Written and maintained by CASRAI Editorial Board
Published · Last updated
On this page
A dissertation conclusion is the closing section or chapter in which you state what the study found, say what those findings mean, acknowledge their limits and point to what should happen next. It answers one question for the reader: what do we now know that we did not know at the start? The University of Southampton Library frames exactly that question as the test of a good dissertation conclusion, and it is the thread this guide follows.
Below you will find the purpose of the section, an ordered structure built from two university writing guides, a short annotated example, the mistakes those guides warn against, a checklist and a note on how placement differs by discipline. The data table at the end of the page lines up what each source expects the conclusion to contain. Your own institution’s regulations and your supervisor’s instructions override any general guidance, including this page.
Purpose of the dissertation conclusion
The conclusion brings the project to a close without claiming that the field is closed. Southampton’s guidance asks the writer to revisit the primary research purpose, meaning the research question or hypotheses, then to restate each main finding briefly and synthesise them, and finally to say what the work means for the field or for professionals who use it. The aim is closure for the reader, while leaving room for further work.
Oxford’s Lifelong Learning writing guide describes the same job as a sequence of moves: summarise, describe practical applications where the field is an applied one, evaluate the contribution and limits, and suggest future work. It also notes that where these moves appear varies by discipline, which matters for deciding whether you need a separate chapter at all (see the discipline section below).
A conclusion differs from the sections around it in one respect. The abstract and introduction are where the core argument is first stated; the results and discussion are where evidence is presented and interpreted. Southampton warns against stating the central argument for the first time in the conclusion and against introducing new primary data or new theories there, since those belong in the results or discussion. If you are still deciding how the whole document fits together, the chapter outline template shows where the final chapter sits, and what a dissertation is explains the overall form.
How long should a dissertation conclusion be?
Southampton’s guidance gives a rough rule: the conclusion is about 5 to 10 per cent of the total word count, usually nearer 5 per cent. That is one library’s rule of thumb, not a universal requirement. Your department, graduate school or degree regulations may set a different figure or none at all, and a conclusion folded into a discussion chapter may be only a paragraph or two (see the discipline section). For overall length expectations see dissertation length.
Structure of a dissertation conclusion
Two shapes are described in the sources, and they fit together. Southampton calls its shape an hourglass, the reverse of the introduction: the first part is narrow and the second part broad. Oxford calls it an inverted funnel that also runs from narrow to broad. Combined, the order below works for most research dissertations. Treat it as a default to adapt, not a template that every institution requires.
- Restate the aim. Return to the research question, aims or hypotheses (Oxford move 1.1; Southampton’s first element). Southampton notes that the findings in this narrow first part answer the problem, often by restating it.
- Summarise the approach and what was done. Oxford lists the methods (1.2) and the work carried out (1.3) as optional parts of the summary. Keep this to a sentence or two; the methods chapter already has the detail.
- State the key findings or points established. Oxford move 1.4; Southampton asks you to restate each main finding briefly and then synthesise them. Synthesising means saying what the findings add up to together, not listing them one by one.
- Give practical applications, if your field is applied. Oxford move 2 covers real-world problems the findings address (2.1) and recommendations to policy-makers or industry (2.2). Southampton adds that recommendations, if included, must be consistent with the findings, and that they may sit in the discussion chapter instead.
- Evaluate the contribution. Oxford move 3: the implications or contribution (3.1), the most valuable aspects of the work (3.2) and its limitations (3.3). Southampton frames this as the wider impact for the field or for professionals.
- Suggest future work. Oxford move 4 proposes further work that addresses the limitations identified in 3.3. Southampton’s second, broad part likewise returns to the introduction’s themes, future research and real-world relevance.
- Close. End on the significance of the work rather than on a trailing caveat. Southampton advises against implying that the area is finished.
The University of Rochester Medical Center’s thesis guide supplement for its statistics PhD describes a final chapter titled “Conclusions and Future Work” with a similar content list: a discussion of the findings within a unified framework, the limitations of the research and of any new methods, the future work and new questions the research raises, and a brief recap of the state of the field at the start, a summary of the main results and an explanation of the contribution. That is one programme’s requirement, and it shows how a department can spell out the same elements in its own words.
Limitations and future work: keep them linked
Oxford observes that limitations and future work are often closely linked, because the suggestions in move 4 respond to the limitations stated in move 3.3. It also describes two ways of arranging them: a parallel style, in which each contribution is paired with a limitation, and a block style, in which contributions are stated together and limitations together. Either is acceptable; what matters is that each limitation you name has a matching, concrete next step, so that the reader sees a plan rather than an admission. Oxford also notes that modal verbs such as may, could and might are used to hedge positions in the applications, evaluation and future-work moves, which keeps claims proportionate to the evidence.
A short annotated example
Illustrative composite written for this guide; it does not quote or describe any real dissertation, author or institution. The paragraph below follows the order above for an imagined study of how first-year students use feedback. Square-bracket notes label each move.
This dissertation has asked how first-year students act on written feedback and whether the format of that feedback changes what they do next. [Restates the question, with a reflective present-perfect opening instead of “In conclusion”.] The study combined a survey with follow-up interviews. [One-sentence method recap.] It found that students acted most often on feedback that named a specific next step, and less often on general comments. [Key finding, stated once, without repeating every statistic.] Taken together, these results suggest that the form of feedback, and not only its quality, shapes its use. [Synthesis: what the findings add up to.] For teaching staff, this may mean writing comments as actions. [Practical application, hedged with a modal verb.] The study drew on a single course and self-reported behaviour, so it cannot show that the format caused the difference. [Limitation, stated plainly.] A comparison across several courses, with observed rather than reported behaviour, would address both points. [Future work tied directly to the limitation.] The work therefore adds a narrower, testable claim to a question that remains open.
Notice what the example leaves out: there are no new results, no new theory and no restated table of statistics. Everything it says was already established in earlier chapters.
Common mistakes in a dissertation conclusion
Southampton’s guidance lists several things to avoid. Each is stated here as that source states it.
- Repeating excessive detail or statistics. Summarise findings; do not re-present results.
- Introducing new primary data or new theories. These belong in the results or discussion chapters.
- Opening with “In conclusion”. The guide calls this cliched and suggests a reflective opening in the present perfect, such as “This dissertation has considered…”.
- Stating the core argument for the first time. Signpost the argument in the abstract and introduction so the conclusion confirms rather than reveals it. The dissertation abstract sample guide shows how the abstract carries it.
- Implying the area is finished. A conclusion gives closure to your project, not to the field.
For a wider list of structural and style problems across the whole document, see common thesis and dissertation writing mistakes.
Dissertation conclusion checklist
Use this as a final read-through. Every item traces to the Southampton or Oxford guidance above, or to the programme guide for Rochester.
- Does the opening return to the research question, aims or hypotheses?
- Does each main finding appear once, briefly, and are the findings then synthesised?
- Can a reader answer “What do we now know that we did not know at the start?” from this section alone?
- Is every recommendation consistent with the findings, and is it included only if your field and institution expect it?
- Are the contribution, the most valuable aspects and the limitations all stated?
- Does each limitation have a matching suggestion for future work?
- Is everything in the conclusion already established in earlier chapters, with no new data or theory?
- Does the length fit your institution’s rule, or, failing a rule, the rough 5 to 10 per cent guide?
- Does the final paragraph give closure without implying the field is finished?
Discipline differences
Oxford’s guide states that the placement of these moves varies by discipline, and that you should check the conventions of your own field. In short:
- Humanities, and some social sciences, often use a stand-alone concluding chapter.
- Medical sciences often place the moves in the Discussion section, leaving a conclusion of only one or two paragraphs.
- Some sciences merge the two into a combined “Discussion and Conclusion”.
Oxford also notes that the practical-applications move applies only to applied fields. If your dissertation is a thesis by publication or follows an unusual format, check your own regulations for what the concluding chapter must cover.
Frequently asked questions
What goes in a dissertation conclusion?
A restatement of the research aim, a brief summary and synthesis of the main findings, the contribution and wider relevance, the limitations, and suggestions for future work. Applied fields often add practical recommendations (Southampton; Oxford).
How long should a dissertation conclusion be?
Southampton suggests roughly 5 to 10 per cent of the total word count, usually nearer 5 per cent. Follow any figure set by your institution.
Should the conclusion include new information?
No. Southampton advises against new primary data or new theories in the conclusion; they belong in the results or discussion.
Is the conclusion the same as the discussion?
Not always. Oxford notes that in some fields, such as many medical sciences, the discussion carries most of these moves and the conclusion is very short, while others combine the two into one section.
Should I write recommendations?
Only if your field and institution expect them. Southampton describes recommendations as optional, consistent with the findings, and possibly placed in the discussion chapter instead.
Sources
- University of Southampton Library, “Writing a conclusion (dissertation)”: library.soton.ac.uk/writing_conclusiondissertation, read 2026-11-14.
- University of Oxford, Department for Continuing Education, “Writing Conclusions”: lifelong-learning.ox.ac.uk/writing-conclusions, read 2026-11-14.
- University of Rochester Medical Center, statistics PhD thesis guide supplement: urmc.rochester.edu thesis guide supplement, read 2026-11-14.
Data: Elements of a dissertation conclusion across three university guides
Compares which conclusion elements each of three university writing guides names (Southampton Library, Oxford Lifelong Learning, Rochester Medical Center statistics PhD supplement) and where Oxford says each placement occurs by discipline. Compiled from the guides' own text; 'Not stated' means the source does not mention the element.
Period covered: 2026.
| Element | Southampton Library | Oxford Lifelong Learning | Rochester Medical Center (statistics PhD) |
|---|---|---|---|
| Restate research aim, question or hypotheses | Yes: revisit the primary research purpose | Yes: move 1.1 | Not stated as such; recap of the state of the field at the start |
| Summarise main findings | Yes: restate each briefly, then synthesise | Yes: move 1.4, key points established | Yes: summarise main results |
| Recap of methods and work carried out | Not stated | Optional parts of the summary: moves 1.2 and 1.3 | Limitations of new methods are discussed |
| Practical applications or recommendations | Optional; consistent with findings; may sit in discussion | Applied fields only: move 2 | Not stated |
| Contribution or wider impact | Yes: impact for the field or professionals | Yes: move 3.1 and 3.2 | Yes: explain the contribution |
| Limitations | Not stated as a separate element | Yes: move 3.3 | Yes: limitations of the research |
| Future work | Yes: in the broad second part | Yes: move 4, addressing the limitations | Yes: future work and new questions raised |
| Suggested length | About 5 to 10% of word count, usually nearer 5% | Not stated | Not stated |
Sources
- University of Southampton Library: Writing a conclusion (dissertation) (accessed 2026-11-14)
- University of Oxford Lifelong Learning: Writing Conclusions (accessed 2026-11-14)
- University of Rochester Medical Center: statistics PhD thesis guide supplement (accessed 2026-11-14)
Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.







