How to Prepare for a Thesis Defense (Viva)

First: find out which kind of defense you're having

"Thesis defense" means two quite different events depending on where you are, and preparing for the wrong one wastes weeks.

• The UK-style viva voce is a closed oral examination. You normally face two examiners — one internal to your university, one external — neither of whom has supervised you. UK universities commonly describe vivas as running roughly 90 minutes to three hours. There is usually no presentation and no audience. Some institutions also appoint an independent or non-examining chair: a procedural role who has not read your thesis, takes no part in the academic decision, and is there to make sure the process is fair and that you get a break.
• The US-style dissertation defense usually opens with a public seminar — often around 45–60 minutes — that colleagues and friends may attend, followed by closed questioning from your committee. You then step out while the committee deliberates.

Formats, time limits and protocols vary by institution and even by department, so treat every generic guide (this one included) as a starting point and your own regulations as the authority.

Week 6–5: read your regulations, then read your thesis

Download your university's research degree regulations and find two things: the list of possible outcomes, and the deadline attached to each. Knowing that "minor corrections" is a normal, expected result — with a defined window to complete it — removes a surprising amount of the fear.

Then re-read your thesis in one or two sittings, as an examiner would, with a pen. You are not proofreading. You are building three things:

• A one-page map. For each chapter: its job in the argument, its single main claim, and the evidence behind it.
• An errata list. Every typo, wrong cross-reference and mislabelled figure you spot. Bring it. Handing examiners a list of errors you already found reads as command of your own work.
• A weak-points list. The paragraphs you'd rather they skipped. They won't. Every one of them needs a prepared, honest answer.

Week 4: prepare the four questions behind every question

Examiners ask dozens of questions, but nearly all of them are probing four things. Prepare these properly and you can improvise the rest.

1. What is your original contribution?

The single most predictable question in any defense — and the one most candidates answer vaguely. Aim for one sentence a non-specialist could repeat: "Before this thesis, X was measured only in laboratory settings; I show it also holds in Y, and that the effect weakens when Z." Then have a two-minute version with the evidence.

2. Why this method, and not the obvious alternative?

For every major design decision, know the alternative you rejected and why. "I used semi-structured interviews rather than a survey because the construct wasn't stable enough to pre-code" is a defensible answer. "It's what my lab does" is not.

3. What are the limitations?

Name them before the examiners do, and state what each one does and doesn't undermine. Candidates who over-claim get taken apart; candidates who calmly bound their own claims rarely do. If you wrote a strong discussion section, most of this material is already drafted.

4. Where does this sit in the literature — and what next?

Be ready to name the three or four papers your work is in direct conversation with, including one you disagree with and why. Expect "if you started again tomorrow, what would you do differently?" — a question about judgement, not regret.

Week 3: know your sources cold

Examiners read your reference list closely; it is a fast map of whether you know the field. Two failure modes come up again and again: a foundational paper missing, and an entry that doesn't match the source it claims to be. Work through your bibliography and confirm each entry resolves to a real, correctly described work — especially anything you added late, pulled from another paper's reference list, or drafted with AI assistance.

Week 2: run a mock viva

Nothing else in your preparation does as much work. Ask two people — ideally one outside your immediate subfield, because that's what an external examiner effectively is — to read at least your abstract, introduction and conclusions, and question you for an hour. Practise thinking out loud rather than memorising answers; a scripted response collapses at the first follow-up. Record it if you can bear to: most people discover they answer a different question from the one asked.

Week 1 and the day itself

• Bring a clean copy of the thesis with tabs on key figures, tables and your errata list.
• Listen to the whole question before answering. A pause is fine; examiners read it as thinking, not floundering.
• If a question is unclear, ask them to rephrase it. That is a normal move, not a weakness.
• Concede real points. "That's fair — the sample doesn't support that claim, and I'd soften it" ends a line of questioning that arguing would extend.
• Ask for a break if you need one, especially in a long viva.

Understanding the outcome

Typical outcome categories are: pass with no corrections; minor corrections; major corrections; resubmission for re-examination; award of a lower degree such as an MPhil; and fail. The middle categories are the ordinary ones — at many UK universities minor corrections are due within about three months for full-time candidates and major corrections within six to twelve months, though the exact list and deadlines are your institution's to set.

Get the correction requirements in writing before you leave, and treat them the way you'd treat reviewer comments on a journal submission: a numbered list, an action for each, and a short note where you respectfully disagree.

Frequently asked questions

Keep reading

• How to Write the Discussion Section of a Research Paper
• How to Respond to Reviewer Comments
• How to Write a Literature Review Without Fabricating Sources
• Free Citation Generator