Enough is enough! How much is enough?

Dec 03 2013 Published by under Uncategorized

Here's a quick poll for you readers who are in academia or have been. In your university, how much work does a PhD student need to do before he or she is allowed to graduate? Do you need a set number of papers, do they have to be published or is a bunch of unpublishable data sufficient? Is this something fixed or is it determined on an ad hoc basis? Should there be a set criteria for number of publications? Whatever you can eke out in five years? Do you agree or disagree with your institutions policy? Please mention what stage you are in your career in your comments.

Now, órale, discuss...

87 responses so far

  • DrugMonkey says:

    Strict published paper requirements are part of the exploitation game.

  • bashir says:

    No set requirements with regard to papers. In theory the prelims and dissertation lead to that. but they don't have to. I am fine with that I suppose. I don't think the faculty want to link graduation to pubs per se, though it ends up that way effectively given the typical prelim and diss requirements.

    I'm a postdoc

  • For my program, the bare minimum was a first author paper in a journal with an impact factor >5, and 2-3 mid-author papers. The committee decided when it was "enough" - a 4th year student with these was generally not allowed to graduate yet, generally because they seemed to receive a project well underway if they could meet the requirements in that minimum time frame. If you had a high risk project that fizzled out into nothing, like I did, you were just stuck until you had enough things working for a first author paper. As a 7th year grad student, I hated that rule - anyone in my program would have agreed that when it came to technique mastery and experimental design, I had accomplished more than all those who graduated before me, I was just stuck with a very bad project for a grad student. Now as a postdoc reflecting back, however, I'm glad I had to stick around extra time to get that first author paper - I'm positive I would not have gotten my F32 funding if I had no first author papers to list.

    • namnezia says:

      Wow, it's insane that not only did they require 3 publications, but one in a high-ish impact factor journal! And that they pulled this sort of crap to 4th year students that met the quota. I had a second year student start from scratch and end up with two papers on decent journals in 2 years. In any case, I think it is beneficial to the student to get first author publications and an important part of the training process to write a paper and learn to negotiate the review process and do not necessarily view this as exploitative as DM does.

      • DrugMonkey says:

        While I agree there is value in publishing, you are ignoring my key point is the hard and fast rules. Sugar scientist exhibits the classic case. Should have been passed much before.

        Grad school is *the* time for risking all on a project that may not work out- but PHD qualification should have nothing to do with success as measured by published papers. Publishable work, not published work. With allowances for a good job on a negative finding or a dry well. That is my standard.

        • ETD says:

          I can see the middle ground here for sure. Communicating/publishing is a critical step that has to be part of the learning process for students. But as you say, there needs to be allowance for time served well, negative results, and unlucky projects (not to mention crap projects due more to cruddy PIs then poor students).

          • namnezia says:

            Sometimes however, negative results might be due to lack of effort (sometimes I said) and it is committee's job to determine this. But from the point of view of other students, if someone graduates after 6 leisurely years with no published papers, folks who are busting their asses to get pubs will say this might devalue their PhD. So that's *something* to be said for fixed standards.

          • drugmonkey says:

            One graduate student lucks into a large, fast moving and very well-published laboratory. Is put on a can't-miss project where everything goes right and maybe ends up learning only a few techniques. She gets her required papers in 4-5 years.

            Another graduate student goes into a small, poorly funded laboratory and works her ass off developing assays on the cheap, finding all the blind ends and learning multiple novel approaches. Six years in and a single first author paper is only a potential possibility.

            Which "devalues" the rest of the cohort's PHD? I guess opinions would vary on that.

            My experience suggests that within one graduate program, and probably within on *laboratory*, the idea of trying to make sure everyone has had to suffer as much as the next student is laughable.

    • drugmonkey says:

      I'm glad I had to stick around extra time to get that first author paper - I'm positive I would not have gotten my F32 funding if I had no first author papers to list.

      While you are not wrong that it is better if one sails out of one's PhD program with many high IF first-author papers, this misses several critical points.

      The idea that qualification for a PhD is entangled in a direct way with enhancing relative competitiveness for future career goals is idiotic. This is not a trade school we are talking about. The PhD is a degree that is awarded for work completed in the doctoral program. Do we only award MDs to people who specialize in particular sub-areas of medical practice? or to those who are in the top X%ile of their medical class? Only award the JD to law school students who have done the right internships to have a shoo-in at a white shoe firm? no, we don't. The PhD, if anything, should be even LESS vocationally-directed than the JD and MD.

      If we require a certain number of publications, then this shifts the training toward getting publications rather than learning how to do science or following one's own ideas. At the training stage, I favor inculcating learning and personal initiative over the tactics and strategy of hitting publication count numbers.

      Requiring publications (and therefore extending what should be 4 yrs of training to 6+) stokes an arms race. More first-author, more JIF points, more total pubs. on and on. Where would it end? What bright line can we draw that says when enough is enough?

      It cannot escape your attention that the faculty involved in graduate training programs both define the terms of "sufficient work" and benefit tremendously from the underpaid labor of the graduate students. And let us face it, the 5, 6, 7th year graduate student is far more likely to be highly productive for that PI than is the 2nd year noob student. Especially with the increasing desperation to finish driving her to work 14 hr days, 7 days a week, without vacations. So training faculty are hopelessly in conflict here and cannot be trusted one inch when they start on about standards (that, btw, they never had to meet in their training programs) and the way it "has to be". They are the ones who can insist that a society journal pub or methods paper will never sully their lab and so the standard for a grad student is glam-or-bust.

      Which brings me to the ever ratcheting inducements in the training arc to CHEAT. To fake data, to falsify...to screw over labmates...to do whatever it takes to get a Baby Glam or a Glam first-author paper.

      • namnezia says:

        "Do we only award MDs to people who specialize in particular sub-areas of medical practice? or to those who are in the top X%ile of their medical class? "

        No, but your performance during med school does affect where you do your residency and what subspecialty program you are able to get into. So not all MD degrees are equal. Likewise, not all PhD degrees are equal. For some their performance as a PhD will, in many cases, determine the track they will follow later in their careers. If you come out of your PhD with 5 publications you will feel the long-term benefit of this down the line than if you produce no publications. But again, that is up to the student in question.

  • PinkGlitteryBrain says:

    No official requirements. Typical minimum is three chapters, where chapters = publications or publication-like. They don't have to be already published.

    • Travis says:

      No official requirements. Typical minimum is three chapters, where chapters = publications or publication-like. They don't have to be already published.

      I'm not sure if they are official or not, but this is what was expected in my program as well.

    • outoftune says:

      The same general guideline exists unofficially here in my department.

      (Current PhD student, engineering, in the Netherlands)

  • LM says:

    In my experience (physics department), there is no set publication threshold. Ideally, to earn a PhD, grad students should provide a contribution that could plausibly be credited as a first author paper (in some non-fraudulent journal). Whether the work is published like this (or combined with something different, or possibly not even published) is not as important. (A separate question is what is in the student's best interest if he/she wants to continue in academia.) In addition to this, I think grad students should have to deal with scientific peer review to some extent - if they don't publish a paper, they should be presenting their work at conferences.

    I don't think that this is actually as high a bar as it sounds. In the cases I knew where students graduated without publications, they certainly could have written up some of their minor results, but they (and their advisors) did not think it worth the time.

    I'm a postdoc.

  • shademar says:

    No publication requirements (also...no formal defense) but you have to show that you've contributed something novel to scientific progress. At least three thesis chapters, with two showing novel and potentially publishable/published work.

    Quals (2nd year) are much more stringent and tend to be the stage that washes people out.

  • ETD says:

    No set requirement in my program, was determined by the committee. Seemed to give some flexibility and leeway for the committee to be forgiving if student got a bad project/bad luck. However, it did leave some students feeling like goalposts moved too much from one student to the next.

    Similar program at same institution had a hard two pubs rule - by my observation they seemed to have more students master out and more people hang on for overly long degrees. Like Sugar Scientist described, some of these people were WAY more qualified than people who graduated years before them.

  • TG says:

    Our grad. school program required a minimum of 1 first author publication, although you were encouraged to have more. I've seen one student successfully defend with only a second author as he was nearing the 7 year limit they set.

    (As an aside, does anyone know how many institutions set a time limit to defend? I thought that was strange to have as you can't predict how well/bad a project may go.)

    I'm a postdoc.

    • ETD says:

      My institution had a kind of time limit. They actually put a sunset on required coursework - students had to get permission from the graduate school to extend the credits past 7 or 8 years or else repeat classes.

  • RX says:

    No official requirements for my PhD program, it's up to the PI.
    My lab is crazy. Here's the requirement: total first author impact factor: 30, total pages of paper: 20. The first graduate of my lab got 1 Neuron and 1 Nature Neuroscience paper. All the rest graduates tend to follow this pattern.

  • Dr Becca says:

    Where I went to grad school, it was very department specific. Some had paper requirements. Mine was one of the more lenient ones, and I'm pretty sure there were people who defended without any papers, but in the end it was up to each student's committee. At the time of my defense I had 1 high-ish IF 1st author published, one smaller 1st author in review, and data for another small 1st author that I put off getting published until a few years into my post-doc.

  • usymmons says:

    I am a postdoc since October this year. The institute where I did my PhD did not have a formal requirement (ie publications) to submit or defend your thesis. If you did have (I think) at least two papers (one first author) you were allowed to apply for a cumulative thesis (writing a short intro/summary of your work and attaching your papers), but I think they abolished this category now. Also, my institute really pushed to have students finish within 4 years (we're talking about the European system here). Afterwards, you were allowed to stay for a "bridging postdoc" position for another year. More specifically this meant that if you did not submit your thesis within 4 years there was no way your PI could pay you (knowing you'll have a problem paying your bills is a really good incentive, but trust me it's the kind of stress you don't want to have if you're anyway stressing about your thesis). Moreover, it really sucked, because a) to be eligible for the top grade at your defense you needed to have at least one first-author publication, b) if your project was ambitious and slow (eg I did old-school mouse genetics, making my own transgenics from scratch) you would stay on for an that entire year of bridging postdoc, and still leave without publications (or papers just submitted) - unfortunately you may only apply for many postdoc fellowships for 2 years after having completed your PhD and you need a first-author publication for virtually all fellowships.

  • I am a postdoc, graduated in 2011...

    There were no publication requirements when I started grad school, so those were the "rules" I was held to, however several years later, 2007-8ish they instituted a 1 first-author either published or submitted before defending rule. But they still gave committees a bit of flexibility to waive this if for example the student ended up on a big multi-author higher impact paper with a lot of their data as co-first or second or multiple mid-tier co-authored publications (in my field mid-tier defined as IF 5-10)...

    The policy seems reasonable to me especially since committees can take into account a student's goals in deciding whether to let them finish or make the stick it out if academia is their goal since I agree with others that obtaining fellowships or even postdoc offers in decent labs nowadays is dependent on publications as evidence of productivity.

  • miko says:

    PhD (small institute): "Expectation" of 2 first-author papers in "good" journals, with no defining beyond that, and ultimately decided by committee. I know of cases where not enforced, usually for good reasons (risky project, longer-term project). And yes, by good they meant JIF > x, with no agreed upon x.

    Postdoc (ILAF): No rules that I know of, mostly PI dependent I think.

    Current: I don't know of any requirements, too new to know unwritten expectations.

  • jmcin9 says:

    (Currently Postdoc) No set requirements in the department I graduated from, individual committees pretty much set the requirements. Mine wanted 1-1st author published, and 1-2 others at least submitted when I handed in my dissertation. Other middle author papers not directly related to my project didn't count for them. However several other students graduated without any papers, or 1-2 middle author only papers.

  • Marcel says:

    At the UdG, there are two ways to present a thesis: based on publications, or as a "monograph" format. See:


    If based on publications, then depending on the doctorate program there is a minimum number of papers.

    • Mum says:

      I like that the journal requirements are based on quartiles. It is still based in IF, but at least it transfers well between fields.

  • Zen Faulkes says:

    Career stage: now tenured, so my Ph.D. was long ago, in a distant land...

    I was required to write a thesis. That was it. There were no requirements it be a certain length, that a certain number of papers come from it, or any particular time frame.

    In my head, I think of a master's thesis as about equal to one paper, and a doctoral thesis as about equal to three papers.

  • DJMH says:

    No formal req but one first author pub was usually at least submitted by the time of defense. Definitely "pity" PhDs whose projects had gone nowhere but who had put in 6-7 yrs were routinely graduated. Often committee would ask about future plans...if you weren't pursuing academia it was accepted that the bar was a bit lower, I think.

    • drugmonkey says:

      Why is that a "pity" PhD, DJMH?

      • DJMH says:

        Just using the term I heard for it. But, frankly, I think that the advisor/committee/student has screwed up pretty badly if the student never gets a first author paper, no matter how low-IF the journal. Publication is part of the experience of understanding how to do science: how to pitch and present your data, and how to synthesize the literature.

        Main point being, the pub should be for the student's benefit. Advisor's benefit incidental. If you can keep those boundaries in mind, requiring one first-author paper doesn't seem like an unreasonable standard to me.

        • Bunbury says:

          "...if the student never gets a first author paper, no matter how low-IF the journal. "

          this is in principal true, but some PIs rather publish nothing than go for a lower IF journal. In that case you are in bad shape as a student.

          • DJMH says:

            Agreed, which is why it is the job of the committee to lean on the PI to publish in whatever journal works, and the responsibility of the department to be clear about this expectation.

  • mk says:

    We have no publication requirements, although our thesis should be considered publishable. We have a fairly small number of courses to take, so we can focus on our research. Requirements include a comprehensive which must be defended (options are either a review paper or a grant proposal, and must be complete by the end of our second year), a prospectus, and the final thesis. I think a typical length is around 4 years. Usually we just keep on going until we have enough data for our thesis.

  • eeke says:

    I got my PhD in the '90's. There was no written requirement for publications, but everyone who graduated from the program I was in had multiple first-authored papers. For someone to reach that stage without having at least a submitted manuscript for publication is partly a failure of the mentor. I think at least one publication should be required, and if it looks as though the student's project is failing, it should be caught by the thesis committee or supervisor early on. To allow a student to flail about for 5+ years with no papers is inexcusable.

  • odyssey says:

    Old graybeard type.*

    The dept I'm in requires a first authorship manuscript published, in press, or at least submitted. I disagree with that policy for the reasons DrugMonkey has outlined above. Aside from that, the dissertation committee has to agree the student is ready to graduate. The publication requirement came about because we had a couple of committees graduate students who shouldn't have been. Really shouldn't have been. Sadly it's easier to enforce a publication requirement than corral slackass faculty...

    * Tenured full prof.

  • Jim Thomerson says:

    My PhD was in 1965. No requirement for papers. I was lauded for getting my dissertation in publishable shape the summer after I received the PhD. I later published a small paper on a side study I did whilst working on the PhD.

    I went directly to Assistant Professor tenure track, and had six publications of various natures my first year, and a major non dissertation publication the next year. I received early tenure and promotion after three years. Secret is to get a job at a small regional start up university.

  • Dave says:

    In the UK (top 10 school), there was no formal requirement and I would say it was split 50:50 between those that had at least 1 first author vs. those that had nothing going in to the final stretch. The thesis and viva are the primary basis of assessment still. I had two firsts in the three years that I was studying - one IF>4 and one IF>6 - and several co-author papers in journals with similar IFs. But I was publicly roasted in my viva voce for over 6 h (with a lunch break!), so I'm not sure what difference it made to the outcome of my PhD examination.

    But......it absolutely mattered when it came time to find a post-doc!

  • dr24hours says:

    Mine had no official requirements. I had a single first-author paper in press with a respectable mid-tier journal.

  • […] has initiated an interesting conversation on the criteria for awarding a PhD in the sciences. A commenter over there alleged a set of rules […]

  • AJ says:

    I'm a postdoc with a PhD date of last fall.

    My PhD department also had no hard stipulations, though that tended to vary between individual labs. We did have a hard deadline of 7 years, though, where if you finished your seventh year and still hadn't graduated you had to retake all your classes. Needless to say, nobody stayed around that long.

    I also have to agree with DrugMonkey re: hard manuscript guidelines being exploitative. I took 5.5 years to finish and currently have a second-author and a first-author ready for submission and waiting on my pokey former PI. My project turned out to be nightmarish, as trying to determine the structure of an unstructured protein is plain old masochism at its finest. However, the "what doesn't kill you makes you stronger" adage holds true and I've become great at troubleshooting...and also have a dissertation full of futile attempts. I don't think my lack of first-author papers makes me any less of a PhD level scientist, just a less lucky one. It has been a disadvantage when applying for F32 funding, but everything including your hair color and shoe size is a disadvantage when funding success rates are in the single digits.

  • AA says:

    Is the min requirement norm to have 1-2 first author pubs only? I used to work for a guy in my undergraduate and he required at least 1 first author pub per year, medium-IF rating (i.e. best of sub-sub-field journal). It's not an unreasonable standard.

    No offense, but how can you spend 7 years in grad school and not get at least one pub? Yes, projects don't work, things screw up. Ok so perhaps you were working on a high risk CNS-level project that failed... it's still your fault for putting all eggs into one basket. Isn't learning how to anticipate/evaluate risk level of projects and plan contingencies part of the PhD process? If you know you are doing a high risk high reward project, you should get at least 1-2 safe side projects, so you don't end up with *nothing*!!!

  • bsci says:

    It was lab-specific in my program, but the department did have oversight to reduce abuses. The oversight was that a student's committee (including primary mentor) needed to sign off on the steps needed to complete the degree. This set clear expectations and limited the PI's ability to set unreasonable goals or to later shift the goal posts.

    The standard in the lab where I got my PhD was 3 separate publication quality projects. I think everyone I know got at least one submitted and other published before graduation.

  • Matt says:

    At my Uni (UK, Russell Group) there were no formal requirements for publications or anything like that. The time limit was four years if you were full-time. Most people were funded for 3, so all data collection would have to be done by then. The standard was that the work should be publishable, not necessarily publish. Course, it helped if you were published already, because then it's hard to argue that it wasn't publishable work. Most people would have one or two papers out. I don't think where they were mattered. I'm a post-doc now.

  • Anka says:

    I'm a first year postdoc. My school's rule was one first author paper accepted (not necessarily published). I know people who actually defended but couldn't get their degree until they'd actually gotten that acceptance notice.

    However, standards vary widely by lab and committee. In my lab, the general rule was 3 first-author papers, however I know someone who got out with two. I had 3 first-author papers in my fourth year, but was not allowed out because "I was only a fourth year, and my work had potential for a higher tier journal on the next paper." So, I ended up getting out in my fifth year with 4 first author papers, as well as one more first author paper and one second author paper in prep.

    For comparison, another person in my lab and in my year got out in his 5th year with one published first-author paper in a good journal and one first author paper in prep. He was super lazy, and my boss just wanted to shuffle his butt out of the lab because he didn't want to keep paying him, even though normally he would make the guy hold out for a 5th paper. (Needless to say, I found this rather enraging).

  • Junior Postdoc says:

    There was no formal requirement in my program, but the majority of students have at least one first author paper in a good journal when they graduate (e.g. J Neurosci). This generally seems to be the unwritten rule for most committees and the head of the program.

    I actually wonder if a formal minimum paper requirement would make for a more reasonable standard for graduation and protect students from the drawbacks of their PI's CNS worship. It is very hard to convince my grad school advisor to publish anything with an IF below PNAS lest he be spurned from the cool kid table at the next HHMI meeting.

    • ETD says:

      I don't think formal minimum would help protect students if PI was only chasing IF. Really good friend is in 8th year of program that requires two 1st author. Had a project ready 4 years ago that would have provided 1st paper, but PI wouldn't pub because he didn't think it had high enough impact potential. Paper has since been scooped so he can't go back and publish the work, and my friend is still spinning around trying to meet the minimum pub requirement.

  • As a current student at an R1:
    There's a 3-week comprehensive exam. You write a mock grant proposal based off of one of three broad areas related to your field of work. For instance, if someone was looking at a typically developing population, a prose may ask about proposing a study that includes a specific atypical population. That exam is then reviewed by your 5-person committee and a recommendation of your PhD candidacy is made.

    Then there is an oral proposal of your dissertation (or third 1st author publication if you choose to defend three related 1st author publications) followed some time later by a defense. Then depending on your defense and work up to the defense, you are bestowed the PhD title.

    This is pretty normal. However, the relationship between advisor and student here is pretty close -- no lab rotations and collaborations either come from advisor recommendations or your own. My advisor is particular in setting us up for post docs in some specific labs we have history with (good labs if you're into developmental cognition). But that's supplementary to the degree (we think -- no one has deviated until this year.)

  • Besides the number of papers requirement (which wasn't a thing in my department, although 2 or 3 papers were pretty typical), did anyone else actually read their university's guide as a student to see what was formally required?

    I ran into a lot of old rules that were no longer enforced -- for example, at the University of Illinois supposedly all PhDs (even in the sciences) were supposed to demonstrate at least reading proficiency in at least one foreign language -- the idea being that reading papers in French and German was an important skill (and before WWII it was, I suppose).

    • DrugMonkey says:

      I require my trainees to have proficiency in reading Chinese journals, JB. Get with the times!

      • This is actually something that would be useful! Or alternatively, I wonder, given the number of Chinese nationals who are grad students in the US, if anyone asks them to do journal club presentations on notable research published in Chinese for the benefit of the non-Sinophones.

    • namnezia says:

      I think my uni recently got rid of the foreign language requirement.

    • atcgphd says:

      My PhD granting institution got rid of that foreign language rule circa 2000 - yes that's right, 2000 - and not without some resistance. An astonishing number of old fogeys maintained it STILL might be necessary, at some point, to read those French/German/Russian journals ... and THEN what would the poor student do?

      No mention whatsoever of Chinese journals. Talk about out of touch.

  • Jim Thomerson says:

    I used Spanish for my MS, German and French for my PhD. We had a superstar type who failed the French exam 14 times. The rules were changed about then to reduce the foreign language requirements.

    I read my PhD institution catalog and found that a laboratory course in organic chemistry was expected of entering grad students in my future department. It my MS institution I found and took a three hour with lab "Organic Chemistry for Agriculture Majors". In initial advisement, I was asked, "Well, young man, have you had a laboratory course in organic chemistry?" "Yes Sir, I surely have!" and here I am, an Emeritus Professor.

  • phagenista says:

    Assistant Professor, R1

    No formal requirements in my program. I agree with the three publishable (not necessarily submitted/published) units rule of thumb, and with giving allowances for failed experiments, political instability or natural disasters in a field site, etc.

    As near as I can tell, the younger a scientist is in their career, the higher the bar is for the dissertation. Older faculty know they passed that guy 5 years ago, 10 years ago with something /even worse/ than what the grad students or postdocs think was the pity PhD they awarded in the (hopefully short period of time) you've been in the program. I voted no on a very thin defense recently, and the rest of the committee, all more senior than I, voted in favor even though they agreed with all of my criticisms.

  • GMP says:

    In my opinion, a well-supervised PhD student in STEM fields will be able to publish as first author. That means that, if you have them on a high-risk project, it will be balanced out by bread-and-butter ones for safety. I require 3 papers from the thesis for graduation as the minimum (this is not unreasonable in my field), with the understanding that people wanting to go into academia should have more.

    Btw, we had a heated discussion in the comments over at my place about 2.5 yrs ago on pretty much the same topic
    None and We're Done. There was a parallel thread over at DM's Science Blogs place that I can't find now, where things went a little heated in the comments; I think this one is related The Essential Character of the PhD

  • Ola says:

    PhD in '97 in Europe. No papers upon completion of lab work, but 3 written up alongside thesis and published 6 mo. to 1 yr. after graduation (all in mid rank journals, 1st author, cumulative impact factor about 20). This was back in the day when the stipend covered 3 yrs. of work. So, the boss squeezed as much lab work as possible out of you, the you had to write up on your own dime (i.e. write all day, bus tables at night, tend bar on weekends). This resulted in large delays between formally "finishing" lab work and being thrown out of the lab, and actually graduating - in my case 9 months. I actually started a post-doc before my formal PhD oral exam - had to take a day off work to travel back for it. I guess it was just assumed I would pass. My first paper from post-doc (rushed out the door to support a grant app) was actually my 2nd paper overall, then followed by the other 2 papers from the thesis work.

    Oh my how things have changed.

  • thankascientist says:

    at my PhD institution/department, the rule was you had to have "made a significant contribution to knowledge, at least part of which is of a quality and quantity worthy of at least two publications." I don't necessarily agree with the "at least two" criterion, because I think a single paper (or contribution worthy thereof) can be a significant contribution to knowledge. The idea was to allow those who were scooped in the later stages of their PhD to graduate without starting over-- a policy I agree with in practice, in part because I've seen people benefit from it and go on to do great things.

    I'm a relatively recent postdoc, graduated in 2012.

  • LincolnX says:

    Our programs require two first author papers in a "publishable" format, wrapped in a scholarly Introduction and a Discussion that integrates the findings in the context of the larger field.

    Our dissertation chair is chosen from outside of the degree granting department so that the advisor, while still important, is not the only word on when enough is enough. In fact I've seen a couple of instances where this structure has helped nudge a student toward finishing up when the advisor was a little reluctant.

    After having mentored over a dozen students, my limited observation is that those who take it to the step of actual publication (and who write NRSA or NSF-style fellowships) are far more successful in getting tenure track positions than those who do not.

    While the more cynical might portray this as "trade school" mentality, I think it is an appropriate apprenticeship practice that promotes the best chance for success (and yes, this is a value statement but I do think we should try for that). More than that, the students realize this and want more opportunities to publish and write. With what I just went through to get my competitive renewal, in my view an emphasis on publishing is not misplaced.

    But yeah, requiring specific impact journals is bullshit, elitist and exploitative.

    • Mum says:

      I wonder if all of these places with "submitted" requirements are, at least in part, to blame for some of the very shitty papers I have been asked to review. Every now and then I get some papers that you truly wonder if the authors are serious.

      Perhaps they allow submissions because of the ridiculous time for review and crap reviews we sometimes get. You don't want to hold someone's degree over some asinine demand by Reviewer #3 (often while the student is required to keep paying tuition). But still, submitted could mean nothing at all.

  • qaz says:

    We just added a requirement of at least one "real" first-author paper submitted. We added this to the requirements after NIH yelled at us about one of our training grants. We made it submitted so papers could be under review at Glam mag. Thesis committee to determine that paper is real and not just random bogosity. It was easier to add the requirement than to corrall slack profs.

    • qaz says:

      Note - pub doesn't need to be submitted to GlamMag. Just needs to be a real paper.

    • DrugMonkey says:

      Could you expand on what sort of criticism of your training grant this move addressed?

      • qaz says:

        The fact that our students "weren't publishing enough first author papers in fancy enough journals".

        We'd had a couple of students graduate without pubs. (At least one I know was let out by a slack PI and easy committee, but I think there were others who wanted to go into fields that didn't need pubs. [e.g. not academia])

        We had a site visit and the funny thing is that one of the people who complained said that his students didn't publish in GlamMagz either, but he felt ours should...

        After a long discussion in the program on receiving the comments, we settled on one "real" pub submitted. We said "real" to prevent bogus journals and abstracts from counting. (Basically, if the journal is pubmed or the equivalent, we were OK with it.). We also said submitted because we didn't want to drive a student/PI pair to submit to PLoSONE when they had a shot at N/S. We also put in place new stricter membership rules to prevent slack committees.

        At this point, it seems to be toughening up our students, which is a good thing. I think the real questions are going to come whether we really won't graduate someone who takes N years and doesn't have a pub yet. I'm not sure if the program has the guts to be that nasty. Certainly, the younger students are getting more "publication" focused.

  • Ella says:

    We didn't have a formal requirement in our department. It seems like the longer you were around the more likely you were to leave without papers - not so much a 'pity PhD' as a 'we don't really want to pay you any more and we're not sure how else to get you out the door'.

    In our lab the unspoken rule was you'd have a first-author submitted somewhere by the time of your defense. Which is totally reasonable given your thesis should be in publishable shape anyways - if you weren't scooped in the process. Most faculty were more concerned about the contribution and quality of your work to the field than number/IF of pubs.

    First year postdoc.

  • laurakat2 says:

    Current Graduate student in bioinformatics/genetics.

    We are required to have one first author publication (no IF restriction). For the field this an exceptionally low bar. We routinely have students exit rotations with a publication (or at least co-author). I don't know if it's just because LPUs are popular in the field, or just that our flavor of computational work can be fast.

    Most graduate with at least 4-5 first author pubs, and numbers of total publications in the 10-25 region are fairly common.

  • AA says:

    With all the talk of only 1-3 pubs before graduation, I'm wondering if this is field specific?

    In my field, newly hired TT profs have at least 10 pubs from grad school and postdoc. To be competitive you need about 10-20 pubs, and over 20 pubs are not unheard of.

    So seriously, you think 1-3 pubs from one's PhD is enough to compete in the market today?

  • Icee says:

    We don't have any formal requirements. Any informal ones are left to the PI but there are people who graduate (masters and PhD) with dreadful theses. My intellectual understanding greatly increased from writing my dissertation as a monograph but my advisor and some others have not made publishing a priority. Depending on career goals, that can be really bad for students and when there's NO emphasis on publication it can hamstring our future opportunities.

    I like leaving the decisions in the hands of the committee rather than journal reviewers, but to have a very strongly worded list of "suggestions" for papers submitted/published would have been nice, and would give some PIs the kick in the pants they need to shepherd the pubs out the door instead of letting manuscripts languish on their desks for a year, and exacerbating their students' imposter syndromes.

    I have been a postdoc for 6 mos and I'm working to the bone to get all my PhD papers published within 1 year of graduation (3 of 7 published, 2 submitted, 2 to submit before Feb. Ugh.) I had NO first-author pubs even submitted when I defended (I had one bazillionth-author pub from a rotation).

    Also I'll be 2nd auth for a Master's student who is writing up now. She already told me that she won't put any effort in to publishing beyond writing her thesis because our advisor hasn't been prompt with edits. She said "I'm going to med school so I don't care if I get papers. If he wants it published, he can write it". Guess who's gonna end up getting it into publishable shape: me, because I want the work to get published and I want the paper to be decent. This is where some higher mores would be beneficial.

  • Jim Thomerson says:

    In my particular little subfield, I needed to read papers in French, German, Spanish, and Portuguese. Which I could do with papers in that area.

  • Dr. R says:

    My program (Microbiology in the US) had strong recommendation (but not absolute requirements) for 2 first author publications and 1 presenation (could be a poster) at a national or international conference. Honestly, I thought these were great guidelines. There are obvious exceptions (students stuck with an unlucky project or a bad PI), but in general I think you are doing a disservice to the student to ask them to leave and be competitive for post-docs or funding without first author publications.

  • Just defended my dissertation last month.

    my institute is very case by case. One person had 4 or 5 first author papers, and another from the same lab had 0-1. Their goals were different: research academia vs teaching only. I think the mentor and student should have a good long talk about future goals and go from there. I want to go into research academia, so i cared a lot about getting a good postdoc position. So for me 'real' stuff came first, then the degree. For example, I wrote and submitted an NRSA F31 first, then defended that as my proposal.
    I took 7.5 years to finish, but that is basically average at my institute, and I have 2 1st auth published, and 2 1st auth written, an encyclopedia article and a review paper written. Plus 3 middle author papers. My strategy worked pretty well as I got called for interviews each place I applied including one that wasn't even hiring and I am starting a good postdoc position in January.
    but to expect this from someone who does not want to go into research academia is ridiculous. It would be great for a PI to have a yearly 'what are your goals?' meeting with each grad student to prevent exploitation and wasted time.

  • ### says:

    Communicating our discoveries is an essential part of being a scientist, and writing and revising papers should be an essential part of a Ph.D. education. I appreciate the idea of not having hard quotas to accommodate the unlucky-but-hardworking student, but those exceptions should be rare, and possibly even require a approval by the DGP or someone besides the advisor.

    In my graduate program, the standards slipped and more and more students graduated with "publishable" chapters that then never were published. This was due to student visa term limits, to the availability of industry jobs in bioinformatics, and to ambivalent PIs. In addition to not properly educating those graduates, it led to resentment from the other students and affected the morale and reputation of the program.

  • […] awarding PhDs. But I only realized how extended the spectrum of variability is when I came across a survey by Namnezia a couple of months ago. Apparently, some universities strictly require one or more (first author) […]

Leave a Reply