater benefits from their mentor (DuBois & Neville 1997, p.233). Conversely, short-duration matches (lasting, say, a few months) can be a contributory factor when a programme or match has been unsuccessful (Foster 2001).

Nonetheless, not all programmes with positive results have run for a long period of time. For instance, the Challenge for Change project in Wellington, New Zealand (NZ) runs for only 20 weeks, with improvements noted about two-thirds of the way through (Cardy 1999, p.7).

The length of matches amongst Australian mentoring programmes generally ranges from around three months up to an unlimited amount of time. Some programmes set a standard maximum initial period of time (or number of hours) for the mentoring relationship to develop, and then review the relationship to determine whether it should be continued.

Many Australian programmes operate on the basis that matches should last for three to four months, with a number of the programmes commenting that they see this as an ideal length of time.

Alternatively, there are some Australian programmes that operate mentoring relationships for a longer duration, ranging from 12 months to an unlimited length of time. These programmes regard a longer period of time as being more beneficial to the young person.

The general view among many Australian mentoring projects is that the length of the relationship will tend to be variable, depending on a range of factors such as the needs of the young person, the resources available to the project, the nature of the relationship, and the levels of staff turnover.

Australian mentoring projects that are more likely to set shorter periods for contact (or to be more exacting about specifying the duration of contact) tended to have both of the following features:

more formalised projects run in connection with juvenile justice operations. This is the case overseas also (see eg South Dakota Department of Corrections 2001, pp.1-4).
projects which use paid rather than volunteer mentors. Funding constraints appeared to be a factor here.

Frequency of contact

The frequency of contact will largely be determined by the nature of the project and the activities undertaken by the pair. Contact must be regular and consistent, with the ideal frequency of contact suggested as once per week.

The general view from both the literature and the consultations is that mentors need to have frequent and consistent contact with mentees, preferably once per week.

In terms of the ideal frequency of contact, many of the general mentoring guidelines from overseas suggest only that contact between the mentor and the mentee be regular and consistent (National Mentoring Working Group 1991, p.1; Barron-McKeagney et al 2001, p.135). However, the overseas literature is more prescriptive, recommending that contact should be once a week for periods ranging from one to six (but typically several) hours (Youth Justice Board n.d.; OJJDP 1998, p.16; de Anda 2001, p.99; Sherman et al 1998; p.129; South Dakota Department of Corrections 2001, p.2). This was also the view of the Australian mentoring projects consulted for this study.

According to the overseas literature a high level of contact (at least 12 hours per month) is one of the key elements contributing to positive results (Morley et al 2000, p.21; DuBois & Neville 1997, p.232). For instance, the evaluation of the American Across Ages programme found that the level (or number of hours per week) of mentoring had a significant impact on mentees. Mentees who had spent six hours per week or more with their mentor demonstrated positive differences on a number of measures, such as rates of absence from school (Foster 2001, p.24).

In Australia, an external evaluation of the Bushlaw Action Plan Partners project in Port Pirie, SA concluded that some participants were disappointed with their Action Plan Partners (mentors), and infrequent contact appeared to be the significant factor here. For instance it was reported that some Partners do not seem to contact their clients and are difficult to contact when the client needs them (Scholz 1999, p.18). The evaluation concluded that it was important for Partners to be proactive and ensure that they have regular contact with clients and are available to them (p.26).

Another factor that is stressed in the literature is that the continuity of the mentoring relationship is the most significant way in which the mentor can support the mentee, rather than rigid requirements concerning frequency of contact per se (Ellis 1997, p.54). This is regarded as a key element in a successful mentoring relationship.

Among the overseas programmes reported on in the literature, mentors typically meet with the young person for between two to four visits per month for several hours (Howitt et al 1998, p.46; Grossman and Garry 1997, p.4).

This is consistent with the experience in Australia. Around a third of the mentoring programmes surveyed reported that their mentors and mentees meet once a week or more, around half meet approximately once a week, and the remainder meet two to three times per month.

According to both the overseas literature and the Australian consultations, there may be a correlation between the length of the mentoring relationship and the frequency of contact. That is, projects that last for a short amount of time will frequently have more intensive contact over that period, whereas relationships that last for more extended periods may involve less frequent meetings. Of the Australian mentoring projects surveyed, the programmes reporting that contact occurs at least twice a week tended to be programmes with relationships running for six months or less.

Parental involvement

Mentoring projects should engage parents or guardians of the mentee in the project where possible, including obtaining their permission, and informing them as to the nature and extent of the mentoring relationship.

The overseas literature regards it as preferable to attempt to involve the young person's parents or guardians in the project, at least to some degree (Benioff 1997, p.34). Nonetheless this literature indicates that parental involvement in mentoring projects can vary. While some projects only require parental written consent, others try to include parents in all aspects of the project, including selection and approval of the mentor and attendance at regularly scheduled activities. Most overseas projects operate somewhere in the middle of this spectrum (OJJDP 1998, p.17).

Some overseas projects arrange specific meetings for the mentors and parents of the mentees, to make the parents feel included and to provide them with more information about the programme (although these may not always be well-attended) (Benioff 1997, p.48).

Regardless of the level of involvement, the family and/or parents of the mentees need to be informed and supportive of the programme for it to be fully effective and operate in the intended manner (Anon 1999; National Mentoring Working Group 1991, p.1). Consent by the family or guardian of the mentee is essential, and the literature recommends that no programme should operate without ensuring this (California Mentor Initiative 2000, p.1).

However, the overseas literature also reports that parental involvement is difficult for many projects to achieve. This is for various reasons, including a lack of interest from mentees' parents or guardians, a lack of resources from the programme, or the belief of a range of participants (parents, mentors, staff and mentees) that the project is an activity for the young people alone.

Parental involvement was found to be the least successful aspect of the Dalston Youth Project in England. A number of mentors reported that parents were suspicious of them, saw them as an authority figure (or as a representative of the justice system or social services), and found meeting with parents generally difficult (Tarling et al 2001a, p.27). It was suggested that project staff could do more to pave the way for mentors in terms of parental involvement. Many parents also reportedly believed that the project should be reserved for the young people, resulting in poor attendance at parents' meetings (although many did attend the graduation night) (Tarling et al, 2001a, p.49; Tarling et al 2001b, p.1).

The Australian consultations and literature suggest that it is common for mentoring projects to attempt to engage or work with mentees' families, although there is variation in the emphasis placed on this factor. There was some acknowledgement in the consultations that involving the family of mentees can be difficult at times, as has been found overseas. For instance, family members may feel threatened by the prospect of their child being assigned a mentor, and regard it as an implied criticism of them and their parenting abilities. Of course, in many cases there may be various problems in the family, and this is a key reason the young person needs a mentor.

It also appeared that involving or liaising with the mentee's family members to at least some extent is particularly important for Indigenous projects. For instance, the AFSP in Perth reported in consultations that when selecting a mentor, a worker would always speak to the family to ensure that the person is acceptable to them.

The family has to accept the person - so they know who's looking after their kids. (Stake-holder)

Some Australian mentoring projects include a component of family support in their activities. For example, the worker with the Brayton Mentoring Initiative in Shepparton, Victoria worked with mentees' families to develop trust and help them to support the mentee. On occasion the worker also provided counselling to the family or referrals to other agencies for more intensive support (ARTD 2001, p.A23).

The Coffs Harbour/Clarence site of One2One in NSW employed a part-time family support worker for some periods of the project, to support families of mentees through individual and group work, crisis intervention and family mediation. The external evaluation of the programme concluded that this was a valuable component of the model, particularly since the project was located in a rural area with few alternative support services available for young people and their families (ARTD 2002, p.66). In consultations the YWCA noted that since many of the mentees' families were chaotic and dysfunctional and with a high level of conflict, they required support in their own right in order to enable the mentoring programme to function effectively.

Activities undertaken by the pair

According to the literature, activities undertaken by the mentor/mentee pair should be designed to enrich and enhance youth opportunities and experiences (OJJDP 1998, p.12). Evans and Ave note that mentoring relationships focused on having fun, respecting the young person's views and engaging them in decision-making and activities are more successful than those programmes that are prescriptive, and require the young person to undertake certain activities with the mentor as more of an authority figure (Evans & Ave 2000, p.47).

The overseas literature notes that the majority of mentoring projects focus on sporting and leisure activities (Long & Dart 2001, p.72). Ave et al report that particularly effective activities are those that are fun, but also that focus on some type of new learning, such as life skills or a hobby (Ave et al 1999, p.81). Activities should also be special and inexpensive (Benioff 1997, p.47).

The types of activities undertaken or recommended overseas include: taking walks, attending plays, movies, sporting events or school activities, playing catch, visiting the library, washing the car, grocery shopping, watching TV (Grossman & Garry 1997, pp.2-3), homework assistance, going out for a meal or snack, going to a museum, concert or festival, or just spending time together (Morley et al 2000, p.22).

Australian programmes have followed this trend towards low-cost, fun and informal activities. Activities include fishing, walking in a park, going for a fast food meal, attending movies and sporting events, educational activities, shopping, picnics, going to the beach, watching videos, cooking, going for a drive, rock climbing, ice skating, bowling, going to a theme or other park, the production of artworks, creating Koori art, attending theatre and car shows. It was reported that the activities undertaken by the matched pair will depend on the interests of the young person (or the mutual interests of the matched pair) and the aims and nature of the project.

During the fieldwork, some mentoring projects stressed the importance of mentors and mentees participating in no/ low-cost activities. This was in order to maintain the emphasis on the development of a quality mentoring relationship, to teach the young person that they can find enjoyable activities which do not involve great expense, to avoid financial dependency on the mentor, and to avoid considerable financial strain for the project (Crime Prevention Unit, SA Attorney-General's Department 2002, p.18).

The Stage One Report on the Panyappi project in Adelaide (prepared by the government funder) found that there was confusion among some mentors, young people and referring agencies between the role and purpose of mentoring and a one-on-one activity programme. This had led to inappropriate expectations. For instance, some young people expected their mentor to treat them to a paid activity such as the movies or go-karting. The report notes that the difficulties with reimbursement of costs had had some beneficial impact in forcing mentors to find ways to spend time with young people at minimal cost (Crime Prevention Unit, SA Attorney-General's Department 2002, p.18).

Most mentoring activities are undertaken by the matched pair alone. However, some projects incorporate group activities centred on, for instance, structured educational/career or community service activities (OJJDP 1998, p.16), or field trips or cultural events (de Anda 2001, p.99; Morley et al 2000, pp.22-23). Many overseas programmes have both separate and group activities.

This was the case with some of the Australian programmes also, such as One2One in NSW. The external evaluation of the project concluded that the group activity component of the project was highly valued by both mentees and mentors, and provided opportunities for socialising, fun and games in a safe and supportive environment (ARTD 2002, p.viii).

Some Australian programmes also focus on conducting very specific tasks to provide a focus for the mentoring relationship. Examples include:

The Relationship Violence - No Way! project in Adelaide, which uses a variety of peer education techniques including one-to-one mentoring to educate both male and female victims of domestic violence about domestic violence issues (Friedman 1999, p.18; Friedman 2001).
The Men Mentoring Men (MMM) programme in the ACT, which uses a model centring on mentees repairing bikes in a bike repair workshop.

Some overseas programmes have an initial residential component, designed to bond the mentors and mentees, take the young people out of their home environment, and participate in a range of team building and adventure activities (Benioff 1997, pp.40-43). These are generally designed to engage young people and mentors with the project and the activities tend to build confidence and self-esteem (Tarling et al, 2001a, p.38; Tarling et al 2001b, p.2).

Some of the Australian programmes similarly contained a residential camp-style component, focusing on wilderness-based activities for both the matched pairs and combined groups of mentors and mentees. Examples included the New Trax Programme in Melbourne, and the Bush Law Action Plan Partners project in Port Pirie, SA.

For example, the New Trax Programme provides an intensive 11-day life skills programme for young offenders on community-based orders. The age range covered is 17-23 years, although the majority of participants are aged 17-19 years. The programme involves outdoor camp activities (a ropes course and a three-day wilderness camp), as well as courses on a variety of issues such as behaviour management and drug and alcohol issues. Highly qualified tutors and group leaders are employed to run these activities. Men serving as prisoners in Won Wron Prison have also been used as group leaders in the programme (Upper Yarra Community House n.d., pp.1, 4).

Although the participants are not formally matched with a prisoner as a mentor, much of the work involves informal mentoring, with the prisoners working with one or two young offenders at a time. The mentoring between the prisoners and young people continues after the formal programme on an informal basis. Mentors and mentees make contact with one another using the telephone and prison visitation facilities.

In addition, some overseas mentoring programmes use mentors to take on part of the case management of the young person, working with them on their case plan, attending appointments with them, and assisting them in reaching these specific goals. Again this is also the case with some Australian mentoring programmes, such as those run by the NSW DJJ and the WA Ministry of Justice. For instance activities undertaken with mentees in the former scheme can include developing or maintaining connections with school and Technical and Further Education (TAFE), attending appointments with, and acting as an advocate for, the young person with various agencies such as Centrelink, and providing court support.

The Mentoring Knowledge Base in the United Kingdom (UK) suggests a number of techniques for keeping young people interested in the programme, including setting clear goals, allowing the young person to choose the meeting place/activity, planning long-term fun projects, ensuring each meeting ends on a positive note, and providing an awards scheme for the young person (Mentoring Knowledge Base n.d.). It is important that the mentees have some input into the activities undertaken by the pair (Ave et al 1999, p.81).

Differences in the activities for young men and women

In the Australian consultations it was observed that there can be differences in the types of activities or approaches that will appeal to young men and women. The general view was that there is a tendency for young men to be less forthcoming in talking about their feelings and problems than young women. It was also reported that young men might tend to be more interested in activities such as sport, whereas girls might be interested in activities such as artistic or creative pursuits, health or (different) sporting activities, as well as just chatting.

It was therefore suggested that it can be preferable to engage young men in an activity such as sport and have back door strategies to encourage them to talk. For instance, mentors reported that a common approach might be to take the young man to a sporting event, and then he might start talking about problems in the course of general conversation while driving to or from the event.

I didn't even ask him direct questions - then I could address issues as he brought them up. (Mentor)

I was working with a half-Samoan, half-Aboriginal guy. Especially in rural areas they tend to be football crazy. For me the connecting point was footy. As soon as he understood I was trying to make him the best thing he could be, he started to open up. So on the way to having a kick, he opened up. You need to meet them on their interest or get them interested in something. (Mentor)

On the other hand, girls might be more open to sitting down and having a chat at an earlier stage of the mentoring relationship. In fact, this chatting may be a crucial element in getting the mentoring relationship off the ground and developing it further. Indeed it may be important to do this even before engaging the young women in other types of activities such as sport.

For instance, one of the female mentees consulted who was very enthusiastic about her female mentor noted that:

We talk about personal things... She's like a friend... She treats me like an adult - she tells me things. She's not just talking about drugs, it's more like normal life - for example, she discusses her pregnancy with me.

It was stressed, however, that as with all aspects of mentoring and matching, it is important to tailor the mentoring approaches to the individual young person.

Nature of the relationship between the mentor and mentee

Consultations indicated that on the continuum from worker to friend, mentors will generally be located much closer to the friend end than the other professionals in the life of that young person. Some of those consulted (including mentees and mentors) even described the mentor role in terms of being a friend. Nonetheless it was clear that mentors cannot be regarded simply in these terms, and that all parties involved need to be clear about the distinction between the two.

You can be a good friend, but you need to be a leader, you don't just act like their mate - their mates are often into offending. (Mentor)

You need to set boundaries - for example, they'll test your limits. They need to know you're setting guidelines. (Mentor)

One of the key elements of a successful mentor is knowing the boundaries. They're not [the mentee's] boss or case workers or best friend - that's a fine line we need to tread. (Mentoring Project)

The necessity of maintaining this distinction was also emphasised by the external evaluation of the Bushlaw Action Plan Partners Project in Port Pirie, SA, which stated that:

It is... important that [mentors] convey the right message to their clients and act as a role model rather than a 'pal'. (Scholz 1999, p.26)

Ending the mentoring relationship

Programmes should have a clear policy on how and when to end mentoring relationships. This should include the mentor meeting with the mentee and discussing the close of the relationship. It should be kept in mind that while a relationship may have an official end, the relationship may continue on a casual or indefinite basis.

The overseas literature reports that despite the difficulty and potential disappointment for the mentee, many programmes do not have specific practices on closure of a mentoring relationship. Mentoring relationships should commence with all parties being aware that the relationship has a formal end, but that the relationship may continue beyond that time if the parties wish (Mentoring Knowledge Base n.d.). In many cases, the end of a formal mentoring period will not mean the end of the relationship, with some mentors continuing to meet their mentee informally on a regular basis (Benioff 1997, p.48).

The overseas literature states that it is important for the mentor to discuss the end of the formal relationship with the mentee well in advance, and the two should decide how this will be done in practice (Benioff 1997, p.48). Relationships can end for a range of reasons, such as the young person outgrowing the mentor, the end of a designated time limit for the relationship, the mentor dying, one of the pair relocating, one of the pair being too busy, or simply that the pair do not get along (Ave et al 1999, p.82).

The literature also states that closure systems, at a minimum, should comprise an exit interview with each party and a meeting of the pair. There should be a clearly stated policy on continued contacts for the pair and assistance for the mentee in deciding any future plans (California Mentor Initiative 2000, p.4). Appropriate procedures will vary depending on how long the match has lasted and why the match is ending. Ideally, closure should include:

a clear explanation of the reasons for the end of the match
an opportunity to vent negative feelings about the relationship ending
a celebration of the good things about the relationship, such as revisiting a favourite activity
respect for, and acknowledgment of, the feelings expressed, using a non-blaming stance
a face-to-face meeting where mentees can say their farewells, if possible and appropriate
rituals where appropriate to culture, age, and situation eg a gift, or a trip out somewhere
the provision of counselling or other support if a mentee is having difficulty coping with the loss of a mentor (Ave et al 1999, p.83).

Mentors need to be advised on how to draw the relationship to a successful close by tapering off to replace meetings with telephone calls, and making agreements on contact in a crisis situation. Further contact past the official end of a relationship should also be negotiated and/or discussed with the young person's parent or guardian (Peel & Calvert n.d.).

The Australian consultations did not specifically address the issue of closure of the mentoring relationship. However there was general acknowledgement that the length of and process for ending the relationship needed to be made very clear to the young person from the beginning of the relationship, in order to avoid or minimise any sense of disappointment or betrayal by the young person. Some programmes also reported that an informal relationship between a mentor and mentee might continue beyond the end of the formal relationship. This appeared to be more likely with programmes using volunteer mentors.

Monitoring and supervising mentors

Provision of supervision, feedback and support to mentors by project staff that is consistent, timely and regular (although not too burdensome) is essential for any programme. Support can be in the form of group meetings, telephone contact or face-to-face interviews. The frequency can be determined by each individual project, but an average of once per month (at least in the early stages of the relationship) is recommended.

The literature and the consultations clearly demonstrated that it is vital for project staff to provide ongoing supervision and support to mentors, and that both mentors and mentees need to understand their respective roles and be supported through monitoring and review procedures (Prince's Trust 1999; Morley et al 2000, p.21; Grossman & Garry 1997, p.4). Supervision of the mentoring relationship is regarded as critical to a programme's success (Sipe 1996, p.10).

The literature and consultations identified two key reasons why regular supervision and support of the mentoring relationship is so important:

It helps mentors to build trust and develop positive relationships, as well as assisting them to negotiate the challenges in the mentor/mentee relationship (Foster 2001, p.10). The nature of the target group means that sometimes mentors may be dealing with quite challenging or complex behaviour or issues by the young person. Grossman and Garry note that many mentoring relationships would have faltered or failed without the nurturing and support of caseworkers (Grossman & Garry 1997, p.4).
The organisation needs to satisfy accountability requirements towards their clients. Mentees are often very vulnerable young people with multiple difficulties. They have also experienced many negative relationships with adults previously. There are child protection issues to consider as well in terms of the safety of the young people involved. In some instances the auspicing agency may have a specific statutory duty of care towards their mentees.

The literature suggests that monitoring procedures for mentoring programmes should include consistent scheduled meetings with staff, a framework for ongoing feedback, input from all stakeholders, written records, and a process for managing grievances and other issues (Mentoring Australia 2000; California Mentor Initiative 2000, p.4).

Support and supervision for mentors can take a number of forms, including telephone support, mentor support groups and meetings with caseworkers (Sipe 1996, p.10).

There is some contention in the literature regarding how much supervision is required or recommended for mentors by project staff. The level of supervision available will vary, depending on factors such as the nature of the project, the capacity of staff and the perceived needs of the relationship. The general view in the literature is that an average of once per month is preferable, at least in the initial stages of the match. For example, supervision in the US BBBS programme (for generally at risk young people) entails contact with all parties in the first two weeks of the match, and maintenance of monthly contact with the mentor and parent/child for the first year (Grossman & Garry 1997, p.3).

Nonetheless, DuBois and Neville found that mentors who engaged in more frequent contact and supervision with programme staff, or faced more programme requirements, reported less emotional closeness with their mentees (Dubois & Neville 1997, p.232).

The majority of Australian mentoring projects surveyed have an established system in place to support mentors. For some this includes ad hoc on-call access to project staff (eg Great Mates in Perth). Others have more structured processes, such as meetings that vary in frequency, between twice weekly (eg the Bush Law Action Plan Partners project in Port Pirie, SA) and monthly (eg Men Mentoring Men Project in the ACT).

One2One in Parramatta, NSW, is an example of a more supervision-intensive project. It was reported that supervision involved weekly telephone contact with mentors, fortnightly contact with the young people and their families, a three month review followed by a six or 12 month review (depending on the duration of a matched relationship), and annual reviews conducted thereafter. In addition, monthly volunteer group support meetings were held and ongoing training offered.

On average, Australian projects tend to meet on a fortnightly basis with mentors, or note only that their level of contact with mentors is regular. Generally, however, all Australian projects consulted, were aware of the need to supervise the mentoring relationship in an ongoing manner, and whether this occurs in an organised or an ad hoc fashion, there are systems in place to facilitate this.

The importance of providing mentors with adequate supervision and support is demonstrated by the findings of the report on the Panyappi Project in Adelaide for Indigenous young people. Mentors reported feeling very isolated, and that limited team and individual support was provided. They also had limited contact with other mentors, which again was felt to contribute to their isolation. The majority of mentors reported that they regularly went outside the project to obtain support. Some also said they did not feel trusted to perform their roles, and this undermined their motivation and confidence to work. As a result of this low motivation, regular work avoidance was reported to be practiced by some mentors in the project (Crime Prevention Unit, SA Attorney-General's Department 2002, p.12).

Training and orienting mentors

Mentors should be required to undergo at least 20 hours of training prior to being matched with a young person. Training should be provided by people with appropriate qualifications and include an overview of the programme, clarification of commitment requirements, boundaries and limitations, crisis management and problem solving, communication skills, and a summary of policies and guidelines.

The overseas literature recommends that mentors receive adequate training and orientation, and again this is cited as a crucial factor in an effective mentoring programme (Singh & White 2000, p.29). This is echoed in the Australian literature also (ARTD 2002, p.51; Crime Prevention Unit, SA Attorney-General's Department 2002, p.14). Training and orientation ensures that mentors have a comprehensive understanding of their role, and helps them to develop realistic expectations of what they can accomplish (Foster 2001, p.10).

Levels and styles of training tend to vary depending on the intensity of the project and the available resources. Effective training for mentors focusing on communication, limit-setting skills, relationship-building and youth interaction is one key aspect of mentoring projects that is believed to significantly contribute to positive results (Morley et al 2000, p.21; Grossman & Garry 1997, p.4; Sipe 1996; p.9).

Training should be provided by accredited trainers, and include skills development, cultural sensitivity and appreciation training, an overview of the programme, clarification of commitment requirements, guidelines for participants, relationship management strategies, job and role descriptions, confidentiality information, communication skills development, a summary of policies and guidelines, and problem solving resources (California Mentor Initiative 2000, p.3; Mentoring Australia 2000). Training may also include information regarding adolescent development, behaviour modification skills, listening skills, identification of drugs and drug use, mediation, non-judgemental counselling skills, and anger management (OJJDP 1998, p.18).

As with the orientation programme which should be in place for mentees (see below), orientation for mentors should include an overview of benefits and rewards, the level of commitment expected and a review of programme policies (National Mentoring Working Group 1991, p.2; California Mentor Initiative 2000, p.2).

Recommended training times are variable. Some range from eight to 10 hours (Rogers & Taylor 1997, p.129) while others recommend a minimum of 20 hours (Youth Justice Board n.d.; Mentoring Knowledge Base n.d.). Some programmes have up to seven full days of training for mentors (Trailblazers n.d.).

McLaren describes one model of mentor training that has proved successful in Hawaii, whereby mentors are trained over 18 hours to identify certain desirable behaviours and attitudes in their mentees, and reward these particular behaviours with small personal and material reinforcement (McLaren 2000, p.72).

Benioff reports that the training for the Dalston Youth Project in England lasts for four days, with ongoing support meetings and continuing training sessions throughout the programme (Benioff 1997, pp.50-52). Stage 2 of the project involved training over two weekends (Tarling et al, 2001a, p.23; Tarling et al 2001b, p.2).

Australian projects were not specifically asked about training for mentors, but a number discussed this issue in consultations. There was recognition that structured training sessions of mentors both initially and on an ongoing basis can be beneficial and important for mentors.

Programmes typically provide at least some initial training of mentors before they commence work. This training may concern, for instance, strategies for dealing with difficult situations with mentees, available referral networks and topics which mentors may come across in dealing with mentees, such as drug misuse.

Generally mentors in programmes run by government agencies appeared to have more access to regular training. Some programmes reported that they do not provide as much training to mentors as they would ideally like due to funding constraints. Smaller and regional/rural settings also face some limitations in the types of training available through other agencies locally.

Notes

7 This is based on the US Big Brothers Big Sisters (BBBS) programme.

Mentors

Recruiting mentors

When recruiting mentors, a project must provide a clear and realistic idea of the benefits and expectations of mentoring, including the level of commitment required.

There is no good practice in recruiting mentors. The mode of recruiting will depend largely on the type of people who are required as mentors (eg ethnicity, interests, age, etc). However, personality is considered to be more important in selecting mentors than physical, social or racial characteristics.

Recruitment strategies

Recruitment of mentors is one of the key issues faced by mentoring programmes. Programmes that recruit mentors too hastily or without care are doomed to failure (National Criminal Justice Reference Service 2000).

The overseas literature provides the following advice in relation to recruitment of mentors:

Identify characteristics of the programme that could positively or negatively effect its ability to recruit particular groups of mentors, such as setting, transportation, characteristics of the mentees, or the level of commitment required. This assessment should include an identification of the groups that are most likely to respond to recruitment efforts (Jucovy 2001, p.3).
Jucovy suggests two basic principles that should provide a starting point for a recruitment plan: maintaining broad name recognition both in the community and amongst groups to be targeted for recruitment; and creating a written recruitment plan, including goals, timelines and budgets (Jucovy 2001, p.1-3). The recruitment plan can outline strategies to improve the number and variety in the mentor pool, such as ongoing marketing and public relations (where resources allow), and strategies that outline the benefits and expectations in a realistic manner (Mentoring Australia 2000; see also California Mentor Initiative 2000, p.2).
Use a range of recruiting strategies, to reach both a broad variety and large number of potential mentors. Effective options cited in the literature include:
- word-of-mouth recruitment, which Sipe states has been found to be the most effective strategy by most mentoring programmes (Sipe 1996, p.52)
- presentations and links with businesses and community groups such as churches and sport teams (Jucovy 2001, pp.6-8; Novotney et al 2000, p.7; Sipe 1996, p.13)
- media advertising in a range of publications, particularly those targeted at groups the project wishes to recruit eg people from different ethnic backgrounds (Jucovy 2001, pp.6-8; Novotney et al 2000, p.7; Sipe 1996, p.52)
- displaying flyers in community public places such as libraries or gyms (Novotney et al 2000, p.7; Mentoring Knowledge Base n.d.; OJJDP 1998, p.20).

These strategies should use a clear mentor job description setting out all the requirements of the mentors, and qualities/attributes the project is seeking (Jucovy 2001, p.4).

The fieldwork demonstrated that there is no single effective recruitment strategy for mentoring programmes, and that strategies will vary significantly between different projects, depending on such factors as:

the nature of the target group, eg exclusively Indigenous (such as the Finding Yourself programme in Melbourne)
any specific characteristics sought in mentors, eg a sporting or artistic background (such as Whitelion in Melbourne), or prison inmates (such as the New Trax Programme in Melbourne)
the geographical location of the project, eg metropolitan (such as Great Mates in Perth), regional (such as the Bushlaw Action Plan Partners project in Port Pirie, SA, and the Brayton Mentoring Initiative in Shepparton, Victoria) or state-wide (such as the NSW DJJ Mentoring Scheme)
whether the project is using paid or volunteer mentors.

Some projects which were well-established and associated with juvenile justice agencies (such as the programme run by the WA Ministry of Justice) reported that standard media advertising could be one of a range of effective ways to recruit mentors.

On the other hand, consistent with overseas experience, most projects consulted during the fieldwork reported that they had found word-of-mouth recruitment to be the most effective recruitment strategy. This involved actively developing and using good networks of mentors and/or other agencies (or specific individuals in agencies) to keep an eye out for potential mentors and encourage them to apply. The mentoring projects which use this approach argued that it is far more effective to identify the very specific qualities required in a mentor than more generic strategies such as public advertising. This confirms the finding of the external evaluation of One2One in NSW that recruitment is:

Best achieved through word of mouth and connections in local communities rather than mass promotions. (ARTD 2002, p.vii)

The experience of Great Mates in Perth provides another example. The project reported that in the early stages of implementation it publicly advertised a meeting for those interested in becoming mentors. A large number of people responded - but the vast majority were found to be unsuitable to work as mentors. The process was also very labour-intensive in requiring notification of unsuccessful applicants. The project now recruits mentors through word-of-mouth (through both the programme's networks and current mentors) and has found this to be a far more effective strategy. It is of interest that a number of the programme's mentors are young men in their late teens or twenties, who have been recruited through networks of church-based youth activities. This is a demographic group from which mentoring programmes often find difficulty attracting mentors.

Recruitment difficulties

According to both the literature (Sipe 1996, p.13) and the consultations, recruiting sufficient numbers of appropriate mentors is an area where many projects face problems, and is an ongoing challenge for many programmes.

Australian mentoring projects reported varying degrees of difficulty when it came to recruiting mentors, with approximately even numbers of projects commenting on either the ease or difficulty of recruitment.

The following factors appeared to be associated with greater difficulties in recruiting and retaining mentors among Australian mentoring programmes:

a shorter period of operation, and a small number of matches. As the external evaluation of One2One in NSW concluded, developing mentoring programmes may need to reach a critical mass in terms of the mentors recruited and matches made in order to generate further applications from potential mentors (ARTD 2002, p.53)
stand-alone projects
lack of effective networks of other agencies/individuals to assist with identification of potential mentors
use of volunteer mentors. It should be noted however that staff recruitment and particularly retention appeared to be a common problem for projects using paid mentors as well. This was due to factors such as lack of a guaranteed regular income (the most significant issue), the part-time nature of the position, and the low pay. During the consultations, two Indigenous projects, the AFSP in Perth and Panyappi in Adelaide, reported that the fact that the mentoring positions were only part-time had made it more difficult to recruit mentors. In the case of the AFSP this was only reported to be a problem in relation to recruitment of male mentors, since women were more likely to be seeking or satisfied with part-time work due to family commitments.

The experience of One2One in NSW provides an illustration of these recruitment difficulties. Recruitment of volunteer mentors was the most significant barrier faced by the programme (ARTD 2002 p.viii). A much smaller number of mentors was recruited than anticipated, and this was a particularly acute problem for the Parramatta site. Only around one in 20 people who made an initial inquiry ended up becoming a mentor. A range of possible reasons for the low recruitment levels were identified, including:

concerns by potential volunteers about working with the target group of young offenders
failure to reach a critical mass of volunteers who could then generate other applications by mentors. While this critical mass was at least approached in the Coffs/Clarence site, this was not the case in Parramatta. The report concludes that for a mentoring project to be viable, this critical mass would be some 20-30 volunteers (p.vi)
covering too broad a geographic area
lack of promotional material and strategies specifically tailored to One2One in the Parramatta site. The generic material for the YWCA's BSBB programme for at risk young people generally (eg postcards and banners on public buses) was used instead. This approach was criticised as unsuitable, failing to target appropriate mentors, being unappealing to young offenders, and failing to distinguish the project from the BSBB programme
length of the time commitment involved (12 months)
project staff having skills in case work rather than marketing
lack of follow-up in some instances of people making initial enquiries
using volunteer rather than paid mentors
not reimbursing for fuel costs (in the Coffs/Clarence site this was remedied at a later stage of the project) ( pp.53-54).

Mismatch between the demographic characteristics of mentors and young people

The literature and consultations indicated that ideally mentors should come from a wide range of backgrounds and experiences (Mentoring Knowledge Base n.d.). In addition, while it is not regarded as essential to match mentors and young people on demographic factors such as sex and ethnicity, this may often be regarded as desirable in practice.

One of the common challenges experienced by mentoring programmes both in Australia and overseas is that there may be a mismatch between the demographic characteristics of mentors and mentees. For instance, mentoring programmes commonly find that while mentees are predominantly young males from low socio-economic backgrounds, mentors tend to be female, older, from a higher socio-economic or educational background, and not from an ethnic minority. For instance among overseas mentoring projects:

The US JUMP programme evaluation found that the mentors were most likely to be white females, despite efforts by projects to recruit a diverse range of people to act as mentors. More than half the mentors were white, and 62.8 per cent were women despite the majority of mentees being non-white (OJJDP 1998, pp.29-30; Novotney et al 2000, p.4). Mentors were generally well educated, with 83 per cent having completed at least some college education (Novotney et al 2000, p.4).
The PLUS programme operating in Oakland found that the majority of mentors were female, white and from human service backgrounds (Howitt et al 1998, p.46).
70 per cent of mentors in the Dalston Youth Project in England were female, and most mentors were aged between 20 and 35. In contrast to many other projects, more than 65 per cent were from minority ethnic backgrounds (Tarling et al, 2001a, p.12; Tarling et al 2001b, p.2). Just over three-quarters of the volunteer mentors were in full-time employment, and 85 per cent reported a history of volunteering (Tarling et al 2001, p.21).

Some (but not all) Australian mentoring projects have experienced similar difficulties concerning the demographic profile of the mentors they attract. For instance, the mentors for the Parramatta site of One2One in NSW were predominantly young, single professional women, while the Coffs/Clarence site had a more diverse profile of people including a broader age range and mix of marital statuses (although still predominantly female).

The external evaluation of the state-wide NSW DJJ Mentoring Scheme also reports that an enduring difficulty experienced by the programme was maintaining a supply of mentors to meet the varying demands for services from particular client groups. These included clients from various non-English speaking backgrounds, clients with more specialised needs and clients in rural areas (where there may only be sporadic demand) (ARTD 2001, p.vii).

The overseas literature suggests particular recruitment strategies to encourage recruitment of mentors from ethnic minorities:

advertising in ethnic-targeted media (Benioff 1997, p.23)
employing project staff from the ethnic background of the mentee. It is reported that it is particularly difficult to recruit mentors from Indigenous or non-English speaking backgrounds if none of the project staff are from that background themselves. For example, a project operating in NZ reported difficulty in attracting Mäori mentors, especially in locations where there were no Mäori staff members (Ave et al 1999).

One US programme which had experienced difficulties recruiting mentors from ethnic minorities overcame this problem by reducing the required length of commitment and the frequency of visits required. It also hired a community relations director to recruit from diverse communities, and formed an African American Ambassadors' Council, in which business and community leaders lent their support to recruiting efforts and participated in some events designed to recruit mentors (Morley et al 2000, p.25).

The Australian consultations indicated that projects focusing on Indigenous young people (which generally seek to employ an exclusively or predominantly Indigenous pool of mentors) may need to adopt specific recruitment strategies.

For instance, word-of-mouth strategies seem to be particularly important. Panyappi in Adelaide promoted the project and available positions via the youth and community services sector and the Aboriginal community rather than through standard processes such as media advertising. It was felt that this would be more effective and encourage applications from people who were actively involved in the community (Crime Prevention Unit, SA Attorney-General's Department 2002, p.11). The AFSP in Perth also reported using the same approach in consultations.

A stakeholder also noted from her observations of Indigenous-specific programmes that it is important to have an Indigenous coordinator for such initiatives, as it can be difficult for a non-Indigenous coordinator to recruit Indigenous mentors. This is consistent with the experience overseas reported above in relation to ethnic minorities and Indigenous people.

Rural areas

It is reported in the literature that recruiting mentors may be easier in metropolitan locations, since in rural areas transportation, geographic and poverty issues can provide barriers to recruitment (Morley et al 2000, p.26; Ave et al 1999). In consultations with the state-wide NSW DJJ Mentoring Scheme, it was reported that there are several challenges for recruitment/location of mentors in rural areas. These include:

the transitory, mobile populations of some rural areas (eg Wagga Wagga) can make it harder to recruit mentors
the necessity for mentors to be more multi-skilled than those in metropolitan areas
the large amount of travel that can be involved in meeting with mentees
deciding where to base mentors in sparsely populated areas.

On the other hand, the experience of several other Australian mentoring projects located exclusively in regional/rural areas suggests that this does not necessarily provide a barrier to recruitment. Both the Bushlaw Action Plan Partners project in Port Pirie, SA and the Brayton Mentoring Initiative in Shepparton, Victoria have been successful in recruiting mentors, and the key factor here appears to be the establishment of good networks of referral agencies. For instance the external evaluation of the Brayton Mentoring Initiative attributes the success of the project in recruiting male mentors and making matches despite a relatively small pool of potential volunteers to:

the high profile of the auspice agency and the worker in a small rural community, and the informal approach to selecting mentors and making matches. (ARTD 2001, p.A21)

Likewise, the regional Coffs Harbour/Clarence site of One2One in NSW had much greater success in recruiting mentors than the metropolitan Parramatta site, and this was felt to be due to the strong partnerships developed by the former site with key referral agencies (ARTD 2002, p.55).

The Bushlaw Action Plan Partners project in Port Pirie, SA reported that the small size of the community made it easy to find mentors who are respected in the community. On the other hand, it meant that particular care was required in selecting good mentors, since:

if we put on a bad mentor, it would not be regarded well in the community, and we would be crucified more.

Characteristics of mentors

There are a number of characteristics that programmes should focus on when recruiting mentors, the most important elements being the ability to listen, a non-judgemental attitude, flexibility, respect for and ability to relate to young people and reliability/consistency.

There is agreement in both the literature and the consultations that personality traits are more important than social or demographic characteristics in selection of mentors.

Personality traits

There was agreement in the literature and consultations concerning the types of personality features which should be sought in mentors.

The most important of these are felt to be:

the ability to listen
a non-judgemental attitude
flexibility
respect for and ability to relate to young people
reliability and consistency. In the consultations it was stressed that mentees have often had many relationships with adults who are unreliable or who have let them down. Failing to follow through on a relationship can be more damaging for the young person than not having a mentor at all (Anon 1999).

Other valuable qualities mentioned include:

commitment to the full course of the programme
patience
not forcing disclosure, and having the capacity to gradually develop trust
the ability to convey respect and dignity to the mentee
the ability to see solutions and opportunities in the young person's terms
acceptance of the mentee's social class, culture and family
awareness of ethnicity and gender equality issues
a clear understanding of the role, boundaries and responsibilities of the mentor
willingness to consult with programme staff for help and advice
understanding the need for fun
openness (Rogers and Taylor 1997, pp.133-4; Foster 2001, p.10; National Criminal Justice Reference Service 2000; Youth Justice Board n.d.; Jones-Brown and Henriques 1997, pp.225-256; de Anda 2001, p.102).

Other desirable qualities frequently mentioned in consultations were passion and enthusiasm.

Tarling et al (2001a, pp.29-30) identify several further good practice qualities in mentors which essentially draw on the characteristics identified above:

Recognising the importance of giving mentees praise and encouragement, but also questioning and challenging inappropriate behaviour as a friend as opposed to an authority figure. This can involve techniques to help the young person to see their actions from other perspectives.
Mentors being conscious of the need to encourage mentees to take decisions, set themselves achievable goals, and participate in decision-making within the relationship (such as selection of activities etc).
The mentoring relationship has to be reciprocal - a one-sided relationship is not likely to be successful or lasting (Tarling et al 2001a, pp.29-30).

In contrast, the literature reports that characteristics of less effective mentors include tending to set goals and tasks too early on, adopting an authoritative role, emphasising changes in behaviour, and having difficulty meeting with the young person regularly and consistently (Council on Crime in America 1997, p.18; Sipe 1996, pp.7-8).

Screening procedures

Mentors should be rigorously screened prior to being matched with a young person. As a minimum, screening processes for mentors should include at least one personal interview with project staff, a criminal record check, a reference check and an application form.

Both the literature and the informants consulted for this project stressed that selection and screening of mentors is one of the most critical elements of any mentoring project (Rogers & Taylor 1997, p.132; McLaren 2000, p.72). Screening should be rigorous, and can take up to eight months to complete all the prescribed checks and processes (Sipe 1996, p.13).

Screening is important as it is designed to exclude both those who are not fully committed and appropriate to the programme, and those who may pose a safety risk to children (Morley et al 2000, p.21; Grossman & Garry 1997, p.4). It has been found both in Australia and overseas that services working with young people (particularly those who are more disadvantaged or vulnerable) can tend to attract sex offenders and others with inappropriate motivations. For instance, Whitelion - based in three juvenile justice institutions in Melbourne - reports that it screened out some men who had expressed an interest in working with the programme when it was discovered that they were in fact wanting to groom young female inmates to become prostitutes.

The literature recommends that mentor screening include referee checks, basic background information, past work experience (both paid and voluntary), history of contact with young people, information on why the person wants to become a mentor, the amount of time they can commit to the programme and information about any criminal record (Benioff 1997, p.25). Procedures for screening should include an application process and review, a face-to-face interview, reference checks with at least two referees who should be able to vouch for the applicant's suitability to work with young people, criminal history checks, a home visit, suitability criteria, and successful completion of pre-match training and orientation (National Mentoring Working Group 1991, p.2; California Mentor Initiative 2000, p.3; One-to-One n.d.; Mentoring Knowledge Base n.d.; Peel & Calvert n.d.; Mentoring Australia 2000).

It is also recommended that projects have a set of written eligibility requirements for mentors, and use this as a tool to determine the suitability of mentors during the screening process (National Mentoring Working Group 1991, p.1).

In Australia, the screening procedures of mentoring projects range between:

very rigorous processes eg One2One in Parramatta required an initial interview, completion of an MMPI2 psychological profile, provision of four references, a main interview, attendance at a training weekend, a police check and a home visit
very informal and casual procedures, with no formal application process.

Most projects lie somewhere in between, with most requiring the completion of an application form, some references, at least one interview, and at least some training.

All projects have an idea of the type of person they are seeking as mentors and generally screen people according to these preconceived ideas. Nonetheless some projects do maintain written eligibility requirements; these tend to be more formal in projects run by government agencies.

For instance, the NSW DJJ Mentoring Scheme requires that mentors have:

a drivers' licence
knowledge of local communities and government and non-government agencies
an ability to relate to, engage, and motivate young people
awareness of cultural issues and ability to impart this knowledge to young people
police clearance and a working with children check.

Excluding mentors with criminal records

The literature recommends that police and/or criminal checks should be carried out on all potential mentors prior to any contact with young people (South Dakota Department of Corrections 2001, p.2; OJJDP 1998, p.12). Sex or violence offences, offences against children or very recent criminal offences should all preclude a person from becoming a mentor.

Australian projects adhere to the general principle of conducting criminal record checks, with more than 90 per cent of the projects consulted asking applicants to declare any criminal history, and conducting either state or national police checks. The only Australian projects that do not conduct criminal record checks are:

the New Trax Programme in Victoria, discussed below
the Finding Yourself programme in Victoria, where the mentors are slightly older young people, and the project states that it is aware of any criminal record through either personal knowledge of the mentor or because the mentor has been referred to the programme by juvenile justice.

Although all projects should be aware of any previous criminal offences committed by potential mentors, many overseas and Australian projects feel that reformed ex-offenders can provide excellent role models for young people (Benioff 1997, p.25; California Mentor Initiative 2000, p.2; Youth Justice Board n.d.; One-to-One n.d.).

For instance, one informant consulted for this project had previously worked with the New Deal mentoring programme in England for the long-term unemployed (which is now being rolled out across the country). The programme included a number of clients with a history of offending. One of the key things they found with this group was that:

Mentors who are themselves ex-offenders were the relationships that worked. If you matched them with someone who didn't have this, it didn't work.

For some Australian and overseas mentoring projects, people who have previous criminal records are actively encouraged to apply, or represent the exclusive pool from which mentors are drawn.

An example is the New Trax Programme in Victoria, which draws mentors solely from the inmate population of HM Prison Won Wron, and is obviously already aware of mentors' criminal histories. The programme has stringent selection criteria for mentors to be accepted into the programme, including:

a demonstrated high level of successful participation in personal development programmes
an assessment by the Review and Assessment Committee at the prison that the prisoner poses no risk to the community or themselves
a demonstrated high level of appropriate behaviour during their prison terms
good communication skills
a commitment to the aims and objectives of the programme
eligibility (or pending eligibility) for leave periods from prison (Upper Yarra Community House n.d., p.6).

Sex offenders are not permitted to participate in the programme.

The prisoners used in the New Trax Programme provide informal mentoring during and after various camp-style activities. An internal evaluation of the programme concluded that the programme had met with considerable success and that the involvement of prisoners had been a critical factor in this and added a dimension that was previously unachievable (Upper Yarra Community House n.d., pp.1,4).

The following comment by one of the mentors in the programme illustrates the potential benefits of using mentors with a criminal history to, in effect, warn young offenders what their life will be like if they continue along the same path:

I had empathy with their needs, and want in life. This gave the young men a real life experience and forced them to face reality which they had not wanted to do. Many of the young men wanted to know how I tried to stop using drugs and committing crimes... By being older and having knowledge and empathy with these young people they start to realise using drugs ends up either killing them or spending many years in prison and it is not a glamorous or rich life which they thought it would be but a life of emptiness and regrets and that they have a chance to take a different road and not end up like myself. ...[It] brought home the real world to these young men which I believe helped them to re-evaluate their lives and to stop using drugs. (Upper Yarra Community House n.d., p.16)

In consultations it was reported that in selecting prisoners to be involved in the programme:

We try to pick the mean-looking crims. The more serious crimes [they've committed], the better - usually murder or armed robbery. The more serious the crimes, the greater the impact it has on the kids.

It was suggested by those consulted for this study that in assessing whether a person's criminal record should be a bar to employment as a mentor, various factors should be considered. These include:

how long ago the offending behaviour was
whether the applicant has lied about their record
whether the person has turned their life around and dealt with their previous problems which led them to offend. As the external evaluation of the Bushlaw Action Plan Partners project in Port Pirie, SA notes:

Although the 'been there done that' message helps young people relate to the Action Plan Partner [mentors], 'been there and still doing that' is not appropriate. (Scholz 1999, p.26)

The issue of applicants having criminal records was felt to be particularly pertinent for projects wishing to recruit Indigenous mentors. This is because Indigenous people are disproportionately likely to have a criminal record compared to non-Indigenous people. Excluding people with a criminal record altogether may mean that it is extremely difficult to recruit mentors from an Indigenous background. For instance, the Logan Area Youth Justice Service in Brisbane (run by the Department of Families) reported that it attempted to establish a mentoring programme for Indigenous young offenders in the Woodridge area. However, the majority of the applicants (all Indigenous) were excluded due to their prior criminal history. Although criminal history is examined on a case-by-case basis, it was not felt that the applicants could be accepted on this occasion, due to the nature and/or recency of the offences. The applicants who were accepted (all female) obtained more permanent positions within the local area.8

A particularly flexible approach to the issue of criminal history may therefore be required for projects with a primary or sole focus on Indigenous mentees/mentors.

Projects also need to consider the practicalities of criminal screening, as these procedures can often take some months to complete. Issues such as whether to let applicants commence training before criminal record checks are complete (Benioff 1997, p.25), or how far other elements of the screening should progress while checks are being sought should be considered carefully by project staff, and appropriate policies should be put in place.

Matching

The matching process should be based on a clear and consistent policy. While sex and ethnicity may be considerations, the primary factors in matching should be the interests, needs and goals of the young person.

There was agreement in the overseas literature and the Australian consultations about the following aspects of the matching process.

The process by which mentor/mentee pairs are matched is an integral part of any mentoring project, as the success of any mentoring scheme is ultimately based on the personality match between the two people (Youth Justice Board n.d.; National Criminal Justice Reference Service 2000). Generally, it is felt that match characteristics have a significant influence on the perceived benefits of the programme for both mentors and mentees (Novotney et al 2000, p.5). For instance the evaluation of One2One in NSW concluded that:

The quality of the matching contributed significantly to a match performing and required considerable skill and knowledge about the mentor and young person by the project worker. (ARTD 2002, p.35)

Nonetheless, some overseas researchers have commented that matching is not a critical element for many programmes (Grossman & Garry 1997, p.4), with one even suggesting that matching is the least critical element for success (Sipe 1996, p.11).

There is no science or formula to the matching process, and each mentoring project's procedures will differ according to their particular programmes and goals. However, considerations should include demographic characteristics such as sex and race/ethnicity, personality, interests, and the needs and goals of the young person (National Foundation for Educational Research 2001, p.5; Mentoring Knowledge Base n.d.; Benioff 1997, pp.36-37; see also ARTD 2002, p.33). Some of those consulted for this project likened mentor matching to a dating agency, where two people can be matched on similar interests etc but this will not necessarily guarantee that they will click in practice.

Practical considerations, such as geographic location and mentor mobility, should also be taken into account in matching mentors and mentees (National Foundation for Educational Research 2001, p.5).

Regardless of how each project decides their criteria for matches, the matching process should be consistent and based on appropriate criteria (California Mentor Initiative 2000, p.4).

The behaviour of the mentor and ensuring matches based on mentor/mentee similarities and preferences in attitudes and activities are much more important to the success of a mentoring relationship than characteristics such as age, sex and ethnicity (Foster 2001, p.10; Jones-Brown & Henriques 1997, p.221; Benioff 1997, pp.36-37; Tarling et al, 2001a, p.19; Tarling et al 2001b, p.2). This was also the conclusion of the evaluation of One2One in NSW, which examined the quality of the project's matches in some detail (ARTD 2002, p.35).

A variety of issues or sources of information are often used in assisting with the matching process for mentoring projects both in Australia and overseas, such as:

attempting to assign a mentor to a young person based on perceived need. For example, if a young male mentee is living with his father and has no adult female role model, he may be paired with a female mentor in an attempt to provide him with a broader range of influences
projects which have a residential component prior to the pairing of mentors and mentees (such as wilderness or bush camps) may observe interactions during the activities as a means of establishing appropriate matches (Tarling et al, 2001a p.27; Tarling et al 2001b, p.2)
formal application processes for mentees, such as completing an application form or attending a short interview where the project managers can explore the young person's interests, personality, needs, and background, in order to ensure a more suitable match (Benioff 1997, pp.31, 34-35)
by the above process or otherwise, taking requests of mentees into account in terms of background, interests, or sex (Benioff 1997, p.36-37). Some programmes give mentees a choice of features such as age, sex, ethnicity or interests when matching them with a mentor - this is more likely to be the case for programmes catering for older teenagers (Trailblazers n.d.). However it is noted that this process can lead to disappointment and unrealistic expectations by the mentee, and is generally not recommended in the literature.

Matching on demographic characteristics

Race and ethnicity

Should mentees ideally be matched on race, ethnicity or cultural background?

Matching on demographic characteristics such as sex, race, ethnicity or cultural background is generally not regarded as important to a successful match. Demographic factors generally tend to be an issue considered in making a match rather than an over-riding and significant consideration. Sex and ethnicity matches usually seem to be made wherever possible both overseas (Morley et al 2000, p.25) and in Australia.

There is no current research that would support an assessment as to whether same- or cross-ethnicity matches are more effective (OJJDP 1998, p.35). Sipe (1996, p.59) reports that both types of matches were equally likely to be successful in the US BBBS programme (for generally at risk young people).

On the other hand, Novotney et al (2000, p.5) found in another US programme that when youth and mentors were of different ethnicity, the mentors reported that they perceived significantly less improvement than same-ethnicity matches in a range of areas including school attendance, avoidance of drugs and alcohol, avoiding fights, staying away from gangs, not using knives or guns, and avoiding friends who start trouble (Novotney et al 2000, p.5). Mentors paired with young people of the same ethnicity felt that they understood their mentee better than those in cross-race matches (Novotney et al 2000, p.5). However, it should be noted that this study only dealt with results as perceived by the mentors, rather than any clear data on recidivism or school performance levels.

Alternatively, there is some evidence to suggest that cross-race matches may be preferable. Rogers & Taylor cite a study where it was found that helpers and clients (not necessarily mentors/mentees) were seen as less helpful and the helpers less committed when both parties were of the same race (Rogers & Taylor 1997, p.131).

In the Australian consultations there was discussion by Indigenous projects and other stakeholders as to whether Indigenous young people should be matched with an Indigenous mentor. The general view was that this may indeed be valuable for some (or even many) Indigenous young people. For instance, an external evaluation of the AFSP in Perth reported that for 65 per cent of the 20 participants surveyed, having an Aboriginal person as a mentor was important to them (Social Systems and Evaluation n.d., p.32). In the consultations, staff of the programme also expressed the view that generally it was preferable to have Aboriginal mentors because it's working with your own people and knowing how Aboriginal people live, think and their culture. Having the Aboriginal family networks could also make it easier to work with the families of offenders and find the offender or others when required.

Similarly, the Finding Yourself programme in Echuca, Victoria, which uses exclusively Aboriginal mentors to mentor Aboriginal young people in an isolated Aboriginal community, reports that we don't find a problem with Koori kids not wanting to work with Koori mentors. The programme feels that it is important to have Aboriginal mentors, since many of the mentees need a fair bit of cultural development, particularly those who have been fostered into non-Aboriginal families.

Nonetheless, there was also agreement in the consultations that it should not be assumed that it is always appropriate to match an Indigenous young person with an Indigenous mentor. For instance, some Indigenous young offenders may be quite alienated from their communities and not feel motivated to work with someone from their own background. As an example, a youth service in Victoria reported that it had attempted to establish a mentoring project for Indigenous young people, but it had never got off the ground. One reason given for this was that the young people did not appear to be interested in working with an Indigenous mentor. In fact, the young people appeared to be more interested in working with the service's young African-American worker, who they perceived as being cool.

Can mentees be matched on ethnicity or race in practice?

It has been found both overseas and in Australia that even if it is desired to match on ethnicity, in practice this may not always be possible due to difficulties in recruiting mentors from the appropriate minority background. Mentoring programmes can tend to attract a disproportionate number of mentors who are white, middle class, female etc. Therefore overseas programmes have found that often young people from ethnic minorities may either face long waits before being matched or quite often end up being matched with a mentor from a different ethnic background (Singh & White 2000, p.29).

For instance, Novotney et al (2000, p.4) found that of the first matches in the US JUMP programme, ethnicity was able to be matched in 58.3 per cent of cases, and when matches were not of the same race a white mentor was matched with a non-white youth 77.8 per cent of the time.

Matching on race/ethnicity has also been an issue for both mainstream and Indigenous-specific programmes in Australia. For instance, One2One in NSW had four Indigenous young men in the programme, but no Indigenous mentors (ARTD 2002, p.34).

The external evaluation of the AFSP in Perth also found that not all the mentors appointed had been Aboriginal, and some stakeholders were dissatisfied with this. The report concludes, however, that given the demand for mentors clearly outstrips supply of Aboriginal people willing to take on this role, non-Aboriginal people should be able to be mentors so long as they are accepted by the Aboriginal community and the families and offenders concerned. For instance, some of the non-Aboriginal mentors appointed were married to Aboriginal people and therefore accepted into the Aboriginal community (Social Systems and Evaluation n.d., pp.31-32).

For some racial or ethnic groups, it may also be important to consider not only the background of the mentor and mentee, but also the particular family grouping to which they belong and even the model of mentoring adopted. Evans and Ave (2000, p.47) note that mentoring programmes are based largely on an American-Caucasian cultural ethos.

For instance, a NZ project found that many mentees wanted to be matched with a mentor from their whanai or hapu (family or community group). Family groups often preferred to take a collective approach, providing a mentor-rich environment rather than individual mentor opportunities. Many Mäori adults were also prepared to act as mentors, but felt that additional training was unnecessary or inappropriate (Ave et al 1999, p.39).

The importance of considering the family groupings of the mentee and mentor obviously has relevance to Indigenous mentoring projects in Australia also. However, the consultations suggested that, if anything, having Indigenous mentors with close family/community ties with the mentee may be inadvisable or unachievable.

This has been a clear lesson from the AFSP in Perth, which is run by the WA Ministry of Justice. When the programme was first established in 1996, the aim was that a significant family member would be identified and contracted to act as a mentor in accordance with a prescribed management plan. Legal jurisdiction would be retained by the coordinator of the programme. This aim was informed by the findings of a survey of Aboriginal families in the northern suburbs of Perth which had been conducted prior to the implementation of the initiative. The findings of this survey included:

Families wanted to be dealt with on a family basis rather than as a homogenous group by Aboriginal groups, organisations and government bodies.
The families were able to clearly identify significant family members who are the decision-makers and authority figures to whom members of the family and the extended family respond.
Families want to become more involved in decisions in relation to their young people.
Concern about the loss of authority and the inability of family members to affect the behaviour of young members of the family (Social Systems and Evaluation n.d., pp.1-2).

However, an external evaluation of the AFSP found that this aim was not achievable in practice (and following the evaluation this aim was in fact abandoned). In around 90 per cent of cases the mentor was not a family member. There was general agreement that it was unrealistic to expect family members to become mentors (Social Systems and Evaluation n.d., p.29).

In consultations with the AFSP it was also reported that in those cases where there had been a family member as a mentor, the family connection brought it unstuck. Several reasons were given for this and the reluctance by family members to become involved as mentors:

The mentor and mentee having too intimate a knowledge of each others' business.
The shame element - young offenders don't want to be shamed in front of their family members, including having to disclose their offending behaviour.
The less the family knows about the offender the better.

The difficulties which can arise when Aboriginal mentors have a family or other close connection with the mentee was also noted by other informants consulted for this study and the report on Panyappi in Adelaide. Panyappi mentors who had worked with relatives as mentees reported that while this carried some advantages in that they already knew the mentee through family connections, it also created difficulties and increased pressure on them since there was often an expectation by the mentee that the mentor would either bend the rules or give them preferential treatment in some way. Nonetheless, the mentors did not necessarily feel that they should not mentor relatives - rather, that there just needed to be recognition and monitoring of this issue.

To complicate matters further, it was also reported in consultations with the AFSP that due to the frequency of feuding amongst different groupings of Aboriginal people, it may also in some cases be inappropriate to appoint a mentor from another camp or grouping to the mentee. The project reported that it is therefore always important to mention the mentor's name to the offender's family to obtain their approval before appointment.

Sex

Similar to the findings on ethnicity matches, the value of cross-sex matches or same sex matches is unclear (OJJDP 1998, p.34). In the US JUMP programme, the male mentees matched with female mentors reported liking their mentors and feeling that they were as helpful and understanding as the male mentees matched with male mentors. Nonetheless, the male mentees reported greater benefits with respect to avoiding drugs and gangs than boys who were matched with females. Similarly, female mentors who were paired with boys reported significantly less improvement than their male counterparts in a range of areas, including avoidance of drugs and alcohol, avoiding fights, staying away from gangs, not using knives or guns and avoiding friends who start trouble (OJJDP 1998, p.39; Novotney et al 2000, p.5).

The external evaluation of One2One in NSW also reported that, while sex-matching was not a factor in whether a match performed or not, the role modelling impact of the match was most obvious in the matches between young female mentors and young female mentees.

Most overseas projects use sex as a strong determining factor when making a match (OJJDP 1998, p.34). Many matching decisions for overseas and Australian projects which consider the sex of the mentor and mentee are based on the perceived needs of the young person in terms of having a range of role models. (This may of course mean that a mentor of the opposite sex to the mentee is regarded as most appropriate in some cases).

In practice, as with ethnicity, sex matching for overseas and Australian programmes may not always be feasible given that programmes tend to have high proportions of female mentors and male mentees. In the US JUMP evaluation, sex was able to be matched in 85.1 per cent of first matches (Novotney et al 2000, p.4).

Nonetheless the only firm rule in relation to sex matching which appears to be followed both overseas and in Australia is to consistently avoid matching female mentees with male mentors (Benioff 1997, p.37). Only 4.1 per cent of matches in the US JUMP programme consisted of matching a female youth with a male mentor (Rogers & Taylor 1997, p.135; Novotney et al 2000, p.4). This reluctance to match female mentees with male mentors is generally based on the desire to protect the mentee from any inappropriate behaviour on the part of the mentor, given the companionship nature of the relationship.

One mentoring project noted in consultations that female young offenders often have a history of physical or sexual abuse, and it is generally preferable to match them with female mentors from this point of view also.

Age

In terms of age of the mentor, the Australian consultations suggested that this is generally not regarded as a significant factor, and will depend on the needs of the specific young person. In Australia it appears that although a small number of projects have a significant representation of young people as mentors (eg Great Mates in Perth), it is more common for mentors to be at least in their 30s or older. This is consistent with overseas experience. For instance the average age of mentors for the OJJDP JUMP programme of mentors is 35 (OJJDP 1998, p.30).

Some Australian informants argued that having younger people as mentors (who are generally harder to recruit) can be beneficial in that the mentor may find it easier to relate to the mentee and have more interests in common. For instance, the MAYT Mentor Programme for Indigenous young people in Adelaide reported that:

We find mentors in their early 20s tend to be the most effective - they can relate to [the young people] most effectively.

Similarly, the Finding Yourself programme in Echuca, Victoria, uses Indigenous mentors aged 15-19 years to work with Indigenous young people from Cummerangunga (an isolated Aboriginal community across the border in NSW). All of the mentors have themselves gone through the programme as mentees. The programme feels that having young Indigenous mentors is a major advantage, since:

It's easier to relate to the young people, and they use appropriate language.

On the other hand, other informants felt that older mentors can have various advantages. These include:

being very reliable and stable within the community (Rogers & Taylor 1997, pp.128-129)
having adequate life experience to deal appropriately with any challenging or difficult issues or behaviour
providing good role models for young people with poor relationships with their own parents.
Some kids need old fogeys like me. In my instance it's a very good match because he hasn't had a father figure for a very long time. He hasn't had the level of tolerance and understanding from an older person - that's what he's had from me. It wouldn't be as powerful if it was a younger person. (Mentor)

I'm looking for older people [as mentors] who can understand the plight of young people today. They've often experienced some problems in their life, and have felt the pain but come through it. They can say to the young person 'I know how you feel, but maybe if you did X you could change'. (Mentoring Project)

Some support for the benefit of using older people as mentors can be found in the experience of Panyappi in Adelaide. The report on the project stated that it has experienced a very high turnover of staff, and much higher than originally anticipated. One of the key reasons identified for this (by both mentors and others) was that the mentors were young and had limited experience, which translated into a lack of confidence in performing the challenging work required of them (Crime Prevention Unit, SA Attorney-General's Department 2002, p.13). However, this should not necessarily be interpreted as meaning that young mentors will always be unsuitable, given that a variety of other problems were also identified with the project, including inadequate support and supervision of mentors (ibid, p.12). The report concludes that Panyappi's experience indicates the need for a high level of support for mentors, particularly in the early stages if the mentor is inexperienced or lacks confidence to perform the role (ibid, p.19).

Paid versus volunteer mentors

There is no evidence to indicate whether it is preferable to employ paid or volunteer mentors. However, it is good practice to reimburse mentors for agreed costs and out of pocket expenses relating to the relationship (such as fares, mileage, entry fees etc). Receipts and/or proof of purchase should be required for reimbursement.

The overseas research suggests that it is more usual for mentoring programmes to employ volunteer rather than paid mentors (Singh & White 2000, p.27; Youth Justice Board n.d.).

In Australia, projects are split fairly evenly down the middle, with half the projects surveyed paying mentors a stipend or (usually) an hourly rate, and half using volunteer mentors. Of those projects who pay their mentors, hourly rates range from $14-$20 per hour, with the average hourly rate of just over $16.

In consultations there was considerable disagreement as to the appropriateness or wisdom of using paid versus volunteer mentors. Many of those consulted had strong views one way or the other.

The key argument used by those stakeholders and projects who favour volunteer mentors was that the young person has enough paid workers in their life, and it is psychologically vital that the young person knows that the mentor has chosen to spend time with them.

The fact that mentors are unpaid can be a positive reinforcement for the young person, especially if there are a number of welfare agencies involved in their lives. We have had young people comment that the mentoring experience is more significant due to the mentor not being paid to spend time with them. (Mentoring Project)

As a practical matter - although this was not mentioned in consultations - another argument in favour of volunteer mentors is cost. Many projects which use volunteer mentors, particularly those run by community-based agencies, simply do not have the funding to pay mentors.

The projects and stakeholders who felt that it is preferable to pay mentors raised several arguments:

Paying mentors makes them feel valued, promotes a sense of responsibility to the project, and ensures that the project can demand an appropriate level of accountability that may be more difficult with volunteer mentors. The latter point was particularly stressed by mentoring programmes run by agencies with statutory responsibility for the young people they mentor.

Payment for mentors values their skills and inputs. It also allows for a professional contractual relationship of fee for service. (Mentoring Project)

Great Mates believes that it is in the best interests of a young person if the workers are paid. This creates a sense of mentor responsibility to Great Mates, ensuring that the young person receives the best possible care and is in keeping with the Great Mates expectations. (Mentoring Project)
The target group of young offenders or young people at risk of offending is a difficult and challenging one to work with, and specific skills are required to undertake this work effectively. This should be recognised by paying mentors. General community members may not have the willingness or skills to work with this target group.
It is easier to recruit mentors. The external evaluation of One2One in NSW reported that one of the reasons for the low rate of referrals was that young offenders did not seem to be regarded as an appealing target group for potential mentors, and many did not pursue the process due to concerns about the target group. Some stakeholders in both Parramatta and Coffs Harbour/Clarence also felt that it would have been easier to recruit paid mentors (ARTD 2002, pp.52-54).
It attracts a better quality of applicant.

I've worked a lot in the voluntary sector - you often get people as volunteers who don't have anything else to do. (Mentor)
It improves equality of opportunity among potential applicants. That is, people with lower incomes may be able to apply if they will be paid for their time. People who can afford to work on an unpaid basis may not necessarily be from the wide range of backgrounds ideally desired for a mentoring programme.

You don't want wealthy middle class people wanting to work with the deserving poor. (Stakeholder)

Overseas research shows that often mentoring projects have a disproportionate number of mentors from more advantaged backgrounds.

As noted by the report on Panyappi in Adelaide, projects working only with Indigenous young people mostly use paid (Indigenous) mentors, for two key reasons:

the difficulties in attracting mentors who can commit extended periods of time to a young person, as noted above
recognition that employment and the associated training of mentors can contribute to the skills development of those employees, and enhance their opportunities for further employment in the youth and community services sector (Crime Prevention Unit, SA Attorney General's Department 2002, p.3).

A further reason for this was also mentioned in consultations ie that it is important to recognise that mentors are performing an important role for their communities and adequately compensate them for this. Many Indigenous people take on numerous community responsibilities for which they are not paid.

Reimbursement for expenses

There is a clearer view from the literature and consultations concerning the reimbursement of out-of-pocket expenses. The general view from both was that mentors should preferably be reimbursed for limited and previously agreed expenses.

For instance the Mentoring Knowledge Base states that it is good practice to reimburse volunteer mentors for agreed out of pocket expenses, for items such as travel costs, tickets, entrance fees, food purchased during a meeting (to a maximum) and the costs of any special equipment. It is argued that this will promote equality of opportunity among mentors and can assist in a project retaining volunteers. Receipts and/or relevant proof of purchase should be shown for projects to reimburse funds (SOVA n.d.; Mentoring Knowledge Base n.d.).

One of the reasons for the difficulty experienced by One2One in NSW in recruiting mentors was not reimbursing out-of-pocket expenses (ARTD 2002, p.54).

Almost 70 per cent of the Australian projects surveyed reported that they did reimburse their mentors for agreed expenses, with the most common expenses met including travel/mileage, small entertainment expenses, food and beverages, and phone calls. Some projects, such as the Finding Yourself project in Echuca, Victoria, choose to put on activities for the matched pair, so that the mentor does not incur any expenses.

Prompt reimbursement of expenses and provision of other supporting resources appears particularly important to consider for Indigenous projects, given that Indigenous mentors may be particularly likely to be on low incomes. For instance, the report on Panyappi in Adelaide found that even though a condition of employment had been that mentors have their own roadworthy vehicle and were prepared to use this for work purposes (with reimbursement for mileage), in practice these vehicles were not always available for use due to the mentor's socio-economic circumstances. Mentors were also dissatisfied with the time delays involved in receiving reimbursement for expenses and fuel. They argued that this was unsatisfactory given that they were on low wages and could not afford to absorb the costs themselves. The report recommended that a vehicle be provided for the project and that expenses be reimbursed promptly (Crime Prevention Unit, SA Attorney-General's Department 2002, p.7-8).

Notes

8 There is an intention to re-examine whether the project can be established in 2003.

Mentees

Voluntary versus mandated clients

It is preferable to only accept young people voluntarily into a mentoring project.

Some overseas mentoring programmes are voluntary, with the mentees choosing whether or not to attend (Long & Dart 2001, p.72). Others are compulsory, or ordered by the court or a similar authorised body (Benioff 1997, pp.30-31).

In Australia, most mentoring programmes only accept mentees into the programme on a voluntary basis. Only three projects (the New Trax Programme in Melbourne; the Bushlaw Action Plan Partners project in Port Pirie, SA; and the Brayton Mentoring Initiative in Shepparton, Victoria) accept mentees as part of a mandatory court order.

According to the literature it is preferable for participation by young people in mentoring programmes to be voluntary (Benioff 1997, p.30). This was the general view from the consultations also. It can be difficult to effectively engage a young person in a close relationship such as mentoring if the young person is unwilling to be involved. Even Australian projects accepting young people on court orders felt that it is important to give young people a choice as to whether to be involved in the programme. However, as noted by the evaluation of the NSW DJJ Mentoring Scheme, it is perhaps questionable whether participation can truly be regarded as voluntary when the young person is on an order (ARTD 2001b, p.vi).

Recruitment processes

The Australian projects consulted drew their mentees from a range of sources. Eighty-six per cent of all projects consulted (18 of 21) drew some of their mentee referrals from the juvenile justice system, and for 15 of the 21 this was their primary referral source. Other major sources of referrals were community organisations (52 per cent), self-referral (38 per cent), or 'other', which included schools, government departments, and family members (67 per cent).

Factors associated with adequate levels of referrals

Another challenge faced by at least some Australian mentoring projects has been recruiting sufficient numbers of young people into the programme. Consultations indicated that programmes are less likely to experience this where the following factors are in place:

good referral sources
the programme particularly appeals to young people
young people are proactively encouraged to participate.

Good sources of referrals

The overseas (Benioff 1997, p.32) and Australian literature notes that young people do not tend to refer themselves into mentoring projects. For instance, the external evaluation of the AFSP in Perth reports that for none of the participants consulted had participation in the programme been their own or their family's idea. For most participants the idea had come from their juvenile justice officer, and for some the coordinator of the programme had suggested it (Social Systems and Evaluation n.d., p.12).

This indicates the importance of building strong ties with, and promoting the project to, potential referring agencies (such as probation officers, social workers, police, youth workers, teachers, youth workers etc) who come into contact with potential participants.

Consultations likewise indicated that Australian mentoring projects which have had greater success in recruiting young people have well-established and effective networks to provide a regular and adequate source of appropriate referrals. Programmes which have a guaranteed source of either all or most of the referrals required have an advantage here, such as programmes connected with:

juvenile justice agencies eg the AFSP and the Juvenile Justice Mentoring Programme (both run by the WA Department of Justice) and the NSW DJJ Mentoring Scheme
juvenile justice institutions eg Whitelion in Melbourne - we've got a captive market
child welfare agencies eg the FAYS Mentor Programme in SA.

This is not to say that other programmes that do not have these automatic referral mechanisms cannot attract sufficient numbers of young people into the programme. It does suggest, however, that such projects need to give careful attention to establishing good referral sources (and preferably multiple sources).

The importance of strong referral networks is demonstrated by the experience of several Australian mentoring projects: One2One in NSW, Panyappi in Adelaide, and the AFSP in Perth. These are discussed in turn below.

One2One is an example. While the Coffs/Clarence site received adequate referrals (around 30-40 referrals annually, consistent with around 20-30 matches annually), the Parramatta site received less than half this level of referrals. The key factor identified for this was the development of strong partnerships by the Coffs/Clarence site with the police Youth Liaison Officers (YLOs) and Conference Administrators/Convenors (CAs), including provision of information and feedback about the programme. This had not occurred for the Parramatta site. The much smaller numbers of YLOs/CAs in the Coffs/Clarence site was one reason for this (ARTD 2002, pp.55, 60).

Panyappi in Adelaide provides another illustration. It was reported that there have been minimal referrals into the project from community-based agencies. To try to address this issue, one of the project mentors had worked directly with a city-based street work team to promote direct liaison between the project, a key youth service and young people in the inner Adelaide area. The report on the project notes that:

This reflects the need for the project to establish credibility and trust with the community and youth services before the project is accessed and recommended. It has been hypothesised by some people [that] the internal difficulties experienced by the project have harmed the community's perception of the project and may effect their preparedness to refer [to] or even support the project. (Crime Prevention Unit, SA Attorney-General's Department 2002, p.15).

The external evaluation of the AFSP in Perth, run by the Ministry of Justice, also highlights the importance of developing and maintaining good referral networks. Of the 56 Juvenile Justice Officers and Community Corrections Officers consulted, 41 per cent had made referrals to the programme, with the number of cases referred ranging from 0-12 per officer. Nearly 18 per cent of the officers who had not referred a case came from areas with no access to the programme, or only very recent access to the programme. Ten officers felt they did not know enough about the programme to make a referral, and again these officers came from areas without access to the programme or where the programme was new. This demonstrates that programmes are likely to get more referrals where they have been established longer and where key referral points are familiar with the programme (Social Systems and Evaluation n.d., p.19).

The evaluation also noted the importance of providing adequate feedback to referral sources about the outcomes for the client. Some dissatisfaction with this process was expressed by the officers surveyed. The report concluded that this lack of feedback had contributed to the lack of clarity in relation to case management (Social Systems and Evaluation n.d., p.ii).

Programmes that particularly appeal to young people

The literature states that mentoring programmes should be developed on the basis of the mentees' needs (Mentoring Australia 2000). Consistent with this, consultations indicated that young people are more likely to be attracted to mentoring programmes which have a hook of offering mentors, activities or other aspects of the programme which might be particularly appealing to young people, and in particular young offenders. This can help make mentoring appear cool to young people with a history of offending and multiple problems who might otherwise not find the prospect of hanging out with an adult attractive. This appears particularly important for older teenagers, and for young men. Examples of youth-friendly hooks offered by Australian mentoring programmes include:

high profile mentors/coordinators
focusing on activities which particularly appeal to young people
proactively encouraging young people to participate.

High profile mentors or coordinators

Mentors or coordinators with a high public profile can be very appealing to young people, particularly disadvantaged young people such as juvenile offenders.

The high profile has a lot to do with it - this person can't possibly be interested in me - why would my hero take time out [to spend time with me]? ... For a disadvantaged young person, mixing with all these VIP people, it's pretty important. (Stakeholder)

An example is the Whitelion programme in Melbourne which uses high profile mentors from the sporting, artistic, media and business communities as role models and mentors for young inmates in juvenile justice institutions. It was reported that such mentors are very appealing to young people in the centres, and this was apparent from consulting with some of the young people who had developed informal mentoring relationships as a result of the role modelling programme. As the coordinator noted, 'I get young people coming up and saying to me I want a mentor'.

Another example is Great Mates in Perth, whose coordinator is a well-known American basketballer. It was again reported by mentors and other stakeholders that this is an advantage in attracting young people into the programme, raising the profile of the project in the community, and developing links with and support from high profile sporting clubs/personalities.

Doors open for him all the time. (Mentor)

The links developed with sporting clubs and personalities are regularly used to support the programme and provide some of the activities conducted with mentors. For instance, mentees are taken to meet sporting personalities, or attend sporting events for free. Sporting personalities are also used to promote Great Mates eg by giving presentations at schools.

Focusing on activities which particularly appeal to young people

Another way to attract young people is to focus on activities which particularly appeal to them. Examples of this are provided elsewhere in this report, such as:

working on bikes, eg Men Mentoring Men in Canberra
sporting activities, eg Great Mates in Perth, Whitelion in Victoria.

Proactively encouraging young people to paticipate

Another factor stressed in consultations was the importance of making proactive attempts to 'sell' the project to potential mentees by project staff and other referral agencies, particularly by those whom the young person may already know and trust. The need for this is demonstrated by the experience of the AFSP in Perth. The external evaluation of the programme found that most of the mentees consulted did not have very high expectations of their mentors before commencing the programme and, in fact, 25 per cent were not even sure what to expect from having a mentor. The evaluation found that in practice many mentees got more from their mentors than they expected (Social Systems and Evaluation n.d., p.13).

Mentoring projects that are integrated into other youth service programmes, and have therefore had contact with the young person for other reasons, can be well-placed to sell projects to potential mentees.

The importance of other word-of-mouth recommendations of the programme by both workers in other agencies and young people who have been through the mentoring programme was also stressed in consultations. This process in turn depends on the programme generating sufficient matches to enable this to take place. For instance the MAYT Mentor Programme in Adelaide reported that:

We've had no problem getting young people in at all. In fact we've had the reverse problem - we've had to [knock back some young people]. Positive word-of-mouth seems to be the key factor - kids who've had mentors tell others and recommend it - that's really important.

According to the overseas literature, one way to promote mentoring programmes to potential mentees is to provide an orientation programme, including an overview of benefits and rewards, the level of commitment expected, and a review of programme policies (National Mentoring Working Group 1991; California Mentor Initiative 2000, p.2). Benioff (1997, p.33) suggests that young people be encouraged to come to an induction evening where they can learn more about the programme before deciding whether or not to become involved (if the programme is voluntary). This can be made more appealing to young people through the use of colourful fliers, the offer of free refreshments, the fact that they may bring a friend, and youth-friendly timing and locations.

The experience of One2One in NSW illustrates the importance of the above points. There were markedly lower numbers of young people referred into the programme than anticipated, particularly in Parramatta. Apart from the failure to develop strong partnerships with referral sources in Parramatta, other key reasons for this identified by the external evaluation included:

lack of specific promotional material in the Parramatta site. The only material available was a brochure promoting the YWCA's BSBB programme (for generally at risk young people) rather than One2One for young offenders. Apart from the fact that the material did not describe the programme, it was also regarded as unappealing to young people in the target group. In Coffs Harbour/Clarence targeted One2One promotional material was developed.
YLOs and CAs reported that most young people attending cautions and conferences were not interested in being referred to the programme, and could not be persuaded by the Convenor, YLOs or family members to participate in the programme. CAs felt that this was a particular issue since young people attending conferences were typically older, more influenced by peers rather than adults, and further entrenched in the juvenile justice system (ARTD 2002, pp.56-60).

Screening/eligibility of mentees

It is generally accepted as good practice not to accept any mentees who may pose a threat to their mentor, or who may require more professional support (eg for mental health problems).

A number of mentoring projects overseas and in Australia have policies on the characteristics of the mentees accepted into the programme. The consultations suggested that when selecting mentees, the most critical feature Australian mentoring programmes look for is young people without appropriate role models whom it is felt would benefit from such a person in their lives.

Some overseas and Australian projects also exclude young people from their programme for a range of reasons. Of the 21 mentoring programmes included in this study, nine reported that they did exclude some groups of mentees. The remaining 12 either did not exclude any mentees, or had not come across the issue. Of the projects that do exclude some young people, the reasons given included that the young person:

is not in the target group, eg not Indigenous, outside the age range etc
has a lifestyle not suited to mentoring, eg they are homeless or have no fixed address, or use drugs
has exhibited sexual or physical violence
has needs that are unlikely to be met by the programme, eg the young person has serious mental health issues
is generally high-risk, uncooperative or inaccessible.

This is consistent with the recommendation in the literature that generally young people should not be accepted into a mentoring programme if they may pose a threat or could cause harm to their mentor, or may need more professional support eg due to mental health issues.

Characteristics of mentees

Age

Overseas mentoring projects cater for mentees at a range of ages. For example:

The OJJDP's review of the JUMP programme found that the mentees ranged in age from 5 to 18 years, with the average age range being 12-14 years (OJJDP 1998, pp.24-25).
The Dalston Youth Project in England covered young people aged between 11 and 14 (Tarling et al, 2001a, p.v; Tarling et al 2001b, p.2).
The One-to-One Programme in London runs mentoring for people aged 16-21 (One-to-One n.d.).

In Australia, the majority of projects cater for the age range of 10-18, or 10-17, with the youngest mentee aged 8 (for the Finding Yourself programme in Echuca, Victoria), and the eldest aged 24 (Big hART in Melbourne).

Sex

Most projects in Australia cater for both male and female mentees. One project consulted had exclusively females as its target group (Big hART in Melbourne), while three of the projects only included males (the New Trax Programme in Melbourne, the Grassmere Youth Mentoring Programme in Melbourne, and Men Mentoring Men in Canberra).

Race/ethnicity

Some overseas studies report that most of the young people in mentoring programmes are from ethnic minority groups (OJJDP 1998, pp.24-25).

Amongst the Australian projects surveyed, 86 per cent included Indigenous young people as mentees, and 48 per cent included young people from non-English speaking backgrounds. However, many of these were mainstream programmes.

There were, however, some projects with an exclusive focus on Indigenous young people. For instance these included: the AFSP in Perth, three SA programmes (Panyappi and the MAYT Mentor Programme in Adelaide, and Bush Breakaway in Ceduna) and the Finding Yourself programme in Echuca, Victoria. There were no projects with an exclusive focus on young people from non-English speaking backgrounds.

Social and family characteristics

The consultations suggested that mentees in Australian programmes have the typical characteristics of young offenders generally, such as:

a history of offending or risk factors for offending
poor family relationships and a lack of appropriate role models
a low socio-economic background
being predominantly male
having multiple other problems, eg a history of substance abuse, family violence, poor educational achievement, low literacy, behavioural problems and so on.

For example, many of the mentees in the Brayton Mentoring Initiative in Shepparton, Victoria had learning difficulties, anger and behavioural issues, and lacked basic social skills, and some had an intellectual or psychiatric disability (ARTD 2001a, p.A21).

Similarly, the mentees in the New Trax Programme in Melbourne had characteristics such as the following:

95 per cent unemployed or not studying
85 per cent used hard drugs on a daily basis
38 per cent lacked secure accommodation
38 per cent had psychiatric problems
57 per cent had family conflict issues
85 per cent had previous criminal offences
65 per cent had anger problems
54 per cent had problems with violence (Upper Yarra Community House n.d., p.6).

The above findings are consistent with the overseas research which shows that:

Mentees are likely to come from families living in relative poverty (Howitt et al 1998, p.46; Ave et al 1999) and single parent households (Novotney et al 2000, p.4).
Mentees generally lack a male parental figure (OJJDP 1998, p.25).
Mentees have often lived in homes characterised by violence, criminal activity (Ave et al 1999, p.26), and parental drug or alcohol abuse (Novotney et al 2000, p.4).
Mentees often report that they could not talk with family members, and that they had become involved in the mentoring programme because they wanted someone to talk with and share time and activities with (de Anda 2001, p.100).
Mentees typically have friends with a range of problems including delinquent behaviour (Novotney et al 2000, p.4), problems at school, social/family problems, delinquency, drug use, and teen pregnancy (OJJDP 1998, pp.26-27).

Specific issues for mentoring programmes servicing young offenders

In the course of the scoping consultations for this study, it appeared that there are numerous mentoring projects operating around Australia for generally at risk young people, but relatively few covering the specific target group of young offenders or young people at risk of offending. (The focus of this project did not enable a systematic or comparative analysis of these broader mentoring programmes.)

Informants were asked to comment on why this might be the case. The general view was that young offenders or young people at risk of offending very much represent the hard end of the market to deal with in a mentoring programme. It may be less appealing (to both service providers and potential mentors) to provide mentoring services to this group, compared to the broader category of at risk young people.

Young offenders may be more difficult to deal with than those who are simply at risk. They may be older, exhibit more difficult (or dangerous) behaviour, have more severe and entrenched problems, be difficult to engage and less responsive to the potential influence of a mentor (some young people can be disappointment after disappointment), and some may even be regarded as past the point of early intervention.

This does not necessarily mean that mentoring programmes should not be run for young offenders or young people at risk of offending, but rather that programmes need to carefully consider the implications of the target group and develop programme elements and strategies accordingly.

The experiences of some Australian projects illustrate these challenges. For instance, the Brayton Mentoring Initiative in Shepparton, Victoria focused on connecting young people into local community networks, education, training and local business. However, there was some reluctance by key agencies to be involved with the project, due to fear and misunderstanding of young offenders and insufficient resources to support young offenders who had multiple difficulties and high needs. This presented an initial barrier to placing clients in employment, education and training. This was a particularly critical issue in a regional location such as Shepparton, due to the limited number of agencies available, which limited the capacity to shop around. It was also reported that some local police were viewed as unsympathetic towards the clients of the programme, which further alienated them from their community (ARTD 2002, p.A23).

Nonetheless, the project ultimately managed to successfully engage the support of key agencies through two main strategies:

Performing an active role in community education, promoting the project in the community, recruiting mentors, and educating employers and service providers (including Indigenous agencies) about young offenders, their needs and behaviours.
Developing the capacity of the agencies to support the project's clients, by providing active support of the young people in the early stages of their work with service providers, employers and mentors, and building a reputation of availability to respond to these key players as required (eg following up clients who did not turn up to work) (ARTD 2001a, pp.A23-24).

The external evaluation of One2One in NSW likewise notes the difficulties of providing mentoring programmes for young offenders.

It is clear that servicing this target group is more costly than that of mentoring programmes which target younger age groups, who are less down the offending pathway. A lesson from the pilot is that programmes for young offenders take more work in recruiting and supporting mentors, in gaining appropriate referrals, and in providing pre-match activities to keep engaged young people who have been referred. Such programmes are also less likely to achieve economies of scale in terms of numbers of cases per project worker. Both these lessons are consistent with the limited evidence to date on costings from programmes in the USA and England and Wales. (ARTD 2002, p.70)

Similarly, an informant who worked with the New Deal mentoring programme in England reported that the sub-group of offenders required far more intensive work than the other long-term unemployed clients in the programme. This created some resource difficulties for the programme, and required them to renegotiate their budget for the project several times.

It is of interest to note, however, that some mentoring projects have quite deliberately and successfully targeted hard end young offenders. An example is the Bushlaw Action Plan Partners project in the regional area of Port Pirie, SA, which includes a camp component as well as follow- up mentoring. The project reported that in its initial stages, the young offenders with the greatest problems (the big gang leaders) were specifically targeted for inclusion in the programme. This was then followed by a targeting of the offsiders to the first group. The project reports that this was a very effective strategy.

We changed the big ones . . . these were the king pins. There was a clear and dramatic drop in the crime rate after that in this area. We knew that the reoffending rate of these specific kids had dropped quite dramatically. ... Once we'd intercepted the core offenders, we found that we were getting young people [who were less entrenched in offending].

Evaluation

Mentoring projects should engage in thorough and ongoing data collection, and be subject to regular external evaluation. This is particularly critical for projects in the earlier stages of development.

The importance of evaluation

The CPB has identified evaluation as an important component of crime prevention programmes. For instance this was recommended in the 1999 report on early intervention approaches to crime prevention (National Crime Prevention 1999, pp.93-99), and is emphasised and discussed in more detail in the more recent report, Principles for Evaluating Community Crime Prevention Projects (National Crime Prevention, 2002). This latter report notes that community crime prevention projects:

are frequently not evaluated, but when they are, the evaluations typically are quite limited and contribute far less to decision-making about the programme than they might. Yet, an evaluation of a community crime prevention programme can provide valuable information about the way it was conducted and its context, outcomes and efficiency, helping managers and others to plan programme changes (p.7).

Consistent with this, there was clear agreement in the literature (Youth Justice Board n.d.; Benioff 1997, p.67) and the consultations with both mentoring projects and external stakeholders that monitoring and evaluation is essential for any mentoring programme to determine effectiveness and impacts, particularly in the earlier stages of implementation.

The overseas mentoring literature recommends that, as a rule, information and data should be collected and retained throughout the programme, including personal data for all participants, administrative information, programme data, output measures, and outcome measures (Youth Justice Board n.d.).

Lack of evaluations of mentoring programmes

In practice, most mentoring programmes both overseas and in Australia have not been formally evaluated. In terms of overseas projects, the OJJDP states that there have been few studies that provide concrete and measurable evidence of mentoring effectiveness (OJJDP 1998, p.7). McLaren (2000, p.70) states that mentoring is one of the least studied approaches to dealing with young offenders.

Most of the Australian projects surveyed (16 out of 21) had undergone some form of evaluation (internal or external), but only eight had undergone or were undergoing an independent external evaluation. A number of the evaluation reports, both external and internal, are also not publicly available.

In consultations it was felt that external evaluations are preferable to internal evaluations, for a number of reasons:

They provide a more reliable and independent assessment of a programme, and hence have greater credibility (which may be important for the purpose of funding applications).
People skilled in evaluation can be employed, whereas staff with the skills for providing mentoring services rarely have well-developed evaluation skills.
It is easier for all parties involved to provide data to an independent party, particularly where some negative views or data are being provided.

It is also important that appropriate data collection mechanisms be established to collect information as required for any evaluation conducted. Two of the major external evaluations of Australian mentoring programmes (the AFSP in Perth, and the NSW DJJ Mentoring Scheme) comment that the evaluation was significantly hampered by the lack of established data collection mechanisms in place to provide data to analyse. For instance, the report on the NSW DJJ Mentoring Scheme concluded that:

In many respects the Mentor Programme is unevaluable. Data collection and reporting on programme outputs has not been sufficient in most cases to provide for effective monitoring, evaluation or planning. (ARTD 2002, p.vi)

This highlights the need to build evaluation processes and mechanisms into any mentoring project from its very inception. Ideally this would be through use of external evaluators, who can establish data collection mechanisms at the beginning, monitor data collection, conduct independent data collection, and analyse the results. For instance, this was the case with the evaluation of One2One in NSW (ARTD 2002). Nonetheless this can be a comparatively costly exercise, particularly for projects with low or uncertain funding.

Reasons for lack of evaluation

There are a number of reasons for the lack of evaluation of mentoring projects both overseas and in Australia:

the significant barriers that can inhibit an effective assessment of such a long-term, subjective approach to crime reduction and general behavioural alteration. These difficulties include a need for long-term study, the lack of consistency in outcome measures used, and a general lack of comprehensiveness in existing evaluations (Foster 2001, p.11). These are common difficulties encountered in evaluating any early intervention approach to crime prevention
the strong reliance on participant observation (and its subjective nature)
the voluntary nature of many programmes
the difficulty of measuring the impact of the mentoring component in programmes involving a range of initiatives, the small sample sizes available, and the short timeframes or pilot nature of many programmes (Foster 2001, p.25)
funding issues - when funding is very limited, often evaluation is regarded as a lower priority than other project elements (Foster 2001, p.11).

Consultations suggested that in Australia, funding constraints and the fact that many projects have only been running for a relatively short time have been the key factors inhibiting greater emphasis on evaluation. It is of interest that the external evaluations which have been conducted have been funded by government agencies (of programmes either run or funded by the agency).

Increasing the emphasis on evaluation

The above finding indicates that if evaluation is to be given greater emphasis in Australian mentoring programmes, governments will need to take the lead in promoting and funding this activity. In consultations it was suggested that one way to do this is for funders to allocate specific resources to evaluation when funding mentoring programmes.

It is also important that when an evaluation is conducted, particularly of an external nature, that there is a commitment to promptly releasing the report. Several key evaluation reports either could not be obtained for the present project, or were only available after marked delays.

The effectiveness of mentoring programmes

Mentoring programmes for young offenders and young people at risk of offending both in Australia and overseas typically have aims such as:

reducing crime and other anti-social behaviour by young people
reducing other problematic behaviours such as drug and alcohol use
improving self-esteem, social skills etc.

Programmes also typically aim (either implicitly or explicitly) to achieve long-term and sustainable changes in relation to these areas.
The general lack of evaluation material available on mentoring programmes significantly limits the conclusions that can be drawn about the effectiveness of mentoring to achieve the above aims. This is particularly true in the Australian context, where there are only four external evaluation reports providing some evidence in relation to outcomes of mentoring projects (ARTD 2002; Social Systems and Evaluation n.d.; ARTD 2001b; Scholz 1999). For one of these reports only the Executive Summary was available (ARTD 2001b). In addition a further internal evaluation report was obtained, which provides very brief findings only in relation to mentoring under the New Trax Programme in Melbourne, and does not provide details concerned how the outcomes reported such as reduced drug use were assessed (Upper Yarra Community House n.d., p.6). Nonetheless, as with other aspects of mentoring, there is general consistency between the findings of the overseas and Australian research.

From the overseas and Australian literature the following conclusions can be drawn:

Mentoring is a promising but unproven strategy.
Mentoring can achieve positive outcomes for some young people.
Mentoring is only suitable for some young people.

Each of these is discussed in turn below.

Mentoring is a promising but unproven strategy

There is some conflict in the literature regarding the effectiveness of mentoring as a strategy to reduce juvenile offending. From the literature, particularly those overseas reports that review a range of material on mentoring, the following conclusions can be drawn:

There is insufficient evidence at this stage to definitively prove that mentoring is an effective strategy to reduce crime or produce any other outcome. However Singh and White (2000, p.27) comment that mentoring is rarely criticised, despite this fact.
Nonetheless, various mentoring programmes both overseas and in Australia have reported some positive outcomes in terms of reduction in offending and so on. These outcomes are discussed in more detail below.
On balance the most appropriate conclusion from the overseas and Australian literature appears to be that mentoring should be viewed as a promising crime prevention strategy. This is the conclusion reached by McLaren (2000), Sherman et al (1998) and Singh & White (2000) following their extensive reviews of the overseas literature (although McLaren qualifies this by saying at best).
Two of the above researchers suggest that drug use (McLaren 2000; Sherman 1998), alcohol use and violence (Sherman 1998) are the areas where mentoring can have the greatest impact. However, some Australian mentoring projects have experienced difficulties engaging young people with chronic drug problems in mentoring projects. This was noted in consultations with the Juvenile Justice Mentoring Programme run by the WA Ministry of Justice, and has been reported by the Brayton Mentoring Initiative in Shepparton, Victoria in relation to chronic marijuana users (ARTD 2001a, p.A.23).
Where positive outcomes have been reported from mentoring programmes in both Australia and overseas, these are generally of a short-term nature. Evaluations of mentoring programmes are generally conducted over relatively short periods after the intervention has taken place. Overall there is a dearth of evidence of long-term impacts of mentoring programmes, both overseas (Singh & White 2000, p.27) and in Australia. It is therefore not known whether, where positive impacts have been observed, these impacts have been or can be sustained over a longer period of time. This is a common problem with evaluations of many early intervention initiatives in the crime prevention field.
There is some overseas research that suggests there are more effective options to reduce youth recidivism than mentoring. Blechman et al (2001) reports that skills training produces lower rates of recidivism and is more cost-effective than mentoring.

Mentoring can achieve positive benefits for some young people

There is some evidence in the overseas and to a lesser extent the Australian literature that some mentoring programmes have produced positive outcomes for young people, at least in the short periods covered by evaluations. The measures on which positive findings have been reported are generally consistent between the overseas and Australian literature.

Reduced offending

Some mentoring programmes both overseas and in Australia report a reduction in offending by young people who participate in mentoring programmes. For instance, Harman (2000, p.112) reports on a range of programmes involving mentoring in NZ, after which juvenile crime fell 43 per cent (with a 74.4 per cent reduction in crimes committed by those aged 13 or under). An evaluation of the Partners Against Crime programme in Detroit likewise found that recidivism amongst juvenile offenders participating in the programme was reduced by 38 per cent compared to the control group, and was 50 per cent lower than offenders who had been offered a position in the programme but turned it down (National Criminal Justice Reference Service 2000).

Consistently in Australia, some mentoring projects report reduced offending by mentees. For instance, the evidence from mentees, their families, mentors and project staff consistently indicated that One2One mentees in NSW had reduced their offending while participating in the programme. In Coffs Harbour/Clarence it was reported that the local police had generally taken their focus off the mentees in the programme (ARTD 2002, p.30).

Similarly 70 per cent of mentees in the AFSP in Perth reported that they had stopped offending as a result of having a mentor. Sixty per cent of mentees also reported that having a mentor had had some impact on their lives, most commonly keeping them out of trouble (20 per cent of mentees) (Social Systems and Evaluation n.d., pp.13-14).

An internal evaluation of the New Trax Programme in Melbourne notes that eight per cent of the participants in the programme re-offended, but it is not known over what length of time this was tracked or how this was assessed (Upper Yarra Community House n.d., p.6).

Completion of juvenile justice orders

Another area discussed in the Australian literature is completion of juvenile justice orders. The external evaluation of the Brayton Mentoring Initiative in Shepparton, Victoria found that almost all (94 per cent) of the young people in the programme were on a juvenile justice order, none were breached for non-compliance, and 10 per cent were breached for further offences (ARTD 2001a, pA.23). However, no figures are presented against which a comparison could be made as to whether this could be regarded as a favourable outcome.

The external evaluation of the AFSP in Perth concluded that the available evidence did not allow any conclusions to be drawn about the success of the programme in improving completion rates of community-based orders (Social Systems and Evaluation n.d., p.11).

Reduced substance misuse and other risky behaviours

Some evaluations of mentoring programmes also report reductions in substance misuse (drugs and/or alcohol) or other risky behaviours. For instance Beier et al (2000, p.330) report that adolescent mentees in an American mentoring programme were significantly less likely to participate in four of the five measured risk behaviours, including carrying a weapon, illicit drug use, smoking and sex with more than one partner in a six-month period. Similarly Mathieson (1997, p.11) cites a range of statistics on the effectiveness of mentoring in the US, including a finding that mentored youth are 46 per cent less likely to initiate drug use (and 70 per cent less likely for minority youth), and 27 per cent less likely to initiate alcohol use (see also Rogers & Taylor 1997, pp.138-139).

In Australia, the internal evaluation of the New Trax Programme in Melbourne reports that of the 85 per cent of mentees in the programme identified as having a drug problem (ie using hard drugs on a daily basis), 64 per cent had significantly reduced or stopped their drug use after participation in the programme, and seven per cent had not. As noted above, it is not known on what evidence this conclusion was based (Upper Yarra Community House n.d., p.6).

Increased participation/performance in education, training and employment

Another finding reported by some mentoring programmes is either greater participation or better performance in education, training and/or employment. For instance, the JUMP programme in the US reports a number of successful outcomes, including improvement in school attendance (30 per cent) and academic improvement (30 per cent) (Jones-Brown & Henriques 1997, p.220; see also South Dakota Department of Corrections 2001, p.1-1).

Similarly, in Australia the external evaluation of the Brayton Mentoring Initiative in Shepparton, Victoria found that the project was successful in increasing mentees' participation in employment, education and training, from 36 per cent overall on entry to the programme through to 85 per cent on exit or currently. The biggest area of improvement was employment, with the proportion in employment increasing from three per cent on entry to 39 per cent. This latter figure included 24 per cent who were also in training. The rates of participation in education also increased from 24 per cent to 39 per cent. These results were similar for Indigenous mentees also (ARTD 2001a, pA21-22).

The internal evaluation of the New Trax Programme in Melbourne also reports that 53 per cent of participants were either employed or had returned to study after completion of the programme, compared to 95 per cent who had been unemployed and/or not studying beforehand (Upper Yarra Community House n.d., p.6).

Improved self-esteem, social/communication skills and personal relationships

A number of mentoring programmes report improvements in more general qualities or skills in mentees such as enhanced self-esteem and social/communication skills, and improved relationships with others such as family and peers (see Grossman & Garry 1997, pp.5-6). For example in NZ, Cardy (1999, p.7) reports on an intensive (twice a week) mentoring programme where positive changes in the mentees were observed after 16 weeks. Changes included higher self-esteem and confidence levels, goal-setting, improved personal relationships with family and friends, and increased participation in sport and school activities. Similarly the US JUMP programme reports improvements in general behaviour (35 per cent) and increased appropriate interactions with peers (Jones-Brown & Henriques 1997, p.220; see also Rogers & Taylor 1997, pp.138-139).

In Australia, One2One in NSW was found to have led to a number of general positive outcomes for young people, including improved family relationships, improved social and communication skills, increased motivation and stability, improved behaviour, and access or exposure to a range of new experiences, opportunities and ideas (ARTD 2002, pp.30-31).

In the Brayton Mentoring Initiative in Shepparton, Victoria, apart from increased involvement in education, training and employment, the other significant area of progress made by mentees was increased community involvement. Some 88 per cent of the participants were involved in the community on either a regular (25 per cent) or occasional (63 per cent) basis following participation in the programme, compared to 33 per cent on entry to the programme. It was also reported that small improvements had been made in the contact between participants and their families over the course of the project. There had been an increased number of young people moving to independent living, with the number living with their families reducing from 74 per cent on entry to 50 per cent on exit (ARTD 2001a, pp.A22-23).

The AFSP in Perth also found that all nine caregivers interviewed reported positive changes in their children as a result of the programme, and most identified positive effects on the family (Social Systems and Evaluation n.d., pp.15-16). In addition 50 per cent of the Juvenile Justice Officers/Community Corrections Officers consulted felt that participation in the programme had been beneficial for the young person (Social Systems and Evaluation n.d., pp.24).

General satisfaction by mentees and/or other stakeholders with the programme

Some mentoring studies report overall satisfaction by mentees and others with the programme. For example both mentees and mentors in the American JUMP programme viewed the mentoring experience as a positive one, although match characteristics did influence perceived benefits (Novotney et al 2000, p.5).

The external evaluation of the AFSP in Perth (for both young people and adults) found that all of the 20 participants interviewed overwhelmingly endorsed the programme and would recommend it to others like themselves. Of the nine caregivers interviewed, eight also thought the programme was good (Social Systems and Evaluation n.d., p.12, 15).

The external evaluation of the Bushlaw Action Plan Partners project in Port Pirie, SA also found that most of the clients consulted were happy with the assistance provided by their Action Plan Partner (mentor). However, it was felt that some did not get full value out of their Partner due to factors such as infrequent contact. It was concluded that:

Most Action Plan Partners are performing a tremendous job and are being credited for continuous success of their clients post camp. (Scholz 1999, p.26)

Findings from the Australian consultations

In addition to the Australian evaluation reports cited above, there was also an overall view among the project staff, mentors, mentees and external stakeholders consulted for this project that mentoring can make a real difference to young people in a variety of ways, and a number of specific instances were cited where this was felt to have been the case. These differences were on a variety of measures, consistent with those reported in the literature discussed above. For instance a mentor with the Bushlaw Action Plan Partners project in Port Pirie, SA reported that:

When we started we had 95 per cent not reoffending. But we've got a different group now - eg there's more mental health issues. So now it's around 75 per cent not reoffending. The nature of the offending also changes [after participation in the programme] - it gets less serious. I've seen this personally with the kids I've worked with.

Mentoring is only suitable for some young people

While some of those consulted for this project felt that mentoring was suitable for all young offenders or young people at risk of offending, others felt that mentoring may be more suitable for some young people than others. The latter appears to be the more realistic view. There is some acknowledgement in the literature that mentoring cannot reach all young people.

There are two components to this issue:

not all young people will be suitable to refer into mentoring programmes
? mentoring will not work for all young people referred into programmes.

These issues are discussed separately below.

Not all young people will be suitable to refer into mentoring programmes

The evaluation of One2One in NSW provides the most detailed consideration in the literature (from either overseas or Australia) of the issue that not all young people will be appropriate to participate in mentoring programmes. The report found that an estimated 20 per cent of young people attending cautions or Youth Justice Conferences were referred onto the programme, with a higher proportion at the Coffs Harbour/Clarence site (31 per cent) than at Parramatta (14 per cent). The rest were screened out due to either not being interested in participating or not being suitable (ARTD 2002, p.55). The screening process was felt to be effective, based on the experience of the police YLOs and CAs. The evaluation therefore concluded that mentoring is likely to be appropriate for only a minority of those cautioned or conferenced (ARTD 2002, p.vi).

Likewise the external evaluation of the Brayton Mentoring Initiative in Shepparton, Victoria found that only around half of the young people in the project were able to be matched with mentors, even though it had been originally envisaged that all of them would be. The key reasons for this were that the young people were not ready or capable of being matched in such a formal, structured arrangement. The young people who were not matched were provided with intensive support by the project's coordinator. The project also found it difficult to engage young people who were chronic marijuana users, due to lack of motivation (ARTD 2001, pp.A21- A23).

There was also acknowledgement of this issue in consultations.

A lot of young offenders are less willing to do this kind of work compared to generally at risk young people. A lot of young offenders I work with wouldn't be interested. (Mentoring Project)

Mentoring will not work for all young people referred into programmes

The consultations and literature also suggest that, of young people actually accepted into mentoring programmes, the programme may work for many but not all young people. For instance, the Dalston Youth Project in England found that around half the young people in the programme did not engage with it in any meaningful way (Tarling et al, 2001a, p.32; Tarling et al 2001b, p.4). Similarly the MAYT Mentor Programme in Adelaide estimated that:

For a good 60-70 per cent [of clients] it will work - for those for whom it doesn't work, it's not always the mentor, it may be the circumstances.

The evaluation of One2One in NSW likewise found that for those young people matched, most (76 per cent) developed into what were termed performing matches, ie ongoing relationships, which often lasted for some period of time (ARTD 2002, p.vii).

The reasons offered in the consultations for why some young people may be unsuited to mentoring included insufficient stability in their lives to make development of an ongoing relationship feasible and lack of motivation to address their problems - or at least to be brought to that point.

The importance of the latter point was very apparent from the consultations with mentees for this study. This is illustrated by the different experiences reported by two young men who had participated in the Bushlaw Action Plan Partners project in Port Pirie, SA. For one young man who had had a very chaotic life and committed various offences, having a mentor had clearly had a dramatic impact, and he had managed to turn his life around with the guidance of his mentor. The mentor had however come into his life at a stage where the young man had accepted that he had various problems and did not want his life to continue in the same pattern. In contrast, another young man had only seen his mentor on a few occasions before the relationship ended, and appeared to have little interest in the process. He seemed to be quite happy with his offending lifestyle and did not appear motivated to change it.

In summary, the results of the consultations are consistent with the conclusion of the external evaluation of One2One in NSW, that:

The scope of mentoring as an intervention with young offenders is limited. ... This suggests that mentoring should be only one element in any strategy targeting young offenders. (ARTD 2002, pvi)

The impact of demographic characteristics of mentees

There is some limited evidence from the overseas and Australian literature that the effectiveness of mentoring may vary according to the demographic characteristics of mentees.

Age

Some informants consulted for this project felt that the age of the young person can have an impact on the suitability and effectiveness of mentoring programmes. Among those who discussed this issue, there was agreement that generally younger people (say in their earlier teens) are more receptive to mentoring approaches:

Mostly the kids over 16 years old, they don't want another adult hanging around. (Mentoring Project)

[Mentoring is more likely to be effective] for the 12-15-year-olds. The older ones, they get too clever, it's too late to intervene, or they don't want a mentor. (Stakeholder)

The experience of One2One in NSW provides support for this view. Those young people who were referred into the programme were younger on average than young people attending cautions and conferences (ie the pool from which potential mentees were drawn) and those who were actually matched with a mentor were younger again. For example, as at October 2001, for the Parramatta and Coffs Harbour/Clarence sites respectively, the average age of young people in these categories were as follows:

attended cautions/conferences: 15.1 and 15 years
were referred into the programme: 14.3 years in both sites
were matched with a mentor: 12.8 and 13.3 years.

Consultations with mentees and their families also indicated that at least initially it was often the parent/guardian (hoping to get help with problems affecting the young person and the family) who expressed some interest in the mentoring programme and was able to encourage the young person to participate. Conversely, the young people who were not interested in being referred into the programme tended to be older, more entrenched in offending behaviour and much more responsive to peers than adult influences. The report concludes that:

This would indicate that successful referrals occur when the young person themselves are still at an age or a stage where they are responsive to parental influence as opposed to older and/or more streetwise young people for whom their peer group and not their family is by far more influential. (ARTD 2002, p.59)

In consultations the YWCA agreed with this conclusion, and contrasted the agency's experience with One2One with its greater success with their BSBB programme for generally at risk young people. The latter programme generally had younger children involved.

The ideal time to match a young person is 9-13 years old, because they're old enough to be independent from their family but not so old that it's totally uncool to be matched with an adult... and be seen in public with them,... and they haven't developed a pattern of offending. Mentoring is more effective as early intervention. ... It's also harder to find mentors who would be capable and interested [in mentoring this older group].

Sex

There is some evidence from the Dalston Youth Project in England that may possibly suggest that mentoring reduces offending more effectively in girls than in boys. In the project evaluation, it was found that 12 per cent of girls were cautioned or convicted of an offence while on the programme, whereas 54 per cent of boys had been cautioned or convicted of an offence (Tarling et al 2001a, p.59; Tarling et al 2001b, p.4). Nonetheless, given that young women consistently have lower rates of offending than young men, some further analysis of this data would be required before clear conclusions could be drawn as to whether the programme was more effective on this measure for girls than boys.

Race

Singh & White examine the evaluations of a number of US mentoring programmes and conclude that the results suggest that mentoring programmes are especially effective for Indigenous and ethnic minority youth. They report that the latter groups have consistently higher positive results as a result of mentoring than Caucasian participants in areas such as initiation of drug use, school attendance, and improved peer relationships (Singh & White 2000, p.27-28). However, Singh & White also discuss conflicting evidence from other US programmes which have had less success with ethnic minority youth and finally conclude that based on the available US information, the benefits of mentoring programmes for Indigenous and ethnic minority youth are unclear (Singh & White 2000, p.28-29).

In the Australian consultations some argued that mentoring programmes are likely to be particularly effective for Indigenous young people. The external evaluation of the NSW DJJ Mentoring Scheme, which has a particular but not exclusive focus on Indigenous young people, concluded that one of the strengths of the scheme has been improved services for some Aboriginal clients, with skilled Aboriginal people working as mentors in communities around NSW (ARTD 2002, p.vii).

Conclusion

This audit and review of mentoring programmes was based on 21 projects nationally, from an initial group of 24 which were identified in the scoping process. The selection criteria for including these projects is described in the methodology. Essentially, it involved selecting programmes where the primary aim of the project was mentoring, the target group was young offenders or young people at risk of offending and the mentors were community-based individuals rather than people acting in a professional or semi-professional capacity as youth workers.

This report provides a snapshot of mentoring in early 2002. Since then new projects have started and others have ceased operation. From the snapshot the following profile emerges:

62 per cent are run by community organisations
government is the primary funding source for most projects
more than half those surveyed cited insufficient funding as a significant weakness for the programme
almost half the projects are pilot programmes
most programmes have organisational autonomy from government.

Models and organisational structure: good practice

Programme objectives

Some key principles that a service should establish prior to operation include:

a tightly defined target group
sufficient trained staff to run the programme and provide support to participants
clear guidelines relating to the recruitment and screening of mentors
well developed links with a range of local agencies who can support the work of the programme.

Naturalistic versus formal

Mentoring programmes can be divided between formal and naturalistic in their approach, with formal programmes being more regimented. There is no evidence in the literature or from the consultations that one type of programme is better than the other. However, it appears important for formal programmes to avoid simply replicating a youth worker function.

A mentoring project should establish a clear conceptual philosophy regarding the nature of the programme and whether it will take a formal or naturalistic approach to mentoring.

Stand-alone versus integrated projects

A key Australian external evaluation report (of One2One in NSW) indicates that stand-alone mentoring projects may have greater difficulties getting off the ground than integrated projects, and there was general agreement amongst the stakeholders consulted for this project that integrated projects are more likely to be effective. The literature also suggests that services other than mentoring should be available to the clients of mentoring programmes, either via that or other organisations.

Mentoring projects which are integrated into a range of other services appear more likely to be effective than those which are stand-alone.

Staff

Both the literature and the consultations indicate the importance of effective project staff. The need for strong overall coordination of the mentoring programme is essential to its success.

Staff running mentoring projects should be appropriately skilled and have clearly written job descriptions.

Establishment and administration

Establishing mentoring programmes involves effective scoping to determine need. It also requires effective consultation with specific groups, in particular Indigenous people and their organisations.

Strong organisational administration and infrastructure are essential for a mentoring programme to operate effectively.
Accountable and appropriate policies, practices and codes of conduct dealing with a range of issues need to be developed and put in place.

The mentoring process: good practice

The literature and consultations show that specific policies and processes around the mentoring relationship are likely to improve positive outcomes.

Initial meetings

Length of contact

Frequency of contact

The frequency of contact will largely be determined by the nature of the project and the activities undertaken by the mentor and mentee. Contact must be regular and consistent. The ideal frequency of contact suggested is once per week.

Parental involvement

Mentoring projects should engage parents or guardians of the mentee in the project where possible, including obtaining their permission and informing them about the nature and extent of the mentoring relationship.

Activities undertaken by the mentor and mentee

Ending the mentoring relationship

Programmes should have a clear policy on how and when to end mentoring relationships. This should include the mentor meeting with the mentee and discussing the close of the relationship. It should be kept in mind that while a relationship may have an official end, the relationship may continue on a casual or indefinite basis.

Monitoring and supervising mentors

Training and orienting mentors

Finding and employing mentors: good practice

The literature and consultations show that there are a number of practices which should be put in place when selecting and employing mentors.

Recruiting mentors

When recruiting mentors, a project must provide a clear and realistic idea of the benefits and expectations of mentoring, including the level of commitment required.

There is no one 'good practice' in recruiting - the mode of recruiting will depend largely on the type of people who are desired as mentors (such as ethnicity, interests, age, etc). It should be noted that personality is often considered to be more important than physical, social or racial demographics.

Characteristics of mentors

Screening procedures

Excluding mentors with criminal records

A criminal record should not necessarily preclude someone from being a mentor; indeed, some programmes even seek out people with a criminal record to act as mentors. However, any person who has been convicted of sexual offences, any offences against children, any violent offence, or any serious offence within the past five years should not be accepted as a mentor.

Matching

Paid versus volunteer mentors

It is good practice to reimburse mentors for agreed costs and out-of-pocket expenses relating to the relationship (such as fares, mileage, entry fees etc). Receipts and/or proof of purchase should be required for reimbursement.

Selecting mentees and developing a programme: good practice

Recruiting and referring

Projects which have had the greatest success in recruiting young people have well-established and effective networks to provide a regular and adequate source of appropriate referrals.

Screening and eligibility of mentees

There needs to be put in place a screening process to assess suitable mentees. It is particularly important to identify a number of issues in this process.

Voluntary versus mandated clients. The research and consultations show that there are potential problems with taking young people as mentees when they are mandated to attend the programme. It may be preferable to only accept young people voluntarily into a mentoring project, irrespective of whether they are subject to a court-imposed supervisory order.
Violence and the potential for harm. It is generally regarded as good practice not to accept any mentees who may pose a threat to their mentor.
The need for professional assistance or intervention in place of mentoring. Some potential mentees may require far more professional support than can be provided by a mentor (eg for mental health problems or serious drug problems).

The programme activities

There are no specific rules about what works in relation to the activities undertaken during the mentoring process. However, several points emerge from the literature and the consultations.

There is a need to proactively sell the programme to young people.
There needs to be a focus on activities which are appealing to young people.
High profile mentors or coordinators can be particularly appealing to young people.

Mentoring programmes for young offenders

In the course of the scoping consultations for this study, it appeared that there are numerous mentoring projects operating around Australia for at risk young people, but relatively few covering the specific target group of young offenders or young people at risk of offending.

The evidence shows that mentoring with young offenders may be more intensive or complex for young offenders than generally at risk young people (especially in relation to mentoring frequency).
Therefore, programmes need to carefully consider the implications of the target group and develop programme elements and strategies accordingly, especially:
- the greater resources needed for young offenders
- the greater demands on mentors.

Evaluation: an essential part of good practice

Projects should engage in thorough data collection and be subject to regular external evaluation. This is particularly critical for projects in the earlier stages of development.
There is a need to build evaluation processes and mechanisms into any mentoring project from its very inception.
Information and data should be collected and retained throughout the programme, including personal data for all participants, administrative information, programme data, output measures and outcome measures.
External evaluations are preferable to internal evaluations.

Determining the effectiveness of mentoring

From the overseas and Australian literature the following conclusions can be drawn:

Mentoring is a promising but unproven strategy.
Mentoring can achieve positive outcomes for some young people.
Mentoring is only suitable for some young people.
Where positive outcomes have been reported from mentoring programmes in both Australia and overseas, these are generally of a short-term nature. Overall there is a dearth of evidence of long-term impacts of mentoring programmes.

Despite the significant limitations on research and evaluation, some tentative points can be drawn on the basis of some successful mentoring projects.

Reduced offending

Some mentoring programmes both overseas and in Australia report a reduction in offending by young people who participate in mentoring programmes.

Reduced substance misuse and other risky behaviours

Some evaluations of mentoring programmes also report reductions in substance misuse (drugs and/or alcohol) or other risk-taking behaviours.

Increased participation/performance in education, training and employment

Another finding reported by some mentoring programmes is either greater participation or better performance in education, training and/or employment.

Improved self-esteem, social/communication skills and personal relationships

Mentoring is only suitable for some young people

The literature and consultations show that mentoring will only be suitable for some young people. There are two aspects to this point:

Not all young people will be suitable to refer into mentoring programmes. In other words, some young people will be screened out as unsuitable for mentoring.
Mentoring will not work for all young people referred into programmes. There will also be some young people who will not successfully engage in a mentoring relationship.

There is some limited evidence that the effectiveness of mentoring may vary according to the demographic characteristics of mentees.

Age. Generally younger people (in their earlier teens) are more receptive to mentoring approaches.
Sex. There is some limited UK evidence which suggests that mentoring may reduce offending more effectively in girls than in boys.
Race/ethnicity. Based on the available US information, the benefits of mentoring programmes for Indigenous and ethnic minority youth are unclear. However, in the Australian consultations some argued that mentoring programmes are likely to be particularly effective for Indigenous young people.

Indigenous programmes

Information was gathered on a number of Australian Indigenous-specific programmes through both the consultation process and the literature review. These programmes included the AFSP (Perth), the MAYT Mentor Programme (Adelaide), Panyappi (Adelaide), Bush Breakaway (Ceduna, SA) and the Finding Yourself programme (Echuca, Victoria). These data indicated that there are a number of mentoring issues which are specific to Indigenous programmes:

There needs to be sensitivity about the location of the programme - particularly if it is seen as too close to government.
There may be an added need for flexibility in programme delivery if Aboriginal young people are in remote or isolated communities.
Adequate scoping and consultation is particularly important for Indigenous projects, to ensure that they will be acceptable to, and engage with, those communities.
Indigenous projects need to continue to engage with the Indigenous community once they are in operation. This may involve special measures to ensure participation in steering committees.
Projects need to be specifically promoted within Aboriginal communities using measures which are likely to reach those communities.
Some Indigenous projects reported that because the mentoring positions were only part-time it was more difficult to recruit mentors. There are also stronger arguments for the payment of Indigenous mentors given the lower socio-economic status of Indigenous people.
While it should not be assumed that it is always appropriate to match an Indigenous young person with an Indigenous mentor, most Indigenous projects found that it was often beneficial to match Aboriginal mentors and mentees.
Aboriginal family and kinship networks may place significant cultural requirements on how mentoring is organised.

In terms of the specific external evaluation of the AFSP in Perth, some significant results were:

all nine caregivers interviewed reported positive changes in their children as a result of the programme, and most identified positive effects on the family
all of the 20 participants interviewed overwhelmingly endorsed the programme and would recommend it to others like themselves. Of the nine caregivers interviewed, eight also thought that the programme was good.

There is a relatively widespread view that mentoring is likely to be particularly effective for Indigenous young people. This may in part derive from the opportunity mentoring provides for Aboriginal organisations and Aboriginal people to actively participate in the process of working with Aboriginal young offenders.

Recommendations for stage 2

National workshop

Option 1: Organise a national workshop/conference for all mentoring projects and other key stakeholders (eg agencies which fund mentoring projects).

It was clear from the consultations that many mentoring projects have little contact with, or knowledge about, other mentoring projects which are operating within Australia. It could be highly productive to organise a national workshop/conference for all mentoring projects (and funders of such projects) from around Australia to come together to discuss key issues including good practice, evaluation techniques, key research findings etc. Such a meeting would provide the opportunity to develop links between mentoring projects. A report from the national workshop could be prepared for public distribution.

A good practice manual

Option 2: Develop a good practice manual

The development of a manual which outlines good practice would be a direct and practical way to assist existing mentoring projects, as well as new projects which are in the process of development. A good practice manual would provide examples of how to set up a programme and advice about policy and practice issues, recruitment, training, targeting, evaluation etc.

Both Option 1 and Option 2 provide the opportunity for practical assistance to mentoring projects, particularly when the consultations revealed relatively limited knowledge of, or contact between, mentoring programmes. Options 1 and Option 2 could be undertaken jointly.

Evaluation

Option 3: Fund evaluations of existing mentoring programmes.

It was apparent from the consultations that most mentoring projects remain unevaluated, and that where evaluation does take place it is often internal and not rigorous in its evaluative standards.

CPB could fund one or more evaluations of existing mentoring programmes which have not been previously externally evaluated. Such evaluations need to be longitudinal and include measures related to re-offending.

Funding existing programmes

Option 4: Provide funding to existing mentoring programmes.

A further option is to provide funding to existing mentoring programmes, preferably on an ongoing rather than pilot basis. The consultations revealed that many projects are uncertain about their funding and exist with a pilot status. Such a situation is unsatisfactory in terms of developing and maintaining professional and effective service delivery. In providing funding to existing mentoring programmes, preference should be given to projects without any ongoing government funding.

Bibliography and select references

Anon (1999) 'Make a Friend - Be a Peer Mentor.' Youth in Action OJJDP Bulletin. Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programmes, US Department of Justice, Washington D.C.

Anon (2000) 'Mentoring to Koori Youth.' Narrandera Argus. 21 November.

ARTD Management and Research Consultants (2001a) Evaluation of the Young Offenders Pilot Programme. Report prepared for the Youth Programmemes Strategies Section, Youth Bureau, Department of Education, Training and Youth Affairs, ARTD, Haberfield, Sydney.

ARTD Management and Research Consultants (2001b) Strategic Review of the DJJ Mentor Programme. Report prepared for the NSW Department of Juvenile Justice, ARTD, Haberfield, Sydney.

ARTD Management and Research Consultants (2002) Mentoring for Young Offenders: Final Report of the NSW Pilot Programme. Report prepared for the Crime Prevention Branch, Attorney-General's Department and the Department of Juvenile Justice, ARTD, Haberfield, Sydney.

Ave K, Evans I, Hamerton H, Melville L, Moeke-Pickering T & Robertson N (1999) The Mentoring for Children/Youth At Risk Demonstration Project: Final Evaluation Report. Department of Psychology, University of Waikato, New Zealand.

Barron-McKeagney T, Woody J & D'Souza H (2001) 'Mentoring At-Risk Latino Children and Their Parents: Impact on Social Skills and Problem Behaviours.' Child and Adolescent Social Work Journal. 18(2): 119-136.

Beier S, Rosenfeld W, Spitalny K, Zansky S & Bontempo A (2000) 'The Potential Role of an Adult Mentor in Influencing High-Risk Behaviors in Adolescents.' Arch Pediatric and Adolescent Medicine. 154: 327-331.

Benioff S (1997) A Second Chance: Developing Mentoring and Education Projects for Young People. Commission for Racial Equality and Crime Concern, London.

Blechman E, Maurice A, Buecker B & Helberg C (2001) 'Can Mentoring or Skill Training Reduce Recidivism? Observational Study with Propensity Analysis.' Prevention Science. 1(3): 139-155.

California Mentor Initiative (2000) Quality Assurance Standards. California Mentor Initiative, California.

Cardy T (1999) 'Mentors Set a Good Example.' Evening Post (Wellington, New Zealand) 7 July.

Council on Crime in America (1997) Preventing Crime, Saving Children: Monitoring, Mentoring & Ministering. Center for Civic Innovation, the Manhattan Institute, New York.

Crime Prevention Unit, SA Attorney-General's Department (2002) Panyappi Indigenous Youth Mentoring Project. Crime Prevention Unit, SA Attorney-General's Department, Adelaide.

Crowley D & McIntyre S (1999) 'One2One: A Mentoring Project for Young Offenders.' Paper presented at the Children and Crime: Victims and Offenders Conference convened by the Australian Institute of Criminology, Brisbane, 17-18 June.

Cunneen C (2001) The Impact of Crime Prevention on Aboriginal Communities. Institute of Criminology, Law Faculty, University of Sydney.

de Anda D (2001) 'A Qualitative Evaluation of a Mentor Programme for At-Risk Youth: The Participants' Perspective.' Child and Adolescent Social Work Journal. 18(2): 97-117.

DuBois D & Neville H (1997) 'Youth Mentoring: Investigation of Relationship Characteristics and Perceived Benefits.' Journal of Community Psychology. 25(3): 227-234.

Ellis J (1997) 'Volunteer Mentorship Programmes to Prevent and Respond to Troubled Behaviour.' The Alberta Journal of Educational Research. 43(1): 53-56.

Evans I & Ave K (2000) 'Mentoring Children and Youth: Principles, Issues, and Policy Implications for Community Programmemes in New Zealand.' New Zealand Journal of Psychology. 29(1): 41-49.

Foster L (2001) Effectiveness of Mentor Programmes - Review of the Literature from 1995 to 2000. California Research Bureau, California State Library, California.

Friedman B (1999) Relationship Violence: No Way! Young Men and Relationship Violence Prevention Project. Partnerships Against Domestic Violence and SA Department of Human Services, Adelaide.

Friedman B (2001) Relationship Violence - No Way! Project Stage 2: Final Project Report (Young People and Relationship Violence Prevention Project), unpublished report.

Grossman J & Garry E (1997) 'Mentoring - A Proven Delinquency Prevention Strategy.' OJJDP Juvenile Justice Bulletin. Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programmes, US Department of Justice, Washington DC.

Keys Young (2000) Roundtable Report on the Design and Management of Public Space in Terms of Crime Prevention Outcomes. Report prepared for the Crime Prevention Branch, Commonwealth Attorney-General's Department, Keys Young, Sydney.

Harman A (2000) 'Youth Programme Slashes Crime.' Law and Order. May: 111-113.

Howitt P, Moore E & Gaulier B (1998) 'An Evaluation of Comprehensive, Community-Based Delinquency Prevention Programmeming.' Juvenile and Family Court Journal. 49(1): 39-49.

Jones-Brown D & Henriques Z (1997) 'Promises and Pitfalls of Mentoring as a Juvenile Justice Strategy.' Social Justice. 24(4): 212-233.

Jucovy L (2001) Recruiting Mentors. Philadelphia, P/PV.

Jurik N, Blumenthal J, Smith B & Portillos E (2000) 'Organisational Cooptation or Social Change?' Journal of Contemporary Criminal Justice. 16(3): 293-320.

Lemmon A & Bowen S (2001) New Directions in Working With Young People in the Juvenile Justice Residential Centres: A Study of the Whitelion Programme. Charles Sturt University, Wagga Wagga.

Long J & Dart J (2001) 'Opening-up: Engaging People in Evaluation.' International Journal of Social Research Methodology. 4(1): 71-78.

Mathieson L (1997) 'The Birth of Dade County's Juvenile Assessment Centre.' The Police Chief. LXIV(10): 110-115.

McLaren K (2000) Tough is Not Enough - Getting Smart about Youth Crime: A Review of Research on What Works to Reduce Offending by Young People. Ministry of Youth Affairs, Wellington, New Zealand.

Mentoring Australia (2000) Mentoring: Benchmarks for Effective and Responsible Mentoring Programmes. Mentoring Australia Website. Available at: http://www.mentoring-australia.com/benchmark.htm

Mentoring Knowledge Base Website (n.d.). Available at: http://www.mentoringknowledgebase.com

Morley E, Rossman S, Kopczynski M, Buck J & Gouvis C (2000) Comprehensive Responses to Youth at Risk: Interim Findings From the SafeFutures Initiative. Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programmes, US Department of Justice, Washington DC.

National Crime Prevention (1999) Pathways to Prevention: Developmental and Early Intervention Approaches to Crime in Australia. National Crime Prevention, Commonwealth Attorney-General's Department, Canberra.

National Crime Prevention (2002) Principles for Evaluating Community Crime Prevention Projects. National Crime Prevention, Commonwealth Attorney-General's Department, Canberra.

National Criminal Justice Reference Service (2000) Juvenile Justice Bulletin. Department of Justice, Washington DC, February. Available at: http://www.ncjrs.org/html/ojjdp/jjbul2000_02_1/ment.html

National Foundation for Educational Research (2001) Evaluation of the Mentoring Bursary Programme. National Foundation for Educational Research, Berkshire, United Kingdom.

National Mentoring Working Group, convened by United Way of America and The National Mentoring Partnership (1991) Elements of Effective Mentoring Practices: Promoting Responsible Mentoring Programmes

Novotney L, Mertinko E, Lange J & Baker T (2000) 'Juvenile Mentoring Programme: A Progress Report.' OJJDP Juvenile Justice Bulletin. Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programmes, US Department of Justice, Washington DC.

Office of Juvenile Justice and Delinquency Prevention (OJJDP) (1998) Juvenile Mentoring Programme: 1998 Report to Congress. OJJDP, Office of Justice Programmes, US Department of Justice, Washington DC.

One-to-One (n.d.) The One-to-One Project. DePaul Trust Volunteering Centre, London.

Public Broadcasting Service (n.d.) Law and Order Series. Public Broadcasting Service, United States.Available at: http://www.pbs.org/lawandorder/episode1.html

Peel S & Calvert I (n.d.) Mentoring for Young Offenders. The DIVERT Trust, London.

Prince's Trust (1999) Factsheet 8 - Mentoring. The Prince's Trust Policy & Research Unit, London.Available at http://www.princes-trust.org.uk

Rogers A & Taylor A (1997) 'Intergenerational Mentoring.' Journal of Gerontological Social Work. 28(1-2): 125-140.

Scholz M (1999) Client Review: Prepared for the Port Pirie Regional JPET Service, unpublished report.

Sheehan K, DiCara J, LeBailly S & Christoffel K (1999) 'Adapting the Gang Model: Peer Mentoring for Violence Prevention.' Pediatrics. 104(1): 50-62.

Sherman L, Gottfredson D, MacKenzie D, Eck J, Reuter P & Bushway S (1998) Preventing Crime: What Works, What Doesn't, and What's Promising: A Report to the United States Congress. National Institute of Justice, Washington D.C.

Singh D & White C (2000) Rapua Te Huarahi Tika - Searching for Solutions: A Review of Research About Effective Interventions for Reducing Offending by Indigenous and Ethnic Minority Youth. Ministry of Youth Affairs, Wellington, New Zealand.

Sipe C (1996) Mentoring: A Synthesis of P/PV's Research: 1988-1995. Public/Private Ventures, Philadelphia, PA.

Social Systems and Evaluation (n.d.) An Evaluation of the Aboriginal Family Supervision Agreements Programme: Report of an Evaluation Conducted for the Ministry of Justice. Social Systems and Evaluation, Perth.

South Dakota Department of Corrections (2001) Aftercare Mentoring Initiative Manual. South Dakota Department of Corrections, South Dakota.

SOVA (n.d.) SOVA Website.

Tarling R, Burrows J & Clarke A (2001a) Home Office Research Study 232: Dalston Youth Project Part II (11-14) - An Evaluation. Home Office Research, Development and Statistics Directorate, London.

Tarling R, Burrows J & Clarke A (2001b) 'Dalston Youth Project for 11- to 14-year olds: An Evaluation.' Findings No. 158, Home Office, London.

Trailblazers (n.d.) Trailblazers website. Available at: http://www.trail-blazers.org.uk

Upper Yarra Community House (n.d.) New Trax Programme Evaluation 1998-2000. Upper Yarra Community House, Yarra Junction, Victoria.

Ware N & Lucas L (n.d.) Weed and Seed Drug Education For Youth Programme Management Guide. Executive Office for Weed and Seed, Office of Justice Programmes, US Department of Justice, Washington DC.

Youth Justice Board (n.d.) Mentoring for Young People At Risk and Young Offenders. Youth Justice Board, London. Available at: http://www.youth-justice-board.gov.uk/grants/mentoring/index.html

Appendix A

Letter inviting nomination of mentoring projects

21 January 2001

Dear Colleague

Re: A NATIONAL PROFILE AND REVIEW OF MENTORING PROJECTS FOR YOUNG OFFENDERS

The Crime Prevention Branch has commissioned urbis keys young, a social research consultancy firm, to undertake a national audit and review of mentoring projects/programmes for young offenders, including the identification of good practice. We are writing to seek your assistance with identification of the following for this study: mentoring projects (including those regarded as 'good practice'), literature and resource materials, and appropriate key informants to interview.

Mentoring projects are to be included in the study where the following five criteria are satisfied:

mentoring is a significant or primary component of the project, and this has been defined and documented by the project itself (eg in their aims and objectives);
a significant or primary target group of the project is young people who (i) have previously committed criminal offences or (ii) have been identified as being at risk of committing criminal offences;
the target group is young people, including (but not necessarily exclusively) young people aged 18 years or below;
the mentors used are community persons rather than people acting in a professional capacity such as youth workers;
the programme is currently operational.

We request that:

If you are aware of any mentoring projects within the above definition, please send the following information to me by mail (urbis keys young, PO Box 252, Milsons Point, 1565, NSW), email (aniaw@urbis.com.au) or fax (02-99567514) by 15 February 2002: (a) the name of the project, contact details for the contact person (address, phone, email), a brief description of the project, and any other information/comments on the project (eg evaluation reports if available); and (b) if you wish to do so, nominate projects you regard as 'good practice' or innovative in some way, and a brief statement of the reason for this nomination. The consultants are particularly interested in finding out about smaller programmes run by community organisations which may not be as well-known.
If you are aware of any literature about mentoring programmes (including evaluation reports and resource materials such as mentoring guides), please send copies or details of the literature to the above address by 28 February. Material which is unpublished or of limited circulation is of particular interest
If you may be appropriate to interview by telephone as a key informant for the project (or can suggest someone else who is), please contact me by 15 February. A small number of in-depth phone interviews with key informants will be conducted in each State/Territory, with people who can comment (preferably in some detail) on the operation of mentoring programmes in their State/Territory (or nationally), what is regarded as 'good practice' etc.

If you have any questions about the study, please contact me on phone 02-99567515 or by email (see above).

We would also be grateful if you could forward this letter on to any other individuals or organisations who may have an interest in the study.

Thank you very much for your assistance.

Regards

Dr Ania Wilczynski

Senior Research Consultant

Sydney, Melbourne, Brisbane, Adelaide

Level 5 20 Alfred Street (PO Box 252)
Milsons Point New South Wales 2061

Telephone 02 9956 7515 Facsimile 02 9956 7514

urbis Pty Ltd ABN 33 200 814 903 ACN 005 658 293

Appendix B

Postal survey to mentoring projects

Audit and review of mentoring programmes for young offenders: questionnaire

1. BACKGROUND INFORMATION

Name of your programme:
Is this programme:
	Run by a community organisation		[ ]
	Run by government		[ ]
	Other (specify)		[ ]
Auspicing/sponsoring organisation:
Contact person:
Address:
State/Territory		Postcode
Contact telephone:
Contact e-mail:

2. FUNDING

(a) Are the main funding source(s) for this programme:
	Government or mainly government	[ ]
	NGO or mainly NGO	[ ]
	A mix of funding sources	[ ]
	Please give brief details
(b) Approximate annual funding received? $
(c) Is the programme (tick as many as apply):
	a pilot or trial programme?	[ ]
	a programme funded for a fixed time (eg 3 years)?	[ ]
	an ongoing programme?	[ ]
(d) Are you reasonably confident about your funding continuing over the next 2 years or so?
	Yes	[ ]
	No	[ ]

3. OBJECTIVES AND TARGET GROUP

(a) What are the key objectives of the project?
(b) How would you describe your intended target group?
Age group covered:
Gender:
Males only	[ ]
Females only	[ ]
Males and Females	[ ]
Are any particular groups included in your intended target group? (tick all that apply):
Indigenous youth	[ ]
Youth of non-English speaking background	[ ]
Youth who are/have been in custody	[ ]
Youth on a community-based Juvenile Justice order	[ ]
Other (please specify)	[ ]
In practice, does your actual target group differ from your intended target group? If so, how?
(c) What is the geographical area your service covers?
Statewide	[ ]
Capital city only	[ ]
Regional only	[ ]
Country only	[ ]
Other (please specify)	[ ]

4. THE PROJECT

(a) Date (year) the project started?
(b) Current staff numbers:		Full Time	Part Time	Full-time equivalent (total)
(c) Current number of young people being mentored:
	An approximate total of young people mentored during 2001?
(d) Current number of mentors registered:
(e) Is this project 'stand alone' or is it one element of a larger scheme?
	Stand alone			[ ]
	Part of larger scheme			[ ]
	(Name and brief description of that scheme?)
(f) Is mentoring the only activity in this programme?
	Yes			[ ]
	No			[ ]
	IF NO, what are the other main programme activities?
(g) What are the source(s) of referral of young people to your mentoring programme ? (Tick all that apply)
	Juvenile Justice System			[ ]
	Community-based organisations			[ ]
	Self-referral			[ ]
	Other (please specify)			[ ]
(h) Which is the main source(s) of referral of young people to your mentoring programme ?
	Juvenile Justice System			[ ]
	Community-based organisations			[ ]
	Self-referral			[ ]
	Other (please specify)			[ ]
(i) Is participation in the programme mandatory or voluntary?
	Mandatory - on a court order			[ ]
	Mandatory - not on a court order			[ ]
	Voluntary			[ ]
	Other (please specify)			[ ]
(j) Are there any types of young people whom you exclude or screen out from the mentoring programme?
(k) About what level or frequency of contact do your mentors and young people usually have with each other?
	Two or more times a week			[ ]
	Once a week			[ ]
	Two or three times a month			[ ]
	Once a month			[ ]
	Less than once a month			[ ]
(l) How long does the young person/mentor relationship usually last? Is this an 'ideal' length of time? What brings the relationship to an end?
(m) What kind of activities do your mentors and young people typically do together?
(n) How easy has it been for your programme to recruit young people to participate?
	Very easy			[ ]
	Fairly easy			[ ]
	Fairly difficult			[ ]
	Very difficult			[ ]
	IF DIFFICULT, what are the main problems?

5. MENTORS

(a) Are your mentors drawn from a particular group or source or from the general community?
	From sporting bodies/occupations	[ ]
	From other occupational or skill groups (eg artists, photographers etc)	[ ]
	People with particular life experiences (eg family violence)	[ ]
	From the general community	[ ]
	Other (please specify)	[ ]
(b) Briefly, what application/selection processes do you use with mentors? What skills and experience do you (i) require (ii) ideally look for in mentors?
Do you have a process for screening potential mentors for a criminal record?
	Yes - we just ask them to declare this but do not conduct any official checks	[ ]
	Yes - we just ask them to declare this and conduct official State-level checks	[ ]
	Yes - we just ask them to declare this and conduct official national-level checks	[ ]
	Yes - we do not ask them to declare this but we conduct official State-level checks	[ ]
	Yes - we do not ask them to declare this but we conduct official national-level checks	[ ]
	No	[ ]
If YES, please give further details of your process
Are potential mentors with a criminal record automatically excluded?
	Yes	[ ]
	No	[ ]
IF YES, please explain your policy
Do you use any form of psychological testing of potential mentors?
	Yes	[ ]
	No	[ ]
IF YES, please give further details
(c) Are mentors paid or volunteers?
	Paid	[ ]
	Volunteer	[ ]
	IF PAID, approximate hourly rate: $
(d) Do you reimburse any mentor expenses eg, travel, entry fees etc?
	Yes	[ ]
	No	[ ]
IF YES, please give further details
(e) Some people have views about the merits of paid versus unpaid mentors and other financial issues. Do you have any comment about the effectiveness/impact of paying or not paying mentors, reimbursing expenses, etc?
(f) How easy has it been for your programme to recruit and retain suitable mentors?
	Very easy	[ ]
	Fairly easy	[ ]
	Fairly difficult	[ ]
	Very difficult	[ ]
IF DIFFICULT, what are the main problems?
(g) Do you have any system for ongoing support or monitoring of mentors?
	Yes	[ ]
	(brief details please: )
	No	[ ]

6. EFFECTIVENESS

(a) Are there any specific measures or formal criteria that are used to judge the success or effectiveness of the programme overall?
(b) Are there any specific measures or formal criteria that are used to judge the success or effectiveness of the programme for individual clients?
(c) Have there been any internal or external evaluation/reviews of the programme?
	Yes, external only	[ ]
	Yes, internal only	[ ]
	Yes, internal only	[ ]
	No	[ ]
IF YES, Please give details of date, title etc
What were the key findings of the evaluation?
Have you provided a copy of any evaluation to urbis keys young?
	Yes	[ ]
	To be sent	[ ]
	No	[ ]
(d) Is there any other evidence you can use to judge the programme's success or achievements?
(e) What do you feel have been the main strengths or achievements of the programme to date?
(f) And what do you feel have been the main weaknesses or limitations of the project to date?
(g) Generally, would you describe the programme to date as
	Very successful	[ ]
	Reasonably successful	[ ]
	Not very successful	[ ]
	Unsuccessful	[ ]
Comments:

7. OVERALL EXPERIENCE

(a) What have been the biggest challenges or problems faced by the programme to date?
(b) Are there any particular changes that you would make, or things you would do differently, if you were starting this programme again?
(c) Any other advice you would give to other organisations setting up a comparable scheme for young offenders/young people at risk of offending?
(d) Have you had contact with, or advice or input from, any other youth mentoring scheme?
(IF YES, please give brief details)
(e) ANY OTHER COMMENTS you would like to make?

Please send any further documentation or reports (eg mentoring manuals, evaluation reports, annual reports etc) you have available on the programme which has not already been provided to Ania Wilczynski at urbis keys young by either email (aniaw@urbis.com.au) or mail (PO Box 252, Milsons Point, 1565, NSW).

Thank you for your assistance

Appendix C

Questionnaire guides for fieldwork

Audit and review of mentoring programmes for young offenders: discussion guide for project staff

We have found during the course of this project that while there appear to be numerous mentoring programmes operating for generally 'at risk' young people, there are relatively few targeting the specific target group of young offenders or young people identified as at risk of offending. Why do you think this is?
What prompted the establishment of your mentoring programme? What scoping or background development work was conducted before the programme was implemented? What impact has this had on the success or otherwise of the programme?
What factors or programme components have you considered in developing a mentoring programme specifically for young offenders or young people identified as at risk of offending eg in terms of the nature of the mentors, the nature of the mentoring activities, attracting mentees and mentors into the programme etc? How have these differed from the factors or programme components which should be considered for programmes targeting generally 'at risk' young people?
Has your programme been tailored for any specific target groups amongst young offenders eg in terms of age, race, sex, geographical location (metropolitan, regional or remote), disability, offence type, nature of presenting problems, offenders in custody etc? How? How effective has your programme been for those target groups? Why?
What impact have the following issues had on the operation of your mentoring programme and its success:
1. the nature of your organisation (eg government versus non-government, whether it is part of the criminal justice system etc)
2. funding sources
3. whether the mentoring component is stand-alone or one of a range of services provided
4. source of referrals?
How effective have your processes been for (a) recruitment and screening of mentors (b) recruitment of young people?
What have been the characteristics of the mentors you have used in your programme eg in terms of age, personal qualities, whether of a particular profession etc? Have these been the characteristics that you would ideally want? What kind of qualities do you think are required for a good mentor?
What are your processes for matching of mentors and young people? How effective have these been?
If you have processes for monitoring/support of the mentor/mentee relationship once established, how effective have these been?
What links does this programme have with other services? How have these assisted (or hindered) the implementation and effectiveness of the project?
Do you regard your project (or components of your project) as demonstrating 'good practice'? Why or why not?
Is there anything that could be done to improve the operation of your mentoring programme? Are there any plans to implement these improvements?
Have there been any aspects of the programme that you have modified over time to improve its operation and effectiveness?
What are the essential features/components which would need to be in place for your mentoring programme to be sustainable in the longer term? Are these features/components in place now?
What evidence of positive (or negative) outcomes has there been for your project eg in terms of recidivism, the development of personal skills etc?
Do you think that overall mentoring programmes for young offenders are an effective approach for dealing with young offenders and young people identified as at risk of offending? Why or why not?
Do you think there should be more, less or about the same emphasis placed on providing mentoring programmes in the future?

Audit and review of mentoring programmes for young offenders: discussion guide for stakeholders

As you are aware, the project is only examining mentoring projects which have a significant or primary target group of young people who have either committed criminal offences or been identified as being at risk of committing criminal offences. The following questions relate to programmes of this nature.

You may not be able to answer all of the questions below - if not, we will concentrate on those questions which you feel able to respond to.

We have found during the course of this project that while there appear to be numerous mentoring programmes operating for generally 'at risk' young people, there are relatively few targeting the specific target group of young offenders or young people identified as at risk of offending. Why do you think this is?
Do you have any comments about the following aspects of mentoring programmes for young offenders, in terms of (a) what represents good practice (b) difficulties experienced by mentoring programmes and the reasons for this?
1. sources/nature of funding
2. the nature of the organisation providing the programme (and whether it is formally connected to the criminal justice system)
3. whether the mentoring component is stand-alone or one of a range of services provided
4. effective establishment of the programme
5. recruitment and screening of mentors
6. characteristics of mentors (eg age, personal qualities, whether of a particular profession etc)
7. sources of referrals of young people as mentees
8. recruitment of mentees
9. the use of voluntary versus paid mentors, and where voluntary the availability of discretionary funds to cover costs of outings, transport etc
10. ecruitment and retention of paid staff to coordinate the programme
11. the process of matching mentors and mentees
12. monitoring/support of the mentor/mentee relationship once established
13. nature of the activities conducted and quality of the relationship developed between mentor and mentee
14. frequency of contact and length of the mentor/mentee relationship
15. links between the programme and other services
16. evaluation of the programme.
Are there any other aspects of the development or operation of mentoring programmes which you regard as either (a) good practice or (b) typical difficulties for mentoring programmes?
What are the essential features which must be in place for a mentoring programme to be sustainable in the longer term?
What could be done to improve the operation of existing mentoring programmes?
What factors or programme components need to be considered in developing a mentoring programme specifically for young offenders or young people identified as at risk of offending eg in terms of the nature of the mentors, the nature of the mentoring activities, attracting mentees and mentors into the programme etc? What differences are there are there from the factors or programme components which should be considered for programmes targeting generally 'at risk' young people?
Do you think that overall mentoring programmes for young offenders are an effective approach for dealing with young offenders and young people identified as at risk of offending? Why or why not? What evidence of positive (or negative) impacts on the young person is there in terms of recidivism, the development of personal skills etc?
Are there specific target groups amongst young offenders for whom mentoring is particularly effective or ineffective eg in terms of age, race, sex, geographical location (metropolitan, regional or remote), disability, offence type, nature of presenting problems etc?
Do you think there should be more, less or about the same emphasis placed on providing mentoring programmes in the future?

Audit and review of mentoring programmes for young offenders: discussion guide for mentors

Can you each tell me a bit about your backgrounds, and why you wanted to become mentors?
Can you also each tell me who is the young person who you are working with, how long you've been seeing them, how often you see them, what kinds of things you do together?
What do you think of the mentoring programme overall?
What have been the good things and the bad things about the programme and how it operates? Is there anything you would change about it if you were running it?
Has participating in the programme made a difference to the young person you are working with? How? Has it had any impact on their offending behaviour? Have they learnt any new skills or more effective ways of dealing with particular situations? Can you give me any examples of situations where you've seen this with your young person?
Has there been any benefits - or disadvantages - for you as a mentor?
What are the qualities you think you need to be a good mentor?
How good was the match/matches between you and the young person/people you have mentored?
What kind of young people do you think mentoring is most effective for?
Have you got any comments about the following aspects of the mentoring programme,
1. in terms of (a) what represents good practice (b) difficulties experienced by mentoring programmes and the reasons for this?
2. the nature of the organisation providing the programme (and whether it is formally connected to the criminal justice system)
3. whether the mentoring component is stand-alone or one of a range of services provided
4. effective establishment of the programme
5. recruitment and screening of mentors
6. characteristics of mentors (eg age, personal qualities, whether of a particular profession etc)
7. sources of referrals of young people as mentees
8. recruitment of mentees
9. the use of voluntary versus paid mentors, and where voluntary the availability of discretionary funds to cover costs of outings, transport etc
10. recruitment and retention of paid staff to coordinate the programme
11. the process of matching mentors and mentees
12. monitoring/support of the mentor/mentee relationship once established
13. nature of the activities conducted and quality of the relationship developed between mentor and mentee
14. frequency of contact and length of the mentor/mentee relationship
15. links between the programme and other services
16. evaluation of the programme.

Audit and review of mentoring programmes for young offenders: discussion guide for young people

Introduce self, explain nature of project.

We'll be using these interviews to get an idea what the young people who've used the programme think about it and mentoring generally. Just to explain how we'll be using what you tell us, we might use some quotes from you, but they would be kept anonymous - that is, we won't put your name with the quote, so it's kept confidential.

As you know, we'll be giving you $25 at the end of the interview to thank you for taking the time to talk to us.

Obtain informed consent.

Can you tell me a bit about (a) yourself, your family etc and (b) who your mentor is, how long you've been seeing them, how often you see them, what kinds of things you do together?
How did you end up in the programme?
What do you think of the mentoring programme? Has it made a difference to you? How? How would you describe your mentor?
Like at school - if you had to give the programme a mark out of ten, what would you give it?
What have been the good things and the bad things about the programme?
How good was the match between you and your mentor?
What kind of person makes a good mentor?
Has seeing your mentor helped you learn anything new or better ways to deal with things?
Has the programme helped you stay out of trouble with the law?
What would you do if you were in charge of the project? Would you make any changes to it?
Would you recommend the project to your friends if they were in the same position?

Early Intervention Youth Mentoring Programmes

An overview of mentoring programmes for young people at risk of offending

Foreword

Contents

Abbreviations

Executive Summary

Objectives

Methodology

Overview of mentoring programmes

Models and organisational structure: good practice

Programme objectives

Naturalistic versus formal

Stand-alone versus integrated projects

Staff

Establishment and administration

The mentoring process: good practice

Initial meetings

Length of contact

Frequency of contact

Parental involvement

Activities undertaken by the mentor and mentee

Ending the mentoring relationship

Monitoring and supervising mentors

Training and orienting mentors

Finding and employing mentors: good practice

Recruiting mentors

Characteristics of mentors

Screening procedures

Excluding mentors with criminal records

Matching

Paid versus volunteer mentors

Selecting mentees and developing a programme: good practice

Recruiting

Screening and assessing mentees

The programme activities

Mentoring programmes for young offenders

Evaluation: an essential part of good practice

Determining the effectiveness of mentoring

Reduced offending

Reduced substance misuse, and other risky behaviours

Increased participation/performance in education, training and employment

Improved self-esteem, social/communication skills and personal relationships

Mentoring is only suitable for some young people

Indigenous programmes

Recommendations for Stage 2

National workshop

A good practice manual

Evaluation

Funding existing programmes

Introduction

What is mentoring?

Background to the research

Objectives of the project

Structure of this report

Methodology

Definition of mentoring

Key components of the methodology

Literature review

Initial scoping consultations to identify mentoring projects

Postal survey

Fieldwork

Face-to-face consultations

Supplementary telephone and face-to-face interviews

Meeting with crime prevention officers

Data analysis and reporting

Some caveats

Notes

Organisational context of mentoring programmes

Location of mentoring projects

Auspicing organisation

Funding

Formal versus naturalistic mentoring programmes

Programme objectives

Stand-alone versus integrated projects

Staff

Establishment and administration

Initial scoping

Scoping and consultation for Indigenous projects

Strong support structure

Initial meetings