Working with students writing their first research papers has gotten me thinking about the writing process. Sometimes people don’t know where to start, or once they start, get lost in the details and lose the big picture.
I’ve written out two related short guides. The first has some general information on what to put in different parts of a paper; these are not hard rules and there are all kinds of wonderful, beautiful exceptions. The second is the order that I like to write/plan the sections in. I like this order because I feel like the paper falls together as I go. They’re written to stand alone, so there is some repeated information.
(1) “Recipe” for a paper
Intro: Everything in the Intro should lead into the question, like a funnel.
– Start big/wide. What are the big concepts that you’re going to discuss? Provide enough theoretical background to understand the topics covered in the discussion. Stay focused on your subject area though. For example, starting with Darwin is probably too big.
– Narrow in. E.g. What species are you going to use? Why use them? Provide relevant background on their biology and, if applicable, research done in this area with your species in the past.
– Get specific. What question(s) will you address with this paper?
– What did you do and why, with enough detail that it can be repeated, but not so much that it’s overwhelming. For example, the height of the counters in the lab is not necessary, but the sizes of the Petri dishes or the amount of reagents might be.
– What did you find? Present this in the order you plan to discuss it in.
– What do each of the results mean? Walk through each result and perhaps discuss what you might have predicted vs what was found. Brainstorm all possible interpretations for each result (or group of results).
– What do results + interpretations mean for your original question(s)? How do your results fit into the current literature? How have you expanded on what was known?
– What’s left to do in this area? What are the next steps/future directions?
(2) Order for writing and planning papers:
1) Start with your goal(s)/question(s). What are your predictions? What do you need to know to assess your predictions?
2) Methods: What do you need to do to get the data necessary to assess your predictions? Write this section out as early as possible because otherwise you might forget details.
3) Analysis: Do the statistical tests you need to get the information from your data that will allow you to assess your predictions.
If you’ve already done the analysis and are on to writing, I’d still start with 1 (objectives) to stay grounded on what the paper needs to focus on, but 3 (analysis) is redundant and 2 (methods) can be written at any time (hopefully it’s already at least partially done).
4) Results: What order do you need to present the results in so that the story/answer to the question(s) naturally unfolds? Sometimes I find it helpful to make a presentation, or explain the results to a friend/colleague. When there are lots of analyses that are interrelated in potentially complicated ways, I sometimes create independent “figures” for each unique result that include the main message/conclusion of each, and tape them to a wall; that way I can easily move them around to see which order makes the most sense.
5) Discussion: explain what each result means, and what they mean when viewed as a whole. What does it mean on a small scale (i.e. your specific question) and on a larger scale (i.e. put it in the context of the literature on your topic… what have you added to the scientific community with this contribution?).
6) Discussion (Future Directions): given what you’ve just contributed to the scientific community, what needs to be done next to build our understanding of the universe in this area next?
7) Introduction: Now that you have your discussion, this is where you put any information/background that is necessary to understand what you did and what your conclusions were. Start with big concepts and narrow down to the specific part of the universe you’re planning to help clarify (your questions/goals).
8) Abstract: 1-2 sentences to summarize each section.
9) Revisions, revisions, revisions.
As I said, there are loads of exceptions to this guide. For example, I read a really great paper that started with an experiment, but in the Discussion the authors included an advancement to theory on the same topic, but wasn’t immediately applicable to the experiment’s results. It read well but was more of a 2-for-1 deal. Another example, some journals include the methods in the appendices, or if you’re writing a scientific note, then all of the sections will flow together and there may or may not be an abstract.
Ultimately you’ve got to do whatever works best for you and the story you’re trying to tell, but hopefully this will be helpful!