Refine the words to spell success

The end of your PhD may be in sight but, warns Harriet Swain, the first draft is only the start of the writing-up process. Editing, revision and a careful look at presentation are key to producing a good thesis

Thank God that's over. Four years' worth of PhD research down on paper and now all that's required is a bit of an edit. So maybe it is not over at all?

In his book Authoring a PhD , Patrick Dunleavy, professor of political science at the London School of Economics, says it takes up to six months to reorganise the elements in a first draft to produce the final text. This will involve drawing out the intellectual themes of your work and ensuring that they run consistently through the thesis title, abstract, introductory chapters and conclusions. You need to think about how your thesis will look to readers and to assess what your research has achieved. He also advises that you think about how your project, your discipline and the wider intellectual world have changed or are likely to change since you came up with your working title. "Is the current title going to have the same fashionable connotations it once did? Is it going to stand up in the future?" he asks.

You will need to make your thesis more readable by checking spelling, grammar and phrasing, as well as font sizes of headings and subheadings, the referencing format and the labelling of figures and illustrations.

Lisa Ganobcsik-Williams, founding co-ordinator at the Centre for Academic Writing at Coventry University, says you will need to leave a good few weeks for all this. "Editing is notoriously picky work, for which one must have untired eyes and a well-rested brain," she says.

In the rush to meet a submission deadline, you can be too involved in the content of the thesis to be able to edit the sentence structure and phrasing effectively, she warns. One way to overcome this is to set aside a couple of weeks to check mechanical issues such as spelling. By the time you return to editing, you should be removed enough from your research baby to see where changes are needed.

Mary Lea, senior lecturer in teaching and learning at the Open University, suggests that you read your work aloud because it helps you to appreciate whether your writing makes sense and to think about connections between different parts of the text.

She says that it is a good idea to pay attention to any sentences that refer back to something else and to make sure the reference is clear. It is also important to check citations for ambiguity. "You cannot rely on the examiners knowing the work of all the authors you are referring to," she says.

Ganobcsik-Williams recommends saving these small-scale edits until you have finished your thesis, but she says you shouldn't leave everything to do with presentation until the last minute. She advises that you should find out what the conventions are for writing a thesis in your discipline early in the process by talking to your supervisor and looking at theses produced by others. It saves time if you have adhered to your university's guidelines for thesis presentation in early drafts. Writing your thesis should be a constant process of drafting and revision, she says. Between ten and 20 drafts of each chapter is typical.

Lea says that if you revise and edit your work as you go you will get a clearer understanding of what you are trying to say.

She advises always keeping a potential reader in mind. While your supervisor knows about your research and is familiar with your literature and methodology, the examiners will need to be persuaded that your approach is valid.

Rowena Murray, reader in the educational and professional studies department at Strathclyde University and author of How to Write a Thesis , says that you will need to explicitly state how your work makes a contribution to knowledge. You should point this out in your conclusion and look back through all the chapters to ensure that they add up to this conclusion. Some postgraduates feel their argument is coherent and do not want to spell it out, she says. Others are worried that their contribution to knowledge is too modest or obvious. Both sentiments could affect the success of the thesis.

Dunleavy says that the concluding sections of the main chapters need to link to themes discussed in the opening chapter. "The theme that each conclusion links to should be wholly relevant to the specific materials in that chapter and also adapted to the role that the chapter plays in the thesis as a whole," he advises.

He warns against trying to cover all the themes in every chapter, which can look too plodding. Instead, you need to ensure that each chapter does a discrete and distinctive job, effectively building the thesis. "Assign one function to each chapter and make sure that this role does not overlap with those of its neighbours." His tip is: "Say it once; say it right."

The main text of the thesis should be about four fifths of the formal university word limit, leaving the rest for the bibliography, notes, appendices, preface and acknowledgements.

If you need to cut, Dunleavy advises trying to make edits in the introductory chapters to safeguard the core of the thesis. He also stresses the need for an explicit cut-off date, especially if you are dealing with current events. "Trying to finish a thesis about events that are still going on or are not yet capable of being evaluated is an effort doomed to failure," he warns. If you find yourself making substantial changes and dropping what you thought were important concepts in favour of new themes at this stage, don't worry, he says. You can see what works and what doesn't only when you see the complete thesis. But you should not lose faith in what you have done or be tempted to go off in another direction entirely, he adds.

You may not be quite finished, but the end is in sight.

Further information

Patrick Dunleavy, Authoring a PhD: How to Plan, Draft, Write and Finish a Doctoral Thesis or Dissertation , Palgrave Study Guides, 2003.

Rowena Murray, How to Write a Thesis , second edition, Open University Press, 2006.

Top Tips

Leave yourself time

Check for coherence and repetition

Spell out your contribution to knowledge

Think about your reader