Abstract and Introduction
This study adds to the limited research on the potential importance of the quality of the relationship between adult prevention service providers and youth participants in enhancing social skills and strengthening prevention outcomes. Study subjects were drawn from seven prevention programs funded under a Youth Mentoring Initiative by the Center for Substance Abuse Prevention in the Substance Abuse and Mental Health Services Administration. These programs maintain a relationship-based service focus but use a variety of one-on-one, group, volunteer, and paid staff service formats. Study results showed that youth who perceived a higher level of trust, mutuality and empathy in their relationship with providers experienced significantly greater improvements in social skills (i.e., cooperation, self-control, assertiveness, and empathy) than program participants who perceived a lower quality relationship with adult providers. These findings underscore the importance of recruitment, training and supervisory practices that promote staff and volunteer skills in achieving high quality relationships with youth participants regardless of the specific intervention strategy. Editors' Strategic Implications: Practitioners and policymakers should review the authors' findings about the importance of individual adult skills in building protective mentoring relationships. The impact of relationship quality, rather than setting, suggests that the scope of effective prevention practice can be broadened beyond the confines of formal prevention programming to any place in which caring and skilled adults interact with youth.
Positive relationships between the service provider and the service recipient have long been acknowledged as a potentially important contributor to intended outcomes in a variety of helping services. For example, it is recognized that successful counseling involves process factors that are common across different counseling models; the counselor's ability to establish an open, trusting, and collaborative relationship with the client is chief among these (Austin 1999). Empathic understanding and unconditional commitment have also been identified as components of a positive counseling relationship (Austin 1999). Whereas the quality of the provider's bond with the participant has been shown to be effective in a variety of service settings, there are strong reasons to predict its importance for achieving positive outcomes, specifically in prevention programs. Two supporting theoretical models are related to adult and youth relationships in prevention practice; these are briefly elaborated below.
The Caring Adult Model. A supportive family environment is one source of positive adult relations with youth, but resiliency research has also shown that a caring adult outside the family can also have a protective influence. In traditional communities, these "natural mentors" were a regular part of the community fabric. In modern society, both supportive family members and natural adult members in the community are less available to young people (Rhodes 2002). Relationship-based programs that target youth are often designed to compensate for the reduced availability of adult family members and natural community members through activities that connect youth with caring adults, typically in one-on-one mentoring relations. The question for prevention research has been to determine to what extent organized relations with adults can produce protective effects for young people. It is also important to identify those characteristics of adult and participant relationships that are important for achieving prevention outcome objectives.
Developmental and Prescriptive Relationships. The distinction between "developmental" and "prescriptive" relationships of mentors and protégées (also called mentees) is a common theme in mentoring research that punctuates the important link between the quality of provider–participant bond and service success (Benard 1996a; Tierney et al. 1995). A developmental relationship is characterized by caring, respect and positive expectations, and opportunities for participation and contribution by the protégé (Benard 1996a). They focus initially on developing friendship and trust. Once trust is established, they can begin focusing on other topics and objectives. Within this relationship, the ability to address specific issues of concern or importance to a youth will grow. The program does not specifically target or directly address pre-determined problem areas (Tierney et al. 1995).
In contrast, in prescriptive relationships the adult role is to guide youth toward values, attitudes, or behaviors "that the adult deems positive" (Benard 1996a, p. 8). Prescriptive mentors are problem-focused and orient mentoring activities toward fixing these problems. These relationships are based on the belief that the adult has more life experience and wisdom and that it is the adult's responsibility to provide guidance for the youth, who implicitly does not have these advantages.
From an outcome perspective, research has shown a clear and consistent difference in the results achieved through prescriptive and developmental relationships between the adult and child. Most clearly, the research has found that the prescriptive orientation does not produce desired results. In programs identified as prescriptive, mentoring relationships are not associated with positive program outcomes, nor are they associated with sustained mentoring matches (Benard 1996a). Some studies have found that matching youth with mentors in prescriptive relations has iatrogenic effects; it is actually significantly associated with negative outcomes such as reductions in protective factors (e.g., Flaxman and Ascher 1992; Slicker and Palmer 1993).
Adding to the understanding of why developmental relationships contribute to positive outcomes, Rhodes (2002) has identified mechanisms through which caring adults positively impact the social-emotional and cognitive development of adolescent youth and how they act as positive role models and advocates (Rhodes 2002). These benefits are achieved through empathy, positive feedback and meaningful conversation. However, research supports a conclusion (Herrera etal. 2000) that the ability to engage in the specific and sustained activities necessary to engage youth requires prior establishment of a relationship in which
"...the two people involved feel connected–that there is mutual trust and a sense that one is understood, liked, and respected. Without some connection, the dynamics that make mentoring relationships effective are unlikely ever to occur" (Rhodes 2002, p. 36).
In summary, research increasingly demonstrates that adult-youth relationships characterized by mutuality, trust and empathy are a fundamental contributor to positive prevention outcomes (Rhodes 2002).
Purpose of the Study. In prevention practice, the primary focus of "relationship-based interventions" (Rhodes et al. 2002) has been the promotion of traditional, one-on-one, adult to youth mentoring programs. However, the literature indicates that traditional one-on-one programs have limited effects in achieving strong prevention outcomes. Catalano et al. (2004) reviewed rigorously evaluated youth development programs and found just 16% to have significant effects on outcomes. We reviewed 25 universal or selective youth development programs (out of a pool of 161) selected for strong evaluation designs (control or comparison groups), acceptable standard of statistical proof, adequate methodological detail to allow assessment of study soundness, and evidence of significant effects. Two mentoring programs, Big Brothers, Big Sisters (BB/BS) and Across Ages, met these criteria. Although the BB/BS study showed statistically significant differences between participant and comparison youth on pre-post tests, group difference effect sizes of .05 were well below the typical standard of .20 as a small but meaningful effect (DuBois et al. 2002a).
As Rhodes (2002) has clearly demonstrated, there is a strong link between the demanding implementation requirements of successful mentoring programs and the tenuous and inconsistent reports of strong outcomes. Successful mentoring requires careful implementation, including mentor recruitment, successful matching, mentor support, achieving lasting mentoring matches, and supporting effective mentoring activities. Establishing strong interpersonal relationships is an important ingredient in mentoring, but too often this ingredient is seriously diluted by the inability of programs to achieve the demanding requirements of a strong one-on-one mentoring program. As a result, many providers turn to alternatives that maintain a relationship-based focus yet depart from traditional one-on-one mentoring.
The purpose of this paper is to report youth outcomes from participation in traditional and non- traditional mentoring type interventions funded by SAMHSA's Center for Substance Abuse Prevention (CSAP). As the previous program descriptions indicate, the programs in the study delivered relationship-based services in a variety of group settings. Understanding the potential role of the quality of provider–participant relations in these settings will produce knowledge that can apply to any relationship in which adults provide prevention services to youth. Practical experience with traditional one-on-one mentoring suggests that relationship-based interventions may require the development of alternative formats for building mutual, trusting, and empathetic relationships between caring adults and youth if these interventions are to be taken to scale and serve large numbers of youth at risk. This study was a first step in providing evidence on the contribution of quality adult relationships with youth to program success in achieving prevention objectives and social skills in a broad range of prevention interventions and settings.
The Description of the Study Programs. The CSAP Youth Mentoring Initiative, in both its first cohort (referred to as Project Youth Connect) and in the second cohort studied here, was designed to provide an opportunity for grantees to implement relationship-based prevention interventions in formats that modified the traditional one-on-one mentoring relationship. In the request for applications, the initial program design stipulated that grantees plan a ten-month intervention that provided a minimum of two hours per week of one-on-one youth contact with an adult. However, this requirement was not fully achieved in either grantee planning or implementation.
The funded interventions offered life skills groups, homework assistance, recreation and other interventions that varied widely across programs. In some of programs, staff were paid to deliver one-on-one services to youth, and in a few others, volunteers were recruited from the community. Program length varied from 26 to 47weeks, and average intensity ranged from one to four and a half hours per week. Sites were located across the country, primarily in urban locations (six of seven sites). Accordingly, these programs provide an opportunity to study the degree to which the quality of relationships between the adult provider and youth participants affects program outcomes across a variety of intervention strategies and formats.