Centre for the Advancement for Learning and Teaching
Writing the Research Proposal
Amanda Everaert, Language and Learning (CALT) Faculty of Information Technology, Room 135 Bld 75
Linda McIver, Faculty of Information Technology
November 2006
Copyright © Monash University
Acknowledgements:
This booklet is based on work by Harriet Searcy and Linda McIver with acknowledgements to Sally Knowles, Murdoch University, WA.
CONTENTS
1. Introduction...................................................................................1
2. Getting Started..............................................................................3
3. Research Methodologies..............................................................4
4. Components of the Thesis...........................................................5
5. Steps in Writing the Thesis..........................................................7
6. From Proposal to Thesis: Structure............................................8
7. Proposal Sections.........................................................................9
8. The Research Context................................................................11
9. Sample Proposal.........................................................................13
10. Analysis of the Sample Proposal...............................................18
11. Useful References.......................................................................20
1. Introduction
The Honours degree is regarded as a kind of research apprenticeship. For this reason, and particularly in the technological and scientific disciplines, the Honours student is usually assigned a project by the supervisor. A
postgraduate degree by comparison, marks the possession of advanced knowledge in a specialist field, and the student is expected to make some substantial, independent contribution to the field.
The thesis the postgraduate student writes can be defined as ‘a proposition which expounds a particular point of view’: it is never just the simple
recording of research. In the honours thesis, your examiners will be looking for a demonstration of your knowledge of the chosen field, and especially: • understanding of the issues and developments in the research area • mastery of experimental procedures and design • mastery of techniques of data collection
• mastery of appropriate analytical procedures and interpretation.
A thesis is not the same as a research paper or technical report. It is usually longer, with a particular focus on the Literature Review component. Overall, it will be judged not only on its technical contribution (originality, technical competence, data analysis, algorithm design and/or software/hardware implementation), but also on the quality of the discussion.
University Requirements
The following information briefly compares thesis requirements at the Honours, Masters’ and Ph.D. levels:
Honours
Length of the thesis is open but is approximately 10,000 to 20,000 words (very long). Length is difficult to prescribe as it depends on the work involved, the degree to which the writer needs to explore other research areas, and
experience. It should be no longer than it needs to be in order to discuss what you have done, why you did it and what you have concluded from the results of your work.
There are a number of possible models, but the second, below, is the more common (at least in terms of Monash theses examined): 1. The student shows evidence of wide reading and demonstrates
understanding in the area. 2. The student shows evidence of wide reading and understanding and
builds an appropriate software product.
Master’s
A major Master’s thesis should show ‘independence of thought and
demonstrate [. . .] the ability of the candidate to carry out research in the selected field.’
A minor thesis should show ‘wide reading and understanding in an area related to the course of study being taken for the degree.’ The length of both theses is approximately 20,000+ words.
Writing the Research Proposal
1
Doctor of Philosophy (Ph.D.)
This thesis should present ‘a significant contribution to knowledge and demonstrate the candidate’s capacity to do independent research.’
It is difficult to specify the length of a Ph.D. thesis. One suggestion however, is that 250 double spaced pages would represent a maximum length.
Academic Writing
Style
It is wise to format and write your thesis in a fairly conventional style. If you are writing a postgraduate thesis and you know who your examiners might be, you may be able to break conventions or include more semi-formal commentary or graphics. Many examiners however, have fairly traditional notions of
appropriate academic style and therefore, you need to prepare accordingly.
Use formal academic writing style; avoid clichés, colloquial expressions and rhetorical questions. Edit each piece of writing carefully; do not rely on a computer program to make subtle grammatical distinctions or to use correct Australian (British) spelling!
Motivation: What do you want the reader to take away?
Make sure you clearly convey the ‘problem’, why it is worth solving and why it is worth reading about. One of the first questions most supervisors want
answered in any piece of project-related writing is ‘why am I bothering to read this?’
Few projects are purely theoretical. Try to relate your project to its real world outcomes: Who will it help if you solve this problem? What will it make easier/ cheaper/faster/more efficient? Can you make the motivation so clear that a non-specialist will understand why your work is useful?
NB: It is the responsibility of all students to make sure they understand the relevant university regulations and departmental guidelines. Please refer to these for full information.
Writing the Research Proposal 2
2. Getting Started ____
1. Clarify expectations with your supervisor and continue to do this throughout your honours/postgraduate year.
Identify and understand the research problem you are investigating. Identify the hypothesis you are going to test.
Determine the appropriate research methodology for testing the hypothesis. Keep in mind the uncertain, changeable nature of research; as you work through your project your aims and expected outcomes may change. Be flexible; remain open to surprises and prepared to change direction if necessary.
2. Read all conventions/regulations/departmental/guidelines covering your candidature.
Be sure of the conventions and regulations governing postgraduate studies at Monash.
Check with your supervisor and honours and masters coordinators about departmental/examiners' expectations.
3. Read other theses in the field.
Get a grasp of the standard you need to aim for.
Ask your supervisor to recommend a good thesis which relates to, or
provides a good perspective on, your topic. Discuss observations with others.
4. Prepare the Research Proposal. Set goals and establish deadlines.
Define the boundaries of your research so that you can manage the time
available to you. Honours students, remember that you will be combining coursework with other thesis requirements throughout semester 1.
5. Start writing early.
It is advisable to start writing as early as possible before the research for the entire thesis is completed. This purpose has been partly built into the Honours course in the form of an assessable Literature Review to be
completed before the end of semester 1. Postgraduates too, despite their longer time-frame, should also start the writing process early, especially if they have any concerns about the quality of their writing or if they have had little opportunity to write academic English.
There are many reasons for writing early:
It can help identify any language or conceptual problems so these can be
corrected early.
It gives a sense of achievement and progress early in the course.
It allows you more time to revise and improve writing, as well as reflect on ideas.
Writing the Research Proposal
3
3. Research Methodologies
Once you have identified the research problem and hypothesis you wish to investigate, you need to decide how you will evaluate the hypothesis or artifact created; in other words, the research methodology you will use. Computer Science is a new discipline and uses a number of different research methodologies originating from other disciplines.
Mathematics
• Formal proof of hypothesis from premises; for example, proof of the correctness of an algorithm.
• Defining and exploring a formal framework; for example, the Chomsky
hierarchy of formal languages
Engineering
• Building and evaluating a system; for instance, a new compiler. • Building and evaluating a piece of hardware.
Social sciences
• Surveys or focus groups for evaluating the usability of a user interface.
Natural sciences
• Modelling emergent behaviour of complex systems.
This list is not exhaustive. It is important that you and your supervisor identify a suitable research methodology for your project.
Exercise:
Identify the appropriate methodology or methodologies for your project. If none of the above are appropriate, describe the methodology you propose to use.
Writing the Research Proposal
4
4.
Components of the Thesis The following are standard components of most theses. Some are obligatory, but variations are possible within in the body of the thesis itself. Consult your supervisor in any case.
Obligatory Preliminaries
Cover page: displays thesis title, candidate's name, qualifications, thesis details, department, university, date
Declaration: states that the material has not been used for other awards from other institutions, and that all sources are acknowledged
Abstract: an approximately 300 - 500 word synopsis of the thesis, which
summarises the work and its findings in order to orient the examiners/ and then, later, other readers, to the thesis as a whole
Table of Contents: lists chapter titles including all major divisions, major sub-divisions, and other components
Lists of Tables/Figures/Illustrations: included as appropriate
Optional Preliminaries
Acknowledgements: recognises the help given by family, supervisors, or anyone who offered support or advice
Dedication: dedicates the thesis to a particular individual/s
NB: The Declaration normally follows immediately after the Cover Page. All of
the above sections are placed before the Introduction and the body of the thesis. The Contents Page only lists the Introduction and sections following the Introduction, including the Appendices.
Thesis Proper:
Introduction
This chapter provides a clear statement of the topic/problem under investigation. It generally includes:
• contextual background information
• a brief theoretical survey (usually the most recent/ relevant studies) • a statement of the problem or ‘gap’ in the research • a statement of aim/s
• brief methodology details and a brief description of your specific contribution (model, software etc.) • an Outline of chapters or Thesis Plan
The introduction is designed for the general reader. There is no need to go into great detail as you do this in the chapters following the Introduction. It is possible to include a literature review in the Introduction, if it is very short. The general trend however, is to treat it separately in the chapters that follow.
Writing the Research Proposal
5
Literature Review
This section may be made up of one or more chapters, and includes coverage of related work. It examines the theoretical background to the research topic from remote to relevant, and/or current studies. A good literature review evaluates research in the field as well as summarises it. Note that the term ‘Literature Review’ is not conventionally used as a section/chapter heading on the Contents page. You should choose a heading which reflects the content.
The rest of the thesis is made up of sequential chapters that focus on the thesis topic itself; that is, ‘the current study’ and its implementation. These chapters contain the student’s original research contribution.
Methods
Coverage of Methods may involve one or more chapters. In them, you should provide:
• a detailed description of the problem, and a justification of the solution • consideration of alternative solutions
• details of the system design and implementation
Results
This chapter often combines Results and the Discussion together, depending on the number of results covered. Your results can be reported in the form of tables, calculations, examples, or extended description.
Discussion
Discussion of the results will usually involve comparison with previous studies, explanations of the significance of your results, and claims about the success or otherwise of the methods involved.
Conclusion
This chapter:
• summarises the major findings, including unanticipated difficulties • considers the overall significance of the research, including any limitations
• concludes with recommendations for future developments, and/or suggestions for further work
Other Components:
Appendices: graphs, tables, data collection sheets, user's manual References/Bibliography: records all works that have been consulted Glossary of terms/Index: lists key terms/topics referred to
Writing the Research Proposal 6
5.
Steps in Writing the Thesis Choose a topic
Ð
Analyse the topic and assess its suitability (conduct a preliminary lit.
search, and/or preliminary experiments, review other theses)
Ð
Make a preliminary plan
Ð
Formulate a research proposal/timetable with practical
work milestones
Ð
Prepare the contents page
Ð
Survey reading and other resources, record bibliographic
information and make notes
Ð
Refine the plan
Ð
Draft the body
Ð
Draft the conclusion
Ð
Draft the introduction
Ð
Compile the appendices & references
Ð
Get a perspective on the thesis
Ð
Write abstract
Ð
Proof-read final draft
Ð
Submit thesis
NB: Note that the ordering of parts will naturally vary according to your personal needs and style.
Writing the Research Proposal
7
6. From Proposal to Thesis: Structure The Research Proposal: is the first formal piece of documentation for FIT4005 initiates the process leading to the submission of the thesis signifies the intention to perform, and write up, research aimed at solving
a specific problem explains the value of the research; why it is worth doing initiates the structure of the thesis but does not contain a results and
discussion section explains the relevance of the project in a way that non-experts can
understand
PROPOSAL STRUCTURE Table of Contents Introduction Research Context
Research Plan and Methods
THESIS STRUCTURE
Contribution of the Research
Introduction Literature Review System Design and
Implementation
________________ ________________ References (Appendices)
Writing the Research Proposal
8
Results and Discussion Conclusions and RecommendationsReferences
Appendices
7. Proposal Sections
Aim, Scope and Title
You need to be very clear about your aim/s before you begin writing the
Research Proposal. Your aim is the impetus which drives your work; without a clear aim you will not be able to decide on a title for your thesis, nor write an effective Introduction or Conclusions.
1. Write your aim/s:
To investigate deduction processes for intelligent learning database systems Too vague: in order to achieve what?
2. Write a problem or hypothesis statement: Problem:
Current deduction facilities for intelligent learning database systems are
inefficient this will help clarify your specific aim… Aim:
To create more efficient deduction facilities for intelligent learning database systems…
Hypothesis:
Graphical user interfaces are more usable than a textual interface this leads naturally to your aim… Aim:
To determine whether graphical user interfaces are more usable than a textual interface…
3. Define scope:
Problem: …by combining the techniques of fuzzy matching and probability
estimation
Hypothesis: …for scheduling software
4. Link title to aim:
Combining fuzzy matching and probability estimation to increase deduction efficiency in intelligent learning database systems.
Are graphical user interfaces more usable than a textual interface for scheduling software? Example:
Problem: Current approximation algorithms for the Maximum Induced Planar Subgraph Problem are sub-optimal.
Aim/Scope: The aim of this project is to design and analyse new approximation algorithms for the Maximum Induced Planar Subgraph
Problem. The behaviour of the algorithms will be compared with some existing approximation algorithms in terms of running time and performance.
Writing the Research Proposal
9
Title: The Design and Analysis of Approximation Algorithms for the Maximum
Induced Planar Subgraph Problem
Exercise: Draft a problem statement, aims statement and title for your thesis.
Introduction
Provides the background information a reader needs to understand your proposal. It:
sets your research in context; establishes the field in which you intend to work
introduces the problem and its importance
clearly states the aim/s and scope of your proposed research
Definitions/Glossary of Terms
May be included if you are using terminology which may be unfamiliar to the reader.
Research Plan and Method
Explains clearly how you intend to approach the work under the following headings:
Research method
Current project status (if work is already in progress)
Proposed thesis chapter headings (resembles a table of contents) Timetable (in table form) Facilities required Deliverables
References
is a section of the report so is given a section and page number lists all sources cited in the text
follows a recognized style; eg, Harvard, Bibtex
Writing the Research Proposal 10
8. The Research Context
Purpose
The purpose of the Research Context or literature review component of the proposal is:
• to provide theoretical frameworks for the research • to show results of related previous studies
• to show how ‘the present study’ fits in with previous research by: o making evaluative comments where appropriate o identifying limitations
o pointing out the gap your work will fill
Start reviewing the literature as soon as possible after discussions with your supervisor. Although it is not necessary to provide a lengthy or detailed review, it is a good idea to read as widely as possible in the early stages of research. To do so will give you:
• the knowledge about your field of inquiry necessary to justify the proposed
research
• clear ideas about suitable procedures for the research • more time to conduct other aspects of the research
Selecting sources
What should you read and note?
The research context provides an historical overview of the theory and
research literature that relates to your specific research topic. At the proposal stage, you should aim to review literature from recent journals, online
information banks, Internet resources, thesis and dissertation abstracts, and major books in the field if these are relevant. Your use of references needs to show the relationship between any previous research and your proposed research. As a rule of thumb, for an Honours thesis, you should consult a minimum of ten references, though some topics may not generate that number at the proposal stage. If you are having trouble narrowing down the scope of the review, consult your supervisor.
What constitutes a strong reference? How can you judge the quality of a paper?
Evaluate source material against the following criteria:
Citations: Who else cites it, and what do they say about it?
Originality: Is this a literature review style paper, or original research?
Try to cite original work whenever possible, or at least check that the original is accurately represented.
References: Who does the writer cite? Have they covered the area
thoroughly and understood it? Do their explanations, assumptions and facts agree with other established literature in the field?
Relevance: How is the information relevant to your own work? (Make sure you convey this clearly in your own writing)
Writing the Research Proposal
11
Relevant references will generally fall into one of the following categories:
Work which explains/describes the field (often a lit-review style paper) Work which your work will build upon, eg; work which had to be done/problems which had to be solved before you could do your
project, the steps which built the state of the art in the field up to where it is now, and from where your work will begin Work that your work will use, eg; methodologies
Work that has tackled the same problem from different angles
(sometimes the difference will be slight)
How can you be sure you have covered the literature in your field? The short answer is that you can never be entirely sure. However, there are strong clues. When all your sources cite each other, and there are few new references in the papers you are reading, you are probably doing well. You should find that the work begins to converge; new papers you find will tend to cite papers you have already found.
How do you know what the seminal work is in your field; the papers you must find? Easy; the work that everyone else cites. In most fields you will find that there are a few papers that nearly every subsequent paper refers to. Make sure you read these papers all the way through and do not rely on other writers’ interpretations. Even papers that represent the work accurately do not always represent it in its entirety.
Organising your reading
Begin by first developing a list of subject headings which relate to the
research topic. Using the headings as your focus, read the newer references before the older ones, and the general references before specific ones. This approach is likely to provide a good understanding of current problems in specific and related research areas, together with an understanding of the interrelationships between the particular research areas.
It is important when reading the literature to take careful notes and to keep an organised file of references (eg. card file, loose-leaf folder, computer files). When taking notes be sure to:
• include all bibliographical details, including page numbers
• distinguish quotations from paraphrases and your own comments
Referencing
The library’s Online Tutorials explain and give examples of the referencing styles used in IT. You can also learn how to use Bibtex here. http:www.lib.monash.edu.au/tutorials/citing
Writing the Research Proposal 12
9. Sample Proposal
School of Computer Science and Software Engineering Monash University
Bachelor of Computing Honours (0194)
Caulfield/Clayton Campus
Research Proposal - Semester 1, 1999.
Deduction of Induction Results
Name (ID: 100001)
Supervisors: ----------------
-----------------
NB: While the section headings of this document should be duplicated in your own Honours
project proposal, the exact format need not be. So don’t get tense about it!
Writing the Research Proposal 13
1.0 Introduction
1.1 Aims and Significance
Knowledge acquisition from databases can be achieved through an induction process. This produces knowledge in the form of rules or decision trees from individual data. Using deduction, the learnt
knowledge is then applied to solve new problems. However, there are difficulties associated with particular deduction types. Of the three possible outcomes in making a deduction,
1. no match; ie. where no conclusions can be made
2. single match; ie. where only one conclusion can be made 3. multiple match ie. where multiple, possibly contradictory, conclusions can be made
the no match and the multiple match are considered problematic. By combining the techniques of fuzzy matching and probability estimation, this project aims to create more efficient deduction facilities for intelligent learning database systems.
2.0 Research Context
Existing techniques for handling the no match case are based on probability estimation including the default class method (Clark and Ng, 1989; Wu 1993) and the Measure of Fit (Michalski, Mozetic, Hong and Lavrac, 1986). The Estimate of Probability (Michalski et al., 1986) is generally used for the multiple match case. These methods give good results where problem domains contain no real-valued attributes. However, in the case of real-valued attributes, the situation is more complicated. The standard approach for induction of these latter attributes is to discretise each one into a certain number of well-defined intervals. The problem with such discretisation is that people often refer to domains with some degree of fuzziness. For example, an individual could classify someone who is 170cm. in height as either “normal in height” or “tall” depending on his/her outlook.
Alternatively, instead of discretising the domain into fixed intervals, some sort of curve could be applied to give a fuzzy border. As a result, a value would belong to more than one interval at the same time. Discretising the domain in this way is termed fuzzy matching. Results can be classified in terms of degrees of membership of the interval. Hence the case of no match could now have a result, and the case of multiple match would give results with varying degrees of membership.
Fuzzy logic has recently been combined with Neural Networks to give improved problem solving results (Kosko, 1992). Experiments with HCV, an efficient induction algorithm, (Wu, 1993) have produced encouraging results in terms of accuracy and rule
compactness (Wu, 1995, 1996), based on experiments performed with data sets from the University of California at the Irvine
Repository of Machine Learning Databases and Domain Theories. However, in the case of no match, no rule fires and in the case of multiple results, there is no indication of preferred result when more than one rule fires. Wu (1995, 1996) has already
undertaken research in the area of the interpretation of induction results and the present study will extend on this work as outlined below.
Writing the Research Proposal
14
3. 0 Research Plan and Methods
3.1 Research Methods
The project will be developed from existing research on expert systems, machine learning and noise handling. Fuzzy matching will be combined with probability estimation to give deduction results for the cases of no match and multiple match. A library of fuzzy membership functions will be set up where the user can specify fuzzy borders and select
membership functions. A model of the system will be developed in C++. Experiments will be conducted using data from the University of
California at the Irvine Repository of Machine Learning Databases and Domain Theories.
Note that how data is classified into membership functions can give varying results. Thus the fuzziness of the borders for classification of information will lead to several experiments. In addition, finding domain specific information to run such experiments with, and interpret, the resulting fuzzy borders, may also prove problematic.
3.2 Proposed Thesis Chapter Headings
1. Introduction
(a) Purpose of Research (b) Objectives of Research
2.
Intelligent Learning Database Systems (a) Rules by Induction
(b) Deduction of Induction Rules (c) Probability Estimation
3. Fuzzy Interpretation
(a) Membership Function (b) Implementation (c) Results
4.
5. Bibliography
6.
7.
Appendix B Programs Appendix A Sample Data Conclusion and Future Work
Writing the Research Proposal 15
3.3 Timetable
Date Week No. Activity 10 March 2/13 First Draft Research Proposal to Supervisor 16 March 3/13 Research Proposal Finalised 20 March 4/13 Commence Literature Review 10 April 7/13 Prepare Draft Literature Review 27 April 8/13 First Draft for Literature Review to Supervisor 4 May 9/13 Literature Review Finalised 8 May 10/13 Thesis Chapter 2 – Current Situation 15 May 11/13 Commence Data domain Search 5 June 1/6 Commence Coding 29 June 4/6 Prepare for Seminar 1 6 July 5/6 Seminar 1 10 July 6/6 Draft Progress Report to Supervisor 24 July 2/13 Thesis Chapter 3 – Fuzzy Interpretation 10 August 4/13 Commence Progress Report 17 August 5/13 Draft progress Report to Supervisor 24 August 6/13 Progress Report Finalised 1 September 7/13 Finish all Coding and Data Testing 4 September 8/13 Thesis Chapter 1 – Introduction 12 October 12/13 Thesis First Draft to Supervisor 13 October 12/13 Prepare for Seminar 2 18 October 13/13 Seminar 2 20 October 13/13 Finalise Thesis 26 October 14 Thesis Final Product to Supervisor
3.4 Special Facilities Required
The facilities offered Honours students at Monash University (Campus) are sufficient to do the research.
4.0 Relevance of the Research
Existing work on deduction tends to exclusively focus on the cases of no match and multiple matching using probability estimation. This project will combine fuzzy logic with probability estimation to give improved deduction results for the cases of no match and multiple match. Given that businesses today accumulate enormous amounts of information which often go unused, the design of more efficient deduction facilities has intelligence gathering potential for intelligent business databases.
Writing the Research Proposal 16
5.0 References
Clark, P. & T. Ng. 1989. The CN2 Induction Algorithm. Machine Learning, 3: 261-283.
Kosko, B. 1992 Neural Networks and Fuzzy Systems: A Dynamic Systems Approach. Prentice-Hall.
Michalski, I. J. Mozetic, J. Hong & N. Lavrac. 1986. The multi-purpose incremental learning system AQL and its testing application to three medical domains. In Proceedings of AAAI 1986, 1041-1045.
Wu, X. 1993. The HCV induction algorithm. In Proceedings of the 21st ACMI Computer Science Conference, 168-175.
Wu, X. 1995,1996. ARC Application for Initial Support. Department of Software Development, Monash University, Melbourne.
NB: This list of references could be considered rather short in a typical proposal. At least 10 references are required. Check with your supervisor.
Writing the Research Proposal 17
10. Analysis of the Sample Proposal
Sample proposal: Deduction of Induction Results
Cover page
Contains the following information: department/school, university name, course name; eg. Bachelor of Computing (Honours), course code, campus, document type, date, title of document, candidate’s name and ID number, names of supervisor/s
Table of contents
List of numbered sections with headings and sub-headings
Introduction
This section of the proposal covers three areas: the proposed area of study, the problem background and the aims of the study.
The general principle observed in most academic writing is to start with the general and then work down to the specific. The introduction should begin by locating the field of research for the reader including brief background details, and then briefly introduce the specific problem area (or ‘gap in the research’). It is important here that the writer briefly indicate the limitations/difficulties specific to the study. This is because these limitations represent the gap/s in the research field which the proposal is attempting to fill. The sample writer signals this gap with the statement ‘Of the three possible outcomes in
making a deduction, . . . the no match and multiple match situations are considered problematic.’ From these points, the writer can then logically derive the aim/s of the project.
Research Context
The second section of the proposal focuses on the literature relevant to the research problem. In fact, it operates like a miniature Literature Review with five separate works being cited overall in the sample. The writer first overviews the research area by examining some of the classical
methodologies used and their limitations. The specific problem which the project examines is then signalled by the statement: ‘However, in the case of real valued attributes, the situation is more complicated.’
The second paragraph then opens up a discussion of possible solutions and supporting detail: ‘Alternatively, . . .’ . In the third paragraph, brief details are given about current experiments directly relevant to the research area. In the sample, the writer has included references to these, eg. ‘Experiments with HCV, an efficient induction algorithm, (Wu 1993) have produced encouraging results in terms of accuracy and rule compactness (Wu 1995).’ In other words, the writer is signalling the specific work/s that are the main basis of the proposed research. Typically, a researcher will seek to improve or extend previous research experiments. In this sample, the writer makes this move in the statement: ‘Wu (1995, 1996) has already
undertaken research in the area of the interpretation of induction results and the present study will extend on this work as outlined below.’
Writing the Research Proposal
18
Research Plan and Methods
One or more paragraphs. Opens with general statement/s about the research methods proposed; eg. ‘This project will be developed from existing research on expert systems, machine learning . . .’ Further statements might focus, for example, on design, software implementation, or proposed sources of data; eg. ‘A library of fuzzy membership functions will be set up where the user can specify fuzzy borders and select membership functions. A model of the system will be developed in C++.’ It includes a final paragraph which comments on possible limitations or potential
difficulties. In the sample, the writer points out how the particular nature of ‘fuzzy borders’ drives the need to conduct several experiments. In turn, the writer anticipates that finding appropriate data and interpreting these borders may prove problematic.
In the sub-sections that follow, the writer provides further information about his or her intended process:
Thesis Outline
A provisional table of contents with numbered sections. This is a very useful step in thinking about what you know or do not know in terms of the research.
Timetable
This is a list of tasks or activities involved and projected/required time for their completion. Various possibilities exist, but a table is the simplest. Do not waste time impressing readers with Gant or PERT charts at this stage.
Special Facilities Required
In this section, the writer needs to briefly state here whether departmental facilities are sufficient for the project; eg. ‘The facilities offered for Honours students at Monash University, Clayton, are sufficient to do the
research.’ If this is not the case, the supervisor will be the best judge about how you should proceed and the appropriate information to state here.
Relevance of the Project
The relevance of the project to the field should be outlined in a way that can be easily understood by interested non-experts. This could be revised later as part of the Introduction to your thesis.
References/Bibliography
The References section lists all sources the writer consults and cites in the text of the proposal. A Bibliography is a record of all works, books, journals or ephemera (ie; manufacturers’ specifications, etc.) that the writer consults in order to write the proposal, whether cited or not.
Writing the Research Proposal
19
11. Useful References
Anderson, J. & Poole, M. (1994) Thesis and Assignment Writing, (2nd edition), John Wiley and Sons, Brisbane.
Commonwealth of Australia, Style Manual for Authors, Editors and Printers, (4th edition), AGPS, Canberra.
Leedy, P.D. (1974) Practical Research: Planning and Design (4th Edition), Macmillan, New York.
Lewins, F. (1988) Writing a thesis: a guide to its nature and organisation, (2nd edition), Faculty of Arts, ANU, Canberra.
Mauch, J. E. & Birch, J. W. (1983) Guide to the Successful Thesis and
Dissertation.Conception to Publication: a Handbook for Students and Faculty, Marcell Dekker, Inc., New York.
Phillips, E.M. & Pugh, D.S. (1987) How to get a Ph.D.: A handbook for students and their supervisors, Open University Press, Milton Keynes; Philadelphia.
Turabian, K.L. (1987) A Manual for Writers of Term Papers, Theses, and Dissertations (5th edition), University of Chicago Press, London.
Writing the Research Proposal 20
因篇幅问题不能全部显示，请点此查看更多更全内容