A simple question (or two) on training graduate students

Sep 13 2014 Published by under Postgraduate Training, Tribe of Science

This one is directed at my Readers who have supervised graduate students through the PhD in their laboratory.

Have you ever, at any point, thought that you should not train more than replacement value (n=1)?

If not, why not? What influences in your life have shaped your decision to train graduate students? What is YOUR purpose here?

69 responses so far

  • Jim Woodgett says:

    Nope. Where in the graduate school admissions docs does it say that upon successful graduation with a PhD you will be entitled to a faculty position? I try to prep my students to better compete for what they want out of life (be it academic stream or any other stream). If I look at the outcomes of the 12 PhD students who have graduated from my lab, 5 are in academia as profs (and 7 of 18 postdocs) but the others are all doing pretty well. Isn't that the goal? To train and empower people to be able to pursue paths? If none of my trainees had gone on to academic positions but were happy working in other areas, then have I failed? I don't think so. Are the trainees that are now profs happier than their colleagues who didn't? I kinda doubt it.

    It is a Ponzi scheme if we are selling students on a pathway to replace us. We should categorically not be doing that! I had a great experience as a grad student as, I would guess, did many other profs. Perhaps that also contributes to our wishing to give others similar opportunity?

  • drugmonkey says:

    So that philosophy would be "train as many as possible"?

  • We mustn't forget that we are not going out into the world and forcing unwitting innocents into graduate school to do our bidding and service our research interests. Rather, what we are doing is admitting enthusiastic and hopeful people that seem determined to go to graduate school into our programs at their voluntary request. In other words, "training fewer" students means "rejecting more" students. That is not to say that we shouldn't use our better judgement and admit those with characteristics and potential that match their actual ambition and the demands of the field, but it's not a simple calculus that admitting fewer graduate students is without effect on or solely protective of them. It is of effect because reducing the overall graduate student population means increasing the pool of people who have to find other gainful pursuits in life through some other means (this is not horrible, but is a fact).

    With this in mind, I've never set a target number (I've had as few as 1 student at a time and as many as 9 at a time). I matched my interests with the desires of those who were applying and managed things that way.

    Using a much more severe calculus, I would have to start winding down my training now, as I have likely already managed to replicate myself at least once. This would mean not working with many talented students in the future who themselves likely have a lot of potential.

  • qaz says:

    I have never considered that my purpose in life is to be a cog in the scientific machine and then replace myself once when I die. [By the way, if you believe this n=1 thing, then you should retire as soon as you train that one student, which I don't think even you are arguing.] Maybe that's because I consider myself a university professor, and thus a teacher, rather than simply a "researcher". Which is a whole other debate we have had before and will again.

    Two answers:

    First, very simply because good graduate school training is a good way to learn how to think for yourself. (I like to say that college is where you learn to learn from others and graduate school is where you learn to learn from yourself. If there are no answers in the back of the book to check, how do you ensure that your results are real?) This means that graduate training is a good provider of critical thinking training that is useful in many other jobs and skills. In fact, I would argue that PhD training is the best way to learn that one can learn to do anything - the number of people with a PhD in field A who are doing major work in field B is very impressive. (This is the arrogance of the avout in Neal Stephenson's Anathem. ) It's not the only way to learn these sorts of skills, but it's a pretty safe way to do it. Of course, this means we have to stop thinking of graduate school training as "votech" training to make faculty members. Rather, graduate school is a way of learning how to learn.

    And, anyway, graduate school is a pretty good job. So I don't have the same "exploitation" issues that you seem to. Yes, it's poorly paid relative to what equivalent training provides, but it includes a lot of freedom of project, time, and goal. (For the record, this is true of most of academia. Most professors are paid poorly relative to what they could make in other professions, but they have more freedom - particularly with tenure - and more self-determination than most other professions.)

    As far as I'm concerned, I don't worry about how many graduate students I train total. What I am concerned about is the number I am training at any one time so that I can make sure that I give them the best training I can. (What is the optimal lab size? Too small and you lose the intra-lab dynamic interactions. Too large and people get lost.)

  • Jim Woodgett says:

    @dm " So that philosophy would be "train as many as possible"?"

    Nope. Attract the students you feel have the best potential and give them an opportunity to pursue their interests and hopefully excel. I don't think it is good practice to crank students through a PhD mill if I don't think they have the skills, attitude and determination to do well - and who simply see the PhD as a thing to do, a qualification. A PhD by itself is a lousy qualification. They all look the same on the outside - a defended thesis. But they range enormously in quality. Earning the privilege to call oneself a Dr. means very little in terms of future options. It's necessary but not sufficient. It's what you achieved in the quality of your PhD, not that you earned it through working for 4-6 years. We know that and I hope prospective students do too.

  • GAATTC says:

    In my 13 years as a PI, I have trained (or am training) 7 PhD students. None have replaced me, so to speak, and I think only one wants to. Here is the breakdown:

    Student 1: Stats guy in clinical Dept.
    Student 2: Bed and Breakfast owner/mushroom farmer
    Student 3: PUI
    Student 4: in 2nd post-doc (wants academic research position)
    Student 5: PUI**
    Student 6: (current) wants PUI
    Student 7: (current) wants medical school

    **In my opinion, she had the greatest potential to do well in academic research, but jumped at the chance to work at a PUI having seen me and her post-doc advisor fight low success rates.

    So nobody has replaced me yet, and only one may. I'm sure you have all been to meetings and then go to dinner and some established prof has invited all has academic progeny. They take 40 spots and established prof looks out over his children, grandchildren and great grandchildren, and is content. Those days are coming to an end for sure.

    So I'm not contributing to the population boom yet, but would be OK if at least one of my students went the academic research route.

  • Brain says:

    In the first few years I never considered myself part of the problem. I did warn new graduate students I was taking on that the path ahead would be brutal and that the numbers would not favor them getting a faculty position. Two things have started to change my mind. It seems that it is much harder new PhDs to find postdocs (which was not my experience 10 years ago) and it is also much more expensive to support graduate students (at our university over $40K per year if they don't TA). I'm now thinking about trying to prioritize post-docs over grad students since they don't cost much more. I'm not sure that this helps the broader problem of too many phds. For me the staff route is out of the question since even someone out of college costs 60K after benefits.

    I do think that one aspect being left out of this conversation is that for many grad students (myself, and I think my trainees), grad school is a fun experience in which you get to do science and make discoveries. Maybe you'll end up with a job that did not require a phd at some point in the past, but I think many consider the experiences valuable.

  • drugmonkey says:

    Followup question for those vociferously defending their reproductive strategy: is there *any* need to consider the obvious PhD glut? Should we as a profession ever attempt to control PhD production in any way? Would an immediate 15% cut in class admissions hugely ruin everything?

  • drugmonkey says:

    Regarding replacement- I am not using this in the sense of exactly your job. At all. This is about population control.

  • anonymous postdoc (shrewshrew) says:

    The number of PhDs employed as faculty is a small minority of the total. But unemployment of any kind is far lower for PhDs than the population average. I think these data suggest that simply training a PhD student is not exploitative, as PhDs either benefit from the degree in hiring situations, or are exceptionally bright people in general, or both.

    In any case, deciding to train fewer PhDs is usually presented as synonymous with deciding to reduce the number of graduate students in a given laboratory, or reducing graduate school admissions. I am not sure that these are the only points of population control. I think that it should possibly be made harder to achieve a PhD.

    Simply having a PhD is not a reliable indicator of being an exceptional researcher. I think we all know PhDs that are great people, but mediocre scientists. Being able to get a PhD is largely a matter of getting into a PhD program - then there are vanishingly few terminal masters granted.

    Instead of admitting fewer people, perhaps we should be admitting the same number of people and then winnowing them out during their training? Instead, as things currently are, this winnowing happens after a person gets a PhD and has committed serious years to the trajectory. Not that I want this to become a "bitter postdoc" rant, because I'm not bitter. It just strikes me as an inefficiency in the system.

  • BikeMonkey says:

    It is astonishing how few people in the profession of generating PhD scientist are able to see the glut. How many excuses are raised to justify behavior.

    I hope you take the opportunity to consider that you might, just *might*, have been doing something with unanticipated bad consequences.

    Tragedy of the Commons.

  • BikeMonkey says:

    Shrewshrew- see sunk cost fallacy. Good luck with your approach.

  • Bobchickenshit says:

    My glorious opinion is that there's no such thing as "replacement value," just sets of hands to carry out my brilliant research. This is obviously because no one can replace me. The more people that I can pressure to constantly work - including weeknights and weekends - the better I can sleep. I even patrol the lab on weekends and have mandatory lab meetings on Saturday mornings just to keep people motivated!

    Proper mentoring can only happen when the soul has been broken. My multitudes of trainees usually end up as research-track employees in my enterprise (because they obviously lack the skills to become independent), or take tenure-track positions at other, less-prestigious universities, where they benefit from my name more than their publication records. That is because the papers that they left on my desk as they were leaving my lab are destined to remain there until I see fit to release them (usually when things are slow and I need a few extra publications for the year). Serves them right for leaving me! Ingrates!

  • Bobchickenshit says:

    That said, ALL of my trainees have remained in academia.

  • I don't think PIs should train exactly one graduate student to replace them. There are plenty of professions where a PhD and the associated training is an asset. However, PhD candidates should be made well aware of their options and not be deluded into thinking that they are sure to get a postdoc and a faculty position. I think postdocs are a different matter. Postdoctoral training is not really an asset outside of academia.
    Another thing to consider is the nationality of graduate students and postdocs and whether they are planning to go back to their home countries. You may not be training your replacement, but rather a replacement of someone in a less developed country. In that way you are doing technology transfer, which I think is beneficial to all parties involved in the long run. All that said, the glut is a problem and it should be kept in the back of your mind when you are hiring.

  • Bobchickenshit says:

    When I die, WHO can replace me? What a ridiculous post, Drugged Monkey.

  • Bobchickenshit says:

    Oh, no, wait - there's Henry, who is now an officer at the NIH. He always bartends for me at my private parties during the annual Society meeting.

    Then, there's Julia, who left to have kids and raise vegetables on an organic farm in California. Typical woman! Can't handle the pressure.

    So, out of 150 trainees, 2 have left academic science. My scorecard is pristine, just like my grant applications.

  • Brain says:

    If you really want to see a field in crisis check out veterinary medicine. They are cranking out more DVM's and charging them more money for a field with declining demand.

    http://www.nytimes.com/2013/02/24/business/high-debt-and-falling-demand-trap-new-veterinarians.html?pagewanted=all&_r=0

  • hpmcmf says:

    hahahaha, somebody truly suffers from an inferiority complex. pathetic to say the least.

  • Bobchickenshit says:

    I tell you one thing, though.: that Julia was an excellent grope. I bet she's a saggy mess now, with 2 kids and a third on the way.

    Nothing gets you women like a couple of R01s. And, they just keep getting younger. They may not be capable of scientific inquiry, but they do make the lab pretty.

  • hpmcmf says:

    go away chickenshit

  • rxnm says:

    I have the same question for couples who have more than 2 children. New thread?

  • BikeMonkey says:

    Or any kids at all rxnm. Word up.

  • Bobchickenshit says:

    hpmcmf: I have friends on your grant's study section.

  • jipkin says:

    If an individual PI is choosing between grad students and postdocs, I fail to see how them selecting a postdoc alleviates anything. Just means that somewhere else there's now more demand for PhDs in order to get those postdoc jobs.

    DM can you explain why cutting grad school admissions (your pet idea) would be helpful? Fewer grad students doesn't guarantee more TT positions (though increased odds for those that remain in the long run). In my simple head it really just means fewer people doing science, and therefore less science getting done. Unless somehow 15% of PhDs is not a significant fraction of the overall effective academic scientific labor force (seems unlikely....).

  • Morgan Price says:

    n=1 is not a good idea because a PhD is good training for lots of R&D jobs and teaching jobs. But system-wide, fewer grad students would mean more long-term research staff, not more postdocs. I think this could have a lot of advantages . Grad students are not very productive in their first few years, and not very happy in their last few years. Also, they are not so cost-effective once you account for tuition. A lot of people do want to stay in school or try out science for a few years after college, without necessarily planning to become professors or PIs, but a much shorter 3- or 4-year PhD experience could meet that demand.

  • Comradde PhysioProffe says:

    Should minor league baseball be "culled" because there is a "glut" of minor leaguers and almost none of them make it to the majors?

  • Kevin. says:

    As long as the net result is no more NY Yankees, I'm all for it.

  • BikeMonkey says:

    Straight up, Kevin, straight up.

  • MoBio says:

    Bobetc's sarcasm aside--there is a glimmer of something hidden there.

    DM: would you ask Brian Kobilka (or pick your favorite luminary) if he/she should consider not training graduate students because they already greatly exceed N=1 (though it's not yet clear whether the "replacement" is in that mix)?

  • theLaplaceDemon says:

    Shrewshrew - "Instead of admitting fewer people, perhaps we should be admitting the same number of people and then winnowing them out during their training? Instead, as things currently are, this winnowing happens after a person gets a PhD and has committed serious years to the trajectory. Not that I want this to become a "bitter postdoc" rant, because I'm not bitter. It just strikes me as an inefficiency in the system."

    The problem with winnowing during grad school is deciding HOW to winnow. How do you separate the good scientists on bad projects from the bad scientists? The people who are closely interacting with that student are probably going to be people with a vested interest in having them around. How do you account for neglectful mentors? No one wants to tell their colleague "we're letting your student pass because we think they've done remarkably well under your mediocre mentorship."It's also possible for someone to be on a project that uses a technique that they are simply not suited for - but it took them a few years to figure that out, and now it's too late too switch and still graduate on a reasonable timeline. Do you give the PhD to someone who is a great bench scientist, but lacks the other skills necessary to succeed in academia? How do you evaluate that?

    Also, winnowing during graduate school would make an already stressful, intense experience even more stressful - not necessarily the most conducive environment to doing good science.

    I think it would probably be better to admit less students overall than to winnow during graduate school. Obviously, some bad scientists will still slip through the cracks, but hopefully with less students programs can devote more resources to training the ones they do have.

  • Comradde PhysioProffe says:

    The winnowing occurs during post-doctoral training, which makes perfect sense in light of the goal of identifying those scientists that are the best bet for taking a chance on allocating the resources to start their own labs.

  • hpmcmf says:

    @MoBio

    “Bobetc's sarcasm aside--there is a glimmer of something hidden there”.

    Maybe. I, however, think that there are topics in which sarcasm, and the extent of it, can only serve to feed the mentality and behavior that such sarcasm is intended to highlight.

    I can’t understand how DM and PP, so committed to influencing positive change in academic research by bringing up discussion on critical issues, allow such comments to stay. But then, everybody is different in the way to approach problems, as well as in expectations.

  • Someone says:

    The winnowing occurs during both graduate and postdoctoral training. Have you ever noticed how many students leave programs after only a year or 2?

    It is highly problematic, though, if say, a promising scientist ends up with the bad end of the stick during postdoc - either a bad project or an unsupportive mentor (or both). With regard for the latter, a postdoc can get all the unofficial mentoring that he/she wants, but at the end of the day, the postdoc's official mentor is the person who chooses to provide sufficient funds for the postdoc's project (at least for the first 2-3 years), support the postdoc's salary (even on training grants, the mentor's word counts), write letters on behalf of the postdoc, and publish the postdoc's papers. This dependency is perhaps even worse now, given the current funding climate, unfortunately.

  • Grumble says:

    The decision is being made for us - "we as a profession" don't have to do anything. At my med school, for instance, the number of grad student admissions has been cut to about half of what it was a couple of years ago.

    That said, I've never considered it to be an ethical dilemma that some (even most) students won't get faculty positions. Giving them a free education that gets them an advanced degree is not in any way abusive.

  • CAYdenberg says:

    You know perfectly well what the answer is. Why not explicitly blog your thoughts on the matter, instead of asking incendiary quetions?

  • Comradde PhysioProffe says:

    At my institution, a substantial amount of institutional funds were recently committed to subsidize graduate education, with the intended and achieved outcome of increasing the number of graduate students we admit. This is done both to increase our yield of the top applicants we compete with our peer institutions for and to increase our ability to recruit and retain the best faculty.

    BTW, sorry if you think this is "face rubbing", but these are facts about what a certain subset of peer institutions are doing. The extent to which it might offend or annoy you doesn't make it any less real. As a group these peer institutions are small in number, as are the numbers of grad students in the aggregate that they train, and who are disproportionately successful at ultimately securing faculty positions.

  • theLaplaceDemon says:

    @Someone - Legitimate points, but postdocs are in a much better position to evaluate potential mentors and projects than a new graduate student is.

  • bacon says:

    Help me think about this--I recognize that there is an oversupply of PhDs who are searching for permanent academic positions. Why is this a bad thing? I recognize it reduces "success" for the competing individuals, but I'm trying to see how that has long-term negative effects on science as an enterprise. The oversupply may have more to do with directing PhD students towards academic careers and therefore keeping demand for these careers extremely high. But I'm not sure that having fewer individuals with science PhDs is a good thing.

    I disagree that being a graduate student is exploitative. Students here make almost $30K a year--more if they win a fellowship--which is more than the research technicians with similar qualifications.

    I am currently training three grad students. I waited a while before taking any on--I felt that as a young faculty person I needed to be able to guarantee support for a student and offer a decent project. So I waited until year 5 to take rotation students. I had one student get a masters after 2.5 years, which I think was a good choice on her part, for a lot of reasons. The others are still in grad school, with different career plans, and I try to support whatever their career plans are (e.g., setting up one to teach at a local community college since she ultimately wants a teaching gig at a small college). I don't have plans to stop training students any time soon, but I do urge them each to consider career alternatives to academia.

  • toto says:

    "But system-wide, fewer grad students would mean more long-term research staff, not more postdocs"

    OhWaitYou'reSeriousLetMeLaughHarder.jpg

    "This is done both to increase our yield of the top applicants we compete with our peer institutions for and to increase our ability to recruit and retain the best faculty."

    Difficulty: in some parts of Europe, recruiting your own PhDs as faculty is seen as cronyism ("copinage") and is a common talking point. This may be a result of the different funding mechanisms.

  • rxnm says:

    'The winnowing occurs during post-doctoral training, which makes perfect sense in light of the goal of identifying those scientists that are the best bet for taking a chance on allocating the resources to start their own labs."

    Oh, is that what we're doing? Carry on, then.

    Because it seemed to me like we were cloning douchebags.

  • "This is done both to increase our yield of the top applicants we compete with our peer institutions for and to increase our ability to recruit and retain the best faculty."

    Difficulty: in some parts of Europe, recruiting your own PhDs as faculty is seen as cronyism ("copinage") and is a common talking point. This may be a result of the different funding mechanisms.

    What the fucken fucke are you talking about? The reason recruiting plenty of outstanding graduate students is important for recruiting and retaining the best faculty is that the best faculty want to have access to the best, most talented, driven, and creative grad students to train.

  • Alex says:

    Yeah, CPP's department isn't trying to grow their own. They're just trying to offer a good selection of top strains.

  • Cynric says:

    Interesting angle on this from the UK: our research councils are all moving to a "training programme" model for funding PhDs now, in contrast to the old model of letting the universities distribute a block grant to faculty as they saw fit.

    As part of this reorganisation, the Councils have insisted on training in transferable skills, which can include a compulsory placement for 3 months in a non-research environment (e.g. NGOs, publishing, broadcasting etc.) specifically because of the dearth of academic/industrial positions.

    Needless to say this has provoked howls of outrage from the PIs, who are losing valuable lab time on "pointless" distractions. More interestingly, the students don't like it either (based on anecdotal conversations with our cohort - none of whom I supervise). Both students and PIs obviously value the PhD as a chance for focussed study on a topic of interest and development of scientific research skills, not as a worthwhile qualification that improves employability, or an exercise in honing critical thinking skills for "general use".

  • Cynric says:

    And for what it's worth: I take on grad students very reluctantly, and strongly favour postdocs.

    This has had consequences for my career, as the university sees training PhDs as a core part of its remit and so I have had to justify my resistance on a few awkward occasions. I'm sure the fees that come with the students are incidental...

  • qaz says:

    Wait - @Cynic - shouldn't that n=1 issue apply to postdocs as well? Why is training extra postdocs better than training extra grad students?

    If anything, a postdoc is farther along than a graduate student and is going to have fewer options when they fail to find a faculty job. College training makes sense to me (learn from others). Grad school makes sense to me (learn from self). [Could this be done a cycle earlier (learn from others in HS, learn from self in college)? Yes. Is it? No.] But what is the purpose of postdoctoral training? Beyond a holding pattern? (BTW, many fields get away without forcing people to do postdocs. Watching certain fields that have made the transition from gradschool->faculty to gradschool->postdoc->faculty within my lifetime, it looks like postdoc appears when the faculty job market becomes saturated.)

  • Cynric says:

    I don't favour postdocs for any reasons related to pipeline or replacement of self. I just prefer it as I find graduate students a lot of work, as I expect to give them a lot of focussed attention (as I assume they will need to make more useful errors as part of their development).

    I also think the whole postdoc "training" moniker is a cynical ploy that decreases the professionalism of scientists. Postdocs are autonomous workers, and no more or less likely to require training than any other professional entering a new job and finding their existing skills aren't a perfect match. I expect them to learn from their mistakes more quickly and with less input from me, and to bring new skills to my lab.

    So, on reflection, it's a rather selfish preference - but it does also mean I'm employing existing PhDs in a job that they want to do, rather than adding more PhDs to the entry pool.

  • MoBio says:

    To newish faculty re: students

    Musing about the discussion I offer up a slightly different perspective:

    When you begin training students you will not know whether/if your students are 'successful' for 8 or more years (4-5 yrs for PhD then 3+ for postdoctoral training).

    What this means is that you'll likely train a set of students but won't know how successful they will be.

    And it may be a long, long time before you get feedback.

  • qaz says:

    Cynric - Is that a long-term job that you are employing these PhDs in? (I suspect not.) And are you paying them a wage comparable to what they can get elsewhere? (I suspect not.) If not, how do you justify that? (Given that you have rejected the "training" explanation for postdoc exploitation.)

    BTW: These uncomfortable questions are just that. I am as guilty as anyone of training more postdocs than are necessary to replace me.

    I justify that to myself in two ways. First, by believing that my special snowflake training is superior to others. As with any variation and reproduction process, not everyone gets to replace themselves. The scientific workforce evolves as do species - by changing the population of variants. If the skills and resources I can provide (including scientific, conceptual, writing, and networking) that I can teach are better fits for the current ecosystem, my trainees will do better than others'.

    And, second, as my postdocs know, I put a huge amount of effort into providing the training and support they need, and will go to the mat for them. Given that the current system requires postdoc time, I do my damnedest to make it worth their while. (But then, I suppose I'm more of a "K" reproducer than an "R" reproducer.)

  • Drugmonkey says:

    qaz- and don't forget the historical analysis within (sub)fields.

  • drugmonkey says:

    BTW: These uncomfortable questions are just that. I am as guilty as anyone of ...[insert any particular point].

    One thing that may not be entirely clear to my newer or less frequent Readers is that my questions about how we should be behaving as a profession are being asked of myself as much as they are of any segment of y'all.

    When I come at this from the perspective of Labor exploitation, I recognize that I play a part in this scam. Of course I have my own little excuses for why I do things slightly better. As does everyone.

    I recognize that different labs, job types, subfields, etc come with different paths of ease - for example ready access to grad students paid for by something external to the lab, smart and motivated undergrads or, yes, grant funding to pay techs and/or postdocs with. The particulars of our jobs dictate which of these Labor forces we rely on most heavily....and even if we are able to choose any-of-the-above so as to make ourselves proud of our least-evil choices.

  • More than one of my former trainees currently are PIs of their own labs. Is this "unfair" to others who have zero former trainees running their own labs?

  • drugmonkey says:

    I make a critical distinction between postdocs and graduate students, CPP. Training postdocs is dealing with the PhD glut. Training graduate students to the PhD is increasing the problem.

    So it would depend what sort of "former trainees" we are talking about. IMO.

  • rxnm says:

    "I make a critical distinction between postdocs and graduate students, CPP. Training postdocs is dealing with the PhD glut. Training graduate students to the PhD is increasing the problem."

    Eh...I'm not sure about this. There is *something* to the argument that getting a PhD has value outside academic research institutions. There are non-scientific industries that actively recruit science PhDs, for example. A smattering of the usual suspect "alt careers" require a PhD.

    There is fuck-all point to doing a postdoc.

    Regulate numbers by making trainees (phd and postdoc) line items in research grants, abolish F awards.

  • drugmonkey says:

    There is *something* to the argument that getting a PhD has value outside academic research institutions. There are non-scientific industries that actively recruit science PhDs, for example. A smattering of the usual suspect "alt careers" require a PhD.

    Yeah, those are getting over-subscribed too. What we are left with is only the creeping of credentialism.

    Think of how at present a random Communications BA has a leg up over a GED for a retail job and advancement within same. The PhD advantage is rapidly becoming this. Unnecessary for the job, but a convenient way to screen out the field between more-deserving and less-deserving based on credential but not capability or anything else.

  • qaz says:

    I don't understand how training postdocs is "dealing with the PhD glut". It is simply moving it one step up in the evils of creeping credentialism. If anything, the PhD glut appears because postdocs are the cheap labor that science works on. (A postdoc is, supposedly, less work for a PI than a graduate student. [Ha! Your mileage may vary - mine certainly has.] But generally a fully fledged postdoc will hit the ground faster than a newly hatched graduate student.) I feel I could make a case for graduate training beyond the faculty track. But what job and other skills does a postdoc gain that are appropriate to anything beyond the faculty track?

    As I mentioned earlier in this conversation, I've seen fields that have changed within my academic lifetime from "no one does postdocs" to "only losers do postdocs" to "everyone does a postdoc". As far as I can tell, the only correlation I could find was the saturation of the field with faculty. (And I suppose the saturation of the alternate career options which happened at about the same time.)

    PS. There are specific cases where I can make an argument for postdoc as a very useful training. Such as when you have a math or physics PhD and want to do neuroscience, so you need to learn to do experiments. Or when you need to learn a specific complex technique that will take two to three years to get up and running. But that's the exception, not the rule.

  • drugmonkey says:

    I've seen fields that have changed within my academic lifetime from "no one does postdocs" to "only losers do postdocs" to "everyone does a postdoc".

    I have too and in the field I am thinking of it has recently evolved to "only losers fail to have grant money in hand when applying for tenure track jobs".

    Apart from the postdoc glut situation, we agree entirely that there is no reason scientists cannot be transitioned to independent running-of-programs right out of graduate school. IMO. I was VERY disappointed to learn how much hand-holding and limited-appointmenting there was in the Early Independence program

  • Cynric says:

    Cynric - Is that a long-term job that you are employing these PhDs in? (I suspect not.) And are you paying them a wage comparable to what they can get elsewhere? (I suspect not.) If not, how do you justify that? (Given that you have rejected the "training" explanation for postdoc exploitation.)

    As you guess, the answers are No and No.
    The weaselly justification for this is that I have very little freedom to set salaries, despite being the grant holder (a UK universities thing, as we have fixed salary scales). The honest answer is that I offer what I can, advise to the best of my abilities, and actively encourage my staff to apply for personal funding during the project period. I see this as the least bad situation, given that I do think we underpay scientists. Too long of an argument for here, but I think pay scales should taper steeply, so that junior staff get paid much more earlier but advance more slowly. Financial worries during young scientists' most creative period is counterproductive, and hampers research progress.

  • jipkin says:

    "I make a critical distinction between postdocs and graduate students, CPP. Training postdocs is dealing with the PhD glut. Training graduate students to the PhD is increasing the problem."

    wat. Where do you think the postdocs you train come from? If you really want to address the glut, run a lab powered by staff scientists and undergrads.

  • Drugmonkey says:

    How does that help with the current PHD population?

  • Alex says:

    To me, the critical choke-point is PhD production. The PhD takes a college grad, a person who could go down any number of paths, and specializes them. It acculturates them to academia at a crucial juncture in life, when they are (usually, but with all cases of non-traditional students respectfully noted) young.

    Once they come out of that, it's not like the world is falling all over itself to hire them into non-postdoc positions. So, yeah, go ahead and hire a postdoc. You're giving them a job. A postdoc could (in principle) leave tomorrow, and they still have their title, and they have whatever they published. A PhD student is different--if they leave before they finish, there will be a milestone that they didn't reach.

    To me, the psychological impact of being a postdoc was way less significant than the psychological impact of graduate study.

    So, if you limit PhD production, the postdoc treadmill will eventually sort itself out. And until it does, at least you're giving the postdoc a job. So go ahead and hire them, and treat them fairly. But think long and hard about PhD production.

  • Why is everyone ignoring the reality that there are institutions, departments, and PIs that should be producing fewer PhDs and institutions, departments, and PIs that should be producing more?

  • drugmonkey says:

    Because the hit rate of those "institutions, departments and PIs" that think they are hawt stuff are not improved in any way. Sure, you can cherry pick the anecdote of 1 but then try and show why you or your training program differs from the next PI or training program with as much arrogance as your / your program. Doesn't add up. and that is what you need, prediction rather than post-hoc justification, to prove that you (and your ilk) need to be encouraged and the next loser needs to be cut off from PhD production.

  • Actually, we just computed statistics for our PhD graduates over the last two decades, and the percent who are currently PIs is nearly triple the national average.

  • Alex says:

    So you guys should be graduating 1 PhD per year instead of 1 every 3 years? 🙂

  • jipkin says:

    My point DM is that I don't understand why you'd want to limit the current PhD population at all. Like, what's the plan from that point onward. All I've heard is

    1. Decrease number of current PhD students (by decreasing admissions, say).
    2. ???
    3. ???
    4. ???
    5. Science is better off!

    Someone connect the dots for me.

  • rxnm says:

    "Actually, we just computed statistics for our PhD graduates over the last two decades, and the percent who are currently PIs is nearly triple the national average."

    Proving meritocracy. Slow clap.

  • rxnm says:

    the circular logic of elites really takes my breath away sometimes.

  • Dude, somebody's gotta train the faculty of the future.

Leave a Reply