Guest post by Meredith M. Course* and Irini Topalidou.**

The career success of graduate students and postdocs (referred to here as “trainees”) is largely dependent on just one or two principal investigators (PIs). PIs influence how trainees think about science and whether they choose to stay in the field. The fundamental purpose of mentorship is for a trainee to intellectually and professionally benefit from the mentor’s experience and network. Good mentorship benefits PIs and institutions, too: well-mentored trainees will produce higher quality research, ultimately leading to labs that run smoothly and productively. Because PIs are largely responsible for defining our academic culture, who we put in these positions of power and what we expect of them deserve careful consideration.

Though a PI’s mentorship ability plays a crucial role in shaping the field, it is often undervalued and inadequate.1 Not every PI possesses the understanding of what it means to be an effective mentor or knows how to apply these principles; some PIs have good intentions but simply don’t know how to train, guide, and support trainees; others do not prioritize good mentorship, and have no incentive to. Unfortunately, almost every academic institution— and many of their trainees— have suffered due to bad mentors. Given its significance, why is poor mentorship a persistent issue in academia? Major contributing factors include: (1) a lack of training for mentors, (2) the overwhelming demands of a faculty job, (3), a hiring and promotion method that ignores mentorship abilities, and (4) institutional protection of PIs at the expense of their trainees.

The issue develops well before a scientist becomes the head of a lab, since training to become a PI rarely includes learning to mentor effectively. During the doctoral and postdoctoral years, the chief goal of a future PI is to publish several quality, first-author papers—an expectation that does not require or promote mentorship abilities. This emphasis on research may result in outstanding experimentalists, but it does not create effective mentors.

After focusing on benchwork as a postdoc, a young PI experiences dramatic changes in responsibilities; they must suddenly excel as a mentor, manager, grant writer, and leader. Provided with little guidance, PIs are left to discover on their own, by trial and error, how to accomplish these tasks. Unfortunately, the “guinea pigs” of these trials are their trainees. Without being set up for success, overstressed PIs may end up misguiding or even mistreating their trainees.

Furthermore, because the main criterion for hiring and promoting a faculty member is the candidate’s quality of research and not their mentorship skills, institutions accumulate PIs who are not necessarily good at being mentors. In turn, this inadequacy can result not only in unsatisfied trainees, but also PIs who feel insecure or unhappy in their positions. The fact that the institutions entrusted with educating our future scientists do not require evidence of successful mentorship calls into question their role in properly training the next generation of scientists.

Because trainees depend on their PI’s guidance and support for professional success, they have difficulty finding where to turn to when a PI is either passively or actively failing to provide this support. Institutions often protect PIs because they provide grants, have tenure, or both. Institutional protection of PIs, coupled with the lack of a reporting system for trainees, can trap trainees in lab environments where they feel disempowered and alone.

The pervasiveness of poor mentorship in academia suggests that the issue is not a case of individual bad actors; rather, it is a systemic problem. Seen through this lens, we contend that mentorship reform should not fall solely to individual PIs, but to the institutions that both prepare and employ them. Institutions can begin to address this problem if they: (1) recognize the issue, (2) implement mentorship training for postdocs and PIs, and (3) use selection and evaluation methods that provide incentive and accountability.

First, the academy at all levels should recognize that capable mentorship is necessary for their trainees to have healthy and productive educational experiences, and for institutions to represent centers of educational excellence. Understanding that the responsibility of quality mentorship lies at multiple levels within the system—such as university leadership, department chairs, program directors, as well as PIs— is the first step toward improved mentorship.

Mentorship training should begin during the postdoctoral period. Postdocs interested in faculty positions should be encouraged to mentor trainees, and they should be offered mentorship training to improve their understanding of what it means to be a good mentor. This training and experience can help postdocs determine whether a position requiring mentorship truly fits their skills and interests. Alternatives to faculty positions should be considered and equally encouraged, especially for postdocs who are uninterested in mentorship.

Once hired, junior PIs should also receive mentorship training together with guidance and feedback from experienced PIs. It is crucial that this training be mandatory, as those who are unaware that they lack mentorship abilities are the least likely to seek it out. Mentorship training should be included in tenure requirements as part of the PI’s protected time, to signal to PIs that the institution prioritizes mentorship.

Equitable and inclusive strategies need to be part of mentorship training also, as trainees from underrepresented groups generally receive less mentoring than their well-represented peers, which prevents them from benefitting professionally in the same way.2 Institutions should educate their PIs about this issue and outline how they expect them to combat it. Fortunately, high-quality, evidence-based mentorship trainings and tools are available freely (Entering Mentoring3 and The Science of Effective Mentorship in STEMM4 are excellent starting points), therefore institutions do not need extra resources to implement them.

Finally, institutions should realign their selection and retention criteria by requesting evidence of a candidate’s mentorship abilities for faculty hiring and systematically evaluating the PI’s ability to mentor at promotional junctures. After a PI is hired, feedback from both experienced PIs and trainees should be taken into account—and be taken seriously, which means that institutions need regular and standardized mentoring evaluations, such as the freely available Mentoring Competency Assessment.5 In addition, trainees who are struggling under poor mentors deserve human resources personnel to turn to when they need advice, advocacy, and protection from poor mentors. Requiring mentorship compacts6, 7 is another easy and free way for programs and departments to protect trainees, improve accountability for mentors, and assess mentors’ performances.

One of the chief responsibilities of a PI is mentoring future scientists, yet frustration over inadequate mentorship is frequently lamented among trainees. This is not just a failing of individuals, but signs of a system that sets up PIs to fail. Here, we’ve outlined several reasons why institutions accumulate poor mentors, and several steps they can take to ameliorate this issue. It is imperative that institutions take charge of the mentorship environment that they provide to trainees and implement selection, evaluation, and training paradigms that focus on the mentorship ability of those training our future scientists. The good news is that improved mentorship does not need to involve starting from scratch: quality, evidence-based, and free resources already exist. They deserve increased use and prioritization. Effective mentoring will not only improve the experiences of trainees, but also benefit individual PIs, institutions, and the scientific community as a whole.

*Division of Medical Genetics, University of Washington, Seattle, WA, 98195, USA.

**Department of Biochemistry, University of Washington, Seattle, WA, 98195, USA.


1A message for mentors from dissatisfied graduate students

C Woolston (2019)

Nature 575, 551-552

DOI: 10.1038/d41586-019-03535-y

2NIH Scientific Workforce Diversity Office

3Entering Mentoring: A Seminar to Train a New Generation of Scientists

J Handelsman et al. (2005)

4The Science of Effective Mentorship in STEMM Online Guide

National Academies of Sciences, Engineering, and Medicine

5The Mentoring Competency Assessment: Validation of a New Instrument to Evaluate Skills of Research Mentors

M Fleming (2013)

Acad. Med. 88(7), 1002-1008


6Mentorship Compacts/Contract Examples

UW Institute for Clinical and Translational Research

7Ten simple rules for developing a mentor–mentee expectations document

KS Masters and PK Kreeger (2017)

PLoS Comput. Biol. 13(9): e1005709.

DOI: 10.1371/journal.pcbi.1005709

Guest posts are contributed by members of our community. The views expressed in guest posts are those of the author(s) and are not necessarily endorsed by the Genetics Society of America. If you'd like to write a guest post, e-mail

View all posts by Guest Author »