Youth Mentoring as a Means of Supporting Mental Health for Minoritized Youth: A Reflection on Three Theoretical Frameworks 20 Years Later

Abstract

Youth mentoring as a field of study has grown immensely in recent years, with hundreds of peer-reviewed research articles on the subject. A key driver of this interest is the demonstrated ability of youth mentoring to support positive mental health for minoritized youth. Three central theoretical models, published nearly twenty years ago, drive the majority of this body of research: the systemic model, the relational model, and the mechanisms of mentoring model. The present paper examines these theoretical models through conversation with their authors and presents their reflections and insights, the contexts in which these models were originally written, and the similarities and differences among them. By understanding the origins of these three influential theoretical models, what they center, and what they do not center, we can begin to consider the ways in which the body of work on youth mentoring is framed. Ultimately, these analyses and reflections outline future directions for the field and a forthcoming updated conceptual model of youth mentoring that centers issues of equity and social justice.

1. Introduction

Youth mentoring, as in relationships with caring non-parental adults who provide support for a young person, is a flexible, wide-reaching support that has been demonstrated to be associated with a wide range of positive outcomes [1,2,3,4], including the promotion of well-being for minoritized youth [5]. Youth mentoring as a research field has grown rapidly in the past 20 years, with hundreds of articles and several books written on the subject in that time. Three central theoretical models are most commonly referenced in these studies: a systemic model of youth mentoring [6], relational processes in youth mentoring [7], and a model of the mechanisms of change in youth mentoring [8]. These three models work in tandem with one another to describe how youth mentoring works to support mental health for minoritized youth.

While these three foundational models laid the groundwork for the youth mentoring field, it is important to recognize the philosophical orientations that are centered, and not centered, in each as a way to reflect on the knowledge gleaned by mentoring scholars to date. It is also important to consider what these models contribute to the youth mentoring field almost 20 years after their initial publications. The three authors of this manuscript are scholars who have recently entered the field and benefited from these three foundational models. We are thankful to Keller, Spencer, and Rhodes for their generosity in speaking to us about the origins of these models and, in turn, the origins of the field. Indeed, they engaged in their own mentoring as they spoke with us.

1.1. Systemic Model of Mentoring

Keller [6] introduced the systemic model of mentoring framework. The model depicts how the focal youth/mentor relationship sits within a broader network of relationships, namely those between the parent and the mentoring agency, the mentoring agency and the mentor, and the mentor and the parents. Please see Figure 1 below for a depiction of this model.

The success of the child’s mentoring relationship depends on the quality of these secondary relationships among the adults, which should thus be included in any study of the primary mentoring relationship. Indeed, these important secondary relationships have been demonstrated to influence the effectiveness and quality of the primary relationship and potentially impact the outcomes of said relationship [9].

1.2. Relational Processes in Youth Mentoring

While Keller’s model can be seen as the 4000-foot view of the social context of youth mentoring, Spencer’s model provides the most up-close depiction. The relational processes in the youth mentoring model [7] are built on qualitative research aimed at describing the relational processes that act as precursors of change in youth mentoring relationships. According to this model, there are four core relational processes that are necessary for a successful mentoring relationship: (a) authenticity, (b) empathy, (c) collaboration, and (d) companionship. While authenticity is defined as when mentors and youth are able to express their genuine feelings and be real with one another, empathy refers to the mentor’s capacity to understand life circumstances from the adolescents’ perspective. Collaboration, the third essential precursor according to the model, refers to mentors and youth both having a say in the relationship and working together towards a common goal. Companionship, in turn, refers to the emotional well-being that results from mentors and youth liking each other and enjoying each other’s company. Please see Figure 2 below for a depiction.

1.3. Mechanisms of Youth Mentoring

Rhodes’s mechanisms of mentoring model of youth mentoring reflects a combination of elements from the previous two models and adds important details like interpersonal history and community context [8,10]. The final result is an equation-like depiction, which can be seen in Figure 3 below [11]. We refer to this model as the “mechanisms of mentoring model”, as it is the only one of the three models to include an equation-like series of events that lead to theorized outcomes (social–emotional development, identity development, and cognitive development). Rhodes’s model complements Spencer’s work by highlighting mutuality, trust, and empathy as essential precursors to positive change while also reflecting Keller’s perspectives on how mentoring can influence parental and peer relationships, depicted as mediating outcomes.

The mechanisms of mentoring model has been highly influential in the youth mentoring literature [4,12,13,14]. The original book proposing the model has been cited over 700 times, and of the three models included in this paper, this model is most often used as a guiding framework for studies evaluating mentoring programs.

2. Materials and Methods

While we know that young people generally benefit from youth mentoring and the growing body of literature around youth mentoring, it is important for us as a field to consider the foundations that our studies build on and where these foundations come from. By understanding the origins of these three influential theoretical models, what they center, and what they do not center, we can begin to consider the ways in which the body of work on youth mentoring is framed. Ultimately, these analyses and reflections can lead the field to an updated conceptual model built on the original three but also the needs of the youth mentoring field 20 years after these models were first published in light of what has yet to be centered.

Thus, this paper seeks to (1) reflect on the genesis of these three foundational theoretical models, how they were developed, and what philosophical orientations they represent, (2) present the continued philosophical differences among the three models and their authors, (3) evaluate which values have been centered in the theoretical models that make up the groundwork of the field, and (4) present ideas for future directions from both the three foundational authors’ perspectives and our analyses.

We sought to understand the context of the foundational theoretical models of the mentoring field. We selected the three models included in this paper, which we will refer to as the systemic model, the relational model, and the mechanisms of mentoring model. We acknowledge that other models have come out since these seminal works, but we assert that these are the foundational ones in the literature. As one purpose of reflecting on the history of the field is understanding what we do and do not know based on early decisions, we thought these models should be included for their early publication relative to the field and their continued use in more recent studies.

In order to reflect on these models, their application, and their impact on our knowledge base, we reached out to the main authors of the three models: Tom Keller for the systemic model, Renee Spencer for the relational model, and Jean Rhodes for the outcomes-oriented model. There was no real incentive for these authors to have a conversation with us on the subject, other than a continued contribution to the collegiality often seen in our field.

All three models’ authors responded to an initial email asking for their time. All three authors of this paper were present for each of the three conversations. Conversations lasted around 60 min. They were not recorded, but all three authors of this paper took active notes throughout. The conversations followed a semi-structured format, wherein a few questions were asked to all three models’ authors (e.g., how did the theoretical model come about? What has the model contributed to the field? What do you think is most relevant now about the model and what would you change?). The rest of the conversations, however, flowed from these few core questions.

The three authors of this paper met after each conversation and exchanged notes on the topic. One author then combined all notes to decipher common themes. The other two authors provided feedback in an iterative fashion until the outline of the present paper was finalized. A preliminary draft of the present paper was then presented to the three interviewees, who were invited to make comments and edits throughout.

3. Results

3.1. Models Were Born from Clinical/Practice Experience by Early Career Scholars Bringing in Theories from Many Different Literature Bases

3.1.1. Born from Clinical and Practice Experience

Before going back to school for his PhD, Tom Keller worked at a Big Brothers Big Sisters chapter for approximately five years. Tom noted: “we had wise colleagues who really understood relationships and were very astute when it came to dealing with people, doing screening and assessment [15]”. Despite this depth of practice knowledge underscoring the importance of all relationships involved, from the relationship between Match Support Specialist and youth to the one between youth and their parent, this network was not addressed in the literature. As someone who worked at BBBS, Tom felt that that Match Support Specialist role was important in helping to “make or break a mentoring relationship”. He also knew that a lot of time was spent in working with the parent and the family as a whole [15].

Like Tom, Renee had a wealth of practice experience before starting her doctorate as a clinical social worker working in private practice and psychiatric hospitals [16]. “Some of my relationships went really well, some not so well, and I was deeply interested in what that was about [16]”. She noted that sometimes interventions seemed to work in one client relationship but not in another, and she wondered, “what are the core aspects of the relationship that are associated with that [16]?” She had always loved working with adolescents, the “truth tellers” who challenge us in profound ways. It was not until she met Jean Rhodes, who suggested that she could study a relationship’s progression from the beginning with all of its core aspects she was so keenly interested in, that she studied mentoring [16].

Jean approached her work with mentoring in a much more student-oriented way. She read George Albee’s work on psychopathology and noted that it looked much like an equation (e.g., linear, one concept plus another in the context of a third would produce this result [17]). She wanted the burgeoning field of youth mentoring at the time to consider their work in an equation-like approach so “that was how I began to think about the model [18]”.

3.1.2. Gaps in Both Academic and Practice Literature Bases

Motivation for Renee’s work grew out of both her clinical experience and exposure to the risk and resilience literature. A primary finding of the burgeoning risk and resilience literature at the time she started her doctoral studies was that young people faced with significant adversity who have a supportive relationship with an adult often had much better psychosocial outcomes than those without such a relationship. This finding had been replicated over and over with different populations (e.g., international populations, young people whose parents had major mental illness, kids of divorce, etc.). “When I asked a professor who was an expert in risk and resilience research what about those relationships makes such a difference”, Renee noted that the response was, “we don’t really know [16]”.

Tom also took his experiential knowledge into his PhD program, where he started learning about the theories behind relationships. The theories gave him a language and framework for what he and his colleagues had already been seeing and doing at the agency [15]. His ultimate motivation for the systemic model was that if he had known some of these theories and their application to mentoring earlier, he believes he could have been more effective in his BBBS work [15].

3.1.3. Theoretical Bases behind Models

Before authoring their respective models, Renee and Jean were both taken by the idea that qualities of a helping relationship itself promoted positive outcomes, regardless of the modality used within the relationship. Jean’s model, although inspired by Albee’s work on psychopathology, was also influenced by attachment theory and the idea that a mentoring relationship could be a corrective experience. She was also influenced by Vygotsky’s theories of cognitive development and Waterman’s identity theory [18,19,20].

Renee’s work, initially inspired by gaps in the resilience literature, also brought in new feminist and relational theories of psychotherapy, which were much more strength-based and less pathologizing and acknowledged the structural forces at work in people’s lived experiences in comparison to more traditional views on the helping relationship [7,16,21]. Relational psychoanalysis and relational cultural theory acknowledged that rather than relationships being a part of development, relationships are the mechanisms through which development happens, including the relationship between the client and the therapist [16,22]. “If you compare CBT with interpersonal therapy”, Renee noted, “there are more within group differences in effectiveness than there are between groups [16]”. Cross-cutting factors shaping the client/therapist relationship itself are thus probably more important than the modality used. Carl Rogers [23] posited that common factors in a therapeutic relationship, such as empathy, authenticity, and positive regard, were essential to positive outcomes independent of the modality used by therapists [16]. These were thus the starting place for Renee’s model emphasizing essential relational processes that were needed for effective mentorship.

Tom developed his foundational model using a family systems perspective and general systems theory. Tom’s systemic model of mentoring thus has a much wider lens of the mentoring intervention, including the essential relationships that surround the mentoring relationship (e.g., relations between the parent and the youth, the match support specialist and the parent, the mentor and the parent, etc.). This complex and integrated system of interdependent relationships is born directly from family systems theory, which emphasizes how changes in one relationship can affect other relationships [24]. In turn, family systems theory draws on general systems theory regarding principles for the operation/maintenance and coherence/continuity of a relational system [24,25].

3.1.4. Early Career Exploration

Because a purpose of this paper is to reflect on the context of these initial models, it is also important to note that these models were a part of early career exploration for its three authors. When Tom published his model, he was a new professor “without a lot of mentoring data” and wrote a conceptual paper for a special issue [15]. Renee’s model, similarly, was born out of a “working paper in grad school comparing relational theories of psychological development and therapy and applied risk and resilience literature” [16,26]. Her model was also born directly from her dissertation work, in which she described the key motivation as “if I listen to mentors and youth in relationships that have gone the distance, what might I learn about what’s happening [16]?”.

3.2. Models Have Helped Support New Learning in the Field

3.2.1. Social Network Perspective

After publishing the systemic model of mentoring, Tom was asked to co-author a chapter in the second edition of the Handbook of Youth Mentoring on the social network perspective [9]. He sees that as an “outgrowth” of the systemic model, as the larger social network includes the few key players seen in the original model [15]. This larger social network being recognized could center familial capital—informal mentors from within extended- and fictive-kin networks—as an important mechanism of change for minoritized youth [27,28]. The systemic model has been used in recent years to understand how mentor–parent collaboration is associated with mentoring relationship quality [29], how program staff promote mentoring relationship quality and youth outcomes [30,31], how mentoring affects parent well-being and family functioning [32], and how mentors help youth improve help-seeking, parent–child relationships, and participation in extracurricular activities [33]. The conceptual work on the social network perspective has also grown, with publications that now consider how this perspective may be helpful in promoting a more asset-based perspective on mentoring minoritized youth [34].

3.2.2. Essential Relational Processes

The core relational processes from Spencer’s original model continue to be studied and expanded on. Empathy, for example, has been unpacked to include both perspective-taking and adaptability, and it is associated with satisfactory and longer-lasting mentoring experiences for young people [35,36,37]. While the other relational processes continue to be valued in the literature (see [33,38] as examples), this line of work has been expanded to also consider mutuality [39] and classic forms of social support [40].

In our conversation, Renee also reflected on the working alliance as another essential precursor to positive change. The working alliance, an essential bond and alignment of goals between mentor and youth, is crucial. “You might feel like you’re connected with someone, but that you are not going in the same direction. Both are needed for therapeutic alliance, and in mentoring”, Renee stated [16]. Work that centers relational processes should thus consider the working alliance as an important addition moving forward.

3.2.3. Early Match Closure

Both the relational and systemic models of mentoring have supported new knowledge around match closure. Motivated by the Test of Time study [41], the STAR study investigated why some formal mentoring relationships terminated prematurely, collecting survey measures on empathy and other key relational processes and conducting qualitative interviews after matches ended to understand the reasons [42]. Unlike previous studies that focused primarily on qualities of the mentee that were associated with early match closure [43], the STAR study focused on the secondary relationships as represented in the systemic model. Notably, the systemic model was used as an analytic tool for the qualitative data, wherein each individual relationship as depicted in the model was analyzed [44].

3.3. Similar and Different Philosophical Orientations among Models and Their Authors

3.3.1. Alignment among the Three Models

All three authors agreed that there was alignment among the models, with each having its own emphasis. Renee’s relational model focuses on the critical features of the mentor–youth relationship. Tom’s systemic model focuses on how the mentoring relationship interacts with (influences and is influenced by) the other relationships involving the mentor, the youth, and supporting partners (parents and program staff). Jean’s mechanisms model focuses on how the mentoring relationship affects youth outcomes. While Tom’s systemic model of mentoring has the widest lens, encompassing the primary relationship and all other essential relationships to support a match, Jean’s is a slightly more targeted version. Jean’s mechanisms of mentoring model is the only one that includes how to attain a positive outcome for youth, which is “important, really important” to the literature, said Renee. She noted that her own model, the relational model of mentoring, focuses even more specifically on the processes that occur between the mentor and the youth as centrally important to the study of mentorship.

3.3.2. Continued Differences Seen among the Models: How Does Mentoring Help Kids?

The Mentor Joining the Young Person’s World

The biggest point of difference among the three models and their authors is how the authors believe mentoring works in the lives of the youth it benefits. In Tom’s systemic model and expanded social network model [9], mentoring provides opportunities for a mentor to join the world of the young person and for the mentor to expand the world of the young person. He reflected on the balance of these two points, asking in our conversation, “is the point of mentoring for the mentor to go into the child’s world and understand and support that, or is it for the mentee to go into the mentor’s world, get exposed to new things, and make new connections?” [15]. For his model, these are, collectively, the two mechanisms that drive positive change.

In our conversation, Tom went on to highlight some value- or ethics-based considerations in mentoring, particularly around confidentiality between the mentor and the mentee. “If the mentor also builds a relationship with the parent”, Tom asks, “what are the boundaries around information shared by the child?” Tom then reflected on the ties between the systemic model and family systems theories around issues like over-involvement and enmeshment. Finally, he raised questions regarding which of the many relationships in the model should be prioritized and how to keep everything balanced. All of these reflections are built off of the same value for him, that mentoring is a joining of and expansion of the youth’s social world [15].

The Mentor as a Mental Health Paraprofessional

For Jean and the mechanisms of mentoring model, mentors positively impact the lives of youth by acting as paraprofessionals providing an alternative to mental health support. She noted that mentoring is a culturally adaptive intervention, in that those who do not want a therapist for themselves or their child due to stigma or cultural relevance can get a mentor instead. Jean highlighted that models of mentoring often ask for a referring issue (e.g., depression, bullying), just like other mental health interventions, and they can be conceptualized as an alternative to professional care [18]. Indeed, young people who are referred to mentoring have double the rates of depression, anxiety, ADHD, and likelihood of living in poverty compared to young people who are not referred to mentoring [45].

For Jean, this idea that a referring issue addressed by a mentor acting as a paraprofessional is what makes effective mentoring is paramount. “If all we want are non-specific, activity-based friendships, then we should look at the new meta-analysis of after school programs by Kirsten Christensen [18]”. This meta-analysis shows the same overall effect sizes as one-on-one mentoring interventions, but they are able to achieve a much more scalable youth to adult ratio [46]. We as a field tend to pull out anecdotal success stories, she goes on to say, where non-specific friendly mentoring relationships helped one young person, while downplaying the relatively “low overall effect sizes or the fact that nearly 40% of relationships terminate early [18]”. The future of the mentoring field, in Jean’s perspective, is to think of formal mentoring programs more as an extension of paraprofessional services. That still implies building strong relationships, but it also means understanding what prompted the referral and then providing mentors with evidence-based training, strategies, and support to address it.

The Mentor as a Companion and Champion

Renee, in contrast, believes that non-specific friendship and companionship are actually the drivers of positive outcomes. Companionship came from the literatures on both social support and attachment, and it is essential to the relationship as a whole. She said that the more outcomes-focused we become in our study of mentoring, the more likely the notion of companionship will be overlooked or dropped altogether. “What kids are looking for”, Renee said, “is to have fun. In the context of that companionship, there is a lot of growth that can happen”. She continued to say that people feeling better is undervalued as an outcome over other more concrete measures.

Paraprofessional versus Companion

Both Renee and Jean, being long-time colleagues, note the direct conflict in their ideas of what drives mentoring.

“Jean does not like the friendship model of mentoring, but friendships matter a lot. When mentoring programs focus on getting the relationship right, that’s a really good investment. I’m not saying there isn’t a place for outcome driven work. We know that when people have connected relationships wherein they feel love, cared for, we know that good things happen for people. We don’t need super specified outcomes to consider mentoring effective. A connected relationship is a really important outcome. I wouldn’t change that emphasis, despite (Jean’s) compelling argument in Older and Wiser about effective sizes being too small. She is not wrong, but relationships still matter [16]”.

Jean also noted the disagreement in our conversation around the role that friendship or a working alliance plays in making mentoring impactful for young people, saying: “there has to be a working alliance; that is always going to be true. Sometimes people misunderstand me: a strong working alliance is necessary but not sufficient [18]”.

3.3.3. Continued Differences Seen among the Models: Manualization versus Individualized Support

As she writes about in Older and Wiser and discussed with us, Jean believes that the future of mentoring is about becoming more specific so that we can harness relationships in ways that bridge the unacceptable gaps in the number of people needing and actually receiving effective mental health and other services, particularly in marginalized groups [47]. This could include harnessing WISE interventions, which are short-term, effective, “psychologically precise” approaches to a range of issues that lead to strong outcomes [18]. “Because there are fewer professional boundaries, mentors may actually be better than therapists at understanding the broader context. For example, if a kid is anxious, it may have nothing to do with irregular thought processes. It may have to do with neighborhood threats on the way home from school or fighting parents, etc.” A recent meta-analysis led by Kirsten Christensen shows that relationships that are targeted to a specific referral are twice as effective [18,43]. “We could be incredibly powerful as an intervention, more than pediatric psychologists, if we embraced and trained our volunteer mentors as paraprofessionals [18]”.

While she feels strongly that mentors should be equipped as paraprofessionals with training in modalities like CBT, she is more agnostic about the future of case management in mentoring. “Case management is a tricky one”, she said. “It is a very hard job. You’re bridging a lot of complexity… I think all of those relationships matter but I don’t think we’ve identified how we can make them work better [18]”.

In alignment with his systemic approach to thinking about mentoring, Tom has a much wider view on the role of manualization in supporting matches. He noted in our conversation that in the early 2000s, there was a lot of talk about Evidence-Based Practice. It was a “hot topic”, with a lot of manualizing. He wanted the systemic model to be used in a less prescriptive way. “You can’t work from a cookbook”, he noted. He always leans towards giving workers a more conceptual approach than exact instructions on what to do next. “General principles are more helpful for workers [15]”. He continues this work of balancing helpful frameworks and theories with the autonomy that practitioners need in their work. In his ongoing annual Summer Institute, supporting practitioners by connecting them with researchers, he tries “to emphasize the theory as much as the research”. People come up with models and frameworks because there are so many judgment calls that need to be made. Models, thus, should be used as a frame of reference.

4. Future Directions for the Field

4.1. Invest in the Importance of Context

One point of highest agreement among the scholars was that context matters in the application of all three models. “Kids and their contexts are all so different”, Jean noted [18]. “Context” here ranges from the families young people are a part of, recognizing that familial capital—particularly for minoritized youth, and the larger systems with which young people access helpful relationships with adults [28,48]. Tom noted that the systemic model is being used more and more alongside a growing recognition of the importance of social context and surrounding relationships when examining a mentoring relationship [15]. When Renee was asked what, if anything, she would change about the model moving forward, she noted: “I think the basics would stay at the heart of things, but the contextual factors matter too [16]”. She goes on to note that the relational model was built on matches that had the same gender identity, but often different racial, ethnic, and class backgrounds. When considering how mentoring works in cross-race or other cross-cutting relationships, there may be an even more important emphasis on qualities, such as empathy [16].

Tom talked about what a full application of the systemic model may glean in terms of the importance of context. “I don’t think I or anyone else has really done a full systems analysis of a match [15]”. Even in using the systemic model for analyzing qualitative data, like in the STAR project [41], the authors did not actually ascertain patterns of communication (e.g., reciprocal, parallel, circular) among the individuals in the multiple dyads of the mentoring system or the degree to which two individuals work together or at cross-purposes, “all that family counseling stuff [15]”. He believes analyses like these would yield new information on the importance of context in a match.

4.2. Continue to Highlight Important Relational Processes

There are a few key qualities of a mentoring relationship that will always merit attention. Renee called for a shift away from mutuality (“a young person is not just as responsible in the relationship”) to collaboration as a clearer term for what the psychotherapy literature is talking about and calling mutuality [16]. Empathy should continue to be “at the heart of things [16]”. “If people can do that”, Renee said, “a relationship can weather a lot of mistakes, misfires, and have the opportunity to get back on track [16]”. Jean agreed that important relational processes should stay in any future iterations of the mechanisms of mentoring model, saying “I would keep some of it: there has to be a working alliance. That is always going to be true [18]”. She would, however, strive to be “more scientific” on what exactly contributes to strong working alliances and referred us to Mark Karver’s work on the subject [18,49].

Importantly, when reflecting on the outcomes-oriented model, Jean noted: “the model jumps from working alliance to development, but the first thing that has to happen is an attenuation of the specific issue that led to the referral”. Often, youth come in with a specific referring issue (e.g., depression, anxiety, friendship, academic issues, “whatever has gotten them off track” [18]). The issue of attenuation, the step between working alliance and development, as a mechanism “is very dependent on the presenting issue”. In our conversation, she gave example programs, such as for a newcomer immigrant child who is feeling isolated and “acting out” because they are struggling with a new language and culture. That is different from dysregulation that might arise from mental health issues. “If the mentor and the program are really good at understanding the context and basis for referral, then this underlying mechanism can be more effectively addressed”. The mechanisms, she noted, are embedded in family, community, and culture [18].

4.3. Focus More on Expectation Alignment

As a future direction for the mentoring literature, Renee would focus more on expectation alignment as an essential ingredient to a working alliance. She noted that this step was only apparent when studying failed relationships. In “successful ones”, as seen in her dissertation work, there was an alignment of expectations and the ability to adjust those expectations over time. This was also expanded on in a longitudinal study of mentoring relationship development [36]. Indeed, mentor pairs from different racial, ethnic, or socioeconomic backgrounds seem to have a harder time aligning expectations, due in large part to mentor bias [44]. This lack of alignment thus disproportionately impacts minoritized young people and the relationships they are a part of.

4.4. Invest in Program Infrastructure

Utilizing the systemic model helps researchers “zoom out” and consider the program that the formal mentoring relationship is supported by. Tom’s recommendation for future directions is to emphasize the importance of program infrastructure to effectively foster positive mentoring experiences through training, match support, and closure procedures. “For example”, Tom says, “the systemic model would suggest the importance of training and orientation for mentees and caregivers as well as mentors [15]”. This recommendation goes hand in hand with expectation alignment, suggesting that training and support could help staff work more effectively with participants. This investment recommendation also builds on the idea of mentors as paraprofessionals: “we need to truly professionalize the staff (adequate training, compensation, respect) before we can support paraprofessional mentors [15]”.

4.5. Be Specific about What Type of Mentoring We Are Studying

Jean noted in our conversation that “it has been a mistake that we lump formal and informal together” and that continuing to do so will make it difficult for us to effectively harness formal mentoring as a “quasi-therapeutic practice”. This conflation of the informal and formal mentoring literatures is “one reason our field has yet to invest in intervention science [18]”. Informal mentoring relationships, although not thoroughly discussed in this paper, are the product of “luck, shared interest, and circumstance” and should be recognized as a separate experience altogether [18].

Similarly, Jean calls for a separation of peer mentoring models and models that focus on adult mentors. Peers are “incredibly effective mentors”, as noted in Michael Karcher’s work on peer mentoring that yields high effect sizes [18,50]. “It is because we don’t trust peers to go off and do it on their own. We provide lots and lots of structure, which we do not for adults” [18]. Peer mentors also come with a level of credibility and relevance; when comparing seasoned psychologists to college students and volunteers in para-therapeutic roles, paraprofessionals were more effective [51]. Jean also noted a recent meta-analysis of peer mentoring, led by Samantha Burton, which yielded relatively large effect sizes, presumably because of the higher levels of training and supervision.

Lastly, she argues that as we become specific about what type of mentoring we are researching, we need to lift Youth-Initiated Mentoring (YIM) up, as a “whole other intervention needs voice” [18]. YIM is a model wherein young people select mentors from their established social networks, as opposed to being matched with a mentor they likely do not know based on like qualities or interests [52]. YIM honors the people that are already in youths’ lives. In this model, young people are “given permission” to forge relationships with those already deemed important to them, and this should continue to be a focus for the field.

5. Conclusions and Next Steps

The three foundation models of the youth mentoring field have guided 20 years of important work outlining the development of mentoring relationships and the ways in which mentoring benefits minoritized young people. While relational processes within the primary relationship and acknowledgement of the impact of secondary relationships were included in this groundwork, Rhodes’s mechanisms of mentoring model has guided the majority of the published studies on the subject. In conversation with the three foundational authors, we note ongoing differences in their philosophical orientations, particularly around how a mentor helps a young person, whether by joining the young person’s world, by acting as a paraprofessional, or by providing a supportive, growth-promoting relationship and championing the youth. This central philosophical difference is then reflected in the three theoretical models and the resulting work utilizing each.

While the models’ authors offer future directions for the field, including a continued investment in context and relational processes coupled with specificity in the types of mentoring we study, the authors note other possible directions. As a means of updating these models for youth mentoring 20 years later, issues around oppression, social justice, trauma-informed approaches, and youth-driven identity development should be central [48]. In addition, we call for a valuing of the social network of the young person, the assets a young person and their family bring to the relationship, and the many different mechanisms through which mentors can benefit young people [48]. We hope that including modern foci of this nature will center youths’ voices, youth-defined outcomes, and an overall focus on equity. We believe that the groundwork reflected on in this paper provides tall shoulders for us to stand on, and we hope to continue to develop this important line of work to directly benefit young people.