Therapeutic Mentoring for Multiply Marginalized Youth: A Grounded Theory Study of Mentors’ Adaptation During the COVID-19 Crisis

The Campus Connections (CC) therapeutic youth mentoring program provides multiply marginalized youth with access to resources (e.g., highly trained undergraduate student mentors, therapeutic interventions) to thrive. In this study, the researchers investigated the distinctive encounter that mentors and their mentees within the CC program experienced as a result of the COVID-19 pandemic. The qualitative data collected from the mentors’ weekly written reflections, and semi structured interviews were coded using Grounded Theory Method (GTM). Findings describe an adaptive process with three connecting themes of: Initial stress response of mentors, transition to online programming, and the mentorship process. The findings have important implications for youth mentoring programs using virtual platforms.


Introduction
Mentoring is an intervention where a non-parental adult and youth are paired together and form a relationship built on encouragement and support Lund et al., 2019;Rhodes et al., 2006;Sacco et al., 2014;Weiler, Boat, & Haddock, 2019). Mentees benefit most from the mentorship relationship when the mentor is supportive and encouraging rather than controlling (Rhodes, 2005;Weiler, Chesmore, et al., 2019) and the mentor serves as both an advocate and teacher for the mentee (DuBois et al., 2011). Youth mentees report many positive changes when participating in mentoring programs, such as improvement in self-worth, social acceptance, levels of intimacy, communication, trust, and academic performance (Anastasia et al., 2012;Keller & Pryce, 2012;Raposa et al., 2019;Rhodes, 2005).
Mentors also report positive outcomes when participating in youth mentoring programs, such as personal growth, interpersonal skills, academic success, civic skills, professional development, and emotional responses (Weiler et al., 2014). Additionally, mentors benefit from the dyadic relationship by gaining a new awareness of dynamics in society and an increase in self-esteem and interpersonal skills (Weiler et al., 2013). It is imperative that mentors receive adequate training prior to pairing with a mentee (Anastasia et al., 2012), and that they receive on-going guidance and support (Keller & Pryce, 2012). Training might include topics like maintaining safety, program rules, and problem-solving (Anastasia et al., 2012;Cavell et al., 2009). Many universities are now building youth mentoring programs to address the needs of "at-risk" youth (Stark et al., 2020;Weiler et al., 2013).

Therapeutic Mentoring
More specifically, therapeutic mentoring programs are a way to support youth who engage in risky behaviors, have experienced trauma, or have developmental concerns (Johnson & Pryce, 2013;Sacco et al., 2014). Traditional mentorship programs focus on relationship-building as the mechanism of change that helps mentees to work toward their goals, improve academic success, and increase social and emotional wellness (Parra et al., 2002). Some youth mentorship programs include a therapeutic element by employing mental health professionals (e.g., counselors, social workers, psychologists) to address the mental health needs of youth participants (Stark et al., 2020;Sacco et al., 2014;Weiler et al., 2013). In youth mentoring programs that include a therapeutic element, the mentor serves as a role model and advocate and works with the therapist to create a safe environment where youth can form a secure attachment to the mentor and therapeutic program (Sacco et al., 2014).

Virtual Mentoring
Virtual or electronic mentoring refers to mentorship programs that use online platforms to facilitate communication between mentor and mentee. Although the literature related to digital mentorship is limited, the significant effect of mentorship via this medium is well documented. Cultivating the mentor-mentee relationship in the virtual setting can enhance mentees' social and life skills, improve academic grades, and cultivate leadership skills (Agyemang & Haggerty, 2020;Holden et al., 2013;Shpigelman, 2013). Conducting programming via virtual media could increase mentee access to mentorship opportunities when socioeconomic and geographic barriers are present (Agyemang & Haggerty, 2020;Radlick et al., 2020). In addition, virtual mentoring is flexible and accessible for youth with differing abilities (Burgstahler & Crawford, 2007). Mentors also experience positive growth as a result of engagement in virtual mentorship, with individuals reporting increases in self-confidence, self-awareness, listening skills, communication skills, and leadership qualities (Rekha & Ganesh, 2012). Regardless of the medium through which mentorship occurs, the process includes several key components.

Campus Connections Mentoring Program
Campus Connections (CC) is a therapeutic youth mentoring program that exemplifies a sustainable approach to providing at-promise youth with mentoring services while simultaneously providing therapeutic support. This creative university-based mentoring program targets multiply marginalized youth (ages 10-18), prioritizing the inclusion of youth from communities who experience social inequity and oppression: Lesbian, Gay, Bisexual, Transgender, Queer, plus (LGBTQ+) youth; those in the juvenile justice system; and youth who experience socioeconomic disadvantage. While participating in CC, youth are paired one-on-one with highly trained undergraduate student mentors from a variety of university disciplines. Mentor-mentee pairs are assigned a counselor who is a graduate student in the master's in counseling program; the counselors provide direct therapeutic services to youth mentees and their families, and support mentors in navigating the mentor-mentee relationship. Each youth cohort meets for 12 weeks per semester, engaging in weekly 4-hour structured sessions with activities that aim to enhance connection with others and community, foster social-emotional growth, increase study skills and other academic abilities, and cultivate positive identity development (Krafchick et al., 2019).
Youth participants are described as multiply marginalized youth who can thrive when appropriate resources are accessible (Cyrus, 2017) and changes occur "in institutional structures that create and maintain inequality" (Swadener & Lubeck, 1995, p. 3). CC's youth participants are predominantly members of historically oppressed communities who continue to experience multiple intersections of oppression and discrimination. These youth are often labeled "at-risk" due to mental health symptoms or difficulty persisting in certain environments. Referencing these youth as "multiply marginalized" acknowledges that these difficulties can be attributed to systemic factors rather than individual pathology.
To address these systemic variables, CC faculty and counselors help mentors to identify areas where systemic change is needed, and support mentors' advocacy efforts to effect change in youth's school and/or community environments. In addition to these advocacy efforts, counselors facilitate discussion with youth and mentors during social justice focused activities aimed at enhancing understanding of the role of privilege and oppression in access to resources, and day-to-day experience (Krafchick et al., 2019). As part of their participation in CC, mentors are enrolled in a 3-credit hour 16-week fieldwork course. Mentors receive foundational training during this course, with community members and program faculty facilitating trainings in trauma, violence, and resiliency; culturally responsive practice; implementing reflective listening skills; and boundaries and ethics. While engaging in CC programming, mentors also complete weekly reflective journal entries. These trainings, coupled with reflective practice exercises, help mentors develop the awareness, knowledge, and skills necessary to understand when and how to best support and advocate for their youth mentees.
The therapeutic aspect of CC responds to youth participants' need for mental health support. Incorporating mental health professionals (e.g., supervised counselors-in-training) distinguishes CC from traditional mentoring programs, adding resources to positively support both mentors and mentees (Sacco et al., 2014;Stark et al., 2020;Weiler et al., 2013). The CC university-based therapeutic youth mentorship program attends to the key elements identified in Rhodes and colleagues' (2006) model of youth mentoring. We used qualitative methods to explore CC mentors' experience as they transitioned the mentorship process from in-person traditional format to the virtual setting due to COVID-19.

Current Study
We used Grounded Theory Methods (GTM) to explore the unique experience of mentors in the CC program as they adapted to challenges and changes associated with the COVID-19 pandemic. Ten mentors (n = 10) elected to participate in the study, completing one semi-structured audio-recorded interview, and providing access to their weekly written reflections (completed as part of the 3-credit hour fieldwork course associated with the CC program). Multiple methods were employed to ensure consistency, analytical rigor, and the accuracy and trustworthiness of findings.

Methods, Recruiting, & Participants
Mentors who participated in the study were enrolled in the undergraduate Fieldwork in Human Services course at a mid-sized university in the midwestern United States. In fulfillment of the requirements for the course, mentors attended each CC session, completed course assignments and activities, and served as a mentor for their assigned youth mentee for 12 weeks. A total of 22 undergraduate students were enrolled in the course and served as mentors during the Spring 2020 semester. The faculty member teaching the fieldwork course invited all mentors to participate in this study, and 10 of the 22 mentors (45%) consented to participate.
Of those mentors who consented to participate in the study, five (50%) described their ethnicity as Black/African American, Native American, Hispanic/Latino, or Asian. Although the demographic questionnaire did not inquire about mentors' sexual orientation, one mentor (10%) voluntarily disclosed during the semi-structured interview that they are queer and are a member of the LGBTQ+ community. Table 1 summarizes the demographic information for the mentors who participated in this study.
Multiple mentors described themselves as first-generation and non-traditional college students. Those students who described themselves as non-traditional shared challenges to adapting to the virtual environment due to difficulties learning to use technology and struggles to "fit in" with their peers. Data Collection Data used in this study included a brief demographic questionnaire, mentors' weekly written or video-recorded reflections, and audio-recorded individual interviews. A total of 10 participants who served as mentors in the CC therapeutic youth mentoring program during the Spring 2020 semester elected to participate in the study. Nine participants consented to have their weekly reflections from the Fieldwork in Human Services course used as data, and 7 participants completed the audio-recorded interview. Interviews were conducted during weeks 10 and 11 of CC programming (out of 12 weeks total), immediately following the conclusion of the CC session. Each participant took part in one 40-60-minute semi-structured interview facilitated by a member of the research team. We included counselor education faculty, and master's in counseling students in the final year of their degree program. All interviews used the same series of open-ended questions for each participant in order to provide consistent focal points for data collection (Charmaz, 2006).

Fall 2021
Each weekly reflection and audio-recorded interview was transcribed and stored in Word document format in a password-protected online cloud, accessible only to members of the research team. All participants selected a pseudonym at the time of the interview; this pseudonym was used to identify each participant. Data were anonymized, with no connection between participants' identifying information and their interviews or weekly reflections. A hard copy of the Institutional Review Board (IRB) approved consent form was signed prior to the beginning of CC programming and kept in a locked file cabinet in the CC program coordinator's locked office.

Data Analysis Procedures
Data analysis was completed using GTM. This approach consists of intentionally applying an inductive reasoning process through systematic review of data by multiple researchers, ultimately leading to a potential theory and/or process-based model that emerges from the data (Charmaz, 2008;Dillon, 2012;Strauss & Corbin, 2008). The implementation of GTM aligned with a social constructionist paradigm, as each of our perspectives was considered an active ingredient in the concepts and processes that emerged from the data (Charmaz, 2008). During data analysis, each of our reflexive processes was recorded in theoretical memos. Researchers use theoretical memos to record their "hard thinking about [the] relationships" between open codes, selective codes, and categories (Urquhart, 2017, p. 166).
The GTM follows a 4-stage process, beginning with open coding, progressing to axial coding, then selective coding, followed by theory formation (Dillon, 2012;Strauss & Corbin, 2008). Three members of the research team completed this 4-stage process independently, reviewing all of the data while completing theoretical memos (Urquhart, 2017) and constructing an evolving codebook (i.e., list of identified codes and their meaning; Creswell, 2013). The use of memos enabled us to constantly compare our evolving impressions and understanding of the concepts and processes emerging from the data (Corbin & Strauss, 1990). During the initial stage, we engaged in open coding by immersing ourselves in the data, reviewing each document multiple times, and recording in the margins descriptive terms or phrases that captured emerging themes. We then engaged in axial coding, reviewing the data again and noting relationships between descriptive terms or "open codes" identified during the first stage of the process. The third stage of analysis included selective coding, when we re-reviewed the data while identifying relationships between selective codes. Core categories were identified that explain the relationships between selective codes, and more concisely conceptualize the results. The final stage of the analysis process was theory formation. During theory formation, we reviewed our coded data, considered emerging processes and overarching relationships, and recorded our impressions in theoretical memos (Dillon, 2012;Strauss & Corbin, 2008).

Methods of Verification
Our independently coded documents, theoretical memos, and code books were uploaded to the secure online cloud and shared with two other members of the research team who served as peer reviewers. The peer review process was used to improve the trustworthiness of results and further develop emerging concepts and the formation of theory (Given, 2008). The peer review process and the use of more than two sources of data (i.e., reflections and individual interviews) ensured multiple triangulation (Denzin, 2012;Thurmond, 2001). Triangulation of researchers and data sources was used in this study to "add rigor, breadth complexity, richness, and depth" (Denzin, 2012, p. 82) to the understanding of the phenomenon.
Peer reviewers read all coded documents, theoretical memos, and code books, noting relationships between concepts and emerging processes or theory. They compared and contrasted codes and concepts identified by each initial reviewer and assessed the amount of overlap between initial reviewers' conceptualization of emerging themes. The overlap between codes and concepts identified during the initial review was approximately 85%, with some differences in phrasing and word choice. Peer reviewers agreed with all selective codes listed, and clarified the process emerging from the data as reflective of "adaptation" to new challenges with the aim of enhancing wellness (of self and mentee) and connecting with their mentee.

Findings Adaptation & Mentors' Initial Response
Adaptation is the primary category and process that emerged from the data. Mentors described how their experience as students, within their families, and in CC shifted due to the COVID-19 health crisis. For CC, the need for improvisation and adaptation affected all levels of programming, from the faculty teaching the CC fieldwork course, to the mentor/mentee relationship and mentorship process. Mentors also described the effect of COVID-19 on their personal lives (i.e., time spent with family and engaging in self-care), and experience as a college student. In line with the GTM, findings are illustrated using code tables, with relationships between concepts explained in narrative format (Urquhart, 2017).
Adaptation is a core category that is linked to multiple themes. The first theme, or selective code reflective of the adaptive process, was "initial stress response of mentors." Table 2 illustrates the relationship between this category, selective code, and associated open codes. Mentors' initial responses to changes associated with COVID-19 reflect a grief process characterized by shock, anxiety, sadness, and fear. These emotions arose in response to multiple losses, including loss of connection with mentee, loss of connection with friends and family, and experiences of isolation. Feelings of anxiety and fear also arose in response to fear of the unknown.

Category
Selective Code Open Codes Adaptation Initial stress response of mentors Grief and loss due to: • Loss of connection with mentee (no longer being able to see mentee in person) • Loss of connection with friends, and family (no longer being able to see friends and family in person) • Experiences of isolation Shock, anxiety, sadness, fear regarding changes Anxiety and fear regarding unknown

Adaptation & Transition to Online Programming
Adaptation remained the consistent core category that characterized each selective code. The "transition to online programming" is the second selective code that emerged from the data. This selective code includes mentors' adaptation to virtual facilitation of online programming (i.e., social justice activities, academic success time, mentor family time). Table 3 illustrates the relationship between this category, selective code, and associated open codes. Mentors described this transition as requiring them to attend to mentees' non-verbal cues in a new way, as their visual perception of their mentee was limited to their face and upper body visible in a two-dimensional computer image. Mentor BE describes this challenge as impairing their ability to "pick up small details that my mentee would appreciate." Mentors shared that the transition requires more effort and energy on their part in order to facilitate conversation with their mentee, as their mentees had new distractions in the home environment (e.g., game systems or "Xbox," television, "online videos," and family members) and were acclimating to new routines.
Additional challenges to the adaptive process included learning how to bond with their mentee in an online environment. Mentor Luke described "one-on-one time" with his mentee in the virtual breakout room as "crucial to maintaining and continuing to develop our bonds with our mentees." During the month of in-person programming prior to the COVID-19, mentors described bonding with their mentees during mealtime by "playing tic-tac-toe" and having "genuine laughs" together. While in-person, mentors also describe engaging in arts-based activities with their mentees. Mentor Private Dancer stated that she and her mentee, "sat and talked for a bit and I told her to draw a picture of herself. She drew a circle and wrote the word 'trashy' above it." Mentor Private Dancer then took the opportunity to reframe her mentee's belief about herself by "[taking] a darker marker and writing 'truly beautiful' and then finished the picture with my not so talented art skills and then we had a discussion about what her best characteristics are." Translating these in-person bonding experiences to the virtual setting required creativity.
When compared to in-person mentorship, mentors and mentees reported lower levels of trust and rapport and described the relationship as taking longer to develop. Yet, with care and time, a community can develop and be maintained online similar to face-to-face mentoring (DuBois & Karcher, 2005). In our study, multiple mentors describe using online games to interact and bond with their mentee. Mentors also described using one-on-one time in virtual breakout rooms to "catch up on how life has been." Mentors Paris and Marie shared that, although it was difficult to not see their mentees during the couple of weeks between the start of the COVID-19 pandemic and initiation of online programming, they were able to resume building connection with their mentees in the virtual setting.
While some mentors described the virtual setting as riddled with new distractions, others described their mentees as less distracted in the virtual setting. Mentor Luke described his mentee as distracted and engaging in behavior that distracted other mentees during in-person programming. He shared that this behavior decreased in the online setting: I think in some ways it was more beneficial to be able to be in a separate video chat room with our mentees, at least it was for [my mentee]. He tends to get distracted by others, mentor and mentee alike, so getting some one-on-one time with him where he doesn't have the chance to get distracted and wander off to another part of the classroom [was good]. Mentors also shared their concerns for their mentee's wellbeing, as the transition to the virtual setting enabled exposure to mentees' life at home. Mentors described witnessing arguing between mentees' household members (overhearing these altercations during teleconference CC sessions) and seeing mentees' lack of resources and unmet needs. The ethical implications associated with mentors' exposure to mentees' home environment and difficult circumstances were addressed by CC counselors and faculty. Mentors reported feeling emotionally taxed by the empathy and concern they felt for their mentees following exposure to the mentees' challenges at home.

Table 3
Adaptation & Transition to Online Programming

Category Selective Code Open Codes Adaptation Transition to online programming
Reading mentee non-verbal cues Facilitating conversation with mentee New normal, less time, and altered timeline Mentee response, distractions in environment Improved accessibility: Mentee thrived online (more than in-person) Online as a new method of connection and bonding Exposure to mentee home, reporting concerns regarding child welfare

Adaptation of the Mentorship Process
The COVID-19 crisis necessitated adaptation of the CC mentorship process to online programming. Table 4 provides an outline of the relationship between the category of adaptation, selective code mentorship process, and associated open codes. The open codes illustrate a four-part process that emerged as mentors adapted their approach to working with youth. The first phase speaks to setting the stage for the mentor-mentee relationship. The second phase describes the foundation of connection between mentors and mentees. The third phase constitutes the working stage of the relationship, and the fourth and final phase includes the termination process. Note. ACEs = adverse childhood experiences.

Phase 1: Setting the Stage
A key component of the CC program is mentors' enrollment in the Fieldwork in Human Services course. The first three class sessions were dedicated to engaging mentors in reflective practice regarding their own beliefs, biases, and training. Reflective practices included weekly reflective journals with prompts eliciting mentors' application of course material to the mentorship process and relationship with their mentee. Prior to the COVID-19 crisis, the mentors who participated in this study received foundational in-person training regarding the following topics: multicultural competency and cultural humility; responding to violence on campus (e.g., active shooter drills); mentors' responsibilities and role as mandatory reporters; understanding of mental health issues; basic counseling skills; and the effects of adverse childhood experiences. Mentors then used the self-awareness, knowledge, and skills to establish connection and develop a working relationship with their mentees.

Phase 2: The Foundation
Connection is the core theme uniting the open codes associated with Phase 2 of the mentorship process when adapting to changes associated with the COVID-19 crisis. Mentor BE described connection in CC as, "the foundation of the relationship between an adult and individual to allow guidance and situational awareness and to allow the mentee to develop and grow in an environment [where] they feel safe." In line with Rhodes and colleagues' (2006) model of youth mentoring, these findings confirm the need for mutuality, trust, and empathy as the foundation for a strong connection in the mentoring relationship. Mentors shared their use of humor and kindness to cultivate connection with their mentee in the virtual setting. They also reported employing their basic reflective listening skills, acquired during the Fieldwork course, to develop rapport with their mentee. The connection between mentors was also important, as Mentor Paris Northwood described the relationship between mentors in mentor "families" (family units within CC comprised of multiple mentor-mentee pairs): "as a family we've grown, and I could go to them about anything regarding Campus Connections or something personal. I feel like a lot of [them] are super kind and caring." The connection between mentor and mentee and among mentors in their mentor family units was the foundation for the therapeutic element of CC programming.

Phase 3: The Working Phase
The working phase of the mentorship process included engaging in activities related to social justice themes and academic success. While participating in these activities, mentors continued to use their reflective listening skills and knowledge in the virtual setting. Social justice activities focused on developing mentee awareness and understanding of how their experience is affected by society's response to their intersecting identities (e.g., ethnicity, ability status, gender identity, sexual orientation, religion or spirituality, etc.). A portion of each CC session was devoted to academic success. In the virtual setting, mentors identified mentees' academic goals, explored challenges to achieving these goals, and created a concrete plan of action. Prominent focus areas for academic success included supporting mentees' ability to focus during online learning, access devices necessary for online learning, and mentee attendance in online classes.
Mentors described consistent attention to mentee strengths to help youth cope with the challenges and stressors associated with the COVID-19 crisis. Mentor Private Dancer shared an interaction with their mentee's parent that reinforced the positive impact of affirming their mentee's strengths: [My mentee's mother] said that she loved how encouraging I was to [my mentee], and that I really helped her to believe in herself. She knew she was smart, but she kept saying things [about herself] that would bring her down. I think those negative thoughts were making her feel less than what she actually was. Mentors also describe a transformation of their self-perception over the course of CC. Mentors described personal and academic growth during CC both prior to and during the COVID-19 crisis. Open codes emerging from the data that speak to this growth process included mentors learning to attend to physical wellbeing with at-home activities (e.g., "yoga," "exercise"); social self-care via teleconference platforms (e.g., "Facetime" and virtual "girls' group"); and setting aside quiet time for learning to use technology and access online course materials. Mentors also shared their adaptive process in response to changes in the college experience. They articulated feelings of "isolation," the need for a "community" of fellow students, and their desire for a routine associated with study time and class-related activities. Mentors also described growth in self-awareness regarding tendencies to respond to youth in certain ways, and intentional use of reflective listening skills. Multiple mentors planned to pursue careers in the helping professions, and all reported planning to use listening and responding skills learned in CC in their future careers. Multiple mentors described learning the importance of advocacy for youth, particularly those from disadvantaged communities, and their plans to pursue careers that would enable them to advocate for youth.

Phase 4: Termination
The termination phase included the required components for the Fieldwork in Human Services course and the interactions between mentor and mentee as they said goodbye. Mentors completed a letter to their mentee following prompts provided by the course instructor. These letters were then provided to each mentee. During graduation night (the final session of CC) mentors and mentees engaged in "shout outs" to one another, sharing their final thoughts and feelings about their experience in CC and relationship with their mentor/mentee. During the interviews for this study, mentors summarized their experience in CC; Mentor BE stated "[Being] a mentor has changed my experience. I realize that a person can affect another individual's life in a greater way than we believe, even in a short time."

Discussion
The primary objective of the study was to understand the unique phenomenon experienced by the mentors and mentees within the CC program as a result of the COVID-19 pandemic. Results and emerging themes highlight the importance of adaptation as a consistent driver of change throughout the experience. In line with current research, the transition to conducting mentorship activities in the virtual setting is associated with several challenges, from difficulty connecting emotionally in virtual environments, to limited time and space for private discussions with mentees (Ettekal & Agans, 2020). These challenges required mentors to adapt their approach, learning new ways to connect with their mentee, while also working through their own emotional response to the COVID-19 crisis.
An important consideration, based on the findings, relates to preparation to deal with ethical concerns. As a result of shifting to virtual formats, mentors were exposed to different situations than originally expected, which sparked discussions related to ethics. Virtual methods allowed mentors to observe mentee home environments. This initiated discussions in supervision with mentors regarding welfare and safety of mentees. Although these observations did not imply ethical violations, they increased the complexity of the experience and allowed for discussions of these topics to ensure the most effective provisions of services. This also illustrated the need for support, regardless of format; initial training in ethics and best practice guidelines might be insufficient if solely geared toward in-person services. Organizations providing training and support for mentors might benefit from incorporating virtual technologies to facilitate ethical decision-making and review as well as an adaptive perspective on applying ethical decision-making models.
Another factor addressed through the findings of the study related to feeling confident regarding the interpretation of verbal and nonverbal expression via virtual formats and the development and maintenance of the relational bond between mentor and mentee. Relationship development can be complicated in virtual formats and, although different strategies might be required to develop the mentor-mentee relationship, it can be done in ways that benefit both (Abdelhamid et al., 2020). Although this study's participants had the benefit of initial relationship-building in an in-person environment, it is unknown how the relationships would be different if initiated solely through virtual formats. For example, participants reported they would prefer more time and opportunities to build and enhance mentoring relationships.
In this study, initial stages of shock, anxiety, sadness, and grief were replaced by new forms of interaction, which allowed for contrast and comparison with earlier interactions.
These changes forced both mentor and mentee to revisit and respond with new adaptive strategies for connection and relational engagement. These adaptations came with the identification of potential concerns and challenges that required ongoing support for both mentors and mentees. Future research in virtual mentoring should target longitudinal results of mentoring in a virtual setting compared to in-person. Additionally, researchers should explore how the needs of mentees are met in a virtual setting compared to in-person to guide improvements for virtual mentorship.

Limitations
The limited sample size (n = 10) and data sources examined for this study narrow the generalizability of findings. Although the findings cannot be said to reflect a pattern of experience among all mentors engaging in virtual mentoring, the reader is encouraged to consider the applicability of results based upon their own experience (Shenton, 2004) and the context of this study in relation to existing literature.

Implications for Practice
An implication for practice is that training in adaptability and resiliency could help facilitate a thoughtful perspective and the ability to cope and adapt well when unexpected circumstances occur. For example, in the current study, the COVID-19 pandemic caused a mandatory transition from in-person to virtual service provision. Through this study, it was found that mentor-mentee relationships can have transformative properties, both in-person and via virtual environments. Further, the shared experience of responding to unforeseen circumstances can enhance the mentor-mentee relationship.
Other similar mentorship programs might benefit from reviewing their initial training practices and contingency planning to allow space for adaptive responses to novel circumstances. It is crucial to have the needed resources, time, and energy to provide ongoing support to participants. Programs might benefit from normalizing and validating participant experiences of transition as well as brainstorming possible scenarios they might encounter in in-person and/or virtual environments. For example, procedures and the effects of reporting incidents of abuse and neglect might differ based on providing services in person versus through virtual methods.
Other similar mentorship programs might benefit from tracking participant responses and adaption over time. In the current study, participant responses focused mainly on the initial effect and immediate experiences with transition early in the COVID-19 pandemic. Participant perspectives are likely to change or become more adaptive over time due to the duration from the initial point of transition or stressor.