Tuesday, March 29, 2011

Pushing Out Papers While a Pushover

Because I am trying to get as many papers out as possible before I deliver about 3 months from now, I have been in intense writing mode and getting increasingly angry with myself for making the same mistake I have seen many of my colleagues make:

Letting the student graduate before all the papers that we have agreed on are written up and submitted.

It's always the same -- there is a job opportunity, or some other time-sensitive extraneous impetus, so the student must graduate sooner rather than later. The student pleads and swears to high heaven that they will wrap up all the still-looming papers afterwards, if only you let them finish. And then they graduate and the papers never happen.

My colleague across the hall has lost a number of papers like that: the student leaves, stops keeping contact, and the papers are never even drafted. With a close collaborator a similar situation happened recently: a student received a nice job offer from a big company, with one half-written paper still in the works. The student graduated and left months ago, and has since been completely incommunicado. We have dropped the project altogether, as the collaborator and the new student haven't been able to make sense of the data left behind (what was measured and what not, what exact conditions etc.) A lot of work is simply lost because the collaborator does not have the time or the money to repeat everything with the new student; the experimental data is quite interesting and somewhat counterintuitive. My postdoc and I have been working on the theory, which we will also drop since there now isn't sufficient experimental data to confirm that the counterintuitive phenomenon is indeed happening and is not an artifact.

Although I promised myself I would never do that to myself -- let the student graduate before all his/her obligations to the group have been fulfilled (the papers we have agreed on are written up and submitted), it turns out I am as much of a pushover as the next faculty, if not more. I am facing a similar situation with two of my group members, one former student but still here, postdoc(k)ing temporarily, one still a student.

I let the temp postdoc graduate at the end of 2010 because we figured a couple of months would not mean much, and graduating in 2010 (sooner) looks better on his CV than 2011 (later). He was going to stay as a temporary and part-time postdoc while interviewing for jobs and wrapping up two more papers. Instead, in the 3 months he's been here after the PhD, all he's been doing is cramming for his interviews, going on interviews, and he started training to do experiments with some of my experimental collaborators. I have been paying him this entire time. He only just gave me a pathetic draft -- unworthy of a second-year grad student, let alone someone experienced in writing papers -- of what's supposed to be the crown jewel paper from his thesis, which clearly demonstrates he doesn't give a rat's ass about it any more. Thank you, GMP, for having been my advisor and paying me as an RA for 5 years; now that I have the PhD, even though you are still paying me, I thought this might be a good time to spit in your face.

The other, current graduate student just came back from the APS March Meeting last week, where he spoke with his desired future postdoc advisor, who said he'd take the student but they should make it ASAP as he's got some flex funds expiring in the fall two years from now. So now the student, who has been taking lots of classes, more than what he needed and for longer than he needed to in order to get an MS in a "purer" discipline along the way, all of a sudden wants to crank out 3 papers by the end of the year (yeah, like that's gonna happen with his writing speed) and go do the postdoc. I am completely furious. I am sorry, but I am under the impression that if I pay you as an RA for years, the least you can do is have some time after you are done with classses to do the goddamn research and produce some science that the nice funding agency paying you can show for all the money it expended on you. But that's just me being deluded, right? The only way all these papers are going to happen is if I write them from scratch and he's permitted to work on loose ends at his new position. What's the likelihood of the latter happening?

I know there is an overwhelming sentiment in the blogosphere that faculty are universally bad to students, selfish, unyielding, providing insufficient guidance and oversight, alternatively providing too much oversight/micromanaging/stifling students, or being unaccommodating of the students desires to do this, that, or the other.

But the longer I am faculty, the more cynical I am forced to become. Graduate students can be really quite selfish, too: they are there to forward their own career goals, often willfully ignorant of who actually pays their stipend, that those stipends are not prizes for their awesomeness, but rather money from the federal government for which the advisor's group has to do a certain amount of work. They ignore what goes into acquiring said money, that it's not anybody's God-given right, and that a failure to produce papers damages their group's prospects for doing science in the future.

When the advisor is unyielding about graduation and holds it contingent on certain papers being ready, the advisor is perceived to be a selfish tyrant. Well, when the student leaves all the work behind and never look back after having promised to complete it, I fear the student is a selfish liar.

The PhD advising relationship is a symbiotic one: it's supposed to benefit both the advisor's group and the student's career goals, and it assumes everyone involved is an adult. It's not a parent-child relationship, where the child (student) can count on unconditional love and support no matter how self-centered they are. Failure to produce papers -- and I assure you I am not a GlamourMag-obsessed maniac -- is indeed a grave one, for everyone involved. The work does not, in fact, exist until it has been published in a peer-reviewed journal.

Dear faculty colleagues, yes, you can indeed be too much of a pushover: you want to do the right thing and be fair and accommodating of the students' future, but then they end up screwing you over. I, for once, am now considerably less likely to accommodate future requests for great flexibility in graduation dates. I cannot afford falling out of grace with program officers because I am not delivering on promised papers. And I will not keep letting other people down, whose CV's and/or future job prospects may also depend heavily on the work that never gets published. Feel free to call me a tyrant. I consider it a case of "Once bitten, twice shy."


Anonymous said...

You are SO right, GMP! I have 2 manuscripts on my desk that I just can't get out for similar reasons. The problem is if I try to get students to stay until they are done, I have to keeping paying their salary/tuition. I'd be happy to do this if they were doing what needed to be done, but if they continue to procrastinate and don't finish the paper, it's just money down the drain. A colleague of mine who is a terrific scientist and excellent mentor (in my opinion) did the following: s/he sat down with the 5th year student and said "how long do you need to get this paper finished?". Student said 2 more semesters, so s/he agreed to pay tuition & stipend for 2 more semesters on the condition that the student finish the paper and that s/he would not give permission to graduate unless the paper was submitted. Two semesters went by, the student did all kinds of stuff other than science and (surprise!) the work was not done. The student did not graduate on time and had to pay tuition for an extra semester out of their own pocket. It sounds harsh, but there has to be some accountability.

Namnezia said...

If someone is leaving the lab with unfinished papers, I request they:

1. Leave all their data analyzed and compiled in one place in an easy to read format.
2. Make final figures.
3. Write a detailed methods section.

From that point I can write the paper if I have to.

Miss MSE said...

The students to most recently leave my group have left with papers unsubmitted. However, usually this was because said paper drafts had been sitting on our PI's desk for months *before* their defense or scheduled departure date. It needs to be a two-way street: the student needs to produce drafts in a timely fashion, but without feedback, the papers are going nowhere fast.

GMP said...

Anon at 11:57, I hear you. The financial aspect of student procrastination is getting ever more detrimental to a group's operation with less federal money available for everyone and getting harder to get; as you said, you cannot just throw money down the drain, waiting for them to be done at snail pace. What your colleague did does sound harsh, but I am sure he/she has come to realize that's the only way.

Namnezia, I think many people (me included) nominally request the same things as you. But once the student leaves and you get to writing (and actually look at the data, figures, and the methods section in great detail), more often than not there still need to be substantial revisions, figure changes, and rewrites even to the technical section, and often there is a need for additional or repeated experiments/calculations (at submission and/or resubmission). These issues can be particularly bad on collaborative papers, where each figure and the text undergo multiple changes before they are truly final, and the writing can take many months. [For instance, I have a collaborative (3 PI's, 8 authors total) Nature Progeny paper in the works, which we've been writing for nearly a year now, it is on its 8th or 9th set of "final figures".] I am happy if the requirements above have worked for you 100% of the time, but in my experience they are usually sorely insufficient.

@Miss MSE: It needs to be a two-way street: the student needs to produce drafts in a timely fashion, but without feedback, the papers are going nowhere fast
Agreed. Still, I find that most students are happy to have the draft just sit with the advisor without so much as a nudge (guilt-trips work on advisors too). I know I was constantly bugging my advisor until he'd return the corrections; I try to be prompt about returning them myself, even more so if a student shows interest in getting them back.

Bashir said...

Huh. Are these students leaving for academic or industry jobs? Why would those still in academia leave potential papers undone? Perhaps I am missing something, I think it's pretty common in my area to have things to finish up post graduation. I've never heard of anyone dropping papers unless they left academia.

Ragamuffin said...

I don't really get why the post docs/students wouldn't want to finish writing their own papers even after they leave. It is obviously beneficial to them as well, especially if it results in first-authorship. I am leaving my boss of three years to begin graduate school, and there will be one last unfinished manuscript before I leave which I will finish writing from my New Destination. Symbiosis.

JaneB said...

You are so right. I have a LOT of trouble with this, the more so as the UK system does not require that a single paper be in a publishable state before the student graduates... it has to be two-way, but the students so often fail to deliver.... even my "biggest successes", who are both faculty members now, still have papers joint with me from their PhDs which they 'might get to one day'. What a waste of my time, energy and money! Some days I really would rather pay for technicians, at least it'd be clear what I could expect!

Anonymous said...

A professor at my grad institution has the same problem. Her solution is to tell grad students that they relinquish 1st-authorship if they leave without finishing the paper. That is, they become 2nd or 3rd author, depending on how many other lab members step up to finish the paper. But of course, that only motivates people who are staying in academia, and even then, only to a point.

GMP said...

Bashir, Ragamuffin, a number of my students go into industry. Even those who go into academia, there is great variability in terms of how much the postdoc advisor allows them to work on papers from the PhD (postdocs are typically 2 years, 3 tops, so there is the expectation that you'll ramp up quickly and start cranking out papers for your new boss ASAP)...

JaneB, I hear you. It's a source of major frustration for me too.

Anon at 3:37, thanks for the comment, that's a very elegant solution! But as you say, I am sure it works only to a point though...

Gears said...

I'm with MissMSE on this one. Sure, the student should finish up the papers but I've seen more cases of terrible advisors that are too busy for their students than grad students unwilling to do yet even more work for next to no salary.

I haven't spoken to my PhD advisor in at least 6 months and the last time was to get his signature for my thesis. It's been over 18 months (if not more) since we've last had a technical conversation. I've found I'm better off sending him an email and saying "If I don't have a response from you within 2 weeks, I'm submitting it anyway." I don't get any responses and I submit it anyway.

And, I'm in a 4 month transition period right now (mini-PD before TT position starts). I've managed to get one paper accepted, 2 others are submitted, written 4 conference proceedings abstracts, and 2 whole proceedings. Yet I've had no discussions with my advisors about this but they still get credit because their name is on it.

If a student hasn't published anything, then it's ok for an advisor to be strict to a point. But if they've published 3-5 papers (or whatever is average for the field), well then I guess that student is just average. But an advisor holding a student's graduation over their head because they haven't published an above-average amount is an asshole advisor move.

GMP said...

It's been over 18 months (if not more) since we've last had a technical conversation.
Yet I've had no discussions with my advisors about this but they still get credit because their name is on it.

Gears, I am sorry about this, but I don't think most advisors are like that. On the upside, you have a TT position lined up, so I presume your advisor is not just busy but famous? If that's the case, it sounds like tit for tat: your advisor gets some credit on papers without directly advising you, and you get some of his reflected glory which can propel you you in the academic world well beyond your merit alone. Moreover, at least your advisor is not in the way of you submitting and publishing: I have collaborators who insist that everything be written "just so," yet don't have the courtesy to return corrections on collaborative papers for many months -- they are essentially keeping each paper hostage, and I don't even work for them, we're supposed to be peers! And your advisor is paying you as a postdoc in the transitional period, so he doesn't sound like pure evil to me... So I'd say count your blessings, things could be much worse.

Regarding how much students should publish: I think the advisor and a student have to have a talk well before the defense (as in, 1+ years before) regarding what else needs to be done for the student to finish. That includes discussing provisionally which papers still need to get out, and these can be discussed periodically. When the student is done with the work agreed upon, he/she should be able to get out.

About the number of papers: you cannot expect all students to get out with a minimal number of papers. Some projects take a longer ramp-up time, some are conducive to a faster publishing rate. It's about a coherent chunk of work being done within a reasonable time and with the available money. I set a minimum of 3 first-author journal papers from the dissertation work for graduation, but a good student (and particularly an ambitious one!) should publish more (within the same time as his peers) because otherwise they are really not competitive once they leave. And extraneous conditions also come to play such as how the funded project has been performing up to then, who else in the group is working on it etc.

Bottom line is: there are asshole advisors, but not as many as it may seem to students -- not all nonideal advisors are really assholes. Your perspective on what constitutes an ideal advisor (or an asshole) is bound to change after a few years as a faculty.

DrugMonkey said...

Straight up, GMP, Straight up

Ace said...

I came here from twitter re: "you have a TT position lined up, so I presume your advisor is not just busy but famous?"

I don't think GMP would mean to say someone can get a TT job *only* bc the advisor is famous. Advisor definitely has an influence on your career but only if you've a really good CV to begin with.

It is of course possible that advisors have different issues for being late to work on papers or not do it at all. My Phd advsior was notoriously absentminded and disorganized. He also did not let me submit papers without working on them, but yet never got back to me in a reasonable time frame (several months). I had to hound him physically to get work done. He did have shortcomings as an advisor, but he was not an asshole. He had different, excellent qualities as an advisor. Noone is perfect.

The silver lining is, if you could do good work with a non-hands-on advisor, you will find the transition to PI-dom easier.

Dr. Girlfriend said...

OMG that makes me so angry because I would have loved a PI who was more publication focused! I wish you were my PI.

I guess it does not make you feel any better to know that there are students who are pushing their unambitious PIs to get something (anything) research published.

The dissertation is supposed to consist of publishable material. Ideally they should be reformatted published papers, but in certain circumstances (being scooped; extra projects) I believe they can be unpublished. I also think a student should be able to graduate without having submitted the papers BUT (big BUT) they should have at least have written their work up in their dissertation.

It should be relatively easy to extract a paper from a PhD dissertation chapter. If a student does not have enough chapters in the dissertation the committee should not allow them to graduate - period.

I graduated (reluctantly) with unpublished papers, but my dissertation was already long enough. I think this is acceptable, plus I wrote those chapters in the hope that they would get tidied up and published! Even though I am not in research I still want those papers...grr

Anonymous said...

I am a grad student now. Before that, I was an RA. Before leaving my RA job, there were some stuff to be done on a paper. Unfortunately, just before leaving, my boss verbalized her thought that it is my fault that our paper was delayed. Other strong words were said, making me understand that, for right or wrong reason, this woman will never write a high-spirited letter of recommendation for me. So I just put everything together on a computer and said, here's you draft, here's your processed data, here's your raw data, here's the program in case you can find somebody to generate more data. I never wrote to her. I feel sorry, because she seemed nice until then. But I am nobody's dispensable tool, and I did not emigrate to have people play with my future. Nothing was published in the three years since I left. It tells me that my work was difficult enough, that nobody could pick it up, or even start it from scratch. I would never pick up where I dropped it, although it would be a great paper from a famous, Nobel-level PI. It just doesn't work this way.

Gears said...

GMP, Actually, I would understand it if my advisor was a rockstar professor. However, that is so far removed from the case, it's laughable.

And yes, when I needed Prof's signature on my thesis/graduation I got it. Also, whatever was written in those letters of recommendation didn't hurt my chances for a tenure track (especially give that I got one). So I basically got what I needed most. But his research group got a lot out of me too.

Currently, I have twice as many journal publications as the other ~10 PhD students/Postdocs in the group COMBINED! You'd figured if one of your students asked for support and/or mentoring, and they were producing stuff for you, you'd be more than willing to accommodate s/he.

And yes, I have seen your side of the argument. I currently have a MS student who has graduated and is supposed to write a paper on work that I advised. He's in industry now and has limited time to work on it. Would I like him to do it? Yes. Do I expect him to do it at this point? No. On a tenure track, one paper can make your career early on. But if the lack of one paper breaks your career, especially when you're a tenured faculty member, you've got other problems than that.

I would like to ask a more difficult question (basically the same thing every PhD student thinks). Is the reason you can't/won't write the papers because you don't know enough about the research? Because I know that's the case with my Prof. There's no chance in hell any of those papers gets published, even with all the time in the world, all of the data, notes, and lab notebooks, and all the figures made. If that's the case, then pay attention to their research more and you won't have that problem. I could write my MS student's paper for him if I was that desperate.

Alice said...

Gears, it just doesn't sound like your prof cares. I know what you mean as I knew profs like this too. But in the end, your work is your work. PIs always get on papers. You sound like you have some resentment over it and it's understandable. But it's also useless. What's done is done. Good for you to have done all this in your circumstances. After a while, when you move on to your fancy (crazy!) TT job, eventually you will not care. Good luck!

GMP said...

And yes, when I needed Prof's signature on my thesis/graduation I got it. Also, whatever was written in those letters of recommendation didn't hurt my chances for a tenure track (especially give that I got one). So I basically got what I needed most. But his research group got a lot out of me too.

Currently, I have twice as many journal publications as the other ~10 PhD students/Postdocs in the group COMBINED! You'd figured if one of your students asked for support and/or mentoring, and they were producing stuff for you, you'd be more than willing to accommodate s/he.

Gears, no one is questioning your technical excellence (a courtesy you do not extend to others in your comment). I had a baby in my first semester of graduate school, yet also outpublished everyone in my group by a factor of 2-3 and finished in a little over 4 years. I am very grateful to my advisor for giving me the freedom to pursue the topics I wanted so I could be very productive, and I am happy that he had the funding to support me through my varied endeavors. I learned a lot from him, even if we didn't discuss the nitty-gritty details of my day-to-day calculations; I have great respect from him.

In contrast, you seem to consider your advisor a worthless ignoramus. First, there is no way a worthless ignoramus can sustain a 10-person group. Second, if your advisor truly were a nobody who commanded no respect in your field, as you seem to think, his recommendation would mean squat during your job search. It seems like you have a lot to be grateful for: someone who paid you and gave you the freedom to be so productive, who wrote a good letter for you, and is still paying you while you a waiting to start a TT position.

I would like to ask a more difficult question (basically the same thing every PhD student thinks). Is the reason you can't/won't write the papers because you don't know enough about the research?

Who says I can't or won't write the papers myself? That's exactly what I do when it's clear the student is not getting anywhere or has left already without finishing papers. However, when the student is here, holding raw data and figures hostage, and has promised over and over to deliver a first draft, then yes I expect a first draft from him. I am very involved (as in -- very invasive in terms of writing/edits) in every paper that gets out of my group. Trust me, it would be much faster and easier for everyone if I wrote all the papers myself, but that's not the point;
drafting papers is part of the PhD training.

Implying that your advisor (or I or any other PI) doesn't know enough about the research only reveals that you don't really know yet what a PI's job entails. As has been noted by DrugMonkey (can't find the link right now) and others many times, students/postdocs always overestimate their importance in research and underestimate that of their advisors.

Anonymous said...

Haven't had time to read the other comments, so this might be redundant.

When I first read about this mistake your other colleagues have made, I thought you were referring to the consequences for the postdoc.

My PhD adviser 'graduated' me before I thought I was ready; I had published a few things, but only one of the papers that was supposed to be in my thesis was actually in any stage of review. (One chapter was already dead/scooped, and the others weren't quite there.) I defended feeling like a fraud and started my postdoc feeling like a fraud. Two of the chapters got published in my first year as a postdoc, but one I'm still working on a year later. I mean actively working on, with objectives for every week; the methods are crazy hard and new, and I've been tweaking and waiting for results and tweaking and am just going to publish whenever this round is done, and leave the rest for other people/papers.

I'm just representing the flip side. My adviser hasn't used these methods and, I suspect, might doubt me when I say I've been working on this stuff continuously since leaving. I'm sure I care more than anyone about getting this paper out.

Cherish said...

Actually, my observation is that it's half and half in both cases.

Half of the profs bring it on themselves, giving absolutely no feedback (yet claiming all the credit), and then complain that students aren't independent/driven/what-have-you enough. The other half are involved (although sometimes that's not a good thing). I've seen two cases of advisors giving no input, even to the extent that they wouldn't respond to emails or requests for meetings. While I would not say that most are this way, it's probably close to half. So I do think Gears has a point: I have several friends whose advisors take credit for these brilliant students, when those students would have done as well (or even better) regardless of their advisor, who did the work and published pretty much on their own.

On the other hand, I'm sure there are flaky students, too. I didn't see it as much in my PhD program, but there were a good number who left my MS program without getting papers written up. And usually, it was not at all for lack of involvement of the advisor.

I wish there were a way to match bad advisors with bad students so that we could all be appropriately happy or miserable. :-)

Anonymous said...

"Gears said...
I'm with MissMSE on this one. Sure, the student should finish up the papers but I've seen more cases of terrible advisors that are too busy for their students than grad students unwilling to do yet even more work for next to no salary. "

Per GMP's comment, I'm sure there are a few really terrible advisors out there, but not as many as you might think. I can see how students might end up with the impression that PI's sit on papers - here's a case in point. One of my postdocs is an incredibly efficient, clear and elegant writer. Her first draft (given to me) still needed a little editing for content and organization, but no errors in spelling, grammar or basic rules of scientific writing. We got this paper submitted very quickly. In contrast, another trainee gave me a first draft that was incomplete, obviously not edited for spelling or grammar, had paragraphs with no topic sentences, etc. I gave some general and some specific comments back. I got a second draft with only the specific comments fixed, but all the general flaws were not addressed. I sat on that for a while as I pondered whether or not to write it myself or whether I had the patience to teach a semester's worth of basic composition in a few weeks.

If your manuscript is sitting on your advisor's desk, take a long careful look at it, have a few other colleagues who are good writers give you feedback and try again.

Miss MSE said...

Anon@4-1-11-10:56, In my particular case, he openly admits he hasn't actually opened the document file since I sent it to him. Several months ago... It's also been through several rounds of editing between me, a postdoc and a third member of the group. I believe this qualifies as sitting on a paper.

GMP said...

@Anon at 10:56, I hear you. It indeed takes longer to return corrections on a paper that's structurally unsound and where major rewrites are required. I ranted about them here (i.e. students who consider comments such "revise thoroughly" an indication that nothing should be revised).

Miss MSE, I am sorry your advisor takes so long to return corrections. I guess I am pretty prompt (will usually return in a matter of days or a few weeks, depending on how busy I am) but I have long-term collaborators who take forever so that really stalls all collaborative papers. Let me tell you what works with one particular collaborator who takes forever and is a control freak (everything has to be just the way s/he wants it). We would agree when s/he would bring the rewrites back, if s/he didn't, I would insist on a new date, and would keep insisting until the corrections were returned. The point here is to guilt-trip the person, which means you have to keep reminding him/her, no judgement, just constant pressure and constant, polite reminders that you are indeed waiting for the response. The student we coadvised (my collaborator and I) said that as a result of all this gentle but constant pressure our joint papers were coming out much more quickly than those from the collaborator's group alone (where the collaborator was the only faculty, and the students felt, much like you do, that there was nothing to do but wait). So that tells you pressure works; but, you could say that it needs another faculty to apply it. My PhD advisor was usually very good about returning corrections, but when he wasn't, I was constantly pestering him. I don't know how your advisor is, but if you think he could be guilt-tripped into returning corrections, I think you should go for it. I think most faculty appreciate proactivness (is this even a word?) in their students. Gentle, but constant pressure. Good luck!

Anonymous said...

GMP, you did the right thing. Though your comments indicate otherwise, perhaps in the back of your mind you realized it is completely inappropriate to delay a student's graduation for the sake of your lab's productivity. Per the National Academies:

"Powerful forces can work against making good progress. You or other faculty might seek to retain students as they become more proficient. That is an unfortunate conflict between your desire to maximize productivity in your own research and your duty as a mentor to support a student's timely progress. **Your primary obligation is to the education of the student.**"

(from Adviser, Teacher, Role Model, Friend: On Being a Mentor to Students in Science and Engineering, p. 34, though I recommend all faculty read at least this entire chapter...)

Anonymous said...

I wouldn't say exactly that graduate students "procrastinate" and work at "snail pace." Many PhD advisors are "nice" just as long as students have something that they want. My PhD advisor sucked the life out of me. It took me six years to finish my PhD. During this period I had to work as a teaching assistant (every semester), mentor undergraduates (sometimes 3 per semester), write grant (or portions of grant) applications, and tried to do research (among other things), while my advisor spent most part of his days masturbating at home (I am sorry about the language, bu I think that best describes his behavior).

GMP said...

Anon @ Oct 15, there are jerk advisors, and maybe you have one, I don't know. Often students don't actually know what it is that faculty do all day (so it seems like jerking off), whereas there is a lot of work regarding running a research program that goes on behind the curtain. Anyway, I don't know your advisor.

But there are most definitely a number of students who procrastinate or work at a snail pace. My students are nearly always on RA's (if they TA, it's a semester or two over the course of their PhD). So when a student is done with classes, paid as a full RA, and still no research progress for weeks or months, that means they are not working hard enough.