My last post generated a fair amount of commentary both here on this site and on Reddit. It seems that many people have experienced exactly what NIH Director Francis Collins described: they’ve been made to feel like failures for leaving academia. If the vast majority of PhDs and postdoctoral fellows will not become tenure-track academics, then we should be embracing non-academic careers as the default pathway for most trainees. This requires a huge cultural shift away from seeing trainees as generators of science and toward viewing trainees themselves as the product. Making supervisors and institutes accountable for the trainees they produce, keeping lab sizes down to a size where meaningful mentorship can be maintained, and recognizing the value of non-academic careers are all key to making this shift successfully.
Tracking former students and postdocs in a meaningful way
A recent phenomenon at granting agencies has been to track the outcomes of students and postdocs, both in terms of how long they are in the lab as well as what they are currently doing for a job. A big question that spawns from this is whether particular professions or outcomes are more or less valued? If so, who decides if it’s more valuable to create a PhD-level patent lawyer vs. a sessional instructor vs. a professor? I am not sure how this information is used and I worry that it is simply to ask the question “How good is this professor at creating new professors?”
As I mentioned in my last post, the product of a university should, above all else, be its people. We consistently fail this goal by nearly exclusively valuing the production of papers and patents irrespective of what happens to the trainees involved in producing them. Instead we should be measuring a successful research PI by evaluating – in a meaningful fashion – their training and teaching abilities.
Bigger is not always better
One of the most frustrating things about measuring the “productivity” or “success” of a lab is that it is almost always done as a cumulative exercise. Rarely do you ask the question, “What is this professor’s productivity per lab member?” Not too long ago, there was an eye-opening study that showed research productivity plateaued at $750,000 of research funding and got noticeably worse as funding went up (as measured by number of publications and their average impact factor). Even this, though, did not break down the production per person, though it can be reasonably assumed that better funded labs have more people.
Big labs produce more papers, that’s very true – but how many careers are buried in the wake of such “productivity”? It would do grant evaluators well to ask how many trainees and employees does each lab have and how is the lab’s publication record distributed over those people. Anecdotally, I can cite several examples of small labs with excellent productivity that get crushed in grant evaluations for having a thin publication record – a “publication per lab researcher” metric would do such labs a great service and push the heads of larger labs to ensure that everyone in their group is being taken care of properly.
Seeing the forest for the trees and the Selfish Gene
On the note of taking care of one’s trainees, I fail to understand why professors don’t see “non-academic” career options as valuable to them. Yes, professors can have tunnel vision when it comes to doing things that benefit their lab moving forward, but surely minting new academics is not the only way to have a positive working relationship with your former trainees.
If you produce a journal editor, might thaey not end up working at a journal in your field? Would you rather have them respect the lab for the way it is run and the science that emerges from it or that they be bitter about their final few months/years and be spreading bad vibes throughout their new circles?
If you produce an industry researcher, might they not end up working for a company in your field? Good relationships with companies have often sprouted collaborations that benefits both the academic and industrial partners both through shared reagents, shared expertise, and good product development opportunities. Even the most selfish professors should be able to see this logic and be keen to have students of all career motivations leave their labs as happy as possible.
Changing the perception
I have long wondered whether people are better motivated by negative or positive reinforcement. Should we reward those professors that invest in training students and postdocs with diverse career goals or should we penalize those that neglect their university duties? As it currently stands, there does not appear to be much reward for those who invest in training and teaching and there appears to be a sizable cohort of professors who are not well-liked by their trainees.
I suggest measuring output based on all the trainees that pass through a lab by noting where they go and how well they were supported and I would also incorporate productivity per researcher into evaluation metrics. Such measures would stimulate professors to consider carefully those that they take on board and I believe would bring down overall lab size of large labs and increase the productivity per research dollar.
In response to the question, “What type of research does your university do” I always answer, “My university doesn’t do research. My university creates spaces where researchers do research”. Similar concept. I like how the role of the unviersity is to produce scientists not science. You rightly point out that as institutions we are measured (in part) on the science our scientists produce (especially in a REF 2014 UK).
Just today I got an e-mail from my PI inviting us trainees to a consultation session with a funding agency “because we are its most valuable product”. Clearly you are on the same wavelength…
Somewhat related, I would be interested to know what criteria one would use to determine how well a PI performs at training trainees.
A very insightful post Dave, as usual, thanks!
I dare note that there has been a bit of whinging in my field about NSERC’s overemphasis on HQP during the evaluation of DG proposals. But NSERC seems to be already asking some of the questions that you suggest they consider when they are weighting this criterion (see the Contributions to the Training of Highly Qualified Personnel section at http://www.nserc-crsng.gc.ca/NSERC-CRSNG/Policies-Politiques/assesscontrib-evalcontrib_eng.asp).
Having recently flipped through “The Parrot Problem Solver” by Barbara Heidenreich (the Cesar Millan of the parrot world), I do think positive reinforcement may be a better motivating factor. It works for parrots, so why not academics!
Great thoughts here, Dave.
Why do you think the focus on content/ideas “research” vs. people/skills persists? Can we attribute this solely to history, inertia, culture, and incentive structures, or are there other factors?
My thought: A public shift in emphasis from research output to personnel output (“HQP”) will result in diffusion of institutional power away from faculty and toward students/trainees/postdocs. Without being melodramatic, I think this is an important consideration.