dissertation Writing research：Organising Your Dissertation

Taken from the Good Practice Guide:
Organising Your Dissertation
The guidelines below are quite detailed and specific. For some theses, it will be possibleto follow them to the letter. For others, the particular subject matter might require adifferent approach, which will require a different number of sections and a different wayof dividing up what you wish to put in. An example of this is that, in theses containinglanguage data, you may need a section separate from your HOW TO WRITTING ESSAY methodology section inwhich you describe the language/linguistic structures that you are dealing with. Useyour common sense about this; it should all fall into place quite easily if you rememberthat your aim is to be clear at all times, and constantly bear the reader in mind in yourwriting and organization. This is the time to look at other theses and see how topicssimilar to your own have been set out in the past.
The TitleThe title should give a clear indication of what the thesis is about for the sake of otherscholars wishing to read it; however, it should not be so narrow that, if the focus ofyour inquiry shifts slightly during execution, it will no longer be an appropriate title. Thefollowing guidelines are taken from the Faculty of Education's handout for M.Ed.students and the points made offer helpful advice for all writers of research projects:"The title should be brief, usually not more than 15 words. It should be in goodEnglish, properly punctuated, without any unnecessary technical terms. It isseldom necessary to include such words as 'A study of...' or 'An investigationinto...' at the beginning of the title or words near the end like 'with specialreference to...'. Phrases such as 'a comparative study...' should be omitted
unless the whole point of the title would otherwise be lost. Normally a title whichsimply links two apparently unrelated concepts with the word 'and' isunsatisfactory, e.g. 'Leisure and school'. The title should not describe thesample, the design, the tests, etc. but should indicate the area of study and thetype of child or institution studied so as to be cross-classified by library users
(e.g. 'cognitive', 'young children', 'nineteenth century')..."These suggestions, of course, are general, and there may be reasons for departingfrom the general rules.
Abstract
Although this comes first, you should write it at the end. It should consist of a brieffactual account of the study, giving the important details and including an indication ofthe findings.
Acknowledgements
Make sure you include everybody who made the work possible; notably subjects,services, schools, (they may be appearing anonymously in your text so make anappropriate anonymous acknowledgment) not only your supervisor but other academicstaff who may have helped with references, statistics or even providing data, technicalstaff who have assisted with or provided equipment, perhaps your fellow students fortext-checking, ideas, or being controls in your experiment, and the typist.#p#分页标题#e#
aken from the Good Practice Guide:http://www.man.ac.uk/goodpractice/
Doing this is a salutary experience, because it reminds you how much collaboration isrequired to complete even a small project.
Aims & IntroductionThis is a short section, usually a few pages long, in which you set out the context andaims of your project including the specific research question(s) which your project
addresses. It can include rather universal statements about why you think this topic isnteresting/needs researching. Then you can set out in a rather general way what thestudy is about and how it fits in to the research context. You will of course cover all thisin the required detail in other sections, but this general introduction is necessary forthe reader, and writing it may be a good way of getting your own thoughts running.Writing down aims at an early stage will also help you design your experiment, as ithelps you focus on the most relevant aspects of your study.If you write it first, be prepared to re-vamp it when you have finished, to make sure itis still relevant.
The Literature Review
Having chosen your topic, you need to read around it and keep careful note ofreferences, e.g. on a card index which you might later hand to the typist if it is in thecorrect form.
There are several different ways of finding all the literature that you need to include.You may find that your search may be more manageable and relevant if you onlysearch articles of the last 10 years. You may know some titles already from the courseor from your supervisor individually. These serve as a 'way in' to the topic. Read otherpapers referred to in the text. Ideally, use as recent an article as possible to provideother references. When you read a cross-referenced article, jot down where you firstsaw it cited, either on your file card or with your notes and record what the first authorsaid about it. That way, you can begin to sort out the cross-connections in the field.The more you read, the easier it will be to determine what is considered importantwork in our own chosen area. For example, so-called 'seminal' articles may be referredto in almost every paper you read. Also, you will be able to develop your own critical
powers if you notice carefully how researchers evaluate each other's work.
A second method is to scan through books and journals in your chosen area. Looking atthe contents page and then the abstracts of, say, the last five years' editions of theBritish Journal of Audiology and a few others, is a simple but very effective way ofgetting started. Remember to look at both the bound volumes of periodicals and theunbound current issues.
Yet another method is to use the various bibliographical services on offer in the Library
- e.g. citations indices, collections of abstracts, computer searches for writings in yourtopic area. The library staff will help you to choose what is appropriate for your aimsand explain how to use the resources. However, if you choose this method, you mustdo it early, as it takes a long time, for several reasons. Obviously you will not have thetexts to hand, as you do in the other two methods, so allow yourself at least two hoursinitially and preferably a half-day to find the titles you have turned up in the library.Also, you will come across books or journal articles which are not stocked by our#p#分页标题#e#
Taken from the Good Practice Guide:http://www.man.ac.uk/goodpractice/library. You can order them from Inter-Library Loan (there is a charge) but remember
there will obviously be a time lag between ordering and obtaining such material.Literature reviews will vary greatly. In some cases, there will be a few highly relevant
studies that you wish to summarise in detail. In other cases, there will be a huge areawhich is slightly relevant but does not have direct bearing on your own particular study.Do be selective and don't include all and sundry; e.g. if you are measuring stapedialreflex latencies as part of your experiment, do not include all that has been written ontympanometry. Also do not go into detail on old work which has long been superseded.If you are in any doubt about the relevance of some of the literature to your piece ofwork, you can write a sentence or two explaining why you think a certain area isrelevant, at the beginning of the section where it is summarised.Ideally, your review should be evaluative and critical of the studies which have aparticular bearing on your own. For example, you may think a particular study did notinvestigate some necessary aspect of the area, or that the authors failed to noticesome problem with their results. You may feel hesitant about criticising published work,but there is no need to; just because a study has been published, that does notnecessarily mean it is superior to the one you are doing - nor that other authors areinfallible!
Your supervisor will advise on the scope of the literature review, but one generalcaution: if you write the literature review early on, you must check it again later in thewriting-up process to make sure it is all relevant. Don't include everything you haveread just because you have read it. Try and structure your written review into sectionsor even separate chapters to make it more readable.
One good reason to get the literature well under control at an early stage is so that youcan check on all the references. It is incredibly easy not to write all the appropriatedetails on your card, and sometimes you may require a page reference, to make adirect quotation, or refer to a particular table of results. You will probably havephotocopies of the papers you are referring to most closely, but if you do not, there isnothing more infuriating than finding a missing page reference or incomplete quotationwhen you get to the typing stage - and then to find the book itself is out on loan fromthe library. None of this should happen, of course, if you are meticulous aboutreferences.
Materials, apparatus & subjects
In this chapter you should accurately describe the hardware, software and human
elements of your experiment. For equipment, give model numbers of equipment and
explain how samples were obtained. When describing subjects, remember to include all
relevant details. These may include age, hearing loss, educational background,
language level and possibly other features, depending of course on your topic.#p#分页标题#e#
Calibration of equipment should be included in this chapter, giving the standards to
which you are working. You are recommended to put block diagrams of your calibration
test rigs in the text but the calibration data may be placed in an appendix.
Taken from the Good Practice Guide:
http://www.man.ac.uk/goodpractice/
Page 4
Method
This is probably the most important chapter of your thesis and could contain
materials/apparatus/subjects as a subsection rather than this being a separate chapter.
It is usually a good idea to write this chapter before you start data collection and let
your supervisor read it. It can then be modified and time will be saved rather than
starting off your experiment and later realising your mistakes. Even so, there may be
problems your supervisor has not envisaged and your method may need further
modification.
You may begin your method chapter with your overall hypothesis and then proceed to
your description of what you did, (or intend to do - however, do write in the past
tense).
When describing your method, it is also usual to write in the passive form (e.g. 'Twenty
children were tested...' rather than 'I tested twenty children...'). There may be
occasions when you need to write 'I'; or you can use 'the author' if you wish to retain
distance. For example, you may include such details as 'All the children tested were
pupils of the author at XY school.' Don't worry if your method chapter sounds
repetitive; this is often necessary in the interests of an unambiguous description.
If your method is drawn from another study, or replicates another person's work, make
sure you refer to this. If the method of proceeding is particularly complex, a summary
table of events or a flow chart might help to clarify it. For some topics, this might be
the appropriate place to include a brief description of the statistical analysis performed
on your data. Others topics might require this to be included in 'Results and analysis'.
Results
In this chapter, the classical approach is described: to merely report results, without
comment. This does not mean that you do not need any text to describe data
presented in tables and figures. Explanatory text should be used to identify and
describe the tables and figures of relevance. Tables of numbers are useful, but if these
are copious and not all interesting, be selective and include the rest in appendices.
Your tables should each be given a heading and be numbered. Below each table, you
should briefly summarise the statistical results, e.g. 'A paired 'T' test showed there to
be a significant difference between groups A and B at the 5% level (see Table X).' You
will need another run of numbered headings for figures. Other methods of presentation
of results may also be particularly useful, e.g. graphs, histograms and pie charts. Make#p#分页标题#e#
sure they are labelled (including all axes) and numbered.
Not all theses work out quite like this; in many projects involving language analysis, it
may not always be possible (or advisable) to report results without comment.
Nevertheless, however 'messy' your data looks, it is a good idea to start off with the
structure suggested above. With language projects, you may have to make some
rather difficult decisions about how much language data to include within the text. The
key principle is, as always, clarity for the reader. If in doubt, ask your supervisor.
Taken from the Good Practice Guide:
http://www.man.ac.uk/goodpractice/
Page 5
Don't worry if your results aren't what you predicted or don't show anything you
expected. Such results are just as valid (some might say more valid!) as predictable
ones. Your thesis will not be failed because your experiment did not have a particular
outcome. For example, if hearing aids A and B gave the same results despite their
different frequency responses, that is a valid product of your method, not evidence of a
wrong method. Don't be tempted to pick out interesting bits of the findings to try and
prove a point if the evidence isn't there - it's called 'data-rigging' and only charlatans
do it!
Discussion
Here, you summarise and comment on your results. This is likely to be a fairly lengthy
chapter and shows how you are able to relate your findings to your own ideas and the
work of others. You should discuss the shortcomings of your project, why you think
your results are what they are and what might have been done in a more ideal setting.
Don't forget to give references, even though you may have given the same ones
several times before.
In this section you will discuss the extent to which your research question(s) was/were
answered, whether your initial hypothesis was supported or not, giving some more
qualification and explanation. You may also wish to relate your finding to practical
applications in the real world. However, don't be tempted to make over-reaching and
grandiose claims for such a small piece of research, this will merely give the impression
of naivete.
Conclusions
Your main conclusion should be to say whether your initial hypothesis was correct or
not, though you may need some qualification here:
e.g. "In conclusion, the null hypothesis was disproved for groups A and B, but for
group C, insufficient data was obtained."
Then briefly relate this to existing published research.
Reference List
The aim of referencing at all is to make sure the person reading your thesis can trace
all the information you quote by going back to your original sources. This should
usually be a straightforward matter if you include a bibliography entry for every
reference in the text. The best way to check is to have a spare photocopy of your work,#p#分页标题#e#
go through the text and highlight all the references. If you don't want to bother with a
spare copy, a small pencil cross in the margin for each reference can be made. Having
done this, you can go through your reference cards and sort out the works that you
have actually referred to in the text.
It isn't a good idea to use secondary references in your work, i.e. don't rely on
somebody else's version of a given study but always go back to the original where
possible. On rare occasions, you might not be able to, e.g. an author might quote an
unpublished or otherwise generally unobtainable document. In this case, include both
the secondary reference and its source in the text, and include both references in full in
the reference list:
Taken from the Good Practice Guide:
http://www.man.ac.uk/goodpractice/
Page 6
Smith (1967, quoted in Brown, 1989 : 126) claimed....... (references in the back
then in full for Brown AND Smith).
The example above includes the page number (126) so that the reader can turn
straight to the citation in Smith, if wished. Please note also that when you quote an
author's exact words, the page number should be given, e.g:
... Reviewing the area of central auditory dysfunction, these authors conclude
there is sufficient evidence to support the view that delayed language and
classroom difficulties are sometimes the result of an underlying central auditory
dysfunction.... (Bamford and Saunders, 1991: 244)
You might like to consider the reasons for using a quotation rather than paraphrasing
an
author's views, as described in 9.1.2. In general, you should use quotations rather
sparingly, and only when you feel this is essential: for example, the author may have
used a particularly apposite phrase, or you may wish to quote a complicated conclusion
to avoid being unclear or misrepresenting an author. The text should be predominantly
your own writing, with occasional quotations: it should not look like a sequence of
quotations bound together by your comments.
The list of references at the end of your thesis should follow one type of accepted
referencing system. Again, all the research papers and books you have read will use
such a method. The university does not insist on one particular system; the important
point is to be consistent and include all the details as required. References consist of:
i) the name of the author(s)/editor(s) (remember to indicate which by adding
(ed) or (eds) in brackets) for editors;
followed by (for books):
ii) date (this sometimes appears nearer the end of the citation)
iii) title of book
followed by:
iv) Place of publication and name of publisher (these may be the other way
round but BE CONSISTENT yourself).
v) Sometimes there is an edition number. You need to include this because the#p#分页标题#e#
pages may be numbered differently in a second edition and the reader of your
thesis may hunt in vain for a particular reference you have given.
Taken from the Good Practice Guide:
http://www.man.ac.uk/goodpractice/
Page 7
EXAMPLES:
Bamford, J and Saunders, E (1991) Hearing impairment, auditory perception and
language disability. (2nd edition) London: Whurr Publishers.
Volterra, V. and Erting, C.J. (1990) (eds) From gesture to language in hearing and deaf
children. Berlin/Heidelberg: Springer-Veriag.
--------------------------------------------------------------------------------
For journal articles, the author is followed by:
i) date
ii) title of paper
iii) name of journal (best avoid abbreviations, but if you use them, use the
standard ones AND include a list of your abbreviations with a key at the
beginning of the reference list)
iv) volume number, possibly part number, and page numbers. Note that
you must give page numbers for a journal article.
EXAMPLE
Tyler, R.S. Wood, E.J. and Fernandes, M. (1982) Frequency resolution and hearing loss.
British Journal of Audiology, 16, 45-53.
--------------------------------------------------------------------------------
Notice that the titles of books and journals receive the same type, (italics in this case).
The titles of journal articles do not.
Some papers appear in edited collections. If you only refer to one paper in an edited
collection, the most economical type of reference is thus:
EXAMPLE
Wightman, F.L. (1 981) Pitch perception: an example of auditory pattern recognition. In
Getty, D.J. and Howard, J.H. (eds) Auditory and visual pattern recognition. New Jersey:
Lawrence Eribaum (3-25).
--------------------------------------------------------------------------------
However, if you are referring to more than one paper from the edited collection, you
may wish to conclude after... and Howard, J.H. (eds) - and then include the full citation
for Getty and Howard as a separate entry in the bibliography. Note, too, that, as with
journal articles, you should include page numbers here.
The above illustrations were all in one standard format. Here is the same reference
given in different formats:
Taken from the Good Practice Guide:
http://www.man.ac.uk/goodpractice/
Page 8
EXAMPLES
Piaget, J. (1952) The language and thought of the child. London: Routiedge and Kegan
Paul.
PIAGET, J. 1952: The language and thought of the child. London: Routiedge and Kegan
Paul.
Piaget J 1952 The language and thought of the child. Routiedge and Kegan Paul,
London.
Finally, you may wish to make a reference which appears not to have a written source:#p#分页标题#e#
... It has been suggested that (J. Bloggs, personal communication)
--------------------------------------------------------------------------------
You may find it difficult to believe that anybody reading your thesis will be so interested
in these minutiae. They will. Don't forget that your supervisor and examiner will know
most of the published work in the field already and anything that you refer to, they will
either know already or want to know more about.