«The Unwritten Rules of PhD Research Gordon Rugg Marian Petre THE UNWRITTEN RULES OF PHD RESEARCH THE UNWRITTEN RULES OF PHD RESEARCH Gordon Rugg and ...»
A third opening gambit which can be misconstrued is asking you to give a brief overview of your thesis. One perfectly understandable response to this is anger: surely if the examiners have read your thesis properly they shouldn’t need a brief overview of it? Again, not so. The brief overview is a pretty good way of getting a nervous candidate to open up: they usually become so absorbed in the topic that they forget their nerves. It’s also a good way of ﬁnding out which aspects of the thesis they ﬁnd most important, which tells you a lot about their professionalism. In some cases, it’s also useful for ﬁnding out just what the thesis is supposed to be about, if it’s been written up in a particularly vague or unreadable style. Quite often, a thesis will contain stuff which looks promising but which is poorly structured and badly described. If the content is good, then a few skilful changes can make a surprising difference to the thesis; if the content is as bad as the style, then the whole thing needs to be consigned to a nameless pit and erased from human memory. Either way, this opening gambit helps the examiners towards a decision.
It’s tempting at this stage to break the third golden rule, by panicking and blurting out the truth about things you did wrong. Don’t give in to temptation. Instead, give a clear, previously prepared overview, listing the main ﬁndings and the main ways in which this work is a contribution to knowledge. You might want to use a few gambits of your own to show that you know the rules of the game – for instance, a discreet reference to your latest paper in a major journal, or a mention of ongoing work with a major ﬁgure in the ﬁeld. Don’t overdo it, though – remember that you’re a candidate, not an examiner.
168 THE UNWRITTEN RULES OF PhD RESEARCH The mid-game After the opening exchanges the real business begins, and vivas begin to differ. Examiners have four main options available to them in terms of outcomes: straight pass, pass subject to minor changes, pass subject to major changes and fail. They have many more options available to them in terms of how they conduct the viva itself, and it’s important to remember that a viva can feel like a grim interrogation to you, but end up with a straight pass.
The straight pass At one end of the cosiness scale is the situation where the examiners make it clear from the outset that there’s little doubt this will be a straight pass, and then ask a few questions for courtesy’s sake before heading off with you and your supervisor to lunch/coffee/the pub. You may be asked to ﬁx three or four typos, for form’s sake. This outcome typically occurs when you have a good supervisor, when you and your supervisor both know how the game is played, and when you’ve done a thoroughly good job on your dissertation.
Minor changes Next down the scale is the viva where you are asked a few technical questions to check on speciﬁc points. You probably won’t be able to answer all the
questions either to your satisfaction or to theirs. Don’t let that shake you:
nobody can answer all the possible sensible questions, or have done all the relevant reading. An example of this is as follows. You have done a PhD which involves a new taxonomy of human error types. You have done the essential reading (Reason, Hollnagel, Rasmussen and so on). You have also done some good further reading by looking at the literature on formal taxonomic classiﬁcation in biology. The external examiner asks you whether you have read the literature on use of multidimensional statistics for classiﬁcation. You haven’t.
What do you do? Well, for a start, you don’t panic. Although the question is a valid one, you have already gone beyond standard best practice in your area by bothering to read up on formal taxonomy. The leading literature on human error isn’t based on multidimensional stats and cluster algorithms. Although you could in principle have investigated that route, it’s only one of many routes which you could have taken, and it’s physically impossible to take all of them. Saying that you could have taken it is very different from saying that you should have taken it. So, you can bounce the question back at the external by making some of the points above, and politely asking what they think that the multidimensional approach would offer to this area. You can then get into a debate with them which allows you to demonstrate your ability, which is the main point. To conclude this example, if you make a good case in the debate, THE VIVA 169 you may be asked to add a sentence saying that you chose not to use multidimensional stats, and giving the reason for not using them: a pretty minor change.
Major changes A more serious situation would be where the examiner’s question identiﬁes a serious area of ignorance on your part. A good example of this comes from statistics. Suppose you have looked at different groups’ perceptions of how car crashes are shown in the popular media, and have gathered some numeric data on the respondents’ perceptions. You write this up in neat tables and mention in your discussion that the difference between two groups is signiﬁcant, and that another ﬁnding is highly signiﬁcant. The examiners reach this point in your thesis and the external examiner says that you have described results as signiﬁcant and as highly signiﬁcant, but haven’t mentioned the tests you used or the p values involved.
None of the options at this point are good. The least bad is that you tell them which tests you used and what the p values were; the examiners will wonder silently what sort of idiot you were to omit this information and will be on the lookout for further signs of idiocy, but you may well get away with simply adding the missing information. The two other main options are about as bad as each other. One is that you tell them that you use only qualitative methods and don’t agree with quantitative methods on principle; you then look like an idiot for choosing a non-qualitative external examiner, and for having gathered quantitative data but not analysed it quantitatively. This issue relates to some serious debates about which skills in a discipline are essential rather than optional, which go beyond the scope of this book: to use an extreme (and genuine) example, should someone be allowed to graduate in French if they refuse on ideological grounds to use standard French spelling and punctuation? If you don’t approve of cabinets, that’s fair enough, but to undertake an apprenticeship in cabinet-making when you feel that way is a decision that falls somewhat short of being sensible.
The last main option involves your admitting that you have never heard of statistical tests or p values, and asking what p values have to do with signiﬁcance anyway. This approach might be perfectly acceptable in some disciplines, but in others where statistics form a core skill (e.g. experimental psychology) it would be disastrous. If you’ve ended up in a situation where your external is asking this sort of question and where your answer is an admission of total ignorance, then you’ve made a serious mistake in your choice of external, in your choice of discipline or in your approach to your subject. The best that you can hope for with either of the two latter options is some serious revisions. Even if you wriggle out of this particular question, the session will have shifted from a fairly routine check to a serious investigation of whether or not you deserve to pass, 170 THE UNWRITTEN RULES OF PhD RESEARCH and there’s a strong likelihood that one of the next questions will sink you fair and square.
The fail At the bottom of the cosiness scale is the viva where the examiners think that you’ve made a serious mess of things, and where you exacerbate matters by being gratuitously offensive, ignorant and/or stupid. Usually, but not invariably, these cases occur when the student submits their thesis against the advice of the supervisor. A typical example might be a part-time PhD student who has a fairly inﬂuential day job as a manager, and who cobbles together a dissertation topic on job satisfaction based on reading textbooks and professional magazines, then goes on to conduct a badly designed questionnaire and/or some badly designed interviews, and who talks about ‘getting all this airy-fairy academic stuff out of the way’. (Yes, this sort of thing has really happened, and more than once.) There isn’t much that can be salvaged from such cases. The literature review is too simplistic to lead to an interesting research question; the methods are too boring to form the basis of a decent rewrite; the data will probably be untrustworthy or uninformative because of the ﬂaws in the methods. Work of this sort will fail, and deservedly so. Again, if you think that cabinet-making is overrated and that MFI ﬂatpack furniture is just as good, then you’re entitled to that opinion but you would be pretty silly to apply for an examination as a would-be master cabinet-maker and bring along a poorly assembled ﬂatpack as your alleged master piece.
The end game By the end of the session, the examiners will probably have reached a conclusion about what to do with you. ‘Probably’ because they will need to check with each other and reach an agreed verdict. They will politely ask you to leave, and will do whatever examiners do while you are pacing around in the corridor, feeling nervous.
What examiners do is to check with each other, reach an agreed verdict and have some breathing space. If your thesis is a clear pass or a clear fail, then they will check explicitly that they all agree. If it’s a clear pass, they won’t necessarily summon you back 30 seconds after you’ve gone out; quite a lot of examiners believe in keeping the candidate waiting for a few minutes, on the grounds that it is good for the candidate’s soul.
If there’s consensus that you’ll have to make changes, then nice examiners will draw up a clear list. This might be a longhand list written there and then, or it might be emailed to you a day or two later.
If there’s not consensus, then the examiners have to slug it out among themselves. At this point, you can reap what you have sown much earlier. For example, if you have produced a couple of decent journal papers out of your work, then that demonstrates that your work is of adequate professional quality, THE VIVA 171 which strengthens the case for anyone wishing to argue that your thesis contains good stuff, even if it’s been badly written up. Similarly, if you have done a solid literature review, then that shows that you have a proper professional knowledge of your area, even if your data collection was a bit tatty. Both these examples involve starting early; they’re not something you can cobble together in the last week. It’s a good idea, a year or so into your PhD, to read your institution’s regulations and then get someone knowledgeable to translate them into English for you, so that you can ﬁnd out about things like indications of acceptable quality in the thesis. If there’s something about ‘publishable in a peer-reviewed journal’, for instance, then getting a couple of publications in a peer-reviewed journal will help your case in the event of a debatable verdict after the viva.
When all of this has been done, the examiners will reach an agreement with your supervisor about the next stages, and summon you back. Some will say the magic words: ‘Congratulations, Dr Smith’; others won’t. Strictly speaking, you aren’t Dr Smith until you’ve formally graduated, so don’t read anything too much into it if they don’t use those words; externals are the sort of people who will probably know about the distinction between ‘doctor’ and ‘doctorandus’ and who may phrase their greeting accordingly. The rest of the session will quite probably be a bit hazy. If you and/or your friends have been efﬁcient, there may be champagne ready. The examiners may or may not participate in the festivities; the day will end and in the fullness of time the ﬁrst day of the rest of your life will dawn.
So, that’s what happens from the examiners’ point of view. What happens from your point of view? This section is briefer, partly because much of it is covered above and partly because it’s also covered in depth in all the other books on this topic.
Before the viva
Early in your PhD, discuss with your supervisor whether or not to go for a journal publication or two. Supervisors, and disciplines, vary in this regard.
When you write-up, allow plenty of time to do a decent job, and pay particular attention to displaying your cabinet-making skills in the thesis. Choose an appropriate external in a sensible way: by this stage in your career, you should have a reasonable idea of the relevant rules of the game and of the main relevant players.
The week before In the days before the viva, reread your thesis and your data, plus some of the key literature. You’ll probably be utterly sick of all of them long before this 172 THE UNWRITTEN RULES OF PhD RESEARCH stage, so reward yourself with chocolate, or whatever currency works for you. Organize a mock viva, including a presentation at the start, with at least one mock examiner who knows the craft skills of doing a good viva and presentation. If they’re good, they won’t be gentle with you in the mock viva, and they will give you constructive feedback about what to change for the real event. A lot of this is likely to involve blood in the water – you’ll probably panic gently in the mock session and blurt out needless admissions of weakness such as: ‘I know now that I should have used a larger sample size’ rather than something like ‘My next study will extend this and will use a larger sample’.
The day before The day before the viva, check that you are sure where and when the session will take place (there may have been a last-minute change which you missed).
Line up a trusted friend to be available for you during the viva day. They will probably organize a tactful bottle of champagne and glasses in a way which doesn’t look too much as if it is tempting fate, and whisk you away if it all gets too much – even a pass with minor changes can leave some candidates feeling like a thoroughly wrung dishcloth. Do not go out for some serious drinking the night before: you’ll need to be fresh on the day. Check that you know the examiners’ names, titles and main publications. This is not to help you grovel; it is to show professional courtesy and to improve your chances of anticipating the direction of their questions. Make sure that you have appropriate clean clothing for the next day and a hard copy of your thesis. Ask your supervisor politely if they will make sure they have a copy of any required changes, in case you forget in the excitement of the moment. (They will probably be planning to do this anyway, but it doesn’t hurt to make doubly sure.) On the day On the day, turn up in plenty of time, and have something to do while you wait. Don’t be offended if you aren’t introduced to the examiners, invited to join them for lunch or if you are kept waiting before the session begins. All of these are perfectly routine features of the viva process and reﬂect neither discourtesy nor inefﬁciency. You might, for instance, be kept waiting because some idiot has been illicitly using the viva room for an unofﬁcial seminar, and has had to be evicted, leaving the examiners with the thrilling job of rearranging all the furniture.
At the start of the viva, be polite and do your exercises for staying calm if
you need to. If you aren’t asked to do a presentation or an overview, don’t fret:
the mock session won’t have been wasted, because it will have helped you to pull together your thoughts about the thesis in a way which will come in very useful during the viva.
THE VIVA 173 The ﬁrst question or two will probably be fairly light, and used for breaking the ice. With these and the subsequent questions, you need to remember the three golden rules (don’t lie, don’t try to be funny and don’t panic and blurt out the truth) plus a couple of other things.