writing process

So, the basic question here is about your writing process. What does it look like? Rough sketch is great. Detail is fine too.

i have one that mostly works for me, but it feels like it needs at least one more tweak* to be both allow me to be productive on the level i expect and to be sustainable. i’m not exactly sure where that/those tweak(s) best fits. So, care to share your model? I sketched out the gist of mine after the jump.

  1. Have an idea. I think this one’s worth mentioning because of the number of times that i’ve seen (or started) work that doesn’t really know what it’s about. And there’s little that’s more frustrating than reading (or writing) something that doesn’t know what it’s trying to say.
  2. Do preliminary analysis to wrap my head around what i think that the paper might end up looking like.
  3. Write an abstract and a page or two of the basics of that idea.
  4. More reading.** It’s at this stage where i typically take some time away from the actual paper itself and go back to the literature to get a good handle on what i understand the state of the literature to be, make sure i have a solid theoretical motivation for what i’m tackling, etc.
  5. Write a first draft of the front end. This is typically a really rough sketch of the ideas behind what i’m up to and doesn’t have much in the way of grammatical correctness or linearity of argument. It’s basically a mind-dump of all the things that i think could be relevant to the end product. Often this includes ~2x as much info as i think will actually be pertinent to the “final” version of the paper. There is very little censoring of ideas at this point. The one thing that this generally concludes with is a clear statement of the primary hypothesis/es.
  6. Go back a do what i think will be the “real analyses” based on those hypotheses.
  7. Fill in a few sentences of findings, etc for the first draft.
  8. Make an outline of what i think the paper will actually look like. Now, i know that a lot of people use outlines at a lot of varying stages in their writing process. i tend to find that i’m not really ready to come up with one until well into the life of a paper. i don’t think i’m alone in this.
  9. It’s often at this stage that i also come up with some additional analyses that will be necessary to complete the story of the paper, so here’s when i typically do those.
  10. Starting with a blank document, the first draft in another window and the outline in a third, write a second draft. I say write specifically because, often times this draft borrows very little direct language from the first draft (in large part because of how unstructured / mind-dumpy i allow those “first drafts” to be).
  11. Quick revision of the “second” draft.
  12. Share with collaborators / writing group.
  13. Revisit, revise, etc. Depending on the feedback, i’ll typically jump back in to the above either at step 6 (to conduct new, or re-do, some analyses) or 8 (to restructure the argument), depending on the nature of the feedback i get.

Putting this on paper also raised another question for me about the life-stages of a research paper. When do you feel like your work is ready for presentation (e.g., at a conference, etc.)? I actually have two answers to that for my process, and differentiating between the two largely depends on whether the paper is among the first from an entirely new project (somewhere between 11/12) or an additional paper from a long-running project (somewhere around 7/8).

*One more for now. I suspect it’s something i’ll fiddle with for as long as i’m at this. I guess i’m saying i still feel like i’m one more tweak away from a solid first draft of what the process ought to look like, if that makes sense.
**Obviously, there’s a lot of relevant reading that pre-dated step 1 here, but i don’t necessarily think of that as part of the writing process. Maybe i should, but that’s likely a discussion for another time. This is the stage i’m at on a big chunk of a sociology of science project i’m part of these days, and hence the bulk of the content in my summer reading bleg post. Incidentally, when i wrote up that list, i was surprised to see that i didn’t have any of my go-to writing books in the pile i brought with me for this first 6 weeks away this summer. So additionally, do you have any favorites of those?


2 Responses to writing process

  1. Michael Bishop says:

    Where is the part where you spend months playing with your data without much focus… maybe I should shave a little off that step 😉

    I like Becker’s book on Writing for Social Scientists and Williams on style, but I’m no great stylist myself.

  2. pmk says:

    1. Start with a novel question. I usually start with something I’ve seen in life, and figuring out where that challenges conventional wisdom or (at least some strain of) academic research. I try to use this to derive competing hypotheses.
    2. Also start with a novel approach or novel data. Think of using some data in a way that folks haven’t used it before. Or maybe a new method.
    3. Write up ~1 page that articulates the main arguments I want my data to adjudicate. This is usually the point in an article’s life that I submit it for presentation at a conference. Only about 1/3rd of my papers ever get more refined than this before the actual presentation.
    4. Spend 2 years playing with data. Maybe write a grant. Mostly focus on other papers that started at step 1 about 2 years prior. And check facebook/twitter.
    5. About 6 months after I’m truly sick of the project and rue the day I had the idea, I start writing up the results, drafting the front end, and complaining about it all day long. Usually, my collaborators also complain that I have taken so long. All we do is complain.
    6. After 3-6 months of drafting and revising, I send it out for review. If it is rejected, I refer to the anonymous reviewers as idiots, and resubmit the paper to another journal that has now risen in my estimation, because it may still recognize the genius that is my work. Hubris is under-rated as a valuable survival strategy when submitting papers that seem to attrack rejection.
    7. I read the first chapter of Becker’s book on writing for social scientists, but then I discovered that twitter and facebook are much more pleasurable ways to waste time when I should be working. See step 4.

Leave a Reply

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

WordPress.com Logo

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

Google+ photo

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

Twitter picture

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

Facebook photo

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


Connecting to %s

%d bloggers like this: