Thesis Defense Questions: What Gets Asked and How to Answer

The short answer

Thesis defenses ask six questions in some form almost every time: summarise your work, why this method, what are the limitations, what is your contribution, how does this relate to a specific author or study, and what would you do differently. Everything else comes from your own text — examiners read your thesis and ask about the parts where the argument is thinnest. Prepare by finding those parts yourself first. The defense is not a test of whether you know the field. It is a test of whether you understand the decisions you made and the limits of what they can show.

That framing matters for how you prepare. Reading more literature is the instinct and it is mostly wasted effort. Re-reading your own thesis with a pen, marking every sentence you could not currently justify, is the work that pays.

This post is about preparation. If what you are feeling is dread rather than uncertainty, the more useful pieces are thesis defense anxiety and imposter syndrome at the defense — come back here once the panic has somewhere to go.

What examiners are testing with each question

Almost every question in a viva is a proxy for one of four things. Recognising which one is being probed tells you what a good answer contains:

• Do you understand your own method? Not whether it was the best possible method — whether you know why you used it and what it can and cannot deliver.
• Do you know the limits of your evidence? The single most common source of lost marks is a claim that outruns the data behind it.
• Can you locate your work in the field? Where it sits relative to what already existed, and what it adds.
• Can you think on your feet about your own subject? This is what the unexpected question is for, and it is why a thoughtful admission of uncertainty scores better than a confident guess.

If you are ever mid-answer and unsure where to go, ask yourself which of these four the examiner is actually after. It usually redirects the answer within a sentence.

The six questions that come up almost every time

1. Summarise your thesis in three minutes. Often the opener. Have a version that works at one minute, three minutes and ten. It should contain the question, the method, the main finding and why the finding matters — in that order.
2. Why did you choose this method? Covered below, because it is the question candidates handle worst.
3. What are the limitations of your study? Also below. Have four ready.
4. What is your contribution? The question that most often produces a vague answer from a candidate who has a perfectly good one.
5. How does your work relate to the work of X? X is usually someone central in your literature review, or someone the examiner has published on. Know the three or four names your argument leans on well enough to say where you agree and where you diverge.
6. What would you do differently? A generous question, not a trap. It invites you to show methodological judgement. Answer with a specific design change and the specific problem it would fix, not with a general wish for more time or more data.

Two more appear frequently enough to prepare: what surprised you? and where would you take this next? Both are easy marks if you have thought about them for ten minutes beforehand and awkward if you have not.

Why did you choose this method?

The weak answer describes the method. The strong answer explains the choice — which means naming the alternative you rejected and why.

A structure that works for nearly any field:

1. What the research question required. A question about how many needs different tools from a question about why or how.
2. The alternative you considered. Name a real one you thought about.
3. Why you rejected it. Access, sample size, timescale, ethics approval, the nature of the phenomenon.
4. What your choice cost you. Every method trades something away. Saying what yours traded away is the part that reads as maturity.

Note that a constraint is a legitimate reason. Choosing interviews because a survey could not reach the population in the time available is a real methodological argument, and stating it plainly is stronger than inventing a theoretical justification after the fact. Examiners have supervised enough theses to recognise a reconstructed rationale.

What are the limitations of your study?

Prepare four, from four different categories, and know which one is the most serious. The categories:

• Sample. Size, selection, self-selection, a single institution or single country.
• Method. What your instrument cannot detect. Self-reported data, a single coder, no control group, cross-sectional rather than longitudinal.
• Scope. The time window, the sector, the sources you excluded and why.
• Your own position. In qualitative work especially — how your involvement shaped what you saw and what you did about it.

Then, for each one, be able to say what it does and does not undermine. A limitation that narrows what you can generalise is different from one that threatens the finding itself, and the ability to tell those apart is exactly what is being assessed.

The failure mode here is not having too many limitations. It is treating them as an apology. State them as boundaries of a valid result, not as flaws in a defective one.

What is your contribution?

Candidates freeze on this because they hear it as a demand for something groundbreaking. It is not. At bachelor and master level a contribution is usually one of these:

• Applying an existing framework to a context where it has not been tested.
• Evidence on a question where the existing literature disagrees.
• Data from a population or setting that is under-represented in the field.
• A synthesis that clarifies a body of work nobody had put side by side.
• A negative or null result that narrows what other people need to test.

Write your answer as one sentence: before this thesis, X was unknown or untested; this work shows Y. If you cannot fill that sentence in, spend an hour on it before the defense — it is also the sentence that improves your abstract, your introduction and your conclusion.

How to prepare in two weeks

Two focused weeks is enough for most bachelor and master defenses.

Week one — re-read your own thesis as a critic.

• Read it start to finish with a pen. Mark every sentence you could not defend for sixty seconds. That list is your real question bank.
• Write the one-sentence version of the question, the finding and the contribution. Word perfect, because these three you will say out loud.
• List your four limitations and rank them. Assume the worst one will be asked about.
• Re-read the four or five sources your argument depends on most. Not the whole bibliography — the load-bearing ones.
• Check your own numbers, tables and figures. Being asked where a figure came from and not knowing is a bad moment that is entirely avoidable.

Week two — rehearse out loud.

• Answer each of the six questions aloud, timed, aiming for two to three minutes each. Out loud is the whole point: an answer that is clear in your head can fall apart in speech.
• Get one person to play examiner, ideally someone outside your field. If they cannot follow your explanation, tighten it.
• Record one run and listen back once. Once is enough, and it will show you the filler words and the places you trail off.
• Practise the transitions: how you buy three seconds, how you ask for a question to be repeated, how you close an answer instead of letting it fade.
• Confirm the logistics with your department — length, slides, who attends, whether you may bring notes or a printed copy of the thesis. Most institutions allow the printed copy, and having it in front of you is worth more than most preparation.

How to say you do not know

Every candidate hits at least one question they cannot answer. It is expected. What is assessed is what you do next.

The pattern that works has three parts:

1. Acknowledge the limit plainly. That is not something my data can answer. No apology, no waffle.
2. Give what you do have. Based on the pattern in chapter four, I would expect — flagged clearly as an expectation, not a finding.
3. Say how you would find out. Testing that would need a longitudinal design, or a second coder, or access to the population I could not reach.

Three related moves worth having ready. If you did not understand the question, ask for it to be rephrased — this reads as care, not weakness. If you need a moment, take it out loud: that is a good question, let me think about it for a second. And if an examiner corrects you and they are right, say so and move on; arguing a point you have lost costs far more than the original error.

What does not work is guessing in a confident voice, agreeing with a critique you have not understood, or answering a different question because it is the one you rehearsed.

The mistakes that cost marks

• Overclaiming. Defending a broad generalisation from a small study, under pressure, when conceding the limit was available. The most common way a good thesis gets a worse mark.
• Not knowing your own thesis. Being unable to locate a section or explain a figure you wrote three months ago. Re-reading fixes it entirely.
• Reciting a memorised answer that does not match the question actually asked. Examiners hear it immediately and follow up harder.
• Answering the tone rather than the substance of a question that sounded sharp. Almost always misjudged.
• Running long on the presentation and eating into the discussion. Time it, and time it out loud.
• Treating limitations as confession. They are the boundaries of a valid result. Say them like it.

A last thing worth holding on to: if your thesis was accepted for defense, your examiners already believe the work is defensible. The viva is the conversation about a piece of research you know better than anyone else in the room. Prepare the six questions, mark the thin paragraphs, rehearse out loud, and let the rest be a conversation.