As an editor of a scientific journal, I’ve had a long history of tending to focus on the Methods and Results Sections (especially the figures and their legends) at the expense of the Discussion Section and the Introduction. This in the context of the current way we organize scientific articles.
That said, I find myself facing the challenge of mentoring doctoral students in how to write a paper for publication. The difficulty is that even though I might discount the Discussion as an editor, still as an author, they do need to be written.
It is easy to communicate what not to do:
Don’t simply regurgitate the Results with the addition of speculative commentary and citations.
Don’t write a mini-Review paper–confusing that distinguished scientific form for the function of Discussion.
Don’t hype your results beyond what they clearly say. Especially avoid phrases like “to our knowledge, the present work represents the first….”
But what should go into a Discussion?
Seems to me that the biggest clue come from the word “discussion”.
Here are the dictionary definitions:
1. Consideration of a subject by a group; an earnest conversation.
2. A formal discourse on a topic; an exposition.
So let us take the first definition. Imagine a group that includes you, the author, and additionally your readers (including potentially your reviewers). The Discussion section should represent the consideration of the Results where your written words represent that complete consideration as if all the members of the group were in fact participants.
Hence, you might imagine the Discussion to represent the stylized representation of that group consideration were it to actually take place in a seminar room, complete with slides and pointed questions.
You present that familiar slide which summarizes your findings (each one corresponding to a result). Your audience members ask scientific questions (just as they would during a seminar) and you do your best to answer them. The above written into a Discussion section, amounts to a formalized discourse (the second definition from above)–but really amounts to just the imaginary “minutes” of what was presented and what was asked during our imaginary talk where you are presenting to the group (your readers).
Thus one ought to order the paragraphs of the Discussion section such that the most important finding goes first. Then it is put into context–why it is the most important finding. Most important is the implicit inclusion of what might be construed as reasonable scientific critiques on the finding (this is the inclusion of your imagined audience) along with your answers to those critiques. It is within this imagined give and take that appropriate referral to the relevant literature becomes important.
To do the above clearly requires both some sense of imagination based heavily in reality. You’ll write a better Discussion if you’ve actually presented your real data to a real audience!
And then on to the next most important finding.
The trick is to embody all of this “discussion” in your Discussion without it all sounding like the minutes of the local PTA meeting. Rather the “minutes” are transformed into a formal exposition (second definition).
One ends with a summing up of what the significance of the Results taken in their totality might mean to the field–with speculation kept to a real minimum.
Jim
Discoverers and Discoveries 