Wednesday, 3 September 2014

The unspoken art of publication (why I suck at it)

My PI has set ambitious publication goals for our lab following the summer of data collection. Knowing her track record in the past, I have no doubt they are achievable. This is in fact one of the main reasons I came to work with her. For I also have no doubt I will find this just as difficult, if not more so,  than learning the technique of our experiments.
I've been doing scientific research off and on (at this point, mostly on) for nearly ten years now. The first piece of research for which I thought "I would like to publish this" was my master's thesis back in 2004. It wasn't published. I entered my PhD, after a 2 year break from academia, knowing that I should have three publications by the time I was done to be competitive on the job market. I never published anything until after I submitted my dissertation, and I still don't have three publications (one in print, one in press, one revise and resubmit I need to both revise and resubmit). So I've been intending to publish for as long as I've done research. But good intentions only take you so far.
My lack of publications has already nearly cost me a career in science. I was unemployed for just under a year after completing my PhD. As I asked about jobs, and for advice on getting them, my non existent publication record was always the thing that stumped people. I gave seminars, I led workshops, I presented at conferences. I got a lot of encouragement about my research, and always the same warning: none of that was worth much as long as none of it was published. At my first interview, the first thing I mentioned was that I had just submitted a paper. My interviewer's relief was palpable. In fact, my lack of publications was held against me even earlier: one of the comments from my unfunded NSF DDIG was that, as I had no publications, I was an unknown quantity when it came to research. At the time, I felt the comment was unfair. After all, I was only a third year graduate student. With time, I have come to see the comment for what it was: a warning to me from the reviewer.
If I'm going to make it in this gig, I need to understand why, despite having the intention to publish, I've been so bad at it so far. And I think a lot of that boils down to learning a set of skills that are specifically relevant to publication, distinct from research. Coincidentally, as I was begining to write this post, this very topic of discussion was brought up by the esteemed Drs Isis and Drugmonkey. So, at the risk of repeating what has already been, here are my thoughts on things I've come to understand have hampered my ability to publish.
I do not find writing hard. I don't particularly like writing papers (or my dissertation), but I write well, and quickly, and generally produce first drafts that are pretty solid from a basic readability point view. I do hate making figures (I really really do), but as I become better at using ggplot, that's getting less of an issue. Yet writing has, so far, probably constituted a relatively minor portion of my time as a researcher. Why is that the case?
1) Re-learning how to write. This one only occurred to me recently, as a result of the discussion over at Drugmonkey's. It struck me that learning to write collaboratively, to accept and embrace criticism, to be willing to submit anything less than your best work for comment, is not taught when writing undergraduate or high school essays. You don't get a do-over for those: no matter how constructive the comments of your teacher, errors in style, argument, structure, analysis amount to lost grades. Certainly in the British University system, one has no opportunity to have one's written work critiqued and fix the errors before marking. Thus, succesful undergraduates are the ones who have learnt to go over their work with a fine toothed comb and anticipate all possible critiques. It makes sense they will be prickly in their response to criticism: any thing they missed or overlooked is a rebuke of their hard earned skills. I know people will respond "but you're not undergrads anymore". To which my response is: you were there too once. Recognise the patterns and help the trainee unlearn them. For what's it's worth, that process for me started outside academia, when writing website copy. My initial reaction to being critiqued was not great. I've learnt, finally, to view writing as a cooperative exercise, to embrace critiques and view my drafts as suggestions. It's taken a long time to get there.
2) Knowing when what you have is publishable. This, I think, is the one that stumped me for many side projects. I simply had no idea what "publication standard" was. My master's thesis was probably borderline publishable as it was at the time, yet I didn't know it. My first year rotation project in graduate school, I was told, was unpublishable. Yet a paper came out just a few years ago asking the same question, using the same methods, with the same conclusion, though looking at a different taxon. Maybe my project wasn't publishable as was, but it was clearly a damn sight closer than my rotation advisor thought. This, I think, is one where advisors are crucial. As graduate students in particular, I think we tend to assume our work is sub par. It takes encouragement to realise that no, our work is decent.
3) Publishing what you have, not what you want to have. I submitted five conference abstracts while in Grad school. All had good data, most of which will be part of the publications I will eventually get out of my dissertation. Yet, of the ones that would eventually be part of my dissertation, I wrote none of them up as I went. Why? Because I had a plan for the papers I wanted to publish, and until I had all the pieces of that plan, I wouldn't publish anything. Unfortunately, one of the pieces of that plan didn't come together until my last ten months in grad school. Thus, I delayed publishing results that were publishable until after everything was done. This was foolish.
4) Don't worry that your paper will be bad. If you have spent time developing your hypothesis, if you can argue your methods, if you have done your background reading (which, if you have defended your thesis proposal you should be able to do), your paper will not be bad. It may not be the best. It may only be worthy of a society level journal. That doesn't matter. It will exist, and it will be OK, Don't worry about some journal club somewhere tearing it apart. And remember that, at some point, you will find someone who will demolish a paper you have held as a paragon of scientific integrity. 5) Do not seek "accepted with minor revisions". Don't, it isn't worth the effort, Trust me, I've done it. I much prefer the paper I have that was accepted with major revisions than the one that was almost accepted as is. I am excited about my revise and resubmit. Peer review, as much as it can be jarring, is an occasion to make your paper better. "Major revisions" is just another round of editing.
6) Speaking of which: find mentors who will read and critique your drafts constructively. Reach out to a broad circle of people. Thank each person willing to read your draft. Remember you are the author, so have final say, but give each commenter a fair and judicious hearing.
7) Middle authorships are important. My advisor in grad school had few side projects to share with me. I carried the weight of all my research and my papers. This makes me proud of my dissertation, but being middle author on a lab paper early on would probably have made me more comfortable about publication. So mentors: involve your newbie trainees sufficiently in projects that they can be middle authors. It's a great, low stakes way for them to learn the arcana of publication. And trainees: god damn it, ask. Find some skill you can offer for a project and ask if you can be involved. Get that middle author paper.
8) Seek out and be grateful for collaboration. Two people can work faster on all aspects of a paper than one. You must do the heavy work of your dissertation and postdoc projects, yes. It does not follow you must do all the work, and do it alone.
9) Don't chase glam. Your first paper should be aimed squarely at a respectable journal in your field. Don't sell yourself short, but don't pin your hopes on a longshot, not until you have a few good pubs under your belt. Again, a published paper in a decent society journal is worth more than a CNS paper in prep.
To mentors, I would say this: none of this is obvious to your trainees. Nor will they learn it by osmosis. Talk to them, sit them down, explain the process to them. Demand drafts of them. Give them drafts to edit. This is the very definition of tacit knowledge. If you do not seek to make it explicit, you cannot expect your trainees to know it,

No comments:

Post a Comment