Wednesday, March 6, 2019

Your mentor can make or brake your academic career

[This is a copy of my blogpost, previously on]

Barely a day goes by when I don't see a mention or active discussion of abusive PIs on Twitter. This isn't surprising. I've personally heard countless horror stories about dysfunctional labs. Sometimes people are lucky to get out. Sometimes they suffer for years and tolerate the abuse. Often, they leave academia. Given how common these stories are among the people we know, if you follow 300 or more scientists on Twitter, your feed will be full of them. Here are just a few snippets from conversations I noticed in the past few days.
Academia has a particular problem when it comes to mentoring young scientists. As I've written before, we don't hire based on mentoring talent, we don't train faculty how to manage labs, and we don't reward good advisors when it comes to grants and tenure. This is a disastrous setup that results in mostly mediocre mentors, lots of terrible ones, and by chance, only some who are terrific.
The other problem is that scientists often pick labs without paying enough attention to the mentoring quality of the PI. Postdocs get a one-day visit, with virtually no ability to assess in that day what their relationship with the future advisor would be like. Not all students rotate and can evaluate the personal fit with the professor and avoid the terrorizing ones. And even entering students who do rotate frequently underestimate just how essential a good mentor is to a successful career. 
In my second year, I asked my advisor Jasper Rine why students some times knowingly join labs where the PI is a clear nut case. Jasper shrugged. I then asked, "For how long does your graduate advisor matter?" Jasper replied, "For your entire academic career." I can't emphasize this enough. Getting a good postdoc, recommendations for fellowships, and everything that follows - you can do all of this without a great advisor, but it's so much harder.
Of all the decisions we make in the decade or more of our training, picking the graduate/postdoc advisor may be the single most important decision that influences whether or not you enjoy your next six years, whether or not you finish the PhD, and whether or not you end up in the job that you love.
Here is a fantastic guide from Professor Barres, aptly called "How to Pick a Graduate Advisor".

[comments below]

I like this and find this a fascinating topic (and one I've been thinking about from the flip side for quite a while as I start to build my lab). My only suggestion is that it may be worthwhile to note how extremely personal this decision is. One can have a bad relationship with a mentor simply because personalities don't jive. I'm sure if you were to visit any successful lab, you would find a wide-range of responses to the question "Are you satisfied with the PIs mentoring ability?" And that is a much more nuanced context to navigate in a limited time frame than if the PI is simply a jerk with a bad reputation.
Great post! On a really important topic. For me, the single best advice on this topic (actually applies everywhere) is that it’s much less important what you do than who you do it with. Somehow, I’ve seen so many students who simply don’t follow this advice. I just don’t know why. The funny thing is that all the guidance services for incoming students explicitly give them this same advice, and students choose to ignore it, often to work on the “hot new topic”, or to work with on the topic that they felt they really wanted to study. I have seen the latter several times even just recently. One even said “I know you told me that, and I ignored it. I wish I hadn’t.” Perhaps it’s just the natural folly of youth. So in some sense, one of the primary enablers of the monster PIs are the students themselves! Postdocs are a whole other ball of wax for the reasons you mention.
One other point is that mentors can become worse over time (or, I suppose, better, although I think that’s less common). As other demands for time and the number of trainees increase, then I think a good mentor can just get overwhelmed and end up bad. For this reason, I’m honestly not sure why any student would work for anyone other than a junior faculty member. Yes, blatantly self-serving… :)
I think that bad mentorship does get punished, although perhaps not as badly as it ought to. Amongst junior faculty, the ones with dysfunctional labs are much more likely to be denied tenure–I’ve seen it. At the senior level, you end up with an unmotivated group that tends to not publish as much. I think that this is a place to apply pressure: if granting agencies kept track of papers/dollar and average number of papers per group member, it would be a way to make sure that PIs are taking care of all their people. An imperfect way to do it, but better than nothing. Overall, the thing is that being a bad mentor leads to lower productivity, and that’s bad for everyone, including the PI. There are counterexamples, but I believe this to be generally true.
I also think the mentor/mentored relationship is quite complex, and even with good intentions on both sides, there can be the perception of conflict of interest.Here are some thoughts I had.
Although I was not writing about the bad monster PIs, and they are definitely out there.

No comments:

Post a Comment