In our latest blog, we provide guidance on how to address the challenge of crafting a suitable Discussion section in scientific writing. Researchers often say that constructing the Discussion section is the hardest part of the paper-writing process. Clair Geary, a 5th year PhD student in the Immunology and Microbial Pathogenesis program at Weill Cornell, recently published her first first-author paper and explained to us the difficulty of writing a good Discussion, saying, “For the other sections, it was clear to me what needed to be included and in what order, but for the discussion there were tons of options for what to write about and I had trouble deciding what points were worth emphasising and what types of connections would be most interesting to readers.”
So, what are the main aims of the Discussion in scientific writing and how should you present this vital section when preparing a paper for publication?
Explain how your study has moved knowledge in this field forward
You’ve laid out the research problem in the Introduction and then clearly presented the results you collected during your research. The Discussion section of your scientific paper is now the platform for you to explain how your study has moved the reader’s knowledge of this field forward.
You do this by describing how your research takes readers from the challenges and issues outlined in the Introduction (point A) to their new understanding of the topic (point B), using the results you’ve just described. That done, you must also consider taking your audience to point C by discussing the implications of the findings for clinicians and policymakers as well as their limitations. At point D, you finally suggest future research directions and potential clinical applications.
In the first paragraph of your Discussion, briefly recap the latest understanding in the field and emphasise the novel aspects of your study. Do not repeat in detail information previously presented in the Introduction and Results sections. In particular, avoid simply repeating the numerical values presented in the latter. Instead, offer readers a brief summary of the research problem and state your contributions to the body of knowledge.
Next, aim to interpret the strengths and weaknesses of your principal findings. Critically analyse your findings and don’t be afraid to provide evidence for why they extend the literature or refute previous findings. While it is important to relate the presented findings to those of other relevant studies, however, refrain from overly citing previous work. This is not a literature review and only those papers that are directly relevant to your results – either because they support your findings or because they contradict them – should be mentioned. Take this opportunity to describe how your study moves the field forward but don’t go overboard on why your paper is superior.
It’s now time to discuss the implications of the findings for clinicians or policymakers. For example, which potential clinical applications might your findings point to? However, you are entering dangerous ground at this point and “most editors and readers will appreciate you being cautious, not moving beyond what is often limited evidence. Leave readers to make up their own minds on meaning: they will anyway” .
Next, delineate the limitations of your findings and suggest future research directions. Are your findings generalisable to other geographical regions or cultural contexts? Are your methods able to be repeated in future experiments? Finally, use the last paragraph of this section to draw overall conclusions and summarise.
The tone in your Discussion
Setting a balanced tone in your Discussion is one of the greatest challenges for researchers, and this has led to some debate about the extent to which this section should be structured . Docherty and Smith complain that authors are guilty of seeing the Discussion as a chance to “sell” the paper, stating that researchers are under such pressure to grow their publication records that they might intentionally or unintentionally “introduce bias by emphasising the strengths of the study more than its weaknesses, reiterating selected results, and inflating the importance and generalisability of the findings” . They add that “commonly authors go beyond the evidence they have gathered and draw unjustified conclusions” .
Skelton and Edwards take a different angle by promoting the “desirability of speculation”  in the Discussion section, which should, they argue, be discursive as its title suggests. According to them, “every discussion is obliged to ‘go beyond the evidence.’ Every paper must reach a conclusion that is not contained in its results. And not all statistically significant findings have clinical relevance. In quantitative research, therefore, a central aim of discussions is to reinterpret the significant as relevant—and that requires subjective interpretation of data.”
Hess’ advice is that “if you feel compelled to speculate, be certain that you clearly identify your comments as speculation” such as by preceding any statement with “We speculate that …” . Further, to ensure a balanced tone, our recommendation would be to seek out the Discussion sections in your target journal and aim to provide a similar structure and style.
The Discussion section is the opportunity for you to explain to your audience the merits of your study and its unique contribution. The best advice is not to pass up that opportunity by using unclear language, vague terms and overly speculative conclusions. As with the rest of the paper, aim for grammatical accuracy as well as a consistent set of terms and specific terminology. Your readers will not thank you for suddenly introducing a new abbreviation or concept at this late stage of proceedings. Modern academic writing also prefers the active voice.
According to Sanli and colleagues, the length of the Discussion section should not exceed the sum of other sections (Introduction, Materials and Methods, and Results), and it should be completed within 6–7 paragraphs. They add that “each paragraph should not contain more than 200 words” . This is sage advice for scientific papers that comprise 4000–5000 words.
Further, a combination of tenses is acceptable. For example, prefer the past tense to summarise the findings, whereas the present tense may seem more appropriate when interpreting the results or describing the significance of the findings. Indeed, a mixture of tenses might even be found in the same sentence in Discussions. Moreover, the future tense can be used to make recommendations for further research or provide future avenues.
In essence, taking the time to generate a well-crafted Discussion benefits not only the quality of your publication but also the wider scientific community. By clearly placing your findings in context, you ensure that their contribution and implications are considered fully.
References Docherty, M., & Smith, R. (1999). The case for structuring the discussion of scientific papers: Much the same as that for structuring abstracts. BMJ, 318(7193), 1224.
 Skelton, J. R., & Edwards, S. J. (2000). The function of the discussion section in academic medical writing. BMJ, 320(7244), 1269-1270.
 Hess, D. R. (2004). How to write an effective discussion. Respiratory Care, 49(10), 1238-1241.
 Sanli, Ö., Erdem, S., & Tefik, T. (2013). How to write a discussion section? Turkish Journal of Urology, 39(Suppl 1), 20.