The Care and Feeding of Your PI: A tip for grad students and postdocs

March 30, 2011

GMP has a hilarious LOL/sob post up over at Academic Jungle in which she laments becoming PI Pushover.

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 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)…. in the 3 months he’s been here after the PhD…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.

Of course, the academic blogsphere is entirely made up of hardworking trainees with distant, out of touch PIs (on the one hand) and PIs who are highly engaged mentors cursed with lazy-ass trainees (on t’ other).
So the comments are lining up accordingly, just like they will do with this post after I publish it.
But here’s the thing Dear Hardworking Trainee….GMP is right.

Once all the threats have been removed, you suck at getting proper manuscripts written. And this is why we get into these little discussions about whether publishing papers, or generating publication quality manuscripts, should be a criterion for the PhD as a general principle.
Here’s my little bit of advice on the care and feeding of your PI where a manuscript is concerned.
Put in the Effort!
Seriously. If there is one thing a PI hates more than anything it is the absence of evidence that the trainee is working on something. With concentration and focus.
Yes, we are here to train you how to put together a paper and all that. Indeed.
But you are no longer undergraduates six year olds.
The PI expects you to go away and come back with some actual progress. New analyses written….knowledge of a literature that the PI knows exists (or knows)… and you clearly haven’t gotten past 2008 in your PubMed search?…a coherent discussion of the data. Some reflection that you understand that the format of the Journal you have been discussing submitting this puppy to for consideration.
Here’s another little tip from the PI side of the office. You have to get it through your head that you will write COPIOUS amounts of text that may not pass muster. Either for this publication or ever. Write, write, write, write. Explain. Cite. Written academic text is not some precious commodity that needs to be hoarded…it is what you manufacture and excrete. Live it. Love it. Don’t be stingy*.
“I don’t know what he meant or wants so I am paralyzed and can’t write anything” says the trainee. HA! Write. It. Every. Possible. Way. You. Think. He. MIGHT. Have. Meant. That. Comment!
This is the only way to get better and more efficient at a given writing task, for one thing, and it is the kind of evidence your PI is looking for that you are actually working. Basic Principle of Life: It is far easier to edit down than to edit up in length.
There is another purpose which is that this is the way you cement your understanding of a literature in your own mind. The way you cement your understanding of what you are about, scientifically. Mike the Mad Biologist has a weird observation in his response to Science Professor’s observation that biomedical scientists are a whiny bunch. Quoth the Mad One:

The problem is that most of the skills you learn are only useful in…the biomedical sciences. Most don’t learn enough ‘generalist’ skills, such as high level math or serious programming skills, to have other career alternatives if academia doesn’t work out. Worse, many of the skills they learn become obsolete. A decade ago, sequencing was a Ph.D. activity, or at least, an activity supervised very closely by a Ph.D. Now, it’s largely automated, and the machines are mostly run by technicians with bachelors degrees.

If you are focused on bench skills (or even “high level maths” and “serious programming skills”, my Mad friend) then you are missing the point of the PhD. This brings me to the observation that it is not just the case that we evil PIs only train PhD level scientists for technical work. It is also the case that some many trainees show little evidence that they themselves understand that bench work IS just technical work. If you don’t have a conceptual grasp of what you are trying to understand, big picture, you are soooo not ready to be a PI. A scientist, yes indeed, but not a scientist ready to head up an independent research program. You don’t get there only by reading papers. You get there by writing. Writing academic text. Whether that be for a dissertation that will only moulder in the library or for a manuscript ready for publication.
The tip is simple: Put in the Effort.
Your PI will respond accordingly.
Disclosure: In contrast to a prior time when I may perhaps have written a similar set of observations, my trainees are currently getting. after. it. when it comes to writing. I’m the weak link… 🙂
*I have like 7 draft files of a given manuscript or grant applications before I even get withing smelling distance of a reasonably put-together “First Draft”. Just saying.
[UPDATE: Apparently I should have read a little bit deeper into GMP’s recent rantings.]

18 Responses to “The Care and Feeding of Your PI: A tip for grad students and postdocs”

  1. Andrew Says:

    “Basic Principle of Life: It is far easier to edit down than to edit up in length.”
    Maybe my writing is afflicted with logorrhea or I am excessively in love with my own writing, but I actually find that the opposite is true for me. I find editing down (to meet word limits, etc.) very painful, such that filling pages is relatively easy. Although perhaps you mean that it’s easier for a PI to edit down a trainee’s chaff than to create wheat out of thin air.


  2. DrugMonkey Says:

    The latter, although here practice makes perfect as well. I can be pretty ruthless with my own text. [ maybe that’s why I refuse to apply much in the way of editing to my blogging….kind of a reactive outlet :-)]


  3. queenrandom Says:

    Right on. I wonder how this trainee expects to get a good recommendation for grants/future postdocs/etc? Furthermore, it’s his career. His CV. His first authorship. He has a lot to lose by thrusting his paper into someone else’s hands. Why would any trainee let this shit fall through the cracks like that? Publish (and publish quality shit) or perish etc.
    I don’t always agree that the paper has to be 100% accepted and published before graduation; it should be a reasonable case-by-case decision based on the student’s demonstrated ability to give a fuck about hir own career and demonstrated courtesy to get shit done on a short time frame. And the paper should be damn close to publication. None of this first draft has been written but not even discussed or submitted bullcrap.


  4. Ace Says:

    Feed meeeee!!! Feed me Seymour!


  5. Eric Lund Says:

    @Andrew: Actually, your case is consistent with DM’s aphorism. The key word is “edit”. Like you, I often find it painful to edit material out of a paper/proposal/presentation to comply with length limits, but I do get practice at doing so. Seldom do I get to a significantly advanced stage and find that I can choose to add material without taking other material out.
    One of the main difficulties with editing up is fitting the additional material within the narrative of your existing material. For instance, as a reviewer I sometimes see revised manuscripts where entire paragraphs have been added at the request of one or both referees. Often it is painfully obvious that these paragraphs were added as an afterthought, because they are not connected to the original narrative in the paper. It’s not impossible to do, but it takes substantial editing to make it fit. Even adding a sentence or two to clarify a point to a reviewer is IME a nontrivial task.


  6. The problem is laziness or crappiness works in many cases. Sometimes PIs don’t have a choice – they need to get published after all. Other times they just find it easier that way. My PI was like the later, and I was the first student to take the lead and insist I had a go at the first draft. Student are too passive in their own career development sometimes.
    It makes me so angry how many grad students and postdocs do not take the initiative in drafting and writing their own papers. I honestly don’t think you should be first author unless you wrote (or took the lead in writing) the damn thing.
    For me the literature reviews, proposals and paper writing were the best part, so I find it hard to empathise with this reluctance to bring a project to fruition. Still, if you don’t like writing you should have applied for a technical position not a PhD.


  7. Not to get off topic, but I view ‘high level skills’ as a way to have a back up if everything falls apart academically. I agree though, that writing is essential.


  8. Beaker Says:

    Writing a quality manuscript and submitting it to a good journal is in the interest of both student and PI. All PhD-wannabees must be able to do this or they should not graduate. There is no advantage for anybody to have the student skip out early. Nobody cares how long it took for you to get your PhD. They only care about whether papers came out of the work. Any student itching to leave pre-publication is better served by busting their butt to get those manuscripts written and uploaded first. Here is how you do it. Get a good alarm clock, and set it to wake you up early. Eat something and drink a caffeinated beverage, then sit butt in desk chair and write for a few hours before doing whatever else you need to do in your life. No reading, email, blogging, texting twittering, etc until you’ve put in your hours writing. A few months of this, and presto! You will have your manuscripts, your training as a writer, and your committee will agree that you qualify for your degree. Prioritize manuscript-writing over dissertation-writing and consider that the PI will probably help a lot with the manuscripts but not so much with the dissertation. Bonus: it’s a lot easier to land a good posdoctoral position if those accepted/published papers are on your CV before you start your job search.


  9. I sometimes see revised manuscripts where entire paragraphs have been added at the request of one or both referees.


  10. Anonymous Says:

    Nobody cares how long it took for you to get your PhD.
    I beg to differ. Lots of people care how long it took you to do a PhD. Here’s an example: when I interviewed for a tenure track position, a colleague told me that the college dean (it’s a college of engineering) would not agree to sign off on an offer to anybody who took more than 5 years to do a PhD, because in the dean’s experience, shorter PhD duration correlates with better success as a faculty.


  11. Beaker Says:

    shorter PhD duration correlates with better success as a faculty
    My personal experience indicates that, yes, there is some correlation, but it is weak. One guy from my PhD class was a superstar. He graduated in 3.5 years and has continued on the fast track since. Another dude took 8 years and still sucks today. Another guy got out in 4 years because the project was good and his PI was a young high flyer. This person has done done little since. By contrast, another guy tried a whole bunch of risky projects, and most of them failed. It took him 7 years to graduate. But his experience served him well because today he is on the faculty of an elite university. Roger Tsien, Nobel Prize winner, struggled in his PhD because his early projects crashed and burned. It was only his “last chance” project to develop calcium indicators that led to his Cambridge PhD–and the rest is history. Perhaps it is not the same for engineering. An alternative interpretation is that the college dean you mention was an asshole.


  12. Alex Says:

    The best person that I knew in grad school took almost 8 years. He also published a very, very high profile paper on crazy risky work that took a long time to make happen, and he’s now tenured at a very, very fancy place.


  13. GEARS Says:

    You know, if you’re name is on it for the research as the PI and you’re getting credit, you should know the work too!
    I’ll follow up with same thing I responded with on GMP’s blog. How many papers would you actually write if your student left you all of their notes, data, lab notebooks, literature lists, and did the figures for you? If you answer little to none, you don’t deserve to be an author. That clearly shows you no idea what’s really been done and haven’t had enough contact with your student.
    Maybe FSP is right on both counts and this is an indication that BioMed is different. Maybe all BioMed students are whiny and maybe all BioMed PI’s are slave-driver taskmasters.
    If I ever got on such a high pedestal that you and GMP are on, I want my grad students to knock me down from there. Shocking but maybe you’d be even more productive as a PI if you knew what was going on in the trenches rather than in your safe, cozy command center.


  14. There are cases where graduation without first authorship is permissible, such as being scooped or having the work that would have made an acceptable publication being merged into a bigger collaborative project. A PI will prefer one high impact publication to two little ones, and unfortunately joint first-authorship is only acceptable when contribution is equal.
    The student does not have the option to publish as first author of a low impact paper if the PI wants their data as a figure or two on a big paper. This is where the graduate committee comes in! The decision as to whether or not a candidate is ready to graduate should not be left to one person.
    The dissertation, if not made up of reformatted (and more in-depth) publications, should consist of chapters demonstrating work of publishable quality. The dissertation is the true test of writing skills. Many PIs write publications, and even when the student takes the lead it is obviously tidied up to some degree. The dissertation demonstrates the student can (or can’t) write, and it is a pity it is never read. It never fails to amaze how many postdocs cannot write anything worth a damn!


  15. In the US six years was respectable enough, but now I am in the UK I am embarrassed to admit it took that long as a fulltime grad student. Here they encourage graduation in four years and you get booted after five. I’d hazard a guess that over the span of six years publication records are comparable. (I’d like to know figures on that one!)
    Funnily enough, despite being less experienced, postdocs are not considered ‘trainees’ here either. Although tenure-track does not exist and faculty positions are even harder to come by, the long-term career postdoc is an option.


  16. Isis the Scientist Says:

    This smells of an “Isis is right.”


  17. In relation to the generativity of writing itself, please see this excellent post:

    One of my trainees and I discussed a manuscript he was writing and decided on a particular angle. He spent several days generating a complete draft coming at the topic from this angle.
    After I read this draft, I realized that the angle wasn’t optimal, and indicated that we actually needed to come at it from a different angle, which would require substantial rewriting. The trainee got all huffy, angrily accusing me of “wasting his time” by not having pointed him at the correct angle in the first place.


  18. “After I read this draft, I realized that the angle wasn’t optimal, and indicated that we actually needed to come at it from a different angle, which would require substantial rewriting. The trainee got all huffy, angrily accusing me of “wasting his time” by not having pointed him at the correct angle in the first place.” (Comrade #17)
    And he was just a passive partner waiting to be told what angle to take next? Did he not have an opinion, or understand why that particular angle was not working? I usually conceded that my PI was right, but not always. Overtime it developed into a more collaborative process, with my contribution growing more valuable with experience. Certainly I never assumed that ‘the best angle’ was always obvious to even an experienced researcher. Why do so many students want to be told what to do rather than learn how to do it themselves?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: