Preparing for Your Thesis Defence: What to Say and What to Expect

Many students picture the defence as an oral exam in which they must know everything by heart. That picture is wrong. The defence is not a test of what you know; it is a test of how well you can defend the work you did. The committee is not checking how many books you read — it is checking whether you actually own the work.

That shift in perspective changes how you prepare. What you need is not memorisation but a second pass over the decisions you made during the research, and honest answers to the questions most likely to be asked. This article walks through what is assessed in a defence and how to get ready for it.

What is actually assessed

Three things appear in almost every defence. First, command of the work: do you know where each part of the text came from, why you chose this method, and what the data actually say. Second, awareness of limitations: do you know what your work does not show and where it is weak. Third, the ability to answer respectfully: can you stay calm under a critical question and respond, rather than become defensive.

The important point is that the first two cannot be manufactured in the last few days. If you did the work yourself, command and awareness are already there, and the defence only reveals them. If some part of the work was not yours, this is where it shows — and this is why the standing advice is to keep the trail of your work and your draft versions. They remind you of your own decisions, and if someone asks how you arrived at a result, you have a documented answer.

The three questions that almost always come up

First: why this topic? The committee wants to see whether you understood the problem or simply picked a topic that looked manageable. A good answer is short and starts from the problem, not from personal interest: where the question came from, what was missing in the existing knowledge, and which gap your work targets.

Second: what is your contribution? This is not the moment to list everything you did; it is the moment to make the difference between your work and everyone else's clear. A good answer can say "others have said this, and I added that". If you cannot state your contribution in one or two sentences, you probably have not made it clear to yourself yet.

Third: what are the limitations of the work? This is not a trap. The committee wants to see whether you recognise the limits yourself. A good answer has two parts: the real limitation, and what it removes from your conclusions. Admitting a limitation does not weaken your work; claiming there is none does.

Presentation: what belongs on a slide and what does not

Slides are a presentation aid, not a second copy of the thesis. A structure that usually works goes like this: the problem and the research question, a brief review of what others have said, the method, the findings, discussion and limitations, then the conclusion. A few short points per section is enough; if the audience can read the slide, they are no longer listening to you.

Avoid pouring text onto the slide. Full sentences, long paragraphs and crowded tables turn a presentation into a reading exercise and make you dependent on the screen, so the moment a question goes beyond the slide you lose the thread. Keep tables and figures, but only as much as is needed to make one point. Practise finishing within the allotted time as well; defence sessions usually allow a limited slot for the presentation, and overrunning eats into the question and answer period.

Have your key sentences ready

Write a few sentences in advance and rehearse them — not to memorise, but so you do not get lost in the moment. Your contribution sentence, your main limitation sentence, and your closing conclusion sentence. These three are the pillars of your defence, and if they come out smoothly, much of the anxiety disappears.

Devote one of those sentences to the role of AI in your work. If you used a tool, an honest and clear answer is the best route: say which part you used it for and which parts are entirely your own. Hiding this in the room carries more risk than saying it. Remember that under current regulations, the Iranian law on preventing and countering fraud in the preparation of academic works, and its implementing regulation, contain no clause about text produced with AI. There is no explicit ruling, and assessment falls back to institutional regulations and the judgement of examiners. In that space, transparency works in your favour.

One technical point as well: at graduate level, registering the work and running a similarity check before the defence is a mandatory procedure, and IranDoc's similarity system reports how much of your text matches theses, dissertations, articles and the web. It reports the degree of similarity and the source, but it does not say who wrote the text, and it does not assess anything about artificial intelligence.

Hard questions and how to handle them

Some questions you will not be able to answer. That is not a failure and not a disaster; it is part of the nature of research. There is an important difference between two replies: "I don't know" and "I don't know, but here is how I would find out". The first closes the door; the second shows that you did not merely memorise a result but hold the method for thinking about the problem.

If a question falls outside the scope of your work, say plainly that this was not within the bounds of this study and how it could be investigated if needed. If a question concerns a statistical or methodological detail you cannot recall, say you will check the exact figure in the text and follow up, rather than guess. Guessing is the worst option in a defence, because it is quickly exposed and it casts doubt on your other answers.

Stay calm with a sharp critical question too. The examiner's aim is not to humiliate you; it is to see how steadily you stand up to criticism. Instead of becoming defensive, restate the question in your own words once, concede a point if it is fair, and explain why you made the choice you made. Accepting a limitation does not signal weakness; denying it does.

After the defence

The defence does not end when the session ends. The committee usually announces a list of corrections, and the final version has to be prepared on the basis of them. Make those corrections yourself. Handing them to someone else — even to a tool — is exactly where your share of the work fades, and if a question later arises about a corrected passage, you will not be able to defend it.

If you do not understand a correction, ask. Asking what a correction means is a sign of care, not of weakness. Read the final version carefully afterwards as well, to make sure the corrections have been applied everywhere they touch, not only on the page the examiner pointed at. Consistency of the text after revision is something the next reader of your work notices immediately.

A checklist for the night before and the day of the defence

Keep the practical tasks of the last two days simple. Most defence anxiety comes from unfinished tasks, not from the questions themselves. Go through this list once, from top to bottom.

• Rehearse the whole presentation once, with a timer.
• Have the final text, the slides and any earlier corrections with you.
• Say your three key sentences out loud: contribution, limitation, conclusion.
• Prepare your answer about the role of AI in the work.
• Write the main limitations as a short list.
• Re-read the questions raised in any earlier pre-defence session.
• Bring the similarity report and the appendices of the work.
• Arrive a little early and test the presentation equipment before you start.
• Take water and a paper copy of your own notes.
• Talk to the audience instead of reading the text aloud.

If one item on the list does not get done, do not panic; what matters is that you command the work, not that the list is complete. A defence is a professional conversation, and you are the owner of this work.

Conclusion

Do not treat the defence as a memory test. Rehearse the three recurring questions, keep the slides simple, and have your honest answer about the role of tools ready. If you did the work yourself, preparing for the defence is mostly review rather than construction.

Before the session, read your own text once more through an examiner's eyes and notice where questions might arise. If you want to know what each degree level expects from a final project, see how the thesis differs across degree levels, and to keep the last days organised, do not miss the checklist of thesis writing rules. For the wider logic of academic writing, writing a scientific paper and the IMRAD structure is a useful companion.