(This Edge for Scholars post is a co-production of Dr. Josh Fessel and Dr. Jen Heemstra.)

Josh here.  I was recently catching up with a former lab tech who is now in a fantastic PhD program at an excellent research university.  She had just switched labs and advisors, a move that has turned out to be really positive.  As we discussed this, she told me about her previous mentor, and what she told me was simply shocking.  Personal attacks in emails, erratic and unpredictable behavior in the lab, gossiping about lab members to other lab members – all the ingredients of a truly toxic environment were on display.  I couldn’t stop thinking about all the things she’d said, and about the fact that, although somewhat extreme, her experience of having a mentor that comes up short is, sadly, not so very rare.  Worse, this stuff seems more common than what I remember from my time in grad school.  I turned to social media for a quick gut check and was lucky enough to meet Dr. Jen Heemstra, an Associate Professor in the Department of Chemistry at Emory and a paragon of mentorship, trainee advocacy, and scientific awesomeness.  We started talking about this very simple and distressing question:  Are we experiencing a mentoring crisis in science?

The Scope of the Problem

So, how bad is it, really?  A recent study (flawed though it be) found that depression and anxiety are extremely common among graduate trainees, with 39% and 41% of respondents reporting depression or anxiety symptoms, respectively.  The numbers were slightly less for those identifying as men, slightly more for those identifying as women, and way more for those identifying as transgender.  Issues with mentorship emerged as being a very, very big deal across the board.  Most trainees suffering from either anxiety or depression reported unhealthy work-life balance and stated that they were not getting real mentorship, ample support, positive emotional impact, or a feeling of being valued from their mentors.

It’s no better for postdoctoral fellows, by the way.  It’s just that we don’t have as much granularity in the data.  For a heartrending example, read the story of Dr. Francis Dolan and Dr. Oliver Rosten.

Jen suggested that we gather our own data (paragon of science awesomeness, remember?), so we did just that.  Looking at the results, it’s unclear whether the news is good or bad.  In a mostly unscientific poll on Twitter, out of 170 anonymous respondents, 61% reported being satisfied or highly satisfied with the support and mentorship they were receiving.  So, for the majority of trainees, we’re doing pretty well!  That’s the good news.  The bad news is, that means 39% were dissatisfied or highly dissatisfied.  That would mean we’re not doing a good enough job as mentors for almost 2 out of every 5 trainees. Moreover, in an even less scientific poll, we informally asked graduate students what they expected the poll to reveal, and based upon their interpersonal interactions with other students, they universally guessed that at least 50% of students would report some level of dissatisfaction, and were surprised that the numbers in the poll trended so positively.


Criticize the studies, but don’t miss the message.  This is serious.  Even for those who aren’t experiencing diagnosable depression, anxiety, substance abuse, or other such health issues, things aren’t exactly rosy.  After talking with trainees (and more than one program coordinator/director of graduate studies), several themes emerge repeatedly:

  • Mentor neglect, with mentors not really being involved enough in trainees’ projects, or sometimes not being physically present at all
  • Mentors holding out (and holding up career progress) for the “big money and fabulous prizes” – the splashy C/N/S paper, the bigger and better analysis that would cost a ton to outsource but that “would really enhance your dissertation,” etc.
  • Mentors being unaware of trainees’ career aspirations, either because they’ve just never asked or because mentors cling to the belief (shown by the data to be as dead as old dad’s hatband) that if you’re not aiming at a tenure-track PI position in an R1 institution, you’re wasting your time and your mentor’s time
  • Mentors being unaware of significant pressures in the lives of their trainees, including common things like time requirements for coursework, funding difficulties, “real-life” issues like healthcare coverage and expenses or the illness or death of a close family member, and unique training needs (e.g., accommodating specific disabilities)
  • Mentors dismissing or – far worse – participating in discrimination or harassment of trainees in the lab and/or at conferences
  • Mentors not quite knowing how to help a struggling student, if they are even aware that the student is struggling

Why Is This Happening?

Here we’re in a data-free zone at the moment, but hypotheses abound.  Maybe the biggest single contributing factor to the mentorship crisis has to do with what’s required just to keep a lab afloat.  At least one grant going in every cycle to maintain two concurrent active R01-level grants to run even a relatively modest size lab is now the norm.  Couple this with the ever-increasing requirements for manuscripts to be successfully published, and it’s a darn pressure cooker on the good days.  This was definitely not the case when we trained.  We’re all aware that what PIs used to do with the time they now spend writing grants and wrangling with reviewers was, y’know, doing more creative and groundbreaking science, thinking harder about the data and the questions, and maybe even reading more deeply and widely.  What we sometimes forget is that that’s also when a lot of mentoring happened.  My mentor’s door was always open, and we could talk about anything.  This in spite of the fact that he was a famously irascible dude, and he only ever said “Not now, please” one time…when he was writing a grant and was up against a deadline.

Done right, mentoring takes a lot of time and effort.  That’s time and effort that is almost entirely without direct incentive on the part of the PI, particularly when it comes to promotion and tenure decisions.  And who suffers the consequences of bad mentorship?  Not the PI, certainly.  Can you think of one lab that has shut down because the PI was a crap mentor?  I can’t, but I can think of numerous examples of young scientific careers that were hobbled or ruined by terrible mentorship.  And I can certainly point to PIs who are wildly successful by conventional measures but who are also widely known among trainees (especially former trainees) for their abusive or negligent behavior.

This is all made worse by the fact that trainees are in a supremely vulnerable position.  They depend on their mentors to ensure successful completion of their projects and to open doors professionally for them.  They are, in very real ways, at their mentors’ mercies for their current and future professional success.  Who protects trainees?  Certainly there are directors of graduate studies, but there are two big limitations on what they can actually do.  First, the DGS has no real sanction power over any mentor.  Second, the DGS is a scientist in their own right, trying to run their own lab and keep their own grants funded and papers published.  And for postdocs, there’s not even a DGS.  The protections are slim to nonexistent for trainees, even as PIs enjoy a freedom from accountability as mentors that would be unthinkable in most other arenas.  A sports team doesn’t have a winning season, and it’s the coach who’s sweating it, not the players.

We don’t want to dwell overlong on the negative.  Again, 61% satisfied or highly satisfied.  At least some of us are doing some things right.  The goals, then, would seem to be to identify what those things are, what other things we could be doing better, and to communicate those things widely.  We’d love to find ways to double down on successes and to avoid throwing out the baby with bathwater.

Does It Have to Be This Way?

Imagine if we could change the culture to reward great mentoring just as we currently reward other research-related metrics such as papers and awards.  This change can come about in two ways: (1) Top-down change – In any R1 department, research is a top priority, and graduate students and postdocs are the lifeblood of that enterprise.  Department and university administrators should have an active interest in making sure these individuals are receiving adequate mentoring and are being provided with supportive environments in which to work. Many departments do in fact look closely at mentoring when considering faculty tenure or promotion, even going so far as to solicit letters of recommendation from current and former mentees.  What if this were done on a more regular basis, and used as a metric for considering academic rewards? (2) Bottom-up change – Students and postdocs have the power to vote with their feet and only join labs of supportive mentors.  The trick is to make this information available to trainees.  Word of mouth can be somewhat effective, but lab rotations give students a much better opportunity to sample the environment before they commit. Faculty who place a high priority on mentoring are undoubtedly already rewarded by being able to recruit the top students in their programs, but this effect could be further amplified by providing trainees with more information about lab and department environment as they choose their path.

What Can We Do?

We think lots, but we want to avoid the classic mistake of making recommendations without talking to the stakeholders.  So, consider this the first of a short series of posts on the topic of improving mentorship.  In the coming weeks/months, we hope to talk with trainees, mentors, and a variety of other folks with skin in the game regarding training future generations of scientists.  We’ll report back with what we learn.  At the end, we’ll hopefully know what we’re doing right, what we should be doing better, and how to go about doing it better.

It’s long past time for a discussion of what to do for the 39 percent.  Stories of trainees in crisis are increasingly common and easy to find.  To do nothing different means that we’re okay with the suffering and, sometimes, with the ruin or death of our most vulnerable junior colleagues.  Personally, neither of us is okay with that.

Want to live on the Edge?


Join the conversation

Your email address will not be published. Required fields are marked *

Saving subscription status...


I suspect that over time, the ‘bad mentor’ proportion has stayed fairly constant. Awareness of it has increased, but the pressures on the mentors themselves have increased, and so I’ll bet things stay about the same. Some of the ‘bad mentor’ stories I’ve heard have been from folks in labs that were about to collapse, and the mentor seemed on the verge of a nervous breakdown…

For some reason, I’m unable to reply, but I can post a new comment.  This is meant as a reply to Dr. West’s comment:
“I suspect that over time, the ‘bad mentor’ proportion has stayed fairly constant. Awareness of it has increased, but the pressures on the mentors themselves have increased, and so I’ll bet things stay about the same. Some of the ‘bad mentor’ stories I’ve heard have been from folks in labs that were about to collapse, and the mentor seemed on the verge of a nervous breakdown…”
I must respectfully disagree with you, Dr. West.  I don’t think the bad mentor problem has been constant over time.  Awareness *has* definitely increased more broadly, which is a very good thing.  But in talking with lots of graduate students and other trainees over time, this definitely is a problem that is growing, as far as I can tell.  Challenges have always existed, certainly, but the stories I hear now are markedly different (more numerous, more egregious) from the stories I heard in the past.  The pressures on mentors have *definitely* increased as well, and I think that’s a big part of what’s exacerbated the problem.  And though some bad mentors are indeed facing personal and professional crises and those are being visited upon their trainees, there are just as many examples of mentors that are wildly successful by most conventional measures who are truly terrible (sometimes downright abusive) when it comes to mentorship.  I won’t name names, but one could probably find out without a lot of legwork.
I’d be very interested in other people’s thoughts on this.  Speak up, y’all!

C. says:

I am in a senior technical position in the Ophthalmology department of a major university in the west. I don’t have a Ph.D. but have worked closely with many Post docs, grad students and rotating grad students. I have done my fair share of mentoring and crying with students. What I have found true is that department gossip from the lab technicians and post docs is a very good predictor for mentorship quality and for the behavior of graduate students themselves. I am still surprised though. I have seen graduate students and mentors bully and attack other technicians and graduate students and mentors overlook and even condone it as, ” Well, they are successfull.” I have seen the University do nothing to support anyone except the abusive PI and allow abusive PIs, graduate students and post docs to move on to new territory where their abuse continues. 
Mentorship means that the entire lab and department participates in the protection of others and the correcting of poor behavior and ethics. Part of being in a senior position is sort of like parenting. You are molding and shaping and it is a community effort.
I have watched many be permanently broken and never speak up. It needs to change from the NIH down. The message from the students up needs to be, “We stand together in support of each other. We will not stand for bullying, abusive and absent mentors.” Hopefully, the universities hear the message and things improve.

You May Also Like