«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 ...»
• Trouble with writing may indicate problem areas in your research rather than problems in your attitude – if you encounter trouble with writing, then look closely at what you are trying to do and the materials you are using.
• Organize the ideas/concepts/material before you start to write.
• Be precise.
106 THE UNWRITTEN RULES OF PhD RESEARCH What’s ‘enough’ writing?
• A reasonably complete, analytic literature review. (Its authority derives from selecting the literature appropriately, putting the literature into a sensible conceptual structure that points toward your thesis and balancing recognition of gaps and limitations with synthesis of ideas and identiﬁcation of patterns in the literature.)
• Progress on objectives, manifested as new insight supported by evidence.
• Answering the ‘so what?’ question. Articulation of the signiﬁcance of the research ﬁndings, and of the contributions to knowledge. Understanding both their generalizability and their limitations.
Q: Do I need to include everything I’ve done?
A: No. If you decide that a subset of data, or an entire study, does not contribute to the overall composition of the thesis, you don’t need to include it. Note, however, that this does not apply to the situation where one study shows that your initial hunch was clearly wrong: in this case you have to include the study. If you do research sensibly, then you will be phrasing your questions as a series of reductions of the problem space, rather than a search for conﬁrmatory evidence, so this issue should not be a problem for you.
Q: Do I need to include the raw data?
A: That depends on the conventions of your discipline; check with your supervisor and your institutional regulations about PhDs. The usual principle is that the appendices should include examples of what you used at each stage, starting with instructions to respondents, continuing with examples of any data collection instruments that you used, and also showing one or two examples of completed response sheets or whatever it was that you used.
This allows the examiners to check what you did at each step, and to satisfy themselves that you did it right.
Q: How long does the thesis have to be?
A: The real answer is that it is as long as it needs to be to do its job, and no longer. The full signiﬁcance of this answer won’t make much sense to you until you’ve supervised undergraduate projects and taught research methods for a while, so the more immediately helpful answer is to check the regulations and ask your supervisor. The lesser answer is: within the maximum page or word limit set by your institution.
WRITING STRUCTURE 107 Q: I’ve just discovered a mistake in my analysis of the data, two days before I’m due to hand in. What do I do?
A: Good question. Whatever you do, don’t lie. Get in touch with your supervisor immediately and ask for advice about how to handle the corrections. If it’s a major mistake, you’ll need to redo the analysis, for all sorts of practical and ethical reasons. If the mistake is comparatively trivial and you are about to run out of time on your thesis, your supervisor may be able to suggest ways of buying time within The System so that the deadline is not an issue. Another possibility might be a rewrite which simply cuts out that part of the analysis from your write-up completely, if the mistake only applies to a manageably small subset of the write-up.
Q: I’m writing up, and I’ve just discovered that someone else has published something almost identical. What do I do?
A: Don’t panic. It’s usually possible to present the same material from at least two different viewpoints, if you know what you’re doing. Talk to your supervisor about this. You should be able to rephrase your work to take the other person’s work into account, and to differentiate yourself clearly from it. For instance, they might have studied a different social group from the one you studied; if so, you can put more emphasis on the social group aspect of your work, and less on the methodological novelty. You might well be able to get some beneﬁt from comparing and contrasting your results and the other person’s.
Q: I’ve developed writer’s block. What can I do about it?
A: The standard-issue books have plenty of ideas, and there are some in the chapters on writing in this book. Examples include deliberately doing something completely unrelated to writing up; writing something deliberately inaccurate, so that your subconscious rebels and makes you start writing the truth; rewarding yourself with treats; setting yourself small, manageable goals;
and getting a friend to help motivate you.
Q: Can I write-up in the same style that you use in this book?
A: You must be joking. This is the style we use over a cup of coffee; the style we use in our academic articles is very, very different (and a lot less fun, both to read and to write...) 10 Writing style
Writing is like dressing. You use different styles for different purposes. If you are going swimming, you wear swimming gear; if you’re about to do some welding, you wear welding gear. Style, in this sense, is about function, as opposed to ‘style’ in the sense of transient fashion. Unfortunately, many PhD students are somewhere between adequate and dreadful when it comes to academic writing; even those who are good still have to add many writing skills to their repertoire if they’re to perform at a good PhD level. That’s what this chapter is about. It should help you to understand how writing functions as a professional tool, and also to know about how to use appropriate styles of writing for the different things you will be doing as a PhD student. This overlaps with presentation skills, so we have included some coverage of that topic as well.
One important point to understand early is that this is not about using traditional grammar, or about slick presentation – it’s about language as a tool.
Another important point for students who are not native speakers of English is that you don’t need to speak perfect English to be able to write well. Good academic writing is about structure and form, not just about grammar.
The ﬁrst part of this chapter talks about the cabinet-making skills of writing:
how to avoid looking like an idiot, and how to look like a skilled professional.
This includes some detailed worked examples from different disciplines.
Remember that the signs of good cabinet-making can differ across disciplines, so you’ll need to ﬁnd out the signs for your area, rather than assuming that they’re the same as the ones we describe here. The next part then tackles the issue from a different angle.
WRITING STYLE 109 Finally, there are various lists and tips, including a list of FAQs about academic writing in general and writing up the thesis in particular (see also pp. 101 ff ).
Blood in the water Swimming in shark-infested waters is a bad idea if there is blood in the water. It is an especially bad idea if the blood is yours. Much the same principle applies to writing. Critical readers can detect blood in the water a long way off, and will come cruising in at speed looking for a kill. Sometimes they go for a quick kill, but on other occasions they decide to play with their victim for a while ﬁrst. It’s not a pretty sight.
Another analogy for the same problem, which is less eye-catching but more directly applicable, is the wolf-pack. Wolf-packs will run near potential prey, sizing it up for signs of weakness. If an animal looks healthy and capable of looking after itself, they ignore it instantly. If an animal looks ill or weak, they go for the kill. If your writing (or presentation, if you’re giving a seminar or conference presentation) looks healthy and professional, then you will probably be left alone. If there are signs of weakness, then the wolves start closing in fast. Begging a metaphorical wolf-pack for mercy has about the same chance of success as begging a real wolf-pack for mercy if you’re a plump caribou in the middle of a long, cold, hungry winter.
So, what do you do about it?
Step 1: stay indoors until you’re ready The ﬁrst and most simple step is not to go into predator territory if you have open wounds. If your work isn’t good enough, then don’t present it; go back and get it right, instead of presenting inadequate work and making apologies for it. Here is a good example (from a seminar by a colleague): ‘We found the following results... however, feedback from the subjects afterwards indicated that our initial instructions had been ambiguous. When we replicated the initial experiment with revised instructions and a new set of subjects, the results were as follows...’ The subtext here is: ‘I take it utterly for granted that you redo an experiment without hesitation if you have to, regardless of the time and trouble – trying to plead for mercy and present inadequate data isn’t even on the agenda for me.’ The key aspect of the presentation here is that the second sentence follows immediately from the ﬁrst one, without hesitation, and plunges straight into the results without making a big thing out of the fact that the researchers took the trouble to redo the experiment. Somebody who treats this level of professionalism as taken for granted will probably have been meticulous about 110 THE UNWRITTEN RULES OF PhD RESEARCH everything else in their study too, so the predators know there’s not much point in trying their luck.
One thing which novices usually get wrong involves sample sizes and amount of data. Here is another good example, from an MSc thesis: ‘There were three groups of respondents, each with four subjects...’ (followed by description of subject groups). That works out as 12 subjects. Twelve. Not a couple of thousand, or even a couple of hundred. The average novice would at this point start making apologies for the small sample size, with blood gushing into the water at a rate which would have every shark in the neighbourhood abandoning its previous plans and heading inwards for an unscheduled easy meal. This thesis simply described the subject groups and moved on to the next section.
The subtext in this example was: ‘This thesis is about a test of concept. For what I’m doing, I don’t need a large sample size, and I know it. Now, let’s move on to the next thing, which also shows that I know what I’m doing.’ Sample size and amount of data are important; however, the ﬁgures required will vary dramatically depending on whether you are doing data-driven work, test of concept or method-mongering. Not many novices know this. Even fewer know that excessively large sample sizes are often an indication of poor experimental design and inadequate knowledge of inferential statistics, both rich sources of blood in the water. If the previous paragraphs have left you with an uneasy feeling, then you might want to consider reading about types of paper, experimental design and inferential statistics.
Step 2: send out the right signals once you are ready Show you’re a professional: use the language and conventions of professionals in the discipline. If you don’t know what these are, then you need to learn them. Reading this book, and other texts on this topic, is a good start. Make your language and non-verbal signals as different as possible from those of the clueless beginner.
What are conventions for?
Every discipline has its standards for presentation. Some are highly formalized, others are tacit. You discover the standards by looking analytically at the
literature. Conventions are there to:
• embody standards in the discipline;
• make work accessible;
• make methods accessible, comparable and replicable/repeatable;
• make work comparable and help synthesis – putting work in a form that relates to other work.
Show you have something worth saying This is something to be done courteously and in a non-aggressive way. There’s no sense in antagonizing your audience, but there is sense in sending out a signal that you are a professional and that you won’t be wasting their time with second-rate garbage. A handy tip is to think in terms of asking your audience to ‘look at this’ rather than ‘look at me’. If you start by identifying an interesting problem and an interesting possible solution, rather than by telling the audience about something which you ﬁnd interesting and a solution which you think might work, then this is likely to be a good start. (It also has the advantage that if the proposed solution doesn’t work, then that doesn’t reﬂect on your judgement...) A strategy which is often effective is to bring in a potential solution from another discipline. The other discipline needs to be credible in the domain where you want to apply it – this normally means that it has to be more formalized and more mathematical than the domain of application.
It’s advisable to have a good knowledge of the other discipline, since otherwise your proposed solution is likely to be shot down by an expert from that discipline as being just another example of a keen beginner who hasn’t read the literature in depth and who makes classic novice mistakes, only one from a different discipline this time. (You might just get away with superﬁcial knowledge of the other domain in a conference, where there’s a slight chance that none of the audience know that domain. You won’t get away with it in a refereed venue, because the editor will make sure the work is sent to referees from the relevant ﬁelds. It’s also surprising just how knowledgeable academic audiences are about the most unlikely things.) Know your enemies Remember your enemies at this stage. The real enemies are the people you could be confused with. Differentiate yourself from them tacitly, not explicitly. If you mention them explicitly, then you have added an unwelcome item to the audience’s agenda, namely deciding whether you really are different from your enemies, as you claim. If you do it tacitly, then that item shouldn’t be an issue.
The keen beginner One enemy is the keen beginner who hasn’t read the literature in depth and who makes classic novice mistakes. You need to send out a strong signal that you don’t ﬁt into that category. The good old cup of coffee will help you to ﬁnd out about classic novice mistakes. The good old reading through the classic literature back to the year dot will help you with the literature in depth.
It’s surprising what you ﬁnd if you go back more than about 20 years – a lot of researchers are too hungry for fortune and glory to bother doing their 112 THE UNWRITTEN RULES OF PhD RESEARCH homework properly, and to discover that they are reinventing the wheel.
There’s no long-term substitute for reading the literature. The short-cuts in this book are about identifying the best ways to do things efﬁciently and well, not about ways of avoiding the literature. The literature, like research methods, is your friend; the more you have in your repertoire, the more situations you can tackle conﬁdently.
The intelligent layperson Another enemy is the intelligent layperson. If general knowledge is enough to tackle the problem, then you shouldn’t be bothering to tackle that problem.
This is where counter-intuitive results from the literature are useful.
The snake-oil merchant and the self-proclaimed genius Other potential enemies are the snake-oil merchant and the self-proclaimed genius, both of whom peddle their own patent panaceas. Claiming that you have solved a problem which has bafﬂed the best minds in the ﬁeld (including, by implication, the audience) is not something which will endear you to your audience, and is about as advisable as marching into a mediaeval Mongol war camp and shouting, ‘Genghis is a sissy!’ at the top of your voice.
Don’t show weakness or doubt You should never show weakness, apologize or ask for mercy. If you’ve done the work right, then there’s no need to apologize or ask for mercy. If you haven’t done the work right, then don’t present it – go and get it right instead.