TL;DR

• The discussion section interprets your results, not just repeats them.

• Start with your main finding, stated plainly and directly.

• Connect every claim back to existing literature in your field.

• Acknowledge limitations honestly — editors expect it.

• End with implications, not a summary of what you already said.

Most student researchers spend weeks on their methodology and results, then rush the discussion section in a single sitting. That is a mistake. The discussion is where editors decide whether your paper contributes something real to your field. It is where reviewers look first when they want to understand what you actually found. A weak discussion section can get a strong paper rejected. A clear, well-structured one can carry a paper with modest results all the way to publication.

The confusion is understandable. You have already presented your data. What else is there to say? The answer is: quite a lot. Your results tell readers what happened. Your discussion tells them what it means, why it matters, and how it fits into the broader conversation your field is already having.

This guide walks through how to write a discussion section that does all of that, step by step, in plain language. If you have already drafted your results, you are ready to start.

What a Discussion Section Actually Does

A discussion section interprets your findings in the context of existing research. It explains whether your results support or challenge what other researchers have found, identifies what your study adds to the field, and is honest about what your study cannot prove. It is not a second results section. It is an argument — a reasoned case for why your findings matter.

Many students treat the discussion as a place to restate their results in slightly different words. Editors notice this immediately. A discussion section that simply says "the results showed X, which means X happened" adds nothing. The question you are answering in this section is not "what did I find?" It is "so what?"

Think of it this way: your results are evidence. Your discussion is the argument you build with that evidence. Every claim you make should be connected to something — either your own data, a cited study, or a recognised principle in your field. Unsupported assertions weaken the section and raise red flags for peer reviewers.

If you are still working on earlier sections of your paper, the guide on how to write a literature review for your field will help you build the foundation your discussion will draw on.

How to Write a Discussion Section: The Opening Paragraph

Open your discussion section by stating your main finding directly, in one or two sentences. Do not restate your entire results section. Identify the single most important thing your study found and say it plainly. This gives reviewers and editors an immediate anchor for everything that follows.

Here is the structure that works. In your first paragraph, state your primary finding. In your second paragraph, compare that finding to the existing literature. Did your results align with what previous studies found? Did they contradict something widely accepted? Either outcome is valuable — but you need to address it explicitly.

For example, if you conducted a study on sleep patterns among adolescents and found that weekend sleep duration was significantly longer than weekday sleep duration, your opening should state that finding clearly. Then, in the next paragraph, you would connect it to prior work — citing, for instance, studies published in journals such as Sleep Medicine or Journal of Adolescent Health that have examined similar patterns. You are placing your finding in conversation with what already exists.

If you are preparing to submit your paper and want to understand what goes into the submission package beyond the manuscript itself, the post on how to write a cover letter for journal submission covers that process in detail.

How to Connect Your Results to Existing Literature

Connecting your results to existing literature means citing specific studies and explaining precisely how your findings relate to theirs. Broad statements like "this aligns with previous research" are not enough. Name the studies, name the authors, and explain the relationship clearly.

This is where many student researchers struggle. They know their results. They have done a literature review. But they have not built the bridge between the two. Here is a simple way to approach it:

1. Identify the two or three most relevant prior studies in your field.

2. For each one, write one sentence explaining what that study found.

3. Write one sentence explaining how your finding either supports, extends, or challenges it.

4. If your finding contradicts prior work, offer a possible explanation — a difference in methodology, sample size, or context.

This process forces you to be specific. Specificity is what separates a discussion section that earns respect from one that reads like a placeholder.

If you want support identifying the right journals for your paper and getting structured feedback on your draft before submission, joining the Publication Compass waitlist puts you in line for a platform built to help student researchers do exactly that.

Acknowledging Limitations Without Undermining Your Work

Every study has limitations. Acknowledging them is not a sign of weakness — it is a sign of intellectual honesty, and peer reviewers expect it. A limitations paragraph that is too vague or too brief will raise doubts about whether you understand your own methodology. A limitations paragraph that is clear and specific actually strengthens your paper.

The Committee on Publication Ethics (COPE), which sets standards for ethical academic publishing, emphasises that transparency about study limitations is a core element of research integrity. Hiding or minimising limitations is considered a form of misrepresentation.

Write your limitations paragraph using this sequence:

1. Name the limitation specifically — for example, a small sample size, a non-random sample, or a single geographic location.

2. Explain why that limitation exists — often it is a practical constraint, not a flaw in your reasoning.

3. State what effect it may have on the generalisability of your results.

4. Briefly note how future research could address it.

This approach shows reviewers that you understand the boundaries of your own work. It also gives your paper a longer shelf life — future researchers can cite your limitations section when designing follow-up studies.

Understanding your methodology deeply makes this section much easier to write. The post on how to write a methodology section for a science paper is worth revisiting before you draft your limitations.

How to Write a Discussion Section Conclusion: Implications and Future Directions

Close your discussion section by explaining what your findings mean for the field, not by summarising what you already said. Implications should be specific and grounded in your results. Future directions should be realistic — name the actual gaps your study leaves open, not generic calls for "more research."

Many student papers end their discussion with a paragraph that essentially repeats the abstract. This wastes the final impression you make on a reviewer. Instead, use this space to answer two questions:

1. What should practitioners, educators, or future researchers do differently because of what you found?

2. What specific question does your study leave unanswered, and how could a future study answer it?

If your study is in the social sciences, the implications might speak to policy, practice, or community interventions. If it is in the natural sciences, they might point toward experimental follow-up or applied technology. Either way, make the implications concrete. "This study suggests that school start times should be examined in relation to adolescent sleep debt" is more useful than "more research is needed on sleep."

For researchers working on social science papers specifically, the guide on how to write a methodology section in social science addresses how methodological choices shape what you can and cannot claim in your discussion.

Common Mistakes That Weaken a Discussion Section

The most common mistakes in a discussion section are repeating results without interpretation, making claims that go beyond what the data supports, ignoring contradictory evidence, and writing implications that are too vague to be useful. Each of these signals to editors that the author has not fully engaged with their own findings.

Here is what to check before you consider your discussion section finished:

• Does every paragraph interpret, not just describe?

• Is every major claim connected to a cited source or your own data?

• Have you addressed at least one finding that surprised you or did not fit your hypothesis?

• Is your limitations paragraph specific and honest?

• Do your implications go beyond "more research is needed"?

If you can answer yes to all five, your discussion section is in good shape. If not, those are the paragraphs to revise first.

One more thing worth checking: your abstract should reflect what your discussion concludes. If they are misaligned, editors will notice. The post on how to write an abstract journal editors read walks through how to keep both sections consistent.

Frequently Asked Questions

How long should a discussion section be?

A discussion section is typically 15 to 25 percent of the total paper length. For a 4,000-word paper, that is roughly 600 to 1,000 words. Length varies by discipline and journal. Always check the target journal's author guidelines, as many specify word limits for individual sections.

Should I repeat my results in the discussion section?

Do not restate your results in full. You may briefly reference a specific finding to anchor your interpretation, but the discussion section should focus on what the results mean, not what they are. Reviewers have already read your results section. Repetition signals that you have not done the interpretive work the discussion requires.

How do I know if my discussion section is too speculative?

If a claim cannot be traced back to your data, a cited study, or a recognised principle in your field, it is likely speculative. A useful test: for every interpretive sentence you write, ask what evidence supports it. If the answer is "nothing specific," soften the claim or cut it. Phrases like "the data suggest" are more defensible than "the data prove."

Can I introduce new references in the discussion section?

Yes. It is appropriate and often necessary to cite studies in your discussion that were not in your literature review, particularly if your findings relate to work you encountered after designing your study. What you should not do is introduce entirely new concepts or variables that were not part of your original research design.

How is the discussion section different from the conclusion?

Some papers combine the discussion and conclusion into one section. When they are separate, the discussion interprets findings and situates them in the literature, while the conclusion provides a brief final summary and the most important takeaway. If your journal requires both, keep the conclusion short — two to four sentences is often enough.

Conclusion

Writing a strong discussion section takes more time than most students expect, but the structure is learnable. Start with your main finding. Connect it to prior work with specific citations. Acknowledge your limitations honestly. End with implications that are concrete and grounded in what your data actually shows. Follow that sequence and you will produce a discussion section that gives reviewers something real to engage with.

The discussion section is your opportunity to show that you understand not just what you found, but why it matters. That is the skill that separates a publishable paper from one that gets returned for major revisions. For more guidance on every stage of the research and publication process, visit the Publication Compass blog.