Mentors’ Beliefs About Their Roles in Health Care Education: A Qualitative Study of Mentors’ Personal Interpretative Framework : Academic Medicine

Secondary Logo

Journal Logo

Research Reports

Mentors’ Beliefs About Their Roles in Health Care Education: A Qualitative Study of Mentors’ Personal Interpretative Framework

Loosveld, Lianne M. MSc; Van Gerven, Pascal W.M. PhD; Vanassche, Eline PhD; Driessen, Erik W. PhD

Author Information
Academic Medicine 95(10):p 1600-1606, October 2020. | DOI: 10.1097/ACM.0000000000003159


Having a committed mentor is valuable to students in health professions education. Not only can mentors promote mentees’ academic development, performance, satisfaction, and success, but they also can help them cope with the conflicting demands of career development and private life.1–6 How mentors shape their practice is influenced by their personal beliefs about the goals and purposes of mentoring, what valuable mentoring activities to engage in, who should decide on the focus of the mentoring relationship (mentor or mentee), and which strategies and methods should be used to enact these beliefs. Given the idiosyncratic and contextualized nature of these beliefs,6 mentors have their own unique ways to interpret and enact their mentoring.2,7–9 Stimulating mentors to make explicit their beliefs and critically examine them can raise awareness of why mentors act as they do in particular situations and, if necessary, allow them to regulate these beliefs to engage in more suitable mentoring. Because there is a paucity of studies of the personal beliefs of mentors in health professions education,6 we know little about the content of these beliefs and how they bear on the way mentors develop and enact their approaches to mentoring. Inspired by Kelchtermans’ “Who I am in how I teach is the message,”10 we set out to identify how mentors’ “self”11 shows in their mentoring. More specifically, this research aims to contribute to the development of theory on mentoring and mentoring beliefs,6 thereby taking into account the complexity of the work as well as its idiosyncratic and contextualized nature.

Becoming a mentor is not just a natural extension of being an experienced teacher. It requires specific knowledge and skills.12 Consequently, expertise acquired in lecturing or clinical teaching, for instance, does not automatically transfer to one’s work as a mentor. Learning how to be a mentor therefore requires faculty development initiatives that specifically target the mentoring role. This approach resonates with the growing research base underscoring the need for faculty development initiatives aimed at acquiring, maintaining, and improving mentoring skills.1,3,5,12–20

This need for faculty development initiatives, combined with the notion that how mentors mentor is based on individual beliefs, strongly suggests that training mentors cannot go without identifying and tapping into those beliefs.10 Indeed, researchers have argued that “professional development should not only concern instrumental knowledge, but also involve one’s normative assumptions … as these are enacted in practice.”21 Hence, we should ask mentors to reflect on their personal mentoring practice because doing so will serve a dual purpose. First, it will make mentors aware of their beliefs and allows for adjustment if needed. The mere act of thinking about what they do, how they do this, and why they do it this way may itself foster development.11,14,15 Second, it may allow training to be personalized because beliefs can be incorporated in mentor training programs, for example, through role-plays, individualized coaching, or by using cases from the mentors’ workplace.21

This study is specifically concerned with identifying mentors’ beliefs. To operationalize mentors’ beliefs, we used Kelchtermans’ concept of the personal interpretative framework.10 Initially, based in grounded theory,22 this conceptual framework was developed as a set of sensitizing concepts that could be used to reconstruct and analyze the professional development of teachers through their biographical career stories. The initial framework was developed based on an exploration of the literature.23 Although originally constructed from research on the professional development of experienced primary teachers in the early 1990s, Kelchtermans’ notion has recently been taken up successfully in research on the identities and practices of other educational groups, including beginning teachers, teacher–educators, and educational leaders.23,24 The personal interpretative framework acts as a lens through which teachers make sense of and act in the professional situations in which they find themselves.10,24 Rather than being static, the content of the personal interpretative framework is dynamic and results from the meaningful interactions between the individual teacher and his or her professional working context (including the curriculum, collegial environment, student population, available resources, and faculty development initiatives).

Figure 1 illustrates the personal interpretative framework in context. Its 2 main dimensions are professional self-understanding (PSU) and subjective educational theory (SET).23 The first dimension, PSU, refers to how teachers conceive of themselves in their role and comprises 5 components: self-image, self-esteem, task perception, job motivation, and future perspective. The first component, self-image, is descriptive; it denotes how teachers typify themselves in their teaching role. Although largely based on self-perception, it also reflects and incorporates feedback from others (including teachers’ students and colleagues). The second component, self-esteem, is evaluative, revealing how teachers evaluate their practice by answering questions such as how well do I do? The third component, task perception, mirrors what teachers feel they should or should not do as teachers. Fourth, job motivation is a conative element, which articulates the reasons teachers choose to enter, continue, or leave their profession. Future perspective, finally, refers to teachers’ future expectations of their job.

Figure 1:
The personal interpretative framework, from a study to explore mentors’ beliefs about their role in health care education. Eighteen faculty who mentor undergraduate students in the Medicine, Biomedical Sciences, and Health Sciences programs at Maastricht University, Maastricht, the Netherlands, participated in semistructured interviews between December 2017 and January 2018, to reconstruct their personal interpretative frameworks. The framework is a continuous interaction between mentors and their professional working context. It consists of 2 dimensions: professional self-understanding (PSU) and subjective educational theory (SET), which mutually interact, as indicated by the double-headed arrows. Both PSU and SET consist of multiple components, respectively describing the what and how of mentoring.

The second dimension of the framework, SET, encompasses teachers’ individual system of knowledge and beliefs about teaching that informs their decisions on how to act in particular situations. Just like PSU, SET is intrinsically personal as teachers ponder, what works best in this particular situation? and why would this work? Teachers derive their SET from a combination of personal experiences, hearsay, and best practices learned from colleagues, as well as from more formal sources such as research or professional development support.10

For professional development of mentors, it is important that mentors reflect on their beliefs about mentoring. Professional development by mentors cannot solely be aimed at changing knowledge and skills25; without tapping into mentors’ individual beliefs, the process would not yield durable professional development. Hence, to improve mentoring, we must not only teach mentors how to mentor but also help them to gain insight into their individual mentoring beliefs, so that they can adopt desirable mentoring characteristics2,5,6,26 in the mentor–mentee relationship. To enable them to gain this insight, in this study, we reconstructed mentors’ mentoring beliefs through the personal interpretative framework.



Using a multiple-case study approach,27 we conducted semistructured interviews with mentors to explore the content of their personal interpretative framework. A “case” refers to a single mentor in interaction in his or her professional working context (see Table 1). Professional working context could, for example, be interactions with colleagues, conversations with mentees, faculty development initiatives, or self-directed initiatives to learn more about mentoring. We chose to define our cases as an interaction of mentor and professional working context because it aligns with our theoretical frameworks’ notion of the personal interpretative framework being in constant interaction with the current professional working context.

Table 1 - Mentoring Characteristics of 3 Educational Programs at the Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, the Netherlandsa
Program Mentoring as integral part of curriculum Duration of mentoring program, years Mentees per mentor per academic year, no. Use of development portfolio Assessment by mentor Mentors in sample, no.
Medicine Yes 3 5 Yes Longitudinal competency assessment via portfolio 4
Biomedical Sciences Yes 3 9–16 Yes Longitudinal competency assessment via portfolio 12
Health Sciences No 1 9–13 No Assessment of academic writing 2
aEighteen faculty who mentor undergraduate students in the Medicine, Biomedical Sciences, and Health Sciences programs were interviewed between December 2017 and January 2018 in a study to explore mentors’ personal interpretive framework.


Given the exploratory nature of this study, we sought to maximize relevant heterogeneity in participants’ career trajectories and professional working contexts. We purposefully selected28,29 18 mentors from Maastricht University in Maastricht, the Netherlands. The mentors were selected from 3 undergraduate programs, specifically Medicine, Biomedical Sciences, and Health Sciences, that clearly differ in the extent to which mentoring forms an integral part of the curriculum, the duration of the mentoring program, the number of mentees assigned to each mentor, the use of portfolios to support mentoring, and the type of assessment by the mentor (see Table 1). Mentors were faculty at Maastricht University providing mentorship to undergraduate students. Mentors engage in preparatory mentor skills training and attend hands-on information sessions throughout the year. They are expected to meet with their mentees 3 to 5 times per year, on both an individual basis and during group sessions. The individual meetings between mentors and mentees are usually centered around personal and professional development goals of the mentees. In both the Medicine and Biomedical Sciences programs, mentoring is a longitudinal part of the study program, integrated in programmatic assessment and supported by e-portfolios.30,31 At the time of the interviews, all participants had completed at least 1 academic year of mentoring. Consequently, they had all completed preparatory mentor training and had gone through at least 1 complete cycle of mentoring, so they had experiences and examples to reflect on during the interviews.


We recruited participants via email, which led to a sample consisting of 4 mentors from the Medicine program, 12 from the Biomedical Sciences program, and 2 from the Health Sciences program. Twelve of our mentors were female, 5 were male, and 1 did not wish to reveal gender and age. The mean age of the mentors we interviewed was 48.9 years (SD = 7.6).

Data collection

The interview guide (Supplemental Digital Appendix 1 at was based on earlier research32 on the personal interpretative framework and built on the framework in that it covered all its components with a particular focus on mentoring. Open-ended questions were included to encourage deep reflection and storytelling. Before we conducted the interviews with the selected mentors, we tested the guide.23,32,33 L.M.L. interviewed the participants and, in consultation with the team, iteratively adapted the interview probes during the interview period. The face-to-face interviews were conducted in December 2017 and January 2018 and lasted approximately 1 hour. All interviews were audiorecorded, transcribed verbatim, and anonymized before further analysis.

Data analysis

Data were analyzed in a 2-step procedure. First, we did a within-case analysis34 to understand the uniqueness of each case. Second, we completed a cross-case analysis to build a general pattern of explanation and identify differences and commonalities across cases, that is, patterns in mentors’ understandings of their role as a mentor and approaches to the practice of mentoring.35,36

We started the within-case analysis by coding the transcripts. An initial coding scheme37 based on the theoretical framework was complemented with descriptive codes summarizing the content of specific fragments of data. To make sure the meaning of codes was clear and unambiguous, P.W.M.V.G. double-coded 2 of the interview transcripts that were coded by L.M.L. Overlap and differences in the interpretation of codes were discussed between L.M.L. and P.W.M.V.G. until a satisfactory level of understanding of, and distinction between, codes was reached. Based on the coded transcripts that ensued, we were able to compile a synthesis text, (i.e., a biographical case report) that shared a common structure reflecting the theoretical framework. (For the case report template, see Supplemental Digital Appendix 2 at The case reports contained extensive quotes from the interviews (thick descriptions38), thus preserving the richness of the data while at the same time making the large amount of data more manageable for the cross-case analysis.

We started the cross-case analysis34,39 by comparing the biographical case reports. Reports were compared on their summaries, characteristic statements, and illustrative quotes and keywords. Based on the PSU and SET content of these phrases, we grouped reports in different ways until distinctive and meaningful clusters of PSU and SET types could be distinguished. We searched our data for as many different types as possible, but only if they meaningfully differed from each other. In addition, we ensured that we clustered in such a way that the PSU and SET of each mentor fit within 1 of the clustered types. Several rounds of close reading and clustering the biographical profiles resulted in the identification and description of 3 types of PSU and 3 types of SET, which combined into 4 distinct mentoring positions. The 4 positions were subsequently described in detail and extensively discussed by the research team. After completing this process, we read all the interviews once more and checked them for full consistency with the positions identified. To confirm that we interpreted the mentors’ beliefs correctly, we invited all participating mentors to join a session in which L.M.L. explained the full study, described the results, and discussed the meaning of the results with the participants.

We used ATLAS.ti Version 8.2 (Scientific Software Development GmbH, Berlin, Germany) and Microsoft Excel 2016 (Microsoft Corporation, Redmond, Washington) to manage data throughout the analysis.

Researcher reflexivity

Two members of the research team (P.W.M.V.G. and L.M.L.) are mentors in one of the undergraduate programs, and L.M.L. is a mentor trainer as well. This affiliation may have influenced the researcher–participant relationship, as well as the data collection and analysis. In recognition of the relationships of L.M.L. and P.W.M.V.G. to the research, the researched, and the research setting, every step in the process, as well as intermediate findings were thoroughly discussed with the other team members, who had a more distant relationship with participants and did not hold mentor roles in the targeted study programs. Yet as researchers who are knowledgeable in the field, they may have been better able to ask relevant questions and to follow up during the interviews and encourage mentors to name specific examples to illustrate their answers. The academic background of the research team members was diverse, albeit complementary: educational sciences (L.M.L. and E.W.D.), cognitive psychology (P.W.M.V.G.), and pedagogical sciences (E.V.). This diversity contributed to a rich, diligent, and multiperspective interpretation of data but also safeguarded against blind spots.

Ethical approval

Ethical approval was granted by the Ethical Review Board of the Netherlands Association for Medical Education (NVMO-ERB reference number 944).


From our within-case and cross-case analyses, we distilled 3 types of PSU and 3 types of SET, which, when combined, constituted the following 4 mentoring positions, which are summarized in Table 2: (1) facilitator, a service-providing, responsive mentor; (2) coach, a development-supporting, responsive mentor; (3) monitor, a signaling, collaborative mentor; and (4) exemplar, a service-providing or development-supporting, directive mentor.

Table 2 - Mentoring Positions and Mentors Holding These Positions, From a Study to Explore Mentors’ Personal Interpretative Framework, Maastricht University, Maastricht, the Netherlands, December 2017–January 2018a
Mentoring position Dimensions of personal interpretive framework Mentors in sample, no.
Professional self-understanding Subjective educational theory
Facilitator Service providing Responsive 5
Coach Development supporting Responsive 3
Monitor Signaling Collaborative 6
Exemplar Service providing, development supporting, or both Directive 4
aEighteen faculty who mentor undergraduate students in the Medicine, Biomedical Sciences, and Health Sciences programs participated in semistructured interviews between December 2017 and January 2018.

These positions should not be regarded as fixed and stable innate traits but rather as reflecting the different conceptualizations and combinations of mentors’ PSU and SET, as well as different conceptualizations of mentors’ working relationships with mentees, which may dynamically vary across contexts and over time. Therefore, mentors will probably not only display PSU and SET types from their own position but also make use of the other positions, depending on the context.

In the next paragraphs, we will describe each position in terms of the 2 dimensions of the professional interpretative framework, illustrated with quotes from the mentors who participated in the study (M).

Position 1: The facilitator

A facilitating mentor is an approachable first point of entry, a familiar face, whose PSU is centered around providing the mentee a certain level of “service.” This service includes, for example, explaining to the mentee how the academic world works and how to self-direct one’s study, or putting the mentee in touch with others, such as study counselors, the board of examiners, or psychologists. Facilitating mentors see themselves as sources of information, being a stable, ever-present factor during their mentee’s study career:

I always interpreted mentoring as a point of contact students have, something very close. When they enter, everything is new, there are many students, [I can be] the one person they regularly see throughout the year, whom they can always ask questions. Even simple questions, or questions they don’t know whom to ask, they can ask me. (M11)

Although facilitating mentors acknowledge the benefit of using a portfolio for documentation purposes, they tend not to use it to guide their interactions with mentees. Therefore, they will work with the portfolio insofar as it is required by the study program but will not use it as a yardstick for determining the mentee’s study progress.

The SET of facilitating mentors is responsive, meaning that they do not proactively intervene in their mentees’ development but leave it to mentees’ initiative when it comes to getting in touch, asking questions, or making requests. Mentors with a responsive SET can be seen as “available on call” when mentees need them: “I think I’m accessible, but I also think I’m not a difficult mentor, one who asks difficult questions. I much prefer being there for my student at the moment they need me.” (M12)

A mentor with a responsive SET is there for the mentee when needed, stays on top of things, and keeps track of the mentee’s academic progress.

Position 2: The coach

The primary aim of the coaching mentor is to nurture a development-supporting PSU, looking at mentees as unique individuals, and not simply as students or future professionals: “… students should be who they are and accept who they are.” (M04) More specifically, this type of mentor encourages mentees to think about their development in the broadest sense, not only in terms of academic achievements but also with respect to meeting the challenges and demands of one’s personal and future professional life, thereby supporting the development and well-being of their mentees. Because their PSU is development oriented, coaching mentors rely more on the information in the mentee portfolios than facilitating mentors do; portfolios allow them to see their mentees’ learning goals and reflections, see their assessment scores and feedback, and check whether they need additional support. As with the facilitating mentor, the coaching mentor allows the mentee to actively steer and deploy the mentoring relationship toward growth and development, resulting in a responsive SET, where the mentee is in the lead: “I keep pushing, but I don’t pull…. I cannot drag my student into something and hope they follow along.” (M04)

Position 3: The monitor

A monitoring mentor holds a signaling PSU, which means that the monitor can provide suggestions or advice and may deploy his or her own network to the benefit of mentees’ development. Given this signaling goal, a monitoring mentor uses the mentee’s portfolio as an instrument to follow up on how the mentee is doing and to signal potential downward or upward trends in his or her development. Monitoring mentors can then point this out to their mentees:

I feel it’s my role to tell students, “This is not a clever thing to do, you need to make choices here.” But it is not my task to tell students which choice to make…. I can point students to issues they create and ask, “Why do you think so? What is more important to you? You are the one making the decisions here.” (M23)

A monitoring mentor aims to assist the mentee in becoming a reflective learner and to support self-understanding by mirroring the mentee’s behavior and supporting structured, deep reflection.40 According to their own description, a monitor is a critical friend who is in a mutually respectful relationship with the mentee, in which the mentee is often regarded as an equal partner rather than a junior or novice.

The collaborative SET of a monitor entails that, depending on mentees’ needs in the varying stages of their study trajectory, the mentor flexibly adapts the level of support and involvement, which translates into a collaborative relationship with mentees:

Some sort of “let them swim and we’ll stand there holding a lifebuoy.” I think that is better than … just standing there, holding a rescue hook all the time, saying “come over here.”…. In the end, students need to do it. If they want to make use of [mentoring], OK. If not, that’s OK too. (M20)

Position 4: The exemplar

An exemplar usually does not have a clear preference for either a service-providing or a development-supporting PSU but can engage in both. This mentor may normatively use personal experiences when advising a mentee and may have an initiating role in discussing concerns about the mentee’s functioning. Consequently, the exemplar is more directive in his SET compared with the responsive mentors (facilitator and coach) and collaborative mentor (monitor). Rather than “solving” an issue for the mentee, the exemplar encourages the mentees to try to solve issues themselves or refers them to dedicated professionals. He or she does follow up on such referrals but keeps a professional distance. A mentor who acts like an exemplar can be stern toward a mentee who does not meet expectations and can decide to take action on that as well:

Well, I sent one of them off to the student counselor; he was asleep during our meetings. And why? Going out ’til 5 o’clock on Saturday, on Sunday. You don’t do that in my group, buddy. So I told him, “I think you’re overestimating yourself.” So I called the student counselor and managed to get him there. (M05)

Conversely, when mentees do their best or show interest in the professional field of their mentors, they can learn a lot from the personal expertise of these mentors who position themselves as exemplars. They then, for example, take the time to show mentees what they do as academics: “Some [students] are already performing quite well, right? Five minutes in a conversation I thought, ‘Well that’s that, nothing to add anymore, everything is going quite fine.’ And then I took him to see our research facilities.” (M10)


This study aimed to reconstruct mentors’ personal interpretative framework23 on the basis of in-depth interviews with 18 mentors from 3 undergraduate programs. We distilled 4 mentoring positions from our analysis: the facilitator, the coach, the monitor, and the exemplar. Each of these dynamic positions represent a coherent set of normative beliefs about what are meaningful mentoring activities to engage in, who should decide on the focus of the mentoring activity (mentor or mentee), and which strategies and methods should be used to enact these beliefs. Both the facilitator and the coach adopt a responsive approach (SET) to mentoring but differ in terms of the activities they engage in with their mentees. While facilitators aim for a service-providing PSU, coaches tend to focus on development support. Monitors, however, interact with their students on a more collaborative level and help them to recognize and track their progress. Exemplars, finally, have no clear preference for either service provision or development support, and nurture a more directive mentoring SET.

The results of our analysis can be used to support professional development of mentors. Mentors can be asked to analyze and reflect on their mentoring and the context in which it takes place. This analysis and reflection makes them more aware of the beliefs that are the foundation of their mentoring, which can contribute to more durable changes in the professional development of mentors.

It is important to note that the 4 mentoring positions are not intended to be a prescription for what mentoring practice should ideally look like, since mentoring inherently is contextualized and dynamic.6,41,42 Rather, the goal of this research was to provide an interpretative description of the diverse ways in which mentors think about their goals and practices. Implicit in the notion of the personal interpretative framework is the idiosyncratic, yet deeply contextualized and dynamic sense of self mentors bring with them to particular situations. The position mentors prefer, embodying their set of cognitions, guides the way they interpret a particular situation or context and how they act accordingly. At the same time, however, these positions are modified by their interaction with that context, for example, by the type of mentee, formal or informal mentoring, and the requirement to assess students or not. Consequently, positions should not be seen as stable, innate traits, but differing over time and between contexts. When mentors become aware of their current position, they can reflect on it and decide whether this position is still appropriate2,5,6,26 for interacting with their mentee. Mentors can make use of multiple positions instead of rigidly holding on to a certain role. Accordingly, mentors could decide to strengthen their position, which could be supported by, for example, individual coaching or tailored faculty development.

Given the contextualized and dynamic qualities of these positions,6,10,24,41,42 we welcome future research that extends this study to include mentors working in vastly different mentoring programs and settings (e.g., programs with a strong portfolio-based assessment, informal mentoring, programs without preparatory mentor skills training) or mentors at different stages in their mentoring career. In an informal mentoring setting where mentees independently initiate mentoring and only contact their mentor when they feel they need to, for instance, we might expect mentors to adopt a signaling-responsive mentoring position.

In this study, the personal interpretative framework10 offered us a refined analytical tool to reconstruct and visualize the complex reality of mentoring beliefs. The theoretical framework was initially developed for research in primary education, but it has been applied to other fields as well.23,24 We used interviews for reconstructing mentors’ personal interpretative framework. Interviews are time intensive. Therefore, future research could explore more time-efficient methods, for example, surveys, which can be used to reconstruct mentors’ personal interpretative framework.

A limitation to the present study in general is the composition of the sample. For practical reasons, that is, limited availability of mentors in the 3 programs, we interviewed mentors who were relatively new to mentoring (< 1.5 years of experience). Because PSU and SET are always in development, mentors’ personal interpretative framework is bound to change over time. For that reason, it would be interesting to explore whether mentors with more prolonged mentoring experience hold positions that are distinct from their less seasoned counterparts.


Although mentors’ positions are inherently dynamic and context specific, being aware of their position can help mentors to understand why they act the way they do in certain situations and how their behavior affects their mentee’s development. It can also help mentors to identify personal learning needs and, consequently, provide opportunities for faculty development. The present research adds to the theoretical knowledge about mentoring in health professions education and can give rise to future innovations in faculty development targeting mentors.3,43 Mentors who acquire and practice their mentoring skills through training and are reflectively aware of how and why they mentor can be highly valuable facilitators, coaches, monitors, or exemplars to their students.


The authors thank all participating mentors for their invaluable contributions to this study. The authors are also grateful to Angelique van den Heuvel for her help in preparing this manuscript and to Stephanie Meeuwissen for sharing her valuable insights on mentoring.


1. Driessen EW, Overeem K, van der Vleuten CP. Get yourself a mentor. Med Educ. 2011;45:438–439.
2. Sambunjak D, Straus SE, Marusić A. Mentoring in academic medicine: A systematic review. JAMA. 2006;296:1103–1115.
3. Kashiwagi DT, Varkey P, Cook DA. Mentoring programs for physicians in academic medicine: A systematic review. Acad Med. 2013;88:1029–1037.
4. Atreya AR, Stefan M, Friderici JL, Kleppel R, Fitzgerald J, Rothberg MB. Characteristics of successful internal medicine resident research projects: Predictors of journal publication versus abstract presentation. Acad Med. 2018;93:1182–1188.
5. Straus SE, Chatur F, Taylor M. Issues in the mentor–mentee relationship in academic medicine: A qualitative study. Acad Med. 2009;84:135–139.
6. Sambunjak D, Straus SE, Marusic A. A systematic review of qualitative research on the meaning and characteristics of mentoring in academic medicine. J Gen Intern Med. 2010;25:72–78.
7. Driessen EW, Overeem K. Walsh K. Mentoring. In: Oxford Textbook of Medical Education. 2013.Oxford, UK: Oxford University Press;
8. Heeneman S, de Grave W. Development and initial validation of a dual-purpose questionnaire capturing mentors’ and mentees’ perceptions and expectations of the mentoring process. BMC Med Educ. 2019;19:133.
9. Sng JH, Pei Y, Toh YP, Peh TY, Neo SH, Krishna LKR. Mentoring relationships between senior physicians and junior doctors and/or medical students: A thematic review. Med Teach. 2017;39:866–875.
10. Kelchtermans G. Who I am in how I teach is the message: Self-understanding, vulnerability and reflection. Teach Teach. 2009;15:257–272.
11. Balmer DF, Darden A, Chandran L, D’Alessandro D, Gusic ME. How mentor identity evolves: Findings from a 10-year follow-up study of a national professional development program. Acad Med. 2018;93:1085–1090.
12. Ramani S, Gruppen L, Kachur EK. Twelve tips for developing effective mentors. Med Teach. 2006;28:404–408.
13. Pfund C, Maidl Pribbenow C, Branchaw J, Miller Lauffer S, Handelsman J. Professional skills. The merits of training mentors. Science. 2006;311:473–474.
14. Aspfors J, Fransson G. Research on mentor education for mentors of newly qualified teachers: A qualitative meta-synthesis. Teach Teach Educ. 2015;48:75–86.
15. Michael O. Mentoring mentors as a tool for personal and professional empowerment in teacher education. Int J Evidence Based Coach Ment. 2008;6:1–18.
16. Gandhi M, Johnson M. Creating more effective mentors: Mentoring the mentor. AIDS Behav. 2016;20(suppl 2):294–303.
17. Sood A, Tigges B, Helitzer D. Mentoring early-career faculty researchers is important—But first “train the trainer.” Acad Med. 2016;91:1598–1600.
18. Tan YS, Teo SWA, Pei Y, et al. A framework for mentoring of medical students: Thematic analysis of mentoring programmes between 2000 and 2015. Adv Health Sci Educ Theory Pract. 2018;23:671–697.
19. Athanases SZ, Abrams J, Jack G, et al. Curriculum for mentor development: Problems and promise in the work of new teacher induction leaders. J Curric Stud. 2008;40:743–770.
20. Johnson MO, Subak LL, Brown JS, Lee KA, Feldman MD. An innovative program to train health sciences researchers to be effective clinical and translational-research mentors. Acad Med. 2010;85:484–489.
21. Vanassche E, Kelchtermans G. Facilitating self-study of teacher education practices: Toward a pedagogy of teacher educator professional development. Prof Dev Educ. 2016;42:100–122.
22. Glaser A, Strauss AL. The Discovery of Grounded Theory: Strategies for Qualitative Research. 1967.New Brunswick, NJ: AldineTransaction;
23. Kelchtermans G. Getting the story, understanding the lives: From career stories to teachers’ professional development. Teach Teach Educ. 1993;9:443–456.
24. Vanassche E, Kelchtermans G. A narrative analysis of a teacher educator’s professional learning journey. Eur J Teach Educ. 2016;39:355–367.
25. Steinert Y, O’Sullivan PS, Irby DM. Strengthening teachers’ professional identities through faculty development. Acad Med. 2019;94:963–968.
26. Johnson WB. On Being a Mentor: A Guide for Higher Education Faculty. 2007.Mahwah, NJ: Lawrence Erlbaum Associates;
27. Yin RK. Case Study Research and Applications: Design and Methods. 2018.Thousand Oaks, CA: SAGE Publications, Inc.;
28. Coyne IT. Sampling in qualitative research. Purposeful and theoretical sampling; merging or clear boundaries? J Adv Nurs. 1997;26:623–630.
29. Patton MQ. Qualitative Research and Evaluation Methods. 2015.4th ed. Thousand Oaks, CA: SAGE Publications, Inc.;
30. van der Vleuten CP, Dannefer EF. Towards a systems approach to assessment. Med Teach. 2012;34:185–186.
31. Van Der Vleuten CPM, Schuwirth LWT, Driessen EW, Govaerts MJB, Heeneman S. Twelve tips for programmatic assessment. Med Teach. 2015;37:641–646.
32. Vanassche E, Kelchtermans G. Teacher educators’ professionalism in practice: Positioning theory and personal interpretative framework. Teach Teach Educ. 2014;44:117–127.
33. Campbell A, McNamara O, Gilroy P. Practitioner Research and Professional Development in Education. 2004.London, UK: SAGE Publications Ltd;
34. Paterson BL. Mills A, Durepos G, Wiebe E. Within-case analysis. In: Encyclopedia of Case Study Research. 2012:Thousand Oaks, CA: SAGE Publications, Inc.; 971–972.
35. Miles MB, Huberman M. Qualitative Data Analysis: An Expanded Sourcebook. 1994.2nd ed. Thousand Oaks, CA: SAGE Publications, Inc.;
36. Ayres L, Kavanaugh K, Knafl KA. Within-case and across-case approaches to qualitative data analysis. Qual Health Res. 2003;13:871–883.
37. Saldaña J. The Coding Manual for Qualitative Researchers. 2009.London, UK: SAGE Publications Ltd.;
38. Geertz C. Thick description: Toward an interpretive theory of culture. The Interpretation of Cultures: Selected Essays. 1973.New York, NY: Basic Books;
39. Perri 6, Bellamy C. Types of research design. Principles of Methodology: Research Design in Social Science. 2012.London, UK: SAGE Publications Ltd;
40. Korthagen F, Vasalos A. Levels in reflection: Core reflection as a means to enhance professional growth. Teach Teach. 2005;11:47–71.
41. van der Vleuten CP. When I say … context specificity. Med Educ. 2014;48:234–235.
42. Mishler EG. Meaning in context: Is there any other kind? Harv Educ Rev. 1979;49:1–19.
43. Pfund C, House SC, Asquith P, et al. Training mentors of clinical and translational research scholars: A randomized controlled trial. Acad Med. 2014;89:774–782.

Supplemental Digital Content

Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Association of American Medical Colleges.