Patricia North & Steven Wilson Purdue University

best practices in youth MentoringPresentation to 5th Annual “Connections Count” SummitApril 8, 2010 Patricia North & Steven Wilson Purdue University

Today’s Agenda • Definitions of Youth Mentoring • Review of Research Literature • Three Key Conclusions • Evidence-Based Best Practices • Conclusions/Implications

Definitions of Youth Mentoring “Mentoring is a structured and trusting relationship that brings young people together with caring individuals who offer guidance, support and encouragement aimed at developing the competence and character of the mentee. A mentor is an adult who, along with parents, provides a young person with support, counsel, friendship, reinforcement and constructive example. Mentors are good listeners, people who care, people who want to help young people bring out strengths that are already there. A mentor is not a foster parent, therapist, parole officer, or cool peer.” Mentor/National Mentoring Partnership, 2010. Retrieved from http://www.mentoring.org/mentors/about_mentoring/ Mentoring is “a relationship between an older, more experienced adult and an unrelated, younger protégé –a relationship in which the adult provides ongoing guidance, instruction, and encouragement aimed at developing the competence and character of the protégé.” Jean Rhodes, 2002, p. 3 from her book Stand by me: The risks and rewards of mentoring today’s youth, Cambridge, MA: Harvard University Press.

Definitions of Youth Mentoring • Most definitions of youth mentoring have three core elements (see David Dubois and Michael J. Karcher, Handbook of Youth Mentoring, 2005, p. 3-4): • The mentor is someone with greater experience or wisdom than the mentee. • The mentor offers guidance or instruction that is intended to facilitate the mentee’s growth and development. • There is an emotional bond between mentor and mentee, a hallmark of which is a sense of trust.

Definitions of Youth Mentoring • Scholars do not all agree on single definition • Varied definitions raise questions: • Must the mentor be older? If so, by how much? • Must the mentor be unrelated to mentee? • Can a mentor be paid? • How long must a relationship last for it to be considered mentoring? • What is the relationship between mentoring and tutoring? (developmental/instrumental mentoring)

Types of Mentoring Relationships • Natural or informal mentoring relationships can (though do not necessarily) evolve out of a variety of roles that adults play in the lives of youth (e.g., extended family members, neighbors, coaches, religious group leaders) • We will focus on more formal programs that are designed to match mentees and mentors. • Such programs aim to foster relationships that benefit youth in the same ways that natural mentoring relationships can benefit youth. Natural/Informal Formal

Research Literature • DuBois, D.L., Holloway, B.E., Valentine, J.C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157-197.

Research Literature: The Method • Meta-analysis combines (quantitatively) results from many individual studies on the same topic to see how similar/different findings are across studies • Meta-analysis gives us an idea of the big picture: by combining many studies, we see how much impact youth mentoring has on a much larger scale (vs. just in single study) • Meta-analysis clarifies which aspects of mentoring programs influence their impact – • e.g., comparing the impact of mentoring programs on different outcomes (academic vs. social) • e.g., comparing programs that had clearly defined expectations for the minimum level of mentor/mentee contact that should occur per week vs. programs that did not set such expectations

Research Literature: The Method • DuBois analyzed evaluation studies of programs meeting the following criteria. To be included in the meta-analysis, a program must have involved: 1. mentoring in a one-on-one relationship; programs where mentoring occurred on a small group basis (1mentor, 5 mentees) were not included 2. mentors who were older or more experienced than the mentees; peer tutoring programs were not evaluated, but programs where older youth mentored younger children were included 3. mentors whose official job responsibility was not working professionally with the mentee (e.g., not a social worker assigned to the youth) 4. menteeswho, on average, were younger than 19 years old 5. assessment of program outcomes either by: (a) preprogram vs. post-program comparison of mentored youth or (b) comparison of mentored youth with youth who did not receive mentoring

Research Literature: The Method • Based on these inclusion criteria, the meta-analysis included: • 60 independent samples (from 55 published studies) • Samples ranging in size from 40 to 1000+ youth • 8 of the 60 samples were from BBBS programs

Research Literature: The Method • Each study also was coded on multiple characteristics divided into 6 categories: 1. report information (e.g., year) 2. evaluation methodology (e.g., type of research design) 3. program features (e.g., program goals, geographic location, setting, compensation of mentors) 4. characteristics of youth (e.g., gender, at-risk status) 5. mentor-mentee relationships (e.g., frequency of contact) 6. assessment of outcomes (e.g., type of outcome, timing of assessment)

Research Literature: The Method • DuBois developed a list of theory-based best practices based on previous recommendations for establishing effective youth mentoring programs. • Based on the meta-analysis, DuBois identified empirically-based best practices. Two of these were not included on the theory-based list. The others confirmed some of the theory-based practices. • Using both procedures, he identified a total of 12 best practices • He divided studies into those programs that • Included at least half of the best practices (6+ out of 12) vs. • programs that included fewer than half (5 or less of the 12).

Research Literature • Three Key Conclusions: • Overall impact of youth mentoring is modest • Programs that incorporated best practices had greater impact on youth • Mentoring can have no impact or even negative impact on youth

Conclusion 1: Overall of youth mentoring impact is modest • “Based on available findings…youth mentoring programs do indeed have significant capacity to reproduce through more formal mechanisms the mentoring relationship between youth and adults.” • HOWEVER, “Results further indicate…that it may be most appropriate to expect the typical youth participating in a mentoring program to receive benefits that are quite modest in terms of absolute magnitude.” The amount of impact is small when compared to evaluations of other types of interventions with youth (e.g., family counseling or social skills training).

Conclusion 2: Some programs had neutral impact to negative impact • Example: Did programs assess the fidelity of program implementation? (i.e., did they assess whether key program principles actually were being followed on an ongoing basis?) • 15 programs did not describe any methods they used to assess the fidelity of program implementation • When combined, these 15 programs had no positive impact on mentees’ academic, social, and/or emotional outcomes (pre vs. post-program, or in comparison to youth who did not receive a mentor) • 45 programs included some assessment of fidelity of program implementation • These 45 programs had modest positive impact on mentees academic, social, and/or emotional outcomes

Conclusion 3: Utilizing best practices enhances program effectiveness • Just more than half (33) of the 60 programs followed the majority (6+ of the12) of the best practices • Programs that followed the majority of best practices had modest positive impacts. • Programs that did not follow the majority had no positive impact (i.e., neutral).

Evidence-Based Best Practices • Recruiting, Matching & Training Mentors • Programs should screen prospective mentors (e.g., background checks, in-person interviews, home visits and/or psychological testing) • Consider recruiting mentors (volunteers) with ‘helping’ backgrounds • Larger impact shown for mentors with helping backgrounds or from helping professions (e.g., teachers, parents/caretakers) • Carefully match mentors/mentees based on a variety of factors (i.e., in addition to considering gender/race, also consider interests, physical limitations or other characteristics of potential match) • Utilize pre-match training to prepare mentors for relationship (e.g., instructor led, group or individual, utilized prepared materials, self study)

Evidence-Based Best Practices • Developing Mentoring Relationships • Set expectations for frequency of contact and length of relationship • Larger impact shown for programs that set expectations regarding frequency of contact • Plan for supervision of mentors during mentoring relationship (e.g., in-person meetings, telephone conversations, mail) • Provide ongoing training for mentors (e.g., cover additional training topics pertinent to youth) • Larger impact shown for programs that continued to train mentors

Evidence-Based Best Practices • Developing Mentoring Relationships (cont’d) • Develop support groups for mentors (e.g., allowing mentors to share challenges and successes of mentoring relationships may prevent mentor frustration/drop out) • Provide structured activities for mentor/mentees (e.g., Halloween party, members from local college football team lead ‘training camp’) • Programs that offered activities resulted in larger impact • Ensure mechanisms for parental involvement; parents should be aware of mentoring relationship and familiar with mentor/activities but not an active part of mentor/mentee meetings • Programs that encouraged support or involvement from parents showed larger impact

Evidence-Based Best Practices • Evaluating Program Implementation & Impact • Monitor program implementation to ensure intended program practices are followed • Programs that monitored program implementation had positive impact on youth

Evidence-Based Best Practices • Planning at the Program Level • Consider the setting (e.g. school-based, community-based) • Research suggests smaller impacts for school-based programs; perhaps due to shorter periods of contact and/or shorter relationship lengths that result from academic schedule (e.g., summer vacation) • For established programs, it may be useful to consider the unique characters of a setting that are beneficial or challenging to fostering mentoring relationships. • e.g., School-based programs can benefit from mentors and mentees being in the presence of peers. This can provide opportunities for mentor to give positive feedback to mentee and/or model behavior in front of peers. • At the community level, this may mean thinking about the mix of mentoring types available to youth. Mostly community based? Mostly school based? Where are the gaps?

Conclusions/Implications • Youth mentoring is not a panacea (culturally, we want to believe it works, but that alone does not guarantee positive outcomes for youth) • There’s no “magic set” of evidence-based best practices (lists will evolve as program evaluation research continues to grow) • As with all best practices, the application of these practices will vary depending on program goals, participants, resources, etc. • Ongoing conversations between program providers and prevention/evaluation researchers about best practices are crucial • Hopefully, the ideas about best practices presented here will promote useful discussion throughout the summit and beyond

Patricia North & Steven Wilson Purdue University