A viva is not an examination of the candidate’s memory, nor a presentation of the thesis. It is an oral defence: the examiners have read the work, they have formed doubts about it, and the candidate is asked to answer them. The questions below are the ones that recur.
What defending means in computer science
In computing the thesis is rarely a single experiment, and the defence rarely turns on one result. A candidate is usually defending a construct, an algorithm, a model, a framework, a system or a formalism, together with the evidence that it works and the claim that it was needed. The examiners probe four things in turn.
That the problem was real. Why this problem, why it was open, and what in the existing literature failed to solve it. A solution to a problem nobody had is the most common weakness in a computing thesis, and it is usually exposed by the question of who else has tried.
That the design decisions were reasoned. Every choice of representation, architecture, parameter and baseline is a decision that could have been made otherwise. The candidate is expected to say why each was made, what the alternatives were, and what was lost by not taking them. “It is what everyone uses” is not a defence.
That the evaluation supports the claim, and no more. This is where most vivas in computing concentrate. The examiners ask what the baselines were, whether the comparison was fair, whether the data were adequate, whether the improvement is significant rather than merely numerically larger, whether it would survive a different dataset, a larger scale or a real deployment. A candidate who overclaims on a narrow evaluation is in more difficulty than one who claims modestly and shows the limits.
That the contribution is stated and bounded. What is new, to whom it matters, and where it does not apply. Examiners respond well to a candidate who names the limitations before being asked, and badly to one who defends the work as though it had none. Conceding a genuine weakness is not losing the argument; it is evidence of the judgement a doctorate is meant to certify.
Two features of the discipline shape the defence. Software and data are part of the contribution, so questions about reproducibility, whether the code runs, whether the results can be obtained again, are substantive rather than procedural. And much of the field publishes in conferences rather than journals, so a candidate is often defending a thesis assembled from papers, and will be asked what the thesis adds beyond the sum of them.
The underlying question behind all of these is the one examiners report asking themselves: does this person think like an independent researcher? It is answered less by having the right answer than by how a difficulty is handled when one arises. Further reading on what counts as a contribution is on the Contribution to the body of knowledge page.
100 possible questions for your PhD viva examination
- How have you evaluated your work? (intrinsic evaluation: how have you demonstrated that it works, and how well it performs)
- How do you know that your algorithm/rules/techniques are correct?
- What are your key findings?
- Which of these findings are the most interesting to you?
- Why are your findings interesting?
- Have you achieved your research aims?
- How have you demonstrated its usefulness for a specific application context?
- How do your findings fit with or contradict the rest of the literature in this field?
- How do you know that your findings are correct?
- What is the area in which you wish to be examined? (particularly difficult and important if your thesis fits into several areas, or has several aspects, or seems to fit into an area of its own as mine does).
- What do your results mean?
- Which topics overlap with your area?
- Is your field going in the right direction? (For example, if everyone`s been concentrating on speed, but the real issue is space - if the issue is time, you can just wait it out, unless it's combinatorially explosive -, but if the issue is space, the system could fall over). This is kind of justifying why you have gone into the field you're working in.)
- What are the main implications of your research for the rest of the field?
- Where do current technologies fail such that you (could) make a contribution?
- Who are your competitors?
- What do you do better than your competitors?
- What do you do worse than your competitors?
- Which are the three most important papers that relate to your thesis?
- Has your view of your research topic changed during the course of the research?
- What are the main issues and debates in this subject area?
- Which of these issues/debates does your research address?
- Who has had the strongest influence in the development of your subject area in theory and practice?
- What published work is closest to yours?
- How is your work different from other works in the literature?
- What are the most recent major developments in your area?
- How do your findings relate to literature in your field?
- Who will be most interested in your work?
- What is the relevance of your work to other researchers?
- What is the relevance of your work to practitioners?
- What is the relevance of your contributions to other researchers? ?
- What is the relevance of your contributions to industry?
- How do you expect X to progress over the next five years?
- How long-term is your contribution, given the anticipated future developments in X?
- Which aspects of your work do you intend to publish – and where?
- Where will you publish your work? (Think about which journals and conferences your research would best suit. Just as popular musicians promote their latest albums by releasing singles and going on tour, you should promote your thesis by publishing papers in journals and presenting them at conferences. This takes your work to a much wider audience; this is how academics establish themselves.)
- What are the weakest parts of your work?
- What are the strongest parts of your work?
- What would have improved your work?
- Where did you go wrong?
- How would your system cope with bigger examples?
- Does your system/technique scale up? (This is especially important if you have only run your system on `toy' examples, and they think it has `learned its test-data’.)
- Where did your research-project come from?
- How did your research-questions emerge? (You can`t just say “my supervisor told me to do it” - if this is the case, you need to talk it over with your supervisor before the viva. Think out a succinct answer.)
- How has your view of your research topic changed?
- What are the motivations for your research?
- Why is the problem you have tackled worth tackling?
- What motivated and inspired you to carry out this research?
- Can you tell us how you came to choose this topic for your doctorate?
- Why have you defined the topic in the way you did? What were some of the difficulties you encountered and did they influence how the topic was framed?
- What`s original about your work?
- What is your original contribution?
- What are the most original parts of the thesis?
- Where is the novelty? (Don`t leave it to the examiners to make up their own minds - they may get it wrong!)
- To what extent do your contributions generalise?
- To what extent would they generalise to systems other than the one you’ve worked on?
- What are the contributions to knowledge of your thesis?
- What have you done that merits a PhD?
- What is the relevance of your contributions?
- Which propositions would you say are distinctively your own?
- How do you think your work takes forward or develops the literature in this field?
- What are the “bottom line” conclusions of your research?
- How innovative or distinctive are your conclusions?
- What are the main achievements of your research?
- What is the implication of your work in your area?
- What does it change in your discipline?
- Have you solved the field`s problem that you claim to have solved? (For example, if something is too slow, and you can make it go faster - how much increase in speed is needed for the applications you claim to support?)
- Under what circumstances would your approach be useable? (Again, does it scale up?)
- What are you the most proud and why?
- What ethical implications to this work?
- How did you deal with the ethical implications of your work?
- What advice would you give to a research student entering this area?
- What have you learned from the process of doing your PhD? (Remember that the aim of the PhD process is to train you to be a fully professional researcher - passing your PhD means that you know the state of the art in your area and the directions in which it could be extended, and that you have proved you are capable of making such extensions.)
- Can you summarise it in one sentence?
- In one sentence, explain your PhD to your mum?
- What is the idea that binds your thesis together?
- You propose future research. How would you start this?
- what would be the difficulties in future work?
- You discuss future work in your conclusion chapter. How long would it take to implement X, and what are the likely problems you envisage? (Do not underestimate the time and the difficulties - you might be talking about your own resubmission-order!)
- Outline where you think future development of your ideas could lead and how this might be done?
- Who are your envisioned users?
- What use would your work be in situation X?
- Under what circumstances would your approach be useable?
- What are the core methods used in this thesis?
- Why did you choose this approach?
- In an ideal world, are there different techniques you`d have liked to use?
- What did you gain from your approach? (You need to justify your approach - don`t assume the examiners share your views)
- What were the alternatives to this methodology?
- What would you have gained by using another approach?
- What were the crucial research decisions you made?
- Looking back, what might you have done differently? (This requires a thoughtful answer, whilst defending what you did at the time)
- How has your view of your research topic changed?
- What advice would you give to a research student entering this area?
- How do your methods relate to your conceptual framework?
- Why did you choose to use those methods of data collection?
- Were you disappointed with your conclusions?
- How else might you have undertaken your research?
- How did you decide upon your research boundaries?
- What was the universe from which your sample was selected and how did you define it?
- What other methods did you consider and why were they rejected?