How to write a dissertation

Richard N Griffiths

Faculty of Information Technology, University of Brighton.

A handout prepared for Information Systems and Computing  students.
 

Abstract

An inadequate understanding of standards of academic authorship is identified in previous undergraduate dissertation submissions.  In order to remedy this, a general schema for organising and presenting the material of a dissertation is proposed.  This is given in two ways; as exemplified by this document itself, and as a series of suggestions.  Variations on the basic schema to account for differing subject matter are indicated.  It is claimed that, whilst this is not the only valid approach that may be taken to the task, using it will at least lead to a comprehensible final product.

Introduction

The Problem

From past experience of reading undergraduate dissertations, and I have read a considerable number,  it has become clear that a significant proportion of final year students are not able to differentiate between the styles of writing appropriate to technical reports, journalistic articles and academic papers.  When they do differentiate, (and the dissertation is intended to be in the latter style) they may spoil what is basically a thorough piece of work, by inadequate structuring.

The precise characterisation of the style of an academic paper is beyond the scope of this document.  The best way to appreciate the genre is to experience it. Read a few!  What is presented here is an attempt to provide guidelines for the gross structuring of such a paper.  Inevitably, there will be topics and approaches for which the detail of the structure proposed is inappropriate, but it should be possible to accommodate these by modification.  At any rate - all papers should have an introduction, a discussion section organised in some way or other  and a conclusion

The Proposed Solution

A skeleton framework for sectioning a dissertation is given here, together with some suggestions for content.  The major sections of a dissertation should be- The requirements of each section are discussed  below.

Discussion

The Identification Section

This simply consists of the title of the dissertation, the name of the author (you) and a stereotyped note as to the status of the document, such as-

 ‘An undergraduate dissertation submitted in partial fulfilment of the requirements of IS315  the Information Technology Applications module of BA(Hons) in Computing and Information Systems.’

This may simply be laid out in print on a sheet of A4, or as some student’s choose, it may be ‘dressed-up’ with coloured graphics, pictures out of magazines, etc.  Over-done, this can create the impression that this is a light weight journalistic article.  The same goes for ‘over producing’ the design and layout of the document.  For certain topics, (e.g., ‘desk-top publishing’) going to town on the graphic design may be appropriate.  For others, restraint in the cause of taste, may be exercised.  Any illustrations used should be there to convey information, not as decoration.

The Abstract

This is the whole paper condensed to a paragraph.  (It could be longer, but one paragraph is usually most appropriate.)  It is important to include three things; the main problem addressed, an outline of the solution offered, and any conclusions reached.  Do not hold back on any of these — you are writing an academic paper, not a thriller, so giving the game away before the conclusion does not matter!

The Introduction

Here you present the question or problem that the dissertation solves.  If it is not obvious what question your dissertation answers, it will be difficult to decide what use it is (and at the end of the day, how to assess it).  Obviously, the scale of the question must be constrained by the resources at your disposal, of which time is a major limitation.  You are not required to be original, or to produce something of world-shattering importance.  (But don't let me stop you if you have got something like that to write!)

In this piece of work you have simply to demonstrate your competence at researching a topic within your professional domain  and presenting your findings in an appropriate manner.  If the answer to a particular question would be interesting to you or even better useful to you, then it is likely to make a good question on which to base the dissertation.  Do not assume, however, that the relevance or interest of the question is immediately obvious to other readers.  In this section, you should also motivate the question, that is, explain why it is interesting and relevant.

A final function for the introduction is to explain how the discussion has been structured.  It may be that several positions in an argument are examined, or a couple of case studies are presented and analysed.  The aim is to make sure the reader understands why you are giving any particular piece of information in the place that you do.

The Discussion

This is the main body of the dissertation.  It is where you explore the question that you have posed and give a chain of reasoning that will justify the conclusions that you present.

It should not be written as one block of text, but should be broken up into relevant sections.  Each section should probably be treated as for the whole dissertation.  i.e. it must have;  a descriptive title, an introduction that explains the question that this section answers and how it does this, discussion paragraphs that give the substance of the material and a conclusion that points out the detailed step that this section justifies — and which is going to be used in the overall discussion.  You may wish to draw attention to these sub-sections by giving them individual headings, but more often, this becomes cumbersome and they are better left simply as paragraphs.

Footnotes should be used sparingly, if at all.  Generally, all points should be made in the text of the discussion — and if they can not, then you must ask if they need to be made at all.

The Conclusion

Here you present your answer to the question posed in the introduction, together with a justification indicating how the chain of reasoning flows from the discussion.  You may wish to indicate any points that were not able to be resolved and suggest lines of further enquiry.

The Acknowledgements

If you received help from anyone in the preparation of the dissertation, then it should be acknowledged here.  If you have not received any help, then you have not approached the task effectively!  Almost certainly library staff will have helped you search indexes, teaching staff may have discussed points with you, contacts in industry may have sent you documents or given you interviews.

The Bibliography and References

It is vital  that you give full references to the literature that you have consulted. This is appropriate in an academic paper and you risk accusations of plagiarism if you do not! Use the Harvard System or other form of referencing. To indicate a reference in the body of the text, the name of the author, page and date should be given, say  (Smith, W; 1984  pp. 321).
Explicit reference should be given for both quoted text, which must  be distinguished by quotation marks, and any paraphrased text.  You may choose to further distinguish quotations by indentation, change of type face, etc.  Where you have abbreviated quoted text, the cuts should be indicated by three dots (...), called ellipsis.

Any source referred to must appear in the bibliography section, giving; the name/s of the author/s, the title, the date of publication, in the case of a journal paper — the name of the journal, in the case of a book — the publisher.  To refer to electronic media, see Li, X & Crane, N.

Conclusion

Guide-lines for structuring a dissertation have been given by example, (i.e., the structure of this document) and have been described.  It is suggested that, whilst not every dissertation will fit neatly into the sections given, most will, and that the result will be understandable as a whole.

Further pointers may be found in books on study skills, of which the library holds several, and also (Bundy, A. et al , 1984).

Acknowledgements

Lyn Pemberton was helpful in pointing me to the Bundy et al. article.
Aaron Sloman's  notes on writing dissertations and theses provided a model for this effort.

Bibliography

Bundy,A., Du Bulay,B., Howe,J. & Plotkin,G. 1984  "How to get a Ph.D. in AI", O'Shea,T. & Eisenstadt,M.  Artificial Intelligence: Tools, Techniques and Applications Harper & Row, New York

Li, X. & Crane, N.  Bibliographic Formats for Citing Electronic Information  [Online].  Available HTTP://www.uvm.edu/~ncrane/estyles/ [1995, June 8].