Hmmm. Not sure I’m happy with that title, my students probably could have done better.
I was talking with a colleague about paper writing the other day. In comparing mental notes, we had completely different experiences of how our papers got written as graduate students. I wrote mine. Well, that is a lie. I wrote the complete first draft. That first draft came back to me absolutely bathed in red pen. Yeah- that’s right PEN. These were the days before you could just hit ‘accept all’ on the track changes function and have a nicely edited draft with the touch of a few buttons. I alternately hated (not really, only figuratively of course), and loved, and hated (only figuratively), and loved my mentor as rounds of drafts were turned in and handed back sometimes with words edited back to read exactly as they had been written in some earlier version. In the end I admired my mentor’s technique with this whole thing- because he/she made paper writing a very valuable learning experience for me, the trainee.
I try to torture teach my students this way now, and I see both the strengths and weaknesses of this approach. The biggest strength I have already mentioned- the learning experience of becoming a better writer and learning the thought process of putting a data, a paper, a story really- together. To build a case, make an argument on paper. The weakness- of this approach is that it can take what feels like forever and a day- as green students cobble together their idea of what constitutes something publishable (with anxious advisers prodding them along). The difficulty, the time and the effort involved for the adviser depends on the language, writing ability, and background knowledge of the student. When English is the second language- having the student write the first draft can mean that the mentor re-writes practically every.single.word. Re-writing at this scale can be incredibly labor intensive for the mentor.
My colleague, on the other hand, never had the experience of putting a paper together and doing the crazy amounts of editing during their graduate training. Oh they may have lightly edited some draft- but the bulk of the text was written at the outset by the mentor themselves. I’m sure that this approach ultimately brings the paper to submission status faster, and students may still learn what pieces of data are needed to put a paper together- but I bet a lot of the learning of scientific writing is lost when papers are written this way. There are clear benefits to the mentor, the student, the lab and the project in being able to publish quickly. What happens though- when the student has to put together a thesis? What happens when they move on to their postdoc and haven’t yet written a whole manuscript from start to finish?
And finally- I wonder how career stage of the PI plays into this… are more seasoned PIs more secure ($s, papers) and not as needy of quick pubs… thus able to let newbie paper writers flounder a little? Does the necessity of as many pubs as quickly as possible that early career stage PIs make them more prone to do the paper writing for their trainees? Or are these factors irrelevant… are we bound to repeat what our mentors trained us to do- do it like they did it. That seems to be how I do it… then again – I think I had a stellar mentor in this respect.
Blue Lab Coats 
I think I’ve done it all ways with trainees- from doing most of the lifting to letting them flail away. My mentors were universally of the let ’em flounder variety so I don’t think it is training. More the fact that the circumstances surrounding every paper is different.
Very interesting. I plan to get a graduate degree in the future. I hope I get to go through the process of putting the paper together. What is the point if you don’t learn all there is to learn?
Wow. I’ve never even heard of the second method. When I was in grad school, I always went from start to finish, sometimes feeling incessantly stuck in the never-ending cycle of edits. But as you say, I came out a MUCH better writer. Unless the second method (PI-doing-first-draft-or-heavy-lifting) is only used in rare circumstances, I can’t image that professors/PIs are doing anything but harm to their students this way.
Pingback: Tweets that mention Did your student write the paper, or did you? « Blue Lab Coats -- Topsy.com
As a grad student and postdoc, I wrote the first draft of everything I published and my mentors made suggestions which I then used to do the subsequent drafts.
As a new PI, I can only wait so long (up to 6 months in one case) before giving up on waiting for the trainees to write the first draft and doing it myself. It’s not what I want to do but if I wait for them the drafts will never get written.
It seems that my mentor is taking the approach that your mentor took with you. I turn in a chapter and I get back tons of handwritten feedback. I do the majority of the editing and we go through many rounds. She explains problem areas, points them out, sometimes shows me an example of how to fix it, and so on. It is a long and tedious process and there are days when I wish she did it like some of my friend’s major profs. In these cases the student writes a draft and the professor does so much revising the paper doesn’t look the same. Basically, the student hits the “accept all” and the process is finished. The number of revisions is much less (the more experienced writer did most of the work) and they get to the “final version” much more quickly. I know that in the long run I will benefit from this process because in the end I will be a better writer.
And she always uses blue or purple or some other color than red to mark up my papers. It seems less harsh 🙂
I’m a new faculty but from a large lab where I did most of the supervising inc
luding manuscript prep. This lab puts out at least 10 papers/yr so as u can imagine I logged in many hours doing this. I find the best approach is to hav them write section by section. We start w figures and legends..then results..then discussion …and then intro abstract title (and methods shoved in ther somewhr). Piecemeal makes it easier for both u and the student and faster.
As a PhD student, I wrote every paper from start to finish that I am first author on. If my advisor added or changed anything in the text, *I* was usually the person to get hung from the rafters by reviewers for that gibberish! I learned, very painfully, how to write papers and proposals during my MS, and my advisor would, painfully, go through PACKS of red pens. I attached a fresh new red pen to every draft. I also bathed in red ink, along with the love/hate back and forth seething jumping for joy over pages turned back. I am sure I took years off my advisor’s life. Paybacks are a bitch, and I’ve had the same experience with my students. I let them flail as much as my own heart can take it. It’s alot easier if you start students off with M&M, then Results.
At the moment I am a graduate student working on a [publishable] paper that is part of the assessment for my Masters. Given my background, the fairly ‘hands-off’ approach of my supervisor seems reasonable. For most students, this would be their first article ever (maybe after some “dummy” versions in other courses, not based on their own data etc) so the hands-off system might cause them to take a long time or give up. Ideally, there would be a bit more guidance than I am experiencing. However, years ago when I was a Research Officer on a specialist psychiatric unit, writing my first article with my boss was an entirely different experience! He had written several papers and a book before I met him and he had developed a very clumsy style of writing. At first he had us each write a rough draft of everything, then combined them and assigned sections to each of us. I usually got the Introduction and results and he got methods and discussion. We would go through endless joint sessions of changing the wording, while arguing and using quite different forms of expression! Eventually it would be up to me to make the thing hang together, write the abstract, tidy the references and send it off. To me, this was TOO ‘hands-on’- he had a really wordy and rather old-fashioned, stodgy style (sorry mate) and I was quite minimalist- even when he had decided on an abstract, I’d go away and change and shorten it! So to me, a middle ground between these would be best- having never experienced a boss/supervisor taking over and writing the whole thing! I DID take over once, writing a short paper overnight and sending it straight to the journal- they didn’t change a word, so our team got a painless publication!
DM- I think I should point out- to your point- that what my mentor did with me may not be what he/she chose to do with each and every graduate student- depending on their experience, goals, writing ability, written English etc. In addition- what other mentors who have a more heavy lifting approach do, may also vary with the student. Thanks for making that point- I get that at some point the work has to be published. I’ve seen some text that I’ve had to edit (not necessarily from my trainees)- where I thought- hey I am not sure how much I can get from this- and I re-wrote very, very heavily. That’s ok and I still have them edit what I have done. Not all manuscripts or students start from the same place. For sure.
My PI is the second sort – PI would rather write the text the first time. This style has not changed since PI got tenure. (FWIW: I am a native english speaker, PI is not)
When I tried sending PI a draft of part of a review article, it was sent back to me with a couple of sentences intact. PI rewrote all the rest, and by that I mean PI threw out what I had written and wrote something completely different. So I gave up on writing from scratch and now just heavily revise the drafts I get from PI (gots lots of experiments to do if PI is going to rewrite the whole thing anyways). 😦 At least PI isn’t bothered by my heavy revisions…
It didn’t help me learn how to write from scratch, but I have been learned to think critically about the points PI is making as we tend to have different perspectives on our research (due to our different scientific backgrounds). Also, I am detail oriented and PI is a big picture person; I’ve noticed we complement each other more if we write in this direction. Still, it’s frustrating at times, but I know the end product has always been better than either PI or I would have produced alone.
Not the approach I would prefer if I had a real say in it.
I firmly believe that a student’s thesis/dissertation should be her own writing (bathed in my red ink, as you describe above), maybe because that’s the way I was taught, but also because many dissertations are basically a collection of manuscripts. If the advisor writes the manuscripts, and the manuscripts are bundled together into a thesis or dissertation, is it really the student’s work? However, I’m anxiously awaiting several students to complete their M.S. theses, and the instant those babies are defended, I’m taking the lead in turning them into respectable manuscripts. So, I guess my philosophy is students write the thesis, then whoever has the most incentive to turn it into a submitted manuscript gets the green light. I suspect that as long I’m mostly advising MS students, I’ll be writing the papers that end up in the journals.
It never occurred to me that you could do it any other way than the “dripping with red” way. I certainly wrote all of my papers. There was some help on the tricky parts, and it did take a long time. My advisor was big on me doing things myself, with help if I ever truly floundered. I think it was the right approach.
I always write the first draft if I am first author. I remember my mentor completely re-writing some of my conference abstracts, but by the time I started writing papers for publication he had shifted from nitty-gritty edits to more general comments.
It never would have occurred to me that a PI could write a paper and put a student’s name on it. Is that what you mean? Or did your colleague simply graduate without any first author papers?
I will have to write all my papers. With the endless editing to the point where the data I used to love, I will hate. That is how it when with my grant proposal and that is how its been for the students in front of me. Thankfully, I am considered an “excellent” writer so apparently I’ve gone through less drafts than others. Which makes me feel very very bad for the others.
I’m a graduate student and I definitely write all of my papers. There is some back and forth and editing afterward, but the writing is still pretty much my own. It probably does help that I speak and write English much better than my advisor does. I like this way of proceeding most of the time although I’d really like for some revising to take place in real time with both of us in the same room going through the paper. I fill it would be a better training opportunity than just receiving the .doc file with all the edits. Plus that way I could maybe write in LaTex instead of word :-).
I do get frustrated sometimes though. My advisor just decided to turn a 8000 word paper into a 3000 word one after a first rejection. I disagree with her on what really needs to be taken out in order for that to happen, but her answer is basically: I don’t have time to delve into the paper and data right now, but don’t take anything out and just make it shorter … There I do wish she would try and thin out by 5000 words without taking any of the data out …
Although I’m new to the whole TT thing (I just finished negotiations and start at the end of the semester), the best thing I learned during my PhD work was to write papers on my own and be able to publish with only minimal involvement from my advisor. I do, however, remember my first draft on my first paper I wrote during my MSc and I was crushed by my advisor. That was a 6 month slog of write – edit -review – write – edit etc. And then, after getting crushed in peer review, learning the painstaking art of making corrections when the reviewers have conflicting opinions.
However, since then, I’ve been able to do all of the writing and peer review issues with only minimal input from my advisor. If it’s a student’s first attempt at writing a paper, they should do the bulk of the work for their experience. I know I’ll be under pressure to produce by the time I have tenure review but the *goal* of academia is teaching…
My PhD advisor wrote the majority of the papers for our lab. We were expected to write M/M and the first draft of the results, but generally just edited the intro/discussion. The reason for this was because it was a better use of our time to be doing more experiments (note that she was TT at the time, now tenured but doubt it has changed). Now, as a postdoc, I’m writing the drafts and making the revisions to all of my papers with input from advisor. I wish I had been more involved in paper writing during my PhD, but at least I’m getting that training now, before heading off to PI land…. As a side note — what is the general trend with grants? In my PhD lab, I rarely had the opportunity to discuss or read, let alone help write, the grants my PI was submitting. It’s much different in my current lab — as a lab (often during lab meetings) we discuss the grants being submitted, specific experiments in those grants, and at least have the opportunity to read and offer opinions on the actual grant itself.
I wrote all my (first author) papers as a grad student. They came back with lots of “did you think of this”, “this needs to go there” etc… but it was very helpful to get over the panic of “it’s a white page”. although, my PI and I sat down and planned the paper out based on the figures first – so I sort of knew what story I wanted to tell.
I think that is the important part – to let the grad person tell the story and see if they see what the paper is going to be about. I have learnt that I can’t decide a title, too flaky for that, and the last part of the discussion/conclusion is usually best as a collaborative thing. That said, I learnt the most from those verbal discussions about the “corrections” to my draft and I wouldn’ve never picked it up as much if I only “tracked changes”.
Well, I guess I still print my papers and read/correct by hand since I feel that I have a better overview of it. Old school…
As as student and then a post-doc, I always compiled the figures and wrote the paper and did the back and forth
nonsenseedits, until it truly was ‘publishable’. As a brand spanking new PI, I hope that I will have the patience to do the same…I just need papers, and I kind of needed them yesterday. BUT, the learning experience of writing and editing a manuscript is invaluable (and painful), but mostly invaluable in the development of a young scientist.My writing experience as a graduate student sounds exactly like yours. I wrote the complete first draft (in sections) and then we went through several rounds of editing that sometimes made me want to pull my hair out. As a result I am great at writing manuscripts and feel completely comfortable with the process. As it turns out, this is a great thing because my post-doc mentor is not a great writer and those who rely heavily on him to get papers out spend significantly more time in the writing process than those of us who can write the paper on our own and only need his input on a broader scale.
As an aside – If i am lucky enough to run my own lab, I will probably use my gradAdvisors method when it comes to students writing papers.
last year writing a grant, i got all sorts of red pen comments on a hard copy of the grant i sent to my PI. when i suggested “track changes” to the word document, she responded “i’m not a luddite!” and we both laughed.
but i learned what i needed to change about my 1st drafts 9and 2nd and 3rd and 16th! drafts) as opposed to the kids down the hall who submitted and abstract to their PI’s and got a final, completely different version with no tracked or red pen changes (radically different, in case you were wondering). at least i knew what i needed to do to step up my game, and they didn’t.
have i mentioned that my PI is awesome?!!!
Definitely depends on both career stage of PI and individual abilities of trainees. I wrote all drafts of all my own papers throughout PhD and postdoc, I floundered a bit at times, and it was the best training I could have got, but my advisors were very senior and busy people. As well, I was pretty good with written communication to start, even if more of a novice with the form of the scientific paper. Result was that my pubs all came out within 6 months of leaving my training labs (because I followed through), whereas my peers less skilled at writing and more dependent on the writing or rewriting efforts of the boss had their papers published 3 years later or not at all.
In my own lab, I have yet to attract a trainee that is a native English speaker/writer, and yes, this makes a difference. I still insist that each trainee write a complete first draft after we work on an outline together, for the training value mainly, but I am simply not at a career stage where I can go back and forth with drafts for years of rewrites, and I have seen enough failed attempts where I have tried to do the red pen approach and the revised drafts are no more comprehensible than the original. In the end I will have to rewrite the whole damn thing anyway, so I have become more likely to short circuit and take over the writing sooner rather than later after that first draft. Still, it depends on the particular trainee, how hard they want to work and what potential they’ve shown, the exact effort I want to put into the training exercises.
I have a first author paper. I remember being angry that I only got to turn in my data and write the Materials and Methods section and even that was edited and re-written. At the end of the day I have to agree that my boss probably did this because he was new and needed to publish ASAP.
I was wondering how often this happens and then I came across your blog. Makes me feel much better. LOL…a little at least.
A very good quality information at all, I am amazed at all the posts you have made. Thank you …..