Rubbishing the Viva

There’s a strange article today in the Times Higher that claims that the UK’s system of examining PhD students is “a scandal” and that it is “way behind the rest of the world”. These comments are from a chap called Ron Barnett (an emeritus professor at the Institute for Education, who explains

“Students can spend five years doing their PhD, present their thesis and come up against the maverick view of an independent examiner and in effect be rubbished,” he commented.

“I’ve seen it happen far too many times,” he told a Westminster Higher Education Forum seminar on the future of postgraduate education, held in London on 17 October.

I have to say I find it hard to reconcile such remarks with the business of examining PhDs as I’ve observed it, in Physics and Astronomy. And I’ve done quite a few over the years; see, e.g., here. For a start, it’s extremely rare for a student to spend five years doing a PhD in my field – the Research Councils put extremely strong pressure on departments to ensure that students submit within four years, and most research students take less time than this to produce their thesis.

But it’s the idea that a maverick external examiner can sabotage a PhD that I find hardest to recognize. If that looks like happening the internal examiner should stand up for the candidate. In fact, here in Cardiff we have an additional safeguard against this sort of eventuality: each viva has a Chair as well as the two examiners. The Chair is just there to ensure fair play and that proper procedure is followed, but is rarely (if ever) called upon to intervene in practice.

I can’t speak for other fields, of course, and it may indeed be more of a problem in other disciplines. Curiously, Prof. Barnett says that he has seen it happen “far too many times”. I wonder how? As internal examiner? In which case he should have stepped in to stop it? If not as internal then in what capacity was he privy to the conduct of a PhD viva? I’m confused.

Anyway, in a couple of weeks I’ll be participating in a PhD examination in another country (Denmark). There the defence is public, and it involves two external “opponents”, but I don’t know whether it is easier or harder for the candidate than the British system so I won’t comment on whether it’s fairer or more rigorous than what we have in the UK. I’m very much looking forward to seeing how it works, actually.

In my opinion, if there is a “scandal” in the system of UK PhD examinations, at least in science disciplines, it’s not the one Prof. Barnett describes. It’s that we produce far too many low-quality PhDs based on dull, incremental research and that, if anything, externals are not tough enough.

There, I’ve said it. No doubt you’ll have a go at me through the comments box!

Advertisements

35 Responses to “Rubbishing the Viva”

  1. Although I agree with you Peter, I’ve seen a viva go badly wrong. A junior, inexperienced examiner coupled with a mentality that anything different to their research must be wrong, definitely scuppered an excellent student’s viva. They even forced the student to spend another 3 days with them at their European institute to ‘prove’ the student was good enough to be awarded the PhD AFTER the viva happened. The Chair and Internal, not knowing enough about coding and the subject area to answer back, let it happen.

  2. Anton Garrett Says:

    It might be more likely to happen in a non-subject like Prof Barnett’s.

  3. One difficulty with the system of public defences (eg within Scandinavia) is that its basically a show. Since the candidate’s family attends this leads to an implicit pressure on the examiners not to push too hard. Furthermore, there is usually also an explicit pressure from the institution that everyone sticks to (a very limited) time; its unusual for the opponent (or examination committee member) to have more than an hour to probe.

    Having examined in both the UK and Scandinavia I would claim that, in general, the UK examination system is more demanding of the candidate. Then again, the quality of the thesis is usually higher in Scandinavia. These observations were made for PhDs in experimental particle physics.

    I’ve heard horror stories about maverick examiners in the UK but never within my field. The worst complaint I’ve heard about physics vivas is that the external examiner sometimes comes unprepared and the internal fails to take up the slack and guide the viva forward.

    • Rob Ivison Says:

      I’ve done a couple of examinations in Denmark and I agree with these sentiments. It takes a harder man than me to ignore the presence of a candidate’s spouse and parents in the audience. In both cases the Danish work was of a higher standard than I usually encounter in the UK, so the subtle pressure wasn’t an issue; it’s there, nonetheless.

      I do appreciate several aspects of the Danish system: they pay an examiner’s fee appropriate to the work involved, and any visit to Copenhagen is a joy.

  4. Matthew Smith Says:

    Peter maybe you are slightly biased in that any viva you have been involved you have ensured has run well? For the vast majority of people the viva seems to be well run. However, on the second day of my PhD a candidate was given a seven hour Viva, while the external did not have any major criticisms of the work. I struggle to see how that is good situation for anyone involved. When mentioning the system in the UK to people in Italy the past week they certainly seem to think it sounds a lot tougher than the process they went through. I have yet to go through mine, but I wonder how coherent I will be after an hour…

    • telescoper Says:

      My viva was about 4 and a half hours, which I think is longer than average, but not excessively so. Seven hours is silly, as everyone would be exhausted long before the end. I suspect the external just liked the sound of his/her own voice. The chair in that case should have intervened – much like a judge in a trial would so – to get things moving instead of getting bogged down one particular point.

  5. Either the viva voce, public defense or whatever is a formality or it is not. If it is a formality, basically a show, then it should be abolished. If not, then a non-negligible fraction should actually fail, but I don’t think that happens. (Though this might be a case of people who are likely to fail pulling out to avoid embarrassment or whatever, in which case it might not be a good idea to abolish it.)

    I think a public talk of some sort is a good idea, if only because it will prevent degrees from being awarded when they are definitely not deserved, or at least make it more difficult for this to happen. A defense with an opponent is perhaps a bit outdated, but does involve getting an opponent who actually knows something about the topic.

    • Certainly the viva is not a formality in the UK as I am aware of cases where someone has failed or has come very close. However, even if it were something of a formality, it’s still not clear that it should be abolished. We have a first-year interview at which stage a student can be taken out of the PhD programme (although I don’t actually know if they are required to leave or advised to do so). We’ve never done so in my institute, but I believe it does happen in some other departments reasonably often. After this point it is assumed that the student is capable of completing a thesis that would pass a viva. Ideally then, the thesis is not submitted until it is in an acceptable state. One could then argue that the viva is unnecessary, but it still plays a role in ensuring that the thesis is indeed in an acceptable state and puts some pressure on the candidate to go and prepare for such an examination. Just because they almost always pass, does not mean it doesn’t have any value.

    • In the Scandinavian universities where there is a public defence, it is indeed extremely rare for a candidate to fail, but there is also a large responsibility put on the institutions not to let a candidate up to defend if they are not sure they will make it. On the rare occasions that a candidate does fail (it happens), it is often because the candidate decided to go up against the recommendations of their institution. I know, anecdotally, of cases where a candidate was passed who should have been failed, but I don’t think this is a particularly relevant measure of the quality of postgraduate education as a whole.

      I do think there is a useful purpose in having a final examination of one kind or another, but I’m less certain that the exact form of that examination matters so much. Either way, the harshness of viva/public defence is unlikely to be the most influential factor in determining how good researchers postgraduate programmes will produce.

  6. Most of the vivas I’m aware of are reasonable, but I am aware of two tat went off the ralis. One involved a field where the external’s mother did the fundamental work, and lasted about 8 hours (this was in biophysics) while the other, in History, was truly bizarre. The viva went okay, and the usual list of corrections was supplied, but when these were completed the external changed their mind and failed the candidate. This case is still working its way very slowly through appeal.

    So I’d say that a few rogue examiners are messing up an essentially workable system. The presence of a chair who is prepared to use their powers would help catch these outliers.

  7. “It’s that we produce far too many low-quality PhDs based on dull, incremental research and that, if anything, externals are not tough enough.”

    is this a comment on the PhDs you have examined or supervised? i’ve not yet seen a PhD thesis at Durham which could be described as incremental – all have had at least 1-2 chapters which were novel and interesting. in part this is a result of the assumption that much of the thesis will have been published in refereed journals, which if nothing else removes the most incremental stuff.

  8. Here in Australia we lament the fact that students have a postal examination, rather than a viva. I would prefer a viva any day.

  9. Bryn Jones Says:

    Perhaps the author of the Times Higher Education article works in a field where a PhD thesis actually does present a thesis – an argument, an opinion – rather than being a presentation of work that attempts to be objective, as in the sciences. This may be why he is so critical of the examination system in a way that I do not recognise.

    I’ve tried to think of cases in British universities where the PhD examination system produced obviously inappropriate decisions, and can’t really recall any. Perhaps this is because the conclusion of the viva voce examination is seldom an absolute pass or fail, but instead the norm is that a thesis requires some corrections before acceptance. There are cases I can think of where a demand by an examiner for a major rewrite (rather than minor corrections) seemed probably inappropriate, but no major injustices.

    I do agree with Peter’s point that British PhDs sometimes fall short of the quality that might be expected. They do not compare favourably with PhDs from some other countries because there is a limit to the research that can be carried out and the training received in only three years. This is a result of the policy by research councils and university departments to train large numbers of people to PhD standard and as cheaply and quickly as possibly.

  10. The British system is one of the most rigorous I have encountered, in terms of the depth to which the examiner can push the student and the fact that there is a possibility of failure. It really is an examination! I had a 5 hour viva myself (with in the end very few corrections, but my examiner pushed me hard on every assumption, and I really felt I deserved that PhD at the end!) and I have now run one of a similar length as an external examiner. We took a break for lunch so that the candidate did not pass out, and the whole process felt friendly and constructive (helped by a good internal). I was also left with a hugely positive impression of the candidate, who was open to all of our questions and discussion points. As a result I would hire him in a second if I had the chance, and now write very enthusiastic reference letters for him – it’s great to get the chance to see how a young scientist justifies and defends their work when their supervisor is out of the room! The Dutch system, by contrast, is largely ceremonial and although it’s fun to troop out in the silly hats and gowns, I honestly feel that it’s not as rigorous. The existence of a €100k bonus from the Dutch government to any department that graduates a PhD student also provides a rather blatant incentive that might encourage a less than rigorous department to push through students who are not up to standard ( know examples where this has happened, although not in astronomy).

    • I’m sure that there are at least indirect monetary incentives in the UK to produce more doctors.

      In any case, Dutch astronomy has one of the highest reputations in the world. They have about the same number of institutes per capita as otherwise comparable countries, but all are top-notch. (The closure of Utrecht of course casts some shadow on this.)

      While the actual graduation stuff in the Netherlands might be ceremonial, the quality of the degree doesn’t depend on it.

  11. Dutch astronomy does have a great reputation, true. However, the financial incentive is there and there is great trust placed in the ethical standards of the department to ensure that theses reach a required standard. Consider the following. At my university, the PhD committee must consist of a majority of members from our university, and the thesis can be passed based on majority vote (this occurs before the ceremonial defence, the committee are asked to give a simple yes/no as to whether the thesis reaches the required standard). Since it is the university that benefits financially from the student passing, there is clearly a conflict of interest. Whilst I have confidence in the ethical standards of my colleagues I would much prefer it if this potential for conflict of interest did not exist.

    • Is there no conflict of interest in the UK?

      • No idea! It’s been a long time since I left. I’d actually be interested to find out what the failure rates are for PhDs across the world, for comparison. If standards are rigorous, then there must be some people who, for whatever reason, don’t make it. I know several examples in the UK where people left with an MPhil, and I’ve yet to encounter any in the Netherlands. Alternatively we Dutch university faculty members just rock!

      • The idea of the MPhil as a consolation prize is, I think, specific to the UK (and might make the introduction of any low-level degree with a similar name more difficult). I don’t think comparing failure rates would provide much information. There are too many difference with respect to motivation for people to drop out without failing, consolation prizes not being counted as failures, consequences of failure, possibilities to try again, difficulty of entry in the first place—not to mention different standards for the degree itself.

      • the incentive (i believe) is to have the student submit their thesis within 4 years, otherwise there is a penalty in the number of studentships provided by STFC in future.

        however, if they fail their viva – then i don’t think that matters…

  12. I think Ian’s correct – we have penalties not rewards.

    In the UK a PhD thesis submitted 3 years and 364 days after the start of the project that fails is deemed a success, while one submitted 3 years and 365 days after the start that wins a Nobel prize is deemed a failure, and the host department is penalised.

    You can probably guess that I don’t think this approach is entirely correct 🙂

  13. If it turns out that ‘I’ve seen it happen too many times’ means that Prof Barnett was himself the supervisor, I suspect his piece tells us more about his peers’ view of his research program than it does about the viva system!

  14. […] much different from the system we have in the UK? Probably, as it is a much more public occasion. Will that necessarily make it better? I don’t know. I was reading the thesis last night, actually, and it’s very interesting […]

  15. My supervisors set up the viva, that lasted 6 hours. The internal had not written the report by himself and the external lied about the methods. And all to hide a misconduct in research. In UK top university. This wouldnt have happened if the viva was open to publics.

    • telescoper Says:

      How could the external have “lied about the methods”? That doesn’t make sense if you wrote the thesis. And how could a viva hide misconduct? By whom?

      • the external examiner made a report evaluating the methods used in the thesis and his judgement did not objectively valued the methodology from a scientific point of view. A misconduct by the supervisory team. The content of the thesis had been sold to a pharma company without knowledge of the student so the viva had to be arranged to be failed so data has officially no value at all. But in fact it has a lot of value.

      • telescoper Says:

        Presumably you appealed the decision?

      • I appealed and the review committee dismissed the case before the hearing. The “academic judgement” of a reviewer will always prevail (even if wrong); UK law.

      • telescoper Says:

        Well, since you’ve so far used three different phoney email addresses I for one am not inclined to trust the veracity of your statements.

      • I see… well, I understand your point; but you may not have been chased by GSK, Astrazeneka and Pfizer and top2 UNV of the world. Have you watched “the constant gardener”? reality is always above fiction.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: