Writing the Methods section as narrative

Image: Once Upon a Time, CC-0 via pixabay.com

We often tell ourselves that a good Methods section allows someone else to replicate our experiments.  I’ve argued, among other places in The Scientist’s Guide to Writing, that we needn’t and shouldn’t expect this function of most Methods sections.  Rather, a good Methods section gives readers what they need to ascribe authority to you as a scientist, and to understand the Results you’ll present.

I get frequent pushback against this idea, usually in connection with prominent hand-wringing over the so-called “replication crisis”. But a couple of weeks ago I gave a talk about writing at Saint Mary’s University (the one in Halifax, Nova Scotia) and I got a different and very convincing kind of pushback.  An audience member whose name I didn’t catch (sorry!) argued that “narrative-for-replication” is an effective way to teach beginning writers to produce Methods sections.  Undergraduates, for example, are told to describe everything they did, in the order they did it, so that their classmates could do the same experiment and get (at least hypothetically) the same result.  Actually, I’m quite sure I was taught this way as an undergrad myself.

This brought me up a bit short, because I still maintain that a Methods section detailed enough to allow real replication is far too long and too detailed to serve the needs of real readers.  And not only that: the narrative of “everything I did in the order I did it” is unlikely to be the easiest structure for the reader to understand.  The logic is appealing, though, as one step in a way to teach writing.   I have to admit that my “authority and understanding” criteria for what belongs in a Methods section is much more abstract than the narrative-for-replication criterion of “everything your classmate would need to repeat what you did”*.

The idea here is essentially that saying too much is a heuristic to make sure you say enough – and I can get behind that.  It’s closely related to the idea of “early writing”: writing Methods when you either propose or execute the experiment, not later when you’re likely to leave things out.


Saying I’m OK with this heuristic doesn’t mean I’m rolling over on the issue of Methods written for replication.  Instead, I’m OK with it because it matches up with another really important thing about writing: the value of getting to a first draft easily and quickly, even if that first draft bears little resemblance to the final product.  That lack of resemblance is crucial.  The first draft should be seen as process, not product.  A first draft may suck (most of mine do), but it exists, and that’s a big part of the battle; I can fix “suck” much more easily than I can fix “not there at all”.  So if writing a Methods draft as narrative-for-replication gets that draft written quickly, that’s terrific.  It just isn’t the end.  Instead, the narrative-for-replication draft Methods is a list of elements from which one can select and arrange bits to make the real Methods.

When we teach narrative-for-replication Methods, then, we need to couple that with something else: the importance of letting go of the 1st draft.  Revision is tough: we’re emotionally invested in the words we worked so hard to produce.  To make things worse, modern word-processing makes a draft look really professional to the eye, even if conceptually it’s really just scribbles.  So we need to teach the importance of revision – and not just superficial tinkering, but wholesale reconsideration and reinvention. Revision is an inextricable part of writing, which is why The Scientist’s Guide to Writing has several chapters on it.

So: teach Methods as narrative-for-replication? No. But teach writing a Methods draft via narrative-for-replication? By all means, and thanks to my anonymous questioner for making me connect the dots that became this post.

 © Stephen Heard (sheard@unb.ca) May 10, 2016

Related posts:

*^In fact, narrative-for-replication is how I pack to travel: I imagine myself going through the day, and pack the reagents and equipment I use.  OK, I think, I take a shower: pack shampoo; I shave: pack the razor…. My suitcase is a Methods section.  Or at least, it is until I realize I can’t zipper the darn thing closed.


4 thoughts on “Writing the Methods section as narrative

  1. NK Simons (@Tritotanus)

    I am really sitting on the fence after reading your post. On the one hand I also love reading papers with a short, concise method section that is easy to read and gives me just the relevant information to understand the results. I also agree with your argument that revisions are an integral part of writing and that it is important to learn how to write for your readers. On the other hand, I dread the thought of “losing” all the details which are relevant for the few readers which want to replicate a study or even just apply the method in another context. I would also argue that writing a narrative-for-replication is an important step to detect potential flaws in your own experimental design or statistical analysis (especially for early-career scientists). Thinking in detail about your methods and writing it down for someone else to follow them, clearly increases your confidence when defending them against reviewers (or against a supervisor who is not familiar with what was done). Therefore, I would argue that both types of method sections are equally important. One obvious solution is to have the detailed methods in a supplement, but then those should be treated with the same care and given the same value as the methods sections in the main manuscript.


    1. ScientistSeesSquirrel Post author

      Thanks, good comment, and good point about methods-as-narrative as a way to think through methods. That’s part of what’s behind my tangential reference to “early writing”, but I kind of threw that away without explaining. I do agree that the online supplement is a great opportunity to have our cake and eat it too – with the caveat that realistically, I don’t expect peer review to pay much attention to the online supplements, and I don’t expect more than about 0.01% of all readers to read them! But still a great place for a more detailed “replication methods”.


  2. Manu Saunders

    Interesting post! Perhaps it’s a matter of quality vs quantity? If we just focus on word length regardless of content, then of course revision is essential to condense methods and make them succinct. But I would argue that a good methods section shouldn’t just be to “ascribe authority to you as a scientist, and to understand the Results you’ll present”. It also needs to ascribe authority to the study. Therefore, the narrative-for-replication detail is necessary, not to be able to replicate, but because by knowing how to replicate, you (the reader) knows what process the author has taken to collect and analyse their data…and you are then in a better place to judge the validity of the results.


    1. ScientistSeesSquirrel Post author

      Manu – yes, interesting point. In my book I talk about the reader knowing you’ve used appropriate methods in appropriate ways, and I think that’s what you’re getting at. I would argue that still doesn’t need a replication level of detail, but you’re right, that’s more about authority for the study rather than the author. I muddled those two a bit in the post. Thanks for commenting!

      Liked by 1 person


Comment on this post:

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 )

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.