Premium Content

Discussion: The most difficult part of writing a paper

Which section of a paper do you find most difficult to write?

Go to the profile of Clio Heslop
Aug 18, 2015

Please sign in or register for FREE


Go to the profile of Laurie Marczak
Laurie Marczak over 2 years ago

I was surprised by the advice in this module to include "This suggests..." at the end of a results paragraph. Previous advice I'd received, including comments from editors and referees, was to keep all elements of discussion out of the results section (with the obvious exception of deliberately writing a combined results/discussion). Perhaps this is journal or discipline specific?

Indeed, I was just revising a paper and cut a sentence in the results section that included "...suggesting that..." with the margin note that the thought should be expanded on in the discussion instead.

Go to the profile of Peter Gorsuch
Peter Gorsuch over 2 years ago

In response to Laurie's point: Good question. This does vary from journal to journal, from discipline to discipline and indeed from person to person. I'm slightly removed from the first two of those things, because I work on papers across the natural sciences and I don't work for a journal - here's my own personal view!

Obviously the results section isn't just a repository of information - that's what figures, tables and of course data repositories are for - so it needs a narrative of some sort. In practice, to explain how the ideas you're presenting relate to each other, you're very likely to need a certain amount of interpretation. For example, if you reached a certain conclusion through a computational study and then investigated the degree to which that conclusion reflects natural systems, you'd need to explain the conclusion that resulted from the computational work, and how you reached it, at some level. If you don't, readers might not really appreciate why you carried out the experimental work and what you were looking for. I think that's really what the "this suggests" recommendation stems from - it's about ensuring that readers can fit each set of results into your overall argument. (If that argument intrinsic to your study design involved bridging any logical gaps by using speculation, that would probably be something you would want to discuss more thoroughly in the Discussion section!)

Having said that, I'm not saying that I think the Results section should be used to explore the various questions that might arise from those results, or extend beyond the scope of the study - the level of interpretation, and especially speculation, should be kept to a minimum in the Results section. I think it's generally fine to refer to findings being entirely consistent with a previously described finding, for example - this drives the narrative along, it tells the reader that this isn't a major point to the paper, and it saves you having to acknowledge this similarity (which is often useful to point out) in a separate point in the Discussion section, where it might just be a bit distracting. But I wouldn't do much else in the way of comparisons with previous studies. Most people also avoid too much discursive critical consideration of the findings (alternative explanations etc.).

So in short, I feel that a little restrained, judicious interpretation is actually quite important in the Results section (and I suspect that this view might be shared by most journals and disciplines), but would generally query the inclusion of interpretation that steps outside what is strictly necessary for presenting the results in the context of the main argument.

Go to the profile of Bart Verberck
Bart Verberck over 2 years ago

Good point. I agree with Peter. The idea of putting a phrase of the type “This suggests …” at the end of a paragraph in the Results section is to keep the story flowing. The function of such a phrase is to indicate or stress why you’ve done the experiments/calculations just described and perhaps why the obtained results necessitate carrying out the experiments/calculations described in the next paragraph (still in the Results section).
Indeed, these statements have a preliminary character, the full conclusions should be reserved for the Discussion section.

Go to the profile of David Rogers
David Rogers over 2 years ago

For other users who are interested in this conversation, the video Laurie, Peter and Bart are discussing is "Writing the Results section" from the module "From Introduction to Conclusions". If you're a subscriber, you can watch the video by following this link:

Go to the profile of Rafael  Zamorano
Rafael Zamorano about 2 years ago

I find difficult to write the introduction. Especially the connection of the background with my findings.
It is also difficult for me to write the final parts of the analysis and interpretations of the results. However, I may have clear thinking for the conclusions.

Go to the profile of Vadim Volkov
Vadim Volkov about 2 years ago


Go to the profile of Kang Liu
Kang Liu about 2 years ago

Introduction and interpretations of the results

Go to the profile of Rafael  Zamorano
Rafael Zamorano about 2 years ago

William Shockley in 1957 wrote a paper: On the statistics of individual variations of productivity in research laboratories.
Shockley won the Nobel prize for the invention of the transistor and later more or less singlehandedly launched Silicon Valley. Shockley wrote the paper while been the director of Bell Laboratories back in the 1950s when this was one of the premier research centers in the world. In that paper, he showed that productivity as measured by total number of publications, rate of publication and number of patents is log-normally distributed.

Shockley suggested that producing a paper was tantamount to clearing every one of a sequence of hurdles. He specifically lists:
1) ability to think of a good problem; 2) ability to work on it; 3) ability to recognize a worthwhile result; 4) ability to make a decision as to when to stop; and write up the results; 5) ability to write adequately; 6) ability to profit constructively from criticism; 7) determination to submit the paper to a journal; 8) persistence in making changes (if necessary as a result of journal action).

Because of the lognormal characteristic in Shockley statistical study, the probability to successfully publish scientific papers is, P= P1P2P3P4P5P6P7P8, the product of the Pi´s, where Pi is the probability to be successful in step i, where 1

Go to the profile of Megan McNellie
Megan McNellie about 2 years ago

I find the Discussion most difficult, esp. how to fit my research (the brick) into the work into the context of the existing research (the wall)

Go to the profile of Ross  Colquhoun
Ross Colquhoun about 2 years ago

The discussion pieces have been very helpful. Thank you My reservation, based on some years of writing and reviewing students' papers was that the introduction should include what was accomplished or what the findings and conclusions were. My understanding was that these topics were better left to the results and discussion sections. My preference, based on identifying the 'gap' in the research to date was that the introduction should conclude with a statement of the hypothesis/es to be tested. I would be interested to hear others views on this issue.

Go to the profile of Olive Onyemaobi
Olive Onyemaobi about 2 years ago

Discussion section. It is a bit challenging to structurally compile the information from the results section and taking special care not to over- or under- hype your research.

Go to the profile of Ivan Palomares Carrascosa
Ivan Palomares Carrascosa about 2 years ago

Introduction, particularly in keeping a reasonable proportion of it focused on the context, existing limitations/challenges in that context, the contribution to overcome them, and the outlook of the paper structure.

Go to the profile of Laura Zhou
Laura Zhou about 1 year ago

How to analyze the data scientifically

Go to the profile of Julian D. Munoz Sierra
Julian D. Munoz Sierra 7 months ago

Discussion, especially when it must be separated from results section.

Go to the profile of Ruth  McKeogh
Ruth McKeogh 5 months ago

I always find the bookends of a paper most difficult. Introducing the topic and then drawing the conclusions without spoilers in the middle of the paper.

Go to the profile of Evaristo Mukunda Malenje
Evaristo Mukunda Malenje about 2 months ago

Discussion. Especially if some experiments did not work as planned making it hard to link the results to the objective

Go to the profile of Lwazi Charles Mahlalela

the results section. Some reviewers will want to always have a  referefence to cement a result and it's meaning. I feel like that removes the originality of the work. What if my results are new. Where then should I get a reference?