How to Write a Dissertation Conclusion That Answers Your Question and Ends Strong

The analysis is done, the discussion is drafted and the deadline is close. Knowing how to write a dissertation conclusion matters now, because the final chapter is your last chance to show what the project achieved. Before you draft a word, settle these questions:

• What exactly does my conclusion need to answer, and in which order?
• How is it different from my discussion chapter?
• How many words should I give it?
• Can I revisit a point I only touched on earlier, or does that count as new information?
• Do my limitations and recommendations belong here or in the discussion?
• How do I finish without a flat last line or a cliché?

The answers draw on writing guidance from the University of Oxford, the University of Southampton, the University of Warwick and other published sources. Rules differ between departments, so your handbook and your supervisor always have the last word.

How to Write a Dissertation Conclusion in Five Moves

A dissertation conclusion answers your research question in plain terms, shows what your findings add, admits what they cannot show and points to what should happen next. Most university guidance builds the chapter in roughly this order:

1. Answer the question. State the answer to your main research question or hypothesis in the first lines.
2. Synthesise. Combine the main findings into one coherent message instead of replaying each chapter.
3. Name the contribution. Say what the reader now understands that the introduction could not yet tell them.
4. Weigh the limits and the next steps. Admit the limitations that matter, then give specific recommendations for practice or further research.
5. Close. End with a final paragraph that finishes the work without suggesting the topic is exhausted.

Oxford’s Academic Writing Hub describes the same content as four moves: summarising the research, applying it in practice (in applied fields), evaluating it and suggesting future work. The labels differ, but the logic is shared. Look back, judge, then look ahead.

Why the Final Chapter Runs in the Opposite Direction to Your Introduction

The University of Southampton’s Writing the Dissertation guide compares the shape of a dissertation to an hourglass. The introduction and literature review move from broad to narrow, the methods, results and discussion stay narrow, and the conclusion moves from narrow back to broad. In practice, that means you restate your focused answer first, then widen to what it means for your field or profession.

Reread your dissertation introduction before drafting. The broad theme that opened the project should reappear in your final paragraph, now with an answer attached. Southampton suggests asking what the reader knows now that they did not know at the start. That single question is a useful test for the whole chapter.

The same guide also suggests writing as if a reader will see only the abstract, the introduction and the conclusion. Those three parts should therefore agree about the question, the answer and the contribution.

Dissertation Conclusion vs Discussion vs Abstract: Where Each One Stops

The three sections overlap in subject but not in job. The Royal Literary Fund’s dissertation guide puts it simply: the discussion explains and argues the main points, while the conclusion restates the most fundamental ones briefly, without the supporting argument. The abstract goes further still, summarising the whole project with even less detail.

Section	Main job	Level of detail	New material?
Discussion	Interprets findings against the literature and builds the argument	High, with evidence and citations	New interpretation, yes
Conclusion	Answers the question, states the contribution, points forward	Brief and general	No new data, sources or arguments
Abstract	Gives a skimming reader the whole project in miniature	Lowest	No

A quick sorting test helps when a sentence feels stuck between chapters. If it needs evidence or a citation to stand up, it belongs in the discussion. If it states what that evidence adds up to, it belongs in the conclusion. For the chapter before this one, see writing a strong discussion chapter, and for the one before that, the dissertation results chapter guide.

Some universities merge the two. Scribbr’s UK guide notes that combined sections are common in shorter research papers, although a dissertation usually keeps a separate final chapter. Check your department’s template before you plan the structure.

What to Include in a Dissertation Conclusion, Move by Move

Answer the research question in the first lines

Open with the answer, not with a reminder that your project looked at a topic. Restate the aim in a clause, then give the finding. The two sentences below come from an invented hybrid-working study, so you can see the difference:

• Weaker: This dissertation examined hybrid working and feedback.
• Stronger: Hybrid working changed how junior analysts experienced feedback, but only where managers set out review criteria in advance.

Scribbr recommends turning your aims and questions into one statement of what you did and found, rather than pasting them in from the introduction. If you set several questions, answer them in the order you framed them in your original research question.

Synthesise findings instead of replaying chapters

Synthesis means bringing separate findings together into one view. Southampton advises giving each main finding a sentence or two, then condensing and connecting them into a short portrait of the project. Group findings by idea, not by chapter. A reliable pattern is cause, condition, limit: A led to B, but only when C was present.

Resist the urge to repeat statistics. Southampton’s own example shrinks a run of sample-by-sample percentages into one headline change, which is the right level of detail here. Keep the full numbers in the results chapter.

State your contribution without overclaiming

Scribbr suggests three ways to show what your work adds: return to the problem statement, refer to the gap you identified in your literature review, or show how your findings confirm or challenge an existing theory. Choose the one that fits your project and say it plainly.

Match the force of your wording to the strength of your evidence. GradCoach cautions that one study rarely proves or disproves anything, so avoid claiming more than your data allows. At the same time, Warwick warns against being too modest, because the conclusion is where you present your contribution.

Pair your limitations with your recommendations

Oxford treats limitations and future research as linked moves. The weaknesses you name become the questions the next study could answer. Choose the two or three limits that genuinely shape how your findings should be read, explain how they might affect them, and then use them to justify what should happen next. If your discussion already covers limitations in depth, keep the conclusion version to a sentence or two.

Make recommendations specific. A line such as “more research is needed” tells the reader nothing. Name the design, population or method that would test your finding further. For example: a longitudinal study that follows the same analysts across a full review cycle could test whether predictable feedback keeps its effect over time.

Scribbr advises phrasing practical recommendations as “shoulds” rather than “musts”, since research informs rather than demands. Skills You Need adds that recommendations for practitioners are strongly advised in management and business courses and some other fields, so ask your supervisor which applies to your subject. Level matters as well. Warwick suggests that a Master’s student might point towards a PhD project, while a doctoral student might point towards post-doctoral work.

Close without closing the topic

Your last paragraph should give a sense of finish without implying the subject is exhausted. Southampton frames the ending as the close of one piece of research that another researcher could build on. Return to the broad theme from your introduction, state what the reader now understands, and stop.

Skip “In conclusion” as an opener. Southampton and Scribbr both advise against it, Warwick calls formula phrases of this kind clichés, and the chapter heading already tells the reader where they are. Avoid saving your main claim as a final-chapter surprise too. Academic writing signals its answer from the abstract and introduction onwards.

How Long Should a Dissertation Conclusion Be?

A dissertation conclusion is usually the shortest chapter in the project. Published guidance gives different ranges, which is why your handbook should settle it. Scribbr and GradCoach both suggest about 5-7% of the total word count, while Southampton gives 5-10% and says most conclusions sit nearer 5%.

Dissertation word count	At 5%	At 7%	At 10%
8,000	400	560	800
10,000	500	700	1,000
12,000	600	840	1,200
15,000	750	1,050	1,500
20,000	1,000	1,400	2,000

Those figures are arithmetic, not rules. Scribbr notes that empirical scientific studies tend to run shorter, while humanities topics and systematic reviews may need more room to tie the argument together. Skills You Need says a conclusion rarely needs more than five pages, though detailed practical recommendations can take extra space. To check your own numbers, try the dissertation word count calculator.

Where the Conclusion Sits in Your Discipline

Placement and naming vary by field. Oxford’s guidance says a humanities or social science conclusion is often a stand-alone chapter that includes every move, whereas in the medical sciences many of those moves sit in the discussion and the final chapter may be only one or two paragraphs. Some science departments combine both into a single “Discussion & Conclusion” chapter. Oxford also publishes tables of contents showing different end sections across departments.

• Stand-alone final chapter: typical of humanities and many social science dissertations, and it carries all four moves.
• Short conclusion after a full discussion: seen in medical sciences, sometimes only a paragraph or two.
• Combined chapter: common in science departments, so the conclusion becomes the closing section of the discussion.
• Conclusions and future work: an Oxford MPLS example titled “Conclusions and Future Work” gives future work its own subsection.

Warwick’s academic English guidance adds another layer for long projects. Each chapter can end with its own short conclusion that reminds the reader what was done and builds a bridge to the next chapter, separate from the overall conclusion. It also recommends reading past students’ work in your department, which is the quickest way to see what your markers expect.

Dissertation Conclusion Example With Notes

The sample below is invented for demonstration and belongs to a hypothetical Master’s study of hybrid working. It is deliberately short, so a real chapter will run longer and cite your own sources.

This dissertation set out to establish whether hybrid working changed how junior analysts experienced feedback from their managers. It has shown that it did, but unevenly. Analysts who worked mostly from home rated scheduled reviews as useful only when managers had shared written criteria beforehand, while office-based analysts relied more on informal comments.

Taken together, the survey and interview findings suggest that the predictability of feedback mattered more than where it was delivered. The survey showed only a small difference between the two groups once review timing was held constant, and the interviews explained why: participants valued knowing when and how they would be assessed.

These findings qualify accounts that treat physical presence as the main driver of early-career development. They also add evidence from a junior group that the literature review identified as under-studied. However, the sample came from three firms in one sector, so the results may not transfer to smaller employers or to roles with less structured reviews.

A comparative study across sectors could test whether predictable feedback keeps its effect where roles are less formal. Employers could pilot written review criteria for remote staff and track whether ratings of feedback change. What is clearer now than at the start of this project is that the debate over where people work may matter less than how plainly organisations tell staff how they are doing.

• Paragraph one states the aim in a clause and then gives the answer. The present perfect (“has shown”) signals the shift to a reflective voice.
• Paragraph two synthesises two data sources into one claim and keeps the numbers out.
• Paragraph three states the contribution through the literature review gap, then names one proportionate limitation with a cautious verb (“may”).
• Paragraph four offers one specific research direction and one practical recommendation, using “could”. The final sentence returns to the broad theme an introduction would open with.

Real examples with analysis appear in the Southampton and Oxford guidance mentioned earlier. Southampton includes an annotated conclusion, and Oxford reproduces abridged conclusions from its own dissertations in the social sciences, humanities and MPLS.

Language That Makes a Conclusion Confident, Not Careless

• Tense. Southampton points to the present perfect as the signal of a reflective register (“This dissertation has considered…”). Warwick suggests following it with the simple past for specific steps, such as what a chapter provided.
• Person. Stay consistent with the rest of the dissertation. Third person is typical, but some disciplines and reflective genres accept first person, so confirm with your supervisor.
• Modal verbs. Oxford explains that “will” and “must” carry the most confident stances, “may” and “might” signal tentative ones, and “could” often recommends a research direction. Pick the verb that matches your evidence.
• Phrase banks. Manchester’s Academic Phrasebank groups conclusion language by function, including restating aims, summarising findings, noting limitations and recommending further work. Borrow the patterns, then write sentences specific to your study.
• Citations. Southampton advises re-citing earlier sources when you revisit them, especially when you cite more than one work by the same author.

Common Dissertation Conclusion Mistakes and How to Fix Them

• Introducing new material. New data, theories or sources belong in the results or discussion. Move them there, or cut them.
• Replaying chapters one by one. Rebuild the chapter around your research question and the two or three insights that answer it.
• Repeating statistics. Replace a string of figures with the headline finding and let the results chapter carry the detail.
• Overclaiming. Swap “proves” and “demonstrates conclusively” for “suggests” or “indicates” unless your design genuinely supports the stronger word.
• Apologising for the work. Warwick and Scribbr both advise against dwelling on weaknesses. State the limit, explain its effect, and move on to what the study did achieve.
• Holding back the answer. Southampton warns against saving your core argument for a final reveal. Put the answer in the opening lines.
• Drifting into opinion. Warwick cautions that the conclusion is not a place for a personal rant. Anchor every statement in your findings.
• Vague recommendations. Name who should act, or which study design, population or method would test your result next.

Adjusting Your Conclusion to Your Research Design

The five moves stay the same, but the emphasis shifts with the design you described in your dissertation methodology.

• Quantitative: say whether each hypothesis was supported, in the order you stated them, with direction and rough size of effect rather than test statistics.
• Qualitative: bring your themes together into one claim and comment on how far the findings may transfer. Avoid adding fresh participant quotes, because a new quote is new data.
• Mixed methods: explain where the strands agreed, where they diverged, and what the combination showed that neither could alone.
• Literature-based or systematic review: summarise the state of the evidence, its quality and the gaps it leaves. This type may need more space, as Scribbr notes.
• Case study: explain what the case shows and how far it applies beyond its own setting.
• Applied or professional fields: Oxford’s second move covers practical applications and specific recommendations to policy-makers, industries or companies, so give it real space.

Final Checks Before You Submit Your Conclusion

• Does the first paragraph answer the main research question directly?
• Could a reader who only sees your abstract, introduction and conclusion tell what you asked, what you found and why it matters?
• Does every claim already appear, in substance, earlier in the dissertation?
• Are your limitations proportionate, and does each one lead to a specific recommendation?
• Is the length within your handbook’s guidance?
• Do tense, person and terminology match the rest of the dissertation?
• Does the last paragraph return to the theme of your introduction?
• Have you removed “In conclusion” and any other stock opener?

Dissertation Conclusion FAQs

How long should my dissertation conclusion be?

Aim for roughly 5-7% of your total word count unless your handbook says otherwise. Southampton allows up to 10% but notes that most sit nearer 5%. For a 10,000-word dissertation, that means about 500 to 700 words.

Can I include new information in my dissertation conclusion?

No. New data, arguments and sources belong in earlier chapters. You can revisit earlier material and cite it again, but the conclusion should rest on what the reader has already seen.

What is the difference between the discussion and the conclusion?

The discussion interprets your findings in detail against the literature. The conclusion is shorter and more general: it answers your research question, states your contribution and looks ahead without re-arguing the evidence.

Should my conclusion include limitations?

Often yes, but briefly. Scribbr advises mentioning important limitations in the conclusion only when the discussion has not already covered them. Follow your department’s template, because conventions differ.

Do I need recommendations in my conclusion?

Suggestions for future research are expected in most guidance. Recommendations for practitioners depend on your field, and Skills You Need strongly advises them in management and business subjects. Southampton notes that some departments place both in the discussion chapter instead.

Should I write my conclusion in the first person?

Use whichever person you have used throughout the dissertation. Third person is typical, while some disciplines accept first person. If you are unsure, ask your supervisor and check a high-scoring past dissertation from your department.

How do I end a dissertation conclusion?

Widen back to the broad theme from your introduction and state what the reader now understands. Give a sense of closure without implying that the subject is finished, and avoid clichés such as “In conclusion”.

Next Steps for Your Closing Chapter

Draft the conclusion once your discussion is stable, then return to the introduction and abstract so all three agree. Scribbr suggests writing the abstract next, while the research is still fresh. Block out the time for those final edits in a dissertation timeline planner before the deadline squeezes them.

If your draft still reads like a summary, rewrite the opening so it answers the research question in two sentences, then rebuild the rest around it. For a second opinion on structure, tone and flow, support with your conclusion chapter is available once your draft is ready.

Sample text, including the example above, is for reference and learning. Use any support in line with your institution’s policy.