Common mistakes and how to avoid them
Alaric Maude Having a topic that is too big.
Beginning PhD students often believe that they must tackle much bigger or hard-toresearch questions than could possibly be answered in a PhD, just because this is the way that questions are framed in the research literature that they read. But professional researchers in universities will typically have many more resources for tackling big issues (such as large budgets, sophisticated research technologies at their disposal, large cooperative research teams, or squads of people to assist them). What is a good question for professional researchers to address is not usually a good question for someone doing a PhD thesis in lone-scholar, no-budget mode. (Dunleavy 2003, p. 20)
But a topic that is too small makes it difficult for you to demonstrate that you have made a contribution to knowledge. Not being clear about your thesis topic. Try the ‘dinner party [pub/disco] test’.
The challenge posed by having to explain your thesis topic can also be a salutary stimulant to clarifying your own thinking. During the course of your doctorate there will be gruesome occasions, at dinner parties or drinks with strangers, when someone turns to you and asks what it is you do. Once you admit to working on a doctorate, your conversation partner’s inevitable fellow-through is to ask about your subject. From this point on you have typically about two minutes to convince your normally sceptical inquisitor that you know what you are doing and that it is a worthwhile thing to be at. … So the ‘dinner party test’ is always a frustrating experience to undergo, and many students feel that it is an impossible one for them to pass. To expect them to be able to capture the essence of their sophisticated and specialized topic, and to convey it in a few lines to a complete stranger, is just absurdly to underestimate what they are about. Yet in my view the test is a good one. If you cannot give a synoptic, ordinary language explanation in two or three minutes of what you are focusing on and what you hope to achieve, the chances are very high that in a very fundamental way you do not yet understand your thesis topic. (Dunleavy 2003, pp. 22-3)
Not being able to express your thesis aim as a question. In a thesis you develop, state and justify a question, and then develop an answer (or a partial answer) to that question. You should be able to state your thesis problem as a single question, but you may also have subsidiary questions that follow from the main one. In addition:
Do not include any elements in your research question that will not be addressed in substantive and (hopefully) original ways by your analysis. Do not have elements of your research analysis or evidence that are not covered by the statement of your key research question. (Dunleavy 2003, p. 276)
Not having a bit of a puzzle to solve.
You define the question: you deliver the answer. This proposition means that every effective PhD thesis should be genuinely personalized in some way. You should take a manageable part of the existing literature’s questions or concerns, and then tailor or modify that topic so as to shape it so that it can be feasibly answered. The way that the question is shaped should be reasonably distinctive, coming at a subject from a personally chosen angle. If you have such a personalized (even mildly idiosyncratic) perspective then it is less likely to be adopted by other researchers during the course of your studies. It is best to try and frame your thesis around an intellectual problem or a paradox, not around a gap. It needs to focus on a set of phenomena that ask for explanation, which you can express as a non-obvious puzzle and for which you can formulate an interesting and effective answer. … Many PhD students from countries or disciplines with more empiricist approaches, or placing more emphasis on intellectual or social consensus, find the idea of problematizing their thesis topic difficult and odd. They often regard their chosen topic as obviously worthy of study or intrinsically interesting and important in commonsense terms alone. They see no puzzle or enigma in front of them, merely an empirical landscape only partially painted by previous authors, which is their opportunity. This is a dangerous state of mind to be in at the start of a doctorate. It is often associated with people picking overly derivative topics important at some previous levels of education, or taking on very conventionally framed subjects from the existing literature which are too large or difficult to resolve in a PhD. Above all, an ‘unproblematized’ thesis topic normally provides students with no worthwhile intellectual focus or protection at the examination stage. It leaves open too many questions along the lines of: what is this thesis for? (Dunleavy 2003, pp. 23-4) Structure your thesis around a paradox, not around a gap — a principle for clarifying the central research question or problematic of your thesis. You should aim to explain a non-obvious puzzle in an original way, not just to produce the first description of something not already (extensively) studied. (Dunleavy 2003, p. 274)
Not delivering what you promise. The conclusion to the thesis must match the introduction. Some examiners read these two chapters first, and then the body of the thesis. Not being able to focus each chapter on a single purpose, such as the answer to a question.
Each of your chapters should do a discrete and distinctive job, well signalled from its start, and effectively building the thesis. … Assign one function to each chapter, and make sure that this role does not overlap with those of its neighbours. (Dunleavy 2003, p. 206) A key orientating device here is a rolling thesis synopsis of three or four pages. This document is for your own use and for your supervisors only. It greatly expands on your chapter plan or contents page by giving a paragraph of writing about what each chapter will say. The synopsis also expresses the main ‘storyline’ of your thesis. You should write your first synopsis as early as possible in your first year. Thereafter it is vital to keep revising it, so that it is permanently up to date and always captures your latest thinking. The whole point of a rolling synopsis is that you should never be writing or
2
working into a vacuum. As you work on one chapter you always need to have a paragraph or so about what later unwritten chapters will cover, and an accessible summary also of the key points made in chapters already written. The rolling synopsis should always concentrate on summarizing your substantive arguments and conclusions — what you have claimed, what you have found out, and what you hope to discover. (Dunleavy 2003, p. 53)
Describing your research methods without explaining how they help to answer the main and subsidiary questions in your thesis. You should be able to explain precisely what methods will be used to answer each question, rather than have a general account of all the research methods used in the thesis. Ignoring the ‘need-to-know’ principle.
‘Need to know’ criterion — a key principle to use in determining how much detail or information to include in your text. Ask: ‘What do readers need to know in order to follow and appreciate my argument?’ Provide only enough set-up or background information to meet this need. (Dunleavy 2003, p. 271)
Early chapters should provide concepts, ideas and methods or tools which can be applied in later chapters, so that the thesis is a connected and cumulative argument. One way to manage this problem is to think about what you would like to be able to say at the end of the thesis (e.g. ‘I would like to be able to say which of two competing theories is the better explanation of some phenomenon, and why’), and then work out what you have to find out and explain to be able to say this, what concepts, ideas and tools are needed, and what is the appropriate starting question that will lead to this answer. Designing your thesis backwards can save considerable wasted time and effort, and ensure that you can actually answer the research question, but this is not something you can do in the first few months of your research. The ‘need-to-know’ criterion should particularly be applied to literature reviews. The material in these reviews should either contribute to developing or justifying a research question, or provide concepts, ideas and tools that are used to develop answers. Literature reviews that are summaries, with little comment, evaluation or attempt at synthesis of different views. Also think carefully about where to put the literature review material. It doesn’t all have to be in one ‘blockbuster’ chapter, and some can go in the findings chapters. The appropriate division will vary from thesis to thesis. Insufficient chapters on your own findings.
So much then for the organization of the whole. But this section is also about the core of your thesis — which may be simply defined as all those sections with high research value-added. The core contributes to originality either by ‘the discovery of new facts’ or by ‘the exercise of independent critical power’. This set of chapters contains all the most substantively new or different sections of your research, the ones that determine if you get a doctorate or not. In a ‘big book’ thesis not all of your doctorate can or should fall into the core. There will also be a certain irreducible amount of non-core materials, composed of: Lead-in material, which introduces and sets up core material for readers so that it is
3
understandable and accessible. Sometimes dismissively labelled as ‘throatclearing’stuff, lead-in sections or chapters always require careful management. None the less they often loom much larger to students in terms of their length, and their writing and rewriting time, than their eventual role in the final thesis would justify. Readers often page through lead-in materials quite quickly, looking mainly for ‘the beef’ to be found later in the core sections. Lead-out materials do the ‘book-closing’ role for large theses, providing an integrating summation or restatement of what has been found, and setting it in a wider context. … To get a doctorate (and to do a good thesis more broadly) the size of the core matters a great deal. You must make sure that there are enough core chapters, and that they are big enough in terms of the total wordage of your thesis, to colour the whole thing as an original piece of work. My suggested rule of thumb for ‘big book’ theses is that 50,000 out of the 80,000 words of main text must be core materials. That is, appreciably more than half of your text should be original-ish stuff, reporting primary research that you have undertaken, or making new and distinctive arguments that you can plausibly claim to have originated or developed. This is a very demanding standard, but a therapeutic one. It throws into sharp focus the need to concentrate on your thesis’s value-added elements. Do not end-load a ‘big book’ thesis, leaving all the good bits squeezed into the last third or quarter of the text, as many people do. A recurring problem in most humanities and social sciences disciplines is that students spend so much time and effort on writing lead-in materials that they create a long, dull, low-value sequence of chapters before readers come across anything original. To check your own plan, count the number of chapters and the number of pages that readers must scan through before they come to the core. Overextending the lead-in stuff will also squeeze out the time needed to do your core research and write it up properly. Long ‘legacy’ chapters (often literature reviews or methods descriptions inherited from your first one or two years of study) also restrict the text space you have available to set out the core properly. (Dunleavy 2003, pp. 49-51)
Misallocating your time in reading and writing. Think of the implications of the advice above. Chapter conclusions which are section-by-section summaries of the chapter. Conclusions should be about what the chapter has contributed towards the development of the thesis. This could be the elaboration of some key questions, the identification of some concepts and tools that will be used in later analysis, or an answer to part of a research question. Too many subheadings. Breaking a chapter into lots of sections and subsections fragments the chapter and makes it harder to develop a coherent argument. Complex numbering systems like 3.2.3 and 3.2.3.1 can produce confusion rather than order, particularly if the headings are in the same font and therefore look the same. Try using subheadings with different font sizes or styles, without numbers, as is normal in journals and books. This provides a better indication of the level and importance of the heading. For example, Geographical Research has only two levels of heading: First order headings Second order headings
4
Repetition.
The ‘say it once and say it right’ approach urges you not to blur the argumentative impact of a single connected set of points about X by dissipating them in dribs and drabs, a little bit here and then again there and somewhere else a third time. Instead you should pull together all the related little ‘x’s into one, big bloc X argument. In weakly organized text this idea can again be a great force for good. Nothing is so corrosive of readers’ confidence in an author than the feeling that they are simply re-encountering material already described in a disorganized text, or are revisiting in only a marginally varied form points made already, perhaps for the third, fourth or fifth time. But some degree of linkaging back and forth across a text is inevitable and necessary. For instance, cross-referencing and short ‘reminder’ passages can often be justified on the ‘need to know’ criterion. Radically overdoing a ‘say it once and say it right’ logic may sometimes push an already well-structured text into inaccessibility, denying readers the ‘warm-up’ links that they need to grasp a wider pattern of argument. (Dunleavy 2003, p. 109)
Thinking that you will remember a good idea, and not writing it down immediately.
Jotting thoughts down whenever you have them is a second seemingly obvious but actually crucial aspect of increasing your creativity. Nothing is so evanescent as your own good ideas, so fleetingly present and so easily lost. One of the most famous social psychology articles sheds light on this issue, focusing on ‘the magical number seven, plus or minus two’. Empirical research shows that on average we can all of us hold only about seven ideas at the forefront of our attention. Very clever people are perhaps able to focus on nine ideas at once, while less adept people (like me) may only be able to concentrate on five ideas at a time. When we are confronted by larger sets or longer lists of ideas we tend to react by randomly dropping some elements from the forefront of our attention. Hence if you think of a lot of ideas without jotting them down, you may appropriately be anxious that you will forget them. (Dunleavy 2003, p. 35)
Carry a thesis notebook around with you, because ideas come at strange times and in strange places. Neglecting to keep a full record of everything you read. Keep a record of what you have read, where it is located, and any comments. See an example on the next page. Build your bibliography as you read, as a Word file. Inadequate backup system. Backup regularly, make two backups (such as an hourly or daily backup on a memory stick, and a weekly or monthly backup on a CD), and keep them in different places to the computer and each other. It pays to be paranoid. Reference Dunleavy 2003, Authoring a PhD: how to plan, draft, write and finish a doctoral thesis or dissertation, Palgrave Macmillan, Houndmills.
5
Community Regeneration Project References Reference Alterman, R. and Cars, G. (eds) 1991, Neighbourhood regeneration: an international evaluation, Mansell, London Badcock, B. and Harris, K. (eds) 1998, Revitalising housing areas: proceedings of the 1998 National Urban Renewal Seminar, Australian Housing and Urban Research Institute, Melbourne Bell, S. 2000, The unemployment crisis in Australia: which way out?, Cambridge University Press, Melbourne. Location Comment Action
File Concluding chapter ALTERMAN has been expanded into Alterman 1995. Limited relevance. Office Requested from DDS 8.8.01
Office
Computer file notes
Completed
Campbell, M. 2000, ‘Reconnecting the File long term unemployed to labour market CAMPBELL opportunity: the case for a Local Active Labour Market Policy’, Regional Studies, vol. 34, no. 7, pp. 655-668.
Notes in LALM folder
Completed
6