Should we stop training PhDs in certain sub-fields?

There's been a lot of discussion in the last few months about the number of PhDs we produce. You may recall JHU trying to get out in front of the issue, but it's still a hot button topic:

The question of reducing the number of PhDs produced is a complex one and there's simply no chance that the overall numbers will be reduced evenly across the board. It seems inevitable that certain fields are going to dry up and blow away and I think this has direct bearing on how we consider training students.

In my own lab we do several things that can more or less be divided into two camps, one of which is more "classical" and the other more cutting edge. Importantly, I have NSF funding for both, so each can be considered a viable enterprise from a funding perspective. However, there is a wild skew in the job prospects for students with training in one vs. the other.

This leads to a significant dilemma. Obviously I feel that both types of work are important and both contribute significantly, but there are just no jobs to do the classical work. And I don't just mean no academic jobs. I mean that training in this particular field leaves you few options outside of an academic job, of which there are none. There is funding out there for this type of work, but it is not accessible if one can not find a faculty job to exploit it.

We've been trending this way for more than a decade. I see it from the labs I know training people in this field. I see it in our regional conference that is bimodally skewed to the old and the students. I see it in the job ads that circulate. Unless you have other significant skills or are willing to leave the counrty, it is excrutiatingly difficult get a job (academic or otherwise) utilizing a PhD in this classical field.

In stark contrast, everyone who spent most of their time in the other side of my lab has left and found desirable (to them) employment. They have skills that are more broadly transferable and that a wider range of potential employers are interested in. Therefore, I am left to wonder whether it is even ethical for me to accept PhD students into my lab to work on the classical stuff, or is it simple labor exploitation. Ironically, I get more applicants interested in this than the cutting edge stuff we do.

Is it time for me to back-burner the classical stuff? Do I need to begin to rethink all of these projects to bring in new techniques that might alleviate this issue (some are emerging, but they are not yet ready for prime time)? Should I be the one deciding or should the student applicant pool decide?

14 responses so far

  • hufbauer says:

    Thanks for the thoughtful post, Prof-like substance. I think that yes, you should be the one deciding whether students get a PhD or not. It is against our (faculty) interests to have MS rather than PhD students, but I think that is a good intermediate path. I agree largely with drugmonkey's sentiment that training PhD students is often exploitative. No we can not and should not have to guarantee them jobs, but knowing most will not get jobs yet taking them anyway is, I think, unethical.

    Students who come out with an MS and still really love science and want to try for it will understand the risks much better, and should have a stronger CV when they enter the job market. Students who decide it is not for them have put in less time, and will have more alternate career paths open to them, often, than people with a PhD.

    My response doesn't really address classical versus new stuff - too hard for me to say not knowing the particular field. I love classical science in general, and think much to much has been left by the wayside already, so I'd keep it if you can!

  • drugmonkey says:

    You should decide.

    If you can get the ones that come for the classical stuff to pick up enough cutting edge skills to make them competitive with the other folks, hey, great. But it is really hard to pull off IME. Getting students out of their comfort zone when that comfort zone exists in your lab, is productive and has the PI's interest in a general sense.....not easy.

  • proflikesubstance says:

    I think marrying the two approaches isn't a great fix just yet. I'm more apt to have people focus on the cutting edge projects and sprinkle in the classical stuff as it applies. That may change with a few technology advances, but we're not there yet.

  • Renewbie says:

    This may be a naive question but why do we need to have students doing so much of the work? Two of the most successful PIs I know run smallish labs staffed primarily or exclusively with postdocs, PhD level staff scientists, and techs. Both have told me that they do it this way because students require a much higher investment of time and are not actually that much cheaper than staff, especially considering that it takes a few years before PhD students are able to produce anything. When I first heard this I thought it sounded odd and even kind of selfish, but now that I am a new PI myself, I get the argument. It also seems like it might be a responsible approach. It seems like one other advantage would be that the career prospects of the trainees and potential trainees would not need to drive the scientific directions as much in a case like this where the science itself is fundable but not marketable at the moment.

  • proflikesubstance says:

    That kind of approach still requires a good amount of funding. Salaries are expensive, which is why most labs are more student-heavy. I'm not saying that is the correct way to do it, just the reality.

  • I think Renewbie is on to something though: as long as funders want lab staff on the cheap, the incentive is to train students to do things that aren't in their own best long-term interests.

    That observation, of course, doesn't solve your particular problem.

  • M says:

    I think there are several possible approaches:

    -Explain to students up front what their job prospects will be coming out with that particular skillset. Make sure they know that they will have to pursue an academic (or national lab, perhaps) job (difficult to do) OR plan to switch fields for the post-doc for example. This might mean fewer students will want to dive into this work, but if they really love that type of science, they might be inclined to continue, while knowing from the beginning that they will need to be thinking about their personal development and career path from the beginning.

    -Focus on training soft skills (writing, scientific thinking, project management, etc.). Yes, one should be focusing on these things, but REALLY focus on them. An amazing student should be able to switch fields with enough talent and soft skills.

    -Have students work on hybrid projects or joint projects, where the classical work is only half-time.

    Students need to be taking ownership for their future from the beginning. This is difficult to do when one doesn't know any better, so I think the best thing to do is to provide guidance. Really talk about job prospects. There ARE jobs out there: science writing, science teaching, program management, and totally random work that has little to do with the PhD but that benefits immensely from a PhD-level education. If the student wants an industry job in their field, then they probably shouldn't do something totally classical/academic. But if the student has an eye to entrepreneurship or alternative careers and is extremely bright, they can make it work. And you should help them to know what it takes.

  • Busy says:

    I refuse to train a grad student in a death end area. At the same time, career management is also the responsibility of the student.

    I agree that professors over-rely on grad students because they are "free" when in several instances a post-doc would be better suited and in the end cheaper. This has the added benefit that it diverts money from producing more PhDs into the pockets of unemployed graduates.

  • Claus Wilke says:

    In my mind, the main purpose of graduate school is to train students to be independent scientists: how to develop research questions, how to test hypotheses, how to analyze data, how to write papers. The specific area doesn't matter that much. It matters much more at the postdoc level, because you'll usually get a faculty (or industry) position in the area in which you're doing your postdoc. Therefore, if graduate students are interested in the classical work, I think it's fine to take them on, as long as you explain to them that they will likely have to switch fields somewhat after they graduate. By contrast, having postdocs work on something with few job prospects is unethical, I think.

    See also this post:

  • Paul says:

    I suggest sending them away for an internship for one of the summers before they get a masters. They will get a clear taste of the job market and may find they enjoy something a little more marketable than what they planned to do.

    I think you are in an ethical gray area, but I also wonder about the funding agency that makes it so easy to"train" students with only academic prospects if there are no jobs in academia. Why does that agency only fund academia? If that's the plan, then the agency really needs to tackle this overproduction of PhDs problem, too. Sounds like the humanities--except they are more sanguine up front about having limited job prospects.

  • proflikesubstance says:

    Paul, I'm not sure how exactly I would "send people away for an internship" or what that really even means. And I should be clear that the training provided during a PhD (at least here) is not Academia Or Bust. My point was that training in this classical field results in fewer job prospects (not just in academia) than training in other work we do.

  • Paul says:

    I think you're fine ethically as long as you make it clear to your prospective students what kinds of jobs are out there and how likely they are to get them. You'd be especially fine if you specifically say that it's a serious issue and that you personally are a bit concerned about the job prospects your students face. That should get their attention.

    As for internships, I know of three examples. One grad student worked down the road at a non-profit doing research in our field. One grad student was sent by a somewhat evil prof to work at a bank (not doing related work, either, so that's the example to avoid). One grad student (me) spent a few weeks each summer working at a national lab. Mine was more like a fellowship than an internship. It was useful to see how a non-academic research group functioned. As it happens all three of us are now working outside academia, and only one of us (the guy who got stuck at the bank) is working outside our field, but still at the national lab where I did my stint.

    Maybe your field doesn't have programs like this. Maybe it could. Maybe my examples are too anecdotal to be of any use 🙂

  • Zil says:

    I wish they had at least talked about job prospects when I was choosing my research area. We were so entrenched in the academic pursuit -- which I do think has intrinsic value, but it's just not always realistic. I got to the end and realized there was no job. I definitely would have picked a different research area and probably a different grad program had I thought seriously about job prospects.

    I wish people would stop telling prospective grad students "you can do anything with a Ph.D." That's the standard line where I'm from and it's complete and total b.s., because what you can't do is get a job in the city you want to live in. Unfortunately, I bought the line an now I'm stuck in a temporary job in the middle of nowhere.

    Talk, talk, talk with students. Be realistic. Let them follow their dreams, but give them an honest, upfront picture of their future.

  • TheBrummell says:

    Regarding internships, a year ago I was in the middle of mine; now I'm near the end of my PhD, with the intention to stay inside the ivory tower and eventually land a job as a professor. As somebody who may be delusional, therefore, please take what I say as only anecdote. I'm in a Department of Soil Science in a Faculty of Agriculture (actually, "Agriculture and Bioresources") at a mid-sized Canadian university that covers pretty much every discipline (we have a med school, a vet school, a theological seminary, a school of business, etc etc). My work might be called "classical" because I have strenuously avoided any hint of practical applications - my colleagues express puzzlement at my work in the Arctic, where there is no agriculture.

    My advisor tries to send every one of his PhD students away for what he calls an internship, a period of between 3 and 6 months away at another lab or institution. The idea is to expose the student to different perspectives and ways of doing the kinds of work we do. His lab - currently he has 10 PhD students - is split between a "classical" area (Microbial Ecology) and a "cutting edge" area (Ecotoxicology). I'm not sure the categories perfectly overlap with what has been described - I think of the two areas more as "Academic science" vs. "Applied science", given the focus of most ecotoxicology research on solutions to problems such as reducing human exposure to various environmental toxicants such as benzene (from fuel spills) or copper/zinc (from mining waste). I'm on the Microbial Ecology side, and while I do plan to stay in academia (see "Delusional", above) I've had many conversations with people of similar qualifications who are happily not in tenure-track academic positions; I do not think my skills leave me unlikely to get a non-academic job should I choose that path in future.

    I went to work with CSIRO in Australia, mainly to access some microbiology / molecular biology tools & techniques they'd developed (so perhaps that's "cutting edge"), and I spent 4 months in Tasmania. It was wonderful, and scientifically valuable for my career, as well. Within the last few years my lab mates have variously gone to Belgium, Denmark, the UK, and a large consulting company in the next province over. We each have come back with the desired expanded perspective as well as a long list of valuable contacts and scientific skills. It's a great program, I'm a huge fan, and when I think about the kind of professor I want to become, I try to think how I could send my future students on similar adventures.

    It's largely a funding issue, I think. Sending people away to work in other labs seems rather expensive - my advisor was charged "cost recovery" for the work I did in Tasmania, which is considerably cheaper than sending samples away for somebody else to process and analyze, but still amounted to around $15K (travel costs were mostly covered by an award directly to me). My advisor is very, very good at winning grants, I do not think I will be in anything like a similar position as far as lab finances go, ever.

    Changing topics, what is the "classical" field that you work on? I have the impression that some currently less-than-sexy fields are due for a resurgance as important experts retire or die. Consider taxonomy - how many biologists have a story about requiring the services of some ancient (and perilously ill) expert lichenologist, arachnologist, or ichthyologist, some long-retired emeritus employee of a national museum (in another country)? Getting specimens identified and sorting out the biogeographical and phylogenetic implications in one's work is a major headache for many people I've talked to over the last few years. I have occassionaly pondered a "what if...?" that puts me on a taxonomy career path from my M.Sc., and from looking around I think such a person - recent PhD in Biological Systematics (specialist in whatever Phylum/Class/Order) would have considerable choice in job placement, and in future collaborative work with many different researchers. Like fieldwork? Go out with the Ecologists! Like a well-stocked lab? Hook up with the Molecular Biologists! Museums may be facing funding crunches of their own, but I (wildly, possibly delusionally optimistically) think they'd save a bit for a taxonomist. The small number of self-described Systematists I've met have all been happy people, at least.

Leave a Reply