Three little-known keys to writing a thesis
Experience and research have uncovered three keys to writing a thesis, as the following quotations (with emphases added) show.
- Have a `thesis' of the thesis
A review of 139 examiners' reports ... revealed that rarely were theses criticised for `bad writing' in the sense that most people understand that phrase. That is, theses were acceptable in terms of the mechanics of presentation: sentence structure, paragraphing, spelling, grammar, etc. They also were not criticised for failing to conform to conventions of the discipline about referencing or presentation of data. What frequently was criticised was the students' failure to take a clear philosophic stance or to reach a conclusion. Examiners called upon students to state clearly their hypothesis and their conclusions. If students adequately communicate the `thesis' of their dissertations, they usually avoid unnecessary length, lack of coherence, repetitiousness and confusion in their writing.
Supervisors need to emphasise throughout students' candidacies that they are striving in the thesis to communicate one big idea; that there should be a `thesis' or centre to which everything in the document contributes. (Nightingale 1992, p. 174)
- Have a research problem which is gradually refined as the thesis is written
Educational research and our own experience ... suggest that it is extremely important for the beginning researcher to define the research problem at a very early stage in the research process. Defining a research problem is often found to be a most difficult and frustrating task. The reason for this lies primarily in the fact that undergraduate students are by and large not compelled to define the problems they work on; such problems are presented to them by lecturers, and the notion that defining and articulating a problem is a demanding intellectual process in its own right is often poorly developed amongst undergraduate students. Yet it is a crucial preliminary step in the research process, and one which the postgraduate student, who has recently emerged from the security of undergraduate life where problems appear to exist self-evidently, must confront and overcome. It if is not, and the research proposal remains vague and ill-defined, the student's subsequent activities of researching and note-taking will lack focus, be more time-consuming than is necessary, and largely ineffective. (Zuber-Skerritt & Knight 1992, p. 196)
- Start writing a first draft early, based on preliminary conceptual maps
Another crucial phase in the research process is the transition from analysis to synthesis; that is, from the collection and analysis of literature or data to the writing of the first draft. Many postgraduate students attest to the psychological difficulties they must overcome before writing of the first draft can proceed; for many the task appears insuperable, and much time can be wasted at this point as the student prevaricates and justifies this prevarication by asserting the need to continue the phase of analysis. Most supervisors have heard the plaintive cry: 'I still haven't read enough!'; this is frequently a symptom of nerves as the awesome moment approaches when the student must lay aside the security of index cards and plunge into the writing phase. Our experience suggests that problems particularly arise when the postgraduate student is unaware of the stages and steps through which research and writing normally proceed. This manifests itself as an attempt to write a final draft without the intermediate steps of constructing a flowchart of ideas (or a conceptual map), writing a first rough draft, revising and editing, and then rewriting. In the attempt to move immediately to writing the final draft, the student becomes preoccupied with the fine details, stylistic niceties and attractive presentation, often at the expense of development of ideas or argumentation; as a result, the writing process is inhibited, and the product is often characterised by unevenness of thought and argument. (Zuber-Skerritt & Knight 1992, p. 200) [That is, constructing drafts of a flowchart of the sections of a chapter or the subsections of a section is useful early in the process of reading the literature, with several consequent revisions. Rarely has a student not read enough to start writing the first draft of these frameworks.]
Another trap for student writers is that they believe they need long periods of time if they are going to try to write anything. Waiting for the significant piece of free time to come along makes procrastination easy. Two helpful strategies are to encourage students to set attainable sub-goals so they use short periods of time efficiently. For instance, rather than trying to write the whole section on methodology, a student could set the sub-goal of writing only the description of a key piece of equipment. Of course, if she or he had been writing all along, there would be at least a rough draft of this which would simply need to be refined.
Another helpful strategy for writers who often face interruptions to their work is to leave themselves `pick-up points'. This means that they do not work until they are at the absolute end of something, but quit when they can still see what will come nest. They jot down a few notes about what they expect to write next, and when they come back, there is no blank page facing them.
Finally, supervisors who have several research students or whose departments have a group of novice researchers should encourage them to exchange drafts of their work frequently. The more commentary, the more often a student is asked, `What did you intend to say here?', the better the chances of a well-constructed thesis.(Nightingale 1992, pp. 176-177)
References Nightingale, P. 1992,
`Initiation into research through writing', in Zuber-Skerritt,
O. (ed), Starting Research - Supervision and Training, Tertiary
Education Institute, Brisbane.
Zuber-Skerritt, O. & Knight, N. 1992, `Problem definition and thesis
writing - workshops for the postgraduate student', in Zuber-Skerritt,
O. (ed), Starting Research - Supervision and Training, Tertiary
Education Institute, Brisbane.
Appendix II
Research proposal structure keyed to the thesis structure
At many universities, candidates in PhD programs are usually required to present a research proposal during or at the start of their candidature. This note provides a suggested outline for a proposal that fits with the structured approach to presenting theses, based on experience and Poole (1993) and Krathowl (1977). The centre headings in capitals are required by QUT, with the recommended side headings being my interpretation of what is required.
As a rule of thumb, the proposal should be a minimum of about two or three pages and a maximum of about seven to ten pages in length (with the list of references and any appendices of support material not being included in this page count), so the estimates of word and page lengths given below are very tentative. The proposal could have about twenty or so references. Sometimes a QUT proposal requires details of a `coursework' unit and an example of a tailored one is provided in a note at the end of this appendix. Please remember to check spelling and to provide page numbers at the middle top of each page.
Any research proposal should be carefully tailored to the organisation asking for it, so the format should below always be adjusted to suit other requirements.
OBJECTIVES OF THE PROGRAM
1 Introduction
The introduction is a five- to ten-line picture of the whole research, showing the major controversies or gaps in the literature which leads to the research problem. This description may become section 1.1 in the final thesis.
The research problem is presented at the end of this section, in italics and indented. Note that readers of a research proposal cannot be expected to know the jargon of every discipline, and so the title and research problem should be expressed in as simple terms as possible, and any specialist terms should be defined in this section as they are introduced.
2 Justification for the research
This section is about one page and justifies the research, usually on four dimensions:
- size of the industry involved,
- gaps in the literature (provide several references in support and refer to section 3 below),
- unusual methodology to be used (provide several references in support and refer to section 4 below), and
- possible benefits of outcomes for policy and for practice.
This section becomes section 1.3 in the thesis.
RELATION TO PREVIOUS RESEARCH
3 Preliminary literature review and theoretical framework
In about two pages, show the major issues and schools in the literature and the gaps in the literature, and then briefly justify some likely research questions (for qualitative research) or hypotheses (for quantitative research) arising from the gaps which may be the focus of data collection and data analysis. A model of the theoretical framework along the lines of Sekaran (1992, chapter 3) would be impressive. This section becomes the later parts of chapter 2 in the thesis. Define key terms as the section progresses or have a definition sub-section.
Incidentally, having read a completed PhD thesis similar to the planned one is a good inspiration and guide for the task ahead.
4 Contribution of the research
A contribution is a change to a body of knowledge created by a research project. This section complements section 2 above about the justification for the research. This section describes the specific outcomes of the research developed in chapter 5, and describes their importance. For example, it discusses a model which will be developed to fill the gaps in the body of knowledge noted in section 2, or a checklist which will be developed for managers who have no guides at present. In brief, this section is specific about likely outcomes and their importance. A candidate could also mention a conference at which a paper about the research could be presented, such as the annual conference of the Australia and New Zealand Association for Management (ANZAM) or the Marketing Educator's Conference. As well, the title of a journal which might publish an article about the research could be mentioned.
5 Limitations
Outline and justify the major limitations that will be placed on the research, for example, industry, level of management, states, etc. No claim for generalisability will be made beyond these limits. This section could be kept to about one third of a page. This section becomes section 1.7 of the thesis.
6 Background of researcher
This is a brief section outlining any pilot studies that the researcher has done, and his or her research qualifications and experience, for example, titles, methodologies and word lengths of dissertations.
RESEARCH METHODS AND PLAN
7 Methodology
This section would be between one half and one page in length. It should be both comprehensive and concise, with references to support its judgements. The methodology usually does not need to be described, merely justified. But again, avoid jargon that non-specialists might not know, or explain or describe what is meant by specialist terms. Topics could include:
- justification of a quantitative or qualitative paradigm;
- justification of the methodology within that paradigm (with preferably an explanation why some alternative methodologies were not used), using terms such as correlational/causal, field studies/field experiments/lab experiments, cross sectional/longitudinal (Sekaran 1992, chapter 4), instruments and especially the unit of analysis, the dependent variable and how it will be measured, sampling frame and size of sample;
- arrangements for access to the data, for example, agreements from people to be interviewed; and
- possible ways of analysing data, for example, conjoint analysis with SPSS or the matrix method of analysing qualitative data suggested by Miles & Huberman (1985).
This section should have at least three references to textbooks or articles about methodology, to justify the proposed steps. Moreover, a proposal for a quantitative methodology should indicate that operational definitions of the constructs in the proposed hypotheses of section 3 above, have been considered (for example, how `firm size' will be measured). In addition, scales and their accompanying statistical test should have been thought through (for example, a rank scale needs a nonparametric test). Tables of these considerations would be helpful.
This section becomes chapter 3 in the thesis.
TIMETABLE
8 Thesis outline
One or two lines per chapter should suffice, especially if the standard five chapter structure will be used. This section becomes section 1.5 of the thesis.
9 Timetable
The timetable could be shown for each chapter, for convenience. The same rules of thumb could be used for time as for length, that is, 5, 30, 20, 25 and 20 percent of the desired word length; for example, chapter 2 would take about 30 percent of the available time - but allow two months at the beginning for settling in and at the end for putting the finishing touches to the whole thesis, and an extra couple of months if there is a lead-in or follow-up study to the main methodology. Table II.1 is a rough guideline for a minimum time PhD which follows these principles. Phillips & Pugh (1987, p. 74) also have a usual timetable for a PhD program which is not very much different to table II.1's. This section and the next ones are in the proposal only and are not in the completed thesis.
Table II.1
Approximate guidelines for writing a minimum time PhD thesis
| Chapter or section | Topic | % | Words | Months |
| 1 | Introduction | 5 | 3,500 | 3 |
| 2 | Literature review | 30 | 21,000 | 6 |
| 3 | Methodology | 20 | 14,000 | 4 |
| 4 | Data analysis | 25 | 17,500 | 5 |
| 5 | Conclusions and implications | 20 | 14,000 | 6 |
|
| 100 | 70,000 | 24 |
RESOURCES
10 Resource requirements
A tentative estimate of direct funding requirements is required, for example, postage for survey mailings. Justifications and sources of estimates are required for each expense item, for example, the date a quote was received from Qantas or a price list of a computer supplier. Printing costs of questionnaires will require estimates of their length and the price per page to print. There should be no surprises in the budget items, for they should flow naturally out of the earlier sections about the aims and design of the research. For example, car hire should not be just costed, but why car hire was necessary rather than public transport should be explained and its use related back to overall aims of the research.
This estimate is not the formal request for the funding, and acceptance of the proposal does not include approval of funding. If outside funding is being used, make it clear that academic integrity will not be jeopardised.
11 Evaluation
Many research proposals require details of how the project will be monitored and evaluated, but these details are not necessary for a PhD proposal. If you want to gild the lily of your proposal by adding an evaluation section you might to mention that `The research will be monitored through weekly or fortnightly meetings with the supervisor and at regular thesis-in-progress seminars. The thesis will be evaluated through normal examination procedures which will be organised by the supervisor.'
List of references
Krathwohl, D.R. 1977, How to Prepare a Research Proposal,
University of Syracuse,
Miles, M.B. & Huberman, A.M. 1985, Qualitative Data Analysis, Sage,
New York.
Phillips, E.M. & Pugh, D.S. 1987, How to Get a PhD, Open University
Press, Milton Keynes.
Poole, M.E. 1993, `Reviewing for research excellence: expectations,
procedures and outcomes', Australian Journal of Education, vol. 37,
no. 3, pp. 219-230.
Sekaran, U. 1992, Research Methods for Business: a Skill-Building
Approach, Wiley, New York.
Note
Adding a `coursework unit' to a proposal is supposed to force the candidate and supervisor to think about what will happen after the proposal is approved. One appropriate unit is an `advanced readings' unit in one or two of the bodies of knowledge which the PhD will cover. The example below of a tailored unit covering the two parent disciplines of market segmentation and advertising illustrates the core elements involved ranging from aims, through objectives, and a program of topics, to assessment details and criteria. The topics are usually selected from a modern, authoritative textbook and recent review articles.
Doctoral Research - Coursework Unit
Coursework unit. Independent readings in the areas of marketing segmentation and advertising.
Objectives.
- To review the review of literature and identify major themes about:
- the process of marketing segmentation, targeting and positioning
- the bases for marketing segmentation, with particular reference to the use of age as a segmentation base
- identification and measurement of market segments
- the use of segmentation in developing marketing and advertising strategies
- advertising appeals and their relationship to geographic, demographic, behaviouristic and psychographic bases of segmentation.
Areas.
Topic No. Week Topic
Section 1: Developing market segments
1 1 Mass marketing versus marketing segmentation
2 2 Introduction to factors used to form segments
3 3 Segmenting consumer and industrial markets
4 4 Requirements for effective segmentation in marketing
(these requirements in advertising are covered in section 3.)
Section 2: Describing, measuring and choosing segments; implementation
5 5,6 Describing segments - customer profiles, size and growth
estimates
6 7 Evaluating market segments
7 8 Selecting market segments
8 9 Positioning strategies
9 10 Choosing and implementing a positioning strategy
Section 3: Marketing segmentation and advertising
10 11 Appeals used by advertisers
11 12,13 Advertising appeals and their relationship to bases of
segmentation
12 14 Advertising effectiveness: measurement and evaluation
Outcomes. A 3000-4000 word report on each section, upon completion of each.
Assessment. The supervisor will give a grade for each report. The criteria used for assessment shall be:
- breadth of reading
- ability to group core ideas in the literature
- ability to synthesise ideas into a new and coherent framework
- ability to communicate and present the report to the reader.
Preliminary coursework reading list
Aaker, D.A., Shansby, J.G. 1982, `Positioning your product', Business
Horizons, May- Jun, pp.56-62.
Aaker, D.A., Stayman, D.M., Hagerty, M.R., 1986, `Warmth in
advertising: measurement, impact, and sequence effects', Journal of
Consumer Research, vol. 12, Mar, pp.365-381.
Abrams, B. 1982, `Middle generation growing more concerned with
selves', Wall Street Journal, Jan 21, p.25.
Albright, J. 1992, Creating the advertising message, Mayfield,
Mountain View.
Bertrand, K. 1989, `Market segmentation: divide and conquer',
Business Marketing, Oct, pp.48-54.
Burnett, J.J. 1981, `Psychographic and demographic characteristics of
blood donors', Journal of Consumer Research, Jun, pp.62-66.
Bonoma, T.V., Shapiro, B.P. 1983, Segmenting the Industrial Market,
Lexington Books, Lexington, Mass.
And so on for a total of 28 references.
Appendix III
Action research in a PhD thesis
Action research as a methodology for management PhD research is relatively rare (Perry 1991). Moreover, although action research has the potential to overcome many deficiencies in social science research, its results are generally viewed as not generalisable (Heller 1986). This appendix reviews a number of issues which candidates using action research might consider when writing their PhD thesis. The appendix attempts to ensure that action research is no longer a marginal backwater depending sometimes on very carefully selected examiners, but becomes a part of the river of PhD research. Action research is outlined in Kemmis and McTaggart (1988a), and Zuber-Skerritt (1991).
Firstly, it is wise to consider the thesis as something distinctly separated from the action research project, that is, the candidate will have two projects - the action research project and the thesis project which uses data from the action research project (Perry & Zuber-Skerritt 1992). The philosophy and processes of action research are broader and more complex than those implicit in most PhDs. In particular, the action research project is relatively unfocused, emphasises practice and has outcomes of reflections which include propositional, practical and experiential (group and personal) knowledge. In contrast to action research, a PhD thesis project usually emphasises an individual candidate's additions to propositional knowledge published in the literature of a discipline. In brief, in the action research project, action research may be an ideology, but in a PhD thesis it is merely a methodology. Writing a PhD thesis about an action research project without acknowledging differences between the thesis and the action research project is difficult.
Provided these differences are acknowledged, the structure of a five chapter PhD thesis can be adapted to PhD research using the action research methodology. For a start, the 'research problem' in chapter 1 of the thesis could be different to the 'thematic concern' (Kemmis & McTaggart 1988b, p. 9) of the action research project; the research problem necessarily refers to practices of a workgroup and is written in terms of the literature of a discipline, but the thematic concern is less restricted. For example, a research problem could be 'How can the senior management team at an open-cut coal mine integrate marketing, operations and financial subsystems in the planning of inventories of mined coal?', and the thematic concern of the senior management group at Pacific Coal could be 'How can our inventory management procedures be improved?' The action research project will probably require multidisciplinary solutions, but it is advised that the thesis should concentrate on only one or two disciplines, to facilitate its examination.
Chapter 2 of the thesis written about an action research project would refer to some unresearched areas of propositional knowledge which are the foci of the data collected from the action research project. However, to be true to the spirit of action research, these propositions should not have been finalised before the action research project began - unlike PhD research using some quantitative methodologies, when the hypotheses should be crystallised before the data collection project begins. Furthermore, chapter 2 could outline the boundaries of practical and experiential knowledge which existed at the start of the action research project. Alternatively, the discussion of practical and experiential knowledge might be restricted to an appendix, if likely examiners are not expected to be familiar with action research methodology.
Chapter 3 could be used to describe the action research project - not to allow replication of the experiment, but to demonstrate the researcher's competence in the action research methodology. The chapter could have sections or refer to appendices which contain the following details of the action research project (Kemmis & McTaggart 1988b):
- the names of group members;
- the group's thematic concern;
- details of the multiple sources of data, for example, dates of meetings and their attendees and matters discussed, reports and letters;
- the distinctions between the stages of the project through its one or more cycles of plan-act-observe-reflect;
- the group's published report of the project - which is written before the thesis is completed and for a different audience from the thesis, for example, this could be a short narrative or a management report;
- the evidence that the group has reflected on processes as well as content, which might be recorded in the group's published report noted above but does not have to be; and
- the nature of the action research, that is, technical, practical or emancipatory (Carr & Kemmis 1986).
As noted above, an appendix might also reflect on the practical and experiential knowledge gained in the action research project, but it would be more usual to include that reflection in the body of the thesis.
Chapter 4 could be used to categorise the data collected in the action research project (not all of which needs to be included in the appendices referred to in chapter 3). This chapter organises the data from the action research project into patterns. Chapter 4 begins the candidate's own preliminary reflection on the action research project and could be divided into sections according to the propositions of propositional knowledge, and into sections for practical and experiential (personal) knowledge if they are to be included in chapters of the thesis rather than in appendices. So the chapter should be written with the ideas to be developed in chapter 5, in the candidate's mind.
Finally, chapter 5 makes conclusions about the full PhD research, linking the data of chapter 4 to the boundaries of the body or bodies of knowledge outlined in chapter 2. A section in chapter 5 entitled `Reflections on methodology' should be included in a PhD thesis which refers to an action research project . Then sections `Conclusions about the research problem', `Policy implications' and `Further research' will conclude the thesis. In PhDs using other methodologies, a chapter 5 section of reflections on the methodology is not required, because those reflections are incorporated into the `Limitations' and `Further research' sections.
In conclusion, an action research methodology can be used in PhD research, but action researchers should be concerned that their thesis may be messy, inconclusive and be unrelated to propositional knowledge published in the literature of a discipline. Use of the adjusted five chapter format for a PhD thesis which has been outlined in this appendix may allay that concern.
References
Carr, W. & Kemmis, S. 1986, Becoming Critical: Education, Knowledge
and Action Research, Falmer, London.
Heller, F. (ed) 1986, The Use and Abuse of Social Science, Sage,
London.
Kemmis, S. & McTaggart, R. (ed), 1988a, The Action Research Reader,
(third edition), Deakin University, Geelong.
Kemmis, S. & McTaggart, R. (ed), 1988b, The Action Research Planner,
(third edition), Deakin University, Geelong.
Perry, C. 1991, 'Action research in management education and
research', Symposium on Action research at the Annual National
Conference of the Australian Association for Research in Education,
Gold Coast, November 1991.
Perry, C. & Zuber-Skerritt, O., 1992, `Action research in graduate
management research programs', Higher Education, vol. 23, pp. 195-208.
Zuber-Skerritt, O. (ed) 1991, Action Research for Change and
Development, Gower, Aldershot.
Acknowledgment: Discussions with Ortrun Zuber-Skerritt, Bruce Frank
and Helen Samujh helped clarify some issues in this appendix.
However, the views expressed are the writer's.
Appendix IV
Frequently referred to pages of Style Manual (Australian Government Publishing Service 1988)
Topic Page
Adjectival possession 290
Bullet points 226
Contractions and abbreviations 96
Dates 172
Foreign words and phrases (for example, i.e.) 105-106
Harvard referencing style 129-148
He/she and gender issues 121
Headings 220
Hyphens and prefixes 81-86
Lists 226
Names of countries 107
Names of people 107
Omissions using three points 88
Paragraph indentation and spacing 224
Possessive apostrophe 80
Punctuation in quotations 94
Quotations 25 and 230
Spacing between paragraphs 224
Tables 231-237
Titles 51
APPENDIX V
Writing an abstract
An abstract should contain (based on Brown et al. 1993):
- 1 What did you do?
The research problem, and the hypotheses - 2 Why did you do it?
Briefjustification for the research problem and the hypotheses - 3 What happened?
Methodology - 4 What do the results mean?
Patterns in the data - 5 What is your work good for?
Conclusions and implications, with special and explicit consideration of the CONTRIBUTIONS
Note: as a rule of thumb, about 250 words is the maximum length of an abstract for a journal article, two pages for an honours and masters thesis, and three pages for a PhD thesis.
Reference
Brown, R.F., Pressland, A.J. & Rogers, D.J. 1993, 'Righting scientific
writing: focus on your main message', The Australian Rangeland
Journal, vol. 15, no. 2.
APPENDIX VI
Writing the introduction to a journal article
Swales (1984) studied introductions of journal articles and established four major moves or steps were required:
Move 1: Establish the field
Assert centrality State current knowledge Move 2: Summarise previous research
Move 3: Prepare for present research
Indicate a gap Raise a question Move 4: Introduce present research
State purpose Outline present research An example and an exercise about constructing an introduction are in Nightingale (1992, pp. 110-116). References
Nightingale, P. 1992, 'Writing about research in the humanities and social sciences', in Zuber-Skerrittt, O. Manual for Conducting Workshops on Postgraduate Supervision, Tertiary Education Institute, University of Queensland, Brisbane. Swales, J. 1984, 'Research into the structure of introductions to journal articles and its application to the teaching of academic writing', in Williams, R. & Swales, J. (eds), Common Ground: Shared Interests in ESP and Communication Studies, Pegamon, Oxford.
