TL;DR

• Restate your thesis in new words, do not copy it.

• Summarise findings without introducing new evidence.

• Connect your work to a broader significance or real-world impact.

• Acknowledge limitations honestly; reviewers expect it.

• End with a forward-looking statement, not a summary sentence.

Most students spend weeks on their methods section and about twenty minutes on their conclusion. That imbalance shows. A weak conclusion can undermine a strong paper, because the conclusion is the last thing a reviewer or reader holds in mind. It shapes how they judge everything that came before it.

The conclusion is not a summary. It is not a place to repeat your abstract. It is the section where you tell the reader what your findings actually mean, why anyone outside your study should care, and where the field should go next. That is a lot to do in a few paragraphs, and most guides skip over the specifics.

This post walks through exactly how to write a conclusion for a research paper, from the first sentence to the last, with the kind of detail that makes the difference between a conclusion that lands and one that trails off.

What does a research paper conclusion actually need to do?

A strong conclusion does four things in order: it restates the central argument in fresh language, it synthesises the key findings without repeating every detail, it connects those findings to a broader context, and it points toward future work or open questions. Each of these is a distinct move. Conflating them is the most common mistake student researchers make.

Think of your conclusion as answering one question: so what? Your methods section explained what you did. Your results section explained what you found. The conclusion explains why it matters. That shift in register, from reporting to meaning-making, is what separates a publishable paper from a school assignment.

Journals like PLOS ONE and Frontiers in Psychology publish author guidelines that explicitly ask for a discussion of implications and limitations. Nature family journals ask authors to address broader impact in their concluding remarks. These are not stylistic preferences. They are structural requirements, and understanding them early saves significant revision time later.

How to write a conclusion for a research paper: a step-by-step structure

Follow these five steps in order. Each one builds on the last. Skipping a step leaves a gap that reviewers will notice.

1. Restate the thesis or research question. Use different words than your introduction. The goal is to remind the reader of your central claim, not to quote yourself. One sentence is usually enough.

2. Synthesise your main findings. Do not list every result. Pull out the two or three findings that most directly answer your research question. Explain how they fit together, not just what they are.

3. Interpret the significance. Why do these findings matter beyond your study? Connect them to existing literature, a real-world problem, or a gap your work begins to close. This is where your conclusion earns its place in the paper.

4. Acknowledge limitations. Every study has them. Naming yours shows intellectual honesty and strengthens your credibility. Keep this brief, one to three sentences, and do not apologise for them.

5. Propose future directions. What should be studied next? What questions does your work open up? This final move signals that you understand your contribution as part of an ongoing conversation, not a final word.

If you are preparing to submit and want structured feedback on whether your conclusion meets journal expectations, joining the Publication Compass waitlist gives you early access to an AI platform built specifically to guide student researchers through this process.

How long should a research paper conclusion be?

Most peer-reviewed journals expect a conclusion between 5 and 10 percent of the total paper length. For a 3,000-word paper, that means roughly 150 to 300 words. For a 6,000-word paper, 300 to 600 words is appropriate. These are not rigid rules, but they reflect what editors and reviewers have come to expect across disciplines.

Length is not the same as depth. A 600-word conclusion that repeats the results section adds nothing. A 200-word conclusion that synthesises findings, names a limitation, and closes with a meaningful future direction does its job well. Aim for density over length. Every sentence should carry new meaning, not restate something the reader already knows from earlier in the paper.

Students writing in the social sciences often write longer conclusions because their findings require more interpretive work. Students in the natural sciences often write shorter ones because the discussion section carries more of the analytical weight. Knowing your discipline's conventions matters. Reading five or six published papers in your target journal before you write your own conclusion is one of the most practical things you can do. You can learn more about choosing the right venue in this guide to how to choose the right journal for your research paper.

Common mistakes that weaken a research paper conclusion

Knowing what to avoid is as useful as knowing what to include. These are the errors that appear most often in student submissions and in first-time author revisions.

Introducing new evidence in the conclusion is the most disqualifying mistake. If a finding, source, or argument did not appear earlier in the paper, it does not belong in the conclusion. Reviewers will flag it immediately, and it signals that the paper's structure was not thought through carefully.

Overstating the findings is a close second. Phrases like "this study proves" or "these results demonstrate conclusively" rarely hold up under peer review. Academic language is precise. Use "this study suggests," "the findings indicate," or "this evidence supports the view that." The hedging is not weakness. It is accuracy.

Ending with a generic sentence is the third common error. Conclusions that close with something like "more research is needed in this area" without specifying what kind of research, on what population, using what methods, tell the reader nothing. Be specific. Name the gap. Point toward a method or question that would address it.

If you are working on your first submission and want to understand the full process from draft to acceptance, the guide on how to publish a research paper as a student covers the broader journey in detail.

How to write a conclusion for a research paper in specific disciplines

The core structure stays the same across fields, but the emphasis shifts depending on your discipline. Understanding those shifts helps you write a conclusion that reads as fluent and field-appropriate to a specialist reviewer.

In biology and the natural sciences, the conclusion typically focuses tightly on what the data support and what experimental limitations exist. Journals like PLOS Biology expect authors to connect findings to mechanistic understanding or applied relevance. Speculation is acceptable but must be clearly labelled as such. You can find discipline-specific guidance in the post on how to publish a biology research paper as a student.

In psychology and the social sciences, conclusions often spend more time on real-world implications. Journals like Frontiers in Psychology and Journal of Adolescent Health value conclusions that connect findings to policy, practice, or clinical application. If your study involved a specific population, your conclusion should address how generalisable the findings are and to whom.

In computer science and mathematics, conclusions frequently describe the practical utility of a proposed method or theorem and identify open problems that remain unsolved. The future directions section carries particular weight in these fields because the work is often foundational rather than applied.

What peer reviewers look for in a conclusion

Peer reviewers read conclusions with a specific checklist in mind, even if that checklist is informal. They are asking whether the author understands what their study did and did not show, whether the claims match the evidence, and whether the paper makes a genuine contribution to the field.

According to the Committee on Publication Ethics (COPE), one of the most common reasons for desk rejection is a mismatch between the claims made in the abstract and conclusion and the evidence presented in the results. Keeping your conclusion tightly aligned with your actual findings is not just good writing. It is an ethical requirement of academic publishing.

Reviewers also look for intellectual humility. A conclusion that acknowledges the boundaries of a study is more trustworthy than one that presents findings as universal. Naming your sample size as a limitation, or noting that your study was conducted in a single geographic context, does not weaken your paper. It shows that you understand the scope of what you have done.

For a full walkthrough of what happens after you submit, including how reviewers assess your work at each stage, the guide on how to submit a research paper to a peer-reviewed journal explains the process from manuscript to decision.

FAQ

How do you start a conclusion for a research paper?

Start by restating your research question or thesis in different words from your introduction. Avoid phrases like "in conclusion" or "to summarise" as opening words. Instead, open with a sentence that signals the shift from findings to meaning, such as "This study set out to examine" or "The evidence presented here suggests." Keep the first sentence grounded in your specific study, not a general observation about the field.

Can you introduce new information in a conclusion?

No. A conclusion should not contain any new data, citations, or arguments that did not appear earlier in the paper. If you find yourself wanting to add something new, it belongs in the discussion or results section. Introducing new information in a conclusion is one of the most common reasons peer reviewers request a major revision.

How is a conclusion different from a discussion section?

The discussion interprets your findings in relation to existing literature and explores what they mean. The conclusion is shorter and more direct. It synthesises the most important takeaways, states the study's significance, acknowledges limitations, and proposes future directions. Some journals combine them into a single section labelled "Discussion and Conclusion." Always check your target journal's author guidelines.

Should a conclusion include citations?

Generally, no. The conclusion is where you speak from your own findings, not where you cite others. If you must reference a source to contextualise a future direction or limitation, one or two citations can be appropriate. But a conclusion heavy with citations usually signals that the author is still in discussion mode rather than synthesis mode.

How do you write a conclusion for a research paper when your results were inconclusive?

Inconclusive results are not a failure. State clearly what the study found and what it did not resolve. Explain why the results were inconclusive, whether due to sample size, methodology, or the complexity of the phenomenon. Then propose what a follow-up study would need to address those gaps. Reviewers respect honesty about null or ambiguous findings far more than overclaiming.

The conclusion is where your paper earns its credibility

Writing a strong conclusion is a skill. It takes practice, and the first version is rarely the best one. Read your conclusion separately from the rest of the paper and ask whether it answers the question: what does this study mean, and what should happen next? If it does not answer both parts, revise until it does.

The steps in this post, restate, synthesise, interpret, acknowledge limitations, and propose future directions, give you a reliable structure to work from. Apply them to your specific findings, write in your discipline's register, and keep every sentence tied to something concrete. That is how a conclusion becomes the strongest section of a paper rather than the weakest. Find more guidance on the full publication journey at the Publication Compass blog.