Those in academic medicine agree that the goals of mentorship are to support the development and retention of faculty who are productive, satisfied, collegial, and socially responsible. No randomized trials of the effects of mentorship have been done, but systematic reviews of the literature on mentorship and career choice in academic medicine1–3 suggest that effective mentorship produces faculty who are more productive (including obtaining more grants and publications than colleagues without mentors), promoted more quickly, and more likely to stay at their academic institution. These conclusions come largely from one systematic review that explored the impact of mentorship on faculty members’ career choices and academic advancement, based on 42 articles describing 39 studies, 34 of which were cross-sectional self-report surveys.1 We identified 11 studies that have been published since that review in 2006, all of which were observational studies.4–14
Leaders at many academic health centers (AHCs) have come to recognize the importance of mentorship, and thus they have shifted from debating whether to support faculty mentorship programs to discussing how to do so.4,15 However, gaps in the literature, identified in the previously mentioned systematic reviews,1–3 make it challenging for AHC leaders to optimally develop mentorship programs and strategies. For example, the systematic reviews1–3 highlighted the characteristics of good mentors16–23 but not the characteristics of effective mentees or what constitutes a successful or failed mentoring relationship.1–3,23 Understanding these concepts is critically important for faculty members who are searching for mentors and for those who want to be more effective mentors themselves. This knowledge also could be used by AHC leaders to design mentoring programs to develop and retain academic faculty.15–26
We designed our study to explore the characteristics both of good mentors and mentees and of successful mentoring relationships. In particular, we were interested in exploring faculty members’ views on and experiences of mentorship across two different AHCs that instituted formal mentorship programs. The Department of Medicine at the University of Toronto Faculty of Medicine has over 400 full-time members in 17 divisions. It implemented a mentorship strategy in 2003 that included encouraging both new recruits to identify a mentor and existing mentors to participate in mentorship workshops. The Department of Medicine at the University of California, San Francisco, School of Medicine (UCSF) has over 500 full-time members and implemented a mentorship program in 2006 as part of a university-wide initiative. The UCSF program has multiple components, including mentorship workshops, awards for excellent mentors, and a mentorship tool kit for mentors and mentees.4
We invited faculty members from the Departments of Medicine at the University of Toronto and UCSF to participate in individual, semistructured interviews to explore and characterize the mentor–mentee relationship. We used stratified purposive sampling to ensure inclusion of participants from each rank (lecturer, assistant professor, associate professor, professor), gender, and academic stream (teacher [80% clinical work and teaching], educator [50% clinical work, 50% scholarly work in education], investigator [50% research], scientist [>70% research], and administrator). From each department of medicine, we obtained lists of faculty members that outlined their rank, gender, and academic stream. We sent letters of invitation to a sample of these participants from each rank and stream; at two and four weeks, we sent follow-up e-mail reminders. We asked faculty members to reply to the letter or e-mails; once they did so, a research associate contacted them to complete a telephone interview.
Data collection and analysis
After we obtained informed consent from each participant, an experienced interviewer (C.M.) conducted a semistructured telephone interview (March to December 2010). We developed domains of inquiry using the results of the published systematic reviews on mentorship1–3 and a recent qualitative study16 as well as through consultations amongst the investigators. Questions included items about experiences with mentorship and elements of a successful mentoring relationship (see List 1). As we analyzed the data (March to December 2010), two investigators (C.M., S.E.S.) added questions to address new themes that were identified (see List 1). We also collected participant demographic information, including institutional affiliation, career path, and gender.
We digitally taped the interviews and transcribed the recordings verbatim, assigning each interview a unique identifier. We completed a content analysis (March 2010 to March 2011) of the transcripts beginning after the first interview and drew on grounded theory, using a process of open, axial, and selective coding.27,28 The goal of these interviews and the analysis of their contents was to develop a synopsis and understanding of mentoring relationships (including the actions of a good mentor, characteristics of good mentors and mentees, and the qualities of good mentorship) as experienced by our participants.
In open coding, two investigators (C.M., S.E.S.) independently read each transcript, identified the themes using a constant comparative approach, and developed codes from these themes. They repeated this process for each transcript. The same two investigators (C.M., S.E.S.) grouped the codes into categories, and axial coding was done to look at the interrelationship of categories, including a consideration of context (such as the university setting), intervening conditions, and consequences. We were particularly interested in comparing and contrasting participants’ experiences across the two institutions.
Sampling continued until no new themes were identified.27,28 We used written memos to provide a record of the analytic process. The memos captured the decisions and results of the analysis, helped to develop propositions, and were compared by the two investigators (C.M., S.E.S.). They shared these memos with all of the investigators, and we deliberately tried to discount or disprove a conclusion drawn from the data. We determined reliability of the categories by the frequency or consistency with which they were indicated by participants in their interviews. We used NVivo 9 software (QSR, Victoria, Australia) for our analysis.29 We received approval from the research ethics boards at the University of Toronto and UCSF.
We contacted 54 participants who responded to our letter and e-mails, and all agreed to be interviewed. All of the participants had experience as mentors and as mentees, and each academic stream and rank from each institution was represented (see Table 1 for complete demographic information). We identified several themes focused on mentorship, including the characteristics of effective mentors and mentees, actions of an effective mentor, characteristics of successful and failed mentoring relationships, and tactics for effective mentorship. We did not identify substantive differences in the participants’ responses by institution.
Characteristics of effective mentors
We identified several personal characteristics of effective mentors. Most commonly, participants mentioned that mentors should be altruistic: “They have a huge responsibility not to transform that potential into what the mentor sees where it should go but to be detached from that and making sure it’s in the best interest of the mentee.” The mentor not prioritizing the mentee’s best interests can lead to a failed mentoring relationship, which we discuss later. Participants also stated that mentors should be honest, trustworthy, and active listeners. All participants identified honesty as a crucial characteristic for effective mentors: “Just being honest and telling someone … you know that this idea is not a good idea or they need to be doing something else.” Similarly, most participants described the need for effective mentors to be active listeners. One participant stated:
I think that the mentor should play the role of listener so it’s important to listen to what the mentee is saying in terms of what their important goals and objectives are when you’re sort of working through a problem as far as trying to give advice. It’s hard not to kind of impose your ideas and what you think would be right for yourself onto the situation but I think a good mentor kind of listens to each individual mentee and tries to give advice … tailored to that specific person and their own goals and objectives with respect to the certain problem.
Being an active listener requires that the mentor is engaged with the mentee during each session, is focused on the issues identified by the mentee, and prompts the mentee to clarify if there is any confusion. Through active listening, the mentor also facilitates goal setting for the mentee.
Participants said that effective mentors usually had substantial mentorship experience and that an experienced mentor becomes particularly important as one’s career progresses. One participant said:
Having a wealth of experience to draw from in terms of prior mentor–mentee relationships that at my stage and my career I wouldn’t be obviously looking for a junior mentor because I myself mentor other people so I would be looking for more senior mentors who have a wealth of experience and can reflect on sort of where I am in my stage of my career and … find someone at the appropriate stage of their career who has had significant personal life experience in the “school of hard knocks” but also has prior mentor–mentee relationships over a number of years or the number of different mentees so that they’ve been able to draw from that wealth of experience.
In addition, effective mentors have professional experience (including networks of colleagues and collaborators) that can facilitate mentee development. One participant stated, “It’s very helpful for junior people to have somebody who knows the system so that they can give good advice ‘cause I think a lot of people don’t really understand the system.” Another participant said, “It’s important to have a mentor who can open doors and can help sort of jump-start and catapult your career.”
Effective mentors also exhibit important relational characteristics, including being accessible and able to identify and support the development of potential strengths and skills in their mentees. One of the key challenges for mentors and mentees is a lack of time, and participants stated that effective mentors ensured that they remained accessible to their mentees, even if they were located at a distance. Although they may not be able to meet in person regularly, effective mentors used e-mail and phone contact to ensure accessibility. One participant said that it is important that a mentor “is approachable and available when they need them.” Good mentors were able to identify potential strengths and limitations in their mentees and promote their career development. For example, a good mentor “understands what the mentee is trying to accomplish in their career, what their limitations are.”
Characteristics of effective mentees
Participants outlined the personal characteristics of an effective mentee. Mentees should be open to feedback and be active listeners. They also should be respectful of their mentor’s input and time. One participant said:
The mentee should listen to the mentor and take them seriously and that doesn’t mean following every bit of advice … if you’re working with someone and they’re giving you advice you know if you kind of ignore all of it then it’s sort of a fruitless interaction and then also kind of another important part is for the mentee to follow through and be timely with things.
This quote highlights other commonly reported characteristics of effective mentees—being responsible, paying attention to time lines, and taking responsibility for “driving the relationship.” Another participant said,
You can’t just go in and be an undifferentiated blob about what you want, you have to really have thought before you go in and meet with your mentor about what the issue is that you need help with and you know it’s much more useful if you bring your own analysis in with you and then the mentor can give you their analysis and you can talk.
A mentee showing respect for his or her mentor includes being respectful of meeting times and prepared for meetings. Participants stated that the best mentees attended mentorship meetings with lists of topics for discussion, including time lines for projects. Moreover, participants felt that mentees needed to be respectful of the competing demands on their mentors, including meetings with other mentees and their own deadlines. In particular, participants raised the concern that if mentees send mentors their manuscripts, grants, or other documents and ask for a quick review without advance notice, it increases the stress on mentors and, in some cases, leads to burnout for the mentors who may provide similar services for many mentees.
Actions of effective mentors
Participants identified several key actions of effective mentors, including providing career guidance, offering emotional support, and focusing on work/life balance. By career guidance, participants meant that effective mentors act as guides rather than as supervisors who direct their mentees’ activities. Participants stated that mentors “need to be guides, be sensitive to the difference between a guide and somebody who forces the student into or the mentee into a particular path,” and mentors “may well offer some advice but recognize that it is only advice, it’s not orders.” Another participant said, “The most important thing is not trying to solve their problems but to help them find solutions.”
Participants included the following as career guidance: advising, advocacy, networking, creating opportunities, goal setting, career monitoring, and helping mentees navigate institutions. For example, one participant said a mentor needs to “provide strategic advice to help the person kind of help themselves in the best way possible,” and many participants outlined the need for mentors to provide “critical feedback.” Several participants commented on the mentor’s role to provide advice: The “mentor’s responsibility is to assist their mentee in terms of their career and to sort of provide advice and support and feedback and you know be a sounding board for the mentee.” Another participant stated that the mentor has to
find out what exactly they’ve [the mentee] been doing and the different aspects of their career in terms of teaching or clinical or research or administration and can review their job description … to make sure they know what is expected of them now … help them develop a long-term plan, like where they see themselves in a year, 3 years, 5 years, 10 years from now and see that they’re getting adequate support to get to where they want to go and if it’s practical.
Participants also mentioned creating opportunities, helping mentees identify potential opportunities, and providing introductions as being critical actions of effective mentors. One participant said that a mentor is “really like a guide, opening doors, giving opportunities, advising about the future, about avoiding mistakes, where to spend time and resources on things that matter instead of trying to do everything … helping to have balance.” Another participant stated that mentors make
introductions to other people, provide information about funding sources, checking in about “well are you where you want to be, do you know what your opportunities are, what opportunities are you trying to get …,” a facilitator of connecting people to sources of information or other people.
Participants also mentioned that effective mentors warn their mentees of potential pitfalls and protect them from harsh interactions. One participant said:
The role of the mentor I think is really to be a guardian angel so what I mean by this is your guardian angel prevents you from hitting yourself when you know something is falling from the sky … moves you. Keeps you out of trouble and also makes the environment suitable for you to grow.
Participants stated that effective mentors should monitor their mentees’ career progress and ensure that they have assistance navigating their institutions. As one participant said, a mentor should “ensure that [their mentees are] progressing adequately in the system and to some degree protect and warn them of pitfalls that others in similar positions have encountered.”
In addition, participants noted that effective mentors should provide emotional support to their mentees, including sharing their own feelings honestly and encouraging their mentees to do the same. Mentors should help their mentees identify what factors may be contributing to their emotional state, such as relevant stressors. One participant stated that mentors “should help [their mentees] learn how to deal with stress, how to do one’s job effectively and keep one’s sanity.” Another participant described mentors as having a “bit of the nurturing role just to kind of give the mentee the message that they are available to discuss other issues, maybe more personal issues that may impact on them and their career.” Participants also said that effective mentors should provide encouragement and proactively “check in with people to see how they’re doing.”
Next, participants stated that mentors need to help their mentees reflect on the appropriate balance between their work and personal life. Although discussions with mentees often focus on career issues, participants outlined the importance of mentors focusing on how mentees should target opportunities and ensure that they were not neglecting their personal lives. One participant stated that a mentor should “try to develop a very clear idea of what the mentee wants and desires in their career path and how that career path and how their work interrelates with their larger personal life and social life.”
By practicing these actions and exhibiting the personal characteristics of an effective mentor that we described earlier, mentors can role model mentorship for their mentees. Participants stated that many people are not naturally effective mentors and that strong role models can positively influence their behavior.
Characteristics of a successful mentoring relationship
Participants identified five key features of a successful mentoring relationship: reciprocity, mutual respect, clear expectations, personal connection, and shared values (see Table 2 for representative quotes).
Characteristics and consequences of a failed mentoring relationship
Reflecting on their experiences, participants identified several contributing factors to ineffective mentoring relationships, including poor communication, lack of commitment, personality differences, perceived (or real) competition, conflicts of interest, and the mentor’s lack of experience (see Table 3 for representative quotes). Most participants either had observed or participated in a failed mentoring relationship. They described the consequences of such a relationship, including failure to obtain a grant, failure to retain a promising junior faculty member, and inability to maintain a relationship with the mentor leading to lack of collegiality in the department.
In most cases, when a mentoring relationship did not work, participants reported finding someone else to provide mentorship instead and stated that the failed relationship “was a good life lesson.” However, the experience led them to be more cautious in approaching potential mentors in the future; for example, several participants spoke to the other mentees of their potential mentors for an independent assessment of the mentor before formalizing the relationship. Some participants also suggested that the mentorship facilitator or the department chair could act as a “broker” when a mentoring relationship is failing, help defuse a potentially tense situation with the previous mentor, and find a new mentor if necessary. However, some participants stated that junior faculty might not be comfortable discussing a failed mentoring relationship with their department chair or with anyone more senior because of concerns around the potential impact on their own career. As a potential solution to this issue, one participant mentioned the possibility of being able to “break up” with a mentor, having a situation analogous to a “no-fault divorce” so that a level of collegiality can be maintained. One of the most serious concerns raised by some participants was that, in extreme cases, the failed mentoring relationship led to junior faculty’s disillusionment with academic medicine and may have contributed to their leaving the institution.
Tactics for a successful mentoring relationship
Participants identified several tactics that mentors can use to optimize their mentoring relationships. As most relationships start in the mentor’s office, it should be an environment where the mentee feels safe and welcome. Next, mentors should establish a communication framework, including what several of the participants referred to as a process of “reiterate and review.” This approach includes ensuring that the mentor and mentee understand the discussion and action items. One participant described it as “making sure that the mentee and the mentor understand what they’re talking about and asking for repetition and clarification to make sure that they understand what we’re talking about.” Another participant suggested
checking back to what the goals from the last session were, asking them if there are new learning goals this time, delving into each one of the learning goals and as time permits at the very end recapping what was communicated, accept and establishing accountability or those things.
Several participants suggested that the mentor use a checklist to frame discussions to ensure that time is spent addressing career, administrative, education, and personal issues. Several participants reported using such a framework to guide their initial meetings with mentees. For example, one participant stated that by the end of the initial discussion, he was able to determine “have I introduced myself, have I explained what my availability is, have I given them my cell number, have I established a list of goals to achieve, and have I accomplished them at the end.”
Finally, participants identified several strategies for enhancing the organization of meetings, including scheduling regular appointments and keeping a list of action items. Participants did not agree on how often meetings should occur, but all did agree that they should be regularly scheduled. The number of meeting times that participants recommended ranged from twice per year to monthly. They also advised communicating via e-mail and telephone regularly between meetings.
Although some studies1–3 have shown the benefits of mentorship, less detail has been available on the characteristics and actions of effective mentors and mentees and on the characteristics of failed mentoring relationships. Our study fills the gaps in the literature on these topics. To our knowledge, it is the largest qualitative study on mentoring relationships and is unique in its inclusion of participants from two large AHCs in North America.
Several studies outlined the characteristics of good mentors,3,23 including personal characteristics, interpersonal abilities, and professional status. In a qualitative study of nomination letters for a mentorship award, investigators concluded that good mentors should exhibit admirable personal qualities, act as career guides, make time commitments, support personal and professional balance, and role model good mentoring.23 Our study provided rich detail on the personal qualities needed to be an effective mentor. Moreover, it went a step further in providing a description of the qualities needed to be an effective mentee. Much of the literature focuses on the mentor, but, as outlined in our study, a mentoring relationship involves two people, and thus its success depends on the characteristics of both individuals. Similarly, we found that both the mentor and mentee should share mutual goals, respect, trust, and commitment to the relationship for it to be successful.
A recent systematic review of the literature described the actions of effective mentors3 but offered little information on the nature of successful or failed mentoring relationships. Our study provided unique details on the characteristics of failed mentoring relationships, including lack of communication, of experienced and knowledgeable mentors, and of commitment to the relationship. We found that competition between mentors and mentees, however, contributes more to a failed relationship. Previous studies and commentaries also identified this competition and an exploitative relationship where either ownership of intellectual property was not clear or the mentor and mentee had competing interests as causes of failed relationships.16,24 Of particular concern is how common the participants in our study perceived these toxic relationships to be, especially given that mentors often serve as role models for their mentees.
Our study has a few limitations. We interviewed faculty members at only two institutions, so our sample size may limit the generalizability of our findings. However, we did include participants from both the United States and Canada, making our study unique. Also, participants may not be entirely representative of the faculty at these institutions. However, our sample of 54 faculty members is large for a qualitative study, and it included participants from all academic ranks and streams. Moreover, we continued sampling until saturation of themes occurred, so we do not believe that we missed any points of view.
Successful mentorship is vital to career success and satisfaction for both mentors and mentees. Our study provides guidance for mentors and mentees on how to create a successful mentoring relationship. In particular, it provides information for mentees on what to look for in a mentor and how to structure their mentoring relationship. In addition, although some faculty are naturally excellent mentors, all require some training to improve their mentorship skills, both to be better mentors and better mentees. We recommend that such training programs focus on promoting the characteristics of effective mentors and mentees that we have outlined here. For example, we created a series of scenarios, based on our findings, that we use for discussion at mentorship workshops; these cases highlight issues such as how to find a good mentor, how to prepare for mentoring meetings, and how to “break up” with a mentor, which is especially important if the relationship is failing.
Several participants in our study observed or experienced failed mentoring relationships, which have potentially significant consequences. On the basis of the results of our study, we recommend a few specific strategies for dealing with failed mentoring relationships, including using the mentorship facilitator or department chair as a mediator, implementing a “no-fault divorce” rule whereby either the mentor or mentee can end the relationship, and developing a workshop on communication and good mentorship for faculty to attend.
There are many gaps in the mentorship literature. For example, no quantitative studies look at the association between a failed mentoring relationship and a faculty member’s promotion, retention, or academic productivity. Similarly, no studies assess the different approaches to mediating failed mentoring relationships. Additional gaps in the literature exist regarding both strategies for effective mentorship, including the impact of mentorship education interventions, and appropriate mentorship throughout a faculty member’s career. Given the importance of mentorship on academic faculty members’ careers and the declining interest in pursuing a career in academic medicine, future studies must address these issues.
Funding/Support: S.E.S. is funded by a Tier 1 Canada Research Chair; M.O.J. is funded by a National Institutes of Health grant (K24MH87220); M.D.F. is supported by a Fulbright research grant and a National Institutes of Health grant (1UL1RR024131).
Other disclosures: None.
Ethical approval: The authors received approval from the research ethics boards at the University of Toronto Faculty of Medicine and the University of California, San Francisco, School of Medicine.
Additional interview questions
* Do you perceive that women have more trouble finding a mentor than men?
* Can you provide examples of this?
* Should mentors be assigned?
* Why/why not?
* Should mentorship be facilitated?
* If so, by whom?
* Are there any categories of people who you believe shouldn’t provide mentorship for an individual (e.g., division director, department head)?
* In this report, we focus only on the responses to a portion of these questions.