The dissertation is the most popular format for the Extended Project Qualification, and the most misunderstood. Students often approach its 5,000 words as a very long essay, then discover too late that examiners are rewarding something different: a question genuinely explored, evidence weighed rather than summarised, and a project visibly managed from first idea to final conclusion. The EPQ is worth half an A level, so the standard expected sits deliberately between school work and university work — and that bridge is exactly what this guide walks you across. It covers shaping a question, structuring the argument, researching properly, writing and referencing at the level examiners reward, keeping the production log that carries so many process marks, and finishing with the presentation. For the wider picture — formats, grading and timelines — see our complete EPQ guide.
One point of honesty before anything else. The EPQ is your own project: it is supervised by your school or college and authenticated as your work, and it must be. Nobody can legitimately write it for you, and any service suggesting otherwise is putting your qualification at risk. What good help can do — and what this guide tries to do — is teach you the method: how to plan, research, structure and revise, so that the project you submit is both genuinely yours and as strong as you can make it.
Start with a debate, not a topic
The single biggest predictor of a strong EPQ dissertation is the shape of its question. A topic — “social media”, “the death penalty”, “artificial intelligence” — cannot be answered, only described, and description caps your grade no matter how polished the prose. A debate-shaped question, by contrast, has competent people on at least two sides, evidence that can be weighed, and a conclusion you have to earn. Compare “The History of the NHS” with “To what extent has the internal market improved or damaged patient care in the NHS?” The first invites a summary. The second demands a judgement, and judgement is where the higher assessment objectives live.
Test a draft question against four checks. Can it be argued both ways using real evidence? Are enough credible sources actually accessible to a sixth-form student, rather than locked behind specialist paywalls? Is it narrow enough to answer in 5,000 words — one policy, one period, one population or one text, rather than a whole field? And does it still interest you, because you will live with it for two terms? If a question fails a check, narrow it: restrict the years, the place, the group affected or the aspect you will judge. Our guide to how to choose a dissertation topic works through that narrowing in detail.
Expect your supervisor to challenge the first version. That is not a setback; refining the question in discussion, and recording why it changed, is itself evidence of project management for your production log.
A workable structure for 5,000 words
There is no single compulsory structure, but a pattern that serves most dissertation-format projects looks like this. Adapt the proportions to your question rather than forcing your material into them.
- Introduction (roughly 400–500 words). State the question, explain why it matters and what is genuinely contested about it, define any key terms, and set out briefly how the dissertation will proceed. Write the final version last, when you know what you actually argued.
- Literature and context chapters (roughly 1,200–1,500 words). Map what is already known and argued: the main positions, the key studies or sources behind them, and where they disagree. This is not a parade of summaries. Group sources by position or theme, compare them as you go, and end the section by identifying the gap or tension your own argument will address.
- Argument chapters (roughly 2,400–2,800 words). The heart of the project, usually two or three chapters, each developing one strand of your answer. Each chapter should state its claim, present the strongest evidence for it, confront the best counter-argument honestly, and reach a small conclusion that feeds the overall judgement. Chapters that merely present “for” then “against” without weighing them waste the format; the weighing is the work.
- Conclusion (roughly 400–500 words). Answer the question directly, in terms of the question’s own wording. State how far, under what conditions, or for whom your answer holds. A conclusion that introduces new evidence, or that sits on the fence your chapters have already climbed down from, undermines everything before it.
- Reference list and any appendices. Every source cited, in one consistent style, plus anything your board requires. These sit outside the 5,000 words in most specifications, but check your own board’s rules.
Notice what this assumes: most of your words go on argument, not background. Students routinely invert that, writing 3,000 words of context and rushing the analysis. Plan chapter word counts before you draft, and hold yourself to them.
Researching a library-based project
A dissertation-format EPQ is a library-based project: your evidence comes from published sources, not from experiments or surveys of your own. That makes the quality of your searching decisive. Start from your question’s key terms and their synonyms, and search deliberately rather than typing the whole question into a general search engine. Use your school or college library catalogue, journal databases your centre subscribes to, and reputable open sources such as university and government sites. Keep a running record of what you searched, where, and what you found — your supervisor and examiner both want to see selection happening, and your log needs those details.
Read with a job in mind. For each source, note its central claim, the evidence underneath it, its date and limitations, and where it sits relative to your other sources: does it support, complicate or contradict them? A simple table does this well. Examiners reward discrimination, so practise judging sources rather than collecting them: a peer-reviewed study, a broadsheet opinion piece and an advocacy website can all be used, but not as if they were the same kind of evidence.
Method still matters even without fieldwork. Deciding your search terms, your inclusion rules — dates, types of source, relevance — and how you will organise what you find is your methodology, and the reasoning transfers directly from undergraduate work: our guide to how to write a dissertation methodology explains how to justify such choices rather than merely describe them. Write a short version of that reasoning into your dissertation’s opening chapters and a fuller version into your log.
Academic writing and referencing at the bridge level
EPQ writing should sound like early university writing: precise, cautious where the evidence is cautious, and always clear about whose claim is on the page. Three habits lift a draft fastest. First, signpost honestly — “Smith argues that…”, “This evidence suggests, but does not establish, that…” — so the examiner can tell source material from your own judgement. Second, quote sparingly and analyse what you quote; a dissertation built from other people’s sentences demonstrates reading, not thinking. Third, prefer hedged, exact claims over sweeping ones: “in the two trials examined” beats “science proves”.
Reference everything you use, in the style your centre specifies — commonly Harvard at this level — and reference as you write, not in a panic at the end. Every paraphrase as well as every quotation needs a citation, because the idea is borrowed even when the words are yours. Keep the list complete and consistent; accurate referencing is straightforward evidence for the “use resources” assessment objective.
The production log: where the process marks live
Students who treat the production log as paperwork to complete in the final week throw away marks they have already earned. The log is the record of your planning, decisions and changes across the whole project, and it carries the process marks: it is the main evidence for AO1 (manage) and much of AO4 (review), while your dissertation itself chiefly evidences AO2 (use resources) and AO3 (develop and realise). A brilliant dissertation with a thin, retrospective log cannot score as a brilliant project, because the examiner cannot see the management that two of the four objectives assess.
Keep it contemporaneously — little and often, dated, as you work. Record your initial ideas and why you rejected some of them. Record the question’s evolution and the reasons for each refinement. Record research sessions: databases searched, terms used, sources found, kept or discarded and why. Record supervisor meetings and what you did about the advice. Record problems honestly — a source that turned out to be unusable, a chapter plan that collapsed, a week lost — and, crucially, what you changed in response. A log in which nothing ever went wrong convinces nobody; examiners trust the project whose log shows real decisions because real projects involve them.
Review as you go, not only at the end. Short reflective entries after each stage — what is working, what is weak, what happens next — build the AO4 evidence steadily and make the final evaluation almost write itself, because you are drawing on judgements made at the time rather than inventing a tidy story afterwards. The final evaluation should address the product against your original aims, what you would do differently, and what the project taught you. Specifics beat superlatives: name the decision you would reverse and the evidence that changed your mind.
The final presentation
The project ends with a presentation to a non-specialist audience, followed in most centres by questions. Treat it as a communication task, not a summary read aloud. Choose the few things a non-specialist needs: the question and why it matters, the shape of the debate, what you concluded and the strongest reason for it, and one honest difficulty from the process. Practise aloud against your centre’s time limit, prepare for the obvious questions — why this question, what was the strongest argument against you, what would you do differently — and remember that a thoughtful answer about your process is itself evidence for the review objective. Visual aids should support the story, not reproduce your chapters on slides.
Get guidance on your EPQ dissertation
If your question will not narrow, your chapters will not balance, or your log has gaps you do not know how to repair honestly, guidance at the planning stage is worth far more than rescue at the deadline. Our EPQ support is coaching, not substitution: we review your question and chapter plan, teach you the research and referencing method, comment on drafts you have written, and supply model material you can study and learn from — the project itself remains yours, as it must. Support is priced from £20 per 1,000 words of material reviewed or prepared. To start, send your brief — your question, your board, your deadline and the stage you have reached — and place an order through the site, email team@unitedacademichelp.uk, or message us on WhatsApp for a quick reply about what help would genuinely move your project forward.
Frequently asked questions
How long should an EPQ dissertation be? The dissertation format is 5,000 words, covering your introduction, chapters and conclusion. Reference lists and appendices are normally additional, but confirm the exact rules with your centre, because boards differ in detail.
Can someone write my EPQ dissertation for me? No. The EPQ must be your own work, supervised and authenticated by your school or college, and submitting someone else’s project risks the qualification itself. Legitimate help is guidance — planning, research skills, structure, feedback on your drafts and model material to learn from.
What makes a good EPQ question? A genuine debate that evidence can settle part-way: arguable on both sides, researchable from sources you can actually access, and narrow enough to answer properly in 5,000 words. If you cannot imagine a well-informed person disagreeing with your answer, the question is probably descriptive.
How important is the production log, really? Very. The log records your planning, decisions and changes and carries the process marks — it is the main evidence for the manage and review objectives. A strong dissertation with a weak log will lose marks that have nothing to do with the dissertation’s quality.
How many sources should an EPQ dissertation use? There is no set number, and examiners reward discrimination over volume. A project drawing thoughtfully on a few dozen sources of mixed types — studies, books, official material, quality journalism — each visibly used, is healthier than a long list the text never touches.
What happens in the EPQ presentation? You present your project to a non-specialist audience and usually answer questions. Cover the question, the debate, your conclusion and one honest difficulty, and practise aloud to your centre’s time limit.